Molecular Imaging: Principles and Practice 1607950057, 9781607950059

Citation preview

MOLECULAR IMAGING Principles and Practice Ralph We is s le de r, MD, P HD Professor of Radiology and Systems Biology Harvard Medical School Director, Center for Systems Biology Massachusetts General Hospital Boston, Massachusetts

Brian D. Ro s s , P HD Professor of Radiology and Biological Chemistry Co-Director Center for Molecular Imaging University of Michigan Medical School Ann Arbor, Michigan

Alnawaz Re he mtulla, P HD Ruth Tuttle Freeman Research Professor, Department of Radiation Oncology and Radiology Co-Director Center for Molecular Imaging University of Michigan Medical School Ann Arbor, Michigan

Sanjiv S. Gambhir, MD, P HD Virginia & D.K. Ludwig Professor of Radiology and Bioengineering Director, Molecular Imaging Program at Stanford (MIPS) Director, Canary Center for Cancer Early Detection at Stanford Chief, Division of Nuclear Medicine Stanford University Stanford, California

2010 PEOPLE’S MEDICAL PUBLISHING HOUSE–USA SHELTON, CONNECTICUT

People’s Medical Publishing House–USA 2 Enterprise Drive, Suite 509 Shelton, CT 06484 Tel: 203-402-0646 Fax: 203-402-0854 E-mail: [email protected] © 2010 Ralph Weissleder, Brian D. Ross, Alnawaz Rehemtulla, and Sanjiv S. Gambhir All rights reserved. Without limiting the rights under copyright reserved above, no part of this publication may be reproduced, stored in or introduced into a retrie val system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording, or otherwise), without the prior written per mission of the publisher. 09 10 11 12 13/PMPH/9 8 7 6 5 4 3 2 1 ISBN-13 978-1-60795-005-9 ISBN-10 1-60795-005-7 Printed in China by People’s Medical Publishing House of China Copyeditor/Typesetter: diacriTech; Cover Designer: Mary McKeon Sales and Distribution Canada McGraw-Hill Ryerson Education Customer Care 300 Water St Whitby, Ontario L1N 9B6 Canada Tel: 1-800-565-5758 Fax: 1-800-463-5885 www.mcgrawhill.ca

United Kingdom, Europe, Middle East, Africa McGraw Hill Education Shoppenhangers Road Maidenhead Berkshire, SL6 2QL England Tel: 44-0-1628-502500 Fax: 44-0-1628-635895 www.mcgraw-hill.co.uk

Brazil Tecmedd Importadora e Distribuidora de Livros Ltda. Avenida Maurilio Biagi 2850 City Ribeirao, Rebeirao, Preto SP Brazil CEP: 14021-000 Tel: 0800-992236 Fax: 16-3993-9000 Email: [email protected]

Foreign Rights John Scott & Company International Publisher’s Agency P.O. Box 878 Kimberton, PA 19442 USA Tel: 610-827-1640 Fax: 610-827-1671

Singapore, Thailand, Philippines, Indonesia, Vietnam, Pacific Rim, Korea McGraw-Hill Education 60 Tuas Basin Link Singapore 638775 Tel: 65-6863-1580 Fax: 65-6862-3354 www.mcgraw-hill.com.sg

Japan United Publishers Services Limited 1-32-5 Higashi-Shinagawa Shinagawa-ku, Tokyo 140-0002 Japan Tel: 03-5479-7251 Fax: 03-5479-7307 Email: [email protected]

Australia, New Zealand Elsevier Australia Locked Bag 7500 Chatswood DC NSW 2067 Australia Tel: +61 (2) 9422-8500 Fax: +61 (2) 9422-8562 www.elsevier.com.au

India, Bangladesh, Pakistan, Sri Lanka, Malaysia CBS Publishers 4819/X1 Prahlad Street 24 Ansari Road, Darya Ganj, New Delhi-110002 India Tel: 91-11-23266861/67 Fax: 91-11-23266818 Email:[email protected] People’s Republic of China PMPH Bldg 3, 3 rd District Fangqunyuan, Fangzhuang Beijing 100078 P.R. China Tel: 8610-67653342 Fax: 8610-67691034 www.pmph.com

Notice: The authors and pub lisher have made every effort to ensure that the patient care recommended herein, including choice o f drugs and dr ug dosages, is in accord with the accepted standard and practice at the time of pub lication. However, since research and re gulation constantly change clinical standards, the reader is urged to check the product infor mation sheet included in the package of each dr ug, which includes recommended doses, w arnings, and contraindications. This is particularly important with new or infrequently used dr ugs. Any treatment regimen, particularly one involving medication, involves inherent risk that must be weighed on a case-by-case basis against the benef its anticipated. The reader is cautioned that the pur pose of this book is to i nform and enlighten; the infor mation contained herein is not intended as, and should not be emplo yed as, a substitute for indi vidual diagnosis and treatment.

Acknowledgments The editors would like to acknowledge the extraordinary contributions of Tania Cunningham, Judy Schwimmer, and Melissa Carlson in the preparation of this te xt. Collaboratively, they assumed the responsibility for or ganization and completion of the chapters of this te xt for which the editors are g rateful. RW BDR AR SSG

iii

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii

1

General Principles of Molecular Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Sanjiv S. Gambhir PART I: MOLECULAR IMAGING TECHNOLOGIES 2

Imaging of Structure and Function with PET/CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

David W. Townsend 3

PET/MRI. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

Marcus D. Seemann 4

SPECT and SPECT/CT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

Brian F. Hutton, Freek J. Beekman 5

Principles of Micro X-ray Computed Tomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

Shaun S. Gleason, Michael J. Paulus, Dustin Osborne 6

Small Animal SPECT, SPECT/CT, and SPECT/MRI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

Neal H. Clinthorne, Ling-Jian Meng 7

Instrumentation and Methods to Combine Small Animal PET with Other Imaging Modalities . . . . . 99

Craig S. Levin 8

Functional Imaging Using Bioluminescent Markers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

Christopher H. Contag 9

Optical Multimodality Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

Arion F. Chatziioannou 10

Fiber Optic Fluorescence Imaging. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

Rabi Upadhyay, Umar Mahmood 11

Fluorescence Tomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160

Vasilis Ntziachristos

v

vi

Contents

12

Endomicroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165

Seok (Andy) H. Yun, Charles P. Lin 13

Intravital Microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176

Thorsten R. Mempel 14

Diffuse Optical Tomography and Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

David R. Busch, Britton Chance 15

Ultrasound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225

F. Stuart Foster, Kevin Cheung, Emmanuel Cherin 16

Molecular Photoacoustic Tomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237

Lihong V. Wang 17

Optical Projection Tomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244

James Sharpe 18

Potential Roles for Retrospective Registration in Molecular Imaging . . . . . . . . . . . . . . . . . . . . . . . 262

Charles R. Meyer, Hyunjin Park, Bing Ma, Boklye Kim, Peyton H. Bland PART II: CHEMISTRY OF MOLECULAR IMAGING 19

Chemistry of Molecular Imaging: An Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277

Silvio Aime, Giovanni Battista Giovenzana, Enzo Terreno 20

Radiochemistry of PET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304

Henry F. VanBrocklin 21

Radiochemistry of SPECT: Examples of

99m

Tc and

111

In Complexes . . . . . . . . . . . . . . . . . . . . . . . . 327

Hank F. Kung 22

Nanochemistry for Molecular Imaging. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337

Yun Xing, Jianghong Rao 23

Newer Bioconjugation Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353

Claude F. Meares 24

Targeted Antibodies and Peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362

Michael R. Lewis, Cathy S. Cutler, Silvia S. Jurisson 25

Hyperpolarized

13

C Magnetic Resonance Imaging—Principles and Applications . . . . . . . . . . . . . . 377

Jan Henrik Ardenkjær-Larsen, Klaes Golman, Kevin M. Brindle 26

Magnetic Resonance Imaging Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389

Elisenda Rodriguez Vargas, John W. Chen

Contents

27

vii

Optical Imaging Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405

Scott A. Hilderbrand 28

Ultrasound Contrast Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425

Mark A. Borden, Shengping Qin, Katherine W. Ferrara 29

Multimodality Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445

Weibo Cai, Xiaoyuan (Shawn) Chen 30

“Click Chemistry”: Applications to Molecular Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471

Neal K. Devaraj, Ralph Weissleder 31

The “One-Bead-One-Compound” Combinatorial Approach to Identifying Molecular Imaging Probes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 481

Ruiwu Liu, Olulanu H. Aina, Ekama Onof iok, Kit S. Lam 32

Chemical Biology Approaches to Molecular Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 497

Stanley Shaw 33

Theranostics: Agents for Diagnosis and Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 509

Jason R. McCarthy 34

Magnetic Nanoparticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 523

Andrew Tsourkas, Lee Josephson 35

Fluorocarbon Agents for Quantitative Multimodal Molecular Imaging and Targeted Therapeutics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 542

Samuel A. Wickline, Ralph P. Mason, Shelton D. Caruthers, Junjie Chen, Patrick M. Winter, Michael S. Hughes, Gregory M. Lanza 36

Aptamers for Molecular Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 574

Bertrand Tavitian 37

Nonclinical Product Developmental Strategies, Safety Considerations, and Toxicity Profiles of Medical Imaging and Radiopharmaceuticals Products . . . . . . . . . . . . . . . . . . . 589

Sunday Awe, Siham Biade, Sally J. Hargus, Tushar Kokate, Adebayo Laniyonu, Yanli Ouyang PART III: MOLECULAR IMAGING IN CELL & MOLECULAR BIOLOGY 38

Overview of Molecular and Cell Biology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 604

Harvey R. Herschman, Hidevaldo B. Machado 39

Systems Biology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 628

Gregory Foltz, Leroy Hood 40

Protein Engineering for Molecular Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 644

Anna M. Wu

viii

41

Contents

Phage Display for Imaging Agent Development. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 660

Kimberly A. Kelly 42

Molecular Imaging of Gene Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 673

Maria Veronica Lopez, Qiana L. Matthews, David T. Curiel, Anton V. Borovjagin 43

Developing Diagnostic and Therapeutic Viral Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 689

Khalid Shah 44

Cell Voyeurism Using Magnetic Resonance Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 703

Naser Muja, Christopher M. Long, Jeff W. M. Bulte 45

Tumor Vasculature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 726

Ambros J. Beer, Gang Niu, Xiaoyuan (Shawn) Chen, Markus Schwaiger 46

Imaging Hypoxia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 756

Ashley A. Manzoor, Hong Yuan, Gregory M. Palmer, Benjamin L. Viglianti, Mark W. Dewhirst 47

Molecular Imaging of Protein–Protein Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 781

Tarik F. Massoud, Ramasamy Paulmurugan, Pritha Ray, Abhijit De, Carmel Chan, Hua Fan-Minogue, Sanjiv S. Gambhir 48

Fluorescence Readouts of Biochemistry in Live Cells and Organisms . . . . . . . . . . . . . . . . . . . . . . 808

Roger Y. Tsien 49

Imaging of Signaling Pathways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 829

Mahaveer S. Bhojani, Brian D. Ross, Alnawaz Rehemtulla PART IV: APPLICATIONS OF MOLECULAR IMAGING

Oncology: 50

Molecular and Functional Imaging of the Tumor Microenvironment . . . . . . . . . . . . . . . . . . . . . . . . . 844

Kristine Glunde, Robert R. Gillies, Michal Neeman, Zaver M. Bhujwalla 51

Novel MR and PET Imaging in the RT Planning and Assesment of Response of Malignant Gliomas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 864

Christina Tsien 52

PET Diagnosis and Response Monitoring in Oncology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 875

Rodney J. Hicks, Richard L. Wahl 53

Magnetic Resonance Spectroscopy Treatment Response and Detection . . . . . . . . . . . . . . . . . . . . 896

Sarah J. Nelson, John Kurhanewicz, Daniel B. Vigneron 54

Diffusion MRI: A Biomarker for Early Cancer Treatment Response Assessment . . . . . . . . . . . . . . 912

Brian D. Ross, Craig J. Galbán, Charles R. Meyer, Alnawaz Rehemtulla, Thomas L. Chenevert

Contents

ix

Cardiovascular: 55

Myocardial Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 925

Heinrich R. Schelbert 56

Congestive Heart Failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 941

Antti Saraste, Marcus R. Makowski, Stephan Nekolla, Markus Schwaiger 57

Molecular Imaging of Atherosclerosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 960

Farouc A. Jaffer, Peter Libby 58

Thrombosis and Embolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 980

Andrea J. Wiethoff, Elmar Spuentrup, René M. Botnar 59

Molecular Imaging of Stem Cells in Myocardial Infarction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 989

David E. Sosnovik, Joseph C. Wu CNS: 60

Central Nervous System Molecular Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1011

Dima A. Hammoud, Andreas H. Jacobs, Martin G. Pomper 61

Neuroreceptor Imaging: Applications, Advances, and Limitations . . . . . . . . . . . . . . . . . . . . . . . . . 1035

Rikki N. Waterhouse, Thomas Lee Collier 62

PET and SPECT Imaging of Neurodegenerative Diseases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1060

Brian J. Lopresti, Victor L. Villemagne, Chester A. Mathis AUTOIMMUNE/IMMUNOLOGY 63

Molecular Imaging of Autoimmune Diseses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1089

Alberto Signore, Marco Chianelli 64

Rheumatoid Arthritis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1108

Lars Stangenberg, Umar Mahmood 65

Autoimmune Diabetes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1130

Diane Mathis, Jason Gaglia 66

Imaging in Asthma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1147

Mikael J. Pittet, Filip K. Swirski PART V: MOLECULAR IMAGING IN DRUG EVALUATION 67

Molecular and Functional Imaging in Drug Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1161

Nicholas van Bruggen, Bernard M. Fine, Markus Rudin

x

68

Contents

PET Imaging in Cancer Clinical Trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1179

David A. Mankoff 69

Magnetic Resonance Imaging in Clinical Trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1192

Jeffrey L. Evelhoch, Douglas L. Arnold, Edward A. Ashton, Barry T. Peterson, Deborah Burstein, Derek L. G. Hill, Chun Yuan 70

Imaging of Gene Therapy: Basis and Clinical Trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1214

Andreas H. Jacobs, Yannic Waerzeggers, E. Antonio Chiocca, June-Key Chung, Juri Gelovani PART VI: OTHER 71

Visualization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1247

David S. Paik 72

Quantification of Radiotracer Uptake into Tissue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1258

Michael M. Graham 73

Mining Genomic Data for Molecular Imaging Targets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1271

Sylvia K. Plevritis 74

Pharmacokinetic Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1284

Sung-Cheng (Henry) Huang 75

Cost-Effectiveness Analysis/Economics of Probe Development . . . . . . . . . . . . . . . . . . . . . . . . . . 1290

Daniel C. Sullivan, Paula M. Jacobs 76

The Regulatory and Reimbursement Process for Imaging Agents and Devices . . . . . . . . . . . . . . 1299

John M. Hoffman Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1326

Preface Over the last decade, the f ield of molecular imaging of living subjects has e volved considerab ly and has seen spectacular advances in chemistr y, engineering and biomedical applications. In a relatively short period of time, comprehensive molecular imaging centers ha ve been established in the US, Europe, and Asia and are increasingly integrated into basic sciences and translational networks. Ne w in vestigators, collaborators, and students drawn into this multidisciplinar y f ield ha ve often expressed the desire and need for an authoritati ve te xtbook. This te xtbook w as designed precisel y to f ill this need. We have been fortunate to recruit over 160 leading authors contributing 76 chapters. Given the multidisciplinar y nature of the f ield, the book is broken into six different sections. Part 1 (Molecular Imaging Technologies) summarizes the dif ferent macro, meso and microscopic imaging technolo gies currently available. Part 2 (Chemistry of Molecular Imaging) is dedicated to re viewing chemical approaches to imaging probe designs for dif ferent types of imaging technologies. This section also contains chapters on the emerging f ield of nanomaterials, chemical biolo gy, and probe design as w ell as signal amplif ication strate gies. Part 3 (Molecular Imaging in Cell and Molecular Biology) contains chapters dedicated to protein engineering, vectors, and pathways. Part 4 (Applications of Molecular Imaging) summarizes the abo ve adv ances in dif ferent clinical disease entities. P art 5 (Molecular Imaging in Drug E valuation) i s d edicated t o i maging i n d rug

development, and P art 6 pro vides chapters on computation, bioinfor matics, and modeling. We hope that the organization of this large amount of information is logical, and we have worked hard to avoid redundancies among chapters. We ha ve also done our best to encourage the use of figures to illustrate concepts and to provide numerous molecular imaging examples. We have striven to make Molecular Imaging the most authoritative and ef fective resource a vailable for the student and newcomer at all levels. If we have succeeded, it is because of the hard work, knowledge, and devotion of our authors and their responses to our critiques. We are g rateful to our institutions and depar tments for the continuing support that has enabled this work. We are mindful of our families and students w ho tolerated our “limited bandwidths” necessary for the timely completion of this edition. We are also g rateful to BC Deck er, PMPH–USA, and the entire administrative staff at our centers for k eeping us in line. In particular we acknowledge the hard work by Tania Cunningham, Melissa Carlson, and Judy Schwimmer . We are optimistic that this book will contribute to the continuing education of a v ariety of professionals and will ultimately aid in the care of our patients to whom all of our efforts are dedicated. Ralph Weissleder, MD, PHD Brian D. Ross, P HD Alnawaz Rehemtulla, P HD Sanjiv S. Gambhir, MD, PHD

xi

Contributors SILVIO AIME, PHD [19] Molecular Imaging Center University of Torino Torino, Italy

AMBROS J. BEER, MD [45] Technische Universität München Klinikum rechts der Isar Munich, Germany

OLULANU H. AINA, D.V.M., PHD [31] UC Davis Cancer Center Division of Hematology/Oncology University of California, Davis Sacramento, CA

MAHAVEER S. BHOJANI, PHD [49] Center for Molecular Imaging University of Michigan Medical School Ann Arbor, MI

JAN HENRIK ARDENKJÆR-LARSEN, PHD [25] The Grove Center GE Healthcare Amersham, UK DOUGLAS L. ARNOLD, MD [69] Montreal Neurological Institute Montreal, Quebec, Canada EDWARD A. ASHTON, PHD [69] VirtualScopics, Inc. Rochester, NY

ZAVER M. BHUJWALLA, PHD [50] The JHU ICMIC Program The Sidney Kimmel Comprehensive Cancer Center Johns Hopkins University School of Medicine Baltimore, MD SIHAM BIADE, PHARMD, PHD [37] Office of Oncology Drug Products Office of New Drugs Division of Medical Imaging & Hematology Products Center for Drug Evaluation and Research U.S. Food and Drug Administration Silver Spring, MD

SUNDAY AWE, PHD, MBA [37] Office of Oncology Drug Products Office of New Drugs Division of Medical Imaging & Hematology Products Center for Drug Evaluation and Research U.S. Food and Drug Administration Silver Spring, MD

PEYTON H. BLAND, PHD [18] Center for Molecular Imaging University of Michigan Medical School Ann Arbor, MI

FREEK J. BEEKMAN, PHD [4] Section Radiation Detection and Medical Imaging Delft University of Technology Delft, The Netherlands

ANTON V. BOROVJAGIN, PHD [42] Institute of Oral Health Research University of Alabama at Birmingham School of Dentistry Birmingham, AL

MARK A. BORDEN, PHD [28] Columbia University New York, NY

xiii

xiv

Contributors

RENÉ M. BOTNAR, PHD [58] Division of Imaging Sciences, King’s College London London, UK KEVIN M. BRINDLE, DPHIL [25] Cambridge Research Institute Li Ka Shing Center University of Cambridge Cambridge, UK JEFF W. M. B ULTE, PHD [44] Division of MR Research Institute for Cell Engineering The Johns Hopkins University School of Medicine Baltimore, MD DEBORAH BURSTEIN, PHD [69] Beth Israel Deaconess Medical Center Boston, MA DAVID R. BUSCH, MS [14] University of Pennsylvania Philadelphia, PA WEIBO CAI, PHD [29] University of Wisconsin-Madison Madison, WI SHELTON D. CARUTHERS, PHD [35] Washington University in Saint Louis and Philips Healthcare St. Louis, MO CARMEL CHAN, PHD [47] Molecular Imaging Program at Stanford (MIPS) Stanford University School of Medicine Stanford, CA BRITTON CHANCE, PHD, SCD (CANTAB.), MD (HON) [14] Eldridge Reeves Johnson University Philadelphia, PA ARION F. CHATZIIOANNOU, PHD [9] Crump Institute for Molecular Imaging David Geffen School of Medicine at UCLA Los Angeles, CA

JOHN W. CHEN, MD, PHD [26] Center for Systems Biology Center for Molecular Imaging Research Massachusetts General Hospital Harvard Medical School Boston, MA JUNJIE CHEN, PHD [35] Cardiovascular Division Washington University School of Medicine St. Louis, MO XIAOYUAN (SHAWN) CHEN, PHD [29, 45] National Institute of Biomedical Imaging and Bioengineering (NIBIB) National Institutes of Health (NIH) Bethesda, MD THOMAS L. CHENEVERT, PHD [54] Center for Molecular Imaging University of Michigan Medical School Ann Arbor, MI EMMANUEL CHERIN, PHD [15] Sunnybrook Health Sciences Centre Toronto, Ontario, Canada KEVIN CHEUNG, MD [15] McMaster University Medical Centre Hamilton, Ontario, Canada. MARCO CHIANELLI, MD, PHD [63] University Medical Center Groningen University of Groningen, The Netherlands Regina Apostolorum Hospital Albano, Rome, Italy E. ANTONIO CHIOCCA, MD, PHD [70] Dardinger Center for Neuro-oncology and Neurosciences James Cancer Hospital/Solove Research Institute Ohio State University Medical Center Columbus, OH JUNE-KEY CHUNG, MD, PHD [70] Seoul National University Hospital Seoul National University College of Medicine Seoul, Korea

Contributors

NEAL H. CLINTHORNE, PHD [6] Division of Nuclear Medicine University of Michigan Ann Arbor, MI THOMAS LEE COLLIER, PHD [61] Advion Biosystems Inc Louisville, TN CHRISTOPHER H. CONTAG, PHD [8] Molecular Imaging Program at Stanford (MIPS) Stanford University School of Medicine Stanford, CA

HUA FAN-MINOGUE, PHD [47] Molecular Imaging Program at Stanford (MIPS) Stanford University School of Medicine Stanford, CA KATHERINE W. FERRARA, PHD [28] University of California, Davis Davis, CA BERNARD M. FINE, MD, PHD [67] BioOncology Genentech, Inc. South San Francisco, CA

DAVID T. CURIEL, MD, PHD [42] Division of Human Gene Therapy Gene Therapy Center University of Alabama at Birmingham Birmingham, AL

GREGORY FOLTZ, MD [39] Swedish Neuroscience Institute Swedish Medical Center Seattle, WA

CATHY S. CUTLER, PHD [24] Research Reactor Center University of Missouri-Columbia Columbia, MO

F. STUART FOSTER, PHD [15] Sunnybrook Health Sciences Centre University of Toronto Toronto, Ontario, Canada

ABHIJIT DE, PHD [47] Advanced Centre for Treatment, Research and Education in Cancer (ACTREC) Tata Memorial Centre Navi Mumbai, Maharashtra India

JASON GAGLIA, MD [65] Section on Immunology and Immunogenetics Joslin Diabetes Center Harvard Medical School Boston, MA

NEAL K. DEVARAJ, PHD [30] Center for Systems Biology Center for Molecular Imaging Research Massachusetts General Hospital Harvard Medical School Boston, MA MARK W. DEWHIRST, DVM, PHD [46] Duke University Medical Center Durham, NC JEFFREY L. EVELHOCH, PHD [69] Imaging Research Merck Research Laboratories West Point, PA

CRAIG J. GALBÁN, PHD [54] Center for Molecular Imaging University of Michigan Medical School Ann Arbor, MI SANJIV S. GAMBHIR, MD, PHD [1, 47] Molecular Imaging Program at Stanford (MIPS) Canary Center for Cancer Early Detection at Stanford Division of Nuclear Medicine Stanford University Stanford, CA JURI G. GELOVANI, MD, PHD [70] M.D. Anderson Cancer Center University of Texas Houston, TX

xv

xvi

Contributors

ROBERT J. GILLIES, PHD [50] Advanced Research Institute for Biomedical Imaging, Arizona Cancer Center Tucson, AZ GIOVANNI BATTISTA GIOVENZANA, PHD [19] University of Piemonte Orientale “Amedeo Avogadro” Novara, Italy SHAUN S. GLEASON, PHD [5] Image Science and Machine Vision Group Oak Ridge National Laboratory Oak Ridge, TN KRISTINE GLUNDE, PHD [50] JHU ICMIC Program The Sidney Kimmel Comprehensive Cancer Center Molecular Imaging Program The Johns Hopkins University School of Medicine Baltimore, MD KLAES GOLMAN, PHD [25] Imagnia AB Malmö, Sweden

HARVEY R. HERSCHMAN, PHD [38] Molecular Biology Institute Jonsson Comprehensive Cancer Center David Geffen School of Medicine at UCLA University of California, Los Angeles Los Angeles, CA RODNEY J. HICKS, MD, FRACP [52] The Peter MacCallum Cancer Centre Centre for Molecular Imaging and Translational Oncology The University of Melbourne East Melbourne, Victoria, Australia SCOTT A. HILDERBRAND, PHD [27] Center for Systems Biology Center for Molecular Imaging Research Massachusetts General Hospital Harvard Medical School Boston, MA DEREK L. G. H ILL, PHD [69] IXICO, Ltd. London, UK

MICHAEL M. GRAHAM, MD., PHD [72] Division of Nuclear Medicine University of Iowa Iowa City, IA

JOHN M. HOFFMAN, MD [76] Division of Nuclear Medicine and Molecular Imaging Huntsman Cancer Institute University of Utah, School of Medicine Salt Lake City, UT

DIMA A. HAMMOUD, MD [60] Radiology and Imaging Sciences Division of Neuroradiology National Institutes of Health Clinical Center Bethesda, MD

LEROY HOOD, PHD [39] Institute for Systems Biology Seattle, WA

SALLY J. HARGUS, PHD [37] Office of Oncology Drug Products Office of New Drugs Division of Medical Imaging & Hematology Products Center for Drug Evaluation and Research U.S. Food and Drug Administration Silver Spring, MD

SUNG-CHENG (HENRY) HUANG, DSC [74] Crump Institute for Molecular Imaging David Geffen School of Medicine at UCLA University of California, Los Angeles Los Angeles, CA MICHAEL S. HUGHES, PHD [35] Cardiovascular Division Washington University Medical School St. Louis, MO

Contributors

BRIAN F. HUTTON, MD, PHD [4] UCL and UCLH NHS Foundation Trust Institute of Nuclear Medicine London, U.K. ANDREAS H. JACOBS, MD [60, 70] Laboratory for Gene Therapy and Molecular Imaging MPI for Neurological Research Klinikum Fulda gAG Köln, Germany PAULA M. JACOBS, PHD [75] SAIC-Frederick Division of Cancer Treatment and Diagnosis Cancer Imaging Program National Cancer Institute Bethesda, MD FAROUC A. JAFFER, MD, PHD [57] Center for Molecular Imaging Research Cardiovascular Research Center and Cardiology Division Massachusetts General Hospital Harvard Medical School Boston, MA LEE JOSEPHSON, PHD [34] Division of Nuclear Medicine Massachusetts General Hospital Harvard Medical School Boston, MA SILVIA S. JURISSON, PHD [24] Department of Biomedical Engineering University of Missouri-Columbia Columbia, MO KIMBERLY A. KELLY, PHD [41] University of Virginia Charlottesville, VA BOKLYE KIM, PHD [18] Center for Molecular Imaging University of Michigan Medical School Ann Arbor, MI

TUSHAR KOKATE, PHD [37] Office of Oncology Drug Products Office of New Drugs Division of Medical Imaging & Hematology Products Center for Drug Evaluation and Research U.S. Food and Drug Administration Silver Spring, MD HANK F. KUNG, PHD [21] University of Pennsylvania, Philadelphia, PA JOHN KURHANEWICZ, PHD [53] University of California San Francisco San Francisco, CA KIT S. LAM, MD, PHD [31] UC Davis Cancer Center Division of Hematology/Oncology University of California Davis Sacramento, CA ADEBAYO LANIYONU, PHD [37] Office of Oncology Drug Products Office of New Drugs Division of Medical Imaging & Hematology Products Center for Drug Evaluation and Research U.S. Food and Drug Administration Silver Spring, MD GREGORY M. LANZA, MD, PHD [35] Cardiovascular Division Washington University Medical School St. Louis, MO CRAIG S. LEVIN, PHD [7] Molecular Imaging Program at Stanford (MIPS) Division of Nuclear Medicine Stanford University School of Medicine Stanford, CA MICHAEL R. LEWIS, M.S., P HD [24] University of Missouri-Columbia Columbia, MO

xvii

xviii

Contributors

PETER LIBBY, MD [57] Division of Cardiovascular Medicine Brigham and Women’s Hospital Harvard Medical School Boston, MA

MARCUS R. MAKOWSKI, MD [56] Imaging Sciences Division King’s College London St Thomas' Hospital London, UK

CHARLES P. L IN, PHD [12] Center for Systems Biology Wellman Center for Photomedicine Massachusetts General Hospital Harvard Medical School Boston, MA

DAVID A. MANKOFF, MD, PHD [68] University of Washington Seattle Cancer Care Alliance Seattle, WA

RUIWU LIU, PHD [31] UC Davis Cancer Center Division of Hematology/Oncology University of California, Davis Sacramento, CA CHRISTOPHER M. LONG, BS [44] The Johns Hopkins University School of Medicine Baltimore, MD MARÍA VERÓNICA LÓPEZ, PHD [42] Laboratory of Molecular and Cellular Therapy Leloir Institute Capital Federal, Buenos Aires, Argentina BRIAN J. L OPRESTI, BSC [62] PET Center University of Pittsburgh School of Medicine Pittsburgh, PA BING MA, PHD [18] Center for Molecular Imaging University of Michigan Medical School Ann Arbor, MI HIDEVALDO B. MACHADO, PHD [38] David Geffen School of Medicine University of California, Los Angeles Los Angeles, CA UMAR MAHMOOD, MD, PHD [10, 64] Division of Nuclear Medicine Massachusetts General Hospital Harvard Medical School Boston, MA

ASHLEY A. MANZOOR, BS [46] Duke University Medical Center Durham, NC RALPH P. MASON, PHD [35] Division of Advanced Radiological Sciences UT Southwestern Medical Center Dallas, TX TARIK F. MASSOUD, MD, PHD [47] Cambridge Cancer Center University of Cambridge Cambridge, UK CHESTER A. MATHIS, PHD [62] PET Facility University of Pittsburgh School of Medicine Pittsburgh, PA DIANE MATHIS, PHD [65] Department of Pathology Harvard Medical School Boston, MA QIANA L. MATTHEWS, PHD [42] Division of Human Gene Therapy Gene Therapy Center University of Alabama at Birmingham Birmingham, AL JASON R. MCCARTHY, PHD [33] Center for Systems Biology Center for Molecular Imaging Research Massachusetts General Hospital Harvard Medical School Boston, MA

Contributors

CLAUDE F. MEARES, PHD [23] University of California, Davis Davis, CA THORSTEN R. MEMPEL, MD, PHD [13] Center for Immunology and Inflammatory Diseases Center for Systems Biology Massachusetts General Hospital Harvard Medical School Boston, MA LING-JIAN MENG, PHD [6] Division of Nuclear Medicine University of Illinois Urbana Champaign, IL CHARLES R. MEYER, PHD [18, 54] Center for Molecular Imaging University of Michigan Medical School Ann Arbor, MI NASER MUJA, PHD [44] Division of MR Research Institute for Cell Engineering The Johns Hopkins University School of Medicine Baltimore, MD MICHAL NEEMAN, PHD [50] Weizmann Institute of Science Rehovot, Israel STEPHAN G. NEKOLLA, PHD [56] Klinikum rechts der Isar Technischen Universität München München, Bavaria, Germany SARAH J. NELSON, PHD [53] University of California, Berkeley Berkeley, CA GANG NIU, PHD [45] Molecular Imaging Program at Stanford (MIPS) Stanford University School of Medicine Stanford, CA

VASILIS NTZIACHRISTOS, MSC, PHD [11] Institute for Biological and Medical Imaging Technical University of Munich and Helmholtz Center Munich, Germany EKAMA ONOFIOK, BS [31] UC Davis Cancer Center Division of Hematology/Oncology University of California, Davis Sacramento, CA DUSTIN OSBORNE, PHD [5] Siemens Molecular Imaging Knoxville, TN YANLI OUYANG, MD, PHD, DABT [37] Office of Oncology Drug Products Office of New Drugs Division of Medical Imaging & Hematology Products Center for Drug Evaluation and Research U.S. Food and Drug Administration Silver Spring, MD DAVID S. PAIK, PHD [71] Richard M. Lucas Center Stanford University School of Medicine Stanford, CA GREGORY M. PALMER, PHD [46] Duke University Medical Center Durham, NC HYUNJIN PARK, PHD [18] Center for Molecular Imaging University of Michigan Medical School Ann Arbor, MI RAMASAMY PAULMURUGAN, PHD [47] Molecular Imaging Program at Stanford (MIPS) Stanford University School of Medicine Stanford, CA MICHAEL J. PAULUS, PHD [5] Siemens Molecular Imaging Knoxville, TN

xix

xx

Contributors

BARRY T. PETERSON, PHD [69] Division of Physiological Measurements Pfizer Global Research and Development New London, CT MIKAEL J. PITTET, PHD [66] Center for Systems Biology Center for Molecular Imaging Research Massachusetts General Hospital Harvard Medical School Boston, MA SYLVIA K. PLEVRITIS, PHD [73] Radiological Sciences Laboratory Stanford University School of Medicine Stanford, CA MARTIN G. POMPER, MD, PHD [60] Johns Hopkins Medical Institutions Baltimore, MD SHENGPING QIN, PHD [28] University of California, Davis Davis, CA JIANGHONG RAO, PHD [22] Molecular Imaging Program at Stanford (MIPS) Stanford University School of Medicine Stanford, CA PRITHA RAY, PHD [47] Advanced Centre for Treatment, Research and Education in Cancer (ACTREC) Tata Memorial Hospital Kharghar, Navi Mumbai, Maharastra, India ALNAWAZ REHEMTULLA, PHD [49, 54] Center for Molecular Imaging University of Michigan Medical School Ann Arbor, MI BRIAN D. ROSS, PHD [49, 54] Center for Molecular Imaging University of Michigan Medical School Ann Arbor, MI

MARKUS RUDIN, PHD [67] Institute for Biomedical Engineering, Electical Engineering and Information Technology University of Zürich and ETH Zürich Zürich, ZH, Switzerland ANTTI SARASTE, MD, PHD [56] Turku University Hospital, Main Hospital Turku, Finland HEINRICH R. SCHELBERT, MD, PHD [55] David Geffen School of Medicine at UCLA University of California, Los Angeles Los Angeles, CA MARKUS SCHWAIGER MD, PHD [3, 45, 56] Klinikum rechts der Isar Technischen Universität München München, Bavaria, Germany MARCUS D. SEEMANN, MD [3] Institute of Radiology and Nuclear Medicine University of Bochum, St. Josef-Hospital Bochum, Germany KHALID SHAH, PHD [43] Center for Molecular Imaging Research Massachusetts General Hospital Harvard Medical School Boston, MA JAMES SHARPE, PHD [17] Catalan Institute for Advanced Research and Education (ICREA) EMBL-CRG Systems Biology Unit Centre for Genomic Regulation, UPF Barcelona, Spain STANLEY SHAW, MD, PHD [32] Center for Systems Biology Massachusetts General Hospital Harvard Medical School Boston, MA

Contributors

ALBERTO SIGNORE, MD, PHD [63] II Faculty of Medicine University of Rome “Sapienza” Roma, Italy

DAVID W. TOWNSEND, PHD [2] PET and SPECT Development Singapore Bioimaging Consortium Singapore

DAVID E. SOSNOVIK, MD [59] Center for Molecular Imaging Research Division of Cardiology Massachusetts General Hospital Harvard Medical School Charlestown, MA

CHRISTINA TSIEN, MD [51] University of Michigan Medical Center Ann Arbor, MI

ELMAR SPUENTRUP, MD [58] University Hospital of Cologne Cologne, Germany LARS STANGENBERG, MD [64] Department of Surgery Massachusetts General Hospital Harvard Medical School Boston, MA DANIEL C. SULLIVAN, MD [75] Duke Comprehensive Cancer Center Duke University Medical Center Durham, NC FILIP K. SWIRSKI, PHD [66] Center for Systems Biology Center for Molecular Imaging Research Massachusetts General Hospital Harvard Medical School Boston, MA BERTRAND TAVITIAN, MD, PHD [36] CEA, Institut d’Imagerie Biomédicale Service Hospitalier Frédéric Joliot Laboratoire d’Imagerie Moléculaire Expérimentale and INSERM U803 Laboratoire d’Imagerie de l’Expression des Gènes Orsay, France ENZO TERRENO, PHD [19] Molecular Imaging Center University of Torino Torino, Italy

ROGER Y. TSIEN, PHD [48] Howard Hughes Medical Institute University of California, San Diego La Jolla, CA ANDREW TSOURKAS, PHD [34] School of Engineering and Applied Science University of Pennsylvania Philadelphia, PA RABI UPADHYAY, BS [10] Center for Systems Biology Center for Molecular Imaging Research Massachusetts General Hospital Harvard Medical School Boston, MA HENRY F. VANBROCKLIN, PHD [20] Radiopharmaceutical Research University of California San Francisco San Francisco, CA NICHOLAS VAN BRUGGEN, PHD [67] Biomedical Imaging Genentech, Inc. South San Francisco, CA ELISENDA RODRIGUEZ VARGAS, PHD [26] Center for Systems Biology Center for Molecular Imaging Research Massachusetts General Hospital Harvard Medical School Boston, MA BENJAMIN L. VIGLIANTI, PHD [46] Duke University Medical Center Durham, NC

xxi

xxii

Contributors

DANIEL B. VIGNERON, PHD [53] University of California San Francisco San Francisco, CA VICTOR L. VILLEMAGNE, MD [62] Austin Health The Mental Health Research Institute of Victoria University of Melbourne Melbourne, Victoria, Australia YANNIC WAERZEGGERS, MD [70] Laboratory for Gene Therapy and Molecular Imaging Max Planck Institute for Neurological Research Klaus-Joachim-Zülch Laboratories of the Max Planck Society Cologne, Germany RICHARD L. WAHL, MD, FACR [52] Nuclear Medicine/PET New Technology and Business Development Johns Hopkins University School of Medicine Baltimore, MD LIHONG V. WANG, PHD [16] Optical Imaging Laboratory Washington University St. Louis, MO RIKKI N. WATERHOUSE, PHD [61] Head, Cancer Imaging and Radiochemistr y Advanced Technology Global Pharmaceutical Discovery Abbott Abbott Park, IL RALPH WEISSLEDER, MD, PHD [30] Center for Systems Biology Center for Molecular Imaging Research Massachusetts General Hospital Harvard Medical School Boston, MA SAMUEL A. WICKLINE, MD [35] Cardiovascular Division Washington University School of Medicine St. Louis, MO

ANDREA J. WIETHOFF, PHD [58] Division of Imaging Sciences King’s College London Philips Healthcare London, UK PATRICK M. WINTER, PHD [35] Cardiovascular Division Washington University School of Medicine St. Louis, MO ANNA M. WU, PHD [40] Crump Institute for Molecular Imaging David Geffen School of Medicine at UCLA University of California, Los Angeles Los Angeles, CA JOSEPH C. WU, MD, PHD [59] Molecular Imaging Program at Stanford (MIPS) Stanford University School of Medicine Stanford, CA YUN XING, PHD [22] School of Engineering University of Dayton Dayton, OH CHUN YUAN, PHD [69] Vascular Imaging Laboratory University of Washington Seattle, WA HONG YUAN, PHD [46] Duke University Medical Center Durham, NC SEOK H. (ANDY) YUN, PHD [12] Wellman Center for Photomedicine Massachusetts General Hospital Harvard Medical Center Boston, MA

1 GENERAL PRINCIPLES

OF

MOLECULAR IMAGING

SANJIV S. GAMBHIR, MD, PHD

Molecular imaging (MI) of living subjects is an emerging field that aims to study molecular and cellular e vents in the intact living animal and human. These events can be as simple as location(s) of a specif ic population of cells or le vels of a gi ven protein receptor on the surf ace of cells. In addition, it is possib le to study more comple x events such as the interaction of tw o intracellular proteins, cellular metabolic flux, or transcription of a set of genes w hen one cell type comes into contact with another cell type. In contrast to molecular processes studied in intact cells outside the living subject (eg, with light microscopy techniques), it is much more dif ficult to longitudinally study the same processes in intact li ving subjects where most cells are located within deep tissues. It is the hope of many MI researchers that most of biology will eventually be ab le to be studied in the intact li ving subject instead of ha ving to remove tissues/cells for further analysis, as is no w commonly done. This will allow the study of simple and comple x processes w hile cells reside in their nati ve en vironment with all molecular feedback loops fully intact. The reasons for monitoring/imaging v arious molecular targets are usuall y related to characterizing a disease process that ma y cor relate with concentrations of one or more of these molecular tar gets. For e xample, the presence of relatively high levels of the somatostatin receptor type 2 (located on cell membranes) in the lungs of a subject may be indicative of the presence of cancer cells in the lung. This may then help guide medical management of the subject in w hich such a molecular signal is detected. Another v ery impor tant reason to study v arious cellular/molecular targets is to help dissect complex underlying biology. F or e xample, one might be ab le to study the migration of a specif ic subset of T-cells (or T l ymphocytes) into a tumor and subsequent acti vation of these T-cells by a T-cell receptor to better understand the details of the interaction of the tumor with the immune system.

An additional impor tant application of MI is in the process of dr ug disco very and v alidation, as w ell as in predicting and monitoring response to v arious types of therapy (see Chapters 51, “Novel MR and PET Imaging in the RT Planning and Assessment of Response of Malignant Gliomas,” 52, “PET Diagnosis and Response Monitoring in Oncolo gy,” 53, “MRS Treatment Response and Detection,” 54, “Dif fusion MRI: A Biomark er for Earl y Cancer Treatment Response, ” and 67, “Molecular and Functional Imaging in Dr ug Development”). Most of MI is perfor med b y introducing a molecular probe (e g, a small molecule) into the living subject, as will be detailed later. Since MI probes are often relati ves of pharmaceuticals and/or interact with the same molecular tar get(s) there are many important links between MI and phar macology (see Chapter 74, “Phar macokinetic Modeling”). The safety of subjects w hen using MI probes is of paramount impor tance. Ideally, one w ould not ha ve to introduce MI probes into the subject to quantitate the molecular targets of interest. However, since in most cases one must f irst introduce the MI probe, it is critical to ensure that this probe does not signif icantly per turb the living subject and certainly that it is not acutel y or chronically to xic to the subject (see Chapter 37, “Nonclinical Product Developmental Strategies, Safety Considerations and Toxicity Profiles of Medical Imaging and Radiopharmaceuticals Products”). MI often ser ves as an impor tant complement to more con ventional anatomical imaging (eg, computed tomo graphy [CT]). Together these techniques can help to impro ve disease management and understanding of biological processes of interest. The process of de veloping new strategies/assays for MI is illustrated in Figure 1 and will be referred to as the MI research chain. This is an iterati ve process, and if successful, leads to a useful MI assa y. Although what is shown is the translation of the assay for eventual clinical application, it is not al ways the case that e very assay is 1

2

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Molecular Targets

Gene expression profiling Proteomics Systems biology

Chemistry

Organic synthesis of probes Combinatorial chemistry Radiochemistry Material science/Nanochemistry

Small/Large animal Models

Imaging

Molecular/Cell Biology

Cell delivery issues Cell efflux issues Cell-Cell interactions Spatial localization of target

Clinical Imaging Biodistribution Animal models of disease Probe Stability/Pharmacokinetics

Computer Modeling

Human pharmacokinetics/ Safety Clinical trials Outcomes assessment Drug efficacy

Multimodality animal imaging Pharmacokinetics Time-activity curves Individual variation studies Image Reconstruction/Quantitation Mathematical modeling Statistical analysis 3-D visualization

Figure 1. Molecular imaging research chain. The process of going from molecular targets to clinical molecular imaging is shown. This research chain is fundamental to the field and is critical for its success. Not all research in the field is intended for clinical translation and in some cases assay development terminates at the computer modeling stage. The process iterates based on lessons learned in a given application, as shown with the dotted line.

intended for clinical applications. Some assa ys may be only intended for animal models to ans wer fundamental biological questions. Man y MI assa ys are already in existence and y et more are being de veloped. Many MI probes and their uses are limited , and y et others ha ve found signif icant clinical use. Although the timeline for developing an assay and translating it for clinical use can be 3 to 7 years, this timeline is decreasing and hopefully can approach 1 to 3 y ears in the near future. To achieve a few well-used assays, the f ield has to be prepared to fail in its attempts at development and evolve from these failures. Each of the components of the MI research chain is reviewed next. The choice of molecular tar get(s) dri ves the entire research chain. Ideal molecular tar gets are present in multiple copies per cell. These are usually protein targets (100 to 1 million copies per cell) but can also be

messenger ribonucleic acid (mRN A, 50 to 1,000 copies per cell). Deo xyribonucleic acid (DN A) is not used because of both its low copy number (which makes it difficult to produce sufficient specific signal), and because imaging of DNA would not allow one to deter mine if a particular gene is being expressed, only that it is present. Clearly, the abundance and specif icity of the tar get for the disease process under study is critical to make the MI assay successful. One fundamental issue for MI is knowing w hich molecular tar gets are rele vant to study for a given set of biolo gical questions or for a gi ven disease management prob lem. In f act, e ven if MI w as ab le to interrogate concentrations of e very potential molecular target and various other events, it would be still difficult to kno w w hich e vents to actuall y monitor/image. It is currently not possible to highly-multiplex (detect multiple molecular targets of interest simultaneously) and as a

General Principles of Molecular Ima ging

HO O HO

OH HO [18]F

A

B

Figure 2. Anatomy of a molecular imaging probe. Most molecular imaging probes are made up of three primary components as shown in the left panel. These include (1) a chemical component that provides specificity for the molecular target of interest (shown in blue), (2) a component that will provide a signal which can be detected (shown in red), and (3) a linker (shown in gray scale), which may or may not be needed. Note that the three components are not shown to scale. Sometimes, the chemical specificity component is much larger than the signaling component (eg, a molecular imaging probe for PET) and at other times, it is much smaller than the signaling component (eg, a targeted microbubble molecular imaging probe in which the signaling component is a relatively large gas-filled microbubble for use with ultrasound). A specific molecular imaging probe is shown in the right panel. This molecular imaging probe is one of the most clinically successful MI probes to date ([18F]-2-fluoro-2-deoxyglucose [FDG]) and is discussed in detail in Chapter 20, “Radiochemistry of PET.” This MI probe has chemical specificity for the glucose transporters which transport it into cells as well as for the hexokinase type II enzyme which phosphorylates the molecule in the sixth position leaving it negatively charged and unable to diffuse out of cells. Furthermore, the phosphorylated FDG (FDG-6PO4) is unable to be further metabolized because it is not recognized by enzymes that normally metabolize glucose-6PO4. The signaling component is the fluorine-18, which is proton rich and decays by production of a positron. The positron annihilates with a nearby electron to produce two 511 keV gamma rays, which have a good probability of making it through tissues to be detected outside the subject.

result, selection of molecular tar gets is critical. Multiplexing is cur rently limited to 3 to 5 maximum molecular targets. If one relaxes the necessity to simultaneously measure the tar gets of interest, then one can perfor m serial measurements (e g, dail y) to deter mine le vels of additional targets. In most strategies, a MI probe must first be introduced into the li ving subject (e g, b y injection into the b loodstream). The “anatomy” of such a MI probe is sho wn in Figure 2 and is usually composed of a chemically specific component that interacts with the intended molecular target (eg, a protein), a signaling component that produces a signal that can hopefull y be detected , and a link er between the two components. MI probes are referred to by many names such as MI agents, imaging agents, tracers, radiopharmaceuticals, radiotracers, acti vatable or smar t probes, constitutively active probes, molecular detecti ves, and molecular spies. Even though a MI probe does provide “contrast” in an image, allo wing one to “see” molecular targets of interest relati ve to a backg round, the ter m

3

“contrast agent” is a poor ter m because most contrast agents refer to nonspecif ic agents that ma y ha ve poorl y defined molecular targets. MI probes can be broadl y categorized as constituti vely acti ve probes or acti vatable probes as illustrated in F igure 3. Radiolabeled probes produce their signal constantl y throughout as radioacti ve decay occurs and are therefore constituti vely acti ve. An activatable probe has the advantage of not producing signal until it interacts with its intended tar get(s), thus leading to a lo w backg round signal. MI probes are adv antageous because the y pro vide molecular specif icity, but the y are also the Achilles’ heel for the MI f ield because of re gulatory issues sur rounding the introduction of a no vel or foreign probe for human applications. It is unfor tunate that we currently do not have methods of “listening-in” on molecular events without first introducing MI probes. This would allow imaging of molecular/cellular processes without de veloping specif ic MI probes and w ould mark edly ease the re gulatory issues. There are a fe w exceptions to this limitation, such as magnetic resonance spectroscop y (MRS), w hich can “listen-in” on a fe w specif ic endo genous molecules (e g, choline) as discussed in Chapter 53, “MRS Treatment Response and Detection. ” Pro gress in this direction without compromising important parameters such as spatial and temporal resolution w ould be gamechanging for the f ield. Fur thermore, as the scientif ic understanding of nor mal and patholo gical processes continues to e volve, leading to a need to image dif ferent molecular tar gets, it is impor tant to ha ve strate gies that can allow development of MI probes for “ne w” molecular targets. Generalizab le strate gies such as engineered antibodies (see Chapter 40, “Protein Engineering for Molecular Imaging”) illustrate ho w MI probes can be de veloped relatively quickly as ne w molecular tar gets of interest are discovered. Chemistry helps drive the MI field as it is pivotal to developing the MI probes. Small molecules, peptides, aptamers, engineered proteins, and e ven more comple x nanoparticles are all possib le MI probes. The MI probe developmental chemistr y can often tak e many months. Ideally, the optimized chemistr y w ould allo w rapid synthesis of the MI probe with high purity so that it can be synthesized at all laborator y, clinical, and research sites that will each perform the imaging. In the cases of human translation, it is impor tant to also perfor m synthesis of the MI probe under good manuf acturing practice (GMP) guidelines. One may wonder while reading the chapters in this te xtbook about the potential to treat the disease of interest using the same probe that is being used to image the disease. If the MI probe can be constructed with a therapeutic component, in addition to the

4

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Constitutively active probes

versus

Time

Activatable probes

Time

Figure 3. Two broad categories of molecular imaging probes. MI probes (red circle linked to a blue partial circle) are shown interacting with their intended blue circle targets, which in this case are located intracellularly and on the cell membrane. Other potential targets (for other MI probes specific for them) are shown as triangles and squares. Constitutively active probes (for PET/SPECT imaging and autoradiography) produce continuous signal, before and after interacting with their target(s), through the decay of the radioisotope. A time delay between injection of the probe and imaging helps to clear the nontrapped probes reducing background signal. Shown in the bottom left panel are some MI probes that produce signal even though they have not interacted with their blue circle target. Activatable probes produce signal only when they interact with their target(s) (eg, near-infrared fluorescent probes for optical imaging). These activatable probes can be thought of as either “light switches” (as in optical probes or hyperpolarized C-13 MR probes that turn from “off to on”) or “dimmer switches” (as in gadolinium MR probes that turn from “dull to bright”). A time delay between injection and imaging helps to achieve sufficient levels of activated probe at the target site where targets are typically enzymes. Background signal is inherently low in this category of MI probe because signal is only produced when the MI probe interacts with its intended targets. Shown in the bottom right panel are some MI probes that have not interacted with their blue circle targets and therefore do not produce signals.

signaling component, possib le link er, and the component specific for the molecular target(s) of interest, then the resulting “theranostic” could be used. In f act, in the field of Nuclear Medicine, the same probe that is used for imaging is often slightl y modif ied (by changing the radioisotope) and then ser ves as a therapeutic (e g, for tumor kill). Also, as discussed later on, the chemical development of multimodality probes enab les the performance of imaging on multiple imaging platfor ms that can span from small-to-lar ge animals and e ven humans. The MI probes can be tested in vitro (with extracts of cells) and eventually in intact cells in cell culture. Testing the probes with intact cells helps to better understand their ability to tra verse the cell membrane, the time

involved for tar geting and clearance from cells, and the potential for nonspecif ic interactions that would lead to increased background signal. Furthermore, use of standard molecular biology techniques allows the modulation of le vels of molecular tar gets (b y transducing the gene encoding for the protein tar get) to test the relationship between MI probe signal and le vels of molecular tar get. Additionally, repor ter genes can be tested and v alidated in cell culture and for such strate gies cell culture testing is critical (see also Chapters 38, “Overview of Molecular and Cell Biolo gy,” and 47, “Molecular Imaging of Protein-Protein Interactions”). Although testing in cell culture is useful, it does not help ans wer se veral critical issues for de veloping an MI assa y. These include i) ho w to deliver sufficient amounts of MI probe to the cells of

General Principles of Molecular Ima ging

interest when the cells reside deep within a li ving organism, ii) if the MI probe does not interact with the tar gets of interest, can it be cleared to reduce background signal, and iii) biodistribution and phar macokinetic issues related to deli very and clearance of the MI probe. F or these and other reasons, the next step is usually to test the MI probe in animal models. Small animal models are con venient to test these and other issues due to their relati vely low cost, highthroughput, ease of handling, etc. Mouse models can be set up with the molecular targets of interest. This can be done by implanting cells with the target(s) of interest or studying mice that spontaneously or otherwise develop the disease e xhibiting the cell/molecular tar get(s) of interest. Alternatively, murine models can be developed by introducing or deleting gene(s) of interest as is done in transgenic/knock-in mice and knock-out mice, respectively. Additionally, the le vels of the molecular target of interest can be manipulated using cur rent or novel pharmaceuticals. Sometimes, small animal models cannot properly reflect human disease so that lar ge animal (eg, porcine) models are used. For example, for many cardiovascular diseases, the porcine model ma y be more appropriate than a rodent model. F or studying neurological diseases, the primate brain ma y be the most appropriate. Also, high-resolution imaging of large animals (e g, rabbits) is also adv antageous for identifying hetero geneity of molecular e xpression within diseased tissue (e g, atherosclerotic plaques) which ma y be used for better e valuation of the MI probe’s ability to characterize disease in humans. Ev en if no animal model is fully reflective of the human disease, the use of animal models is also needed to test for potential to xicity of the MI probe in a li ving subject prior to translation to human studies. Additionally, in the case of MI probes that use a radioisotope for signal production, there is a need to deter mine radiation dosimetry to v arious or gans for e xtrapolation of the dosimetry to humans. The next key step is to test the animal model with the MI probe w hile imaging with the appropriate imaging instrument(s) of choice. Usuall y, the animal has to be anesthetized in order to be imaged unless the anesthesia may interfere with the process being studied or the temporal resolution (how quickly the imaging instr ument can “tak e a picture”) of the imaging instr ument(s) is extremely f ast. This allo ws one to tr uly e xamine the pharmacokinetics and biodistribution of the MI probe. It also allo ws one to optimize routes of administration (usually intravenous), mass of injected probe to achie ve desired signal, signal-to-backg round ratio, time to image

5

the animal, as w ell as se veral other impor tant f actors. Sometimes, the animal has to be imaged both for location(s) and concentrations of the MI probe as w ell as for anatomical information. By combining both the anatomical and molecular images, one can better understand underlying biological/pathological processes of interest. It would be ideal to ha ve a technology that could deter mine very lo w le vels of molecular tar get concentration (eg, picomolar or 10 –12 M), have the ability to follo w just a few cells instead of thousands to millions, have high spatial resolution (sub-millimeter), high temporal resolution (eg, milli-second), be lo w-cost, offer high-throughput, be fully quantitati ve, allo w inter rogation at all depths throughout the subject, and allow measurement of molecular tar gets located an ywhere in the subject and in an y location within the cell. Because no such single ideal imaging technology exists, it is often the case that based on the biolo gical questions being ask ed, the appropriate MI technology must be selected. It is thus impor tant that advocates of each MI technolo gy properl y acknowledge advantages and limitations of a gi ven strate gy and also acknowledge that sometimes a combination of technologies or an alter nate technolo gy is best suited for the biological/clinical question at hand. The next step is to look at quantitati ve issues in the animal model using the images acquired. Quantitation can be of v arious degrees ranging from minimal to absolute. One can attempt to quantitate the amount of MI probe at various sites including the tar get tissues and/or relate the signal from the MI probe back to absolute le vels of the molecular tar get(s) of interest. To perfor m absolute quantitation, it is usuall y necessar y to perfor m dynamic imaging in w hich serial images are tak en to characterize changes in the location(s) of the MI probe signal to produce time-activity curves (see also Chapter 4, “SPECT and SPECT/CT”). A key issue in almost e very molecular image is the requirement that le vels of the signal be ab le to be related back to levels, or even activity in the case of activatable probes, of the molecular target(s) of interest. In addition to quantitation, visualization strate gies can also be de veloped and v alidated so that one can displa y the molecular images in a way that makes it easier to interpret the findings. In man y but not all cases, the f inal steps are to translate the de veloped MI strate gy to clinical applications. This requires obtaining approval from appropriate agencies such as the F ood and Dr ug Administration (FDA) and the local inter nal review board (IRB). Radiation dosimetr y studies (if a radioisotope is used) and toxicology studies in preclinical models are therefore critical. The hope in the initial pilot clinical imaging

6

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

BB

C

D

Figure 4. From molecule to molecular image. The imaging of the location(s) and concentration of MI target(s) of interest starts with subject preparation and introduction of the MI probe into the subject (Panel A). An exception is in MRI where it is typically necessary to perform a pre-scan prior to injecting the agent to account for the “background” tissue signal which can vary greatly within a particular diseased tissue. However, ways of overcoming this limitation are being developed (eg, magnetic particle imaging (MPI)). The MI probe is usually introduced intravenously as shown, but can be introduced by almost any route (eg, orally). The total mass of MI probe introduced is a key issue and in some cases several different probes can be simultaneously injected (each intended for a different target). The next step is to allow the MI probe time to distribute throughout the living subject (Panel B). This allows enough time for the MI probe to reach its target(s) and if needed to clear from tissues where there is no target. This process can be done while the subject is within the imaging instrument or prior to placing the subject in the imaging instrument. Shown are the MI probes reaching their intracellular targets (blue spheres) where many but not all MI probes have bound. Subsequently, the detection of signal from the MI probes distributed throughout the body can now be performed (while the MI probe continues to clear). This may require placing the subject in an external field (Panel C) and/or exciting the distributed MI probes with energy (as in fluorescence optical imaging) so that they may provide a detectable signal. In the example shown, the probes are radioactive and do not require an external field, although one is being produced (as shown by the gray transducer emitting waves) to serve as a reminder of what is needed for different (eg, fluorescence optical imaging) strategies. The production of signal by the MI probes are then detected by one or more detectors placed around the subject. After collection of signal for a period of time, statistically sufficient data may be available to develop projection images (without depth information) and/or tomographic images (with full depth information) to visually represent the distribution of the MI probe throughout the field-of-view studied (Panel D). In the illustrated example, coronal PET images are shown and represent MI probe signal originating from a small lung tumor, the brain, and from the MI probe cleared by the kidneys into the bladder. In this case, low levels of MI signal throughout the rest of the body are also seen. Finally, repeat imaging to characterize changes in the biodistribution of the MI probe, with or without further mathematical modeling of the imaging data, to map biodistribution of the MI probe to the concentration of the molecular target(s) of interest can also be performed.

studies is that the MI probe beha ves similarl y to that observed in the animal model(s). Studying the biodistribution of the MI probe, signal to backg round in tar get tissue sites, signal in non-target tissues, and lack of toxicity in humans are all critical f irst steps. Studying the routes of elimination and metabolism of the MI probe is also usuall y impor tant. A f ailure to pro vide suf ficient signal to background at the target sites will usually lead to stopping further clinical trials. Quite often, the initial MI probe will not be optimal for clinical applications, but it will provide a lot of valuable information that will allow the MI research chain (see F igure 1) to be gin again. It could be that a ne w related molecular target(s) will be selected or more lik ely a new MI probe against the same molecular target(s) will be studied in detail to translate it for clinical applications. After fur ther clinical trials, reimbursement for the MI procedure by insurance carriers (including the Government) is key to help further its clinical use. A mathematical representation for all of MI of living subjects is given by Equation (1). Although highly simplified, this equation helps unify man y of the important characteristics of MI discussed throughout this textbook. The reader is also refer red to F igure 4, which shows how one goes from molecule to molecular image to help fur ther consolidate the infor mation presented. Signal Measured (x´, y´, z´, t, Δt) = Function ( Target Molecule Concentration (Δx, Δy, Δz, t), MI Probe Mass introduced into Subject, Pharmacokinetics of MI Probe, Field Properties Subject is Placed in, Output Signal from MI Probe, Signal Penetration Through Subject, Efficiency of Signal Detection by Imaging Instrument, Subject Preparation ) + Noise

Eq. (1)

Equation (1) needs signif icant discussion in order for the reader to understand v arious important details as outlined next. The signal measured from spatial coordinates x´, y´, z´ represents a measurement made using a physical detector, w hich is often but not al ways located outside the living subject. The detector itself usuall y occupies a

General Principles of Molecular Ima ging

volume but can effectively be considered a signal at some spatial coordinates x´, y´, z´. An imaging detector can also be placed within the li ving subject (e g, a catheter with an optical detector for intraoperative MI). The signal is often measured with man y detectors so that multiple signals emanating from all around the subject can be acquired, or alter natively the signal itself can be tagged with spatial information (eg, spatial encoding with gradients in MRI). This allows in many cases the tomographic reconstruction of images (with multiple vir tual imaging slices through the subject) to determine the spatial distribution of the injected molecular probe. The signal measured has to be collected for some time Δt to collect enough signal to obtain statistically useful data. Just as a conventional photo graphic camera that collects visib le light has to have its shutter open for a small Δt, so does a signal measurement for a MI instrument. This variable Δt determines the temporal resolution of the instr ument and therefore molecular processes that occur over a very rapid time scale relative to Δt will not be measurable or imaged by a given MI technique. The target molecule concentration in some volume element (Δx, Δy, Δz) at some time t is dependent on the underlying biolo gy and histor y of the li ving subject. For e xample, the concentration of a gi ven molecular receptor in the brain of a li ving subject may be dependent on exposure of the subject to a specif ic drug over the last 72 hours. The levels of a specif ic cell surf ace receptor ma y be dependent on transfor mation of the cell from a nor mal to a cancer cell. The parameter t is distinct from Δt, but in most cases it is assumed that during the time-period of signal acquisition ( Δt) the target molecular concentration does not change significantly. The MI probe mass introduced (eg, injected intravenously) is a critical f actor as one might e xpect. If a greater mass is injected , then potentiall y more probe can reach the molecular tar get(s) of interest. This can potentially mean a greater signal from the target molecule site(s) allo wing for shor ter acquisition times Δt. However, as greater masses of the probe of interest are injected, this leads to a greater potential for toxicity to the living subject or possib le signif icant per turbation to the subject. It is also possib le that the signal measured is a function of the concentration of more than one target molecule. For example, if a MI probe has to first be transported across the cell membrane and then bound to an intracellular receptor , then the signal of interest is a function of both the concentration of the transporter molecule and the intracellular receptor concentration.

7

PHARMACOKINETICS OF MI PROBE One of the key difficulties in imaging living subjects as compared to imaging cells e x vi vo is the inability to fully control the beha vior of the MI probe once introduced into the living subject. Whereas in ex vivo studies one might introduce an imaging probe and then simpl y wash away the excess, when studying living subjects no equivalent procedure e xists. Man y f actors including but not limited to chemical proper ties of the MI probe (eg, its lipophilicity), b lood flow, per meability, ability to cross the b lood-brain bar rier, routes of clearance (eg, renal vs hepatobiliar y), and metabolism influence the deli very and biodistribution of the MI probe. The pharmacokinetics of the MI probe is dependent on many of these f actors and can be mathematicall y modeled as detailed in Chapter 74, “Phar macokinetic Modeling.” This modeling is critical to e xtract quantitative infor mation re garding molecular tar get concentrations. It is important to note that it is often necessary to have multiple measurements over time (Signal Measured [x´, y´, z´, Δt1], Signal Measured [x´, y´, z´, Δt2], …, Signal Measured [x´, y´, z´, Δtn]) to be able to fully quantitate levels of target molecules in most cases. This might mean keeping the subject in the field-of-view of the detectors for a significant period of time to perform so-called “dynamic” imaging as opposed to a single measurement to obtain a “static” image.

PROPERTIES OF THE FIELD IN WHICH THE SUBJECT IS PLACED Magnetic, acoustic, and optical f ields in which the subject is placed can allow the signal to be created and/or detected. Magnetic f ields are critical in all for ms of MRI because without these f ields there w ould be no preferential alignment of protons within the body and no w ay to inter rogate the relaxation proper ties of those protons follo wing radiofrequency (RF) e xcitation. As such these f ields are also necessary to monitor changes in the relaxation properties of those protons following the accumulation of a nearby paramagnetic (eg, gadolinium-based) or superparamagnetic (eg, iron-o xide-based) MI probe (see also Chapters 3, “PET/MRI” 6, “Small Animal SPECT , SPECT/CT , and SPECT/MRI” and 54, “Dif fusion MRI: A Biomarker for Early Cancer Treatment Response Assessment”). In molecular ultrasound , the acoustic f ield is needed to deli ver energy to the MI probe (eg, a targeted microbubble) so that it ma y oscillate (due to a gas contained within the microbubble) in the deli vered acoustic f ield and produce a distinct acoustic signal, w hich can then be detected

8

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

(see also Chapter 28, “Ultrasound Contrast Agents”). Radionuclide-based methods do not require an e xternal field because the energy for the signal is “pre-contained” in the radioactive atoms that will produce the signal when they decay. In other cases such as optical fluorescence imaging, the MI probe has a fluorophore w hich must f irst absorb the appropriate w avelength of light in order for the MI probe to produce light of a dif ferent w avelength. In that case, energy has to be deli vered to the MI probe and is not “pre-contained” in the fluorophore.

detectors in order to be measured. If the detectors are outside the li ving subject, there is a cer tain probability that a gi ven signal will reach a specif ic detector. Unfortunately, it is rare for this probability to be high due tothe fact that man y signals do not penetrate through tissues. Some types of signal ha ve a much g reater probability of passing through tissues (e g, gamma ra ys for PET) than others (eg, visible light for optical imaging). The physical properties of the tissues through w hich the signal must penetrate determine the probability of detection.

OUTPUT SIGNAL FROM MI PROBE

EFFICIENCY OF SIGNAL DETECTION BY IMAGING INSTRUMENT

The output signal from the MI probe is k ey as it provides the means by which the spatial distribution of the molecular probe ma y be deter mined. The signaling component of the MI probe ma y consist of a radioacti ve atom (as in single photon emission CT) or a gas that responds with oscillations in an ultrasound f ield (as with microbubb lebased MI probes) (see Chapter 15, “Ultrasound”). Any portion of the physical spectrum as well as sound may be used to provide signal from the MI probe. Dif ferent portions of the physical spectrum will do better in penetrating through tissues, making some techniques more useful at all depths and others much more limited.There are several important issues related to the output signal. Ideall y, the MI probe w ould onl y produce the output signal after it finds and interacts with its intended molecular tar get(s). This would allow for a v ery low backg round signal and more accurate deter mination of the location(s) of the molecular tar get(s) of interest. Such a smar t probe (see Figure 3) is possib le for some MI strate gies (eg, fluorescence optical imaging) and not for others (e g, radionuclide-based strategies). Another key issue is w hether the output signal can be generated onl y once (e g, as with a radioactive atom) or multiple times (e g, a photoacoustic agent that can produce sound as long as it is e xcited and re-excited by light). Linked to this concept is the issue of whether the output signal requires e xcitation or not. It is also possible to de velop MI probes that pro vide multiple types of signals by having multiple signaling components (eg, for positron emission tomo graphy [PET] and optical (see Chapters 9, “Optical Multimodality Technologies,” and 29, “Multimodality Agents”)).

SIGNAL PENETRATION THROUGH SUBJECT The signal being produced b y the MI probe, w hether at the tar get molecule site(s) or not, needs to reach the

Even if a MI probe produces a signal that successfull y penetrates through the tissues of the li ving subject and reaches the detector, it may not be able to be stopped and actually detected b y the detector . In man y cases, a v ery small percentage of the total signal emitted is actuall y detected by the detectors. In PET , as little as 1 to 2% of the total signal produced may be detected. This efficiency of detection is a k ey variable since it will af fect the ability to deter mine v ariables such as Δt so that suf ficient signal can be detected for deter mination of the tar get molecule concentration. Some instr uments attempt to collect the signal from all around the subject b y using multiple detectors, w hereas others do not (sometimes to save costs), w hich leads to a fur ther decrease in o verall instrument efficiency.

SUBJECT PREPARATION It should not be underestimated that the manner in which the subject is prepared can signif icantly impact the measured signal. F or e xample, the ef fects of anesthesia can alter the MI probe phar macokinetics in the body . Shaving hair on the surf ace of a small li ving subject can significantly affect the detected signal in optical imaging. Of course, in some cases, one w ants a specif ic perturbation (eg, introduction of a drug to the subject) to produce a change in the le vels of molecular tar get(s) of interest, however, this is different from subject preparation per se. Keeping in mind these types of issues is key for quantitation and interpretation of the molecular images.

NOISE Noise in the measurement can be due to tw o primar y sources, random or str uctured. Random noise or statistical noise is directl y related to the detected number of

General Principles of Molecular Ima ging

signals from the MI probe. Structured noise refers to nonrandom v ariations in counting rates w hich adv ersely affect the inter pretation and anal ysis of the resulting images (eg, due to or gan/tissue motion, imaging instr ument non-uniformities).

CONCLUSION The f ield of MI of li ving subjects in volves man y biomedical disciplines for its progress to date and further continued evolution. These include physics and engineering for fundamental detector/instr ument design and construction, chemistr y and materials science for development of MI probes, molecular phar macology for optimized delivery and phar macokinetics of MI probes, cell/molecular biolo gy for understanding molecular targets of interest and for repor ter gene-based strategies, advances in genomics, proteomics and high-throughput screening technolo gies for disco very and v alidation of new molecular tar gets, mathematics and bioinfor matics for image reconstr uction and image/data modeling, and clinical medicine for applications of the strate gies for medical management. In addition, man y fields including immunology, microbiolo gy, and de velopmental biolo gy are using and helping to adv ance the v arious MI technologies. The MI f ield is limited b y the amount of currently a vailable trained indi viduals in the v arious sub-disciplines, and there continues to be a shor tage in several of the k ey areas needed to help e volve the f ield (eg, chemistry). Some key factors related to the optimism and caution for the f ield of MI of living subjects require a brief discussion. There continues to be optimism that the

9

molecular specificity provided by MI remains one of its key strengths. As biology and pathology are more full y understood at the molecular le vel, it remains lik ely that MI techniques will for m the basis for inter rogation of all living subjects. Some will argue that we may eventually not need MI because w e will be ab le to detect and treat disease with very specific drugs without requiring knowledge of the spatial localization of the disease. For example, a b lood test might detect a panel of protein biomarkers indicati ve of earl y cancer . Highl y specif ic drugs for this cancer could then be administered without any imaging, and serial b lood biomarker measurements would be used to monitor response of the subject to therapy. These types of in vitro assays have the adv antage o ver MI in that the y can be highl y multiple xed, potentially assa ying thousands of molecular e vents simultaneously. Ho wever, it ma y still be the case that some sites of disease are responding to treatment, whereas others are not; the b lood-based detection will not detect this. It is more lik ely that a combination of strategies, which include MI (e g, measuring blood protein biomark ers and perfor ming molecular/anatomical imaging), will lik ely play an increasing role in clinical disease management for the foreseeab le future. Ne vertheless, it is impor tant to objectively study and discuss the advantages and disadvantages of MI relative to nonimaging strategies as well as to objectively compare various imaging modalities with each other. The MI field is still in its inf ancy, and careful research should help usher in a ne w generation of technolo gies/assays that will continue to fundamentall y change our understanding of biolo gy and patholo gy and the clinical management of patients.

2 IMAGING OF STRUCTURE AND FUNCTION WITH PET/CT DAVID W. TOWNSEND, PHD

Historically, instr umentation for cross-sectional (tomographic) imaging of function, single photon emission computed tomography (SPECT) and positron emission tomography (PET) evolved along a path somewhat different f rom t hat o f a natomic i maging, c omputed tomography (CT) and magnetic resonance (MR) imaging (MRI), and the cor responding clinical studies w ere performed and interpreted separately in different medical departments, nuclear medicine and radiolo gy, respectively. Despite this se gregation, the use of combining anatomic and functional planar images w as e vident to physicians even in the 1960s, 1 preceding the invention of CT. The alignment of tomo graphic images is a comple x procedure owing to the lar ge number of de grees of freedom and without some common features, such as li ver and lung boundaries and boney structures, co-registration may be prob lematic. In addition to simple visual alignment, or the use of stereotactic frames that are undesirable or incon venient for diagnostic imaging, sophisticated image fusion software was developed from the late 1980s onwards.2 For (relatively) rigid parts, such as the brain, software can successfully align images from MR, CT, and PET, whereas in more flexible parts, such as the rest of the body, accurate alignment is more dif ficult owing to the lar ge number of possib le dif ferences (degrees of freedom) between the two data sets. Software fusion is also dependent on matching common features that are e xtracted either from the images (lung boundaries) or from the mark ers placed on the patient. Functional imaging modalities, such as PET and SPECT, often lack reliable anatomic cor relates and ha ve lower resolution and higher noise levels than CT or MR. One way to address the prob lems of software fusion is b y combining de vices (emission and transmission) rather than fusing the images post hoc, an approach that 10

has now coined the term hardware fusion. A combined or multimodality scanner , such as PET/CT , can acquire coregistered structure and function in a single study. The data are complementary allowing CT to accurately localize the functional abnormalities and PET to highlight the areas of abnor mal metabolism. A fur ther adv antage of combined instr umentation is that the anatomic images from CT can be used to impro ve quantitation of functional images through more accurate attenuation and scatter and par tial v olume cor rections; attenuation (absorption) and scatter of the annihilation photons occurs as the y interact with the tissue of the patient. These cor rections are impor tant to achie ve accurate and objective assessment of functional parameters, such as myocardial perfusion, tumor uptak e values, and dosimetry for treatment planning and monitoring response. Since the commercial introduction of PET/CT in 2001, adoption of the technolo gy has been rapid , particularly in oncology. Advances in CT and PET instrumentation ha ve been incor porated into the v ery latest PET/CT designs. A recent ar ticle3 offers an e xcellent overview of PET imaging, and it summarizes major advances in the instrumentation. This chapter will describe some of the early work that led to the commercial exploitation of PET/CT , mo ve forw ard to the cur rent designs, and subsequently discuss the impact of recent advances in CT and PET performance on these designs. Since photon attenuation (absor ption) is the parameter specifically measured by the CT scanner, an algorithm to derive correction factors from the CT images to compensate for the attenuation ef fects in PET images (CT -AC) will be proposed. The practical challenges to implement such an algorithm will also be addressed. The chapter will conclude with a brief re view of the clinical impact of PET/CT.

Imaging of Structure and Function with PET/CT

HISTORICAL CONCEPTS The origins of tomo graphic imaging in medicine date from the 1960s or e ven earlier, whereas fusion of tomographic images w as not investigated systematically until the late 1980s. 2 Following on from the earlier superposition of planar images, 1 the 1990s witnessed the de velopment of tw o principle approaches to image fusion: the software approach and the hardw are approach. The software approach attempts to align tw o image sets post hoc after they have been acquired from two different imaging modalities at tw o dif ferent times. In contrast, the hardware approach combines the instr umentation for tw o imaging modalities and thus acquires both modalities within the same reference frame thereb y ensuring as accurate alignment as possible.

Image Fusion with Software A thorough discussion of the topic is be yond the scope of this chapter. However, it is instructive to review some of the principles of softw are fusion and the e xtent to w hich they can be successful; a thorough re view of softw are fusion methods can be found in Ha wkes and colleagues, 4 and Slomka5 presents specific aspects of merging anatomic and molecular information and of ho w they relate to combined PET/CT design. Fusion of two image sets is achieved either by identifying common landmarks, b y f iducials that can then be aligned, or b y optimizing a metric based on image intensity values. Whatever the method, the number of possible de grees of freedom between the tw o image v olumes defines the comple xity of the subsequent mathematical transformation, the function that, when applied to one image set, converts it into the reference frame of the other image set. For distributions that do not involve a change in shape or size, rigid body transfor mations are adequate. When shears (or a nonisotropic dilation without shear) are in volved, an affine transfor mation comprising a linear transfor mation and translation is indicated. When there are no constraints on the deformation, a nonlinear transfor mation (warp) is used. Although methods involving the alignment of extracted features or fiducials6–8 have shown some success, at least for the brain, most cur rently used methods are intensity based , and images are coregistered by assessing the intrinsic information content. Metrics include intensity ratios 9 and mutual information.10 While such techniques have shown considerable success in aligning images of the brain obtained with CT, PET, SPECT, and MR, the y have been less successful for other parts of the body. Earlier clinical assessment in the lung11 and the pelvis 12 were disappointing, demonstrating a

11

local misalignment of the two image sets of the order of 5 to 8 mm, compared with a re gistration accuracy of better than 2 mm that can be achieved for the brain.13 A recent review14 suggests that softw are fusion can achie ve an accurac y of about half a pix el, or 2 to 3 mm, for some studies although clinical results from more recent generations of fusion software have not been particularly encouraging for example, in recurrent colorectal cancer.15 Software de velopment has nevertheless continued, as illustrated b y the recent pub lication of an automated w arping algorithm to align CT and PET images of the thorax. 16 Commercially a vailable softw are has considerably improved over the past fe w years both in the accurac y of the registration algorithms and in the sophistication of the user interface and displa y. As an e xample, Hermes Medical Solutions (Stockholm, Sw eden) of fers adv anced fusion software for man y clinical applications, including correction of misalignment errors for PET/CT scans, registration of PET/CT scans with MR, re gistration of longitudinal PET/CT studies, alignment of PET and MR scans in Alzheimer’s disease and other for ms of dementia, and registration of SPECT or PET myocardial perfusion studies with CT or MR of the hear t. Fusion software can also play an impor tant role in radiation therap y planning, where PET images are used to def ine the treatment plan.17,18 Fusion of the PET/CT study with the simulation CT scan that is used to plan the radiotherapy treatment can result in modif ication of that standard CT -only treatment plan in a signif icant percentage of cases, par ticularly for disease of the lung. Ho wever, despite considerab le progress, fusion softw are will probab ly ne ver compete with the simplicity and con venience of the core gistered studies acquired on a combined PET/CT scanner.

Multimodality Prototypes The pioneering w ork of the late Br uce Hase gawa and colleagues19,20 in the 1980s set the stage for the hardw are solution to image fusion. The aim of this w ork w as to design a de vice that could perfor m emission (radionuclide) and transmission (X-ray) tomography with the same detector that, in this case, w as se gmented high-purity germanium.20 Although this approach is attracti ve, it is difficult to design a detector that does not compromise performance for at least one of the tw o modalities. The work was signif icant, however, in that it highlighted the strengths of a single de vice that can perfor m both anatomic (CT) and functional (SPECT) imaging. 21 Of comparable significance was the use of the CT images to generate attenuation cor rection f actors (A CFs) for the

12

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

emission (SPECT) data.22 These factors are used to correct the SPECT images for the attenuation (absorption) of the photons as the y traverse the patient from the point of emission to the detector . The device was used for studies in phantoms, artificial constructions used to assess instrument performance, and in animals, par ticularly a study of myocardial perfusion in a porcine model. 23 However, recognizing the dif ficulty of building a detector that w ould operate optimall y for both CT and SPECT , Hase gawa turned to a dif ferent design that comprised a clinical SPECT gamma camera in tandem with a clinical singleslice CT scanner .24 The CT scanner (9800 Quick; GE Healthcare) was positioned in front of, and aligned with, a scintillation camera (600 XR/T ; GE Healthcare). The same bed was used to acquire both studies, and the images were registered by taking into account the axial displacement betw een the CT and the SPECT imaging f ields. After injection of the radiotracer and an uptake period, the patient was scanned first in the CT and subsequently in the SPECT scanner. The CT data were used to generate the SPECT ACFs. The combined de vice w as used to acquire a small number of clinical studies, such as those for quantitati ve estimation of radiation dosimetr y in patients with brain tumor.25 The proposal to combine PET with CT w as made in the early 1990s by Townsend and colleagues independently of the Hase gawa w ork. This concept originated from an earlier low-cost PET scanner design that comprised rotating banks of bismuth germanate (BGO) block detectors that w as de veloped b y Townsend and colleagues at the University of Geneva in 1991. The gaps between the banks of BGO detectors of fered the possibility to incorporate a different imaging modality, such as CT, within the PET scanner. Thus, the concept of PET/CT w as bor n with a 1991 proposal that the components of a CT scanner would be mounted in the gaps betw een the banks of BGO b lock detectors. The suggestion w as also made to use the CT images to generate the PET ACFs.26 The f irst prototype PET/CT scanner became operational in 1998,27 designed and built b y CTI PET Systems in Kno xville, TN (no w Siemens Molecular Imaging) and clinicall y e valuated at the University of Pittsburgh. The design incorporated a single-slice spiral CT scanner (Somatom AR.SP; Siemens Medical Solutions, F orchheim, Ger many) and a rotating ECAT ART scanner (CTI PET Systems). The PET detectors were mounted on the rear of the CT suppor t and the entire assembly rotated as a single unit (Figure 1). The data processing included an algorithm28 to scale the CT images from X-ra y ener gy to PET annihilation photon ener gy (511 keV) and generate the appropriate ACFs (see section “CT-Based ACFs”). Ov er 300 patients with cancer w ere

Figure 1. The first PET/CT prototype evaluated clinically at the University of Pittsburgh. The CT and PET components were mounted on a single rotating support and the data acquired from two separate consoles. The CT images were transferred to the PET console and then used for CT-based attenuation correction and localization.

scanned on the prototype, and the findings are presented in a series of publications.29–31 The results from the prototype demonstrated the impor tance of high-resolution anatom y accurately registered with functional data. The coregistered anatomy localized the functional abnor malities and clarified the equivocal situations, thus improving the accuracy and conf idence of the scan inter pretation. The use of a rapidly acquired, low-noise CT scan in place of a length y conventional PET transmission scan impro ved image quality and reduced scan time.

CURRENT PET/CT INSTRUMENTATION At the end of the 1990s, for a physician wishing to review fused images, the only real option was the software fusion techniques described above in the section “Image Fusion With Software.” Apart from the drawbacks of fusion software mentioned above, access to image data from dif ferent modalities w as f ar from a routine procedure, e ven with picture archiving and communication systems available. Thus, fusion imaging w as typically perfor med for, at most, only a small number of patients. Software fusion packages were, nevertheless, available on many imaging systems and par ticularly those used for radiation oncolhen GE ogy.32 This situation then changed in 1999 w Healthcare launched a dual-head scintillation camera combined with a lo w-power X-ra y tube and detectors,

Imaging of Structure and Function with PET/CT

called the Ha wkeye (GE Healthcare). 33,34 This design features tw o rectangular sodium iodide camera heads with a 350-W X-ra y tube. The Ha wkeye w as the f irst commercial scanner to of fer combined anatomic and functional imaging in a single unit. Less than 2 y ears after the f irst Hawkeye installation, PET/CT scanners incor porating clinical CT and clinical PET perfor mance became commerciall y available. The f irst commercial PET/CT scanner to be announced w as the Disco very LS (GE Healthcare) in early 2001. This was followed a few months later by the Biograph (Siemens Medical Solutions), and then, somewhat later b y the Gemini (Philips Medical Solutions). In the past 7 years, PET/CT designs from all vendors have evolved following the advances in CT and PET instr umentation that is described later . As to the present situation (2008), f ive v endors worldwide now offer PET/CT designs: GE Healthcare, Hitachi Medical, Philips Medical Systems, Toshiba Medical Corporation, and Siemens Medical Solutions. Cur rent designs of fered b y Siemens Molecular Imaging, GE Healthcare, and Philips Medical Systems are summarized in F igure 2. The specif ications and perfor mance of the PET components are v endor specif ic, with the Biograph H I-REZ TruePoint ( Figure 2 A; Si emens Medical Solutions) providing good spatial resolution in 3D with 4 mm × 4 mm × 20 mm lutetium oxyorthosilicate (LSO) cr ystals35; the original Bio graph design (Biograph Classic) w as based on 6.4 mm × 6.4 mm × 25 mm LSO detectors. The Bio graph is cur rently offered with 6-, 40-, and 64-slice CT scanners. The Discovery LS, the original PET/CT design from GE

B

A

Biograph 6, 40, 64

LSO 6.4 6.4 25 mm3 4 4 20 mm3 3D only (no septa) 6, 40, 64 slice CT 70 cm port 21.6 cm axial FOV 4.5 ns coincidence bed on rails

Discovery ST, STE, VCT, RX

BGO, LYSO 3 4.7 6.3 25 mm (BGO) 4.2 6.2 20 mm3(LYSO) 2D/3D (septa) 8, 16, 64 slice CT 70 cm port 15.7 cm axial FOV 11.7 ns coincidence dual-position bed

C

Gemini GXL, TF

GSO, LYSO 3 4 4 30 mm (GSO) 4 4 22 mm3(LYSO) 3D only (no septa) 6, 10, 16, 64 CT 71.7 cm port 18 cm axial FOV 6 ns coincidence bed support in tunnel

Figure 2. Current PET/CT scanner designs from three of the major suppliers of medical imaging equipment: A, the Siemens Biograph TruePoint, B, the GE Healthcare Discovery range, and C, the Philips Gemini series. See Table 1 for the physical properties of the different scintillators; LYSO is lutetium oxyorthosilicate (LSO) with a small percentage of yttrium.

13

Healthcare, combined the Advance NXi PET scanner with a 4-, 8-, or 16-slice CT .36 The Disco very ST (Figure 2B; GE Healthcare) has 6.2 mm × 6.2 mm × 30 mm bismuth ger manate (BGO) detectors in combination with a 4-, 8-, or 16-slice CT scanner; unlik e the Discovery LS, the gantr y of the PET scanner matches the dimensions of the CT scanner . The higher resolution Disco very STE has 4.7 mm × 6.3 mm × 30 mm BGO detectors in combination with 8- or 16-slice CT scanners; the Disco very VCT is an STE conf igured with a 64-slice CT scanner . Although not listed of ficially, the Discovery RX is a research tomograph based on the scintillator LYSO (LSO with a small percentage of yttrium) with detector geometr y comparable to that of the STE, including the retractable septa. The Gemini GXL (Figure 2C; Philips Medical) comprises 4 mm (in plane) and 6 mm (axial) gadolinium o xyorthosilicate (GSO) detector pix els, 30 mm in depth; the Gemini is also an open design with the capability to ph ysically separate the CT and PET scanners for access to the patient. The Gemini GXL incor porates a 6- or 16-slice CT scanner. A recent addition to PET/CT designs is the Gemini TF, the f irst commercial time-of-flight (T OF)PET scanner .37 The Gemini TF has 4 mm × 4 mm × 22 mm LYSO detectors and is combined with a 16- or 64-slice CT scanner . All designs e xcept Discovery LS provide a 70-cm patient por t for both CT and PET. While Discovery and Gemini of fer standard PET transmission sources as an option, in practice, as mentioned above, most, if not all, institutions use CT-based attenuation correction because of the advantages of low noise and shor t scan times that f acilitate high patient throughput. The Gemini and Bio graph acquire PET data in 3D mode onl y, whereas the Disco very series incorporates retractab le septa and can acquire data in both 2D mode with the septa e xtended and the 3D mode with the septa retracted. Since 2001, numerous pub lications ha ve no w documented the benef its of PET/CT compared with PET and CT, with and without software fusion. A good review of the literature prior to September 2006 can be found in Czernin and colleagues. 38

ADVANCES IN PERFORMANCE FOR CT AND PET Multidetector CT Scanners Following the de velopment of single-slice spiral CT scanners in the earl y 1990s, 39 CT performance has experienced a re vival with the adv ent of multidetector

14

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

arrays (MDCT). This was accompanied by increases in X-ray power (from 30 kW up to 60 kW or g reater) and computer capacity for data processing and image reconstruction. Dual and 4-slice CT scanners f irst appeared around 1998 with scan times of 500 ms, followed b y 16-slice (2002) and more recentl y, 64-slice (2004) CT scanners. The increasing number of detector rows (slices) has been accompanied b y f aster rotation times so that state-of-the-ar t scanners can now achieve a full rotation in as little as 330 ms. Spatial resolution has improved from about 10 line pairs (Lp)/cm in 1990 up to 25 Lp/cm or better toda y, with a slice thickness less than 1 mm. A signif icant innovation that will contribute to increased CT perfor mance is the Straton X-ray tube. 40 In the Straton tube, the entire v acuum vessel including the anode and cathode rotates resulting in much more ef fective cooling and heat dissipation, which is a limitation of the con ventional tube. 41 Cooling rates 5 to 10 times higher than for con ventional tubes can be achie ved with the Straton tube that thus result in shor ter rotation times and f aster scans. Since weight is a problem for conventional X-ray sources, the dual-tube Def inition CT (Siemens Medical Solutions) is achie vable because of the Straton tube. After many years of slo w but steady pro gress, the past decade has seen signif icant adv ances in both hardw are and software for CT.

PET Scanners Muehllehner and Kar p3 offer an excellent review of progress in PET instr umentation, including a summar y of the physical performance of the new, fast scintillators recently introduced for PET. This section will summarize some of these advances as they relate to current PET/CT scanner performance.

introduction of new scintillators such as GSO42 and LSO,43 both doped with cerium, impro ved the perfor mance of PET scanners for clinical imaging. Both GSO and LSO have shorter decay times than BGO b y a f actor of 6 to 7, reducing system dead time and impro ving count rate performance, particularly at high activity levels in the field of view (FOV). The physical properties of these scintillators are compared in Table 1. Of even more importance for clinical imaging is the potential of f aster scintillators to decrease the coincidence timing window, thereby reducing the randoms coincidence rate. The increased light output of the ne w scintillators impro ves the ener gy resolution because the increased number of light photons reduces the statistical uncertainty in the energy measurement. However, other ph ysical ef fects contribute to the emission process, and the improvement in energy resolution is not a simple function of the number of light photons. The higher light output also increases the positioning accurac y of a block detector, allowing the b locks to be cut into smaller crystals, thereby improving spatial resolution. BGO, LSO, and GSO do not absorb moisture w hen exposed to air (ie, are not hygroscopic), thus facilitating the manufacture and packaging of the detectors. GSO is some what less r ugged and more dif ficult to machine than either BGO or LSO . LSO has an intrinsic radioacti vity of about 280 Bq/mL, with single photon emissions in the range 88 to 400 keV. Such a radioacti ve component is of little impor tance for coincidence counting at 511 keV, except maybe at very low emission count rates. Sensitivity

PET is intrinsically a 3D imaging methodology, replacing physical collimation required for single-photon imaging with the electronic collimation of coincidence detection. The f irst multi-ring PET scanners incor porated septa, lead, or tungsten annular shields mounted betw een the

New Scintillators for PET

In the 1970s, PET detectors sa w the transition from thallium-activated sodium iodide (NaI(Tl)) to BGO, a scintillator with higher density and photofraction. The photofraction is the fraction of incident annihilation photons that interact in the scintillator through the photoelectric effect; this is the desired process, in preference to Compton scattering that ma y involve multiple interaction points within the detector. While at least one PET scanner design continued to use NaI(Tl) until f airly recentl y, the majority of PET scanners installed during the 1990s w ere based on BGO block detectors. In the late 1990s, the

Table 1. PHYSICAL PROPERTIES OF PET SCINTILLATORS Property

NaI

Density (g/mL) 3.67 Effective atomic number 51 Attenuation length (cm) 2.88 Decay time (ns) 230 Photons/MeV 38,000 Light yield (%NaI) 100 Hygroscopic Yes

BGO

LSO

GSO

7.13 74 1.05 300 8,200 15 No

7.4 66 1.16 35–45 28,000 75 No

6.7 61 1.43 30–60 10,000 25 No

NaI = sodium iodide; BGO = bismuth germanate; LSO = lutetium oxyorthosilicate; GSO = gadolinium oxyorthosilicate.

Imaging of Structure and Function with PET/CT

detector rings. The purpose of the septa was to shield the detector rings from photons that scattered out of the transverse plane, thus restricting the use of electronic collimation to within the plane, a limitation that, while it mak es poor use of the radiation emitted from the patient, limits scattering and allows 2D image reconstruction algorithms to be used. The availability of BGO scanners from 1990 onwards with retractable septa encouraged the use of 3D methodology, at least for the brain, where the net increase of a factor f ive in sensitivity could be realized even with accompanying increases in both scatter fraction and randoms.44 The condition for whole body imaging is f ar less favorable, in part due to the presence of significant activity just outside the imaging FO V in most bed positions. Instead, par ticularly for lar ge patients, 2D imaging has been recommended although higher le vels of the injected biomarker, such as 2-deo xy-2-[F-18]fluoro-D-glucose (FDG), are required to obtain adequate count rates. This situation changed in the late 1990s with the introduction of LSO- and GSO-based scanners that could be operated with shor t coincidence time windo ws (4.5 to 6 ns) and higher energy thresholds (400 to 450 keV) compared with 10 to 12 ns and 350 k eV for a typical BGO scanner . Significantly impro ved w hole-body image quality has been achieved in 3D with a 10 mCi (370 MBq) injection of FDG. A recommended injected dose of 12 to 15 mCi corresponds to operation at peak noise equi valent count rate (NECR) for an LSO scanner in 3D.45 However, since the LSO and GSO scanners ha ve no septa and acquire data in 3D mode only, no comparison can be made with 2D operation. Within the past 2 to 3 y ears, a limited number of LYSO-based scanners with retractab le septa ha ve been evaluated in 2D and 3D operations, and recent publications suggest that 3D operation is no w prefer red o ver 2D operation.46–48 The sensitivity of a scanner can also be impro ved by the addition of more detector material. Planar sensiti vity can be increased b y extending the thickness of the scintillator. For example, a 50% increase in thickness (20 to 30 mm) results in a 40% increase in sensitivity. However, increasing the axial extent by 30% will result in a 78% increase in volume sensitivity (for 3D acquisition with no septa). The latter thus mak es more ef ficient use of the increased volume of LSO although there will also be an increase in the number of phototubes required (and hence increased cost). F ollowing an injection of a radioacti ve tracer, such as FDG, the patient recei ves a radiation dose from all annihilation photons, not just those emitted within the imaging FO V of the scanner . Therefore, the greater the axial co verage, the better use is made of the emitted radiation and the more ef ficient use is made of a

15

given volume of scintillator. For most PET/CT scanners, axial PET co verage is about 16 cm, with one design having an axial extent of 18 cm.37 The most recent design to be announced has an e xtended FOV covering 21.6 cm axially. The latter comprises over 32,000, 4 mm × 4 mm × 20 mm LSO pixels and images 109, 2-mm thick transaxial planes in a single position. Data acquisition is in fully 3D, and the scanner has a peak NECR of around 160 kcps.49,50

Signal-to-Noise

The availability of scintillators that are both fast and have sufficient density to stop a large fraction of incident photons (ie, high stopping power), such as LSO (and LYSO), has revived interest in PET TOF,51 interest that has been further stimulated by the announcement of the f irst commercial PET/CT with TOF, the Philips Gemini TrueFlight (TF).37 The principle of TOF PET is illustrated schematically in F igure 3; positron annihilation occurs in the patient at a distance d − d1 from one detector (Detector A) and d + d 1 from the other detector (Detector B). For photons traveling at the speed of light (c), the arrival time difference between the two photons at the detectors is 2d1/c. Photons originating from the center of the FO V (d1 = 0) obviously arrive in the detectors at the same time. Scanners with f ast scintillators and electronics can measure this time dif ference to within a cer tain resolution. For e xample, for a scanner with a coincidence timing resolution of 500 ps, the spatial uncer tainty on the position of the annihilation is 7.5 cm. This uncertainty is not sufficient to place the annihilation within a 2-mm voxel (and thereby eliminate reconstr uction), but it is superior to having no timing information at all and assigning equal probability to all v oxels along the line-of-response (Figure 3A). Instead , the most probab le location of the annihilation is at the center of the uncertainty distribution in F igure 3B . The TOF infor mation is incor porated directly into the reconstr uction algorithm leading to an improvement in signal-to-noise (SNR). The increase in SNR is proportional to ( D /δ d) , where D is the diameter of the activity distribution and δd is the spatial uncertainty. For a 40 cm diameter unifor m distribution and a 7.5 cm uncer tainty, the increase in SNR is a f actor of about 2.3. As the TOF resolution improves, the spatial uncertainty decreases and the SNR increases b y a lar ger factor. TOF PET w as f irst in vestigated in the earl y 1980s51 with scintillators that w ere fast but did not ha ve good stopping po wer for 511 k eV photons. Interest declined until the recent emer gence of scintillators that

16

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

B

A

Figure 3. A schematic illustrating positron emission tomography (PET) data acquisition with the incorporation of Timeof-Flight (TOF) reconstruction. By measuring the time difference between the arrival of the two annihilation photons, the position of the positron annihilation along the line-of-response (LOR) can be localized with an accuracy dependent on the precision of the temporal measurement: A, without TOF information, the annihilation is located with equal probability along the LOR and B, using TOF information the annihilation point can be localized to a limited range, for example, a 500 ps timing resolution corresponds to 7.5 cm FWHM.

are both f ast and sensiti ve. The new TOF PET scanners based on LSO or L YSO must demonstrate good timing resolution that is stable over time so as to a void frequent detector recalibration. While promising, the clinical impact of TOF PET has yet to be established, although it is anticipated to ha ve a role in the imaging of lar ge patients. This is because the lar ger the diameter of the activity distribution, the g reater the potential increase in SNR for a given time resolution. A more detailed review of the pub lished contributions to TOF development can be found in Muehllehner and Kar p.3 Reconstruction Algorithms

There has been signif icant progress during the past fe w years in image reconstruction methods through the introduction of statisticall y based algorithms into the clinical setting. Pre viously, one of the earliest and most widel y used 3D reconstr uction methods w as the reprojection algorithm (3DRP) based on a 3D e xtension of standard 2D filtered backprojection.52 While this algorithm works well for the lo wer noise en vironment of the brain, the quality for whole-body imaging is less than optimal, particularly when rod source ACFs are applied to lo w count emission data. F igure 4A sho ws a coronal image of a patient with a body mass index (BMI) of 25 reconstructed using 3DRP. Since CT-based ACFs have been applied, the quality is actuall y better than w ould have been obtained

with rod source ACFs. The development of Fourier rebinning (FORE) 53 was a breakthrough that enab led 3D data sets to be accuratel y rebinned into 2D data sets and then reconstructed in 2D with a statistically based expectation maximization (EM) algorithm. The adv antage of the rebinning step is that it accurately compresses a large 3D data set into 2D slices that can then be reconstructed separately using a 2D reconstr uction algorithm. Ev en so, it was not until the accelerated convergence achieved by the ordered-subset EM (OSEM) algorithm54 that those statistical methods became of clinical interest. While FORE and OSEM pro vide impro ved image quality compared with 3DRP, a fur ther advance was the incor poration of attenuation information directly into the reconstruction model in the for m of w eighting f actors: attenuationweighted (AW) OSEM.55 Figure 4B shows the same data set as in F igure 4A reconstr ucted with FORE and AWOSEM; the improved image quality is evident. Further improvement has been achieved by eliminating the rebinning step and implementing OSEM fully in 3D with corrections for randoms, scatter, and attenuation incorporated into the system model. 56,57 The result, again for the same data set, is shown in Figure 4C. Finally, in a recent development ter med high-def inition (HD) PET , the detector spatial response function has also been included in the reconstr uction model. 58 The point-spread function (PSF) v aries throughout the FO V o wing to the ob lique penetration of the detectors b y annihilation photons. By measuring this variability and then modeling the PSF, improved and near -uniform spatial resolution can be achieved throughout the FOV; the improvement can be seen by comparing F igure 4C with the PSF reconstr uction in F igure 4D; all reconstr uctions e xcept 3DRP are unsmoothed. The images in Figure 4 are reconstructed with clinical softw are pro vided b y a specif ic v endor (Siemens Molecular Imaging). Of course, most major vendors provide comparable software capable of producing clinical images of high quality . The VUE point algorithm (GE Healthcare) is an implementation of 3D OSEM that includes cor rections for randoms, scatter , and attenuation and also a z-axis smoothing.The Gemini TF (Philips Medical Systems) has TOF capability; and therefore, the TOF infor mation must be incor porated into the reconve struction.37 For their Gemini scanners, Philips ha implemented a distributed list-mode TOF algorithm (DLT) that is based on a TOF list-mode maximum likelihood approach de veloped b y P opescu and colleagues.59 They pre viously used a ro w-action maximum likelihood algorithm or RAMLA.60 The scatter correction algorithm requires modif ication to incor porate

Imaging of Structure and Function with PET/CT

A

B

C

D

17

Figure 4. A coronal section of an FDG-PET whole-body scan of a patient with a body mass index of 25 acquired in 3D mode with septa retracted and reconstructed using: A, 3D filtered backprojection algorithm with reprojection (3DRP) (7 mm Gaussian smooth), B, FORE + 2D OSEM (14 subsets, 2 iterations; no smoothing), C, 3D OP-OSEM (14 subsets, 2 iterations; no smoothing), and D, HD PET: 3D OSEM with point-spread function reconstruction (14 subsets, 2 iterations, no smoothing).

TOF infor mation. The g reatest outstanding ef fect on image quality and a challenge to reconstr uction algorithms is due mainly to the size of the patient, a signif icant problem given the cur rent levels of obesity among the US population.

CT-BASED ACFS For PET/CT, a reco gnized strength is the a vailability of CT images for attenuation correction of the PET data,28,61 eliminating the need for an additional, length y transmission scan. The use of the CT to generate ACFs not onl y reduces the scan time b y a signif icant amount but also results in more accurate ACFs. Since the attenuation values (µ) are energy dependent, the CT scan at a mean photon energy of ~70 keV must be scaled to PET (511 k eV) energy. The mean energy of a polychromatic X-ray beam is def ined as the ener gy of a monochromatic beam that would give the same linear attenuation as the polychromatic beam inte grated over energy.62 The polychromatic beam also results in beam hardening, the preferential interaction of lo wer energy photons as the beam traverses the body causing the mean ener gy to increase and the corresponding µ values to decrease.

Energy Scaling Algorithm for CT The attenuation of X-rays by tissue depends on the density and the ef fective atomic number (Z eff) of the

material. At these ener gies, the ph ysical processes b y which X-rays are attenuated are the photoelectric effect and Compton scattering. The photoelectric probability varies appro ximately as Z eff4 and scales as 1/E 3 with photon energy (E). The Compton scattering probability has little dependence on Zeff and decreases linearly with E. The linear attenuation coefficient for a given material is expressed by the sum of the two components: µ(E) = ρe{σc(E) + σph(E, Zeff)} where ρe is the electron density and σph and σc are the photoelectric and Compton cross-sections per electron, respectively. Electron density is the number of electrons per unit v olume, although w hen quantum mechanical effects are significant, it is the total probability of finding an electron within a unit v olume. Ho wever, at photon energies above about 100 keV in tissue, the photoelectric contribution is essentiall y ne gligible compared with the Compton contribution, and therefore, the e xpressions for the attenuation coefficient at X-ray energy Ex and gamma energy Eγ are: µ(Ex) = ρe{σc(Ex) + σph(Ex, Zeff)} µ(Eγ) = ρe σc(Eγ) As a consequence of the tw o separate contributions to µ(Ex), a single measurement of µ(Ex) will not uniquely determine µ(Eγ) because, for example, an increase in Z eff could of fset a decrease in ρe resulting in no change in µ(Ex). In general, therefore, a simple energy scaling of

18

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

µ(Ex) is insuf ficient to yield µ(Eγ). By restricting the problem to biolo gic tissues for w hich Z eff are all f airly comparable and noting that the contribution from σph is relatively small even at X-ray energies, changes in µ(Ex) are primarily due to changes in the electron density characteristic of the different tissue types and not to differgic ences in Z eff. Thus, for the limited range of biolo tissues, a single scaling f actor can be used to con vert µ(Ex) to µ(Eγ) for lung, li ver, fat, muscle, and other soft tissues. F or spongiosa and cor tical bone, ho wever, the same scale factor will not apply because of the significant calcium and phosphorous content of bone tissue that result in Z eff different from other tissues. This issue has been addressed 28 by se gmenting bone from soft tissue at a specif ic threshold and applying different scale f actors to the tw o different tissue classif ications, bone and nonbone, corresponding to different values of Zeff. Kinahan and colleagues28 adopted a threshold of 300 Hounsfield units (HU). Subsequentl y, Watson and colleagues62 proposed a mixture model in w hich all tissues with µ < µ(water) are treated as a mixture of air and w ater at v arious concentrations (k), while tissues with µ > µ(water) are treated as a mixture of water and cortical bone. Since this approach limits the composition to a single value for a given µ(Ex), a bilinear scaling function can be defined for biologic tissues, as sho wn in Figure 5. Recent pub lications on CT -based attenuation cor rection for PET also propose a break point at 0 HU ( µ value for w ater)63 although the most appropriate choice ma y be slightl y greater than zero because some soft tissues and b lood

µ(Eγ) (cm−1)

µmix

air–water

k . µcb (1

k) . µw

water–bone

µ(EX) (HU) Figure 5. The bilinear scaling function used to convert CT numbers (Hounsfield units) to linear attenuation values at 511 keV. The attenuation correction factors are generated by reprojecting the µ-map at 511 keV; w = water and cb = cortical bone; k is the concentration of the components of the mixture.

conform to the air–water mix but with densities greater than water for which a break point around 60 HU is more appropriate. The calibration function has been deri ved from phantom measurements and has also been v alidated with patient data. 64 The calibration of the CT scanner ensures that the soft tissue v alues (µ < 60 HU) are independent of the kVp setting of the X-ra y tube. This independence does not apply to bone-like tissue with µ > 60 HU, and therefore, different lines (slopes) are required for each kVp setting. 65 The CT is acquired before the emission data so the ACFs can be generated for the entire v olume. The CT images at ~70 keV are resampled to the spatial resolution of the emission data. Then, the images are scaled v oxel-byvoxel to the ener gy of the emission data b y applying the bilinear scaling function (see Figure 5). The scaled CT images are then forw ard projected to generate ACFs that match the sampling of the emission data. Since the introduction of the PET/CT scanner , CT -based attenuation correction has been a signif icant focus of research to address the v arious possib le ar tifacts. The follo wing sections will re view the status of this w ork and the outstanding challenges that remain for CT -based attenuation correction.

Artifacts Specific to CT-Based Attenuation Correction While the benefits of CT-based attenuation are now well known and documented , a number of challenges ha ve emerged as the technique has become more widel y adopted for PET/CT.66,67 There are tw o main concer ns: (1) the presence of materials in the patient with Z eff values that do not confor m to the basic assumptions in the bilinear model and (2) mismatch between the CT and PET due to patient respiration, cardiac motion, and bowel movement.68 Since the f irst commercial PET/CT installation in 2001, these issues have received considerable attention. Examples of the f irst concer n include metallic objects, 69,70 dental hardware,71 calcified lymph nodes, and intravenous72,73 and oral contrast. 74,75 Materials with high Zeff may even exceed the dynamic range of attenuation v alues measurab le b y CT , and se vere artifacts can be generated in the images. Of par ticular importance in the assessment of head and neck cancer is the presence of dental f illings.71 A number of metal artifact reduction techniques ha ve been in vestigated,76 including modif ied reconstr uction methods 77 and segmentation approaches 78 that can signif icantly reduce the artifacts.

Imaging of Structure and Function with PET/CT

Some characteristic artifacts associated with CT-based attenuation cor rection are sho wn in F igure 6. When slow regular (tidal) breathing is adopted for both CT and PET, respiration ef fects include an apparent displacement of the dome of the liver into the lower lobe of the right lung79 (Figure 6A) creating a cor responding region of apparent activity on the PET scan (ar row). A cur ved re gion of apparent low uptak e (photopenic) at the top of the li ver and spleen in the PET image (Figure 6B) is also observed in some studies. Although such ar tifacts ma y occur for any patient follo wing a tidal breathing protocol, 80 the documented incidence is much reduced for faster, higher performance CT scanners. F igure 6C is an e xample of a study acquired on a 6-slice CT scanner that sho ws no evidence of breathing ar tifacts or misre gistration. The clinical signif icance of respirator y artifacts has been studied for an earl y PET/CT design in a series of 300 patients81 and was found to result in around 2% of incorrect diagnoses.

enhance CT v alues without a cor responding change in density, and it is used in CT to enhance attenuation v alues in the v asculature b y increasing the photoelectric absorption compared with the blood. CT contrast results in a 40% change in attenuation at CT ener gies, whereas, at 511 k eV where the photoelectric ef fect is ne gligible, the presence of contrast has onl y a 2% ef fect or less on attenuation.82 However, if contrast-enhanced tissue pixels are misidentif ied as a w ater–bone mix, the scaling factor will be incorrect and the erroneously scaled pixels may generate ar tifacts in the PET image 83 (Figure 6D, top row). Tens of thousands of PET/CT scans ha ve now been perfor med in the presence of intra venous contrast and experience has shown that contrast administration does not generally cause a problem that could potentially interfere with the diagnostic value of PET/CT.72,84,85 This is lar gely due to the f act that intra venous contrast is fairly rapidly dispersed throughout the v ascular system. An exception may be the passage of the contrast bolus through a major v essel although e ven this does not always generate an artifact on the PET image (Figure 6D, bottom row). Optimized CT protocols ha ve been de veloped for the administration of intra venous contrast that avoid most of the issues. 86 A recent pub lication87 has documented the rate of tw o of the 100 patients studied , where an incor rect management decision w ould ha ve been made because of the use of noncontrast, lo w-dose

Intravenous Contrast

The use of intra venous contrast ma y be indicated w hen the CT scan is performed for clinical pur poses as opposed t o l ow-dose C T p erformed f or a ttenuation correction and localization only. Intravenous contrast contains iodine at concentrations high enough to

A

D

19

B

C

E

F

Figure 6. Potential image artifacts generated from CT-based attenuation correction: A, an artifact due to respiration in which the dome of the liver is displaced into the base of the right lung, B, curved photopenic areas above the liver and the spleen caused by CT and PET mismatch from respiratory movement of the diaphragm, C, an example of a well-registered study that is free of artifacts, D, the variable effects of intravenous contrast showing an artifact on the PET image (top row) due to a contrast bolus and the absence of an artifact on PET (bottom row), E, the effect of oral contrast where the presence of contrast in the GI tract does not cause an artifact on the PET image (arrow), and F, the effect of dental fillings on the CT and PET images.

20

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

CT acquired for localization, and CT -AC onl y. In one case, a cystoscopy was performed in a patient with l ymphoma to e valuate a mass near the b ladder missed b y both the CT and PET scans of the PET/CT ; the mass, seen on the diagnostic clinical CT, was worrisome for bladder cancer , but the c ystoscopy w as nor mal and the mass responded to treatment for lymphoma. The second management change also involved a patient with lymphoma, where a colonoscopy and surgical removal of the neoterminal ileum was required to diagnose nodules due to Crohn’s disease. The nodules w ere well seen on the diagnostic clinical CT but not on the CT or PET scans of the PET/CT. Oral Contrast

Oral contrast is administered to enhance the gastrointestinal tract and the distribution of the contrast material is some what v ariable, both in spatial distrib ution and level of enhancement. Modifications to the basic scaling algorithm ha ve been introduced to distinguish oral contrast-enhanced pix els from bone pix els. 82 As with intra venous contrast, there is little e vidence that the presence of oral contrast results in diagnostic errors of any signif icance.88 Figure 6E shows a patient imaged with oral contrast; enhancement of the colon on the CT image (left; ar rows) sho ws no cor responding artifactual uptake on the PET image (right). Ne vertheless, in some protocols, contrast CT is perfor med in addition to the lo w-dose CT for attenuation cor rection and localization, thereb y increasing the radiation dose to the patient. Ho wever, a lo w-dose w hole-body CT in addition to a clinical CT with contrast over a limited axial range (single PET bed position) ma y involve less radiation dose than a w hole-body clinical CT with contrast. Metal Implants

Dental ar tifacts can be cor rected on CT through the use of no vel reconstr uction techniques, 77 as sho wn in Figure 6F. The uncor rected (left) and cor rected (right) images for CT (top) and PET (bottom) demonstrate that although the reconstr uction algorithm signif icantly improves the CT image, it has v ery little impact on the PET image, v erifying that CT-based attenuation cor rection i s a ctually a r obust t echnique. While m etallic implants, such as ar tificial hip prostheses, can sometimes cause artifacts on CT, this appears to be due more to patient mo vement between the CT and the PET scan

than to the presence of prostheses per se , as demonstrated by Kaneta and colleagues. 89 Even so, it would be somewhat rare for the specif ic pathology under study to be located in the re gion af fected b y ar tifacts from the prosthesis. The nonattenuation corrected image is, in any case, available to resolve ambiguities. Respiratory Motion

Within the past 6 years, the most widely addressed issue related to CT-based attenuation cor rection has been respiratory motion90–93 and the artifacts created by the mismatch betw een CT and PET .94 Rotating 68Ge sources used in conventional PET scanners resulted in a transmission scan that a veraged patient respiration in a w ay compatible with the corresponding emission scan. The use of CT-AC suggests a number of dif ferent protocols that must be in vestigated to resolv e the mismatch problem. For example, the advent of fast, spiral CT scanners made breath-hold CT a reality although clinical images are typicall y acquired with full inspiration to separate lung structures. Such an expansion of the chest does not match a PET scan acquired with shallo w breathing and results in serious attenuation cor rection ar tifacts in the anterior chest w all. The appearance of ar tifacts due to respiratory motion and the spatial and temporal mismatch between CT and PET images has led to an intensive research initiative to identify the best respiratory protocol. A number of dif ferent protocols ha ve been investigated, including: • Continuous shallow breathing for both CT and PET 90 • CT acquired with limited breath-hold o ver diaphragm90 • Breath-hold CT acquired with par tial inspiration90 • Motion-averaged CT over many respiration cycles95,96 • Cine CT acquiring full breathing c ycle per slice 97 • Respiratory-gated CT; PET with shallow breathing98 • Deep inspiration breath-hold 99,100 • Breath-hold CT; gated PET 101,102 • Respiratory-gated CT and PET.103 Currently, the simplest and most widel y used protocol is shallo w breathing for both CT and PET .90 Early single- or dual-slice PET/CT designs showed a high incidence of breathing ar tifacts (F igures 6A and 6B) 80 but with the incorporation of fast MDCT into PET/CT scanners, the incidence of such ar tifacts has been considerably reduced. Ho wever, the CT images still do not exactly match the motion-averaged PET acquisitions

Imaging of Structure and Function with PET/CT

and protocols, such as slo w CT acquisition, ha ve also been in vestigated. The clinical signif icance of these attenuation cor rection ef fects continues to be debated , particularly with respect to lesions in the base of the lung and dome of the liver, where curved photopenic areas are observed (Figure 6B). Displacement of such lesions may result in incor rect localization or , w orse, a f ailure to identify them correctly leading to misdiagnosis. Shallow breathing during PET/CT has been sho wn to be inadequate for the comprehensi ve staging of lung cancer .104 Nevertheless, a signif icant percentage of studies acquired on even a 6-slice CT scanner sho w good registration with shallow breathing. Finally, two other effects can also influence the accuracy of CT-AC: the tr uncation of the transv erse FOV105 and the presence of scattered radiation. Truncation of the FOV arises because CT scanners typicall y have a 50 cm diameter FOV, whereas PET supports 60 cm. Simple software e xtrapolation techniques ha ve pro ved ef fective in extending the CT FOV to match that of PET, at least with accuracy adequate for CT -AC.106,107 Scatter is enhanced by imaging with the arms of the patient in the FOV. However, the short scan times achievable with state-of-the-art PET/CT allo w patients to easil y tolerate imaging with arms raised, reducing the effects due to increased scatter. The exception is head and neck cancer, where the patient is scanned with arms down. Despite the issues discussed above and rare opinions to the contrar y,108 CT-based attenuation cor rection has become the de facto standard for PET/CT although it can be affected by the ar tifacts described abo ve. The advantages, w hich include con venience and shor t acquisition times, largely outweigh the drawbacks. In a small number of studies, quantitati ve comparisons ha ve been made between ACFs generated from standard PET transmission scans and from CT 92,109,110; and although some differences in SUV values have been noted, nothing of diagnostic significance has been documented.

RADIATION DOSE CONSIDERATIONS Patient exposure to radiation from a PET/CT scan is both external from the CT scan and inter nal from the PET injected radionuclide.111

External Dose Dose assessment in CT is challenging and depends not only on the body re gion exposed but also on a v ariety of scan-specif ic parameters including tube potential

21

(kVp), the product of tube cur rent and e xposure time (in milliamp × seconds, mAs), slice collimation, and pitch.112 In addition, the dose also depends on certain technical features of the scanner , such as beam f iltration, beam shaping f ilter, geometry and the acquisition algorithm. Therefore, values for CT patient dose v ary considerably between centers and between scanners. The tendenc y is to o versimplify the situation b y not taking all of these factors into account. For whole body CT scans that extend from the level of the thyroid to the pubic symphysis, the effective CT dose Eext can be estimated approximately as: Eext = Γ CT · CTDIvol where Γ CT = 1.47 mSv/mGy is the dose coef ficient that relates the volume CT dose index CTDIvol to the effective dose. F or a typical set of clinical scan parameters, the CTDIvol is 13 mGy 113 resulting in a total effective wholebody dose of 19 mSv. However, many centers acquire the CT scan for attenuation cor rection and localization onl y, reducing the whole-body dose to as low as 3 mSv or less. In addition, there are a number of strate gies to make better use of the radiation, such as tube cur rent modulation and automatic exposure control.114,115

Internal Dose The internal radiation dose will depend upon the biodistribution and the ph ysical and biolo gic half-life of the biomarker. The dose is e xpressed as the radiation e xposure to the whole body and individually to the various organs. The critical organs are those that receive the maximum radiation dose. The ef fective dose Eint resulting from intra venous administration of a gi ven biomark er with activity A can be estimated from: Eint = Γ · A where Γ is a dose coef ficient computed for the adult hermaphrodite MIRD phantom. The only clinical biomarker of interest is FDG for w hich the dose coef ficient is 19 µSv/MBq,116 although a higher dose coef ficient of 29 µSv/MBq has also been published.117 The dose coefficient holds for standard patients with a body w eight of about 70 kg and is generic rather than patient specif ic since age, sex of patients, and individual pharmacokinetics are not taken into account. In fact, the radiation risk is somewhat higher for females and for younger patients when compared with males and older patients. Age and sex-specific dose coefficients can be found elsewhere.118

22

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

On the basis of the published value116 for the dose coefficient, the average whole-body dose for a typical 10 mCi (370 MBq) injection of FDG is 7 mSv . However, most biomarkers do not distribute unifor mly in the body , and the critical or gan with FDG, for e xample, is the b ladder due to excretion through the urinary system.

Total Radiation Dose The total ef fective dose for PET/CT is the sum of the internal and e xternal doses. F or a full y clinical CT and FDG-PET scan, the effective dose will be around 25 mSv. However, this can be reduced to 10 mSv or less w hen a low-dose CT is acquired for localization and attenuation correction only. In practice, the PET/CT dose to a specific organ will depend upon the exact protocol; for example, if the CT scan does not include the b ladder, the dose to the bladder wall will be due entirely to FDG. For a smaller patient imaged on a high-sensitivity scanner, a lower FDG dose can be used, potentially limiting the effective dose to 5 mSv or less. The worldwide average annual dose due to the natural radioactive background is 2.4 mSv.

THE CLINICAL ROLE OF PET/CT Prior to the introduction of PET/CT, essentially all multimodality clinical imaging w as based on softw are fusion techniques, 14 limited mainl y to the brain. The

introduction of the Ha wkeye scanner (GE Healthcare) in 1999, follo wed less than 2 y ears later b y the f irst commercial PET/CT scanner , has ir reversibly transformed the field of multimodality imaging. From 2001, the sales of PET -only scanners decreased to zero by 2006, completel y replaced b y PET/CT (F igure 7). Currently, in 2008, a w orldwide installed base of o ver 2500 units attests to the rapid adoption of the modality by physicians. The majority of this installed base is in routine clinical operation and there is, at least for oncolo gy, now a g rowing body of literature that suppor ts the accuracy of staging and restaging with PET/CT compared with either CT or PET acquired separately.38,119 Many of these pub lications are within the past 3 or 4 years, and they clearly document the significant impro vements in specif icity and to some e xtent also in sensiti vity, and especiall y in earl y detection of cancer recur rence.120 These improvements are incremental when compared with PET that alone demonstrates high le vels of sensiti vity and specif icity for a wide range of disease states. Impro ved accurac y has been documented for a v ariety of cancers including head and neck, 76,121 thyroid,122 lung,123–125 breast,126,127 esophageal, 128,129 colorectal,15,130 and melanoma. 131 There is also evidence that PET/CT improves accuracy in l ymphoma132 and solitar y pulmonar y nodules, 133,134 in spite of the f act that in l ymphoma the accurac y of PET alone is very high. 135

Figure 7. Shipments of PET and PET/CT scanners for the US market as recorded by the Nuclear Equipment Manufacturers Association (NEMA) for the period January 2002 to October 2007. Note that the figures (in $M) reflect the total revenue for all shipments from which the selling price and individual unit type cannot be determined. Shipments of PET-only scanners declined during this period to zero from January 2006 onwards. The overall market for PET or PET/CT remained fairly constant throughout this period, although since January 2007, with the reduction in reimbursement due to the introduction of the Deficit Reduction Act, sales have declined somewhat.

Imaging of Structure and Function with PET/CT

A

B

Figure 8. Two studies acquired on a Biograph 6 TruePoint TrueV PET/CT scanner. Transaxial sections are for PET (top row) and fused images (bottom row): A, a 50-year-old female patient with a diagnosis of pancreatic cancer (arrow). The images were acquired 94 min after injection of 10.3 mCi of 18FDG. The total scan duration was 10 min with acquisition of five bed positions at 2 min per position. The CT was acquired at 130 kVp and 180 mAs (Siemens CAREDose). B, a 58-year-old female patient with metastatic renal cell cancer (arrow) from an unknown primary. The images were acquired 110 min after injection of 9.7 mCi of 18FDG. The total scan duration was 15 min with acquisition of five bed positions at 3 min per position. The CT was acquired at 130 kVp and 180 mAs (Siemens CAREDose).

In summary, therefore, the impro vement in accurac y of PET/CT compared with PET or CT for staging and restaging is statisticall y signif icant and a verages 10 to 15% over all cancers.38 To illustrate typical state-of-the-art PET/CT scans, Figure 8 shows two studies acquired on a Biograph 6 TruePoint TrueV PET/CT (Siemens Molecular Imaging) with a 21.6 cm axial FO V. Figure 8A shows transaxial PET and fused images of a 50-y ear-old female patient with a diagnosis of pancreatic cancer . The images were acquired 94 minutes after injection of 10.3 mCi of 18 FDG. The total scan duration w as 10 minutes with acquisition of five bed positions at 2 minutes per position. The CT was acquired at 130 kVp and 180 mAs (Siemens CAREDose). The images demo nstrate intense focal uptake of 18FDG in a primar y neoplasm 3.4 × 2.6 cm in size that can be accurately located in the head of the pancreas (arrow). No FDG uptake was identified in any of the proximal nodes although the lik elihood of micrometastases w ould be high. F igure 8B sho ws a 58-y ear-old female patient with metastatic renal cell cancer from an unknown primary. The images were acquired 110 minutes after injection of 9.7 mCi of 18FDG. The total scan duration was 15 minutes with acquisition of five bed positions at 3 minutes per position. The CT w as acquired at 130 kVp and 180 mAs (Siemens CAREDose). The study

23

demonstrates a lar ge FDG a vid peripherall y-enhancing necrotic mass occup ying the anterior mid and lo wer left kidney. The mass is 10 cm in size and appears to in volve the lower pole collecting system (ar row). Another application for which PET/CT is also having an impact is that of radiotherap y treatment planning. Incorporation of FDG-PET images into therapy planning was already taking place prior to the introduction of PET/CT32 using softw are fusion techniques. 136,137 In some cases, the a vailability of the PET images led to a change in treatment plan by redefining the biologic target volume based on FDG uptak e. This w as par ticularly effective for the lung, 138 where reactive changes, such as a par tial or full collapsed lung (atelectasis), could be distinguished from malignanc y as a result of the differential uptak e of FDG. Reasonab le re gistration accuracy at the centimeter level could be achieved locally through the use of f iducials although the software fusion techniques were cumbersome and labor -intensive. From the inception, PET/CT pro vided more con venient and routine access to fused CT and PET images and earl y assessment of the consequences of using PET/CT in planning139–141 was encouraging. More recent surveys18,142 have confirmed the earlier conclusions. Molecular imaging with PET/CT is increasingl y being used to monitor response to therap y,143 for chemotherapy,144–147 for radiation therap y,148,149 and for combinations of each. 150 It has become increasingly evident that simple response e valuation criteria for solid tumors (RECIST) 151 based on anatomic measures of tumor size may not be adequate to accuratel y assess therapy response. The molecular signal is lik ely to be a more sensitive indicator as it reflects tumor metabolism rather than just tumor size. A metabolic change ma y be more suggestive of a response than a size change. The combination of ha ving both CT and PET for monitoring response offers a number of unique possibilities in spite of the technical difficulties associated with CT-based attenuation cor rection. Firstly, the anatomic and the functional volume of the tumor can be estimated, the former from CT measurements and the latter by summing all voxels with a Standardized Uptake Value (SUV) above a threshold that defines malignanc y. Therapy response can be assessed from changes in both these metrics or from a change in the total lesion glycolosis that is calculated as the product of the a verage SUV in the tumor and the v olume.152 The advantage of the CT is that an accurate measurement of tumor volume is available both before and after treatment. It is also helpful and more reliab le to def ine the tumor region-of-interest (R OI) directl y on the CT and then to transfer the same ROI onto the PET image. The boundary

24

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

of the tumor ma y be difficult to deter mine from the PET scan, par ticularly for metabolic responders as the lesion SUV decreases. The CT images ma y also be used to improve partial volume correction, such as by dividing the SUV from the PET image with a reco very coef ficient based on the spherical tumor diameter. Since tumors generally have a complex shape, a more sophisticated par tial volume cor rection method is desirab le.153 Thus, for both technical and practical reasons, PET/CT is continuing to successfully promote the use of PET for monitoring response to different forms of therapy. Cardiac PET/CT applications are in their inf ancy154 and ha ve recentl y encountered a number of dif ficulties. Obviously, the effects of cardiac and respiratory motion are critical for these studies. The problems of mismatch associated with CT -based attenuation cor rection discussed above are potentially more serious for cardiac studies than they are for oncolo gy in that all cardiac studies will be affected rather than just those w hole-body studies with lesions in cer tain sensitive regions, such as the lung. This misregistration results in w hat appears to be perfusion deficits in se gments of the hear t associated with the misalignment. A recent publication155 finds that up to 40% of cardiac PET/CT studies could be af fected by misregistration. A number of different strategies are being de veloped to address this issue, including (1) manual realignment of CT and PET , (2) acquiring a cine CT of the breathing motion and generating an a verage CT for attenuation correction, and (3) acquiring multiple CT scans to ensure at least one matches the PET scan as closel y as possib le. Obviously, the role of PET/CT in cardiolo gy has yet to be determined, but if a strong clinical demand e xists, it is to be expected that transient technical challenges, such as the misalignment issue, will ultimately be solved. A complete review of the status of PET/CT in cardiology can be found in DiCarli and Lipton. 156 Recently, the f irst human MR/PET design has been evaluated following the de velopment of MR-compatib le PET detectors. 157 The design, w hich comprises a PET detector ring inser ted into a 3-T MR, has been used for simultaneous MR and FDG-PET imaging of the human brain.158 While ultimately the aim is to develop a wholebody MR/PET that could f ind applications in oncolo gy and cardiology, cur rent devices are limited to the brain. There have been suggestions that MR/PET could eventually replace PET/CT ,159 although the question is reall y whether adding a PET inser t to MR will attract clinical applications a way from PET/CT . This seems unlik ely since both CT and MR have strengths in specif ic clinical areas, a situation that will probably not change because of the addition of PET . The realization of the full potential

of MR/PET must await the development of a satisfactory whole-body design.

CONCLUSION There is little doubt that, o ver the past 6 y ears, PET/CT has had a growing impact on clinical imaging and particularly s taging a nd r estaging d isease a nd m onitoring response to therap y. Although the technolo gy has been somewhat disruptive in the sense that it has brought together medical specialties that ha ve not traditionall y worked together, the overall impact has been positive. To meet the demand for cross-training of both the technologists w ho operate the de vices and the ph ysicians w ho interpret the studies, guidelines ha ve been pub lished160 and new standards established leading to a somewhat different situation today from the way radiology and nuclear medicine ha ve traditionall y functioned. This trend is likely to continue as other multimodality de vices reach the clinic, including SPECT/CT that w as introduced in 2004 and MR/PET that is cur rently under evaluation for brain imaging.

ACKNOWLEDGMENT This chapter is dedicated to the memor y of Professor Bruce Hase gawa from the Uni versity of Califor nia, San Francisco, a pioneer of the combined imaging of structure and function. He was a friend and colleague and his work was an inspiration to us all.

REFERENCES 1. Wagner HN. A personal histor y of nuclear medicine. London, UK: Springer; 2006. 2. Levin DN, Pelizzari CA, Chen GT, et al. Retrospecti ve geometric correlation of MR, CT, and PET images. Radiology 1988;169:817–23. 3. Muehllehner G, Kar p JS. P ositron emission tomo graphy. Phys Med Biol 2006;51:R117–37. 4. Hawkes DJ, Hill DL, Hallpik e L, Baile y DL. Core gistration of str uctural and functional images. In: Valk P, Bailey DL, Townsend DW, Maisey M, editors. P ositron emission tomo graphy: basic science and clinical practice. New York: Springer-Verlag; 2003. p. 181–98. 5. Slomka PJ. Software approach to mer ging molecular with anatomic information. J Nucl Med 2004;45 Suppl 1:S36–45. 6. Pelizzari CA, C hen GT , Spelbring DR, et al. Accurate threedimensional re gistration of CT , PET, and/or MR images of the brain. J Comput Assist Tomogr 1989;13:20–6. 7. Pietrzyk U , Herholz K, Heiss WD. Three-dimensional alignment of functional and morphological tomograms. J Comput Assist Tomogr 1990;14:51–9. 8. Mountz JM, Zhang B , Liu HG, Inampudi C. A reference method for correlation of anatomic and functional brain images: validation and clinical application. Semin Nucl Med 1994;24:256–71. 9. Woods RP, Mazziotta JC, Cher ry SR. MRI-PET re gistration with automated algorithm. J Comput Assist Tomogr 1993;17:536–46.

Imaging of Structure and Function with PET/CT

10. Maes F, Collignon A, Vandermeulen D, et al. Multimodality image registration by maximization of mutual infor mation. IEEE Trans Med Imaging 1997;16:187–98. 11. Wahl RL, Quint LE, Cieslak RD, et al. “Anatometabolic” tumor imaging: fusion of FDG PET with CT or MRI to localize foci of increased activity. J Nucl Med 1993;34:1190–7. 12. Hamilton RJ, Blend MJ, Pelizzari CA, et al. Using v ascular structure for CT -SPECT re gistration in the pelvis. J Nucl Med 1999; 40:347–51. 13. West J, Fitzpatrick JM, Wang MY, et al. Comparison and evaluation of retrospective intermodality brain image registration techniques. J Comput Assist Tomogr 1997;21:554–66. 14. Hutton BF, Braun M. Softw are for image re gistration: algorithms, accuracy, efficacy. Semin Nucl Med 2003;33:180–92. 15. Kim JH, Czer nin J, Allen-Auerbach MS, et al. Comparison betw een 18 F-FDG PET, in-line PET/CT, and softw are fusion for restaging of recurrent colorectal cancer. J Nucl Med 2005;46:587–95. 16. Slomka PJ, Dey D, Przetak C, et al. Automated 3-dimensional registration of stand-alone 18F-FDG whole-body PET with CT. J Nucl Med 2003;44:1156–67. 17. Dizendorf E, Cier nik IF , Baumer t B , et al. Impact of inte grated PETCT scanning on e xternal beam radiation treatment planning. J Nucl Med 2002;43:33P. 18. Gregoire V, Haustermans K, Geets X, et al. PET-based treatment planning in radiotherapy: a new standard? J Nucl Med 2007;48 Suppl 1:S68–77. 19. Hasegawa BH, Gingold EL, Reilly SM, et al. Description of a simultaneous emi ssion-transmission CT system. Proc SP IE 1990; 1231:50–60. 20. Hasegawa BH, Stebler B, Rutt BK, et al. A prototype high-purity germanium detector system with f ast photon-counting circuitr y for medical imaging. Med Phys 1991;18:900–9. 21. Lang TF, Hasegawa BH, Lie w SC, et al. Description of a prototype emission-transmission computed tomo graphy imaging system. J Nucl Med 1992;33:1881–7. 22. Hasegawa BH, Lang TF, Brown EL, et al. Object specif ic attenuation correction of SPECT with correlated dual-energy X-ray CT. IEEE Trans Nucl Sci 1993;NS-40:1242–52. 23. Kalki K, Blank espoor SC, Brown JK, et al. Myocardial perfusion imaging with a combined X-ra y CT and SPECT system. J Nucl Med 1997;38:1535–40. 24. Blankespoor SC, Wu X, Kalki K, et al.Attenuation correction of SPECT using X-ray CT on an emission-transmission CT system: myocardial perfusion assessment. IEEE Trans Nucl Sci 1996;43:2263–74. 25. Tang HR, Da Silv a AJ, Mattha y KK, et al. Neurob lastoma imaging using a combined CT scanner-scintillation camera and 131I-MIBG. J Nucl Med 2001;42:237–47. 26. Beyer T, Kinahan PE, Townsend DW, Sashin D. The use of X-ray CT for attenuation correction of PET data. Nuclear Science Symposium and Medical Imaging Conference, 1994. 1994 IEEE Conference Record 1994;4:1573–7. 27. Beyer T, Townsend DW, Brun T, et al. A combined PET/CT scanner for clinical oncology. J Nucl Med 2000;41:1369–79. 28. Kinahan PE, Townsend DW, Beyer T, Sashin D. Attenuation correction for a combined 3D PET/CT scanner. Med Phys 1998;25:2046–53. 29. Charron M, Be yer T, Bohnen NN , et al. Image anal ysis in patients with cancer studied with a combined PET and CT scanner . Clin Nucl Med 2000;25:905–10. 30. Kluetz PG, Meltzer CC, Villemagne VL, et al. Combined PET/CT imaging in oncolo gy. Impact on patient management. Clin Positron Imaging 2000;3:223–30. 31. Meltzer CC, Luk etich JD , F riedman D , et al. Whole-body FDG positron emission tomo graphic imaging for staging esophageal cancer comparison with computed tomo graphy. Clin Nucl Med 2000;25:882–7. 32. Caldwell CB, Mah K, Ung YC, et al. Observer variation in contouring gross tumor volume in patients with poorly defined non-small-cell

33.

34.

35.

36.

37.

38.

39.

40.

41.

42. 43.

44.

45.

46.

47.

48.

49.

50.

51. 52. 53.

25

lung tumors on CT : the impact of 18FDG-hybrid PET fusion. Int J Radiat Oncol Biol Phys 2001;51:923–31. Patton JA, Delbeke D, Sandler MP. Image fusion using an inte grated, dual-head coincidence camera with X-ra y tube-based attenuation maps. J Nucl Med 2000;41:1364–8. Bocher M, Balan A, Krausz Y, et al. Gamma camera-mounted anatomical X-ray tomography: technology, system characteristics and first images. Eur J Nucl Med 2000;27:619–27. Brambilla M, Secco C, Dominietto M, et al. Performance characteristics obtained for a ne w 3-dimensional lutetium o xyorthosilicatebased whole-body PET/CT scanner with the National Electrical Manufacturers Association NU 2-2001 standard. J Nucl Med 2005;46:2083–91. Mawlawi O, Podoloff DA, Kohlmyer S, et al. Performance characteristics of a ne wly developed PET/CT scanner using NEMA standards in 2D and 3D modes. J Nucl Med 2004;45:1734–42. Surti S, Kuhn A, Werner ME, et al. Performance of Philips Gemini TF PET/CT scanner with special consideration for its time-of-flight imaging capabilities. J Nucl Med 2007;48:471–80. Czernin J, Allen-Auerbach M, Schelbert HR. Improvements in cancer staging with PET/CT : literature-based e vidence as of September 2006. J Nucl Med 2007;48 Suppl 1:S78–88. Kalender WA, Seissler W, Klotz E, Vock P. Spiral volumetric CT with single-breath-hold technique, continuous transpor t, and continuous scanner rotation. Radiology 1990;176:181–3. Schardt P, Deuringer J, Freudenberger J, et al. New X-ray tube performance in computed tomography by introducing the rotating envelope tube technology. Med Phys 2004;31:2699–706. Kalender WA. Technical concepts. In: Computed tomo graphy: fundamentals, system technology, image quality, applications. Erlangen, Germany: Publicis Corporate Publishing; 2005. p. 47. Takagi K, Fukazawa T. Cerium-activated Gd2SiO5 single crystal scintillator. Appl Phys Lett 1983;42:43–5. Melcher CL, Schw eitzer JS. Cerium-doped lutetium o xy-orthosilicate: a fast, efficient new scintillator. IEEE Trans Nucl Sci 1992; 39:502–5. Townsend D W, Isoardi RA, Bendriem B . Volume imaging tomographs. In: Bendriem B , Townsend DW, editors. The theor y and practice of 3D PET. Dordrecht, The Netherlands: Kluwer Academic Publishers; p. 111–32. Watson CC, Casey ME, Bendriem B , et al. Optimizing injected dose in clinical PET by accurately modeling the counting-rate response functions specif ic to indi vidual patient scans. J Nucl Med 2005; 46:1825–34. Kemp BJ, Kim C, Williams JJ, et al. NEMA NU 2-2001 performance measurements of an LYSO-based PET/CT system in 2D and 3D acquisition modes. J Nucl Med 2006;47:1960–7. Lodge MA, Badawi RD, Gilbert R, et al. Comparison of 2-dimensional and 3-dimensional acquisition for 18F-FDG PET oncolo gy studies performed on an LSO-based scanner. J Nucl Med 2006;47:23–31. Strobel K, Rudy M, Treyer V, et al. Objective and subjective comparison of standard 2-D and fully 3-D reconstructed data on a PET/CT system. Nucl Med Commun 2007;28:555–9. Jakoby BW, Bercier Y, Watson CC, et al. Ph ysical perfor mance and clinical w orkflow of a ne w LSO HI-REZ PET/CT scanner . Nuclear Science Symposium Conference Record , 2006 IEEE 2006;5:3130–4. Townsend DW, Jakoby B, Long MJ, et al. Performance and clinical workflow of a new combined PET/CT scanner. J Nucl Med 2007; 48:437P. Budinger TF. Time-of-flight positron emission tomography: status relative to conventional PET. J Nucl Med 1983;24:73–8. Kinahan PE, Rodgers JG. Analytic 3D image reconstruction using all detected events. IEEE Trans Nucl Sci 1989;36:964–8. Defrise M, Kinahan PE, Townsend DW, et al. Exact and appro ximate rebinning algorithms for 3-D PET data. IEEE Trans Med Imaging 1997;16:145–58.

26

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

54. Hudson H, Larkin R. Accelerated image reconstruction using ordered subsets of projection data. IEEE Trans Med Imaging 1994; 13:601–9. 55. Comtat C, Kinahan PE, Defrise M, et al. F ast reconstr uction of 3D PET data with accurate statistical modeling. IEEE Trans Nucl Sci 1998;45:1083–9. 56. Comtat C, Bataille F, Michel C, et al. OSEM-3D reconstr uction strategies for the ECAT HRRT. Nuclear Science Symposium Conference Record, 2004 IEEE 2004;6:3492–6. 57. Liu X, Comtat C, Michel C, et al. Comparison of 3-D reconstr uction with 3D-OSEM and with FORE+OSEM for PET . IEEE Trans Med Imaging 2001;20:804–14. 58. Panin VY, Kehren F, Michel C, Casey M. Fully 3-D PET reconstr uction with system matrix derived from point source measurements. IEEE Trans Med Imaging 2006;25:907–21. 59. Popescu LM, Matej S, Lewitt RM. Iterative image reconstruction using geometrically ordered subsets with list-mode data. Nuclear Science Symposium Conference Record, 2004 IEEE 2004;6:3536–40. 60. Daube-Witherspoon ME, Matej S, Kar p JS, Le witt RM. Application of the ro w action maximum lik elihood algorithm with spherical basis functions to clinical PET imaging. IEEE Trans Nucl Sci 2001;48:24–30. 61. Kinahan PE, Hase gawa BH, Be yer T. X-ray-based attenuation correction for positron emission tomography/computed tomography scanners. Semin Nucl Med 2003;33:166–79. 62. Watson CC, Townsend D W, Bendriem B . PET/CT systems. In: Wernick M, Aarsvold J, editors. Emission tomo graphy. London, UK: Elsevier Science; 2004. p. 195–212. 63. Burger C, Goerres G, Schoenes S, et al. PET attenuation coefficients from CT images: e xperimental evaluation of the transfor mation of CT into PET 511-k eV attenuation coef ficients. Eur J Nucl Med Mol Imaging 2002;29:922–7. 64. Watson CC, Rappopor t V, Faul D, et al. A method for calibrating the CT-based attenuation cor rection of PET in human tissue. IEEE Trans Nucl Sci 2006;53:102–7. 65. Carney JP, Townsend D W, Rappopor t V, Bendriem B . Method for transforming CT images for attenuation cor rection in PET/CT imaging. Med Phys 2006;33:976–83. 66. Cohade C, Wahl RL. Applications of positron emission tomo graphy/computed tomography image fusion in clinical positron emission tomography-clinical use, interpretation methods, diagnostic improvements. Semin Nucl Med 2003;33:228–37. 67. Bockisch A, Be yer T, Antoch G, et al. P ositron emission tomo graphy/computed tomography-imaging protocols, ar tifacts, and pitfalls. Mol Imaging Biol 2004;6:188–99. 68. Nakamoto Y, Chin BB, Cohade C, et al. PET/CT : artifacts caused by bowel motion. Nucl Med Commun 2004;25:221–5. 69. Cohade C, Osman M, Marshall L,Wahl RL. Metallic object artifacts on PET-CT: clinical and phantom studies. J Nucl Med 2002;43:308P. 70. Goerres GW, Ziegler SI, Burger C, et al. Artifacts at PET and PET/CT caused b y metallic hip prosthetic material. Radiolo gy 2003; 226:577–84. 71. Kamel EM, Burger C, Buck A, et al. Impact of metallic dental implants on CT-based attenuation correction in a combined PET/CT scanner. Eur Radiol 2003;13:724–8. 72. Yau YY, Coel M, Chan WS, et al. Application of IV contrast in PETCT: does it reall y produce attenuation cor rection er ror. J Nucl Med 2003;44:272P. 73. Antoch G, F reudenberg LS, Be yer T, et al. To enhance or not to enhance? 18F-FDG and CT contrast agents in dual-modality 18 F-FDG PET/CT. J Nucl Med 2004;45 Suppl 1:S56–65. 74. Carney JP, Beyer T, Brasse D, et al. Clinical PET/CT scanning using oral CT contrast agents. J Nucl Med 2002;45:57P. 75. Cohade C, Osman M, Nakamoto Y, et al. Initial e xperience with oral contrast in PET/CT : phantom and clinical studies. J Nucl Med 2003;44:412–6.

76. Schafers KP, Raupach R, Beyer T. Combined 18F-FDG-PET/CT imaging of the head and neck. An approach to metal ar tifact cor rection. Nuklearmedizin 2006;45:219–22. 77. Lemmens C, Nuyts J, Faul D, et al. Suppression of metal streak ar tifacts in CT using a MAP reconstruction procedure. Nuclear Science Symposium Conference Record, 2006 IEEE 2006;6:3431–7. 78. Mirzaei S, Guerchaft M, Bonnier C, et al. Use of segmented CT transmission map to avoid metal artifacts in PET images by a PET-CT device. BMC Nucl Med 2005;5:3. 79. Sarikaya I, Yeung HW, Erdi Y, Larson SM. Respiratory artefact causing malpositioning of li ver dome lesion in right lo wer lung. Clin Nucl Med 2003;28:943–4. 80. Romer W, Chung M, Chan A, et al. Single-detector helical CT in PET-CT: assessment of image quality . AJR Am J Roentgenol 2004;182:1571–7. 81. Osman MM, Cohade C, Nakamoto Y, et al. Clinicall y signif icant inaccurate localization of lesions with PET/CT : frequency in 300 patients. J Nucl Med 2003;44:240–3. 82. Carney JP, Townsend DW. CT-based attenuation correction for PET/CT scanners. In: von Schulthess GK, editor. Molecular anatomic imaging: PET-CT and SPECT-CT integrated modality imaging. Philadelphia (PA): Lippincott Williams and Wilkins; 2006. p. 54–62. 83. Antoch G, F reudenberg LS, Egelhof T, et al. F ocal tracer uptak e: a potential ar tifact in contrast-enhanced dual-modality PET/CT scans. J Nucl Med 2002;43:1339–42. 84. Berthelsen AK, Holm S, Loft A, et al. PET/CT with intravenous contrast can be used for PET attenuation correction in cancer patients. Eur J Nucl Med Mol Imaging 2005;32:1167–75. 85. Mawlawi O, Erasmus JJ, Munden RF, et al. Quantifying the ef fect of IV contrast media on integrated PET/CT: clinical evaluation. AJR Am J Roentgenol 2006;186:308–19. 86. Brechtel K, Klein M, Vogel M, et al. Optimized contrast-enhanced CT protocols for diagnostic w hole-body 18F-FDG PET/CT : technical aspects of single-phase versus multiphase CT imaging. J Nucl Med 2006;47:470–6. 87. Gollub MJ, Hong R, Sarasohn DM, Akhurst T. Limitations of CT during PET/CT. J Nucl Med 2007;48:1583–91. 88. Otsuka H, Graham MM, Kubo A, Nishitani H. The effect of oral contrast on lar ge bo wel acti vity in FDG-PET/CT . Ann Nucl Med 2005;19:101–8. 89. Kaneta T, Takanami K, Wakayama Y, et al. High-density materials do not always induce artifacts on PET/CT: what is responsible for the difference? Nucl Med Commun 2007;28:495–9. 90. Beyer T, Antoch G, Blodgett T, et al. Dual-modality PET/CT imaging: the ef fect of respirator y motion on combined image quality in clinical oncology. Eur J Nucl Med Mol Imaging 2003;30:588–96. 91. Osman MM, Cohade C, Nakamoto Y, Wahl RL. Respirator y motion artifacts on PET emission images obtained using CT attenuation correction on PET-CT. Eur J Nucl Med Mol Imaging 2003; 30:603–6. 92. Papathanassiou D, Becker S, Amir R, et al. Respiratory motion artefact in the liver dome on FDG PET/CT: comparison of attenuation correction with CT and a caesium e xternal source. Eur J Nucl Med Mol Imaging 2005;32:1422–8. 93. Bacharach SL. PET/CT attenuation cor rection: breathing lessons. J Nucl Med 2007;48:677–9. 94. Beyer T, Rosenbaum S, Veit P, et al. Respiration ar tifacts in w holebody 18F-FDG PET/CT studies with combined PET/CT tomographs emplo ying spiral CT technolo gy with 1 to 16 detector rows. Eur J Nucl Med Mol Imaging 2005;32:1429–39. 95. Pan T, Mawlawi O, Nehmeh SA, et al. Attenuation correction of PET images with respiration-a veraged CT images in PET/CT . J Nucl Med 2005;46:1481–7. 96. Chi PC, Mawlawi O, Nehmeh SA, et al. Design of respiration a veraged CT for attenuation correction of the PET data from PET/CT. Med Phys 2007;34:2039–47.

Imaging of Structure and Function with PET/CT

97. Alessio AM, Kohlmyer S, Branch K, et al. Cine CT for attenuation correction in cardiac PET/CT. J Nucl Med 2007;48:794–801. 98. Pan T, Lee TY, Rietzel E, Chen GT . 4D-CT imaging of a v olume influenced by respiratory motion on multi-slice CT . Med Ph ys 2004;31:333–40. 99. Meirelles GS, Erdi YE, Nehmeh SA, et al. Deep-inspiration breathhold PET/CT: clinical f indings with a ne w technique for detection and characterization of thoracic lesions. J Nucl Med 2007;48:712–9. 100. Nehmeh SA, Erdi YE, Meirelles GS, et al. Deep-inspiration breathhold PET/CT of the thorax. J Nucl Med 2007;48:22–6. 101. Dawood M, Buther F, Lang N , et al. Respirator y gating in positron emission tomography: a quantitative comparison of different gating schemes. Med Phys 2007;34:3067–76. 102. Martinez-Moller A, Zikic D, Botnar RM, et al. Dual cardiac-respiratory gated PET: implementation and results from a feasibility study. Eur J Nucl Med Mol Imaging 2007;34:1447–54. 103. Nagel CC, Bosmans G, Dekk er AL, et al. Phased attenuation correction in respiration correlated computed tomography/positron emitted tomography. Med Phys 2006;33:1840–7. 104. Allen-Auerbach M, Yeom K, P ark J, et al. Standard PET/CT of the chest during shallo w breathing is inadequate for comprehensi ve staging of lung cancer. J Nucl Med 2006;47:298–301. 105. Beyer T , Bockisch A, K uhl H, Mar tinez MJ . Whole-body 18 F-FDG PET/CT in the presence of tr uncation artifacts. J Nucl Med 2006;47:91–9. 106. Carney JP, Townsend DW, Kinahan PE, et al. CT-based attenuation correction: the effects of imaging with the ar ms in the f ield of view [abstract]. J Nucl Med 2001;42:P56–7. 107. Mawlawi O, Pan T, Cody DD, et al. Evaluation of a new CT truncation cor rection algorithm for accurate quantif ication of PET/CT images [abstract]. J Nucl Med 2004;45:413P. 108. Zaidi H. Is radionuclide transmission scanning obsolete for dualmodality PET/CT systems? Eur J Nucl Med Mol Imaging 2007; 34:815–8. 109. Nakamoto Y, Osman M, Cohade C, et al. PET/CT : comparison of quantitative tracer uptak e between germanium and CT transmission attenuation-corrected images. J Nucl Med 2002;43:1137–43. 110. van Dalen JA, Visser EP, Vogel WV, et al. Impact of Ge-68/Ga-68-based versus CT-based attenuation cor rection on PET . Med Ph ys 2007; 34:889–97. 111. Brix G, Be yer T. PET/CT : dose-escalated image fusion? Nuklearmedizin 2005;44 Suppl 1:S51–7. 112. ICRP 87. Managing patient dose. In: Valentin J, editor. Computed tomography. Oxford, Ne w York: Pub lished for the Inter national Commission on Radiological Protection by Pergamon; 2000. 113. Brix G, Lechel U , Glatting G, et al. Radiation e xposure of patients undergoing whole-body dual-modality 18F-FDG PET/CT e xaminations. J Nucl Med 2005;46:608–13. 114. Kalender WA, Wolf H, Suess C, et al. Dose reduction in CT b y on-line tube cur rent control: principles and v alidation on phantoms and cadavers. Eur Radiol 1999;9:323–8. 115. Kachelriess M, Watzke O , Kalender WA. Generalized multidimensional adaptive f iltering for con ventional and spiral singleslice, multi-slice, and cone-beam CT. Med Phys 2001;28:475–90. 116. ICRP 53. Radiation dose to patients from radiophar maceuticals: addendum 2 to ICRP publication 53. Oxford, New York: Published for the Inter national Commission on Radiolo gical Protection b y Pergamon; 1999. 117. Deloar HM, Fujiwara T, Shidahara M, et al. Estimation of absorbed dose for 2-[F-18]fluoro-2-deoxy-D-glucose using whole-body positron emission tomo graphy and magnetic resonance imaging. Eur J Nucl Med 1998;25:565–74. 118. Hays MT, Watson EE, Thomas SR, Stabin M. MIRD dose estimate report no. 19: radiation absorbed dose estimates from 18F-FDG. J Nucl Med 2002;43:210–4. 119. Czernin J, Auerbach MA. Clinical PET/CT imaging: promises and misconceptions. Nuklearmedizin 2005;44 Suppl 1:S18–23.

27

120. Israel O , K uten A. Earl y detection of cancer recur rence: 18 F-FDG PET/CT can mak e a difference in diagnosis and patient care. J Nucl Med 2007;48 Suppl 1:S28–35. 121. Chen YK, Su CT , Ding HJ , et al. Clinical usefulness of fused PET/CT compared with PET alone or CT alone in nasophar yngeal carcinoma patients. Anticancer Res 2006;26:1471–7. 122. Palmedo H, Bucerius J, Joe A, et al. Integrated PET/CT in differentiated thyroid cancer: diagnostic accurac y and impact on patient management. J Nucl Med 2006;47:616–24. 123. Osman MM, Cohade C, Leal J, Wahl RL. Direct comparison of FDGPET and PET-CT imaging in staging and re-staging patients with lung cancer. J Nucl Med 2002;43:151P. 124. Lardinois D, Weder W, Hany TF, et al. Staging of non-small-cell lung cancer with inte grated positron-emission tomo graphy and computed tomography. N Engl J Med 2003;348:2500–7. 125. Cerfolio RJ, Bryant AS, Ojha B. Restaging patients with N2 (stage IIIa) non-small cell lung cancer after neoadjuvant chemoradiotherapy: a prospective study. J Thorac Cardiovasc Surg 2006;131:1229–35. 126. Fueger BJ , Weber WA, Quon A, et al. P erformance of 2-deoxy-2-[F-18]fluoro-D-glucose positron emission tomography and inte grated PET/CT in restaged breast cancer patients. Mol Imaging Biol 2005;7:369–76. 127. Tatsumi M, Cohade C, Mourtzikos KA, et al. Initial experience with FDG-PET/CT in the evaluation of breast cancer. Eur J Nucl Med Mol Imaging 2006;33:254–62. 128. Bar-Shalom R, Guralnik L, Tsalic M, et al. The additional v alue of PET/CT over PET in FDG imaging of oesophageal cancer . Eur J Nucl Med Mol Imaging 2005;32:918–24. 129. Guo H, Zhu H, XiY, et al. Diagnostic and prognostic value of 18F-FDG PET/CT for patients with suspected recurrence from squamous cell carcinoma of the esophagus. J Nucl Med 2007;48:1251–8. 130. Cohade C, Osman M, Leal J , Wahl RL. Direct comparison of 18 F-FDG PET and PET/CT in patients with colorectal carcinoma. J Nucl Med 2003;44:1797–803. 131. Reinhardt MJ, Joe AY, Jaeger U, et al. Diagnostic perfor mance of whole body dual modality 18F-FDG PET/CT imaging for N- and M-staging of malignant melanoma: experience with 250 consecutive patients. J Clin Oncol 2006;24:1178–87. 132. Freudenberg LS, Antoch G, Schutt P, et al. FDG-PET/CT in re-staging of patients with l ymphoma. Eur J Nucl Med Mol Imaging 2004;31:325–9. 133. Yi CA, Lee KS, Kim BT , et al. Tissue characterization of solitar y pulmonary nodule: comparati ve study betw een helical dynamic CT and integrated PET/CT. J Nucl Med 2006;47:443–50. 134. Kim SK, Allen-Auerbach M, Goldin J, et al. Accuracy of PET/CT in characterization of solitar y pulmonar y lesions. J Nucl Med 2007;48:214–20. 135. Alavi A, Mavi A, Basu S, F ischman A. Is PET-CT the only option? Eur J Nucl Med Mol Imaging 2007;34:819–21. 136. Yu JN, Fahey FH, Gage HD, et al. Intermodality, retrospective image registration in the thorax. J Nucl Med 1995;36:2333–8. 137. Mutic S, Dempsey JF, Bosch WR, et al. Multimodality image registration quality assurance for confor mal three-dimensional treatment planning. Int J Radiat Oncol Biol Ph ys 2001;51:255–60. 138. Bradley JD , Dehdashti F , Mintun MA, et al. P ositron emission tomography in limited-stage small-cell lung cancer: a prospective study. J Clin Oncol 2004;22:3248–54. 139. Schmucking M, Baum RP, Griesinger F, et al. Molecular whole-body cancer staging using positron emission tomography: consequences for therapeutic management and metabolic radiation treatment planning. Recent Results Cancer Res 2003;162:195–202. 140. Ciernik IF, Dizendorf E, Baumert BG, et al. Radiation treatment planning with an inte grated positron emission and computer tomography (PET/CT): a feasibility study. Int J Radiat Oncol Biol Phys 2003;57:853–63. 141. Esthappan J, Mutic S, Malyapa RS, et al. Treatment planning guidelines re garding the use of CT/PET -guided IMR T for cer vical

28

142.

143. 144.

145.

146.

147.

148.

149.

150.

151.

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

carcinoma with positi ve paraaor tic l ymph nodes. Int J Radiat Oncol Biol Phys 2004;58:1289–97. Yap JT, Townsend D W, Hall NC. PET -CT in IMR T planning. In: Mundt AJ, Roeske J, editors. Intensity modulated radiation therapy: a clinical perspective. Amsterdam, The Netherlands; Oxford, UK: Elsevier; 2004. p. 115–30. Weber WA, F iglin R. Monitoring cancer treatment with PET/CT : does it make a difference? J Nucl Med 2007;48 Suppl 1:S36–44. Wieder HA, Beer AJ, Lordick F , et al. Comparison of changes in tumor metabolic activity and tumor size during chemotherap y of adenocarcinomas of the esophago gastric junction. J Nucl Med 2005;46:2029–34. Pottgen C, Le vegrun S, Theegarten D, et al. Value of 18F-fluoro-2deoxy-D-glucose-positron emission tomography/computed tomography in non-small-cell lung cancer for prediction of patholo gic response and times to relapse after neoadjuvant chemoradiotherapy. Clin Cancer Res 2006;12:97–106. Nahmias C, Hanna WT, Wahl LM, et al. Time course of early response to chemotherap y in non-small cell lung cancer patients with 18F-FDG PET/CT. J Nucl Med 2007;48:744–51. de Geus-Oei LF, van der Heijden HF, Visser EP, et al. Chemotherapy response evaluation with 18F-FDG PET in patients with non-small cell lung cancer. J Nucl Med 2007;48:1592–8. Erdi YE, Rosenzweig K, Erdi AK, et al. Radiotherapy treatment planning for patients with non-small cell lung cancer using positron emission tomography (PET). Radiother Oncol 2002;62:51–60. Kong FM, F rey KA, Quint LE, et al. A pilot study of [ 18F]fluorodeoxyglucose positron emission tomo graphy scans during and after radiation-based therapy in patients with non small-cell lung cancer. J Clin Oncol 2007;25:3116–23. Weber WA, Wieder H. Monitoring chemotherap y and radiotherap y of solid tumors. Eur J Nucl Med Mol Imaging 2006;33 Suppl 1:27–37. Therasse P, Arbuck SG, Eisenhauer EA, et al. Ne w guidelines to evaluate the response to treatment in solid tumors. European

152.

153. 154.

155.

156. 157.

158. 159.

160.

Organization for Research and Treatment of Cancer , National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000;92:205–16. Larson SM, Erdi Y, Akhurst T, et al. Tumor treatment response based on visual and quantitati ve changes in global tumor gl ycolysis using PET -FDG imaging. The visual response score and the change in total lesion glycolysis. Clin Positron Imaging 1999;2:159–71. Soret M, Bacharach SL, Buvat I. Partial-volume effect in PET tumor imaging. J Nucl Med 2007;48:932–45. Namdar M, Han y TF, K oepfli P, et al. Inte grated PET/CT for the assessment of coronar y artery disease: a feasibility study . J Nucl Med 2005;46:930–5. Gould KL, P an T, Lo ghin C, et al. F requent diagnostic er rors in cardiac PET/CT due to misregistration of CT attenuation and emission PET images: a def initive anal ysis of causes, consequences, and corrections. J Nucl Med 2007;48:1112–21. DiCarli M, Lipton M. Cardiac PET and PET/CT imaging. New York; London, UK: Springer; 2007. Schmand M, Burbar Z, Corbeil J , et al. Brain PET : f irst human tomograph for simultaneous (functional) PET and MR imaging. J Nucl Med 2007;48:45P. Schlemmer H, Pichler PJ, Wienhard K, et al. Simultaneous MR/PET for brain imaging: f irst patient scans. J Nucl Med 2007;48:45P. Zaidi H, Ma wlawi O, Or ton CG. P oint/counterpoint. Simultaneous PET/MR will replace PET/CT as the molecular multimodality imaging platform of choice. Med Phys 2007;34:1525–8. Coleman RE, Delbeke D, Guiberteau MJ, et al. Concurrent PET/CT with an integrated imaging system: intersociety dialogue from the joint working g roup of the American Colle ge of Radiolo gy, the Society of Nuclear Medicine, and the Society of Computed Body Tomography and Magnetic Resonance. J Nucl Med 2005; 46:1225–39.

3 PET/MRI MARCUS D. SEEMANN, MD

Whole-body imaging is cur rently a topic of g reat interest within the scientif ic community. For tumor staging, diagnostic imaging of the w hole body allo ws simultaneous evaluation of both the region of the primary tumor and the presence of metastases. This evaluation can be perfor med by using se veral dif ferent approaches that include molecular/biochemical imaging techniques such as planar scintigraphy, single photon emission computed tomo graphy (SPECT), positron emission tomo graphy (PET), or anatomic/morphologic imaging techniques such as conventional radio graphy, computed tomo graphy (CT), and magnetic resonance imaging (MRI). (Please also see Chapter 2 “Imaging of Str ucture and Function with PET/CT” and Chapter 4 “SPECT and SPECT/CT” for further discussion). Each of these imaging modalities has specific advantages as w ell as disadv antages in ter ms of, for example, sensitivity, specif icity, accuracy, radiation e xposure, costs, and image acquisition time. The fusion of molecular/biochemical with anatomic/mor phologic information can compensate for many disadvantages and therefore of fers se veral adv antages in comparison to using molecular/biochemical or anatomic/mor phologic diagnostic imaging techniques alone. In the scientif ic literature, the fusion of molecular/biochemical and anatomic/ morphologic information has been shown to improve diagnostic accurac y in identifying and characterizing malignancies, impro ve assessment of tumor stage, therapeutic response and tumor recur rence as compared to visual correlation of the images, allow discrimination of variable physiologic radiotracer uptak e (brain, th yroid gland , f at, striated muscle, myocardium, digestive tract, bone marrow, and genitourinary tract) that can mimic tumor or metastatic lesions from patholo gical uptak e, and help to a void potential f alse-positive inter pretations.1–6 Awareness of the hazards of radiation e xposure has prompted se veral investigators to focus on techniques that enab le w holebody scanning with lo w or no radiation e xposure. The

combination of the tw o e xcellent diagnostic imaging modalities, PET and MRI, into a single scanner of fers several advantages in comparison to using PET and MRI alone and impro ves diagnostic accurac y b y f acilitating the accurate cor relation and e valuation of molecular aspects and biochemical alterations of disease with e xact correlation to anatomic information and morphologic findings (Figure 1). Therefore, the expected diagnostic imaging value of a h ybrid w hole-body PET/MRI scanner is very high.

PET PET imaging relies primaril y on radiotracer uptak e changes for disease detection. PET f acilitates the assessment of molecular aspects and biochemical alterations of a wide v ariety of diseases that are fundamental in the detection of malignancies, evaluation of tumor and tumor stage, and assessment of therapeutic response and tumor recurrence.7 (Please also see Chapter 19 “Chemistr y of Molecular Imaging: An Overview,” Chapter 20 “Radiochemistry of PETm” and Chapter 38 “Ov erview of Molecular and Cell Biolo gy” for fur ther discussion). In general, the accelerated radiotracer acti vity occurs, and therefore can be seen, before anatomic/mor phologic structure changes can be depicted. The main advantage of PET is its high sensiti vity in identifying areas of cancerous in volvement at an earl y stage and distinguishing malignant lesions from benign lesions in most cases. Therefore, PET has become an accepted and v aluable diagnostic imaging tool for patients with cancer . By reducing the probability of overlooking involved areas, it influences the initial staging, restaging after chemotherapy or radiation treatment, and overall management of the disease. The main dif ficulty with PET is the lack of an anatomic/morphologic reference frame. The actual PET scanner allows a total scan range of up to 1,981 mm and 29

30

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A A

B B

C C

D D

Figure 1. 63-year-old woman with cerebral, pulmonary, lymphogenic, osseous, and soft-tissue metastases of a malignant melanoma. Corresponding coronal whole-body 18F-labeled 2-fluoro-2-deoxy-D-glucose positron emission tomography (18F-FDG PET) image (CT-based attenuation correction) (A), whole-body T2-weighted turbo-short tau inversion-recovery (STIR) image (B), whole-body T2-weigthed half-fourier acquired single-shot turbo spin echo (HASTE) image (C), and whole-body T1-weighted Turbo Spin Echo (TSE) image (D).

therefore complete head-to-toe co verage.8 To eva luate molecular aspects and biochemical alterations, PET imaging can be perfor med with dif ferent radiotracers administered sequentially. The most commonl y used radiophar maceutical is a glucose analo g, 18F-labeled 2-fluoro-2-deo xy-D-glucose (18F-FDG). It relies on the detection of an increased rate of aerobic glycolysis. Since the American Food and Drug Administration (FD A) appro ved 18F-FDG in 1997 as a safe and ef fective radiophar maceutical for oncolo gic applications, and since the U .S. Health Care F inancing Administration (HCFA) authorized Medicare in 1998 to reimburse for 18F-FDG PET imaging for cer tain indications, 18F-FDG PET imaging has become an accepted and valuable sensiti ve diagnostic imaging technique for patients with cancer , because it can be used to spot areas of malignanc y and tumor g rowth and to assess tumor stage, therapeutic response, and tumor recur rence. (Please also see Chapter 76 “Re gulatory and Reimbursement Process for Imaging Agents and Devices” for further discussion). The cellular uptake of 18F-FDG molecules is a function of biochemical cell activity and associated with

increased cell tur nover. In most cancers, malignant cells are associated with increased biochemical activity. Therefore, increased uptake of 18F-FDG molecules can be used to spot areas of malignancy and tumor growth. In general, this accelerated biochemical acti vity occurs before anatomic/morphologic str ucture changes. Other imaging modalities, such as CT and MRI rel y primaril y on anatomic/morphologic str ucture changes for disease detection. Whole-body PET imaging with the radiolabeled glucose analo g 18F-FDG has gained widespread acceptance for the staging and restaging of cancer . Furthermore, 18F-FDG PET is a po werful tool for predicting chemotherapy response in cancer more accuratel y than conventional imaging methods. F or e xample, in locall y advanced breast cancer and esophageal cancer, it could be proven that the mean standardized uptak e values (SUV) decreased signif icantly e ven in the second w eek after beginning the f irst c ycle of primar y chemotherap y.9–11 The quantitative 18F-FDG measurement of the mean SUV of the malignancies sho wed a signif icant decrease of radiotracer uptake of 30%. The visual simulated 18F-FDG increase in metastases could be caused b y initial ef fects

PET/MRI

such as neo vascularization and inflammator y cell infiltration within the tumor boundaries and the sur rounding tissue. Thus, it is e xpected that 18F-FDG PET will be useful in reducing the costs of c ytotoxic therapy and the unnecessary side-ef fects of inef fective chemotherap y (Please also see Chapter 52 “PET Diagnosis and Response Monitoring in Oncology” and Chapter 72 “Quantif ication of Radiotracer Uptake into Tissue” for further discussion). 18 F-FDG PET has also been used for research pur poses, because it of fers molecular/biochemical images noninvasively, quantitatively, and repeatedly, not just in humans but also in small animals, using speciall y designed high-resolution small-animal scanning equipment. 12–14 (Please also see Chapter 6 “Small Animal SPECT , SPECT/CT and SPECT/MRI” and Chapter 7 “Instr umentation and Methods to Combine Small Animal PET with Other Imaging Modalities” for fur ther discussion). Neuroendocrine tumors are slo w-growing rare neoplasms of neuroectodermal origin that frequently express specific somatostatin receptors that can be of g reat value in the staging and treatment 15,16 of these tumors. Therefore, this type of tumor entity is imaged b y w ay of somatostatin receptors, and not with the radiolabeled glucose analog 18F-FDG. SPECT using radiolabeled analogs of somatostatin has become the standard in vivo imaging modality of choice in the identif ication and staging of neuroendocrine tumors and thus the most reliable tool for guiding therapy.17–21 Primary endocrine tumors are small neoplasms that arise predominantly in the gastrointestinal tract, pancreas, and lung. The most common metastatic sites are the li ver, l ymph nodes, bone, lung, and peritoneal ca vity. Numerous studies ha ve conf irmed higher sensitivity, specif icity, diagnostic accurac y, and positi ve predictive v alue of somatostatin receptor scintig raphy (SRS) (90%, 80%, 83%, and 100%, respecti vely) compared with mor phological imaging procedures (w hich have a sensitivity of between 50 and 70%, depending on the size and location of the lesion). 17–22 The sensitivity of SRS depends on the high v ariability in the e xpression of somatostatin receptors among the different tumor entities and even within dif ferent lesions in a single patient, 16,23 the limitations of the physical properties of SPECT imaging, and f inally the phar macokinetic characteristics of Indium-111 DTP A-octreotide as a SPECT tracer with high physiological uptake in li ver and spleen, and elimination by the kidne y and also at a lo w percentage b y the hepatobiliary system. (Please also see Chapter 21 “Radiochemistry of SPECT: Examples of 99mTC and 111IN Complexes” for further discussion). A promising somatostatin receptor ligand for PET imaging is [ 18F]FP-Gluc-TOCA (Nα-(1-deoxy-D-fructosyl)-Nε-(2-[18F]fluoropropionyl)Lys0-Tyr3-octreotate). [ 18F]FP-Gluc-TOCA showed a v ery

31

high affinity to human somatostatin receptor subtype (hsst) 2, a moderate affinity to hsst 5, a low affinity to hsst 4, and no affinity to hsst 1 and 3. Its lo w lipophilicity, low liver uptake, rapid renal elimination, and lo w intestinal activity, as well as its f ast and high tumor accumulation, pro vide excellent tumor-to-background ratios.24,25 In medical research, 18F-labeled 1- α-D-(5-fluoro5-deoxyarabinofuranosyl)-2-nitroimidazole ( 18F-fluoroazomycin arabinoside; 18F-FAZA) is a promising radioactive tracer for patients with cancer because it can be used nonin vasively in the detection of tumor hypoxia in humans, as w ell as in small animals. 26 As tumor h ypoxia has a major ne gative predicti ve v alue for local tumor pro gression, lik eliness of metastasis, and overall tumor prognosis in several types of human cancer, the presence of tumor tissue hypoxia is relevant in predicting pro gnosis and response to cur rent radiation treatment. 27 In addition, tumor cell h ypoxia has a negative ef fect on anticancer treatment, gi ven that hypoxic cells are 2 to 3 times more resistant to a single fraction of ionizing radiation than those with n ormal oxygenation le vels.28 (Please also see Chapter 4 6 “Hypoxia Imaging” for fur ther discussion).

MRI MRI is an e xcellent anatomic/mor phologic imaging modality with a high anatomic resolution. Ov er the past decade, major impro vements in MRI ha ve occur red. Improvements in the hardw are of MRI machines ha ve facilitated the de velopment of f ast and ultraf ast pulse sequences as well as the development of phased-array multicoils allo wing for higher signal-to-noise ratio and thus higher anatomic resolution. In addition, introduction of navigator techniques has fur ther enhanced signal-to-noise ratio and spatial resolution allo wing for higher imaging matrices, resulting in improved diagnostic perfor mance in the detection of liver lesions.29 However, image blurring is one potential drawback of this technique if the navigator is incorrectly placed or if the patient has an ir regular breathing patter n. Then, the na vigator technique should be replaced with f ast or ultraf ast breath-hold techniques to avoid respirator y or motion-induced ar tifacts.30 However, breath-hold techniques result in poorer signal-to-noise and limited spatial resolution. In the past, MRI w as used only as a tool to image specif ic regions of the body, due to the prolonged imaging time, limited a vailability of scanning facilities, and extensive costs. The concept that MRI might become the ultimate whole-body imaging tool was initially proposed b y the MRI pioneers Damadian and Lauterbur.31,32 The de velopment of f ast and ultraf ast MRI techniques led to the possibility of rapid w hole-body

32

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

scanning. However, most in vestigators used the body coil for multistation data reception. 33,34 Thus, images suf fered from either low signal-to-noise ratio or poor spatial resolution.35–38 To overcome these limitations, other investigators examined patients on a sliding table platform with an integrated phased-array surface coil.39 Although this approach improves signal-to-noise, the distal limbs w ere not included.40 In addition, the adv antages of parallel acquisition techniques to achie ve higher spatial resolution or shorter acquisition times were not used. The new development of Total Imaging Matrix (TIM) allo ws, for the f irst time, perfor mance of high-resolution w hole-body co verage from head to toe within a single e xamination without the need for patient or surf ace coil repositioning, yielding excellent high-resolution image quality . The basic idea of TIM is the re volutionary matrix coil concept that allows the combination of 76 coil elements with up to 32 channels, a combination that enab les considerab le improvement in both acquisition speed and image quality . The actual MR scanner allo ws a total scan range of up to 2,050 mm and therefore complete head-to-toe co verage. Whole-body MRI is a promising diagnostic modality used for both diagnosis and management in man y diseases, and not onl y for patients with cancer . Ho wever, the optimal imaging protocol and patient preparation for the w holebody imaging ha ve still y et to be e valuated. P erforming MRI without the use of an intra venous paramagnetic contrast medium ma y include the possibility of o verlooking some hypervascular metastases that can only be seen in the hepatic ar terial-dominant contrast-enhanced phase. 41–43 However, Dromain and colleagues 42 found no signif icant difference betw een the hepatic ar terial-dominant contrast-enhanced T 1-weighted images and the f ast spin-echo T 2-weighted images in the detection of hepatic metastases from neuroendocrine tumors. The f ast spin-echo T 2-weighted sequence sho wed a signif icantly higher lesion-to-liver contrast-to-noise ratio than the hepatic ar terial-dominant contrast-enhanced T1-weighted sequence. For mor phologic imaging methods, the w eakness of l ymph node staging is the kno wn lack of reliab le criteria since assessment can be made onl y on the basis of size.44 Lymph nodes can onl y be staged as l ymph node metastases on the basis of mor phologic criteria lik e a nodular shape and a diameter of more than 10 mm (reference standard). Fur thermore, the dif ferentiation of abdominal lymph nodes and small intestine, without using oral contrast media, can be a prob lem. First of all, coronal sequences like T2-weighted turbo-short tau inversion-recovery (STIR), T1-weighted turbo spin echo (TSE) sequence, and T 2-weigthed half-F ourier acquired single-shot turbo spin echo (HASTE) sequence to image the w hole body,

can be perfor med. To a void respirator y motion-induced artifacts, the Turbo-STIR sequence as w ell as the T1-weighted TSE sequence in the chest and abdomen should be perfor med with breath holds. If necessar y, an additional T 1-weighted f ast lo w-angle shot (FLASH) sequence can be used to conf irm the presence of sk eletal metastases. The detection of osseous metastases in the ribs can be a problem. In addition, axial T2-weighted highresolution TSE sequences using prospecti ve acquisition with na vigator technique (respirator y gating to track diaphragmatic and cardiac mo vements) can be obtained from the chest, abdomen, and pelvis. This can be followed by high-resolution axial cross-sectional sequences to focus on the detected patholo gy, and f acultative techniques such as functional imaging, dif fusion and perfusion imaging, spectroscopy, and angiograpy. Finally, MRI is emer ging as a par ticularly adv antageous modality for molecular/ biochemical imaging. (Please also see Chapter 26 “MR Imaging Agents,” Chapter 34 “Magnetic Nanopar ticles,” Chapter 36 “ Aptamers for Molecular Imaging, ” and Chapter 44 “Cell Voyeurism using Magnetic Resonance Imaging” for fur ther discussion). P ossibly in conjunction with rational tar geted therapies, this technique could radically affect the practice of clinical diagnosis and therapy as these technologies continue to mature.45 MR imaging relies primarily on anatomic/mor phologic str ucture changes for disease and tumor detection. After chemotherap y, anatomic/morphologic imaging procedures allow the detection of changes in tumor size and v olume. Reduction of tumor volume as evidence of response to therapy requires a certain time dela y after initiation of therap y, and ma y be masked b y unspecif ic ef fects (e g, edema as a result of necrosis). Whole-body MRI protocols have to be optimized and adapted to allo w for better dif ferentiation of the primary tumor and possib le metastases, especiall y in the abdomen. In general, w hole-body MRI should be performed with intra venous contrast media and in oncolo gic patients in combination with oral contrast media.

PET/MRI Over the past decade, there ha ve been g reat technical improvements in PET and MRI. The combination of these two excellent diagnostic imaging modalities into a single scanner of fers se veral adv antages and impro ves diagnostic accuracy by facilitating the accurate re gistration of molecular aspects and biochemical alterations of disease with e xact cor relation to anatomic infor mation and morphologic findings. PET facilitates the evaluation of molecular aspects and biochemical alterations that are fundamental to the detection of malignanc y and

PET/MRI

assessment of tumor stage, therapeutic response, and tumor recurrence. In general, increased radiotracer accumulation can be seen before anatomic/morphologic structure changes. The primary difficulty with PET is the lack of an anatomic reference frame. The combination of PET with an anatomic/morphologic imaging modality such as CT or MRI can compensate for this disadv antage and offers se veral adv antages in comparison to using PET , CT, or MRI alone. Whole-body MRI produces lar ge amounts of image data, resulting in the possibility of overlooking subtle patholo gic f indings. Fur thermore, anatomic/morphologic imaging procedures do not allo w differentiation betw een viab le tumor tissue and f ibrotic scar tissue. 18F-FDG PET can dif ferentiate viable tumor tissue from atelectases and scars and is therefore helpful in planning radiotherap y of lung carcinoma. The posttherapeutic 18F-FDG PET/CT study can sho w, qualitatively and quantitati vely, decreased acti vity and v olume of all tumor lesions. The cellular uptake of the 18F-labeled glucose analog 2-fluoro-2-deoxy-D-glucose ( 18F-FDG) is a sensiti ve and v aluable mark er for biochemical alterations of cancer cells that is fundamental not onl y in the detection of a wide v ariety of malignancies, but also for prediction of neoadjuv ant chemotherap y response, and is more accurate than anatomic/morphologic imaging methods. A fur ther adv antage of PET is the sensiti ve detection of lymph node metastases of a size smaller than 10 mm that are not unequi vocally assessab le with anatomic/morphologic imaging and can be identif ied by PET, leading to an improvement of this morphologic reference standard. The w eakness of mor phologic l ymph node staging is the kno wn lack of reliab le criteria, since assessment can be made only on the basis of size.44 In the scientific literature, the simultaneous acquisition of coregistered molecular/biochemical and anatomic/mor phologic information has been sho wn to improve diagnostic accuracy of the cancer staging as compared to visual correlation of the images, allo ws the discrimination of v ariable physiologic radiotracer uptake (brain, thyroid gland, fat, striated muscle, m yocardium, digesti ve tract, bone marrow, and genitourinary tract) that can mimic metastatic lesions from patholo gical uptake, and helps to a void potential false-positive inter pretations.1–4,6 The fusion of PET with MRI can compensate for their separate disadvantages, and therefore of fers se veral combined adv antages in comparison to using PET or MRI alone. The expected diagnostic value of the combination of PET and MRI into a single w hole-body hybrid PET/MRI scanner is v ery high. In the case w here PET and MRI are not fused via a single scanner , the repositioning of the patient and time inter val betw een the scans mak es the

33

co-registration and fusion of separatel y obtained images difficult and inherentl y imprecise. 2 Whole-body PET/MRI is a v ery promising diagnostic modality for oncologic imaging and cancer screening in the decades to come, due to the considerably lower radiation exposure in contrast to PET/CT, and the high soft-tissue resolution of MRI (Figure 2). In the literature, 18F-FDG PET/CT and MRI in patients with dif ferent malignant diseases w ere compared. 46 The authors suggest the use of 18F-FDG PET/CT as a possib le first-line modality for whole-body tumor staging. However, they did not perform an image fusion of PET and MRI. The diagnostic value of whole-body imaging modalities PET, CT, MRI, and the image fusion of PET and CT (PET/CT) and PET and MRI (PET/MRI) in the detection of metastases of gastrointestinal neuroendocrine tumors was e valuated in a prospecti ve study .47 PET data w as acquired with a state-of-the-ar t high-count-rate lutetium oxyorthosilicate (LSO) detector Pico-3D full-ring PET scanner of a hybrid PET/CT from the base of the skull to the proximal thigh. As a radiopharmaceutical, a carbohydrate deri vatized F-18-labeled somatostatin-receptor ligand ([18F]FP-Gluc-TOCA = Nα-(1-deoxy-D-fructosyl)Nε-(2-[18F]fluoropropionyl)-Lys0-Tyr3-octreotate) w as used. CT data was acquired with the 16-slice CT scanner of the hybrid PET/CT using a venous-dominant contrastenhanced phase. F or an optimal assessment of the gastrointestinal tract, oral administration of diluted diatrizoate meglumine was perfor med beginning 1 hour before star ting the e xamination. MRI data w as acquired with a 1.5 Tesla w hole-body MRI scanner using the TIM technology.48 MRI w as perfor med with a coronal T2-weighted HASTE sequence, a coronal T2-weighted STIR sequence, a coronal T1-weighted TSE sequence, and a high-resolution axial T2-weighted TSE sequence with na vigator technique. F or the detection of li ver metastases, PET/MRI (100%) and MRI (98.2%) w ere most sensitive, whereas PET/CT (50.9%)( p < .001), PET (49.9%)( p < .001), and CT (37.1%)( p < .001) were significantly less reliab le (F igure 3). In this comparati ve study, MRI was the most sensiti ve imaging procedure in the detection of liver metastases. The main reason for this is the spatial resolution in the submillimeter area and the high soft-tissue contrast. MRI sho wed a lot of lesions with a diameter between 2 and 4 mm, which could not be seen in any other modality. PET, CT, and PET/CT underestimated the extension of liver metastases or even missed the metastatic disease, because of the lo w soft-tissue contrast in CT, the low spatial resolution and physiological tracer uptake in the normal liver tissue, and the dependency on the presence, type, and density of the

34

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

C C

D

E

F F

Figure 2. 63-year-old woman with malignant melanoma. The primary tumor in the left lower arm was surgically resected two years ago. Corresponding coronal centered view from head to diaphragm (A–C) showing a 18F-FDG PET image (A), a HASTE image (B), and a manual fused image of PET and MRI (PET/MRI) (C), with a cerebral metastasis and multiple soft-tissue metastases on the left lateral thoracic wall. Corresponding coronal centered view from the inlet of the thorax to inlet of the pelvis (D–F) showing a 18F-FDG PET image (D), a HASTE image (E), and a manual fused image of PET and MRI (PET/MRI) (F) with pulmonary and osseous metastases.

somatostatin receptors expressed by the tumor lesions as well as the tumor size w hen using PET. Metastases with a low somatostatin receptor density and necrotic metastases showed low to no tracer accumulation in PET. Nevertheless, PET seems to be ab le to gi ve additional information to MRI if there are prob lems with the na vigator technique. For the detection of l ymph node metastases, PET/CT (100%), PET/MRI (97.3%), and PET (91.9%) w ere most sensiti ve, w hereas CT (83.8%; p < .54) and MRI (64.9%; p < . 12) w ere considerab ly less reliable (F igures 4 and 5). PET w as the most sensiti ve imaging procedure in the detection of lymph node metastases. Six l ymph nodes (16.2%) (cer vical, n = 1; mediastinal, n = 2; retroperitoneal, n = 2; and intraperitoneal, n = 1) sho wed a clearl y increased tracer uptak e in PET but were smaller than the reference standard (10 mm) in CT and MRI. Three lymph nodes ( n = 8.1%) (retroperitoneal, n = 1 and intraperitoneal, n = 2) sho wed a size from 11 to 14 mm in CT but no increased tracer uptake in PET. The main adv antage of l ymph node staging with PET is the sensiti ve detection of serotonin-expressing small lymph nodes (< 10 mm), w hich are not assessab le

with morphologic imaging and can be identif ied by PET, leading to an impro vement of the sensiti vity. The detection rate of 100% for l ymph node staging in the combination of PET and anatomic/mor phologic imaging is not a realistic v alue, because it is possib le that indi vidual lymph node metastases sho wed no octreotate accumulation and also were not enlarged. The assessment of lymph node staging is dif ficult and unsatisf actory using morphologic imaging procedures onl y. The weakness of morphologic lymph node staging is due to theknown lack of reliable criteria, since assessment can be made only on the basis of size. 44,47 On the one hand, some lymph node metastases with a size smaller than 10 mm may not be detected, and lymph nodes with a size of g reater than 10 mm can also be caused b y inflammation. Using MRI, abdominal lymph node metastases w ere difficult to differentiate from small and lar ge intestine. F or the detection of osseous metastases, PET (100%), PET/CT (100%), and PET/MRI (100%) w ere m ost sensitive, whereas MRI (66.7%; p < .12) and CT (8.3%; p < .003) were less reliable (Figure 6). PET was the most sensitive imaging procedure in the detection of osseous

PET/MRI

A

B

C

D

E

F

G

H

35

Figure 3. 53-year-old male patient with multiple hepatic metastases of an intestinal neuroendocrine tumor. Corresponding axial Nα-(1deoxy-D-fructosyl)-Nε-(2-[18F]fluoropropionyl)-Lys0-Tyr3-octreotate ([18F]FP-Gluc-TOCA) PET image (A, H). T2-weighted STIR image (B), T2-weighted TSE image (C), HASTE image (D), T1-weighted TSE image (E), arterial-dominant contrast-enhanced T1-weighted TSE image (F), and venous-dominant contrast-enhanced T1-weighted TSE image (G). All MR images are shown with the manual fused image of PET (PET/MRI).

metastases. The metastases seen on CT w ere osteosclerotic. The differentiation of de generative changes of the vertebral column and osteosclerotic bone metastases can be a problem in anatomic/morphologic imaging methods. The missing osseous metastases in the MRI w ere localized in the ribs. The results from this comparati ve study suggested that only the combined use of molecular/ biochemical and anatomic/morphologic imaging procedures achie ved a cor rect tumor classif ication, and the combination of PET and MRI into a single scanner could be a v ery v aluable w hole-body diagnostic imaging tool

not onl y for endocrine tumors but also for oncolo gic tumor staging. 47 Schillaci and colleagues 20 compared the detection of abdominal metastases in SRS SPECT and anatomic/morphologic imaging procedures (ultrasound , CT, and/or MRI) in neuroendocrine tumors and suggested also that onl y the combined use of molecular/biochemical and anatomic/mor phologic imaging procedures achieved a correct classification. Planar scintigraphy, SPECT, and PET are the onl y clinically a vailable nonin vasive imaging techniques that can assess molecular aspects and biochemical

36

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

C

Figure 4. 62-year-old male patient with a lymph node metastases of an intestinal neuroendocrine tumor. Corresponding axial view from the inlet of the thorax showing a [18F]FP-Gluc-TOCA PET image (A), a T2-weighted TSE image (B), and the manual fused image of PET and MRI (PET/MRI) (C). PET identified a small lymph node metastases (5 × 6 × 8 mm).

A

B

C C

Figure 5. 62-year-old male patient with lymph node metastases of an intestinal neuroendocrine tumor. Corresponding axial view from the pelvis showing a [18F]FP-Gluc-TOCA PET image (A), a T1-weighted TSE image (B), and the manual fused image of PET and MRI (PET/MRI) (C).

A

B

C

Figure 6. 62-year-old male patient with an osseous metastasis of an intestinal neuroendocrine tumor. Corresponding axial view from the thorax showing a [18F]FP-Gluc-TOCA PET image (A), T2-weighted STIR image (B), and the manual fused image of PET and MRI (PET/MRI) (C).

alterations that are fundamental to cancer detection, cancer recur rence, and e valuation of therapeutic response using small amounts of radioacti ve labeled molecules in vi vo.49 PET techniques can assess the in vivo biodistribution of man y relevant radiopharmaceuticals and thus contribute significantly and distinctively

to the e valuation of tumors. 50 Small-animal imaging has gained increasing attention in recent y ears as an excellent in vivo evaluation method for molecular biology, oncolo gy, and neuroscience research. Smallanimal models of rats and mice are widel y used in biomedical research for mimicking and studying the

PET/MRI

human condition in health or disease, because of their genetic resemb lance to humans and the feasibility of gene transfer and gene modif ication.51,49 Similar to human h ybrid PET/CT , tumor -bearing small animals could be examined with human hybrid PET/MRI using special scanning and reconstr uction protocols, and this capability is also expected to contribute significantly to research with small-animal imaging. The investigation of cancer in small animals with h ybrid PET/MRI is probably one of the most challenging tasks in nuclear medicine since the str uctures of interest are almost in the same range as the maximum spatial resolution. Hybrid PET/MRI is considered to be par ticularly wellsuited for research pur poses, especially for the e valuation of tumor g rowth and g rowth inhibition f actors; development of new anti-tumor drugs and measuring of anti-tumor ef fects; and cancer treatment response to immunotherapy, chemotherap y and radiation therap y. (Please also see Chapter 67 “Molecular and Functional Imaging in Dr ug De velopment,” Chapter 68 “PET Imaging Clinical Trials,” and Chapter 69 “MR Imaging in Clinical Trials” for fur ther discussion). Cur rent interesting topics include implantation of human tumor cells in small animals and e valuation of tumor g rowth and growth inhibition factors; development of new antitumor dr ugs and measuring of anti-tumor ef fects; and cancer treatment response to immunotherap y, chemotherapy, and radiation therap y using molecular/ biochemical and anatomic/mor phologic parameters. The infor mation from h ybrid PET/MR imaging w ould be of interest for validation of oncologic research studies for the evaluation of new tumor tracers labeled with different positron emitters (F-18, C-11, N-13, and O-15). Combined systems pro vide the ability to accurately cor relate and e valuate biochemical and molecular aspects of cancers with anatomic infor mation and morphologic findings in human clinical routine examinations and are promising for in vi vo animal research. Hybrid PET/MRI scanners pro vide molecular and biochemical information noninvasively, quantitatively, and repeatedly. Hybrid PET/MRI is a highly valuable oncologic imaging modality with the feasibility of in vi vo imaging of tumor -bearing small animals for potential use in oncolo gy research using special scanning and reconstruction protocols. Due to the small dimensions of the str uctures to be imaged in animals, the smallanimal-specific protocols ha ve made it possib le for tumor h ybrid PET/MR images to be clear enough to resolve considerab le hetero geneity of tracer uptak e within the tumors with the help of thin-slice, high-quality MR images. The e xperience with human h ybrid

37

PET/MRI is e xpected to contribute signif icantly to research with small-animal imaging. 50 Although hybrid PET/CT has proved itself clinically as a highl y v aluable oncolo gic diagnostic modality , it might not be the ultimate diagnostic imaging technique since MRI offers several advantages compared with CT and therefore hybrid PET/CT will be in competition with hybrid PET/MRI. 52,53 In comparison to CT , MRI is not associated with radiation e xposure and has a much higher soft-tissue contrast. This has been sho wn to be advantageous in neuroradiological, musculoskeletal, cardiac, and oncolo gic (e g, detection and characterization of focal li ver lesions) applications. The lower radiation dose is important for pediatric applications and repeated imaging in oncolo gic patients. The higher soft tissue is also interesting in head and neck applications. MRI allows for additional techniques such as angio graphy (eg, tumor v asculature), functional MRI (e g, brain activation studies), dif fusion and perfusion techniques within one single examination, virtual endoscopic examinations (eg, bronchoscopy and colonography), and spectroscopy (eg, prostate cancer). The injection of iodinated contrast agents that are potentiall y nephroto xic is not necessary. Ne vertheless, there are limitations in the detection of li ver metastases due to ir regular breathing, the dif ferentiation of l ymph nodes, and the small and large intestine, and in the detection of osseous metastases. The main advantage of PET is the identification of cancerous lesions be yond the diagnostic sensiti vity of MRI, which allows the diagnosis of malignancies at an earlier stage. PET can be used to spot areas of malignancy, monitor tumor g rowth, predict response to therapy, and monitor therapeutic response and tumor recurrence. Malignancies with low or nor mal biochemical activity (eg, mucinous carcinomas, primary renal cell carcinoma, and prostate cancer) in the PET image component may show clearly positive or suspicious f indings in the MRI component of the h ybrid PET/MRI. MRI demonstrates metastases not found on PET . PET imaging is v ery sensiti ve but sometimes suf fers from specificity. MRI allo ws an e xact localization of the lesions, their differential diagnostic clarification, and the visualization of their patho gnomonic mor phologic appearance. A hybrid PET/MRI scanner can compensate for the disadv antages of using PET and MRI separatel y and therefore yields a clear impro vement of diagnostic accuracy by combining two already excellent modalities. The combination of w hole-body PET and w hole-body MRI into a single scanner of fers the ability for accurate registration of molecular aspects and biochemical alterations of a wide variety of diseases with e xact

38

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

correlation to anatomic/mor phologic f indings w hen using these two excellent modalities in oncolo gic applications. A h ybrid PET/MRI w ould be highl y a dvantageous in clinical practice in impro ving the diagnostic value of PET and MRI in identifying and characterizing malignancies and in tumor staging and assessment of therapeutic response and tumor recurrence. The MR images could be used for the attenuation cor rection, could pro vide high-quality anatomic details and the precise anatomic localization of increased radiotracer uptake of the PET imaging, and could be used for measurements of tumor lesions in distance and v olume. The PET images would be used for quantitative analysis of tracer accumulation in tumor lesions. 18F-FDG PET/CT has been sho wn to be v ery effective for early prediction of neoadjuv ant chemotherap y response in patients with metastatic tumor disease w ho underwent examination for therap y monitoring. Anatomic/morphologic imaging methods could re veal an increased necrosis of the primar y tumor and metastases and a progressive sclerosis of the l ytic osseous metastases that could be misinter preted as tumor pro gression. Anatomic/morphologic imaging procedures rel y on anatomic str ucture changes of disease and allo w the detection of changes in tumor size and v olume, but do not allow a differentiation between viable tumor tissue and f ibrotic scar tissue due to treatment. The morphologic criteria of reduction of tumor volume as evidence of response to therap y requires a cer tain time dela y after initiation of therap y and ma y be mask ed b y unspecific effects (eg, edema as a result of necrosis). Thus, it is expected that 18F-FDG PET will be useful in reducing the costs of c ytotoxic therap y and the unnecessary side ef fects of inef fective chemotherap y. Whole-body h ybrid PET/MRI is a v ery promising diagnostic modality for oncologic imaging and for use in cancer screening due to the considerably lower radiation e xposure in contrast to h ybrid PET/CT and the high soft-tissue resolution of MRI in contrast to CT.5,54 For the best pro gnostic stratif ication and to guide the most appropriate therapeutic approach, the use of the most sensitive imaging procedure to detect metastases and evaluate the number of metastases and their distribution should be recommended. An adequate diagnostic cer tainty for the detection of the e xtent of tumor metastases is not attained with any single imaging procedure. The whole-body hybrid PET/MRI scanner will be the state-of-the-ar t diagnostic imaging modality in oncologic applications in the decades to come and also possibly will be used in cancer screening and cardiovascular imaging.

REFERENCES 1. Blomqvist L, Torkzad MR. Whole-body imaging with MRI or PET/CT: the future for single-modality imaging in oncolo gy? JAMA 2003;290:3248–9. 2. Kluetz PG, Meltzer CC, Villemagne VL, et al. Combined PET/CT imaging in oncolo gy: impact on patient management. Clin Positron Imaging 2000;3:223–30. 3. Lardinois D, Weder W, Hany TF, et al. Staging of non-small-cell lung cancer with inte grated positron-emission tomo graphy and computed tomography. N Engl J Med 2003;348:2500–7. 4. Seemann MD . PET/CT : fundamental principles. Eur J Med Res 2004;9:241–6. 5. Seemann MD. Whole-body PET/MRI: the future in oncological imaging. Technol Cancer Res Treat 2005;4:577–82. 6. Shreve PD, Anzai Y, Wahl RL. Pitf alls in oncolo gic diagnosis with FDG PET imaging: physiologic and benign variants. Radiographics 1999;19:61–77. 7. Seemann MD. Diagnostic v alue of PET/CT for predicting of neoadjuvant chemotherapy response. Eur J Med Res 2007;12:90–1. 8. Seemann MD. Human PET/CT scanners: feasibility for oncological in vivo imaging in mice. Eur J Med Res 2004;9:468–72. 9. Biersack HJ, Bender H, P almedo H. FDG-PET in monitoring therap y of breast cancer . Eur J Nucl Med Mol Imaging 2004;31 Suppl 1:S112–7. 10. Brücher BL, Weber W, Bauer M, et al. Neoadjuv ant therap y of esophageal squamous cell carcinoma: response e valuation b y positron emission tomography. Ann Surg 2001;233:300–9. 11. Scheidhauer K, Walter C, Seemann MD. FDG PET and other imaging modalities in the primar y diagnosis of suspicious breast lesions. Eur J Nucl Med Mol Imaging 2004;31(Suppl 1):S70–9. 12. Chatziioannou AF. Molecular imaging of small animals with dedicated PET tomographs. Eur J Nucl Med 2002;29:98–114. 13. Myers R. The biolo gical application of small animal PET imaging. Nucl Med Biol 2001;28:585–93. 14. Seemann MD , Beck R, Zie gler S. In vi vo tumor imaging in mice using a state-of-the-ar t clinical PET/CT in comparison with a small animal PET and a small animal CT . Technol Cancer Res Treat 2006;5:537–42. 15. Kaltsas G, Rockall A, Papadogias D, et al. Recent advances in radiological und radionuclide imaging and therap y of neuroendocrine tumours. Eur J Endocrinol 2004;151:15–27. 16. Reubi JC, Kvols LK, Waser B, et al. Detection of somatostatin receptors in surgical and percutaneous needle biopsy samples of carcinoids and islet cell carcinomas. Cancer Res 1990;50:569–77. 17. Gibril F, Re ynolds JC, Doppman JL, et al. Somatostatin receptor scintigraphy: its sensiti vity compared with that of other imaging methods in detecting primar y and metastatic gastrinomas. Ann Intern Med 1996;125:26–34. 18. Krenning EP, Kw ekkeboom DJ , Bakk er WH, et al. Somatostatin receptor scintigraphy with [111In-DTPA-D-Phe1]- and [123I-Tyr3]octreotide: the Rotterdam e xperience with more than 1000 patients. Eur J Nucl Med 1993;20:716–31. 19. Lebtahi R, Cadiot G, Sarda L, et al. Clinical impact of somatostatin receptor scintigraphy in the management of patients with neuroendocrine gastroenteropancreatic tumors. J Nucl Med 1997;38:853–8. 20. Schillaci O, Scopinaro F, Angeletti S, et al. SPECT improves accuracy of somatostatin receptor scintig raphy in abdominal carcinoid tumors. J Nucl Med 1996;37:1452–6. 21. Schillaci O, Spanu A, Scopinaro F, et al. Somatostatin receptor scintigraphy in li ver metastasis detection from gastroenteropancreatic neuroendocrine tumors. J Nucl Med 2003;44:359–68. 22. Kwekkeboom D, Krenning EP, De Jong M. Peptide receptor imaging and therapy. J Nucl Med 2000;41:1704–13. 23. Patel YC. Somatostatin and its receptor family. Front Neuroendocrinol 1999;20:157–98.

PET/MRI

24. Meisetschlaeger G, Poethko T, Stahl A, et al. Gluc-Lys ([18F]FP)-TOCA PET in patients with SSTR-positi ve tumors: biodistribution and diagnostic e valuation compared with [111In]DTP A-octreotide. J Nucl Med 2006;47:566–73. 25. Wester H-J , Schottelius M, Scheidhauer K, et al. PET imaging of somatostatin receptors: design, synthesis and preclinicalevaluation of a novel 18F-labelled, carbohydrated analogue of octreotide. Eur J Nucl Med Mol Imaging 2003;30:117–22. 26. Piert M, Machulla HJ , Picchio M, et al. Hypo xia-specific tumor imaging with 18F-fluoroazom ycin arabinoside. J Nucl Med 2005; 46:106–13. 27. Nordsmark M, Hoyer M, Keller J, et al. The relationship between tumor oxygenation and cell proliferation in human soft tissue sarcomas. Int J Radiat Oncol Biol Phys 1996;35:701–8. 28. Harrison LB, Chadha M, Hill RJ , et al. Impact of tumor h ypoxia and anemia on radiation therap y outcomes. Oncolo gist 2002; 7:492–508. 29. Martin DR, Semelka RC. Imaging of benign and malignant focal liver lesions. Magn Reson Imaging Clin N AM 2001;9:785–802. 30. Gaa J, F ischer H, Laub G, Geor gi M. Breath-hold MR imaging of focal li ver lesions: comparison of f ast and ultraf ast techniques. Eur Radiol 1996;6:838–43. 31. Damadian R. F ield focusing n.m.r . (FONAR) and the for mation of chemical images in man. Philos Trans R Soc Lond B Biol Sci 1980;289:489–500. 32. Lauterbur PC. Pro gress in n.m.r . zeugmato graphy imaging. Philos Trans R Soc Lond B Biol Sci 1980;289:483–7. 33. Engelhard K, Hollenbach HP, Wohlfart K, et al. Comparison of wholebody MRI with automatic mo ving tab le technique and bone scintigraphy for screening for bone metastases in patients with breast cancer. Eur Radiol 2004;14:99–105. 34. Kavanagh E, Smith C, Eustace S. Whole-body turbo STIR MR imaging: controversies and avenues for development. Eur Radiol 2003; 13:2196–205. 35. Eustace S, Tello R, DeCarvalho V, et al. A comparison of whole-body turbo-STIR MR imaging and planar 99m Tc-methylene diphosponate scintigraphy in the examination of patients with suspected skeletal metastases. AJR 1997;169:1655–61. 36. Eustace S, Tello R, DeCar vallho V, et al. Whole-body turbo-STIR MRI in unknown primary tumor detection. J Magn Reson Imaging 1998;8:751–3. 37. Johnson KM, Leavitt GD, Kayser HW. Total-body MR imaging in as little as 18 seconds. Radiology 1997;202:252–6. 38. O Connell MJ, Hargaden G, Powell T, Eustace SJ. Whole-body turbo short tau inversion recovery MR imaging using a moving tabletop. AJR 2002;179:866–8. 39. Barkhausen J, Quick HH, Lauenstein T, et al. Whole-body MR imaging in 30 seconds with real-time tr ue FISP and a continuousl y rolling table platform: feasibility study. Radiology 2001;220:252–6.

39

40. Lauenstein TC, Freudenberg LS, Goehde SC, et al. Whole-body MRI using a rolling table platform for the detection of bone metastases. Eur Radiol 2002;12:2091–9. 41. Bader TR, Semelka RC, Chiu VC, et al. MRI of carcinoid tumors: spectrum of appearances in the gastrointestinal tract and li ver. J Magn Reson Imaging 2001;14:261–269. 42. Dromain C, de Baere T, Baudin E, et al. MR Imaging of hepatic metast ases caused b y neuroendocrine tumors: comparing four techniques. AJR 2003;180:121–8. 43. Paulson EK, McDermott VG, Keogan MT, et al. Carcinoid metastases to the li ver: role of triple-phase helical CT . Radiolo gy 1998; 206:143–50. 44. Bollen EC, Goei R, v an’t Hof-Grootenboer BE, et al. Interobser ver variability and accuracy of computed tomo graphic assessment of nodal status in lung cancer. Ann Thorac Surg 1994;58:158–62. 45. Weissleder R, Mahmood U . Molecular imaging. Radiolo gy 2001;219:316–33. 46. Antoch G, Vogt FM, Freudenberg LS, et al. Whole-body dual-modality PET/CT and w hole-body MRI for tumor staging in oncolo gy. JAMA 2003;290:3199–206. 47. Seemann MD, Meisetschlaeger G, Gaa J , Rummeny EJ. Assessment of the e xtent of metastases of gastrointestinal carcinoid tumors using whole-body PET, CT, MRI, PET/CT and PET/MRI. Eur J Med Res 2006;11:58–65. 48. Gaa J , Rummen y EJ , Seemann MD . Whole-body imaging with PET/MRI. Eur J Med Res 2004;9:309–12. 49. Schaefers KP. Imaging small animals with positron emission tomo graphy. Nuklearmedizin 2003;42:86–9. 50. Tatsumi M, Nakamoto Y, Traughber B, et al. Initial experience in small animal tumor imaging with a clinical positron emission tomo graphy/computed tomography scanner using 2-[F-18]Fluoro-2-deoxyD-glucose. Cancer Res 2003;63:6252–7. 51. Melder RJ, Brownell AL, Shoup TM, et al. Imaging of activated natural killer cells in mice b y positron emission tomo graphy: preferential uptake in tumors. Cancer Res 1993;53:5867–71. 52. Seemann MD, Schaefer JF, Englmeier K.-H. Virtual positron emission tomo graphy/computed tomo graphy-bronchoscopy: possibilities, advantages and limitations of clinical application. Eur Radiol 2007;17:709–715 (Epub 2006 August 15). 53. Seemann MD . Detection of metastases from gastrointestinal neuroendocrine tumors: prospecti ve comparison of 18F-TOCA PET, triple-phase CT and PET/CT. Technol Cancer Res Treat 2007;3: 213–20. 54. Seemann MD , Gaa J . Images in cardio vascular medicine. Cardiac metastasis: visualization with positron emission tomo graphy, computed tomo graphy, magnetic resonance imaging, positron emission tomography/computed tomography, and positron emission tomo graphy/magnetic resonance imaging. Circulation 2005;112:e329–30.

4 SPECT

AND

SPECT/CT

BRIAN F. HUTTON, PHD AND FREEK J. BEEKMAN, PHD

Today, the majority of clinical procedures using tracers to visualize specific tissue binding sites are performed with planar gamma-camera imaging, single photon emission computed tomo graphy (SPECT), and positron emission tomography (PET). Ev en after the recent e xplosive growth of clinical PET and PET/computed tomo graphy (CT) installations, the imaging of single-photon emitting radiopharmaceuticals with gamma cameras, both in planar mode or with SPECT , mak es up b y f ar the lar gest fraction of clinical nuclear medicine imaging procedures (approximately 95% in Europe in 2007). Man y clinically established tracers for gamma cameras are commerciall y available and used in imaging depar tments. The singlephoton emitting radionuclides used as labels for tracer molecules often have sufficiently long half-lives to allow for long-distance transportation or can be obtained on site via generator systems. Tracers for SPECT can often be readily prepared on site using commercial reagents and kits. Therefore, in contrast with PET , the infrastr ucture associated with cyclotron production is not required. Traditionally, SPECT has evolved using conventional planar systems mounted on some form of rotating assembly, designed for fle xibility of both planar and SPECT acquisition. There is a trend to ward application-specif ic systems, optimized for a specif ic purpose (heart, breast). Another trend in nuclear medicine is toward systems that provide molecular/functional images (PET, SPECT) registered with str uctural/anatomical images obtained with CT or magnetic resonance imaging (MRI). These are of crucial impor tance for research, diagnosis, and patient treatment. Multimodal approaches such as SPECT/CT and PET/CT have already proven to significantly enhance accuracy of diagnosis and patient management in man y cases. The main reason for this is that molecular processes that show up at the site of, for example, a tumor or infection process can be accuratel y localized in an anatomical framework and attributed to a specif ic tissue or an or gan. This combination of both modalities 40

presents added diagnostic information compared to either in isolation. Fur thermore, there is potential to enhance reconstruction and impro ve quantif ication of (local) amounts of radionuclides in the body using the combined information from multiple modalities. Like PET/CT, integrated SPECT/CT devices acquire both emission and transmission tomo graphy with the patient in the same position. Images can be readil y adjusted for differences in for mat and scanner geometr y to overlay the images. Grafting the high spatial resolution capabilities of toda y’s high-speed CT scanners with SPECT’s accurate def inition of disease processes v astly enhances anatomical mapping and localization, mo ving the new dual modality systems directly into a wider range of clinical applications. An impor tant additional adv antage is that with the re gistered CT scan, attenuation correction can be accuratel y applied, which greatly reduces the problems of distor tion and quantitati ve inaccuracies that typically occur with stand-alone SPECT. The goal of this chapter is to acquaint a broad readership with the principles of SPECT and SPECT/CT . In addition, we will attempt to place moder n SPECT in the perspective of past and future SPECT instr umentation and its clinical applications. A primer on the ph ysics of SPECT can be found in Cher ry and colleagues; 1 the reader interested in fur ther detail is refer red to Wernick and Aarsvold.2 A review of nuclear molecular imaging combined with str uctural/anatomical imaging can be found in the study by Cherry.3

IMAGING SINGLE-PHOTON EMITTING RADIONUCLIDES The Gamma Camera The gamma camera, almost al ways based on the so-called Anger principle,4 continues to be the main component of most commerciall y a vailable SPECT

SPECT and SPECT/CT

systems. These systems consist of one to four camera heads mounted on a gantr y so that the y can be rotated around the patient. The main consideration in the design of these systems in the past has been v ersatility of use since in man y applications SPECT has been considered complementary to planar imaging. This is still tr ue in some areas of clinical application (e g, bone scans) although in other areas (e g, myocardial perfusion and brain receptor studies) SPECT is the method of choice. Although dedicated SPECT systems ha ve been de veloped in research centers based on a full ring of detectors (like PET), the y lack v ersatility and compromise the need to position detectors close to the patients at all times; de velopment has therefore centered on adapting planar detectors for SPECT use. To understand the operation of SPECT , it is therefore helpful to have a general understanding of the Anger gamma camera. The basic components of the Anger gamma camera are illustrated in F igure 1. The detector is usually a single scintillation cr ystal, most times made out of sodium iodide (NaI) with thallium impurities of dimension typically 500 × 400 mm and 9.5 mm thick. Gamma photons interact with the cr ystal producing light that emanates in all directions from the point of interaction. The origin of the interaction is deter mined from the light distribution, w hereas the ener gy deposited is propor tional to the inte gral of the light produced. The position and ener gy are deter mined b y first converting the light to a measurab le signal, most times by means of a set of photomultiplier (PM) tubes; they also magnify the small signal that is produced in the photo-sensiti ve la yer at the entrance to the PM tubes. The f inal position and ener gy are then determined electronically (although nowadays this is usually a digital rather than analo gue calculation). Each detected photon therefore is assigned to an ener gy and detector surf ace position. Since photons that scatter lose energy, the energy information can be used to discriminate radiation scattered in the patient from photons that reach the detector without ha ving undergone interaction with tissue; the location simply allows photon counting for a f ine matrix of picture elements (pixels) that cor responds to the detector area. Due to various uncertainties in the detector system, the spatial resolution of the detector itself (intrinsic resolution) is typically 3 to 4 mm full width at half maximum v alue (FWHM) (relating to the spatial distribution of counts from a point source); the ener gy resolution for NaI is approximately 10% FWHM (in this case, relating to the energy distribution for a monoenergetic emitter).

Light guide

41

Scintillation crystal Collimator

C A PM tubes

B D

Figure 1. Cross section through patient and gamma camera. Emitted gamma quanta with different trajectories are shown. A, Ray that goes through collimator parallel with a hole, B, Ray that penetrates the collimator septum, C, Ray that is captured in collimator because angle deviates too much from hole direction and thus not detected, D, Ray that results in scintillation after a scatter event in the patient.

Improvements in the camera’ s intrinsic ener gy and spatial resolution are e xpected with alter natives to PM tubes, for example, using position-sensitive PM tubes or PM tubes with a higher quantum ef ficiency. In addition, several solid state alternatives to PM tubes such as silicon PMs,5 special charge-coupled devices (CCDs) lik e electron multipl ying CCDs, 6–10 avalanche photodiodes, and silicon drift diodes 11 are under development. See Pichler and Ziegler12 for a useful summary of some of these technologies. Also, se veral ne w scintillator materials ha ve been discovered or are under development (eg, LaBr3:Ce and LaCl:Ce), 13–15 which have signif icantly higher light output and density than NaI and other benef icial properties to improve gamma camera performance (see reviews by Moses and Shah 16). Scintillation cr ystals can also be improved by str ucturing the material in such a w ay that light spread is reduced , w hich is par ticularly impor tant when high-resolution light readout is possib le. Examples include monolithic cr ystals with g rooves, pix elated crystals, or tin y CsI needles that are g rown on a special substrate.9 In addition, alter natives to scintillation detectors are being de veloped such as solid state detectors (e g, cadmium zinc telluride [CZT]) that directly convert deposited γ-ray ener gy into electrical signals (see Vavrik and colleagues17 and summar y by Wagenaar18). The elimination of a light-conversion step results in good signal collection with ener gy resolution much impro ved compared with scintillation detectors (typicall y appro ximately 5% at

42

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

140 keV for medical systems that are based on pix elated detectors). This superior ener gy resolution pro vides improved discrimination of multiple radionuclides. An additional benef it is the small size of these detectors, which enab les g reater fle xibility in system design (see recent developments in the section below). A key element of the overall system is the collimator, which consists of a lead or tungsten “hone ycomb” for a parallel-hole collimator, designed so as to ideall y eliminate all photons that are not tra veling nor mal to the detector surf ace. The presence of a collimator limits the direction of the incoming photons. Without this, it becomes extremely hard to determine the origin of detected photons. To gi ve some perspecti ve, for a low-energy high-resolution collimator, the hole dimensions are approximately 35 mm long, 1.4 mm diameter with lead septa 0.15 mm thick. The collimator design largely deter mines not onl y the o verall spatial resolution of the system determined by the hole diameter and length but also the number of counts acquired per unit of activity or ef ficiency of the system. The collimator dimensions can be altered to def ine a dif ferent performance, but increasing ef ficiency b y increasing hole size will result in de gradation of resolution; there is usually a compromise between these two parameters. In addition, thickness of septa needs to be adjusted to minimize penetration w hile imaging higher ener gy radionuclides. Despite the man y de velopments in the Anger camera design since its inception, for most of the presently used commercial systems, the collimator remains the main component that deter mines o verall performance. The lo w sensiti vity means that studies must be acquired for a relati vely long time to accumulate sufficient counts. The only alternative would be to increase the administered acti vity, but this is limited by the radiation dose to the patient. Although parallelhole collimators are b y f ar the most commonl y used collimators, there are alter native systems that are f inding increasing application. Some of these are illustrated in Figure 2. Of par ticular interest are pinhole collimators that have proved ideal for preclinical imaging and various combinations of slits (eg, crossed slits 19 or slitslat collimators 20–22) that are being e xplored in combination with ne w detector designs. The main feature of the pinhole (and slit) collimator is that the acquired image is magnif ied, providing gains in both resolution and sensiti vity pro vided a small v olume placed close to the collimator is imaged , hence the attraction in imaging small or gans (e g, th yroid) and small animals (see Chapter 7 “microSPECT/CT/MRI”).

General Principles of SPECT Acquisition with Multihead Cameras SPECT Acquisition

The standard SPECT acquisition in volves rotation of the detectors with collimators around the patient to acquire a sufficient number of angular vie ws (“projections”) to enable reconstr uction of 3D v olume images. Although alternatives to rotation-based systems ha ve been built (eg, use of many pinholes with stationar y detectors, Rowe and colleagues 23), most principles of acquisition and reconstruction are similar . F or simplicity, the discussion will be limited to a rotating camera system. The cameras are kept as close as possib le to the patient during rotation often with the aid of automatic laser-based contouring control (eg, using a 180° orbit in the case of cardiac imaging). Provided the camera is perfectl y aligned with the axis of rotation, each row of the acquired projection pix els corresponds to a unique slice of acti vity and these projections are used for reconstr uction. Acquisition therefore involves simultaneous acquisition of multiple slices, with each projection angle being acquired for typically 10 to 40 seconds depending on the number of angles and total scan duration. This is quite unlik e PET w here data from all projection angles are acquired simultaneousl y, albeit for a smaller axial extent, or early model CT w here data were acquired very fast for a single slice (multi-slice acquisition on more recent CT systems has closer similarity to gamma camera acquisition). The projection-b y-projection acquisition of SPECT places limits on the ability to acquire dynamic studies although recently proposed designs have less limitations (see later subsection). Alternatively cardiac SPECT can be acquired using electrocardio graph (ECG) gating to evaluate cardiac w all motion, m yocardial thickness, and pump function (ejection fraction). There are some quality control (QC) considerations that are specif ic to SPECT ; these considerations mainl y address potential prob lems that occur with the type of acquisition system. Since indi vidual detectors are rotated around the patient, any focal non-uniformity on the detectors, especially near the center of the detector , will result in a visib le ring ar tifact on the reconstr ucted image. Therefore, careful attention to uniformity with regular QC checks is recommended. The rotation of a relatively heavy detector (due to lead collimator and shielding) in the past has resulted in prob lems of mechanical stability , usuall y reflected in some error in the mechanical – electrical alignment of the system. This er ror in the center of rotation (COR) can result in de gradation of resolution and hence

SPECT and SPECT/CT

43

L P

A

B

C

D

E

F

Figure 2.

Schematic of different collimators. A, parallel hole, B, fan-beam, C, cone-beam, D, pinhole, E, crossed slits, F, slit-slat.

the COR should be checked regularly. Correction normally involves a simple image shift, so the prob lem is easil y rectified. F ortunately, recent systems tend to use f airly sturdy gantries to ensure mechanical stability . Ho wever, note that system calibration becomes critical for pinhole systems where small er rors can be v ery much magnif ied; failure to accuratel y calibrate these systems can lead to extreme deterioration of image quality. Tomographic Reconstruction

There exist two main approaches to calculate the SPECT image volume from projection images (tomographic image reconstruction): (i) anal ytic methods that calculate the image by estimating the in verse of a for mula that represents the image for mation process or (ii) discrete methods that are in principle based on using a matrix representation

of the image for mation process. The image for mation process in this conte xt is ho w photons from positions (x, y, z) in the patient are mapped on projections (x’, y’). Image reconstr uction with anal ytical methods includes algorithms kno wn as “f iltered back-projection” for parallel and f an-beam collimators (Bar rett and Swindel24 and Tsui and F rey25) or for cone and pinhole geometries, methods lik e the F eldkamp algorithm. 26 In a very much simplif ied model of the emission process, the only knowledge of the source activity distribution that can be deduced from the measured projection (for a parallelhole collimator) is the line perpendicular to the detector at the point of interaction, along which the photon must have originated; the back-projection process ef fectively uses this assumption to allocate equal v alues to pix els that lie along this line. The super position of the line intensities determines the pixel intensities for the reconstructed slice.

44

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Imagine first that there is only one radioactive point in the patient; then, in an ideal imaging system, this will result in a single spot on each projection. At the position of the radioactive source, the back projection results in a maximum, but around it, there is a signif icant amount of “reconstructed activity.” In the absence of noise, this blurring effect that occurs b y the summation of crossed lines can be exactly compensated with a special f ilter (referred to as a ramp f ilter). In practice, w here noise is present, some degree of local smoothing must also be applied. The advantage of anal ytical methods is their computational speed, but (i) they have limited robustness to quantum noise that is present in projections, (ii) the y usually do not compensate for image b lurring ef fects, and (iii) they are not really flexible enough to handle the complicated collimator geometries and detector placements that can be designed to e xtract additional information from the object being imaged. Therefore, most modern systems are equipped with iterati ve methods of reconstruction, such as the maximum lik elihood expectation maximization (ML-EM) algorithm (Lange and Carson27), and accelerated versions such as ordered subsets EM (OS-EM).28,29 ML-EM is a statistical algorithm since it tak es into account the characteristics of the Poisson noise in acquired projection pixels. In addition, the algorithms can incor porate models of image de gradation to compensate for these ef fects. Examples of effects being compensated b y iterati ve methods are nonuniform gamma ray attenuation, distance-dependent sensitivity, and scatter.30 Other factors such as spatiall y variant resolution and radiation penetration along the collimator hole edges can also be included 31,32 to

correct for these image de grading ef fects. As a result, statistical algorithms in general produce images with better resolution and ha ve better noise characteristics than analytic algorithms. Iterative Image Reconstruction

The task in iterative SPECT reconstruction is to calculate the 3D distribution of activity based on measured projections. Iterative reconstruction is based on the idea that the calculated 3D activity distribution is close to the real distribution when simulated projections based on the acti vity estimate closel y match projections acquired b y the SPECT camera. The process of iterative reconstruction is the iterative estimation of the best solution to this problem and in volves the repeated application of a set of operations (including the simulation of projections often referred to as forw ard projection) that pro gressively provides a solution that gets closer and closer to a correct estimate of the unkno wn activity distribution. F igure 3B shows the concept of this iterative updating. In practice, before the iterati ve calculations can star t the relationship between the object v oxels and measured projections has to be estab lished. F or a specif ic object voxel A i, one can estimate the probability that a photon will be detected at a specif ic detector pix el P j based on knowledge of the detector geometr y and lik elihood that photons will be emitted from the object (eg, patient); this probability is represented as an element of a matrix M ji. These elements need to be kno wn for each indi vidual voxel-pixel combination (this is often refer red to as the system model or system matrix). The entire set of

Projection space

Object space

Current estimate Ae M11 . A1

M12 . A2

......

M1V . AV

P1

M21 . A1

M22 . A2

......

M2V . AV

P2

A

MU2 . A2

......

MUV . AV

Estimated projection Pe “Compare” e.g. - or/ Measured projection P

Update

Object error map MU1 . A1

Simulation step M Ae

“Backprojection”

“Error” projection

PU

B

Figure 3. Frame (A): equations describing how activity in the object is mapped onto the projection images. Each projection pixel Pj measures the sum of contributions from all object voxels Ai, where Mji is the probability of photons being detected as defined by the system model (see text). The iterative scheme is shown in frame (B).

SPECT and SPECT/CT

measured projection pix els is represented b y a v ector P. The numbers M ji together with P determine the set of equations from which the activity distribution A has to be determined (see F igure 3A). During an iteration of the ML-EM algorithm, the actual estimate of A (which we call Ae) is used to generate an estimate of the projections, denoted with vector P e, simply by carrying out the summations, as presented in Figure 3A, but withAe instead of A. Next, ML-EM uses the ratio P/Pe to calculate errors in the projections from which an object er ror map is reconstructed (the dif ference (P − Pe) is used in alter native algorithms). The error map is then used to update Ae. The generation of a new Pe and the updating of A e often need to be repeated hundreds of times to obtain a good solution. Because of the requirement of man y iterations, acceleration methods to speed up the algorithms ha ve been developed. The OS-EM algorithm 27 is currently the most popular method. It updates the solution based on only a subset of projections rather than recalculating the update using all projections, w hich in volves much less computation. Overviews of the subject of iterati ve SPECT image reconstruction and information on how to car ry out comparisons and update steps during reconstr uction can be found in studies b y Hutton and colleagues 33 and Lalush and Wernick.34 The accurate deter mination of the matrix elements of A can be dif ficult, often requiring comple x calculations and/or measurements that are specific to each different SPECT device. An accurate match of the matrix elements and real detection probabilities has a criticall y important influence on the reconstructed image; the number of iterations and quality of image smoothing for noise suppression are also impor tant. With the de velopment of advanced algorithms that are cur rently available to reconstruct images from comple x geometries, images of superior resolution and quantitative accuracy can be produced.

QUANTITATIVE SPECT Most applications of SPECT rely on qualitative interpretation rather than absolute quantif ication; ho wever, the ultimate goal is to reduce or remove any artifacts and ideally to provide a map of the absolute acti vity concentration in the patient. A number of f actors tend to limit quantitative accurac y (as w ell as diagnostic quality): attenuation and scatter of photons in tissue, the limited resolution of SPECT , and presence of motion. Approaches e xist for the cor rection of all these f actors although there tends to be no standard approach, w hich leads to signif icant intersite v ariability in results. The objective of this section is to pro vide some insight to the

45

problems and an introduction to some of the a vailable approaches for correction. The reader is referred to more detailed coverage of this topic in the re views by Wernick and Aarsvold2 and Zaidi.35

Corrections for Attenuation The naïv e assumption that underlies f iltered backprojection (FBP) reconstruction is that the recorded counts simply reflect the line integrals for the activity distribution; of course in reality emitted photons interact with tissue and so are Compton scattered resulting in attenuation of the emitted radiation. It is con venient to distinguish attenuation as the reduction of measured counts compared with what would be expected in air versus scatter as an increase in the measured counts due to the inclusion of some scattered radiation that still falls in the selected energy window (misplaced compared to the primary events) (see Figure 1). The consequence of attenuation is there will be reduced number of photons originating from depth in tissue, reflected b y reduced reconstr ucted values central to the patient. The situation with non-unifor m attenuation is more complex as the v ariable degree of attenuation leads to inconsistencies in the recorded projections with at times quite serious ar tifacts that are not immediatel y obvious. For example, there is a tendency for increased counts to be reconstructed in areas of low attenuation with a reduction in counts in areas of higher attenuation. In practice, tw o distinct approaches are in widespread use. The Chang correction36 is an appro ximate cor rection, w hich can be sufficient for qualitati ve studies w here there is unifor m attenuation; it simpl y in volves estimating the a verage attenuation at each point in the subject and using this as a post-reconstruction cor rection f actor at the point concerned. Although modif ied Chang algorithms ha ve been developed to deal with non-unifor mities in attenuation, iterative reconstr uction that incor porates a measured attenuation map is preferred for these cases. The measured attenuation map is used to modify the system matrix that is used to estimate the probability of detection at a projection pixel given activity at a specif ic voxel (referred to as Mji in the section on reconstr uction above). Provided an attenuation map is a vailable, the implementation is relatively straightforw ard as the algorithm (e g, OS-EM) is identical except for the different system matrix. A number of techniques w ere de veloped to measure the attenuation in combination with SPECT . They originally involved use of an external radionuclide that was used as a transmission source, ef fectively recording a lo wquality transmission scan similar to CT . Various source configurations were used (see Bailey37 for a comprehensive

46

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

overview). Ho wever, the commercial implementations of most of these systems ha ve been tested and pro ved to be limited.38 Potentially interesting alter natives to these systems involve use of a scanning point source in combination with f an-beam collimators 39 or stationar y point sources combined with cone-beam collimators. 40 In the earl y nineties, Hase gawa and colleagues 41 proposed the use of CT for obtaining the transmission map, in effect the f irst suggested dual-modality system. The f irst commercial system, partly motivated by the need for attenuation cor rection for use in gamma camera coincidence imaging, w as released b y GE Healthcare in 1999 with other companies no w also of fering SPECT/CT systems. These pro vide transmission maps of higher quality than radionuclide methods although there can be prob lems due to motion ar tifacts in the thorax and mismatch of breathheld CT with emission data (see later section SPECT/CT).

Scatter Correction Photons interacting with tissue under go Compton scatter, with deflection and consequent loss of ener gy. Although energy discrimination is used to reduce the number of detected scatter photons, there are scattered photons that cannot be eliminated, predominantly for photons that have undergone a single interaction in tissue. They typicall y constitute 25 to 40% of detected photons. The spatial distribution of scatter from each point of acti vity is quite wide but is constrained to within the boundar y of the patient (unlike the situation in PET w here scatter e vents can lead to coincidence lines of response outside the body boundary). Man y methods for scatter cor rection ha ve been suggested (see re views35,42), but fe w are routinel y applied in practice. The most popular practical approach is the triple energy window method,43 where narrow energy windows are def ined close to the photopeak to estimate the scatter within the photopeak. The use of direct measurement overcomes the limitations of models that mak e simplifying assumptions regarding the scatter distribution, which is a distinct adv antage. A disadvantage is the additional noise propagated if scatter is subtracted from projections (subtraction of tw o noisy datasets results in a corrected image w here the noise amplitude is increased , while the image v alues in the dif ference image are smaller). The alter native is to simpl y add the measured scatter as part of the forward projection step during iterative reconstruction, which significantly reduces the noise. An approach that is gaining in popularity is to model scatter, which is computationally demanding especially if Monte Carlo modeling is used 44,45; ne vertheless, with

improved computer speed and optimized algorithms, these alternatives are now becoming practical. 46

Correction for Effects of Limited Resolution The limited resolution in SPECT not onl y results in blurred images but also a reduction in the observed/measured acti vity concentration for small objects (usually refer red to as the par tial volume effect). This is a direct consequence of the spreading of counts outside tissue boundaries due to poor resolution. Cor rections are therefore concerned with both improving resolution (image shar pness) and contrast and cor recting quantitative v alues. It is becoming quite common to include details of the collimator and detector b lur in the system matrix during iterati ve reconstr uction.47–50 This encourages a solution in w hich resolution is impro ved even if quantitati ve v alues are not full y reco vered. The side benef it of this approach is that reconstr ucted noise tends to ha ve improved characteristics, to the e xtent that reduced (“half ”) imaging time is being adv ocated. However, note that the f inal reconstr ucted data will still be affected b y some de gree of par tial v olume ef fects. The absolute correction of par tial volume effects remains difficult, and most approaches are relying on the availability of high-resolution anatomical data to accuratel y def ine boundaries for the area of uptake. Provided the anatomical data can be accuratel y registered with the emission data and the re gions of interest accuratel y segmented, cor rection for par tial volume losses can be applied on the basis of selected unifor m acti vity re gions51 or for indi vidual pixels.52 This has tended to limit cor rection to specif ic study types (e g, brain perfusion, m yocardial perfusion). Correction for lesions has mainly involved the assumption that lesion shape is spherical, in w hich case precalculated corrections can be applied. Attractive alternatives are iterative deconvolution algorithms that do not depend on the availability of anatomical data. 53 In all cases, an accurate estimate of the resolution must be available. Since resolution is position and object dependent (and depends on the number of iterations in the case of iterati ve reconstr uction), estimation of resolution is not tri vial (see the comprehensive review by Rousset and Zaidi 54).

Motion Correction It is some what ironic that gi ven the sophistication of instrumentation and reconstruction algorithms, correction for patient or or gan motion remains lar gely unsolv ed.

SPECT and SPECT/CT

Despite efforts to minimize motion with positioning aids, patient motion is hard to a void and cer tainly involuntary motion (eg, heart and lungs) cannot be avoided. It is common to use electronic gating to freeze c yclic motion at selected times relati ve to a measured signal (ECG for heart or chest displacement for lung). F or brain studies, the motion can be assumed to be rigid and so correction is more straightforward. This can be achieved using external motion measurement (eg, using optoelectronic or mechanical devices)55–57 where measured motion is incor porated as part of the reconstruction process either using an interiteration registration or once again modifying the system matrix.58,59 Data-driven correction for head motion during SPECT acquisition has also been demonstrated. 60,61 In some respects, the slow acquisition of SPECT can be useful in that it simpl y averages motion effects with a resultant b lurring but minimal ar tifacts. Ho wever, prob lems arise in re gistering SPECT data with data acquired more quickly, such as breath-held CT, since there will be a mismatch betw een the tw o studies. 62,63 Correction for this mismatch is the subject of cur rent research (see section SPECT/CT).

SPECT/CT We ha ve discussed the need for attenuation cor rection, especially in applications w here quantif ication is required. To accurately correct for attenuation, especially where there is non-unifor m attenuation, there is a need for a map of the attenuation coef ficients that can be derived from a CT scan. The first SPECT/CT system was designed specif ically for this pur pose,41 predating the introduction of PET/CT 64 by several years. A comparative study of ef fectiveness of attenuation cor rection clearly demonstrated the v alue of CT compared to the rather imprecise systems a vailable commerciall y using radionuclide transmission sources. 38 The a vailability of anatomical infor mation also aids g reatly in localization, especially in the case of tracers w here uptak e is highl y specific, for e xample, labeled monoclonal antibodies (Figure 4). The availability of the dual-modality system continues to f ind widening application to complement SPECT alone, although acceptance and application has been less dramatic than in the case of PET/CT, where use of CT in combination with Fluorine-18 labeled deoxyglucose (FDG) has had a significant impact on clinical diagnosis. In this case, full diagnostic CT systems are used as standard, whereas at the time of writing, there remains a range of specif ications for the CT component used in combination with SPECT (Table 1). The necessary specifications depend to some e xtent on the application

47

(eg, localization or simpl y attenuation cor rection where contrast detail is less necessar y) although clinical requirements remain some what undef ined (see Hamann and colleagues 65 for a useful comparison). Indeed , the question arises as to w hether a combined SPECT/CT gantry is required 66 or not; alternative means of combining data from SPECT and CT may be more cost effective and suf ficient, especiall y gi ven the indications for SPECT/CT studies; this ma y be achie ved either via software registration (eg, see Eberl and colleagues67) or using separate CT and SPECT gantries. 68 Guaranteeing re gistration between SPECT and CT is best achie ved through use of a combined gantr y, however, it must be noted that the studies are sequential, so there is still a possibility of movement betw een studies. Pro vided care is tak en in patient positioning with a similar bed and patient restraints, softw are re gistration should be possib le (although this assumes the patient is identical during both the e xaminations, w hich can seldom be guaranteed (eg, change in stomach contents, b ladder f illing). Although there are issues in guaranteeing alignment, the use of a single CT unit in combination with multiple cameras would appear to be an attractive solution. There are several considerations in using CT in combination with SPECT, which are outlined below.

Conversion of CT Numbers to Attenuation Coefficients Since CT depends directly on measuring electron density in tissue, the conversion from Hounsf ield Units to attenuation coefficients for a gi ven radionuclide is relati vely straightforward. Unfor tunately, attenuation coef ficients are energy dependent and consequently the required conversion is non-linear . A bilinear function w as suggested for application in PET studies69 and a similar approach is adopted for SPECT radionuclides. 70 A limitation in this conversion is that it results in signif icant over-estimation of the attenuation coef ficient for contrast materials; therefore, there is a tendenc y to acquire a lo w-dose CT for attenuation cor rection without contrast rather than correct for this aber ration (in fact to date contrast material is seldom used in conjunction with SPECT).

CT Artifacts There can be additional CT artifacts that not only influence the accurac y of attenuation cor rection but also interfere with CT inter pretation (see Bockisch and colleagues71). For example, the presence of metal can

48

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 4. Top: SPECT/CT system (Symbia; Siemens Healthcare) with two examples that clearly illustrates the value of fusion imaging to localize increased uptake of radioactive tracer in the foot (Images with permission from Siemens Healthcare). Bottom: low-dose CT, SPECT and SPECT/CT images (Hawkeye 4; GE Healthcare), which illustrate specific uptake of Iodine-123 labeled Metaiodobenzylguanidine (MIBG); without CT, exact localization is difficult. Note also serious artifacts due to bowel motion during acquisition; the artifacts in this case do not interfere with clinical diagnosis. (Images courtesy of University College Hospital, London.)

introduce streak artifacts in CT and even with the available correction techniques they are hard to avoid. A simple approach is to check the presence of suspicious areas on the non-cor rected reconstructions. Streak ar tifacts can also occur due to photon star vation, for example, if imaged with ar ms down using lo w exposure, or due to beam hardening, w hich results from presence of areas of high attenuation. Artifacts can also occur due to motion, w hether patient motion or in voluntary motion of or gans during data acquisition. A f ast acquisition period minimizes patient motion b ut renders the data more susceptib le to in voluntary motions that tend to be averaged in SPECT acquisition (eg, in the absence of gating, cardiac motion can cause ar tifacts on CT).

Registration Accuracy Since CT and SPECT studies are acquired sequentiall y, there is always possibility for misre gistration. A particular concer n in earl y scanners w as the mismatch that occurred due to change in bed height with bed extension; however, newer systems have improved bed supports that rectify this potential prob lem. It is still impor tant to

measure registration between SPECT and CT as par t of the overall QC procedures. An area of particular concern arises due to the practice of acquiring a f ast CT in the lung to minimize motion ar tifacts; the breath-held study freezes the image at a par ticular phase of respirator y motion, unlik e the SPECT study that is nor mally a veraged over a lengthy acquisition period. Although respiratory gating could be applied as in PET , this is not y et common practice for SPECT.

RECENT DEVELOPMENTS Although the clinical use of SPECT is dominated b y rotating conventional gamma cameras, there is increasing interest in developing systems that offer a better compromise between resolution and noise. The fast development of ultra-high-resolution systems for preclinical imaging and the a vailability of ne w detector designs ha ve par tly stimulated this renewed interest. It is clear that designing a system for a specif ic application provides potential for optimization, whereas the multihead planar/SPECT systems are primaril y designed for v ersatility. The motivation for the cur rent de velopments is being dri ven

SPECT and SPECT/CT

49

Table 1. SUMMARY OF SUPPLIER SPECIFICATIONS AT TIME OF WRITING Supplier

Model

No of Slices

Slice Width (mm)

kV

mA

Rotation Time (s)

GE

Hawkeye 4

Siemens

4

5

120, 140

1.0–2.5

23

23–57

Symbia T

1, 2, 6, 16

0.6–19

80, 110, 130

20–345

0.5–1.5

10–517

Philips

Precedence

6, 16

0.6–12

90, 120, 140

20–400

0.4–2.0

8–800

Philips

Brightview

140 (@1 mm)

0.3–2

120

12

60–960

5–80

mAs (360°)

Note that figures are indicative and may differ for the various options offered. Note that the exposure in milli-amp-seconds (mAs) is for a 360° acquisition and will be reduced when using helical scanning (or dose-reduction techniques).

foremost b y the desire to impro ve sensiti vity (or more precisely to reduce scan time) although ef fort is also being directed to impro ve resolution. There ha ve been several recent advances in instrumentation with development of ne w detector materials (with ref inement of the technology to impro ve stability and reduce cost), much emphasis on the search for a replacement of the con ventional PM tube (position-sensiti ve PM tubes, a valanche photodiodes, silicon photon multipliers, lo w noise CCDs), and novel system and collimator designs. To improve sensiti vity, it is clear that one requires more effective use of detectors with either a larger number of detectors sur rounding the or gan of interest or alternative novel approaches to maximize the acquisition counts. F or e xample, in cardiac imaging, there is renewed interest in collimator systems that acquire data only from the heart region using either multiple pinholes or multiple slant holes; in both cases, the objecti ve is to use standard lar ge detectors to acquire multi-angle data from a single position. 72 There are also no vel systems that are specif ically designed to impro ve sensiti vity (Figure 5). The CardiArc (Canton, USA) and MarC systems73 use slit-slat collimators (parallel-hole collimation in the axial direction with pinhole collimation in the transaxial direction). A set of slits is rotated during acquisition so as to acquire multi-angle projections. The system also per mits close positioning of the detector to the patient and impro ved patient comfor t. A further system with similar attention to patient comfort is the D-SPECT system from Spectr um Dynamics (Caesarea, Israel). 74 In this case, a set of CZT detectors is programmed so that each detector rotates on its o wn axis so as to acquire counts primaril y from the hear t region. In this case, ultra-high-sensitivity collimators are used to provide an overall gain of around 8 in the counts acquired from the hear t region as demonstrated in clinical studies. 72 Using this system, cardiac study acquisition is therefore reduced to 2 to 4 minutes. The system relies on proprietar y iterative reconstr uction algorithms

to achieve a reconstr ucted image quality similar to that obtained using a con ventional SPECT system. Fur ther cardiac systems proposed with e ven higher projected performance include use of multiple pinholes75 or multiple slant holes.76 The sensitivity to cardiac activity compared with con ventional dual-head gamma camera systems can be increased by a factor of 10. Finally, cone-beam collimation combined with point source at the focal point of the collimator also has a high potential to impro ve cardiac SPECT (Manglos and colleagues40). When combined with dual or triple detector large f ield-of-view cameras and asymmetric cone-beam designs,77–79 extremely high sensiti vity from the cardiac area can be combined with v ery efficient use of the transmission sources. In this w ay, both gated SPECT data and gated transmission data, as w ell as e xcellent attenuation cor rection, could be obtained on a v ersatile SPECT system. The impor tance of the reconstr uction algorithm should not be underestimated , especiall y in the case of iterative reconstruction. There is a trend toward including increasing complexity in the system model that describes the cor respondence betw een acti vity distribution and detected counts. In par ticular, inclusion of the collimator (and detector) b lurring as a function of source distance from the collimator is pro ving v ery useful in clinical practice. The e xtension of the system model to include collimator blurring results in images with impro ved contrast and superior noise characteristics. As a result, the acquisition time can be halv ed without an y detectab le worsening of image quality compared to con ventional reconstruction (without resolution modeling). Ho wever, note that this requires increased processing time (due to the need for more comple x computation and higher than normal number of iterations). There is also potential to use higher sensitivity collimators to capture further gains in acquisition time.80 Software that offers this approach is now available from all leading suppliers (and is now also being investigated for PET 81).

50

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

C

D

Figure 5. Some new developments: A, D-SPECT (images with permission from Spectrum Dynamics), B, CardiArc (images with permission from CardiArc), C, Cardiac U-SPECT (Reproduced with permission from Beekman and van der Have75), D, Brain U-SPECT (Reproduced with permission from Goorden et al.84).

In preclinical SPECT imaging (see Chapter 7), pinhole collimators have played a central role in achieving high-resolution magnif ication for small objects, providing a v ery ef fective means of achieving submillimeter resolution. Applying similar principles to human studies is not so straightforw ard as achie ving similar magnif ication for the lar ger object size w ould require prohibiti vely lar ge detectors. Ho wever, using detectors with high intrinsic resolution, there is scope for using pinhole or crossed-slit collimators without magnification.82 In fact, using a high number of pinhole apertures with minification can pro vide significantly improved resolution and sensitivity over pinhole systems with con ventional detectors. 82–84 Dedicated systems have been developed for heart and breast imaging. Brain imagers with high resolution can also be de veloped, and the impro vement o ver presentl y a vailable systems

depends very much on the detector resolution a vailable. Demand for the latter tends to be increasing with the introduction of ne w radiophar maceuticals that of fer diagnostic promise (e g, studies of am yloid deposition and various receptor systems).

GENERAL DISCUSSION There are v arious pros and cons for the use of SPECT in comparison with studies perfor med using PET . SPECT uses single-photon emitting radionuclides that tend to have longer half-lives and consequently are more easily distributed than positron-emitting radionuclides. The radionuclides ha ve emissions at dif ferent ener gies so that simultaneous measurement of multiple tracers is possib le, unlike PET where the detected radiation always has energy of 511 keV. Also, specific binding in SPECT tracers can be

SPECT and SPECT/CT

51

Table 2. COMPARISON OF VARIOUS PARAMETERS FOR SPECT VERSUS PET SPECT

PET

Intrinsic resolution

3–4 mm

2–4 mm

Reconstructed resolution

10–15 mm (brain 8 mm)

5–8 mm

Sensitivity

0.03% (dual head)

3.0% (3D)

Energy resolution

~10%

~15–20%

Dual radionuclides

Yes

No

Attenuation (thorax)

× 10

× 20

Scatter

0.35 (140 keV)

2D ~0.15; 3D ~0.45

Random coincidences*

N/A

Proportional to square of singles event rate*

Count-rate (@10% loss)

> 250 k (total)

< 100 k (true coincidences*)

Contributors to patient exposure

Mainly gamma/long half-life

Positron/short half-life

Staff exposure

Low < 2 mSv/yr

Can be high ~4 mSv/yr

SPECT = single-photon emission computed tomography; PET = positron emission tomography. Note that numbers are indicative only and are intended to illustrate differences between the two technologies. *Readers should refer to Chapter 3 “PET/MRI instrumentation” for PET parameter definitions and further details of PET performance.

much higher than those labeled with positron emitters. SPECT still has poorer resolution than PET as cur rently used clinically (as opposed to small-animal systems where the reverse is true). The need for collimation results in poor sensitivity compared to PET although the gap in sensiti vity is nar rowing with the newer technology now available. It should, of course, be remembered that w hat matters is the relati ve reconstr ucted resolution-noise trade-of f, not always directly predicted from the relative counts. Despite criticisms in the past that SPECT w as non-quantitati ve, careful attention to all ph ysical cor rections permits quantification to a high degree of accuracy.85,86 What is limited is the typicall y longer acquisition time often required for SPECT, so patient motion can be problematic and the minimum length of acquisition for dynamic studies is usuall y limited (although less prob lematic with some ne wer designs). Attenuation factors are smaller for SPECT , as is the radiation dose per unit acti vity (for pure gamma emitters). Table 2 summarizes the comparison of human dualhead SPECT and PET general proper ties. Note that the figures presented are representati ve of general use and do not necessarily reflect recent developments in instrumentation (e g, time-of-flight PET or high sensiti vity dedicated cardiac SPECT).

CONCLUSIONS SPECT continues to be widel y used to complement planar nuclear medicine studies, being the method of choice in an increasing number of applications (eg, heart,

brain). The ease of access to single-photon emitting radionuclides with good imaging proper ties and the wide range of suitab ly labeled radiophar maceuticals guarantees continued use. Continuing de velopments in instrumentation and reconstruction and the availability of combined SPECT/CT units suggest an increasing utility in both clinical practice and research.

REFERENCES 1. Cherry SR, Sorenson J A, Phelps ME. Ph ysics in nuclear medicine. USA: Philadelphia, Pennsylvania. Elsevier Health Sciences; 2003. 2. Wernick MN , Aarsvold JN , editors. Emission tomo graphy: the fundamentals of SPECT and PET . San Die go (CA): Else vier; 2004. p. 270–92. 3. Cherry SR. Multimodality in vi vo imaging systems: twice the po wer or doub le the troub le? [re view]. Annu Re v Biomed Eng 2006;8:35–62. 4. Anger HO. Scintillation camera. Rev Sci Instrum 1958;29:27–33. 5. Dolgoshein B, Balagura V, Buzhan P, et al. Status repor t on silicon photo-multiplier de velopment and its applications. Calice/SiPM Collaboration. Nucl Instr um Methods Ph ys Res A 2006; 563:368–76. 6. de Vree GA, Westra AH, v an der Ha ve F, et al. Photon counting gamma camera based on an electron-multipl ying CCD . IEEE Trans Nucl Sci 2005;52:580–8. 7. Beekman FJ , de Vree GA. Photon-counting v ersus an inte grating CCD-based gamma camera: impor tant consequences for spatial resolution. Phys Med Biol 2005;50:N109–19. 8. Meng LJ , Clinthor ne NH, Skinner S, et al. Design and feasibility study of a single photon emission microscope system for small animal I-125 imaging. IEEE Trans Nucl Sci 2006;53:1168–1178. 9. Nagarkar VV, Shestak ova I, Ga ysinskiy V, et al. A CCD-based detector for SPECT. IEEE Trans Nucl Sci 2006;53:54–8. 10. Miller BW, Bradford Barber H, Bar rett HH, et al. Single-photon spatial and ener gy resolution enhancement of a columnar

52

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22. 23.

24.

25.

26. 27. 28.

29. 30. 31.

32.

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

CsI(Tl)/EMCCD gamma-camera using maximum-lik elihood estimation. Proc SPIE, 6142, 61421T, 2006. Fiorini C, Longoni A, Perotti F, et al. A monolithic ar ray of silicon drift detectors coupled to a single scintillator for gammaray imaging with sub-millimeter position resolution. Nucl Instrum Methods Phys Res A 2003;512:265–71. Pichler BJ, Ziegler SI. Photodetectors. In: Wernick MN, Aarsvold JN, editors. Emission tomo graphy: the fundamentals of SPECT and PET. San Diego (CA): Elsevier; 2004. p. 270–92. van Loef EVD , Dorenbos P , v an Eijk CWE, et al. High-ener gyresolution scintillator: Ce3+ activated LaCl3. Appl Phys Lett 2000; 77:1467–8. van Loef EVD, Dorenbos P, van Eijk CWE, et al. High-ener gy-resolution scintillator . Ce 3+ activated LaBr 3. Appl Ph ys Lett 2002; 79:1573–5. Dorenbos P, de Haas JTM, v an Eijk CWE. Gamma ra y spectroscopy with a circle divide 19 × 19 mm(3) LaBr3 : 0.5% Ce3+ scintillator. IEEE Trans Nucl Sci 2004;51:1289–96. Moses WW, Shah KS. P otential for RbGd 2Br7:Ce, LaBr 3:Ce, LaBr3:Ce, and LuI 3:Ce in nuclear medical imaging. Nucl Instr um Methods Phys Res A 2005;537:317–20. Vavrik D, Jakubek J , Visschers J, et al. F irst tests of a Medipix 1 pixel detector for X-ra y dynamic defectoscop y. Nucl Instr um Methods Phys Res A 2002;487:216–23. Wagenaar DJ. CdTe and CdZnTe semiconductor detectors for nuclear medicine imaging. In: Wernick MN, Aarsvold JN, editors. Emission tomography: the fundamentals of SPECT and PET. San Diego (CA): Elsevier; 2004. p. 270–92. Huang Q, Zeng GL. An anal ytical algorithm for sk ew-slit imaging geometry with nonunifor m attenuation cor rection. Med Ph ys 2006;33:997–1004. Walrand S, Jamar F , de Jong M, P auwels S. Ev aluation of no vel whole-body high-resolution rodent SPECT (Lino view) based on direct acquisition of lino gram projections. J Nucl Med 2005; 46:1872–80. Rogers WL, Clinthorne NH, Shao L, et al. SPRINT II: a second generation single photon ring tomography. IEEE Trans Med Imaging 1988;7:291–7. Metzler SD, Accorsi R, Novak JR, et al. On-axis sensitivity and resolution of a slit-slat collimator. J Nucl Med 2006;47:1884–90. Rowe RK, Aarsvold JN, Barrett HH, et al. A stationary hemispherical SPECT imager for three-dimensional brain imaging. J Nucl Med 1993;34:474–80. Barrett HH, Swindell W. Radiological imaging. The theory of image formation, detection, and processing. New York: Academic Press; 1981. Tsui BMW, F rey EC. Analytical image reconstr uction methods in emission computed tomography. In: Zaidi H, editor . Quantitative analysis in nuclear medicine imaging. Ne w York: Springer; 2006. p. 82–106. Feldkamp LA, Da vis LC, Kress JW. Practical cone-beam algorithm. J Opt Soc Am A 1984;1:612–9. Lange K, Carson R. EM reconstr uction algorithms for emission and transmission tomography. J Comput Assist Tomogr 1984;8:306–16. Hudson HM, Larkin RS. Accelerated image reconstr uction using ordered subsets of projection data. IEEE Trans Med Imaging 1994;13:601–9. Hutton BF, Hudson HM, Beekman FJ . A clinical perspective of accelerated statistical reconstruction. Eur J Nucl Med 1997;24:797–808. Tsui B, Frey E, LaCroix K, et al. Quantitati ve myocardial perfusion SPECT. J Nucl Cardiol 1998;5:507–22. Staelens S, de Wit TC, Beekman FJ . Fast hybrid SPECT simulation including ef ficient septal penetration modeling (SP-PSF). Ph ys Med Biol 2007;52:3027–43. Frey EC, Tsui BMW. Collimator-detector response compensation in SPECT. In: Zaidi H, editor. Quantitative analysis in nuclear medicine imaging. New York: Springer; 2006. p. 141–66.

33. Hutton BF, Nuyts J , Zaidi H. Iterati ve reconstr uction methods. In: Zaidi H, editor . Quantitative analysis in nuclear medicine imaging. New York: Springer; 2006. p. 107–40. 34. Lalush DS, Wernick MN. Iterative image reconstruction. In: Wernick MN, Aarsvold JN, editors. Emission tomo graphy: the fundamentals of PET and SPECT. San Diego (CA): Academic Press; 2004. p. 443–72 35. Zaidi H, editor . Quantitati ve anal ysis in nuclear medicine imaging. New York: Springer; 2006. 36. Chang LT. A method for attenuation cor rection in radionuclide computed tomography. IEEE Trans Nucl Sci 1978;25:638–43. 37. Bailey DL. Transmission scanning in emission tomo graphy. Eur J Nucl Med 1998;25:774–87. 38. O’Connor MK, K emp B . A multicenter e valuation of commercial attenuation compensation techniques in cardiac SPECT using phantom models. J Nucl Cardiol 2002;9:361–76. 39. Beekman FJ , Kamphuis C, Hutton BF , v an Rijk PP . Half-f anbeam collimators combined with scanning point sources for simultaneous emission-transmission imaging. J Nucl Med 1998; 39:1996–2003. 40. Manglos SH, Bassano DA, Thomas FD, Grossman ZD. Imaging of the human torso using cone-beam transmission CT implemented on a rotating gamma camera. J Nucl Med 1992;33:150–156. 41. Lang TF, Hasegawa BH, Soo Chin L, et al. Description of a prototype emission-transmission computed tomo graphy imaging system. J Nucl Med 1992;33:1881–7. 42. Buvat I, Benali H, Todd-Pokropek A, Di Paola R. Scatter correction in scintigraphy; the state-of-the-art. Eur J Nucl Med 1994;21:675–94. 43. Ogawa K, Ichihara T, Kubo A. Accurate scatter correction in single photon emission CT. Ann Nucl Med Sci 1994;7:145–50. 44. Beekman FJ, de Jong HWAM, van Geloven S. Efficient fully 3D iterative SPECT reconstruction with Monte Carlo based scatter compensation. IEEE Trans Med Imaging 2002;21:867–77. 45. De Beenhouwer J, Staelens S, Vandenberghe S, Lemahieu I. Acceleration of GA TE SPECT simulations. Medical Ph ysics 2008; 35:1476–85. 46. Xiao J, de Wit C, Staelens S, Beekman FJ . Evaluation of 3D Monte Carlo-based scatter cor rection for Tc99m cardiac perfusion SPECT. J Nucl Med 2006;47:1662–9. 47. Hutton BF, Lau YH. Application of distance-dependent resolution compensation and post-reconstr uction f iltering for m yocardial SPECT. Phys Med Biol 1998;43:1679–93. 48. Tsui BMW, Frey EC, Zhao X, et al. The importance and implementation of accurate 3D compensation methods for quantitati ve SPECT. Phys Med Biol 1994;39:509–30. 49. Pretorius PH, Kingm MA, P anm T-S, et al. Reducing the influence of the partial volume effect on SPECT activity quantification with 3D modelling of spatial resolution in iterative reconstr uction. Ph ys Med Biol 1998;43:407–20. 50. Zeng GL, Gullberg GT, Bai C, et al. Iterati ve reconstruction of Fluorine-18 SPECT using geomotric point response correction. J Nucl Med 1998;39:124–30. 51. Rousset O, Ma Y, Evans A. Cor rection for par tial volume effects in PET: principle and validation. J Nucl Med 1998;39:904–11. 52. Muller-Gartner H, Links J, Prince J, et al. Measurement of radiotracer concentration in brain grey matter using positron emission tomography: MRI-based cor rection for par tial volume effects. J Cereb Blood Flow Metab 1992;12:571–83. 53. Tohka J, Reilhac A. Deconvolution-based partial volume cor rection in Raclopride-PET and Monte Carlo comparison to MR-based method. Neuroimage 2008;39:1570–84. 54. Rousset O, Zaidi H. Correction for partial volume effects in emission tomography. In: Zaidi H, editor . Quantitative analysis in nuclear medicine imaging. New York: Springer; 2006. p. 236–71. 55. Lopresti BJ, Russo A, Jones WF, et al. Implementation and performance of an optical motion tracking system for high resolution brain PET imaging. IEEE Trans Nucl Sci 1999;46:2059–67.

SPECT and SPECT/CT

56. Goldstein SR, Daube-Witherspoon M, Green MV, Eidsath A. A head motion measurement system suitab le for emission computed tomography. IEEE Trans Med Imaging 1997;16:17–27. 57. Fulton R, Hutton BF , Braun M, et al. Use of 3D reconstr uction to correct for patient motion in SPECT . Ph ys Med Biol 1994; 39:563–74. 58. Hutton BF, Kyme AZ, Lau YH, et al. A hybrid 3D reconstruction/registration algorithm for cor rection of head motion in emission tomography. IEEE Trans Nucl Sci 2002;49:188–94. 59. Kyme AZ, Hutton BF, Hatton RL, et al. Practical aspects of a datadriven motion cor rection approach for brain SPECT . IEEE Trans Med Imaging 2003;22:722–9. 60. Feng B , Gif ford HC, Beach RD , et al. Use of three-dimensional Gaussian interpolation in the projector/backprojector pair of iterative reconstr uction for compensation of kno wn rigid-body motion in SPECT. IEEE Trans Med Imaging 2006;25:838–44. 61. Lamare F, Ledesma Carbayo MJ, Reader AJ, et al. Respiratory motion correction in 4D PET/CT : comparison of implementation methodologies for incor poration of elastic transfor mations in the reconstruction system matrix. IEEE Nucl Sci Symp Conf Record 2006. pp. 2365–69. 62. McQuaid SJ, Hutton BF. Sources of attenuation-cor rection ar tefacts in cardiac PET/CT and SPECT/CT. Eur J Nucl Med Mol Imaging 2008;35:1117–23. 63. Tonge CM, Ellul G, Pandit M, et al. The value of registration correction in the attenuation cor rection of m yocardial SPECT studies using lo w resolution computed tomo graphy images. Nucl Med Commun 2006;27:843–52. 64. Beyer T, Townsend DW, Brun T, et al. A combined PET/CT scanner for clinical oncology. J Nucl Med 2000;41:1369–79. 65. Hamann M, Aldridge M, Dickson J , et al. Ev aluation of a lo wdose/slow-rotating SPECT -CT system. Ph ys Med Biol 2008; 53:2495–508. 66. Beekman FJ, Hutton BF. Multi-modality imaging on track [editorial]. Eur J Nucl Med Mol Imaging 2007;34:1410–4. 67. Eberl S, Kanno I, Fulton RR, et al. Automated interstudy image re gistration technique for SPECT and PET . J Nucl Med 1996; 37:137–45. 68. Bailey DL, Roach PJ, Bailey EA, et al. De velopment of a cost ef fective modular SPECT/CT scanner . Eur J Nucl Med Mol Imaging 2007;34:1415–26. 69. Kinahan PE, Townsend DW, Beyer T, Sashin D . Attenuation cor rection for a combined 3D PET/CT scanner . Med Ph ys 1998; 25:2046–53. 70. Blankespoor SC. Attenuation cor rection of SPECT using X-ray CT on an emission-transmission CT system: m yocardial perfusion assessment. IEEE Trans Nucl Sci 1996;43:2263–74.

53

71. Bockisch A, Be yer T, Antoch G. P ositron emission tomo graphy/ computed tomography–imaging protocols, ar tifacts, pitfalls. Mol Imaging Biol 2004;6:188–99. 72. Patton JA, Slomka PJ, Germano G, Ber man DS. Recent technolo gic advances in nuclear cardiology. J Nucl Cardiol 2007;14:501–13. 73. Chang W, Liang H, Liu J . Design concepts and potential performance of MarC-SPECT–a high-perfor mance cardiac SPECT system [abstract]. J Nucl Med 2006;47:P190–1. 74. Gambhir SS, Berman DS, Ziffer J, et al. A novel high sensitivity rapid acquisition single photon molecular imaging camera. J Nucl Med 2008. [in press 2009] 75. Beekman FJ, van der Have F. The pinhole: gateway to ultra-high resolution three-dimensional radionuclide imaging. Eur J Nucl Med Mol Imaging 2007;34:151–61. 76. Xu J, Liu C, Wang Y, et al. Quantitati ve rotating multise gment slanthole SPECT mammo graphy with attenuation and collimator reponse compensation. IEEE Trans Med Imaging 2007;26:906–16. 77. Kamphuis C, Beekman FJ. A feasibility study of offset one-beam collimators for combined emission transmission brain SPECT: a feasibility study. IEEE Trans Nucl Sci 1998;45:1250–4. 78. Beekman FJ. Apparatus for making tomo graphic images. Inter national number WO97/43667, US6324258, European patent EP1007989, and Australian patent AU730166B. 1997. 79. Li J, Jaszczak RJ, Van Mullekom A, et al. Half-cone beam collimation for triple-camera SPECT systems. J Nucl Med 1996;37:498–502. 80. Kacperski K, Hutton BF. Optimal parallel hole collimator for cardiac imaging with iterati ve reconstr uction and resolution reco very. Proc full y three-dimensional image reconstr uction in radiolo gy and nuclear medicine 2007;174–7. 81. Panin VY, Kehren F, Michel C, Casey M. Fully 3-D PET reconstr uction with system matrix derived from point source measurements. IEEE Trans Med Imaging 2006;25:907–21. 82. Rentmeester MCM, van der Have F, Beekman FJ. Optimizing multipinhole SPECT geometries using an anal ytical model. Phys Med Biol 2007;52:2567–81. 83. Rogulski MM, Barber HB , Bar rett HH, et al. Ultra-high-resolution brain SPECT imaging: simulation results. IEEE Trans Nucl Sci 1993;40:1123–9. 84. Goorden MC, Rentmeester MCM, Beekman FJ . Theoretical analysis of multi-pinhole brain SPECT. [Submitted] 85. Willowson K, Baile y DL, Baldock C. Quantitati ve SPECT reconstruction using CT -derived cor rections. Ph ys Med Biol 2008; 53:3099–112. 86. Iida H, Eberl S, Kim K-M, et al. Absolute quantitation of myocardial blood flow with 201Tl abd dynamic SPECT in canine: optimisation and validation of kinetic modelling. Eur J Nucl Med Mol Imaging 2008;35:896–905.

5 PRINCIPLES OF MICRO X-RAY COMPUTED TOMOGRAPHY SHAUN S. GLEASON, PHD, MICHAEL J. PAULUS, PHD, AND DUSTIN OSBORNE, PHD

Like its clinical counterpart, X-ray computed tomography (CT), high-resolution X-ray micro-computed tomo graphy (micro-CT) is a widel y used modality for imaging anatomy in li ving specimens. In this chapter , w e re view the basic physics of micro-CT systems designed for highresolution studies of laboratory animals, the mathematical principles used to de velop reconstr ucted images, the k ey factors that deter mine image quality , and some of the commonly used applications for this technolo gy. Anatomic infor mation pro vided with micro-CT technology is v aluable in molecular imaging applications in at least two specif ic areas: (1) the anatomy provides a physical context or “map” that shows where in the body molecular events are taking place and (2) there are molecular events that ha ve a direct impact on anatomic str uctures that can be imaged using micro-CT.

film cassette to acquire projection images. 3 The X-ra y film was subsequently processed and digitized , providing data sets with sufficient resolution (~150 microns) to reconstruct images of small-animal or gans. By 1984, high-resolution X-ray detector technology had improved, and Burstein and colleagues 4 reported an ~50 mm resolution image of a mouse thorax obtained using a 90 kilovolt potential (kVp) X-ra y source and a 512-element linear ar ray of X-ray detectors. During this period, con ventional CT systems w ere also used to simultaneously image multiple mice 5 with relati vely low resolution (> 800 mm) but v ery high throughput (eight mice at a time, 9.6 seconds per image). In 1987, Flanner y and colleagues 6 brought X-ra y microtomography into a new era with the introduction of

BACKGROUND The typical conf iguration of earl y small-animal computed tomography (CT) systems is shown schematically in Figure 1, where the subject to be imaged is placed on a rotating stage betw een an X-ra y source and a tw odimensional (2-D) X-ra y detector , and typicall y hundreds of 2-D projection vie ws are acquired as the subject rotates. A three-dimensional (3-D) tomographic image v olume is then reconstr ucted using a computer algorithm. Although the principles of X-ra y CT ha ve been understood since the early work of Nobel Laureates Cormack1 and Hounsf ield,2 the de velopment of useful micro-CT systems required that se veral technolo gical advances take place f irst. In the earl y 1980s, the a vailable electronic X-ra y detectors did not ha ve suf ficient spatial resolution to generate useful images of rodents, so some of the early developers used a translating X-ray

54

projection subject on rotating stage x-ray source Figure 1. Schematic diagram of micro-CT system with rotating specimen. The X-ray source generates an X-ray beam that passes through a subject mounted to a rotating stage, and the radiograph, or projection, is captured on the opposite side by an X-ray detector.

Principles of Micro X-ray Computed Tomography

a 3-D imaging system using a 2-D detector consisting of a phosphor plate opticall y coupled to a charge-coupled detector (CCD) ar ray. To acquire a large number of X-ray photons in each micropix el (~2.5 µm × 2.5 µm), these in vestigators used a synchrotron X-ra y source beam line in place of the con ventional X-ray tube. During this time, the F ord Motor Company Research Laboratories also de veloped a 3-D microtomo graphy system for industrial applications using a laborator y X-ray tube for the source and an image intensifier screen coupled to a video readout. As a part of this effort, the scanner was used to study the subchondral (ie, directly under the cartilage) bone architecture in guinea pigs with osteoarthritis,7 human cancellous (ie, spongy , porous) bone, 8 and trabecular (ie, lattice, or f ine matrix) bone str ucture.9 A fundamental contribution of the F ord g roup w as the development of a ne w 3-D “cone-beam” image reconstruction algorithm, that is, the F eldkamp algorithm, which remains one of the most widel y used v olumetric reconstruction algorithms. 10 Rather than treating the X-ray beam as a collection of parallel “f ans,” the F eldkamp algorithm takes into account the diverging, conical nature of the X-ra y beam in its geometric model and is therefore a more geometricall y accurate reconstr uction technique. In the 1990s, a number of groups11–25 developed microtomography systems for high-resolution specimen analysis. Most of these systems used CCD-based detector ar rays, micro-focus X-ra y tubes, and had reconstructed image resolutions betw een 20 and 100 microns. The majority of the studies perfor med using these instr uments focused on high-density tissue,

x-ray source

55

such as bone or teeth, for w hich magnetic resonance imaging (MRI) is less successful. There was signif icant work in the 1990s on mouse genotyping, and se veral genotypes of interest resulted in phenotypes that include bone mark ers, hence micro-CT w as considered a v aluable tool in this effort. For in vi vo small-animal studies, par ticularly lar ge population studies, the scanner conf iguration sho wn in Figure 1 can be cumbersome because the subject must be confined in a rotating carrier designed to prevent soft-tissue or gan motion. Most commerciall y a vailable microCT systems designed for li ve animal studies use the configuration shown in Figure 2, where the X-ray source and detector rotate about a fixed animal pallet. These systems ha ve become highl y sophisticated; using detector elements with up to 16 million pixels, X-ray sources with focal spots less than 10 microns, and the ability to scan a whole mouse in less than 1 minute.

BASIC PHYSICS At i ts s implest, a m icro-CT s ystem c onsists o f a n X-ray source and an X-ra y detector that generate, as the y rotate about a specimen, 2-D projection images.These key components determine the characteristics of the acquired image data and are briefly described here.

X-ray Source The X-ray source typically used in micro-CT scanners (F igure 3) consists of an e xternal high v oltage supply, a f ilament, electron optics, and an anode encapsulated in an evacuated glass envelope. The filament produces a cloud of electrons when heated by an electric current. The electrons are accelerated away from the filament by a potential difference of typically 20 to 150 kV X-rays Cathode (2 )

Be window Anode (1 ) (metal target)

Subject Electrons

2

x-ray detector array Figure 2. Schematic diagram of micro-CT system with X-ray source and detector rotating about a stationary subject.

1

Figure 3. Schematic diagram of an X-ray source. A large voltage potential is placed between the anode and cathode. The cathode then emits a beam of electrons that travel and strike the metal target (anode). When they strike the anode, an X-ray beam is generated and travels out of the housing through a Be window .

56

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

and focused b y the electron optics to a f ine (5 to 50 micron) focal spot on the anode. The electrons decelerate and stop in the anode material (typically tungsten), converting their kinetic ener gy to heat and X-ra ys. Typically, less than 1% of the electron kinetic ener gy is converted to X-rays. The X-rays then typically travel out of the housing through a beryllium window. Beryllium is a metal that is both highl y transmissi ve to X-ra ys and strong enough to pro vide a ph ysical bar rier between the vacuum inside the X-ray source and the external environment. An aluminum filter is normally placed over the exit window to remo ve lo w-energy X-ra ys. The remo val of low-energy X-ray is called “hardening” the X-ra y beam. This is important because most of the low-energy X-rays are absorbed in the body causing additional dose to the subject, and the y can produce streaking ar tifacts around X-ray opaque objects, such as bones. The characteristic energy profile of the emitted X-ray flux is shown in Figure 4. The maximum X-ray energy in the spectrum is given by the expression as follows:

Emax = q × V ,

(1)

where V is the X-ray source operating voltage and q is the fundamental electron char ge. Thus, for e xample, a tube biased with 100 kVp will produce X-ra ys with a maximum ener gy of 100 k eV. The low-energy shape of the

spectrum is deter mined b y the thickness of the aluminum f ilter, with thick er f ilters remo ving more of the lo w-energy spectr um. The characteristic ener gy peaks are associated with the anode material; tungsten anode X-ray tubes produce characteristic energy peaks at 59.3 keV and 67.2 k eV. Note that the a verage energy of the spectrum is typically 30 to 40% of the peak energy.26 The main disadv antage associated with a broad ener gy, polychromatic X-ra y source is that lo wer ener gy (soft) X-rays are preferentially absorbed in the subject. This preferential absor ption adds nonuseful dose to the subject and creates an erroneous relationship between the thickness of a material and the ef fective attenuation coefficient. The result of this is a “cupping” or “beamhardening” artifact in the reconstructed image where the calculated attenuation coef ficient is lo wer near the center of the subject. To overcome these problems, a purely monoenergetic X-ray beam can be generated b y a synchrotron (par ticle accelerator), w hich is a lar ge, expensive, and generally inaccessible device typically used in highly specialized research f acilities. Approximately, monoenergetic beams can be generated using a more standard polychromatic X-ray source and then f iltering the beam to remo ve X-ra ys at undesired ener gies. This will help minimize an y beam-hardening ar tifacts. The disadvantage of using a highl y f iltered beam is the

Simulated X-ray Spectra (Tungsten Anode) Constant Current 5 1 mA

2.00E 1 05 1.80E 1 05

Relative Counts per second

1.60E 1 05 1.40E 1 05 1.20E 1 05

80 kVp 110 kVp

1.00E 1 05

140 kVp 170 kVp (ext) 200 kVp (ext)

8.00E 1 04 6.00E 1 04 4.00E 1 04 2.00E 1 04 0.00E 1 00 0

Figure 4.

20

40

60

80

100 120 Bias Voltage (kVp)

140

160

180

Calculated spectra of a tungsten anode X-ray source as a function of operating voltage.

200

Principles of Micro X-ray Computed Tomography

57

amount of remaining X-ra y flux will be relati vely low, and this will result in long scan times and/or noisy images due to a low system signal-to-noise ratio.

designed using micro-focus X-ra y sources with focal spot sizes ranging from a fe w microns up to ~50 µm to minimize the geometric unsharpness.

Geometric Unsharpness

Power Limitations

An impor tant perfor mance specif ication of the X-ra y source is the focal spot size A. As shown in Figure 5, when X-rays emitted from a focal spot are used to cast an image of an object located at a distance D1 from the focal spot onto an image plane located at a distance D1 + D2 from the focal spot, the resultant image has a magnification m given by the expression:

As pre viously noted , the v ast majority of the ener gy deposited by accelerated electrons onto the X-ra y source anode is dissipated as heat. The maximum power that may be applied to a micro-focus X-ra y source is, therefore, limited by the rate at w hich heat can be removed from the X-ray source target. Flynn and colleagues 25 noted that for stationary tar gets with small focal spot sizes, the heat dissipation is predominantly radial and approximately proportional to the focal spot diameter. They observed that the maximum power for a stationary-target micro-focus X-ray source appro ximately follo ws the empirical relationship expressed as:

( D1 + D2 ) m= . D1

(2)

The blur B associated with the focal spot size is gi ven by the expression:

A D2 B= = A( m − 1). D1

(3)

Note that the b lur in Equation 3 is a measure of the unsharpness in the image plane. It is typical practice to normalize the b lur b y the magnif ication w hen def ining the geometric unsharpness Ug to relate the image error to the geometry of the object being imaged:

U g = A (1 − 1 / m).

(4)

From Equation 4, it is e vident that an unmagnif ied image (ie, the limit in which an object of zero thickness is imaged while in contact with the image plane, and m = 1) will have a geometric unshar pness of 0, w hile a highl y magnif ied image will have a geometric unshar pness approaching the focal spot size. High-resolution micro-CT scanners are

Focal Spot

A

D1

OBJECT PLANE

D2

B

B

IMAGE PLANE

Figure 5. Schematic diagram showing blur (B) due to the finite size of the X-ray source focal spot (A). D1 is the distance from the focal spot to the object and D2 is the distance from the object to the image plane.

Pmax ≈ 1.4 ( A)0.88 ,

(5)

where Pmax is the maximum X-ra y tube po wer in Watts and A is the focal spot size in microns. A survey of specifications from a number of X-ra y source suppliers shows Flynn’s relationship to hold for most commerciall y available X-ra y tubes. F or e xample, a 5- µm focal spot source will typically have a maximum po wer rating of 6, while a 50-µm focal spot source will typically have a maximum power rating of 45 W. Because the total X-ra y flux emitted by an X-ray source is a function of applied external voltage V and the anode cur rent I, this power ceiling (Power = I·V) imposes an upper limit on the available X-ray flux, w hich in tur n imposes a lo wer limit on scan times due to the f act that the lo wer the X-ra y flux in the beam, the longer the X-ra y detector must dwell to collect enough signal to then generate lo w-noise image reconstructions. Most commerciall y available micro-CT scanners have minimum scan times on the order of 1 minute. In clinical CT scanners, this power limitation is alleviated by using rotating anode sources and by pulsing the X-ray source. Rotating anode sources replace the f ixed anode with a disk rotating at speeds in e xcess of 3000 rpm, thereby spreading the heat over a much larger area because the area of the tar get e xposed to the electron beam is al ways changing, allo wing better dissipation of heat a way from the tar get. Pulsed sources limit heat production to only the short periods of time during which image data is being acquired to minimize the heat load. Neither of these approaches has found wide use in commercial micro-CT scanners to date. Even minor wobble in a rotating anode can broaden the focal spot size,

58

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

thereby increasing the geometric unshar pness, and although pulsed micro-focus X-ra y sources are under development, in some cases using recently developed carbon nanotube emitters, 27–29 these de vices ha ve also not yet been widely adopted by commercial micro-CT manufacturers due to lack of general a vailability and some uncertainty about their long-term stability and reliability.

Optimal X-ray Energy For a Poisson-statistics limited detection system, in which a f inite number of X-ra ys are emitted b y the X-ray source, an optimal X-ra y ener gy e xists for best contrast resolution in CT studies.30 Micro-CT data sets are typically Poisson-statistics limited due to the limited X-ray flux emitted b y the X-ra y source and the small detector element sizes. Following the work of Grodzins,30 if a unifor m cylinder of attenuating material is scanned , the contrast resolution limit in the reconstr ucted CT image may be expressed as:

noise = signal

2 D exp(µ D )

)

3

N (Δ x µ

2

,

(6)

where noise is the statistical variation in the reconstructed image, signal is the cor rect v alue for the reconstr ucted image, D is the diameter of the c ylinder, µ is the energydependent X-ra y attenuation coef ficient of the c ylinder material (discussed in greater detail below), N is the number of photons emitted b y the X-ra y source during the study, and Δx is the detector element spacing. At low energies, where µ is large, the contrast resolution is limited by the small number of X-rays penetrating the subject (ie, no X-rays reach the detector). At higher energies, where µ is small, the contrast resolution is limited by the small number of X-rays absorbed in the subject (ie, all of the X-rays pass through the subject and reach the detector). Equation 6 reaches an optimal (minimum) v alue at the ener gy for which µ = 2/D. Flanner y and colleagues 6 have repor ted similar conclusions. Assuming an animal’s gross attenuation characteristics are appro ximately that of w ater, for a 3-cm (mou se-sized) phantom the o ptimal ener gy is approximately 25 keV and for a 5-cm (rat-sized) phantom the optimum energy is approximately 30 keV.

Beam Hardening

energy dependent, especiall y at the lo w X-ra y ener gies preferred for small-animal studies. As the X-ra y beam passes through the subject, the lo wer-energy X-rays are preferentially absorbed near the surf ace, causing the image to be artificially brighter near the edges of the subject. This is the w ell-documented beam-hardening ar tifact.26,31–33 Prefiltration of the X-ra y beam to reduce the lower energy, or “soft,” X-rays can reduce the ar tifact by making the beam more monochromatic, and a number of algorithms ha ve been de veloped to par tially cor rect for beam hardening.31,32 Nonetheless, the effect is difficult to eliminate completely.

X-ray Detector Most commercially available preclinical micro-CT scanners use lar ge area (> 50 cm 2) 2-D detectors with geometries that result in relati vely lar ge ( ≥ 5°) X-ray cone-beam angles. The choice of the cone-beam architecture (2-D detector array) over the fan-beam architecture (approximately 1-D detector array) is preferred in clinical systems because it is primaril y driven by a need to more efficiently collect the low X-ray flux produced by micro-focus X-ray sources. The most widely used detector design employs a CCD detector coupled via f iber optic taper to a phosphor screen (Figure 6). In this design, light produced when a gadolinium oxysulphide (GOS) or thallium-doped cesium iodide (CsI:Tl) phosphor screen on the f ace of the detector is focused onto a smaller CCD detector through the fiber-optic taper. This design takes advantage of the maturity of CCD technolo gy to produce detectors with lo w electronic noise, lar ge (> 107) pixel counts, and e xcellent stability. The method is limited , however, by the cost and size limitations of the fiber-optic taper, by light loss in tapers with lar ge minif ication f actors, and b y a need to correct for distortions induced by the f iber-optic taper.

X-ray

CCD Fiber Optic Taper

The pol ychromatic X-ra y spectr um leads to a second important c onsideration, b eam h ardening. As n oted above, the X-ra y attenuation coef ficient is strongl y

Phosphor Screen

Figure 6. Schematic diagram of an X-ray detector consisting of a charge-coupled device (CCD) coupled via fiber-optic taper to a phosphor screen.

Principles of Micro X-ray Computed Tomography

Other commercial systems use detectors consisting of 2-D complementar y metal-o xide-semiconductor (CMOS) sensor ar rays coupled directl y to X-ra y scintillating materials that con vert X-ra ys into visib le light. CMOS sensor technolo gy has the adv antage of suppor ting the f abrication of sensors with lar ger surf ace areas than CCD sensors. Ho wever, because of the relati ve immaturity of the technolo gy, CMOS detectors tend to have smaller (~10 6) pixel counts, more pixel defects, and poorer signal-to-noise ratios. These shortcomings are balanced, however, by the elimination of the need for a fiberoptic taper. Without a fiber-optic taper, these detectors tend to be less expensive, more compact, and do not require geometric distortion correction. For both detector systems, the spatial resolution is determined b y the characteristics of the scintillating layer and the size of the pix el elements. When used to image a standard line-pair phantom for resolution characterization, widel y used scintillators pro vide resolutions betw een 10 and 20 line pairs per millimeter (25–50 micron resolution). Detector element sizes range from 10 to 50 microns. Together these parameters set the typical intrinsic resolution of widel y used X-ra y detectors between 25 and 70 microns at the f ace of the detector. The resolution of the acquired images can be impro ved significantly through magnif ication. Recall the e xpression for the scanner magnif ication f actor (Equation 1). The unsharpness related to the detector resolution depends linearly on the magnif ication factor,

U d = U D / m,

(7)

where Ud is the magnified unsharpness due to the detector in the image plane and UD is the intrinsic detector unsharpness at the face of the detector. The overall unsharpness U in the image is given by the expression:

U = (U g2 + U d2 )1/ 2 ,

(8)

U = U D [1 / m 2 + (1 − 1 / m)2 A2 / U D2 ]1/ 2 .

(9)

or

Evaluation of Equation 9 shows that at low magnification (m ≈ 1) the image unshar pness is equal to the intrinsic detector unshar pness, w hile at high magnif ication the image unshar pness is appro ximately gi ven b y the focal spot size. The use of magnif ication to impro ve image resolution comes with a linear penalty in f ield-of-view (FOV):

FOV = Detector Size / m.

(10)

59

Thus, the choice of magnification factor is always a compromise between the desired resolution and the maximum size of the animal to be imaged. Frequently, the X-ray source and detector are placed on computer -controlled movable stages to allow the operator to select the magnification factor and FOV on a scan-by-scan basis.

MICRO-CT RECONSTRUCTION Commercially a vailable micro-CT systems typicall y acquire data using a step-and-shoot technique that results in a collection of 2-D X-ra y projection images, each acquired at a dif ferent rotation angle relati ve to the subject. These individual projections contain only 2-D information, but because the y are acquired tomo graphically (radially around the subject), the y can be processed to create a 3-D image volume of the subject. The process of converting the 2-D tomo graphic projections into a 3-D volume is termed image reconstruction. This section presents a summary of the most commonly used reconstruction techniques for micro-CT and also introduces the typical computational platfor ms on w hich micro-CT reconstruction algorithms are implemented. For the purposes of this discussion, it is assumed that a third-generation CT scanner is used , which means that a point-source X-ray source is used. Fur thermore, a 2-D X-ray detector geometr y is assumed , w hich is the most common conf iguration used in micro-CT systems, and this results in a “cone-beam” X-ray geometry. A cylindrical volume is typicall y reconstr ucted, where the axis of the cylinder is aligned with the CT system’s axis of rotation. The diameter of the cylinder is equal to the transaxial dimension of the X-ra y detector di vided b y the CT system’s magnification factor. The length of the c ylinder depends on the type of trajector y, which is described in more detail below. Before reconstruction, the projection data are corrected for backg round or “dark” signal and are nor malized by a “bright-field” projection or a projection image acquired without any attenuating material between the X-ray source and the detector . The dark-f ield projection characterizes charge accumulation in the detector typically due to thermal phenomena unrelated to the image acquisition, and the bright-field projection characterizes the nonuniform system response due to variations in detector sensitivity and X-ray source flux density . F or a single 2-D projection p in a tomographic data set, the nor malization follo ws the expression:

( p − pdark ) pnorm = , ( pbright − pdark )

(11)

60

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

where pdark is the dark-f ield reference projection, pbright is the bright-field reference projection, and pnorm is the resultant nor malized projection. Each nor malized projection has pixel values ranging from 1.0 down to nearly 0, where values of 1.0 represent re gions in w hich the X-ray flux is un-attenuated (ie, no material in the X-ra y path), and v alues approaching 0 represent regions in which the majority of the X-ray flux is absorbed. Each normalized 2-D projection is a measure of the attenuation of the X-ra y flux along the lines of response between the X-ra y focal spot and each pix el in the projection image. F or a gi ven line of response j in the normalized projection pnorm, the X-ra y flux attenuation is described by the Beer–Lambert Law: − µ ( l ) dl pnorm , j = e ∫ ,

(12)

where µ(l ) is the spatially varying attenuation coefficient of the imaged subject along the path l of the line of response. The goal of any X-ray CT reconstruction algorithm is to use the set of projections acquired at multiple angles to calculate µ for each voxel in the image volume. There are a v ariety of techniques that can be used to reconstruct the nor malized projection data. The optimal image reconstruction approach will v ary depending on the geometry of the CT scan and the application at hand. In general, the categories of reconstruction algorithms are characterized by the trajectory used during data acquisition and the computational methods used to generate the images. The sections below f irst outline the dif ferent acquisition trajectories (ie, path of X-ray source/detector around subject) currently in use or under investigation followed by a description of the two general classes of reconstr uction algorithms.

A circular orbit is the simplest of all the possib le orbits, and the resulting data set are also the most straightforward to reconstruct. The disadvantage of the circular orbit with a cone-beam CT system is that the projections suffer from incomplete data, causing errors in the resulting cylindrical reconstruction. Assuming a 2-D square X-ray detector, the theoretical volume of support that can be reconstructed correctly with a circular orbit is a sphere. When data from a circular orbit scan are used to reconstr uct a cylinder, as is typical practice, the image data that lies within the cylinder but outside of the “ideal sphere” are subject to error. The e xtent of the er ror is dependent on the cone angle of the system (ie, the angle of the X-ray axial direction of the CT scanner) and the type of algorithm used.

Helical Orbit To generate a complete or suf ficient set of data in a cone-beam X-ra y CT system, Tuy33 has sho wn that every plane w hich passes through the imaging FO V must also cut through the orbit of the focal point (ie, the center of the 2-D detector) at least once. Although a circular orbit does not satisfy this condition, a helical orbit can meet this requirement. In a helical orbit scan, data are acquired w hen the subject is translated axiall y, that is, in parallel with the CT system’ s axis of rotation, whereas the X-ray source and the detector rotate around the subject (F igure 7). In addition to meeting Tuy’s requirement for data suf ficiency, a helical orbit can reduce ring ar tifacts frequently found in images recon-

Acquisition Trajectories The geometry of the micro-CT scanner hardware and the flexibility of the motion control system (hardw are and software) will def ine the type of trajector y that can be used during the scan. All micro-CT systems must include an ability to rotate the X-ra y source and detector relati ve to the subject. If the motion hardw are and softw are also allow the axial position of the subject to be changed during the micro-CT scan, then the operator has more flexibility in ter ms of def ining the trajector y of the data acquisition. Three trajectories are described below.

Circular Orbit If the X-ray source and detector rotate about the subject and the subject remains axiall y stationary throughout the scan, the data set are acquired with a circular orbit.

Figure 7. The blue arc defines the helical orbit of the X-ray source/detector about the subject. The red line defines the axis of rotation of the computed tomography (CT) scanner. The helical orbit is achieved by spinning the source/detector while moving the animal parallel to the axis of rotation.

Principles of Micro X-ray Computed Tomography

structed from a circular orbit scan. A number of researchers have recently described various methods for reconstructing helical CT data. 34–38

Nontraditional Orbits One alternative approach to acquire a theoreticall y complete data set via helical scan is to acquire data in a circle-plus-line orbit. This trajectory starts with the standard circular orbit described previously, acquiring a full set of X-ray projections as the source and the detector rotate 360 degrees about the subject. Subsequentl y, another set of projections are acquired by holding the source and the detector stationar y (no rotation) w hile translating the subject axially, taking multiple projections along this line (Figure 8). Examples of reconstructions using this type of trajectory are given by Noo and colleagues.39 A second nontraditional trajectory is accomplished when the X-ray source tra vels along a saddle trajector y relati ve to the subject. Although this is less practicall y implemented , and therefore less commonl y used than the pre viously described trajectories, it does have advantages for axially truncated projection data, w hich has been described in detail by several researchers.40–42

Classes of Algorithms Two discrete classes of reconstr uction algorithms are widely used to generate micro-CT images. Analytic algorithms calculate the image data directl y using defined single-pass mathematical methods. These

61

algorithms have the advantage of relative simplicity and speed but tend to produce noisier reconstr ucted images. In contrast, iterati ve reconstr uction algorithms use complex models of the acquisition system, repeatedl y estimate solutions, and compare the estimate with a modeled “ideal” solution until predefined convergence criteria are met. Iterati ve algorithms typicall y produce superior images but can take much longer to execute.

Analytic Reconstruction Algorithms Analytic algorithms, such as those based on f iltered backprojection (FBP), 43 model the tomo graphic data acquisition mathematically via the Radon transform. These models are based on theoreticall y ideal and continuous acquisition conditions. Of course, these assumptions are subject to error: the data acquired are discrete and are collected using nonideal scanners. Because analytic algorithms are mathematically based, they are relatively easy to implement and also have relatively short reconstruction times. One of the main disadvantages of the analytic algorithms is that they do not account for noise in the acquisition system and data. This noise results in lo w image quality f or a pplications t hat h ave l ow s ignal-to-noise ratios, such as nuclear medicine studies (eg, positron emission tomography [PET] and single photon emission computed tomography [SPECT]). Fortunately, X-ray micro-CT scanners can achie ve high signal-to-noise ratios because the X-ra y sources used can generate lar ge numbers of X-ray photons. It is for this reason that analytic algorithms based on FBP have been frequently implemented to reconstruct data from a micro-CT system.Two categories of analytic algorithms ha ve been implemented for micro-CT reconstruction: approximate reconstruction algorithms and exact reconstruction algorithms. Approximate Reconstruction Algorithms

Figure 8. Circle-plus-line orbit. The black circle defines a circular orbit of the X-ray source/detector, and the black line represents movement of the source/detector along the axis of rotation along which additional projections are acquired. The red line defines the axis of rotation.

Approximate algorithms make mathematical assumptions regarding the data acquisition so that the reconstr uction can be implemented easil y and/or in a computationall y efficient manner. The widely used F eldkamp10 algorithm falls into the category of approximate analytic algorithms. For a micro-CT scanner that uses a circular orbit, the Feldkamp algorithm generates a geometricall y cor rect reconstruction only for the center slice of the c ylindrical volume. The geometric accurac y of the reconstr uction diminishes for slices that are not in the center , and the geometric ar tifacts increase as the slices mo ve axiall y away from the center for the FO V. So, for lar ge cone angles, the Feldkamp algorithm suffers from increasingly

62

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

severe geometric distor tions (F igure 9). F or small cone angles (e g, less than 8 de grees), the ar tifacts are small enough to be ne gligible for most CT applications. The advantages of the F eldkamp algorithm include ease of implementation and f ast reconstr uction. Development of methods to impro ve the speed and FO V coverage of the Feldkamp algorithm is a focus of ongoing research. Recently, progress has also been made in generalizing the Feldkamp algorithm to suppor t cone-beam helical orbits, further increasing its usefulness. 35 The speed and ease of implementation adv antages typicall y f ar outw eigh the problems, and therefore the F eldkamp algorithm is almost universally included in a micro-CT reconstr uction tool set. Another recentl y de veloped g roup of appro ximate analytic reconstr uction algorithms that is potentiall y useful for alter native trajectories with possib ly limited views is the so-called “chord-based” class of algorithms. This class of algorithms has been demonstrated to be useful w hen scanning constraints (e g, size of subject, limitation of motion control system) limit the scanner’ s ability to perform a standard trajectory. Also, this class is useful for re gion-of-interest reconstr uctions w here an entire standard trajectory is not needed to co ver a particular region within the body of the subject. A treatment of chord-based algorithms and their noise characteristics is presented by Xia and colleagues. 44 Exact Reconstruction Algorithms

Mathematicians ha ve de voted a considerab le deal of effort in the quest to de velop e xact cone-beam reconstruction algorithms. Although the resultant methods are theoretically valuable and elegant, they have thus f ar suffered from the implicit assumption of ideal conditions for data acquisition. The assumptions of ideal data are

Figure 9. Disk phantom (left) and Feldkamp reconstruction from circular orbit (right). Although the phantom is a stack of parallel discs, the reconstructed image shows a distortion of the disc shapes (nonparallel lines) that increases toward the edges of the cone-beam (top and bottom of image). Images courtesy of QRM website .

of course in valid, and hence, substantial ar tifacts are common w hen reconstr ucting real-w orld data with theoretically exact algorithms.45 The two most commonly referenced exact algorithms were originally developed by Katsevich46 and Grangeat. 47 Exact cone-beam reconstruction remains an area of acti ve research, and ne w variations of both of these algorithms ha ve been recently developed.48,49

Iterative Reconstruction Algorithms An alter native to anal ytic reconstr uction is the use of algorithms based on iterati ve techniques. Iterati ve algorithms take into account the discrete, or digital, nature of the data and generall y represent the data as v ectors and the acquisition process as a matrix. The typical flo w of an iterative reconstruction algorithm is to (1) generate an estimated image solution, (2) forw ard project the solution to generate a calculated set of “pseudo-projections,” (3) compare the “pseudo-projections” with the acquired projections, (4) adjust the estimated image solution based upon this comparison, and (5) repeat the process until a predef ined convergence criteria is met. The most pressing challenge in iterati ve reconstr uction of microCT data lies with numerical dif ficulties sur rounding matrix in version of lar ge data sets and with the time required to complete the iterations. Generally speaking, iterative algorithms are far more computationally intensi ve than anal ytic algorithms and require substantiall y more computing resources. F ortunately, with the continued increase in performance and decreasing cost of com puters, iterati ve techniques are becoming more practical and have recently become standard in clinical nuclear medicine, w here the reconstructed v olumes are relati vely small. Micro-CT image volumes are much larger than clinical nuclear medicine image v olumes because of the relati vely high spatial resolution (ie, larger number of voxels per unit volume), however, and for this application statistical iterative algorithms remain a computational challenge. For example, a typical micro-CT detector might have an array of 512 by 512 imaging elements, or pix els, and a typical micro-CT scan might consist of 360 projections o ver 360 de grees. The probability system matrix that is required for a statistical reconstr uction technique, such as maximum likelihood e xpectation maximization (ML-EM), 50 for a reconstructed image volume of 512 × 512 × 512 would be hundreds of terab ytes (w here 1 terab yte = 1 trillion bytes). Such an o verwhelming prob lem can be made more manageable by taking advantage of geometric symmetries, only loading subsections of the system matrix,

Principles of Micro X-ray Computed Tomography

and di viding up the prob lem across man y computing nodes,51 but the challenge remains daunting. One area of study where iterative reconstruction methods may offer sufficient advantage to justify the use of an iterative algorithm for micro-CT would be in cases where a very low X-ray dose is required. Lo w-dose studies yield data sets with poor statistics, and anal ytic reconstructions of statistically poor data typically have significant artifacts. Iterative statistical algorithms ha ve been sho wn to of fer a clear advantage in producing quantitatively accurate image data from data with poor statistics. 52

Computational Platforms X-ray micro-CT data sets are generall y considered to be large relati ve to clinical applications. Some commercially a vailable micro-CT systems use detectors with up to 16 me gapixels, w hich can produce reconstr ucted image volumes greater than 64 gigavoxels in size. Ev en standard micro-CT systems typically produce reconstructed image v olumes on the order of hundreds of megavoxels in size. These lar ge v olumes are computationally challenging, e ven for f ast anal ytic techniques, such as the Feldkamp algorithm. With no optimization, it can take several hours to reconstruct a typical micro-CT data set on a standard laborator y computer using the Feldkamp algorithm. F or this reason, a tremendous amount of research has been perfor med to accelerate micro-CT reconstruction.

Cluster-Based Reconstruction Platforms As computer costs ha ve decreased, it has become af fordable to assemb le a lar ge number of netw orked computers to share the reconstruction load. These collections of computers are commonly called clusters and are useful to speed up any computing tasks that are highly parallelizable. Clusters are most commonly configured with the LINUX operating system, but Windows™-based clusters are becoming more pre valent as w ell. P arallel implementations of CT reconstruction algorithms for these platfor ms are becoming common because of the highly parallelizable nature of the CT reconstr uction problem. An example of a message passing interf ace (MPI)-based cluster implementation of the Feldkamp algorithm with intelligent focus-of-attention FOV support is presented in the study b y Gregor and colleagues.53 The advantages of clusters are that they are relatively inexpensive, easily upgradeable, and are fle xible in that they can be used for a multitude of computationall y challenging tasks. The disadvantages of clusters are their size and relatively large power requirements.

63

Graphics Processing Units- and Field Programmable Gate Arrays-Based Reconstruction Platforms Specialized hardw are can also be used to accelerate reconstruction algorithms for micro-CT. Again, due to its highly parallelizable nature, the F eldkamp algorithm has been implemented on many specialized hardware platforms from array processors, to f ield programmable gate arrays (FPGAs) 54 and graphics processing units (GPUs). Due to their low cost, high availability, and applicability to CT (and other modality) reconstr uction, GPUs ha ve been a tar get platform for a large number of researchers and companies that are de veloping accelerated reconstruction technology55–57 GPUs, which are typically used for 2-D and 3-D graphics rendering applications, are also well-suited to implement the forward- and back-projectors used in image reconstr uction. A forward-projector takes the cur rent representation of the 3-D v olume and projects it onto the 2-D projection space, and a back-projector does the opposite—it tak es a 2-D projection and back-projects (“smears”) it back across the 3-D v olume. Finally, m any r econstruction e xperts a re t aking advantage of the latest multi-CPU , multi-core PC platforms to accelerate CT reconstruction algorithms by writing C-code that is efficiently multithreaded.58

The Hounsfield Unit As previously noted, the objective of any reconstruction algorithm is to calculate the attenuation coef ficient of the tissue in each voxel of the image volume. For practical reasons, the attenuation coef ficient has historicall y been reported in Hounsfield Units (HUs) by normalizing the measured attenuation coef ficients to the attenuation of w ater and scaling the result b y a f actor of 1000 expressed as:

(µ − µ w ) HU = × 1000. µw

(13)

The HU , also called the CT number , of fers tw o advantages over the ra w attenuation coef ficient. F irst, by nor malizing the v alues of µ obtained with a gi ven scanner to the v alues of µw obtained with the same system, some reduction in measurement v ariation between scanners is achie ved. Second, b y scaling the normalized parameter b y 1000, it becomes possib le to report the quantity as an inte ger rather than as a floating point number , reducing f ile sizes, and increasing computational efficiency.

64

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

From Equation 13, it is evident that regions where µ = 0 (ie, air) will have HU values of −1000, while regions where µ = µw will have HU v alues of 0. This calibration should be repeatable for studies acquired with a given system using a specific X-ray source operating voltage and a specific X-ray filter. However, because attenuation coefficients are strongly energy dependent, any change in a system that changes the X-ra y spectr um will af fect the HU calibration. In practice, it is typically necessary to have a separation calibration for each X-ray tube setting used. Furthermore, e ven w hen a system is properl y calibrated the measured HUs are ener gy dependent. F or example, in a clinical CT scanner operating with a mean X-ray energy of 60 to 70 k eV, measured HUs for cor tical bone are typically ~1000. In contrast, a micro-CT scanner operating with mean X-ra y energy of 25 to 30 k eV will repor t HUs for cor tical bone g reater than 2000. Thus, to compare data from two different scanners, both systems must have acquired the data using similar X-ra y source settings and the X-ra y sources must produce similar spectra. As a practical matter, to perform the HU calibration, a water phantom or w ater equivalent phantom is necessary. This phantom can be any sort of uniform cylindrical object made of a low-density material and filled with distilled water. A centrifuge tube works nicely for this application with small-bore animal scanners. The tube is placed into the scanner FO V and the calibration scan is performed using the same protocol that will be used with the animal. The critical constraint is that the X-ray source voltage and the X-ray filter thickness must be the same for the calibration scan and the animal study.

Micro-CT Protocol Considerations One of the most impor tant advantages of an y small-animal in vivo imaging technology is the ability to continue to study the animal after an imaging procedure. F or this reason, it is critical that the imaging protocol has a negligible effect on the health of the animal and the outcome of the e xperiment. The anesthesia, radiation dose, and contrast media must be carefull y selected to ensure that the study is minimally invasive.

anesthetic agents include isoflurane, k etamine/xylazine, pentobarbital, and tribromoethanol. The current trend is for most imaging studies to be perfor med using isoflurane. Detailed re views of v arious anesthesia protocols for small animals may be found elsewhere.59–61

Radiation Dose Radiation dose to the animal is an impor tant concer n when designing a micro-CT protocol. A number of authors have examined the dose delivered to a mouse or rat during a micro-CT study both empiricall y and using numerical simulations. 11,62–67 Reported dose v alues range from 1 to 15 centig ray (cGy) per scan, with typical v alues of less than 5 cGy per scan. Although the dose le vel for a single scan typicall y f alls belo w the detectable limit for a ph ysiological response, with multiple scans the accumulated dose may have a deleterious effect on an experimental study. The delivered dose depends strongly on the X-ray source settings, exposure times, and number of projections included in the data acquisition.

Contrast Media In clinical studies, iodine or barium contrast media are typically administered orally, intravenously (IV), or rectally to enhance the measured CT numbers of various organs or tissues.26 Similar protocols ha ve been de veloped for smallanimal studies, although protocol design is complicated b y the rapid clearance of most clinical contrast agents in mice and rats and b y the relatively long scan times of most preclinical micro-CT systems. In most cases, some use of a contrast-enhancing agent is required to dif ferentiate between soft tissue organs in a micro-CT study. Frequently used protocols include intraperitoneal (IP) or IV injection of a nonionic w ater-soluble iodine contrast medium (eg, Amersham Omnipaque™-300), IV injection of slo wly clearing b lood pool contrast agents (e g, ART F enestra VC™) and liver imaging agents (e g, ART F enestra LC™), and oral deli very of clinical gastrointestinal imaging agents (eg, barium sulf ate). The increase in CT number pro vided b y contrast media is appro ximately expressed as:

Anesthesia

ΔCT Number µ ʹ′contrast media ≈ , C µ ʹ′

High-resolution in vivo imaging requires that the subject be immobilized during the scan; this is typicall y accomplished b y anesthetizing the animal. Commonl y used

where C is the contrast medium concentration in the tissue (mg/mL) and µʹ′ is the density-normalized attenuation coefficient (µʹ′ = µ/density cm2/g).

(14)

water

Principles of Micro X-ray Computed Tomography

IMAGING APPLICATIONS We now tur n to a re view of some of the more widel y used applications for micro-CT. In general, micro-CT is the tool of choice for studies requiring high-resolution anatomic images of tissues with high relati ve contrast. For e xample, micro-CT is ideall y suited for bone studies due to the high contrast between calcified tissue and soft tissue as w ell as for lung studies due to the high contrast betw een air and lung soft tissue. MicroCT scans of soft tissue or gans are also frequentl y performed, but these studies typically require the use of contrast-enhancing agents. In most cases, MRI is the preferred tool for anatomic imaging of soft tissue. Micro-CT is also frequentl y used to pro vide an anatomic reference for PET or SPECT studies and as a source for attenuation coef ficients to suppor t attenuation and scatter cor rection in PET and SPECT . MicroCT is an attracti ve tool for generating anatomic reference images because the hardw are is relati vely inexpensive and easy to use and scan times can be short, allowing the imager to devote most of the study time to the PET or SPECT acquisition. Because micro-CT intrinsically measures tissue attenuation coef ficients, it is well suited for generating maps for attenuation correction, although, as described belo w, some w ork is required to scale the micro-CT attenuation coef ficients to the energies of the PET or SPECT γ-rays.

Whole-Body Imaging with Micro-CT Perhaps the most commonl y used micro-CT imaging protocol is the whole-body scan. Many commercially available micro-CT scanners can acquire w hole-mouse images in a single orbit and can acquire w hole-rat images with multiple bed positions. Whole-mouse data sets typically have resolutions of 50 to 100 microns, are acquired in 1 to 5 minutes, and deliver ~5 cGy dose to the animal. A representative w hole-mouse image is shown in F igure 10. Two applications for w hole-body data sets are se gmentation of v arious anatomic or gans and structure for numerical anal ysis and calculation of attenuation coefficients in suppor t of PET and SPECT image reconstr uctions. These applications are discussed briefly below.

Whole-Body Tissue Segmentation Although the image in Figure 10 is useful for visualizing the anatom y of a laborator y animal, the reconstr ucted data set does not intrinsically provide quantitative values

65

for the sizes and v olumes of anatomic str uctures. To extract these values from the image data, it is necessar y to se gment the image v olume into dif ferent tissues of interest. If the tissue of interest has high contrast with its surroundings (ie, bone or lung), the se gmentation can often occur automaticall y b y simpl y setting g rayscale thresholds. If the image is noisy or if contrast is poor , a more manual se gmentation is required. This can be tedious and time consuming. Assume an investigator wishes to deter mine the volume of the left kidne y in F igure 11A. In this case, the mouse was given an IP injection of a clinical w ater-soluble iodinated contrast agent before the study. The contrast agent clears through the kidneys, providing good contrast between the kidne ys and the sur rounding tissue. With sufficient contrast, it may be possible to use an automated region-growing se gmentation tool to deter mine the boundaries of the kidney. In this case, the investigator identifies a point within the region of interest and defines upper and lo wer g rayscale thresholds, w hich encompass the range of v alues found in the kidne y. The algorithm then searches for boundaries between the kidney and surrounding tissue in three dimensions. When the data have sufficient contrast to suppor t this process, se gmentation can be accomplished very quickly. Frequently, the contrast betw een the str ucture of interest and the sur rounding tissue is not adequate for currently available region growing algorithms to reproducibly se gment soft tissue or gans. In this case, the quasi-manual tools are typically used, where the operator steps through the slices in the volume and draws the organ boundaries, sometimes with the aid of an edgedetecting tool. F igure 11B sho ws a slice with the kidney boundar y def ined. Once the 2-D boundaries ha ve been def ined in each slice (for e xpediency often the boundaries are manually defined in every fifth or tenth slice and an interpolation algorithm is used to fill in the gaps), the or gan volume has been def ined. Figure 11C shows a rendering of the kidne y surf ace follo wing segmentation. Assuming that the voxel size of the reconstructed image is w ell known (this is almost al ways the case), the segmented or gan surf ace ma y be used to calculate the volume of the or gan. If the density of the or gan is known (most soft tissue has a density close to that of water, 1 g/cm 3), then the w eight of the or gan ma y also be determined. In man y cases, the te xture of the image data within the surface may also be analyzed to characterize the health of the organ.67,68 With the use of contrast agents, most soft or gans can be segmented in a whole-body data set. Figure 12 shows a

66

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

Figure 10. Typical whole-mouse micro-computed tomography (CT) images. In each case, the animal was given an intraperitoneal injection of water-soluble iodinated contrast agent before the CT scan. The images were reconstructed from a tomographic data set acquired with the following paramaters: A, 360 projections (256 × 384 pixels), 360 degrees, X-ray with 80 kVp and 500 µA, reconstructed on 150 µm isotropic voxel grid; B, 512 projections (512 × 768 pixels), 360 degrees, X-ray with 80 kVp and 500 µA, reconstructed on a 84 µm voxel grid.

representative data set with multiple organ segmentations. Volumetric image se gmentation is an area of intense research focus for both clinical and preclinical applications. Ne w tools will cer tainly become a vailable to accelerate and to reduce the user inter vention in the segmentation process.

Whole-Body Imaging for PET and SPECT Attenuation Correction PET and SPECT are both emission tomo graphy technologies in w hich the patient or animal is injected with a γ-ray emitting isotope link ed to a tar geting compound designed to bind to a specif ic tissue or process of interest. As the isotope deca ys, the emitted γ-rays are detected and used to reconstr uct image v olumes in a manner similar to that used for micro-CT image

formation. The γ-rays must pass through tissue along varying path lengths before e xiting the subject and, like the X-ra ys used in micro-CT studies, are subject to attenuation. Unlike micro-CT, where the X-ray attenuation is the source of the image data, photon attenuation in PET and SPECT studies are a source of er ror that should be cor rected. Micro-CT pro vides a means for performing this correction. In PET , the γ-ray emitting isotope is actuall y a positron emitter. The positron quickly annihilates with an electron to produce two 511 keV γ-rays traveling in opposite directions. The PET scanner looks for γ-rays detected at the same time (coincident pairs) and assigns a line of response to each coincident pair . Each e vent that contributes to a PET image, therefore, is associated with tw o γ-rays that in combination travel entirely through the animal. The probability that one of the tw o γ-rays in the

Principles of Micro X-ray Computed Tomography

67

A

Figure 12. Segmented whole mouse. The lungs are shown in blue, the kidneys are brown, the heart is red, and the carotid artery in the neck is pink.

B

C

Figure 11. Micro-computed tomography (CT) image of mouse kidneys. A, raw image, (B) single slice segmentation, and (C) segmented kidney volume.

coincident pair will be attenuated is, therefore, the same as the probability that a photon emitted on one side of the animal will be attenuated as it passes through the animal to be detected on the other side. As previously described, during a micro-CT study an X-ray source and detector rotate about the object acquiring 2-D projections at v arious angles around the object. The ra w CT data are then passed to a reconstr uction program that transfor ms the 2-D projections into a 3-D map of the attenuation coef ficients ( µ-map) for the animal. Once the data is reconstr ucted, it is possib le to then forward project the CT image to create an attenuation sinogram or a map of the probability of attenuation for a number o f ph oton p aths t hrough t he a nimal. After correcting for dif ferences between the ener gy of the CT X-rays and the PET γ-rays, a raw PET data set (PET sinogram) may be attenuation cor rected b y scaling the PET sinogram b y the attenuation sino gram. This process is shown schematically in Figure 13. Unlike P ET c oincident p airs, t he p robability a SPECT γ-ray will be attenuated is dependent upon the depth at w hich the γ-ray is emitted. A number of approximate methods exist to partially correct for SPECT γ-ray attenuation, but the most widely used method today is to incor porate measured attenuation coef ficients into the probability matrix of an iterati ve reconstr uction algorithm. Once again, micro-CT data ma y be used to generate this attenuation data. If photon attenuation coef ficients w ere not ener gy dependent, it w ould be possib le to use micro-CT data sets directly to pro vide PET and SPECT attenuation

68

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 13. Schematic diagram showing the data flow for computed tomography (CT)-based positron emission tomography (PET) image reconstruction.

Tissue, Soft (ICRU – 44)

104

m/r men /r

m/r or men /r , cm2/g

103 102 101 100

102 1 102 2 102 3

102 2

102 1 100 Photon Energy. MeV

101

102

Figure 14. Energy dependent attenuation coefficients for photons passing through water.

correction. Unfor tunately, µ-values are strongl y ener gy dependent as sho wn in F igure 14. Preclinical micro-CT images are typically acquired using X-ray source voltages of around 80 kVp, producing an X-ra y flux with a mean energy of appro ximately 30 keV. In contrast, PET isotopes emit γ-rays with energies of 511 keV, while SPECT isotopes emit γ-rays with energies typically ranging from 100 keV to 300 keV. In their seminal w ork on combined PET -CT imaging systems, Kinahan and colleagues68 identified three methods

to cor rect for the ener gy dependence of the attenuation coefficients. The first approach is to simply scale the attenuation coefficient by a cor rection f actor. The scaling approach estimates the attenuation image at 511 k eV b y multiplying the CT image b y the ratio of attenuation coefficients of w ater at CT and PET ener gies. A single “effective” energy is chosen to represent the CT spectr um. This is the easiest method to implement, but the assumption of a linear relationship betw een photon ener gy and attenuation coefficient is only valid over a narrow range of energies. The scaling method tends to yield er roneous values for materials with higher atomic Z v alues, such as bone. The second approach is to segment the image into up to four dif ferent tissue types (typicall y bone, soft tissue, fat, and lung) and then assign kno wn attenuation coef ficients to each tissue type. This approach tends to be more accurate and has the added adv antage of being noiseless, but it can produce ar tifacts at the abr upt boundaries between tissue types. A signif icant problem, however, is that some soft tissue regions will have continuously varying densities that ma y not be accuratel y represented by a discrete set of segmented values, such as, for example, the lungs, where the density varies by as much as 30%. 69,70 The third approach is a hybrid of the preceding methods. The attenuation image at 511 k eV is estimated b y first using a threshold to separate out the bone component of the CT image and then using separate scaling factors for the bone and nonbone components. This

Principles of Micro X-ray Computed Tomography

69

method is easier to implement than the full se gmentation method and yields superior results to the simple scaling method.

Oncology Oncology research is one of the most acti ve f ields using laboratory animal imaging, and micro-CT has been used to measure tumor v olume and as a suppor ting tool for PET and SPECT studies. 70–80 Micro-CT is particularly useful in the study of lung tumors, w here the natural contrast between the air -filled lung and solid mass of the tumor makes it possib le to readil y identify and measure the v olume of developing tumors and to track response to therapeutic compounds.72,75,76 In these studies, respiratory gating is typically used to minimize blur due to respiratory motion. Micro-CT is also a valuable tool for bone tumor studies,73,78 where the natural contrast of sk eletal tissue suppor ts highcontrast imaging. Xeno graft studies also benef it from micro-CT scans w here the high-resolution images pro vide greater measurement accuracy of tumor volume than traditional assays. When used in conjunction with PET or SPECT studies, micro-CT provides anatomic localization of observed radiotracer uptak e79 and a µ-map for attenuation cor rection. The anatomic reference pro vided b y micro-CT has proven to be par ticularly important when highly specific tracers are used; the absence of a general uptake deprives the SPECT or PET image of an intrinsic anatomic reference, necessitating the use of a reference image to accuratel y identify the anatomic locus of uptake (Figure 15). As noted above, it is impor tant to monitor the deli vered dose w hen imaging tumor de velopment and response to therapeutic intervention to avoid confounding the observed results.

Lung Imaging For some time, CT has been a staple in clinical lung imaging due to the natural contrast that exists between the air and the tissue in the lungs. This makes CT an e xcellent choice for e xamining the comple x and inter twined structures of the pulmonar y system. In laborator y animal studies, high-resolution in vitro imaging allo ws an investigator to e xamine the highl y detailed str uctures of excised lung tissue, 80,81 while in vivo imaging allows for detailed imaging of the lung function. 82–84 The versatility of micro-CT gi ves the in vestigator a wide range of control and tools for use in analyzing pulmonary disease and for research into the function of the lung.

Figure 15. Positron emission tomography (PET)-computed tomography (CT) image (courtesy of Dr. Jamey Weichert, University of Wisconsin, Madison).

In vitro micro-CT imaging is typically performed using very high-resolution acquisition protocols. Dedicated in vitro scanners can provide image resolutions on the order of 5 µm. Because the lung or section of the lung has been excised, longer scan times and higher doses are of lesser concern. The X-ray source voltage is typically set between 60 kVp and 70 kVp with minimal f iltration to maximize the low energy component of the X-ra y spectrum, thereby maximizing the absor ption of X-ra ys in the soft tissue structures of the sample. The high-resolution images are useful for vie wing minute str uctures lik e the parench yma and for mapping of the bronchial netw orks of the lung. These images can also be used for studying pulmonar y diseases, such as emphysema and pulmonary fibrosis. In vivo imaging systems can provide images with resolutions between 10 µm and 15 µm in li ving animals and can also pro vide a considerab le deal of insight into the function of the lung. With the use of physiological monitoring systems to pro vide respiratory gating signals to the scanner, it is possible to image the mechanics of the

70

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

lung as the motion mo ves from full inspiration to full expiration. Although in vi vo imaging studies generall y use protocols that deli ver lo wer doses than in vitro systems, the dose is still high with v alues ranging from 15 cGy85 to 1 Gy83 depending on the desired resolution. Several methods of respirator y gating are typicall y used during in vi vo studies. The simplest method is to detect the respirator y c ycle of an anesthetized animal using a pneumatic sensor 83,86,87 or an optical or infrared sensor.88 The respirator y gating signal is used to trigger the projection acquisition, significantly reducing artifacts due to respiratory motion. This method has the advantage of ease of implementation but typically does not remove all of the motion-related blurring. A second method is to acquire a relati vely large number of ungated projections and then to use image analysis to retrospectively sor t the projections into se veral compar tments associated with dif ferent phases of the respirator y cycle.84,86 This method is the simplest to implement because it requires no gating apparatus during the scan, but it typically does not remove all of the motion-related blurring and it car ries with it the burden of e xcess dose to the animal received during the acquisition of unused projections. The third and largely preferred method is to place the animal on a v entilator and control the respirator y cycle.82–85 By imposing a breath-hold during the acquisition of each projection, blurring due to respiratory motion is vir tually eliminated , enab ling researchers to tak e full adv antage of the resolution of micro-CT scanners (Figure 16).

A

B

Bone Imaging Due to the high relati ve density of sk eletal tissue, bone imaging has long been a key application for high-resolution micro-CT.13,89–92 Key areas of interest include the direct measurement of bone mineral density,93 analysis of trabecular bone str ucture and calculation of trabecular mor phometric parameters, 90 tracking and anal ysis of bone tumor formation,78 and evaluation of bone implant materials. 94 Bone mineral density is typically measured by calibrating the scanner using a tissue equivalent phantom with calibrated reference materials. The measured CT numbers associated with a range of reference materials are recorded and the image data are then calibrated using the derived scale factor. In trabecular bone studies, the trabecular netw ork is typicall y se gmented from the sur rounding cor tical bone before anal ysis. The trabecular lattice is then numerically anal yzed to yield a set of mor phometric parameters including relati ve bone v olume, bone v olume/tissue volume, bone surface to volume ratio, bone

Figure 16. Micro-computed tomography (CT) lung images: A, micro-CT image of an excised mouse lung and B, a transaxial image from an in vivo micro-CT study (courtesy of Dr. Eric Hoffman, University of Iowa).

Principles of Micro X-ray Computed Tomography

surface area/bone v olume, trabecular w all thickness, trabecular wall spacing, trabecular number, and trabecular patter n f actor. These numerical f actors ser ve as indicators of bone quality (Figure 17). Bone metastases are associated with a number of cancers, breast cancer in par ticular, and are readil y detected as regions of lower density in micro-CT studies. Micro-CT has been used to detect and measure the progression of bone metastases in mouse models. 78,95 Optimization of the interface between bone tissue and implants is an area of increasing focus as joint replacements become widel y used in clinical care. The most widely used implant material, titanium, is suf ficiently radiolucent to be studied using micro-CT . A number of groups have used micro-CT to investigate animal response to sur gical methods, bone cement for mulations and implant materials in animal models. 94,96,97 A representative titanium implant study is shown in Figure 18.

FUTURE DIRECTIONS As micro-CT systems are becoming more of a standard tool used in the f ields of genetic, phar maceutical, and disease research, users w ant to use them to routinel y scan lar ge numbers of animals as quickly as possible. This will require new micro-CT developments to enable rapid scanning. The major components that limit the speed at which one can scan using micro-CT are (1) the X-ray source, (2) the X-ray detector, and (3) the rotating gantry. In addition, researchers

A

Figure 17.

71

would lik e to ha ve the ability to perfor m dynamic CT studies on animals, but the relatively slow speed of microCT as compared with the rapid rate of ph ysiologic processes in mice make high resolution (< 50 µm), dynamic micro-CT a challenge to achieve. There is a minimum dw ell time required for each projection acquired during a micro-CT scan to allo w enough X-ra y signal to reach the inte grating X-ray detector . If the po wer of the X-ra y source (and hence the X-ra y flux) can be increased , then this dw ell time can be reduced , and hence, reduce the o verall scan time. There are very powerful X-ray sources available (ie, those used in clinical CT scanners), but the challenge is to increase the po wer while maintaining a small X-ra y focal spot (eg, < 50 µm) to preserve the required micro-CT spatial resolution. Ne w X-ra y source de velopments that increase po wer w hile maintaining small focal spot size will benef it the micro-CT user . Lastl y, pulsed X-ra y sources (e g, cold cathode) impro ve one’s ability to perform gated micro-CT studies w here the hear tbeat or the respiration of the subject is used to trigger the image acquisition at a precise point in the cardiac or respirator y cycle. If the X-ray source can be pulsed (rapidly turned on and off), then there is no longer a need to use a mechanical shutter to control the timing of the X-ra y exposure. Another component that limits overall micro-CT scan time is the readout speed of the X-ra y detector. For example, a 4096 × 4096 CCD-based X-ray detector can take > 1 second to complete the full data readout, which can account

B

Trabecular Structure; in the head of a mouse femur (A) and a vertical section along the length of a mouse femur (B).

72

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

Figure 18. Titanium implant study. Micro-computed tomography (CT) volume rendering (left) and CT slice (right) of a titanium implant in the femur of a mouse.

for substantial time if the micro-CT study has hundreds of projections. If someone is interested in high-resolution studies over a large FOV, then a detector with a large number of imaging elements (pixels) is required, and detector readout can become a significant factor affecting overall scan time. Ne w detectors with f ast, lo w-noise, multipor t readout will improve the speed at which a micro-CT system can scan the subject at high resolution. If one can impro ve the X-ra y source and detector as outlined, then the X-ra y source and detector could be rapidly rotated around the subject while collecting highquality projections. At this point, the limiting f actor in micro-CT scan time can become the speed at w hich the rotating gantry can spin. High-resolution micro-CT systems with slip ring technolo gy would then become a more critical development in this f ield. Slip rings are used in clinical CT scanners because clinical systems require that the gantry be rotated at high speed through man y 360-degree orbits during a study . A slip ring system allo ws the gantr y to continuously rotate because it per mits removal of hardwired cables that can limit the total rotation of the micro-CT gantry. The slip ring interf ace provides a connection path for both power and data transfer, so cables are not required. Once the micro-CT system has a high po wer, highresolution X-ra y source, a f ast detector , and a slip ring

gantry architecture, spiral CT scanning of animals becomes possible. High-speed spiral CT has man y advantages over circular orbit cone-beam CT in ter ms of impro ved image quality and the ability to capture dynamic events in the body. Lastly, the issue of data management in micro-CT is a current challenge. A single micro-CT study of a mouse with 50 µm resolution will create an image volume of ~2 GB. In a high-throughput situation where 20 mice may be scanned in a day, 40 GB of data will be generated per day. After 1 month of scanning, the user has piled up 1200 GB of image data. This creates a substantial data storage problem, but even more importantly, a tremendous data management problem. Researchers w ould lik e to be ab le to quickl y search their micro-CT image database to f ind images of specific mice, images of multiple mice with similar anatomic features, or a collection of longitudinal studies on a single animal, just to name a few examples. The data management tools used by most micro-CT researchers fall short of providing such capabilities. Note that these data management challenges are not limited to the CT modality but are compounded w hen a user also has micro v ersions of PET, MR, and SPECT, for example, and needs to manage multiple image data modalities across their animal population. New software applications are needed that provide efficient ways to store and retrieve useful information from

Principles of Micro X-ray Computed Tomography

the potentially massive micro-CT (PET, MRI, SPECT, etc) image database. Such an application will require innovative developments in automated image analysis, feature extraction, classif ication and image database inde xing strategies for rapid search, and retrieval capabilities.

REFERENCES 1. Cormack A. Computed tomo graphy: some histor y and recent de velopments. J Appl Phys 1963;34:2722–7. 2. Hounsfield G. Diagnosis and management decisions in infections of the deep fascial spaces of the head and neck utilizing computerized tomography. Br J Radiol 1973;46:1016–22. 3. Kujoory MA, Hillman BJ , Bar rett HH. A low-cost computed tomographic scanner. Invest Radiol 1980;15:148–54. 4. Burstein P, Bjorkholm RC, Chase RC, Se guin FH. The lar gest and smallest X-ra y computed tomo graphy systems. Nucl Instr um Methods Phys Res 1984;221:207–12. 5. Miller GG, Dawson DT, Battista JJ. Compute tomographic assessment of radiation induced damage in the lung of nor mal and wr 2721 protected laf1 mice. Int J Radiat Oncol Biol Phys 1986;12:1971–5. 6. Flannery BP , Deckman HW , Rober ge WG, D'Amico KL. Threedimensional X-ray microtomography. Science 1987;237:1439–44. 7. Leyton MW, Goldstein SA, Goulet RW, et al. Examination of subchondral bone architecture in experimental osteoarthritis by microscopic computed axial tomography. Arthritis Rheum 1988;31:1400–5. 8. Feldkamp LA, Goldstein SA, P arfitt AM, et al. The direct e xamination of three-dimensional bone architecture in vitro b y computed tomography. J Bone Miner Res 1989;4:3–11. 9. Kuhn JL, Goldstein SA, F eldkamp LA, et al. Ev aluation of a microcomputed tomography system to study trabecular bone str ucture. J Orthop Res 1990;8:833–42. 10. Feldkamp L, Da vis LC, Kress JW . Practical cone-beam algorithm. J Opt Soc Am 1984; Issue 6, pp. 612–19. 11. Paulus M, Sari-Sar raf H, Gleason SS, et al. A new X-ray computed tomography system for laborator y mouse imaging. IEEE Trans Nucl Sci 1999;3:558–64. 12. Paulus MJ , Gleason SS, K ennel SJ , et al. High resolution X-ray computed tomo graphy: an emer ging tool for small animal cancer research. Neoplasia 2000;2:62–70. 13. Rüegsegger P, Koller B, Müller R. A microtomographic system for the nondestructive evaluation of bone architecture. Calcif Tissue Int 1996;58:24–9. 14. Stenström M, Olander B , Carlsson CA, et al. The use of computed microtomography to monitor mor phological changes in small animals. Appl Radiat Isot 1998;49:565–70. 15. Stenström MOB, Carlsson CA, Carlsson GA, Håkanson R. Methodologic aspects of computed microtomo graphy to monitor the development of osteoporosis in gastrectomized rats. Acad Radiol 1995;2:785–91. 16. Müller R, K oller B, Hildebrand T, et al. Relationship betw een twodimensional and three-dimensional bone architecture in predicting the mechanical strength of the pig mandib le. Technol Health Care 1996;4:113–9. 17. Müller RR, Rüe gsegger P. Micro-tomographic imaging for the nondestructive evaluation of trabecular bone architecture. Stud Health Technol Inform 1997;40:61–79. 18. Johnson RH, Hu H, Ha worth ST, et al. Feldkamp and circle-and-line cone-beam reconstruction for 3-d micro-CT of vascular networks. Phys Med Biol 1998;43:929–40. 19. Corrigan NM, Chavez AE, Wisner ER, Boone JM. A multiple detector ar ray helical X-ra y microtomo graphy system for specimen imaging. Med Phys 1999;26:1708–13.

73

20. Boone JM, Alexander GM, Seiber t J A. A fluoroscop y-based computed-tomography scanner for small specimen research. Invest Radiol 1993;28:539–44. 21. Anderson P, Levinkind M, Elliot JC. In vitro of mechanisms of de- and remineralization of dental enamel. Arch Oral Biol 1998;43:649–56. 22. Reid BK, Cunningham IA. High resolution X-ra y imaging at soft diagnostic ener gies using a silicon photodiode ar ray. Med Ph ys 1993;20:1497–504. 23. Holdsworth DW, Drangova M, Fenster A. A high-resolution xrii-based quantitative volume ct scanner. Med Phys 1993;20:449–62. 24. Holdsworth DW, Drango va M, Schulenber g K, F enster A. A tab letop ct system for high-resolution v olume imaging. SPIE 1990; 1231:239–45. 25. Flynn MJ, Hames SM, Reimann D A, Wilderman SJ. Microfocus X-ray sources for 3-d microtomo graphy. Nucl Instr umMethods Phys Res Sect A 1994;353:312–5. 26. Sprawls P . The ph ysical principles of medical imaging, 2nd ed. Rockville: Aspen publications; 1993, Vol 2, Madison, WI: Medical Physics Publishing; 2000. 27. Cheng Y, Zhang J, Lee YZ, et al. Dynamic X-ray radiography using a carbon nanotube f ield emission X-ra y source. Re v Sci Instr um 2004;10:3264. 28. Zhang J, Yang G, Cheng Y, et al. Stationar y scanning X-ra y source based on carbon nanotube f ield emitters. Appl Ph ys Lett 2005;86:184104–6100. 29. Liu Z, Yang G, Lee Yueh Z, et al. Carbon nanotube based microfocus field emission X-ray source for microcomputed tomography. Appl Phys Lett 2006;89:103–11. 30. Grodzins L. Optimum ener gies for X-ra y transmission tomo graphy of small samples: applications of synchrotron radiation to computerized tomo graphy. Nucl Instrum Methods Phys Res 1983; 206:541–5. 31. Stonestrom JP, Alvarez RE, Maco vski A. A frame work for spectral artifact cor rections in X-ra y CT . IEEE Trans Biomed Eng 1981;28:128–41. 32. Duerinckx AJM, Macovski A. Polychromatic streak ar tifacts in computed tomography images. J Comput Assist Tomogr 1978;2:481–7. 33. Tuy HK. An inversion formula for cone-beam reconstr uction. SIAM J Appl Math 1983;43:546–52. 34. Yan M, Zhang C. Tilted plane feldkamp type reconstruction algorithm for spiral cone beam ct. Med Ph ys 2005;32:3455–67. 35. Defrise M, Noo F, Kudo H. A solution to the long-object prob lem in helical cone-beam tomography. Phys Med Biol 2000;45:623–43. 36. Liu V, Lariviere NR, Wang G. X-ray micro-CT with a displaced detector ar ray: application to helical cone-beam reconstr uction. Med Phys 2003;30:2758–61. 37. Yu H, Wang G. Studies on implementation of the katsevich algorithm for spiral cone-beam ct. J Xray Sci Technol 2004;12:97–116. 38. Noo F, Pack J, Heuscher D. Exact helical reconstruction using native cone-beam geometries. Phys Med Biol 2003;48:3787–818. 39. Noo F, Hoppe S, Dennerlein F, et al. A new scheme for view-dependent data differentiation in f an-beam and cone-beam computed tomo graphy. Phys Med Biol 2007;52:5393–414. 40. Pack JD, Noo F. Cone-beam reconstr uction using 1 d f iltering along the projection of-lines. Inverse Probl 2005;21:1105–20. 41. Zabic S, Hoppe S, Dennerlein F, et al. Resampling density v alues on r-lines into density v alues on a car tesian g rid. IEEE 2006; Proceedings of the IEEE Nuclear Science Symposium, M06–449. 42. Yang H, Li M, Koizumi K, Kudo H. Exact cone beam reconstr uction for a saddle trajectory. Phys Med Biol 2006;51:1157–72. 43. Herman G. Image reconstructions from projections: the fundamentals of computerized tomography. New York: Academic Press; 1980. 44. Xia D, Yu L, Sidky EY, et al. Noise proper ties of chord-image reconstruction. IEEE Trans Med Imaging 2007;26:1328–44. 45. Lee S, Cho G, Wang G. Artifacts associated with implementation of the grangeat formula. Med Phys 2002;29:2871–80.

74

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

46. Katsevich A. Theoretically exact filtered backprojection-type inversion algorithm for spiral ct. SIAM J Appl Math 2002;62:2012–26. 47. Grangeat P. Mathematical framwork of cone beam 3-D reconstruction via the first derivative of the radon transform. In: Louis AK, Natterer F, editors. Mathematical methods in tomo graphy. Vol 1497. Berlin, Germany: Springer; 1991. p. 66–97. 48. Lee SW, Cho G, Wang G. Artifacts associated with implementation of the grangeat formula. Med Phys 2002;29:2871–80. 49. Deng J, Yu H, Ni J , et al. A parallel implementation of the katse vich algorithm for 3-d CT image reconstruction. J Super Comput 2006; 38:35–47. 50. Shepp LA, Vardi Y. Maximum lik elihood reconstr uction in positron emission tomography. IEEE Trans Med Imaging 1982;1:113–22. 51. Benson TM, Gregor J. Framework for iterative cone-beam micro-CT reconstruction. IEEE Trans Nucl Sci 2005;52:1335–40. 52. Elbakri IA, Zhang Y, Chen L, et al. Statistical reconstr uction for quantitative ct applications. IEEE Nucl Sci Symp 2003; 4:2978–80. 53. Gregor J, Gleason SS, P aulus MJ, Cates J. Fast feldkamp reconstruction based on focus of attention and distributed computing. Int J Imaging Syst Technol 2002;12:229–34. 54. Stsepankou D, Kornmesser K, Hesser J, Manner R. FPGA-acceleration of cone-beam reconstruction for the X-ray CT. In: IEEE international conference on f ield-programmable technolo gy. 2004. p. 327–30. 55. Xu F and Mueller K. GPU-Real-T ime 3D Computed Tomographic Reconstruction Using Commodity Graphics Hardware, Physics in Medicine and Biology, 2007;52:3405–3419. 56. Okitsu Y, Ino F , and Hagihara K. Accelerating Cone Beam Reconstruction Using the CUD A-Enabled GPU , HiPC 2008, LNCS 5374, P. Sadayappan et al. (Eds.), Springer -Verlag Berlin Heidelberg, 2008, pp. 108–19. 57. Sharp GC, Kandasamy N, Singh H, F olkert M. GPU-based architectures for CT reconstruction and registration. Phys Med Biol 2007; 52:5771–83. 58. Zeng K, Bai E, and Wang G. A Fast CT Reconstruction Scheme for a General Multi-Core PC, Inter national Jour nal of Biomedical Imaging, 2007. 59. Flecknell PA. Anaesthesia of animals for biochemical research. Br J Anaesth 1993;71:885–94. 60. Flecknell PA. Laboratory animal anaesthesia. San Die go, CA: Academic Press; 1996. 61. Gardner D , Da vis J A, Weina PJ , Theune B . Comparison of tribromoethanol, ketamine/acetylpromazine, telazol/xylazine, pentobarbital, and metho xyflurane anesthesia in hsd:Icr mice. Lab Anim Sci 1995;45:199–204. 62. Obenaus A, Smith A. Radiation dose in rodent tissues during microCT imaging. J Xray Sci Technol 2004;12:241–9. 63. Taschereau R, Chow C, Chatziioannou AF. Current status and new horizons in monte carlo simulations of dose from micro-CT imaging procedures in a realistic mouse phantom. Med Ph ys 2006; 33:216–24. 64. De Clerck N, Meurrens K, Weiler H, et al. In vivo high-resolution X-ray microtomography for liver and spleen tumor assessment in mice. Neoplasia 2004;6:374–9. 65. Boone J , Velazquez O , Cher ry SR. Small-animal X-ra y dose from micro-CT. Mol Imaging 2004;3:149–58. 66. Carlson S, Classic KL, Bender CE, Russell SJ. Small animal absorbed radiation dose from serial micro-computed tomography imaging. Mol Imaging Biol 2007;2:78–82. 67. Gleason SS, Sari-Sarraf H, Abidi MA, et al. A new deformable model for analysis of X-ray CT images in preclinical studies of mice for polycystic kidne y disease. IEEE Trans Med Imaging 2002; 10:1302–9. 68. Kinahan P E, Townsend D W, B eyer T, S ashin D . Attenuation correction for a combined 3-d pet/ct scanner . Med Ph ys 1998; 10:2046–53.

69. Robinson P, Kreel L. Pulmonar y tissue attenuation with computed tomography: comparison of inspiration and expiration scans. J Comput Assist Tomogr 1979;3:740–8. 70. Weichert JP, Moser AR, Weber SM, et al. Radioiodinated NM404-a universal tumor imaging agent? Acad Radiol 2005; Issue 5, p. S58–S59. 71. Badea CT, Hedlund LW, De Lin M, et al. Tumor imaging in small animals with a combined micro-CT/micro-DSA system using iodinated con ventional and b lood pool contrast agents. Contrast Media Mol Imaging 2006;1:153–64. 72. Chang C, Jan ML, F an KH, et al. Longitudinal e valuation of tumor metastasis by an fdg-micropet/microct dual-imaging modality in a lung carcinoma-bearing mouse model. Anticancer Res 2006; 26:159–66. 73. Mouchess M, Sohara Y, Nelson MD Jr , et al. Multimodal imaging analysis of tumor pro gression and bone resor ption in a murine cancer model. J Comput Assist Tomogr 2006;30:525–34. 74. Pickhardt PJ, Halberg RB, Taylor AJ, et al. Microcomputed tomography colonography for polyp detection in an in vivo mouse tumour model. Proc Natl Acad Sci U S A 2005;102:3419–22. 75. Cody DD, Nelson CL, Bradle y WM, et al. Murine lung tumor measurement using respirator ygated micro-computed tomo graphy. Invest Radiol 2005;40:263–9. 76. Kennel SJ , Da vis IA, Branning J , et al. High resolution computed tomography and MRI for monitoring lung tumor g rowth in mice undergoing radioimmunotherapy: correlation with histology. Med Phys 2000;27:1101–7. 77. Weber SM, P eterson KA, Durk ee B, et al. Imaging of murine li ver tumor using micro-CT with a hepatocyte-selective contrast agent: accuracy is dependent on adequate contrast enhancement. J Sur g Res 2004;119:41–5. 78. Winkelmann CT, Figueroa SD, Rold TL, et al. Microimaging characterization of a b16-f10 melanoma metastasis mouse model. Mol Imaging 2006;5:105–14. 79. Chow PL, Stout DB , Komisopoulou E, Chatziioannou AF. A method of image re gistration for small animal, multi-modality imaging. Phys Med Biol 2006;51:379–90. 80. Langheinrich AC, Leithäuser B , Greschus S, et al. Acute rat lung injury: feasibility of assessment with micro-CT. Radiology 2004; 233:165–71. 81. Sera T, Uesugi K, Yagi N. Localized mor phometric defor mations of small airways and alveoli in intact mouse lungs under quasi-static inflation. Respir Physiol Neurobiol 2005;147:51–63. 82. Hoffman E, Chon D . Computed tomo graphy studies of lung v entilation and perfusion. The Proceedings of the American Thoracic Society 2005. pp. 492–98. 83. Namati E, Chon D, Thiesse J, et al. In vivo micro-CT lung imaging via a computer-controlled intermittent iso-pressure breath hold (IIBH) technique. Phys Med Biol 2006;51:6061–75. 84. Badea C, Hedlund LW, Johnson GA. Micro-CT with respirator y and cardiac gating. Med Phys 2004;31:3324–9. 85. Cavanaugh D, Johnson E, Price RE, et al. In vi vo respirator y-gated micro-CT imaging in small-animal oncolo gy models. Mol Imaging 2004;3:55–62. 86. Ford NL, Nik olov HN , Norle y CJ , et al. Prospecti ve respirator ygated micro-CT of free breathing rodents. Med Ph ys 2005; 32:2888–98. 87. Hsu LL, Schimel DM. Computed tomo graphy imaging of lungs in mouse models of human disease—advancing the computing interfaces with physiology. In: 17th IEEE symposium on computerbased medical systems (CBMS'04). 2004. p. 385. 88. Gum F, Agaoglu D, Schild R, et al. Phantom v erification of respiratory gating system for lung and liver. Int J Radiat Oncol Biol Phys 2006;66:S605. 89. Hildebrand T, Rüegsegger P. A new method for the model-independent assessment of thickness in three-dimensional images. J Microsc 1997;185:67–75.

Principles of Micro X-ray Computed Tomography

90. Laib A, Barou O , Vico L, et al. 3-d micro-computed tomo graphy of trabecular and cortical bone architecture with application to a rat model of immobilisation osteoporosis. Med Biol Eng Comput 2000;38:326–32. 91. Engelke K, Karolczak M, LutzA, et al. Micro-CT: technology and applications for assessing bone str ucture. Radiologe 1999;39:203–12. 92. Bagi CM, Hanson N , Andresen C, et al. The use of micro-CT to evaluate cortical bone geometry and strength in nude rats: correlation with mechanical testing, pqct and dxa. Bone 2006; 38:136–44. 93. Schmidt C, Priemel M, K ohler T, et al. Precision and accurac y of peripheral quantitative computed tomography (pqct) in the mouse skeleton compared with histolo gy and microcomputed tomo graphy (micro-CT). J Bone Miner Res 2003;18:1486–96. 94. Akca K, Sarac E, Baysal U, et al. Micro-morphologic changes around biophysically-stimulated titanium implants in ovariectomized rats, head & face medicine. Head Face Med 2007;3.

75

95. Li X, Zanzonico P, Ling CC, O'Dono ghue J. Visualization of experimental lung and bone metastases in li ve nude mice b y X-ra y micro-computed tomo graphy. Technol Cancer Res Treat 2006; 5:147–55. 96. Xing Z, Hasty KA, Smith RA. Administration of pamidronate alters bone-titanium attachent in the presence of endotoxin-coated polyethylene par ticles. J Biomed Mater Res B Appl Biomater 2007;83:354–8. 97. Otsuki B, Takemoto M, Fujibayashi S, et al. Pore throat size and connectivity deter mine bone and tissue ing rowth into porous implants: three-dimensional micro-CT based str uctural anal yses of porous bioacti ve titanium implants. Biomaterials 2006; 27:5892–900. 98. Gleason, S., Osborne, D. Study of CT-Based Attenuation Correction Factors for microPET using kVp Dependent HU Calibration. Joint Molecular Imaging Conference: Pro vidence, RI 2007; 07-A-1153-SMI

6 SMALL ANIMAL SPECT, SPECT/CT, SPECT/MRI

AND

NEAL H. CLINTHORNE, PHD AND LING-JIAN MENG, PHD

Single photon emission computed tomo graphy (SPECT) of small animals is becoming an increasingl y impor tant tool in biomedical research. As in positron emission tomography (PET), it is a functional imaging technique that for ms a three-dimensional image of the distrib ution of a radiotracer injected into a small animal.And like PET, the trend has been to ward the de velopment of SPECT instruments that combine either X-ray computed tomography (CT) or magnetic resonance imaging (MRI). In contrast to PET, however, counting efficiency is generally lower requiring longer imaging times or higher injected doses, w hereas spatial resolution—depending on the needs of the e xperiment—can either be w orse or signif icantly better than PET. Although it has often been claimed that PET is quantitati ve while SPECT is not, there is no reason that SPECT images cannot be quantitati ve if appropriate calibrations and cor rections are made. Moreover, at resolutions in the deep submillimeter re gime, SPECT likely has the advantage over PET in quantitative accuracy. PET images are b lurred b y the f act that positrons tra vel some distance before annihilating, and although “resolution reco very” methods can be used to reduce the resulting par tial v olume ef fects (see Section “Accounting for Degradations: Advanced SPECT Reconstruction”), the y signif icantly increase noise in reconstructed images. A key advantage is that SPECT tends to be easier for many laboratories to integrate. Although many dedicated small animal SPECT instruments are commercially available, often gamma cameras for clinical imaging that have been decommissioned from the hospital can be modif ied for small animal SPECT . Fur thermore, preparation of SPECT tracers is typicall y simpler than PET . Man y radiotracers used in patient imaging are readil y usab le in small animals, eliminating the need for adv anced 76

radiochemical synthesis and the c yclotron typicall y required in PET. The pur pose of this chapter is to pro vide an overview of small animal SPECT applications and instrumentation. Because of the e xplosion of both research and commercial acti vity in SPECT , SPECT/CT, and SPECT/MRI devices for small animals, this chapter has been organized to provide some history and a basic o verview of applications and commercial instruments followed by a description of basic hardware and image for mation methods in SPECT and CT . Design trends in small animal SPECT de vices are then discussed, and this chapter concludes with a brief discussion of future directions in the f ield. F inally, in contrast to clinical SPECT imaging, it is impor tant to note that preclinical SPECT is f ar from an “old” or mature technolo gy; no vel instr umentation and techniques are developing at a rapid pace. Particularly exciting are single photon emission microscop y (SPEM) methods for achieving spatial resolutions of 250 µm and better at useful sensitivity. Many reviews of small animal SPECT applications and instr umentation have been pub lished in the last fe w years.1–6 An excellent treatment of small animal SPECT principles, instrumentation, and methods can be found in the book by Kupinski and Barrett,7 while basic principles of the ph ysics of radionuclide imaging are described in Physics in Nuc lear Medicine.8 More advanced topics on emission tomo graphy are discussed in the book b y Wernick and Aarsvold.9

HISTORY Animal imaging using SPECT has e xisted as long as the modality itself. The Humongotron, an earl y

Small Animal SPECT, SPECT/CT, and SPECT/MRI

SPECT instr ument for patients constr ucted using a standard clinical Anger camera mounted to an old cesium radiotherapy machine, w as used in the v alidation of cardiac SPECT for human subjects through do g heart imaging experiments.10 At nearly the same time, a similar study w as conducted b y Jaszczak and colleagues on the Searle Radio graphics prototype SPECT system.11 Spatial resolution for both de vices was probably close to 10 mm full width at half maximum (FWHM). Soon thereafter , a do g hear t inf arct model was used to v alidate a 72-pinhole time-modulated coded aper ture for human cardiac imaging. 12 Resolution of the device ranged from 3.8 mm FWHM at 4 cm from the aperture to 7.8 mm at 12 cm. The instrument, however, suf fered from the ar tifacts of limited-angle tomography (see Section “Image Formation and Reconstruction in SPECT and CT”). In a study of the temporomandibular joint with bone imaging agent 99m Tc-MDP in rhesus macaques, a single-slice prototype tomograph (SPRINT) for human brain imaging 13 was outf itted with a special imaging aper ture that reduced the transaxial f ield-of-view (FO V) to 10 cm while improving spatial resolution to 4.5 mm FWHM. In the quest for higher spatial resolution for imaging 123 I and 131I tumor x enografts in mice and rats with labeled monoclonal antibodies, an animal SPECT instrument w as de veloped based on the design of the SPRINT II. 14 The imaging aper ture allo wed 1 mm FWHM resolution for 99mTc labeled tracers and 2 mm FWHM for 131I. It w as e ven used briefl y for imaging positron emitting nuclides at high resolution for dosimetry studies in rats prior to the a vailability of modern small animal PET scanners. 15 The “modern” era of small animal SPECT star ted in the earl y 1990s with the introduction of clinical SPECT cameras modified to use pinhole aper tures at high magnifications. The adv antage of this approach, w hich is described in more detail in Section “Basic Principles,” was that the modest intrinsic resolution of the clinical SPECT camera could be de-emphasized and SPECT resolution of 1 to 2 mm FWHM and better w as easily attainable over a small FOV.16,17 The general approach has been extended to provide spatial resolutions toda y that approach 100 µm. Nevertheless, with a lack of w ell-developed small animal models of human disease at the time, interest in small animal SPECT faded. In the past decade, due in part to development of better small animal models (e g, gene knock-ins and knock-outs in mice) and new SPECT radiotracers, and in par t to the de velopment of better SPECT instr umentation, small animal SPECT has gained signif icant momentum and interest is rapidly increasing.

77

RADIONUCLIDES, RADIOTRACERS, AND APPLICATIONS Radiotracers and Applications Numerous radiotracers ha ve been used in small animal SPECT (primaril y in murine models) for applications ranging from central ner vous system (CNS) imaging 18–32 to hear t imaging, 33–40 oncology,41–45 stem cell tracking,36,40,46–49 and others. 50–56 As noted abo ve, one of the advantages of SPECT is that many radiolabeling “kits” are available for clinical SPECT imaging that cross over to use in animal imaging, eliminating the need for an on-site cyclotron and reducing or eliminating the need for radiochemists required in PET. Table 1 lists common SPECT radionuclides along with their half-lives, emitted photon ener gies, and sample radiotracers. By f ar the most common is the transition metal 99mTc, which is obtained in the for m of TcO4− ions through elution of a 99Mo generator.8 Its 140 k eV photon emission and absence of higher ener gy photons is nearl y ideal for SPECT image formation. Because of these favorable properties, it has been used to label a wide v ariety of radiotracers such as MIBI for m yocardial b lood flo w,33 MDP for bone, HMP AO (e xametazime) for cerebral blood flow, and Annexin V for apoptosis. 20 For reference, Figure 1 is a high-resolution 99mTc-MDP bone scan of a mouse, while Figure 2 shows gated tomographic short- and long-axis slices of myocardial blood flow in a mouse heart at end-systole and end-diastole using 99mTc-Sestamibi. Many times, the 6-hour half-life of 99mTc is too shor t relative to the time required to accumulate suf ficient tracer in tissue or to achie ve maximum tar get-to-background ratios. This is often the case with man y labeled peptides and monoclonal antibodies. Typically, 111In with its 2.8-da y half-life is used for labeling these tracers; however, 111In chelation is more challenging as w ell as less stable in vivo, and its higher energy photon emissions are less suitab le for high-resolution SPECT imaging. Despite these limitations, its use in SPECT is widespread. Several iodine isotopes are also widel y used in SPECT. A variety of 125I and 131I labeled compounds are available commercially (eg, see Web sites of P erkinElmer, GE Healthcare, or Biomedical Technologies). Their long half-li ves (59 and 8 da ys, respecti vely) are often adv antageous, but the emission ener gy of 131I is poorly suited to high-resolution SPECT, while that of 125I is not w ell suited for imaging animals much lar ger than mice (or rats for some applications). Ne vertheless, because of the number of a vailable compounds and the fact that very high spatial resolution is possible using 125I,

78

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 1. COMMON SINGLE-PHOTON RADIONUCLIDES, CHARACTERISTICS, AND APPLICATIONS Isotope

Half-Life

Photon Energy (% Abundance)

Notes and Applications

99m

Tc

6.02 h

140 keV (89.1)

Most commonly used SPECT isotope. Generator produced and readily available in nuclear medicine clinics. Numerous radiotracer kits available including MIBI for myocardial perfusion and some tumors, MAG3 for renal scanning, HMPAO for brain blood flow, MDP for bone imaging. Autologous RBC labeling for blood-pool imaging. Labeled Bombesin for tumors and Annexin V for apoptosis. The 140 keV emission and absence of higher energy photons are nearly ideal for SPECT.

201

Tl

3.04 d

68–82 keV Hg X-rays, 167 keV (10.0)

Thallous chloride for myocardial perfusion—largely replaced by MIBI above. Sometimes useful in dual-isotope studies with 99mTc or 123I.

111

2.8 d

171 keV (90.7), 245 keV (94.1)

Monoclonal antibody, protein and peptide labeling, RBC labeling. General cell labeling for tracking.

123

13.3 h

125

59 d

159 keV (83.3), 528 keV (1.4) 27–32 keV Te X-rays, 35.5 keV (6.7) 364 keV (81.5), 637 keV (7.2), 723 keV (1.8)

All iodine isotopes have been used to label a wide variety of molecules such as peptides, monoclonal antibodies, and brain imaging agents such as iodoamphetamine and iodobenzamide. They have also been used as I− for assessing thyroid function (or for radioablation of thyroid tissue in the case of 131I). Clinically, 123I is most often used for diagnostic imaging because of its favorable photon energy, whereas 131I is used for internal radiotherapy applications due to its corresponding β emissions; its high photon energies are not particularly well suited to SPECT. Even the small—abundance of 637 and 723 keV photons significantly degrade imaging performance over 364 keV photons alone. A variety of 125I labeled compounds are commercially available. The low-energy photons of 125 I are readily absorbed by even a few centimeters of tissue but are easily imaged, and research SPECT instrumentation is capable of better than 100 µm resolution.

93 keV (35.7), 185 keV (19.7), 300 keV (16.0)

67

In I I

131

I

8.02 d

67

Ga

3.3 d

Ga-citrate—Hodgkin’s disease, lymphomas, etc. Some acute inflammatory lesions. Presence of 300 keV emission compromises ability to obtain high resolution.

SPECT = single photon emission computed tomography; RBC = red blood cell.

it is likely to become an increasingl y important radionuclide for small animal SPECT . At 159 k eV, the photon energy of 123I is better but relati vely fe wer compounds, such as 123I-IBZM and 123I-FP-CIT, are readily obtainable commercially.28 Of course, tracer synthesis and radioiodination can also be perfor med on site in an appropriatel y equipped radiochemistr y f acility, b ut these resources are likely unavailable in most small animal SPECT installations. One feature of SPECT is that multiple radiotracers can be imaged simultaneousl y and their distributions reconstructed indi vidually if the tracers are labeled with radionuclides emitting dif ferent ener gy photons. Figure 3 sho ws a reconstr ucted and rendered image using 201Tl (~80 keV, rendered as green) for myocardial perfusion, 99mTc-MDP (140 k eV, orange) for bone imaging, and 123I− (159 k eV, b lue) for th yroid uptak e. The SPECT images ha ve also been fused with a CT image of the mouse. Performance in accomplishing isotope separation depends upon the ener gy resolution of

the camera system. In this case, a cadmium zinc telluride (CZT) detector ha ving 4.5% FWHM ener gy resolution at 140 keV was used to achieve good separation between the 167 k eV emission of 201Tl and the 159 keV emission of 123I. Although the choice of radiolabel and tracer is important, tw o additional issues must be considered in small animal SPECT. The f irst is the mass of compound injected. In human imaging, a trace quantity of the compound is injected that does not e xhibit a phar macologic effect, w hereas a similar quantity injected into a rat or mouse may well have an effect.57 The second and related issue is radiation dose to the animal. Funk and colleagues estimated that w hole body radiation dose to mice for many SPECT studies is a signif icant fraction of the LD50/30 (50% mortality in 30 days) and high enough to result in increased gene e xpression.58 This is especiall y relevant if longitudinal SPECT imaging studies requiring multiple tracer injections are anticipated.

Small Animal SPECT, SPECT/CT, and SPECT/MRI

79

Figure 1. 99mTc-MDP mouse bone scan acquired using USPECT-II (Courtesy F. Beekman, MI-Labs).

Figure 3. Mouse injected with 99mTc MDP (orange) for bone metabolism, 201Tl (green) for myocardial perfusion, and 123I (blue) for thyroid imaging. The image is acquired with a single list-mode SPECT acquisition, and the data were energy-discriminated post acquisition to generate three separate SPECT images. The image also shows a coregistered CT scan as a rendered transparent skeletal surface to give anatomic reference to SPECT images (Courtesy K. Iwata, Gamma Medica-Ideas).

ED

ES

Figure 2. Gated 99mTc-sestamibi mouse heart at end-diastole (top panel) and end-systole (bottom panel). Leftmost images are sections across the short-axis of the left ventricle while two orthogonal sections along the long-axis are shown at the center and right (Courtesy F. Beekman, MI-Labs).

Commercial Small Animal SPECT Instruments A number of SPECT and SPECT/CT systems are a vailable commerciall y. By the time this book is pub lished, the offerings will have undoubtedly changed. Our objective is neither to make recommendations of specific systems nor to pro vide an e xhaustive sur vey and comparison of e xisting de vices. Rather , those contemplating the purchase of a SPECT system for small animal imaging should be a ware of se veral things. F oremost is whether the system—or even SPECT for that matter— is suitable for answering the relevant research questions. As opposed to PET , where a single instr ument may be useful for a wide range of tracers in both mice and rats, SPECT performance depends upon how well the system is matched to the imaging task. For example, is high resolution in a small FO V required? Is it necessar y to

follow rapidl y changing dynamic processes? The configuration of most SPECT de vices can be altered to suit the imaging task, but performance optimization will likely require a good w orking relationship with the manufacturer for e ven minor hardw are and softw are modifications. As an e xample of a commercial de vice, F igure 4 shows the FLEX Triumph scanner available from Gamma Medica-Ideas (http://www.gm-ideas.com). The instrument can combine SPECT, CT, and/or PET on a single gantr y. The SPECT system uses CZT as the detector (see Section “Performance Optimization and Design Trends”) and can use multiple image for mation or collimation schemes depending on the animal model and the imaging task. A high-resolution SPECT instr ument from MI-LABS (http://www.milabs.com), the U-SPECT II, is sho wn in Figure 5. It is based on a three-headed scintillation camera (described in Section “Basic Principles”) and multiple pinhole collimators (Section “P erformance Optimization and Design Trends”), which can be changed to suit applications for imaging mice and rats. Bioscan offers two systems, the NanoSPECT/CT with four SPECT detectors and the HiSPECT system (http://www.bioscan.com). The significance of the latter de vice is that it can pro vide small animal imaging capabilities b y adding multi-pinhole collimation to standard clinical Anger cameras. ISE (http://www .isesrl.com) of fers a combined SPECT/PET instr ument

80

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

y

p (s, θ)

Object s

SPECT Camera

θ

x f (x, y)

Figure 4. Gamma Medica-Ideas FLEX Triumph preclinical imaging system. Instrument can incorporate up to three modalities: SPECT, PET, and CT (Courtesy K. Iwata). Figure 6. Rotating camera SPECT image acquisition for a single tomographic slice. SPECT camera and collimator rotate around object and collect a set of line-integral projections p(s, θ) of the distribution of radiotracer f(x, y) in the object.

instrument (http://www.neurophysics.com). As noted , our list is not exhaustive and new small animal SPECT companies and products appear almost monthly.

BASIC PRINCIPLES Image Formation and Reconstruction in SPECT and CT

Figure 5. MI-Labs U-SPECT-II preclinical SPECT system. Interchangeable imaging apertures can be chosen for imaging both mice and rats at high spatial resolution (Courtesy F. Beekman, MI-Labs).

based on YAP scintillation cameras (Section “Performance Optimization and Design Trends”). In addition to its small animal PET systems, Siemens has introduced the In veon SPECT system (http://www .medical.siemens.com). General Electric has in the recent past of fered the eXplore SpeCZT camera, w hich uses a “slit-slat” collimation approach similar to SPRINT II (http://www .gehealthcare.com). Finally, Neurophysics offers the MollyQ SPECT instrument that uses a novel image formation method similar to confocal microscopy resulting in a relatively low-cost

Figure 6 illustrates the principles of SPECT data acquisition. A γ-ray imaging camera and image for mation collimator (discussed in detail belo w) collect a set of projections or line-inte grals of acti vity through the object for each parallel “slice” of a 3D volume. Conceptually, the simplest data acquisition scheme is shown where the detector rotates around the object at least 180° and collects a set of parallel-ray line-integral projections at each view-angle. Tomographic infor mation for the complete v olume is generally obtained b y using a tw o-dimensional gamma camera. Ne glecting the issue of γ-ray attenuation b y the object (discussed fur ther belo w), the set of line-inte gral projections for each slice can be represented as follo ws: R

R

p( s, θ) = ∫ ∫ f ( x, y )δ ( x cos θ + y sin θ − s )dxdy , −R −R

where f (x, y) is the radiotracer distribution in a 2D slice of the object, R is the radius of the maximum FO V, and p(s, θ ) is the projection data inde xed by the vie w-angle of the camera θ and the displacement s from the central

Small Animal SPECT, SPECT/CT, and SPECT/MRI

ray. Reco vering or reconstr ucting the distribution of radiotracer requires collection of projections over the 2D interval [−R, R] × [0, π]. Data acquisition for X-ray CT is similar and Figure 7 shows a fan-beam CT setup. An X-ray detector and X-ray source corotate around the object with the detector measuring the transmission of X-ra ys along each line integral. Using Beer’s Law, the underlying linear attenuation µ(x, y), which we wish to estimate, is related to the projections for each slice by the following: R

R

p(ϕ, θ) = I 0 (ϕ ) ∫ ∫ exp[ − µ ( x, y )δ ( x cos(θ + ϕ ) −R −R

+ y sin(θ + ϕ ) − X s sin ϕ )]dxdy, where I0 is the X-ra y source flux on the detector in the absence of the object and X s is the distance from the X-ray focal-spot to the rotation axis (isocenter). The measured projections for this f an-beam CT e xample are indexed by view-angle θ and offset-angle ϕ on the X-ray detector. As for SPECT , a complete set of CT data requires collecting line-integrals over the interval [−R, R] × [0, π] (where −R ≤ X s sin ϕ ≤ R). SPECT image reconstr uction up until recentl y was accomplished using the filtered bac kprojection reconstruction algorithm. An intuiti ve approach to reconstruction for the parallel-ray geometry is to simply take the projection at each vie w-angle and “smear” it

back across the image at the appropriate angle, accumulating the result at each point in the image estimate.This straight backprojection operation results in an image in which g ross object features are e vident but details are severely b lurred. The Fourier slice or Central Section Theorem shows that the tomo graphic measurement process overemphasizes the low spatial frequency information and that an unb lurred estimate of the object can be recovered by f iltering each projection prior to backprojection with a f ilter having magnitude that increases linearly with spatial frequenc y—ie, a ramp filter in Fourier frequency space.59 The operation can be also be done as a con volution in the spatial domain, and the resulting f iltered backprojection reconstr uction is as follows: π ∞

fˆ ( x, y ) = ∫ ∫ h( x cos θ + y sin θ − s ) p( s, θ)dsdθ , 0 −∞

where h(s) is the spatial domain representation of the ramp filter. The operations of f iltering and backprojection can be re versed with an adaptation of the f ilter. The projection data are first backprojected and are then filtered with the appropriate 2D f ilter representation. Reconstruction of CT data is similar except that (1) measured projection (transmission) data are transformed prior to reconstr uction to give the linear attenuation along each line-integral by the following: p(ϕ, θ) = log

y

p (ϕ, θ)

Obje ct

θ X-ra y De te ctor

x

µ(x, y )

ϕ X-ra y S ource Xs

Figure 7. Typical fan-beam X-ray CT data acquisition in which an X-ray source and a charge-integrating detector corotate around the object and collect a set of transmission measurements along lines through the object. With the transformation noted in the text, transmission measurements are converted to integrals of linear attenuation along each line and can be reconstructed with same method used for emission tomography.

81

I 0 (ϕ ) p(ϕ, θ)

,

and (2) v arious for ms of f an-beam rather than parallelray reconstructions are used. Further information on various f iltered backprojection reconstr uction methods can be found in the referenced literature. 59–61 Although f iltered backprojection methods are still widely used in CT reconstr uction, the y ha ve lar gely been replaced in SPECT and PET applications b y the statistically moti vated methods described in Section “Accounting for De gradations: Advanced SPECT Reconstruction,” which allow for straightforward compensation of de gradations occur ring in emission tomography. Sampling

For the simplest, slice-b y-slice parallel-ra y geometr y shown in F igure 6, the sampling increment Δs between parallel ra ys along each projection and the inter -slice increment is chosen as follows:

82

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Δs


2 πR Δx

, Δθ
25 k eV) used in small animal SPECT ; instead, image formation relies on absorbing collimation in w hich most photons emitted in the direction of the camera are b locked to estimate the directions of those that are detected. More details on ph ysical mechanisms for image formation from photons and their suitability for SPECT can be found in Furenlid and colleagues. 71 The two most popular collimation schemes for small animal SPECT are e xamined belo w; a k ey feature the y share is use of a photon absorbing material. The ideal absorber w ould b lock photon transmission with a v ery thin la yer of material. This can, in f act, be closel y approached at lo wer photon ener gies. Table 2 sho ws several absorbing materials used in small animal SPECT collimators. Attenuation length is the mean free path of photons in the material and the distance at w hich 63% of

84

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 2. ATTENUATION LENGTH AND PHOTOELECTRIC FRACTION FOR COMMON SHIELDING AND COLLIMATION MATERIALS Density (g/cm3)

Material (Z)

Attenuation Length (cm)/Photoelectric Fraction (%) 30 keV

140 keV

171 keV

245 keV

364 keV

Molybdenum (42)

10.28

0.004

97.6

0.20

68.4

0.30

57.7

0.55

36.8

0.86

18.8

Lead (82)

11.34

0.003

95.2

0.04

91.0

0.06

87.9

0.14

79.4

0.32

65.0

Tungsten (74)

19.25

0.002

94.3

0.03

89.4

0.04

85.5

0.10

75.0

0.23

58.1

Gold (79)

19.30

0.002

94.9

0.02

90.5

0.04

87.1

0.09

77.9

0.20

62.6

Uranium (92)

18.90

0.001

95.9

0.02

92.5

0.03

90.0

0.06

83.4

0.16

77.2

photons have been absorbed or deflected from their initial trajectory. At 30 keV, attenuation lengths range from 0.02 to 0.04 mm and collimators allo wing high spatial resolution (< 100 µm FWHM) can be f abricated. At 140 k eV, attenuation length has already increased b y an order of magnitude. More material is needed for ef fective absorption, and collimator feature size must generall y be larger making high resolution more dif ficult to achie ve. This becomes particularly evident at 364 keV. Since attenuation occurs w hen a photon has been either completely absorbed or deflected from its original path, attenuation length is not the w hole stor y. Photons with energies less than 1.022 MeV undergo three significant interactions in matter: photoelectric absor ption in which the photon is completel y absorbed, and Compton or coherent scattering in w hich the direction of the incident photon is merel y changed. 72 Obviously, photoelectric absorption is the most desirable interaction from the collimation vie wpoint. Scattered photons ma y be deflected a way from the camera, ma y be absorbed elsewhere in the collimator , or ma y escape and be detected by the camera, reducing the effectiveness of the collimation. High proton number (Z) materials and lo w energies f avor photoelectric absor ption. As sho wn in Table 2, the fraction of photoelectric interactions decreases rapidly with increasing energy even for high-Z collimator materials. Because of the influence of absorbing material on collimator perfor mance, in par ticular, on the ability to achieve high spatial resolution, man y small animal SPECT instruments have explored use of tungsten, gold, and even depleted uranium in contrast to the lead alloys commonly used in clinical SPECT collimation. 73–75 Parallel-Hole Collimator

Channel collimation, and specif ically the parallel-hole collimator, is presently the most common image formation

aperture in clinical nuclear medicine.These collimators are often used in animal SPECT when good performance over a lar ge FOV is desirab le and high spatial resolution is a secondary consideration. As the name suggests, the collimator consists of a large number of open parallel channels having w alls constr ucted of a suitab le γ-ray absorber . Photons traveling in nearly the same direction as the orientation of the channel will be detected , w hereas those outside the acceptance angle will be absorbed by the channel w alls. F igure 11 sho ws the constr uction of a typical collimator from cor rugated lead sheets as w ell as def initions of parameters used in the e xpressions below. The combination of detector and collimator resolution on the PSF is given by the following expression with the f irst term inside the square-root being the ef fect due to the collimator itself. Note that the collimator resolution depends on the ratio of the hole diameter d to channel length l: d2

RT ≈

l2

( l + z ) 2 + RD2 ,

where additionally z is the distance from the collimator face to a parallel plane containing the source and RD the detector resolution (PSF). The counting ef ficiency (the number of photons detected/those emitted from a point) is as follo ws:

⎛ d 2 ⎞ , η ≈ K ( d / l ) ⎜ 2 ⎟ ⎝ ( d + t ) ⎠ 2

2

where t is the thickness of the septa and K is a normalizing factor depending on collimator hole shape (~0.26 for hexagonal holes in a he x array).8 For collimators used in clinical SPECT at 140 k eV, typical ef ficiency is 2 to 3 photons detected for e very 10,000 emitted (2–3 × 10−4). Significance of collimation ef ficiency will become more apparent in the discussion of noise in SPECT.

Small Animal SPECT, SPECT/CT, and SPECT/MRI

85

De te ctor l

t

d

Figure 11. Typical parallel-hole collimator. Left: parallel-hole collimator principle and definitions. Photons within an acceptance angle determined by collimator parameters will pass through the channels to be detected by the camera, whereas those outside the acceptance angle will be absorbed by the channel. Right: photograph of a section of a parallel-hole collimator showing the channels. Given the parameters written on the collimator face (in inches), spatial resolution is ~8 mm FWHM at 10 cm.

The important features to note regarding parallel-hole collimation are that (1) resolution becomes w orse with increasing distance from the collimator , (2) collimator efficiency is constant ir respective of distance, (3) ef ficiency is related to the square of the PSF width (ie, doubling spatial resolution decreases ef ficiency b y four times), and (4) perfor mance of parallel-hole collimation (and absorbing collimation in general) decreases rapidly with increasing ener gy due to the rapid decrease in absorber performance shown in Table 2. For example, at 1.8% septal penetration, w hich is near the upper end of acceptable, a collimator for 140 k eV can achie ve 5 mm resolution at 10 cm with an ef ficiency of 6.7 × 10−5. To achieve the same resolution at 364 k eV, the ef ficiency drops to 1.0 × 10−5 due to the thick er septa required to maintain the same penetration. A good introduction to parallel-hole collimation is given by Gunter.76 Designing specialized collimators for specific tasks is not out of question for investigators using small animal SPECT. A useful Web-based calculator for collimator design is a vailable on the Nuclear Fields Web site (http://www.nuclearfields.com).

Composite resolution for the pinhole and camera (again, as measured through the PSF width) is gi ven by the following:

Pinhole Aperture

There are several important things to note. F irst, the efficiency depends on the distance of the source point to the pinhole—it drops in inverse square law fashion as the distance increases. And when viewed at an angle θ, the apparent opening of a pinhole aper ture g(θ ) decreases with cor responding loss of ef ficiency. Ev en in the best case, where there is no additional vignetting due to the pinhole geometr y (e g, the k eel thickness in F igure 12), efficiency f alls as cos 3θ. Second, spatial resolution also becomes worse with increasing distance. Attempting to improve resolution b y using a smaller pinhole results in

Although channel collimators are most common in clinical SPECT, the simplest and the most often used collimator in small animal SPECT at present is a “pinhole” aperture in a γ-ray absorbing material. 77 Figure 12 shows a photo graph of a lead pinhole collimator along with definitions of parameters used in the e xpressions for resolution and ef ficiency. Analogous to a camera lens, the image of the object on the detector is in verted b y the pinhole.

2

2

⎛ z + z A ⎞ 2 ⎛ z ⎞ 2 RT ≈ ⎜ deff + ⎜ ⎟ RD , ⎟ ⎝ z A ⎠ ⎝ z A ⎠ in which zA is the distance from the pinhole to the detector, z is the distance from the pinhole to a parallel plane containing the point in the object, and deff is the effective pinhole diameter.78,79 As noted abo ve, absorbers are not perfect so the actual size is some what lar ger than the physical opening and also depends on the absorbing material and actual pinhole shape. A great deal of w ork has been done on pinhole designs for v arious applications and more detail can be found in the referenced work.80–86 The ef ficiency of a simple pinhole aper ture is as follows: η≈

deff2 16 z

2

g (θ).

86

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

De te ctor ZA d e ff

P inhole

Z θ

S ource Figure 12. Pinhole Collimator. Left: pinhole aperture and definitions used in text. Photons emitted from the source either pass through the open aperture or absorbed. Optimum pinhole geometry depends on γ-ray energy. To obtain small effective diameters, pinholes often have a “keel” instead of the absorber thickness decreasing to a knife-edge. Right: example of a simple pinhole aperture in a lead absorber.

loss of ef ficiency proportional to the square of the resolution. Finally, some good ne ws for small animal imaging: effects of the intrinsic resolution RD of the radiation imaging detector on the image resolution can be minimized b y using a lar ge magnif ication ( zA/z). Ne vertheless, in Section “P erformance Optimization and Design Trends,” it is sho wn that signif icant perfor mance increases can be obtained if a high spatial resolution detector is used with a pinhole in conjunction with demagnification (zA < z).

Image Degradations Two signif icant image distor tions were noted above. As a by-product of nonzero ef ficiency, collimation methods introduce an uncertainty in the arrival direction of detected γ-rays resulting in f inite spatial resolution or b lurring of reconstructed images. Fur ther uncer tainty introduced b y the camera adds in quadrature to the collimator resolution. In addition to this b lurring, there are also de gradations introduced by the randomness of radioactive decay and by interactions of the emitted x- or γ-ray photons with the object. This is of course tr ue in PET as well as SPECT. Photon Attenuation

In the case of X-ra y CT, photon attenuation b y the object is important because it encodes the desired signal. Attenuation by the object also occurs in emission tomo graphy, but instead of providing a useful signal, it becomes a distortion that if not cor rected introduces systematic quantitative inaccuracies in reconstr ucted images. For example, reconstructions from a uniformly emitting cylindric object would appear hollo wed out or cupped instead of ha ving uniform intensity.

Although photon attenuation is a signif icant issue in clinical SPECT imaging, it cannot necessarily be ignored in small animal SPECT especiall y if high quantitati ve accuracy is desired. F or e xample, at 140 k eV, the attenuation length of photons in soft tissue is appro ximately 6.5 cm— 63% of the photons will be either scattered or absorbed from this depth. At 1 and 2 cm, the fractions are 14 and 26%, respectively. At 30 k eV, attenuation length decreases to 2.6 cm. The cor responding fractions absorbed or scattered at 1 and 2 cm are 32 and 54%. The effect of attenuation can be compensated in the reconstr uction but requires knowledge of the attenuation distribution.Typically, this can be provided by transforming the map of linear attenuation measured by X-ray CT to the appropriate energy.2,87 Instruments capable of both SPECT and CT simplify matters, but correction can also be accomplished b y appropriately fusing information acquired on separate instr uments. Compton Scatter

As noted, attenuation occurs an y time a photon is either deflected from its initial trajectory or absorbed. As it turns out, attenuation in tissue at 140 k eV and abo ve is almost entirely due to Compton scattering. A Compton interaction is an inelastic process in which the scattered photon always has lower energy. The energy E of the scattered photon as a function of incident ener gy E0 in k eV and scattering angle θ is given by the following: E0

E= 1+

E0 511

.

(1 − cos θ)

A signif icant fraction of scattered photons escape the object and interact in the camera. If these scattered photons

Small Animal SPECT, SPECT/CT, and SPECT/MRI

are used indiscriminatel y in the reconstr uction, additional blurring and loss of contrast—especiall y in small cold regions—will result. However, if the detector is capab le of measuring the ener gy of each detected photon, it can discriminate against those that ha ve Compton scattered. F or example, a typical SPECT study at 140 keV with an Anger camera having 10% FWHM energy resolution might be set up to record onl y those e vents in a ±10% energy window centered at 140 keV. How effective this rejection is in practice depends on both the energy resolution of the detector and the emission energies of the radionuclides. The higher the emission energy, the larger the energy range over which the scattered photon spectrum is spread. For example, photons that have scattered at 90° would lose 1.7, 30, and 151 k eV for incident energies of 30, 140, and 364 k eV, respectively. Thus, although it is relati vely easy to use ener gy to ef fectively discriminate against scatter at 140 k eV and abo ve, it is nearly impossib le for 125I because of its closel y spaced emission energies (see Table 1) and the small ener gy loss with each scatter. Fortunately, at all but the lowest energies of interest, attenuation and Compton scatter in small animals such as mice and rats—w hile not negligible—are small.

Effects of the abo ve distor tions, f inite spatial resolution, attenuation, and Compton scatter can, in principle, be reduced—even eliminated in the limit—b y using an image reconstr uction technique that appropriatel y accounts for these de gradations. Ne vertheless, the “monkey wrench” limiting the ef ficacy of these recovery methods is counting noise in the projection measurements. Radioactive deca y is a random process. Assuming that the half-life is long with respect to the obser vation time, the probability of the emission of a given number of events in a time inter val of duration T is gi ven b y the Poisson distribution: P ( N = n | λ, T ) =

e − λT ( λT ) n n!

,

where λ is the rate of emission. Randoml y selecting emitted γ-rays via a collimator , for e xample, results in another P oisson process but with reduced rate gi ven by η × λ. The mean and v ariance in the number of obser ved events (measured through the collimator) are given by the following:

2 and σ ( N ) = ηλT ,

N = ηλT

respectively, where the angle brack ets denote the e xpectation operation. Note that both the mean number of e vents recorded and the uncer tainty in the number of e vents increase over time. Moreo ver, the v ariance in the number of events is equal to the mean. Rather than the ra w number of recorded e vents, however, in SPECT w e are more typicall y interested in the underlying concentration of radiotracer independent of the measurement time and instr ument efficiency. An appropriate estimator for the rate of photon detection in each projection measurement is as follows: ˆλ ( N ) = N . ηT Calculating the mean and v ariance of the estimated rate, we find that λˆ ( N ) =

N ηT

= λ,

ie, this simple estimator of rate is unbiased and that σ λˆ ( N ) = 2

Counting Noise

87

)

(

σ2 (N ) 2

ηT

2

=

λ ηT

.

The uncer tainty in emission rate estimate decreases with increasing observation time or collimation efficiency. To reduce the standard deviation by half, the product of the observation inter val and collimation ef ficiency must be four times g reater. More details on counting noise can be found in Knoll 72 and Barrett.88

Accounting for Degradations: Advanced SPECT Reconstruction If we assume that the three-dimensional distribution of radiotracer in the object can be represented adequatel y by a f inite number of basis functions—cubic v oxels (volume pictur e elements ) ha ving scale coef ficients λ1,...,λb, for e xample—then the lik elihood of obser ving the measurements—the number of counts yd in each detector channel—given that each is P oisson distributed and statistically independent is gi ven by the product of the individual probabilities, viz: P ([ y1 ,..., yD ] | [ λ1 ,..., λ B ]) T∑ a λ ) ( =∏ D

d =1

b

db

b

yd

(

exp −T ∑ b adb λ b yd !

),

88

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

where the coef ficients adb quantify the probability that a photon emitted from the bth v oxel in the representation of the object is detected in the dth detector channel. Note that the set of coefficients {adb}, often referred to as a system response matrix A, can model man y of the nonideal characteristics of the imaging process including photon attenuation, Compton scatter, and finite spatial resolution, while the for m of the lik elihood function itself pro vides appropriate handling of the kno wn statistical characteristics of the measurements. 89,90 Conceptually, the optimum image is obtained b y choosing non-ne gative v oxel v alues λ1,...,λb that maximize the likelihood of observing the projection measurements y1,...,yD. The solution is usually obtained iteratively using variations of the Expectation-Maximization (EM) algorithm. The EM algorithm for emission tomo graphy introduced by Shepp and Vardi91 and Lange and Carson92 is as follows: λˆ (bk +1) =

λˆ (bk )

adb yd , ∑ (k ) ˆ T ∑ adb d = 1 ∑ adb λ b d

D

b

with superscripts indicating the iteration number for each voxel. Since the abo ve algorithm is slo w to con verge, numerous adaptations have been developed. The ordered subsets expectation-maximization (OSEM) algorithm exhibits faster convergence and is often used in small animal PET and SPECT applications. 93 Despite capabilities for accuratel y modeling the SPECT measurement process, maximizing the ra w likelihood seldom produces a desirab le reconstr uction. Reconstruction generall y entails a trade-of f betw een accuracy, e g, resolution or bias, and precision or v ariance in the estimated voxel values. Maximum likelihood estimation stri ves for perfect resolution or zero bias. Reconstruction error, however, is a function of both variance and bias (and often higher order moments of the error distribution) and is generall y not minimum when bias is zero. P oisson counting noise in the projection measurements introduces v ery lar ge er rors in reconstructed images as one attempts to reco ver resolution or force the bias toward zero. There is an inherent trade-off between bias and v ariance or similarl y between resolution and noise. To limit the v ariance, the solution needs to be regularized, which can be accomplished b y stopping EM or OSEM iterations well short of convergence, through the use of penalty functions,94 the similar Bayesian methods,95 or b y post-smoothing the ra w maximum lik elihood

reconstruction by a desired PSF. Regularization typically leads to an image that has better f idelity with respect to the underlying tracer distribution. Ne vertheless, the method used depends on the purpose for which the images will be used. As noted in the ne xt section, measures of image quality and perfor mance e valuation are f ar from solved problems. Success in reco vering spatial resolution as well as in reducing the ef fects of Compton scatter and attenuation also depends on ho w accurately the imaging system is modeled. Pinhole SPECT , in par ticular, has received extensive attention in this regard.78,96–108

PERFORMANCE OPTIMIZATION AND DESIGN TRENDS Performance Optimization In contrast to PET, SPECT offers many possibilities for optimizing perfor mance through selection of dif ferent FOV sizes, collimators, and detectors. Ne vertheless, performance optimization is a comple x issue requiring the definition of a measure of perfor mance relative to a specific task. For example, is the task one of detecting the presence or absence of a lesion? Or rather, is the task one of quantifying the amount of uptak e in a specif ic region? What is the anticipated distrib ution of radiotracer and how will it dif fer among the animals used in the study? For the case of lesion detection, detectability may be the appropriate measure or perhaps recei ver operating characteristic cur ves that quantify inherent trade-off between false-positives and missed lesions. On the other hand, for quantification tasks some measure of the error is likely more appropriate. As noted in Section “Accounting for De gradations: Advanced SPECT Reconstruction,” er ror can generall y be thought of as being composed of systematic er rors or bias, which is due to f inite spatial resolution, for e xample, and variance, which is introduced both by counting noise and by variations among animals. Mean-squared er ror is the sum of the variance and the squared bias, but other measures such as maximum er ror or mean absolute er ror may be more appropriate depending on the task. To further complicate matters, it is often difficult to categorize the imaging task so con veniently as pure detection or pure quantif ication. After all, reconstr ucted images are typically used to answer a variety of questions. Optimization of medical imaging system performance remains an acti ve research area. A good overview of the rele vant issues can be found in an article b y F ryback and Thornbury109 and an e xcellent

Small Animal SPECT, SPECT/CT, and SPECT/MRI

introduction to perfor mance e valuation methods in Report 54 of the ICRU.110 Numerous methods have been proposed for SPECT performance optimization: refer to the series of ar ticles b y Bar rett for e xamples.111–114 Alternatives ha ve also been proposed for e valuating multipinhole SPECT systems (although the y are relevant to an y SPECT aper ture).115,116 A recent ar ticle by Clarkson and colleagues describes methods for quantifying perfor mance in multimodality imaging systems.117 But ignoring the f iner issues of image quality measurement and optimization, there are some obvious routes to improving SPECT performance.

Design Trends Based on the discussion in Section “Basic Principles,” the most clear-cut path to better small animal SPECT performance is to increase the counting efficiency of the instrument while maintaining spatial resolution. At first glance, the relationship betw een resolution and sensiti vity for both pinholes and parallel-hole collimators seems to work against it—doubling resolution decreases efficiency by a f actor of four. However, note that the object can be surrounded b y multiple cameras to increase ef ficiency with no loss in resolution. This approach has of course been taken in clinical SPECT where it is common to surround the patient with tw o or three Anger cameras. The same approach is used in small animal imaging. The basic idea is to sur round the object with detectors and to fully use the available detector area by projecting as many images of the FO V as possib le onto it without o verlap. Although fan- or cone-beam channel collimators can be used, multiple pinhole aper tures mak e this technique especially straightforward. If M pinholes view the object, efficiency for a point in the FOV equidistant from all pinholes increases by a factor of M while resolution remains

89

the same. Note that as the FO V becomes smaller , more nonoverlapping images can be projected onto the detector and higher ef ficiency can be achie ved. F or e xample, Beekman and colleagues 118 use a 75-pinhole aper ture to project nono verlapping images of a small FO V onto a three-camera instrument. Of course, adding pinholes has limits: at some point, projections of the object star t to overlap on the detector . Imaging systems that allow overlapping projections have traditionally been ter med coded apertur es. As for nonmultiplexing pinhole aper tures, raw sensitivity increases by a f actor of M; however, since there is no w ambiguity associated with the pinhole that passed a γ-ray, each detection car ries less infor mation than in an unmultiplexed system. The increased sensiti vity obtained with additional multiple xing often outw eighs the additional noise resulting from “unfolding” the projection o verlap, and this is especially true if either the overlap is small or the radiotracer is onl y accumulated in re gions of the object that are small relative to the size of the FOV (“hot spots”). Indeed, allowing at least slight overlap in pinhole projections is cur rently the most popular method of increasing ef ficiency w hile maintaining resolution in small animal SPECT. Evaluating effects of multiple xing on perfor mance using statistical methods has been the subject of a number of research projects. 113,115,116 Figure 13 sho ws tw o dif ferent multiple xing apertures: one in w hich relatively few pinholes are used and another having a lar ge number in a uniformly redundant array.119 Although a number of multistep approaches have been used over the years to recover data from multiplexing aper tures, reconstr uction is best done with straightforward adaptation of the maximum lik elihood method presented in Section “ Accounting for De gradations: Advanced SPECT Reconstr uction.” The matrix A merely needs to account for the f act that for each projection measurement photons are arriving through more than

Figure 13. Two examples of multiple pinhole multiplexing apertures. Left: relatively few pinholes in a tungsten absorber. Right: a no-twoholes-touching uniformly redundant array with a large number of open apertures (Courtesy R. Accorsi, Children’s Hospital of Pennsylvania).

90

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

a single aper ture opening. A large number of multiple xing aper ture small animal SPECT systems ha ve been developed.119–122 Although multiple pinhole multiple xing aper tures are used to a g reat e xtent in cur rent small animal SPECT devices, there is an emer ging alter native. Suppose that for a gi ven FOV size and spatial resolution, enough pinholes are placed in a hemispheric shell of absorber such that projections of the FO V on a concentric hemispheric detector just touch. Reducing the radius of the detector , thereb y decreasing zA, and increasing detector resolution to compensate reduces the size of each projection and allo ws more pinholes to be inser ted such that the projections again just touch, which increases ef ficiency without additional multiplexing. But to maintain the same image resolution, the effective size—and therefore the ef ficiency—of each pinhole aper ture must also be reduced in accordance with the for mula in Section “Image De gradations.” It turns out that as zA decreases, the number of allo wable pinholes increases faster than the loss of ef ficiency resulting from smaller pinholes. Each projection of the FOV is demagnified onto a high-resolution detector . If such high-resolution detectors are a vailable, an o verall significant increase in ef ficiency at the desired resolution results. 123 Limits are imposed b y the inability to make pinholes ha ving ever smaller ef fective diameters given the ph ysical limits of absorbers and b y the f act that the high-resolution detectors ma y require the ability to deter mine the γ-ray interaction location in three dimensions rather than just two, but the idea has dri ven much of the cur rent research direction in small animal SPECT instrumentation.

Toward High-Resolution Detectors

Motivated partially by the above idea (but perhaps just as often to make a more compact SPECT instrument), a number of higher resolution, smaller alter natives to the Anger camera are under acti ve development. These can be categorized as direct and indirect conversion detectors. Scintillation cameras are indirect converters; interactions are first converted to visible light photons and then to an electrical signal using a photodetector. γ-Rays interacting in a direct conversion detector such as CZT, on the other hand, generate an electrical signal directl y. As expected, the radiation absorption characteristics of each detector material are important and are summarized in Table 3. The most common detection material used in small animal SPECT is sodium iodide followed by CZT and cesium iodide. A common alter native to the Anger camera w as enabled b y the de velopment of the position-sensiti ve photomultiplier (PSPMT) and pix elated scintillator array. In contrast to the lar ge ar ray of PMTs used in an Anger camera, w hich as a unit has nominal photodetector resolution of 75 mm for a clinical instr ument, PSPMTs typically provide < 1 mm resolution in a significantly more compact package. To achieve 3 to 4 mm spatial resolution in an Anger camera, scintillation light must be spread broadly enough that it is measured by multiple 75 mm diameter PMTs. In comparison, no light-spreading is necessar y for the PSPMT ; the pix elated scintillator merel y channels the light photons to the face of the photodetector. An example using a single large PSPMT and NaI(Tl) array with 2 × 2 mm crystals is shown in F igure 14. Designs using single and multiple PSPMTs have been extensively used in commercial

Table 3. ATTENUATION LENGTH AND PHOTOELECTRIC FRACTION FOR COMMON γ-RAY DETECTOR MATERIALS VERSUS PHOTON ENERGY Material

Attenuation Length (cm)/Photoelectric Fraction (%)

Direct detectors

30 keV

140 keV

171 keV

245 keV

364 keV

Si

0.30

80.8

2.87

5.7

3.16

3.3

3.66

1.2

4.29

< 1.0

Ge

0.014

96.4

0.67

49.6

0.92

37.2

1.40

19.3

1.90

8.4

CZT

0.008

96.3

0.24

77.6

0.39

70.0

0.78

49.8

1.36

29.1

CdTe

0.008

96.4

0.24

77.8

0.38

69.3

0.76

50.2

1.34

29.5

NaI(Tl)

0.037

89.8

0.38

77.8

0.60

69.4

1.20

50.3

2.10

29.6

CsI(Tl)

0.024

90.5

0.26

80.9

0.41

73.5

0.87

55.7

1.60

34.7

YAP:Ce

0.014

96.6

0.64

50.1

0.88

37.8

1.33

19.9

1.80

8.8

LaBr3:Ce

0.013

95.5

0.35

73.0

0.54

63.2

1.02

43.0

1.67

23.6

Scintillators

Small Animal SPECT, SPECT/CT, and SPECT/MRI

91

Figure 14. Example of a compact scintillation camera (disassembled), The camera was constructed from a 5-inch diameter position-sensitive photomultiplier and a pixelated sodium iodide scintillation array.

and research small animal SPECT designs and typically result in compact systems ha ving good performance.124–128 Spatial resolution for these systems depends to a lar ge extent on cr ystal pitch in the scintillator ar ray and is cur rently 1.5 to 2 mm FWHM w hile energy resolution is ~12% FWHM. There ha ve been recent attempts to use CsI(Tl) ar rays with cr ystal sizes as lo w as 0.2 mm separated b y a center -to-center distance of 0.4 mm coupled to a high-resolution PSPMT.129 The in vestigators repor t intrinsic detector resolution as good as 0.6 mm FWHM. One of the advantages of direct detectors as opposed to scintillation cameras is that it is relati vely straightforward to achieve high spatial resolution (at least in 2D) in thick (ie, ef ficient) detectors. Moreo ver, ener gy resolution is generall y superior to indirect con version devices, which allo ws better scatter rejection and better radiotracer separation in multiple isotope studies. The most widely in vestigated direct detector for small animal SPECT is CZT. A CZT detector ha ving ~1.6 mm square pixels and 4.5% energy resolution at 140 keV is shown in Figure 15 and is used in the commercial FLEX Triumph SPECT-PET-CT device from Gamma Medica-Ideas (see Figure 4). Investigators at the Center for Gamma-Ray Imaging at the University of Arizona have developed high-resolution CZT detectors ha ving 380 µm pix els in a 2 mm thick detector with the ultimate goal of realizing SPECT devices having high resolution and efficiency using the demagnification approach.130 Examination of Table 3 reveals that the attenuation length for CZT at 140 keV is 2.4 mm. If the incident angle of γ-rays is lar ge, as it could w ell be in a compact pinhole geometry where zA is small, detectors will

Figure 15. CZT-based SPECT detector consisting of 25 25 5 mm thick CZT modules (upper right) backed by high density low power ASIC readout electronics (upper left). These CZT modules are tiled 5 by 5 to form a 12.5 × 12.5 cm field-of-view 80 × 80 pixel compact gamma camera (bottom). Camera has ~1.6 mm spatial resolution and 4.5% FWHM energy resolution at 140 keV—a significant improvement over the 10–12% typical for scintillation cameras (Courtesy K. Iwata, Gamma Medica-Ideas).

require depth-of-interaction or 3D resolution, w hich is under development. For imaging the lo w-energy emissions from 125I, Accorsi and colleagues have developed a 1 mm thick CdTe detector having 55 µm pixels in a 256 × 256 array read out using the Medipix2 chip for use with multiplexing apertures similar to that sho wn in F igure 13. The detector has an active area of only 14 mm square so that many devices must be tiled together to construct a large detector.131 Although silicon is rarel y considered a good x- or γ-ray detector (its attenuation length at 140 keV is nearly 3 cm), it has several advantages that make it attractive for use at ~30 k eV. It has well-understood properties and is the most w ell-developed semiconductor material. Detectors have been a vailable for char ged par ticles and low-energy photons for decades. It is relati vely straightforward to construct detectors having spatial resolution of nominally 50 µm using doub le-sided or thogonal strip readout. Peterson is e xploring use of a stack of doublesided strip detectors for lo w-energy imaging. 132 Each detector is 6 cm × 6 cm × 1 mm thick with a strip pitch of 59 µm. The attenuation length of 30 k eV photons in silicon is 3 mm; therefore, a stack of detectors must be used to obtain reasonab le ef ficiency. One adv antage of

92

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

using a stack, however, is that it also pro vides depth resolution. Intrinsic silicon contains residual impurities that prevent fabrication of detectors of more than 1 to 2 mm thick; however, it is possible to compensate the effects of impurities b y drifting lithium ions into the detector thereby allo wing f abrication of thick er de vices. Use of 6 mm thick lithium-drifted silicon (SiLi) detectors having 1 × 1 mm elements has been proposed for lo w-energy SPECT imaging.133 Scintillation cameras using solid-state photodetectors instead of bulkier PMTs are also under acti ve investigation for high-resolution imaging. CsI(Tl) scintillators are typically used because of the e xcellent spectral match of the scintillation light with the sensiti vity of silicon photodetectors. As examples, a design that couples a 5-mm thick CsI(Tl) scintillator to a position-sensiti ve avalanche photodiode (APD) has been proposed and estimated resolution is 0.5 mm at 140 k eV.134 Fiorini and colleagues135,136 have coupled a continuous CsI(Tl) crystal 3-mm thick to a miniature ar ray of silicon drift photodetectors and have achieved 0.16 mm resolution with 14% energy resolution at 122 k eV although cooling is required, which can complicate the imaging system. Detection of scintillation light requires lo w-noise photodetectors such as PMTs, drift detectors, APDs, or the new silicon photomultipliers (SiPM), but it has been challenging to construct imaging arrays or position-sensitive photodetectors e xhibiting spatial resolution less than ~0.5 mm. On the other hand , standard CCD and CMOS camera chips are capable of high resolution but their noise precludes use directl y as a photodetector for scintillation light. Ho wever, a ne w de vice, the electron-multipl ying CCD (EMCCD), has become a vailable in the past fe w

years and has been incor porated into high-resolution detectors by several investigators. Like the APD or SiPM, the EMCCD has an inter nal gain mechanism that signif icantly increases the signal-to-noise ratio to the point that scintillation light is detectab le. The drawback is that the devices are small and therefore not w ell matched to FOV sizes used in small animal SPECT . Options for matching the small EMCCD to a lar ger FOV have explored lenses for coupling the scintillation light, 137 fiber optic tapers, 138 and an electronic demagnif ication tube.139 Research Instruments

It is impractical within the scope of this chapter to cover all devices used in small animal SPECT or e ven any in detail. Se veral commercial de vices w ere mentioned above, and w e highlight three research instr uments below. Refer to the current literature for more details on these and other instr uments while noting that dif ferent imaging applications often benef it from v ery dif ferent SPECT designs. 68,99,118,121,122,125,127,130,133,134,139–154 The FastSPECT II de vice under de velopment at the University of Arizona consists of a circular ar rangement of modular scintillation cameras (F igure 16). Each camera is similar to an Anger camera in that it comprises a continuous NaI(Tl) scintillator coupled to 9 PMTs. P osition resolution is some what better than clinical Anger cameras, but a key feature is that good position resolution can be attained close to the edges of each detector . The conventional Anger camera has a lar ge dead area around the edges, resulting in loss of ef ficiency w hen pack ed into ar rays. In contrast, the modular cameras can be close-packed with little efficiency loss.

Figure 16. The FastSPECT II system developed at the University of Arizona. Left: complete unit undergoing calibration. Right: partially assembled instrument showing the arrangement of modular scintillation cameras (Courtesy L. Furenlid, U. Arizona).

Small Animal SPECT, SPECT/CT, and SPECT/MRI

The main feature of F astSPECT II (and with an increasing number of SPECT instr uments) is that the object is viewed from many directions simultaneously by a large number of pinhole apertures (similar to the U-SPECT II, see F igure 5). The simultaneous vie ws satisfy the 3D sampling requirements discussed abo ve (at least for moderate spatial resolution) so that rapidl y changing dynamic processes can be easily followed as in PET. More conventional small animal SPECT instr uments that require translation or rotation of the aper ture do not collect complete data at an y time instant. Although following slowly varying dynamic processes is possible, imaging rapidly changing distributions can be prob lematic with more conventional devices. Meng and colleagues have developed a SPEM prototype with inte grated CT for imaging 125I in mice and in ex vivo specimens.155 The instrument, shown in Figure 17, has two SPEM detectors each consisting of a multipinhole apertures and thin CsI(Tl) scintillator coupled to an EMCCD camera through an electronic demagnif ication tube. To acquire a complete set of SPECT data, the specimen is placed on a rotating stage betw een the detectors. An X-ray source and amorphous silicon X-ray detector are used to acquire simultaneous cone-beam CT images. Spatial resolution for 125I is less than 200 µm, and efficiency in full system can be as high as 1× 10−3. The prototype has been used for tracking radiolabeled T-cells injected into the brain of a mouse. Finally, both SPECT and PET can require an hour or more of imaging time, and in order to minimize motion, animals must typicall y be anesthetized or otherwise restrained. Ho wever, both can alter the functional

93

processes measured with SPECT and this can be especially important in CNS imaging. A team of in vestigators has developed a SPECT instrument that can be used for imaging awake, unrestrained mice. 156 Designed for brain imaging, the de vice uses an IR camera tracking system with retroreflecting spheres attached to the mouse’ s head (Figure 18). P osition and orientation are track ed throughout the scan and used to reconstr uct the acquired data into a reference frame that can then be fused with a CT scan acquired using the same instrument. To limit gross motion, the instrument also incor porates a “bur row” in w hich the mouse feels comfortable.

SPECT/CT AND SPECT/MR SPECT/CT The previous discussions have focused on design of the SPECT portion of a SPECT/CT or SPECT/MRI instrument because at least for SPECT/CT, this is the par t of the instrument that will vary most widely among applications. Adding CT capability to a SPECT instr ument is straightforw ard and is typicall y accomplished b y adding a fixed-anode X-ray tube and appropriate X-ray detector. Operating characteristics for the X-ray source are nominally from 50 to 90 kV and 1 to 5 mA depending on the size of the animal and application. Typically, a 2D CMOS or amor phous silicon flat-panel X -ray detector is used to collect a set of cone-beam projections that are usually reconstructed using the Feldkamp reconstruction63 or v ariations that more appropriatel y handle the incomplete data resulting from the circular orbit. The small cone-angle associated with most CT imaging geometries for small animals signif icantly reduces the sampling ar tifacts noted abo ve resulting from use of a circular source trajector y. Numerous commercial and research instr uments ha ve incor porated CT with SPECT.145,146,157–159

SPECT/MR

Figure 17. A prototype single photon emission microscopy device with integrated X-ray CT. Two electron-multiplying CCD cameras, electronic demagnification tubes, scintillators, and multipinhole apertures (left and right) view a mouse-sized field-of-view. Complete sampling is obtained by placing the mouse on a rotating stage. X-ray tube for CT is located at the bottom of the figure, whereas an amorphous silicon X-ray detector is located at the top.

Performing simultaneous MRI and SPECT is considerably more challenging due to mutual interference betw een the MRI and SPECT hardw are. F irst, electric char ges (eg, electrons) mo ving in a magnetic f ield are subject to a Lorentz force that alters their trajector y making devices with such con ventional PMTs dif ficult—if not impossible—to use in the high static MRI f ield. Ev en direct detectors such as CZT e xperience depth-dependent positioning errors if the electric drift-f ield of the detector is not aligned with the magnetic f ield direction. Another

94

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 18. SPECT/CT instrument for awake, unrestrained mice. Left: SPECT/CT instrument for brain imaging in conscious mice. In addition to SPECT and CT, instrument has infrared-tracking cameras to continuously record position and orientation of the mouse’s head. Right: awake mouse is shown in burrow with retroreflecting targets for IR light (Courtesy D. Weisenberger, Thomas Jefferson National Accelerator Facility).

problem is interference of the g radient switching and RF sequences with the sensiti ve SPECT electronics. Much of this issue can lik ely be resolv ed with appropriate electromagnetic shielding. F inally, the SPECT hardw are—especially the EMI shielding—can interfere with MRI acquisition. This can be minimized b y appropriate design that places RF coils within the bore of the SPECT de vice rather than vice v ersa. As for PET , de velopment of SPECT/MR instruments is an active research area. A CZT SPECT detector inser t for an instr ument under de velopment in a collaboration between investigators at John Hopkins Uni versity and Gamma Medica-Ideas is sho wn in Figure 19. Other research SPECT/MR systems ha ve been developed,160 but as for PET/MR, the field is in its infancy and numerous advances will be made in the coming years.

FUTURE DIRECTIONS Radiation Sensors for Better Integration with Other Imaging Modalities Although the potential of the combination of SPECT and other modalities has been well demonstrated, development of multimodality instr umentation has reall y onl y just begun. Cur rent emphasis has been more on assemb ling multimodal de vices that demonstrate proof-of-concept rather than on achie ving the optimum perfor mance. Taking combined SPECT/MR systems as an e xample, the ideal detectors should be immune to the magnetic f ield induced distor tion and ha ve minimal interference to MR data acquisition. The detectors should also ha ve a high spatial resolution and at least moderate ener gy resolution allowing better inte gration of SPECT infor mation with MR. Given the limited space available inside MR systems, compact γ-ray sensors w ould allow for a more f avorable imaging geometr y for SPECT . As noted abo ve, much research ef fort is no w focused on de velopment of compact, high-resolution SPECT detectors.

Figure 19. MR-compatible SPECT prototype based on rings of eight CZT modules arranged in an octagonal configuration around a sleeve-collimator/RF coil. This system can acquire MR and SPECT images simultaneously with SPECT acquired without any movement (as in FastSPECT II). In the photo above, three rings are populated with 24 modules working as independent cameras. When four rings are populated (32 modules total), the field-of-view will cover an entire mouse. The outer diameter of the system is 12.0 cm to fit standard preclinical MR systems (Courtesy K. Iwata, Gamma Medica-Ideas).

Adaptive SPECT Imaging In addition to adv anced detectors, future SPECT systems will also benef it from adapti ve data acquisition schemes that maximize the ef ficiency for collecting infor mation relevant to a desired imaging task. Ideall y, an adapti ve SPECT system could alter its measurement geometr y in real time based on the imaging infor mation that has been collected and on the given task specified by the user.161 This would help both to of fset the intrinsicall y lo w detection efficiency of SPECT and to mak e aperture selection more robust with respect to the underl ying object. Adaptive imaging has been widel y proposed or demonstrated in

Small Animal SPECT, SPECT/CT, and SPECT/MRI

astronomic,162–164 ultrasound,165,166 and MRI imaging applications.167,168 Recent ef forts ha ve focused on the development of a mathematic framework161 and a prototype system for demonstrating the concept in SPECT .169 To realize the potential of this approach, future in vestigations are needed in the de velopment of (1) meaningful methods for task-based image quality assessment that could be used to guide the self-adaptive process, (2) rapid computation for both reconstruction and evaluation of performance indices, and (3) adaptive detection hardware and associated control mechanisms.

CONCLUSIONS In contrast to clinical SPECT , small animal SPECT is a rapidly expanding field buoyed by the number of radiolabeled compounds a vailable and rapid adv ances in instr umentation. The relati ve ease with w hich SPECT can be incorporated into the biomedical research laborator y is a significant adv antage. Although the f ield of combined SPECT/MR instr uments is still y oung, SPECT/CT is a well-developed technolo gy with se veral instr uments available commercially. In the coming years, expect to see continued expansion of small animal SPECT applications with new and e xisting radiotracers and with commercial availability of advanced concepts such as SPEM, adaptive SPECT apertures, and combined SPECT/MR systems.

REFERENCES 1. Del Guerra A, Belcari N. State-of-the-art of PET, SPECT and CT for small animal imaging. Nucl Instr um Methods Ph ys Res A 2007;583:119–24. 2. Hasegawa BH, Iw ata K, Wong KH, et al. Dual-modality Imaging of function and physiology. Acad Radiol 2002;9:1305–21. 3. King MA, Pretorius PH, F arncombe T, Beekman FJ. Introduction to the ph ysics of molecular imaging with radioacti ve tracers in small animals. J Cell Biochem 2002;221–30. 4. Madsen MT . Recent adv ances in SPECT imaging. J Nucl Med 2007;48:661–73. 5. Meikle SR, K ench P, Banati RB , Kassiou M. Small animal SPECT and its place in the matrix of molecular imaging technolo gies. Phys Med Biol 2005;50:R45–61. 6. Peremans K, Cornelissen B, Van den Bossche B, et al. A review of small animal imaging planar and pinhole spect Gamma camera imaging. Vet Radiol Ultrasound 2005;46:162–70. 7. Kupinski MA, Bar rett HH, editors. Small-animal SPECT imaging. New York: Springer; 2005. 8. Cherry SR, Sorenson J , Phelps M. Ph ysics in nuclear medicine. 3rd ed. Philadelphia (PA): Saunders; 2003. 9. Wernick MN, Aarsvold JN. Emission tomography: the fundamentals of PET and SPECT. New York: Academic Press; 2004. 10. Keyes JW, Leonard PF, Brody SL, et al. Myocardial infart quantification in the dog by single photon emission computed tomo graphy. Circulation 1978;58:227–32. 11. Jaszczak RJ, Murphy PH, Huard D , Burdine JA. Radionuclide emission tomography of the head with 99m-Tc and a scintillation camera. J Nucl Med 1977;18:373–80.

95

12. Rogers WL, Koral KF, Mayans R, et al. Coded-aper ture imaging of the heart. J Nucl Med 1980;21:371–8. 13. Rogers WL, Clinthor ne NH, Stamos J , et al. SPRINT : a stationar y detector single photon ring tomo graph for brain imaging. IEEE Trans Med Imaging 1982;1:63–8. 14. Rogers WL, Clinthor ne NH, Shao L, et al. SPRINT II: a second generation single photon ring tomo graph. IEEE Trans Med Imaging 1988;4:291–7. 15. Wrobel M. Slit Aperture SPECT and Simplified in vitro Methods for the Dosimetry of Positron Emitting Radiotracers. School of Public Health [Ph.D disser tation], Ann Arbor (MI): Uni versity of Michigan; 1988. 16. Jaszczak RJ, Li J, Wang H, et al. Pinhole collimation for ultra-highresolution, small f ield-of-view SPECT . Ph ys Med Biol 1994; 39:425–37. 17. Weber D A, Iv anovic M, F ranceschi D , et al. Pinhole SPECT : an approach to in vi vo high resolution SPECT imaging in small laboratory animals. J Nucl Med 1994;35:342–8. 18. Acton PD, Choi SR, Plossl K, K ung HF. Quantification of dopamine transporters in the mouse brain using ultra-high resolution singlephoton emission tomography. Eur J Nucl Med Mol Imaging 2002; 29:691–8. 19. Andringa G, Dr ukarch B, Bol, JGJM, et al. Pinhole SPECT imaging of dopamine transpor ters cor relates with dopamine transpor ter immunohistochemical anal ysis in the MPTP mouse model of Parkinson’s disease. Neuroimage 2005;26:1150–8. 20. Blankenberg FG, Kalin yak J , Liu LP , et al. Tc-99-HYNIC-annexin VSPECT imaging of acute stroke and its response to neuroprotective therapy with anti-F as ligand antibody . Eur J Nucl Med Mol Imaging 2006;33:566–74. 21. de Win MML, de Jeu RAM, de Br uin K, et al. Validity of in vi vo [I-123]beta-CIT SPECT in detecting MDMA-induced neuroto xicity in rats. Eur Neuropsychophar macol 2004;14:185–9. 22. Garcia-Alloza M, Bacskai BJ . Techniques for brain imaging in vi vo. Neuromolecular Med 2004;6:65–78. 23. Hwang LC, Chang CJ , Liu HH, et al. Imaging the a vailability of serotonin transpor ter in rat brain with I-123-AD AM and small-animal SPECT. Nucl Med Commun 2007;28:615–21. 24. Nikolaus S, Antke C, Beu M, et al. In-vi vo quantif ication of dosedependent dopamine transpor ter b lockade in the rat striatum with small animal SPECT. Nucl Med Commun 2007;28:207–13. 25. Nikolaus S, Larisch R, Beu M, et al. In vestigating the dopaminer gic synapse in vivo. II. Molecular Imaging studies in small laboratory animals. Rev Neurosci 2007;18:473–504. 26. Nikolaus S, Larisch R, Wirrwar A, et al. [I-123]Iodobenzamide binding to the rat dopamine D-2 receptor in competition with haloperidol and endogenous dopamine—an in vi vo imaging study with a dedicated small animal SPECT . Eur J Nucl Med Mol Imaging 2005;32:1305–10. 27. Nikolaus S, Wirrwar A, Antke C, et al. Quantitation of dopamine transporter blockade by methylphenidate: f irst in vi vo investigation using [I-123]FP-CIT and a dedicated small animal SPECT . Eur J Nucl Med Mol Imaging 2005;32:308–13. 28. Pissarek MB, Oros-Peusquens AM, Schramm NU . Challenge b y the murine brain: multi-pinhole SPECT of I-123-labelled pharmaceuticals. J Neurosci Methods 2008;168:282–92. 29. Scherfler C, Donnemiller E, Schock e M, et al. Ev aluation of striatal dopamine transporter function in rats by in vivo beta-[I-123]CIT pinhole SPECT. Neuroimage 2002;17:128–41. 30. Scherfler C, Scholz SW , Donnemiller E, et al. Ev aluation of [I-123]IBZM pinhole SPECT for the detection of striatal dopamine D-2 receptor availability in rats. Neuroimage 2005;24:822–31. 31. Seo Y, Gao D W, Hase gawa BH, et al. Rodent brain imaging with SPECT/CT. Med Phys 2007;34:1217–20. 32. Wall JS, Kennel SJ, Paulus M, et al. Radioimaging of light chain amyloid with a fibril-reactive monoclonal antibody. J Nucl Med 2006; 47:2016–24.

96

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

33. Acton PD, Thomas D, Zhou R. Quantitati ve imaging of m yocardial infarct in rats with high resolution pinhole SPECT . Int J Cardiovasc Imaging 2006;22:429–34. 34. Constantinesco A, Choquet P, Monassier L, et al. Assessment of left ventricular perfusion, v olumes, and motion in mice using pinhole gated SPECT. J Nucl Med 2005;46:1005–11. 35. Liu ZL, Bar rett HH, Ste venson GD, et al. High-resolution imaging with Tc-99m-glucarate for assessing myocardial injury in rat heart models exposed to dif ferent durations of ischemia with reperfusion. J Nucl Med 2004;45:1251–9. 36. Terrovitis J, Kwok KF, Lautamaki R, et al. Feasibility of in vivo cardiac stem cell tracking, by SPECT and PET imaging, using the sodiumiodide symporter as reporter gene. Circulation 2007;116:374. 37. Thomas D, Bal H, Arkles J, et al. Noninvasive assessment of myocardial viability in a small animal model: comparison of MRI, SPECT, and PET. Magn Reson Med 2008;59:252–9. 38. Vanhove C, Lahoutte T, Defrise M, et al. Reproducibility of left v entricular volume and ejection fraction measurements in rat using pinhole gated SPECT. Eur J Nucl Med Mol Imaging 2005;32:211–20. 39. Wu MC, Hase gawa BH, Dae MW. Performance evaluation of a pinhole SPECT system for m yocardial perfusion imaging of mice. Med Phys 2002;29:2830–9. 40. Zhou R, Thomas DH, Qiao H, et al. In vi vo detection of stem cells grafted in infarcted rat myocardium. J Nucl Med 2005;46:816–22. 41. Carlson SK, Classic KL, Hadac EM, et al. In vi vo quantitation of intratumoral radioisotope uptake using micro-single photon emission computed tomo graphy/computed tomography. Mol Imaging Biol 2006;8:324–32. 42. Mathe D, Foldes I. Visualization and quantif ication of mouse thyroid gland 99mTc and 131I-uptake by a small animal SPECT/CT system. Eur J Nucl Med Mol Imaging 2007;34:S398–S398. 43. Mueller C, Forrer F, Schibli R, et al. SPECT study of folate receptor-positive malignant and nor mal tissues in mice using a no vel Tc-99mradiofolate. J Nucl Med 2008;49:310–7. 44. Muller C, Hohn A, Schubiger PA, Schibli R. Pre-clinical evaluation of novel organometallic Tc-99m-folate and Tc-99m-pteroate radiotracers for folate receptor -positive tumour tar geting. Eur J Nucl Med Mol Imaging 2006;33:1007–16. 45. Schipper ML, Riese CGU, Seitz S, et al. Efficacy of Tc-99m pertechnetate and I-131 radioisotope therapy in sodium/iodide symporter (NIS)-expressing neuroendocrine tumors in vivo. Eur J Nucl Med Mol Imaging 2007;34:638–50. 46. Hildebrandt IJ , Gambhir SS. Molecular imaging applications for immunology. Clin Immunol 2004;111:210–24. 47. Kanwar B, Gao D-w, Pan M-h, et al. Pinhole SPECT imaging and quantification of intestinal Cd4+ T cells in the de xtran sulf ate sodium model of murine colitis. Gastroenterology 2007;132:A169–70. 48. Pittet MJ, Grimm J, Berger CR, et al. In vivo imaging of T cell delivery to tumors after adopti ve transfer therapy. Proc Natl Acad Sci U S A 2007;104:12457–61. 49. Scopinaro F, Paschali E, Di Santo G, et al. Bombesin receptors and transplanted stem cells in rat brain: high-resolution scan with Tc99m BN1.1. Nucl Instrum Methods Phys Res A 2006;569:525–8. 50. Cisbani E, Cusanno F , Garibaldi F , et al. Molecular imaging with radionuclides, a po werful technique for studying biolo gical processes in vi vo. Nucl Instr um Methods Ph ys Res A 2007; 571:169–72. 51. de Graaf W, Vetelainen RL, de Br uin K, et al. Tc-99m-GSA scintigraphy with SPECT for assessment of hepatic function and functional volume during li ver regeneration in a rat model of par tial hepatectomy. J Nucl Med 2008;49:122–8. 52. Gambhir SS, Herschman HR, Cher ry SR, et al. Imaging transgene expression with radionuclide imaging technolo gies. Neoplasia 2000;2:118–38. 53. Gotthardt M, Lal yko G, v an Eerd-Vismale J, et al. A new technique for in vi vo imaging of specif ic GLP-1 binding sites: f irst results in small rodents. Regul Pept 2006;137:162–7.

54. Ostendorf B, Scherer A, Wirrwar A, et al. High-resolution multipinhole single-photon-emission computed tomography in experimental and human arthritis. Arthritis Rheum 2006;54:1096–104. 55. Paulmurugan R, Ra y P , De A, et al. Imaging protein-protein interactions in li ving subjects. Trends Analyt Chem 2005; 24:446–58. 56. Pham TQ, Berghofer P, Liu X, et al. Preparation and biolo gic evaluation of a no vel radioiodinated benzylpiperazine, I-123-MEL037, for malignant melanoma. J Nucl Med 2007;48:1348–56. 57. Kung M-P, K ung HF. Mass ef fect of injected dose in small rodent imaging by SPECT and PET. Nucl Med Biol 2005;32:673–8. 58. Funk T, Mingshan S, Hase gawa BH. Radiation dose estimate in small animal SPECT and PET. Med Phys 2004;31:2680–6. 59. Kinahan PE, Defrise M, Clackdo yle R. Analytic image reconstr uction methods. In: Wernick MN, Aarsvold JN, editors. Emission tomography: the fundamentals of PET and SPECT . Ne w York: Academic Press; 2004. p. 421–42. 60. Horn BKP . Density reconstr uction using arbitrar y ra y-sampling schemes. Proc IEEE 1978;66:551–62. 61. Horn BKP . F an-beam reconstr uction methods. Proc IEEE 1979; 67:1616–23. 62. Tuy HK. An inversion formula for cone-beam reconstr uction. SIAM J Appl Math 1983;43:546–52. 63. Feldkamp LA, Da vis LC, Kress JW. Practical cone-beam algorithm. J Opt Soc Am A 1984;1:612–9. 64. Wang G, Yu HY, De Man, B . An outlook on X-ra y CT research and development. Med Phys 2008;35:1051–64. 65. Ayan AS, Metzler SD . Axial resolution of helical-orbit pinhole SPECT with synchronized and unsynchronized motion. IEEE Trans Nucl Sci 2008;55:524–30. 66. Bal G, Zeng GL, Acton PD, et al. Study of different pinhole configurations for small animal tumor imaging. 2004 IEEE Nuclear Science Symposium Conference Record 2004;3133–7. 67. Barrett HH, Gif ford HC. Cone-beam tomo graphy with discrete data sets. Phys Med Biol 1994;39:451–76. 68. Beekman FJ , Vastenhouw B . Design and simulation of a highresolution stationary SPECT system for small animals. Phys Med Biol 2004;49:4579–92. 69. Damiani C, Del Guer ra A, Moretti E, et al. Sampling considerations for high resolution small animal SPECT . IEEE Trans Nucl Sci 2005;52:555–9. 70. Metzler SD , Greer KL, Jaszczak RJ . Helical pinhole SPECT for small-animal imaging: a method for addressing sampling completeness. IEEE Trans Nucl Sci 2003;50:1575–83. 71. Furenlid L, Chen Y-C, Hyunki K. SPECT imager design and data acquisition systems. In: Kupinski MA, Barrett HH, editors. Small animal SPECT imaging. New York: Springer; 2005. p. 115–38. 72. Knoll GF. Radiation detection and measurement. 3rd ed. Ne w York: Wiley; 2000. 73. Kastis GA, Wu MC, Balzer SJ , et al. Tomographic small-animal imaging using a high-resolution semiconductor camera. IEEE Trans Nucl Sci 2002;49:172–5. 74. Tenney CR, Tornai MP, Smith MF, et al. Uranium pinhole collimators for 511-k eV photon SPECT imaging of small v olumes. IEEE Trans Nucl Sci 2001;48:1483–9. 75. Vastenhouw B , Beekman F . Submillimeter total-body murine imaging with U-SPECT-I. J Nucl Med 2007;48:487–93. 76. Gunter DL. Collimator design for nuclear medicine. In: Wernick MN, Aarsvold JN , editors. Emission tomo graphy: the fundamentals of PET and SPECT . New York: Academic Press; 2004. p. 153–68. 77. Beekman F, van der Have F. The pinhole: gateway to ultra-high-resolution three-dimensional radionuclide imaging. Eur J Nucl Med Mol Imaging 2007;34:151–61. 78. Accorsi R, Metzler SD . Analytic deter mination of the resolutionequivalent effective diameter of a pinhole collimator . IEEE Trans Med Imaging 2004;23:750–63.

Small Animal SPECT, SPECT/CT, and SPECT/MRI

79. Accorsi R, Metzler SD. Resolution-effective diameters for asymmetric-knife-edge pinhole collimators. IEEE Trans Med Imaging 2005;24:1637–46. 80. Shokouhi S, Fritz MA, Metzler SD, et al. Design of a multi-pinhole collimator in a dual-headed , stationar y, small-animal SPECT . 2006 IEEE Nuclear Science Symposium Conference Record 2007;2399–402. 81. Smith MF, Meikle SR, Weisenberger AG, Maje wski S. Design of multipinhole collimators for small animal SPECT . 2003 IEEE Nuclear Science Symposium. Conference Record 2004;2291–5. 82. Tae Yong S, Byung-Tae K, Jin Ho J, et al. Optimization of pinhole collimator for small animal SPECT using Monte Carlo simulation. IEEE Trans Nucl Sci 2003;50:327–32. 83. van der Ha ve F, Beekman FJ . Penetration, scatter and sensiti vity in channel micro-pinholes for SPECT : a Monte Carlo in vestigation. IEEE Trans Nucl Sci 2006;53:2635–45. 84. van der Ha ve F , Beekman FJ . Photon penetration and scatter in micro-pinhole imaging: a Monte Carlo in vestigation. Ph ys Med Biol 2004;49:1369–86. 85. Vanhove C, Defrise M, Lahoutte T, Bossuyt A. Three-pinhole collimator to improve axial spatial resolution and sensitivity in pinhole SPECT. Eur J Nucl Med Mol Imaging 2008;35:407–15. 86. Williams MB, Stolin AV, Kundu BK. Investigation of efficiency and spatial resolution using pinholes with small pinhole angle. IEEE Trans Nucl Sci 2003;50:1562–8. 87. Hwang AB, Hasegawa BH. Attenuation correction for small animal SPECT imaging using X-ra y CT data. Med Ph ys 2005; 32:2799–804. 88. Barrett HH. Detectors for small-animal SPECT II. In: Kupinski MA, Barrett HH, editors. Small-animal SPECT imaging. Ne w York: Springer; 2005. p. 49–86. 89. King MA, Glick SJ , Pretorius PH, et al. Attenuation, scatter , and spatial resolution compensation in SPECT . In: Wernick MN , Aarsvold JN, editors. Emission tomo graphy: the fundamentals of PET and SPECT. New York: Academic Press; 2004. p. 473–98. 90. Lalush DS, Wernick MN. Iterative image reconstruction. In: Wernick MN, Aarsvold JN, editors. Emission tomo graphy: the fundamentals of PET and SPECT . Ne w York: Academic Press; 2004. p. 443–72. 91. Shepp LA, Vardi Y. Maximum likelihood reconstruction in positron emission tomography. IEEE Trans Med Imaging 1982;1:113–22. 92. Lange K, Carson RE. EM reconstr uction algorithms for emission and transmission tomo gpraphy. J Comput Assist Tomogr 1984; 8:306–16. 93. Hudson HM, Larkin RS. Accelerated image reconstr uction using ordered subsets of projection data. IEEE Trans Med Imaging 1994;13:601–9. 94. Fessler JA, Hero AO. Penalized maximum-likelihood image reconstruction using space-alter nating generalized EM algorithms. IEEE Trans Image Process 1995;4:1417–29. 95. Qi J , Leah y RM, Cher ry SR, et al. High resolution 3D Ba yesian image reconstr uction using the microPET small animal scanner . Phys Med Biol 1998;43:1001–13. 96. Andreyev A, Defrise M, Vanhove C. Pinhole SPECT reconstr uction using blobs and resolution reco very. IEEE Trans Nucl Sci 2006; 53:2719–28. 97. Bal G, Acton PD. Analytical derivation of the point spread function for pinhole collimators. Phys Med Biol 2006;51:4923–50. 98. DiFilippo FP. Geometric characterization of multi-axis multi-pinhole SPECT. Med Phys 2008;35:181–94. 99. DiFilippo FP, Rif fe MJ, Heston WD, et al. Detached multipinhole small animal SPECT de vice with real-time calibration. IEEE Trans Nucl Sci 2006;53:2605–12. 100. Funk T, Kirch DL, Sun MS, et al. Simulation and validation of point spread functions in pinhole SPECT imaging. IEEE Trans Nucl Sci 2006;53:2729–35.

97

101. Gross KA, Kupinski MA, Hesterman JY. A fast model of a multiplepinhole SPECT imaging system. Proc SPIE 2005;5749:118–27. 102. Hsu CH, Huang PC. A geometric system model of f inite aperture in small animal pinhole SPECT imaging. Comput Med Imaging Graph 2006;30:181–5. 103. Hwang A, Taylor C, Seo Y, et al. Improving the quantitative accuracy of a dedicated small animal SPECT/CT scanner . Med Ph ys 2005;32:2133–4. 104. Israel-Jost V, Choquet P, Salmon S, et al. Pinhole SPECT imaging: compact projection/backprojection operator for efficient algebraic reconstruction. IEEE Trans Med Imaging 2006;25:158–67. 105. Metzler SD, Jaszczak RJ . Simultaneous multi-head calibration for pinhole SPECT. IEEE Trans Nucl Sci 2006;53:113–20. 106. Metzler SD, Jaszczak RJ, Greer KL, Bowsher JE. Angular-dependent axial-shift cor rection for pinhole SPECT . IEEE Trans Nucl Sci 2007;54:124–9. 107. Smith MF, Jaszczak RJ. An analytic model of pinhole aper ture penetration for 3D pinhole SPECT reconstr uction. Ph ys Med Biol 1998;43:761–75. 108. Zeniya T, Watabe H, Aoi T, et al. A new reconstruction strategy for image impro vement in pinhole SPECT . Eur J Nucl Med Mol Imaging 2004;31:1166–72. 109. Fryback DG, Thornbury JR. The ef ficacy of diagnostic imaging. Med Decis Making 1991;11:88–94. 110. ICRU. Medical imaging—the assessment of image quality (repor t 54). Bethesda (MD): Inter national Commission on Radiation Units and Measurements; 1996. 111. Barrett HH. Objective assessment of image quality—effects of quantum noise and object variability. J Opt Soc Am A 1990;7:1266–78. 112. Barrett HH, Abbey CK, Clarkson, E. Objecti ve assessment of image quality . III. R OC metrics, ideal obser vers, and lik elihood-generating functions. J Opt Soc Am A 1998;15:1520–35. 113. Barrett HH, Denny JL, Wagner RF, Myers KJ. Objective assessment of image quality. 2. F isher information, Fourier crosstalk and f igures of merit for task-performance. J Opt Soc Am A 1995;12:834–52. 114. Barrett HH, Yao J, Rolland JP, Myers KJ. Model observers for assessment of image quality. Proc Natl Acad Sci U S A 1993;90:9758–65. 115. Meng LJ, Clinthor ne NH. A modif ied unifor m Cramer-Rao bound for multiple pinhole aper ture design. IEEE Trans Med Imaging 2004;23:896–902. 116. Vunckx K, Beque D, Nuyts J, Defrise M. Single and multipinhole collimator design e valuation method for small animal SPECT . 2005 IEEE Nuclear Science Symposium Conference Record 2006; 2223–7. 117. Clarkson E, K upinski MA, Bar rett HH, Furenlid L. A task-based approach to adapti ve and multimodality imaging. Proc IEEE 2008;96:500–11. 118. Beekman FJ, van der Have F, Vastenhouw B, et al. U-SPECT-I: a novel system for submillimeter -resolution tomography with radiolabeled molecules in mice. J Nucl Med 2005;46:1194–200. 119. Accorsi R, Gasparini F , Lanza RC. A coded aper ture for highresolution nuclear medicine planar imaging with a con ventional Anger camera: e xperimental results. IEEE Trans Nucl Sci 2001; 48:2411–7. 120. Garibaldi F, Accorsi R, Cinti MN , et al. Small animal imaging b y single photon emission using pinhole and coded aper ture collimation. IEEE Trans Nucl Sci 2005;52:573–9. 121. Meikle SR, Fulham MJ, Fulton RR, et al. A prototype coded aperture detector for small animal SPECT . IEEE Trans Nucl Sci 2002; 49:2167–71. 122. Schramm NU, Ebel G, Engeland U , et al. High-resolution SPECT using multipinhole collimation. IEEE Trans Nucl Sci 2003; 50:315–20. 123. Barrett HH, Hunter WCJ. Detectors for small-animal SPECT I : overview of technolo gies. In: K upinski MA, Bar rett HH, editors. Small-animal SPECT imaging. New York: Springer; 2005. p.9–48.

98

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

124. Bradley EL, Cella J, Majewski S, et al. A compact gamma camera for biological imaging. IEEE Trans Nucl Sci 2006;53:59–65. 125. Del Guerra A, Bartoli A, Belcari N, et al. P erformance evaluation of the fully engineered YAP-(S)PET scanner for small animal imaging. IEEE Trans Nucl Sci 2006;53:1078–83. 126. Loudos GK, Nikita KS, Giokaris ND , et al. A 3D high-resolution gamma camera for radiophar maceutical studies with small animals. Appl Radiat Isot 2003;58:501–8. 127. McElroy DP, MacDonald LR, Beekman FJ, et al. Performance evaluation of A-SPECT: a high resolution desktop pinhole SPECT system for imaging small animals. IEEE Trans Nucl Sci 2002; 49:2139–47. 128. Schramm N , Wirrwar A, Sonnenber g F, Halling H. Compact high resolution detector for small animal SPECT. IEEE Trans Nucl Sci 2000;47:1163–7. 129. Cinti MN, Scafe R, Pellegrini R, et al. CsI(Tl) micro-pix el scintillation ar ray for ultra-high resolution gamma-ra y imaging. IEEE Trans Nucl Sci 2007;54:469–74. 130. Kim H, Furenlid LR, Cra wford MJ , et al. SemiSPECT : a smallanimal single-photon emission computed tomo graphy (SPECT) imager based on eight cadmium zinc telluride (CZT) detector arrays. Med Phys 2006;33:465–74. 131. Accorsi R, Celentano L, Laccetti P, et al. High-resolution I-125 small animal imaging with a coded aper ture and a h ybrid pixel detector. IEEE Trans Nucl Sci 2008;55:481–90. 132. Peterson TE, Wilson DW, Bar rett HH. Application of silicon strip detectors to small-animal imaging. Nucl Instr um Methods Ph ys Res A 2003;505:608–11. 133. Choong WS, Moses WW, Luke PN, Tindall CS. Design for a highresolution small-animal SPECT system using pix ellated Si(Li) detectors for in vivo 125 I imaging. IEEE Trans Nucl Sci 2005; 52:174–80. 134. Funk T, Shah KS, Despres P . A multipinhole small animal SPECT system with submillimeter spatial resolution. Med Ph ys 2006; 33:1259–68. 135. Fiorini C, Gola A, Zanchi M, et al. Silicon drift photodetectors for scintillation readout in medical imaging. Nucl Instr um Methods Phys Res A 2007;571:126–9. 136. Fiorini C, Perotti F. Small prototype of anger camera with submillimeter position resolution. Rev Sci Instrum 2005;76:(044303)1–8. 137. Soesbe TC, Lewis MA, Richer E, et al. De velopment and evaluation of an EMCCD based gamma camera for pre-clinical SPECT Imaging. IEEE Trans Nucl Sci 2007;54:1516–24. 138. Teo BK, Shestakova I, Sun M, et al. Evaluation of a EMCCD detector for emission-transmission computed tomo graphy. IEEE Trans Nucl Sci 2006;53:2495–9. 139. Meng LJ, Clinthor ne NH, Skinner S, et al. Design and feasibility study of a single photon emission microscope system for small animal I-125 imaging. IEEE Trans Nucl Sci 2006;53:1168–78. 140. Accorsi R, Autiero M, Celentano L, et al. MediSPECT : Single photon emission computed tomography system for small field of view small animal imaging based on a CdTe hybrid pixel detector. Nucl Instrum Methods Phys Res A 2007;571:44–7. 141. Aoi T, Zeniya T, Watabe H, et al. System design and development of a pinhole SPECT system for quantitative functional imaging of small animals. Ann Nucl Med 2006;20:245–51. 142. Cardi CA, Kar p JS, Zixiong C, et al. Pinhole PET (pPET): a multipinhole collimator inser t for small animal SPECT imaging on PET cameras. 2005 IEEE Nuclear Science Symposium Conference Record 2006;1973–6. 143. Furenlid LR, Barrett HH, Wilson DW, et al. FastSPECT II: a secondgeneration high-resolution dynamic SPECT imager . IEEE Trans Nucl Sci 2004;51:631–5. 144. Hesterman JY, Kupinski MA, Bar rett HH, et al. The multi-module, multi-resolution system (M 3 R): a no vel small-animal SPECT system. Med Phys 2007;34:987–93.

145. Hong KJ, Choi Y, Lee SC, et al. A compact SPECT/CT system for small animal imaging. IEEE Trans Nucl Sci 2006;53:2601–4. 146. Kastis GA, Furenlid LR, Wilson D W, et al. Compact CT/SPECT small-animal imaging system. IEEE Trans Nucl Sci 2004;51:63–7. 147. Kubo N, Zhao S, Fujiki Y, et al. Ev aluating performance of a pix el array semiconductor SPECT system for small animal imaging. Ann Nucl Med 2005;19:633–9. 148. Kundu BK, Stolin AV, Pole J, et al. Tri-modality small animal imaging system. IEEE Trans Nucl Sci 2006;53:66–70. 149. Lackas C, Schramm NU, Hoppin JW, et al. T-SPECT: A novel imaging technique for small animal research. IEEE Trans Nucl Sci 2005;52:181–187. 150. McDonald BS, Shokouhi S, Barrett HH, Peterson TE. Multi-energy, single-isotope imaging using stack ed detectors. Nucl Instr um Methods Phys Res A 2007;579:196–9. 151. Peterson TE, Shok ouhi S, Wilson DW, Furenlid LR. Multi-pinhole SPECT imaging with silicon strip detectors. 2005 IEEE Nuclear Science Symposium Conference Record 2006;2752–6. 152. Walrand S, Jamar F, de Jong M, Pauwels S. Evaluation of novel wholebody high-resolution rodent SPECT (Lino view) based on direct acquisition of linogram projections. J Nucl Med 2005;46:1872–80. 153. Zeng GL, Gagnon D . CdZnTe strip detector SPELT imaging with a slit collimator. Phys Med Biol 2004;49:2257–71. 154. Zeniya T, Watabe H, Aoi T, et al. Use of a compact pixellated gamma camera for small animal pinhole SPECT imaging. Ann Nucl Med 2006;20:409–16. 155. Meng LJ, Fu G, Roy EJ, et al. An ultrahigh resolution SPECT system for I-125 mouse brain imaging studies. IEEE Trans Nucl Sci 2007. [Submitted]. 156. Weisenberger AG, Gleason SS, Goddard J , et al. A restraint-free small animal SPECT imaging system with motion tracking. IEEE Trans Nucl Sci 2005;52:638–44. 157. Zingerman Y, Golan H, Gersten A, Moalem A. A compact CT/SPECT system for small-object imaging. Nucl Instrum Methods Phys Res A 2008;584:135–48. 158. Di Domenico G, Cesca N , Zavattini G, et al. CT with a CMOS flat panel detector inte grated on the YAP-(S)PET scanner for in vi vo small animal imaging. Nucl Instr um Methods Ph ys Res A 2007; 571:110–3. 159. Weisenberger AG, Wojcik R, Bradle y EL, et al. SPECT -CT system for small animal imaging. IEEE Trans Nucl Sci 2003;50:74–9. 160. Goetz C, Breton E, Choquet P , et al. SPECT lo w-field MRI system for small-animal imaging. J Nucl Med 2008;49:88–93. 161. Barrett HH, Furenlid L, Freed M, et al. Adaptive SPECT. IEEE Trans Med Imaging 2008;27:775–88. 162. Gonsalves RA. Phase retrie val and di versity in adaptive optics. Opt Eng 1982;21:829–32. 163. Paxman RG, Schulz TJ, Fienup JR. Joint estimation of object and aberrations using phase diversity. J Opt Soc Am A 1992;7:1072–85. 164. Rousset G. Wavefront sensing. In: Roddier F, editor. Adpative optics in astronomy. Cambridge, UK: Cambridge University Press; 1999. 165. Gauss RC, Trahey GE, Soo MS. Adaptive imaging in the breast. Proceedings of the IEEE Inter national Ultrasonics Symposium 1999;1563–9. 166. Ng GC, Freiburger PD, Walker WF, Trahey GE. A speckle target adaptive imaging technique in the presence of distributed aber rations. IEEE Trans Ultrason Ferroelectr Freq Control 1997;44:140–51. 167. Cao Y, Levin DN. Using prior knowledge of human anatomy to constrain MR image acquisition and reconstruction: half k-space and full k-space techniques. Magn Reson Imaging 1997;15:669–77. 168. Zientara GP. F ast imaging techniques for inter ventional MRI. In: Young I, Jolesz F A, editors. Inter ventional MR. London, UK: Martin Dunitz; 1995. p. 25–52. 169. Freed M, Kupinski MA, Furenlid LR, et al. A prototype instr ument for single pinhole small animal adapti ve SPECT imaging. Med Phys 2008;35:1912–25.

7 INSTRUMENTATION AND METHODS TO COMBINE SMALL ANIMAL PET wITH OTHER IMAGING MODALITIES CRAIG S. LEVIN, PHD

INTRODUCTION Earlier chapters of this book discussed positron emission tomography (PET) as a clinical tool. But in recent years, PET has also emer ged as a po werful pre-clinical research tool and is cur rently used in small laborator y animal research to visualize and track certain molecular processes associated with diseases, such as cancer, heart disease, and neurological disorders, in li ving small animal models of disease. in vi vo small animal imaging PET assays enable very sensitive studies of the cellular and molecular bases of disease in its natural state and may be used to guide the discovery and development of new treatments. Ho wever, to be ab le to visualize and quantify the often subtle preferential accumulation of a PET molecular probe in v ery small str uctures within small animals, such as rodents, requires special very high-resolution PET systems. As a result, during the past decade there has been substantial research and energy devoted to the de velopment of pre-clinical PET systems for rodent research. This work has resulted in the de velopment of numerous research prototypes and several commerciall y a vailable high-resolution PET systems that aim to enhance PETs ability to detect, visualize, and quantify low concentrations of probe interacting with its tar get, to more accuratel y study the subtle signatures associated with cellular and molecular processes of interest. However, in modern biology there is often a desire to measure more than one feature or parameter associated with a disease state to gain a more deep understanding of its nature. PET cannot readily achieve this “multiplexing”

as there is onl y one possible wavelength of photon emissions resulting from the positron annihilation process, and the PET radionuclide often has a relati vely long half-life. Thus, the desire to measure dif ferent disease characteristics, such as gene expression, ligand-receptor interactions, or enzymatic acti vity, using PET almost al ways requires multiple studies to be perfor med on distinct da ys. However, there are other in vi vo imaging modalities, such as X-ray computed tomo graphy (CT), single photon emission computed tomography (SPECT), magnetic resonance imaging (MRI), and optical imaging, that, through a multimodality imaging approach, can provide a range of complementary anatomic, ph ysiological, and/or cellular/molecular infor mation and contrast mechanisms to PET for more powerful characterization of disease states. Thus, there has recentl y been much acti vity to inte grate PET instr umentation with other imaging modalities for in-series or e ven simultaneous measurements of multiple complementary parameters of disease. This chapter begins with a basic re view of the principles of PET and cur rent commerciall y a vailable highresolution, small animal PET instr umentation. It then describes some of the research ef forts under w ay to combine small animal PET instrumentation with other imaging modalities, such as CT, SPECT, MRI, and optical imaging. A more in depth review of PET and other imaging modalities is left to the references and other chapters in this book. If these novel efforts to combine PET with other imaging modalities succeed without signif icant performance compromises of the subsystems in volved, this multimodality approach will likely continue to increase PET’s role in the study of disease and development of novel treatments.

99

100

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

PRE-CLINICAL PET There are se veral desirab le features of PET as a small animal (ie, pre-clinical) imaging modality.1,2 First, as with other biological imaging techniques, such as optical imaging, PET can be used to study the cellular and molecular processes associated with disease in li ve animals. Lik e optical imaging methods that e xploit the phenomena of bioluminescence and fluorescence, PET can detect a v ery low concentration of the probe molecule. Ho wever, unlike optical methods, PET can probe these subtle molecular signals deep (man y centimeters) within tissue, with high spatial resolution and contrast resolution, and thus provide quantitatively accurate spatial and temporal probe biodistribution data. The molecular probes used in PET are typically v ery small, lo w mass, biolo gically rele vant molecules, and thus can readil y reach their molecular target(s) without per turbing the natural states of cells and tissues. Finally, since PET is already a clinical standard of care, successful pre-clinical molecular imaging assays that are proven to be safe can be translated to the clinic. PET is currently used in small animal research to noninvasively study the molecular basis of disease and to guide the development of novel probes and molecular-based treatments.1–30 Many ne w molecular probes labeled with positron-emitting radionuclides and associated PET imaging assays are under de velopment to tar get, detect, visualize, and quantify v arious e xtracellular and intracellular molecules and processes associated with diseases, such as cancer, heart disease, and neurolo gical disorders. 1–30 Thus, there is a continuing need to impro ve PET’ s molecular sensitivity, that is, the capability to detect and quantify the subtle signatures associated with molecular tar gets and processes.31,32

REVIEW OF SMALL ANIMAL PET IMAGING SYSTEM TECHNOLOGY Small Animal PET System Design Issues To visualize and accuratel y quantify the biodistribution of a PET molecular probe in small structures within small animal disease models requires special high-resolution PET systems. During the past tw o decades, there has been substantial research in the de velopment of preclinical PET systems for rodent research.33 This work has resulted in the de velopment of numerous small animal PET research prototypes (e g,34–45) and commerciall y available systems 46–52 (Figure 1). Some of these g roups have made attempts to combine their PET technolo gy with other modalities.

With the e xception of one commerciall y a vailable gas-based multiwire propor tional counter (MWPC) system design, 44,52 nearly all small animal PET system designs (e g,34–43,45–51) use v ariations of the same basic position-sensitive annihilation photon detector design concept, which is essentiall y a miniature v ersion of that used in clinical PET system designs. The system is built from detector modules arranged into a ring. Each module comprises ar rays of long and nar row scintillation cr ystal rods with their small ends coupled to a position-sensiti ve photodetector (Figure 2). Each cr ystal is co vered with a very thin reflector, except for the end f ace coupled to the photodetector, so that each is opticall y isolated from its neighbors. An incoming photon that interacts in one of the crystal rods creates a small flash of light. The light pulse reflects off the crystal faces and exits the end of the cr ystal rod, which forms the basic light signal to be collected by the photodetector . In small animal PET systems, the crystals may be directl y coupled to the photodetector(s), coupled through a light diffuser, or coupled through f iber optics. Coupling scintillation cr ystal to f ibers or f iber bundles before the photodetector is not prefer red because fiber coupling al ways introduces v arying le vels of light signal loss and energy/time dispersion. For higher resolution, narrower crystal rods (eg, < 2 mm) are used (Figure 2), but this tends to trap the scintillation light, w hich introduces light collection variations that depend on the photon interaction point as well as further dispersion in the magnitude and transit time of the light signal reaching the photodetector. These f actors af fect scintillation detector signal-to-noise ratio (SNR), which leads to degradation of multiple perfor mance parameters. To some what relie ve this prob lem, in most pre-clinical PET system designs the crystals are made shor t (eg, < 10 mm), with substantial compromise to photon sensitivity, but with higher and less-varying light collection ef ficiency into the photodetector. The photodetector(s) collect and convert the available light signal into electronic pulses, which are amplified and processed to estimate the incoming interaction location, total energy deposited, and ar rival time of each incoming annihilation photon. The photodetectors currently used in small animal PET system designs are almost al ways photomultiplier tubes (PMTs), in particular position-sensitive PMTs (PSPMTs), w hich contain an ar ray of char ge collecting anodes within a single e vacuated v acuum tube. The signals induced on the anodes can be used to localize the scintillation light flash to within 2 mm or less. A variation of the basic scintillation detector described is to use semiconductor photodetectors, such as avalanche photodiodes (APDs), instead of PMTs to read out the scintillation

Instrumentation and Methods to Combine Small Animal PET with Other Ima ging Modalities

Source size Non-colinearity

J

Subtracted Hammersmith

3 FWHM Resolution (mm)

101

Exact HR (CTI)

SHR-2000 (Hamamatsu)

RatPET (UCLA)

BaF2/TMAE (VUB)

Tomitani

SHR-7700 A-PET (Philips) (Hamamatsu) MADPET HRRT ATLAS (NIH) Donner 600 (Munich) (CTI)(Berkeley) microPET (UCLA) ClearPET APD-BGO YAPPET (Ferrara) (Sherbrooke) Ge eXplore microPET II Focus Light Sharing (b, 2.1 mm) MADPET ll LabPET HIDAC

2

1

Electronic Coding (b, 1.1 mm)

LabPET ll

Individual Coupling (b, 0 mm) Crystal Resolution (d/2)

0 0

1

2 3 Crystal Size (mm)

4

5

Figure 1. Intrinsic spatial resolution (quantified by the full-width-at-half-maximum [FWHM] of a point source response along a given direction) at the center achieved with several existing small animal positron emission tomography (PET) system designs versus scintillation crystal array element size used (source size and annihilation photon acolinearity effects de-convolved). The symbol b parametrizes different degrees of light or charge multiplexing inherent to a PET detector technology that contribute toward spatial resolution degradation. Due to scintillation light sharing, charge sharing in the photodetector or electronic readout, inter-crystal scatter, and positron range, intrinsic spatial resolution often does not reach the fundamental limit set by 1/2 the crystal element width d (represented by the dashed line). Nearly all of these designs have significant resolution degradation as the point source moves away from the system center. Adapted from Bloomfield PM et al.34 Courtesy of Roger Lecomte, University of Sherbrooke.

light. A re view of APDs and other ne w detectors and methods proposed is provided in the study by Levin.32

PET System Performance Issues Performance parameters dictate a PET system’s ability to visualize and quantify a molecular signal in the presence of background.31,32 To understand the effects of PET system design compromises that have been made to achieve current PET-based multimodality system designs as w ell as the ef fects of operating another nearb y modality on PET system perfor mance, here w e re view PET system performance basics. There are se veral important parameters of PET system performance, such as photon sensitivity, spatial resolution, energy r esolution, coincidence time r esolution, and count r ate perf ormance. The ener gy and temporal resolutions as w ell as count rate perfor mance w ork together to def ine the instr ument contrast r esolution, which is the ability to dif ferentiate a subtle probe concentration from background or two slightly different concentration levels of probe in adjacent targets. The photon sensitivity, spatial resolution, and contrast resolution work together to define the molecular signal sensitivity of a PET instrument.31,32

2x2x10 mm3

1x1x10 mm3 Photodetector

Figure 2. Nearly all small animal positron emission tomography (PET) system designs use long and narrow scintillation crystal rod elements (left) to assemble an array that is coupled end-on to a position-sensitive photodetector (right). The crystals rods are typically optically isolated with a thin inter-crystal reflector. The PET system comprises rings of these fundamental module subunits.

Photon Sensitivity

In radionuclide imaging, the photon e vents and resulting signals are collected and processed one at a time rather than running in photon integration mode, which is used in X-ray imaging. The system photon sensitivity is the fraction of all coincident 511 k eV photon pairs emitted from the imaging subject that are recorded b y the system and is also refer red to as the coincidence photon detection efficiency. This parameter determines the statistical quality of image data for a given acquisition time. Photon sensitivity impacts image quality because it influences the noise le vel of images reconstr ucted at a desired spatial

102

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

resolution. Photon sensiti vity in PET is impro ved by (1) increasing the probability that emitted photons will traverse detector material, which is known as the geometric efficiency and (2) b y increasing the lik elihood that photons tra versing detector material will be stopped and counted, ter med the intrinsic detection ef ficiency. The geometric efficiency is enhanced b y bringing the detectors as close as possible to the body, and covering the subject with as much detector area as possib le; these factors decrease the chance that photons will escape without traversing detector material. Ho wever, bringing the detectors closer to the subject can lead to position-dependent parallax positioning er rors (hence loss of spatial resolution uniformity) due to photon penetration into the detector elements 32 (Figure 3). The intrinsic detection efficiency is improved by tightly packing the detector elements to gether with little or no spaces, using denser , higher atomic number ( Z) and thick er (longer) detector elements to improve the 511 keV stopping power. Typical small animal PET detector system photon sensiti vities range from < 1% (one coincidence photon pair collected for every 100 emitted) to several percent. Spatial Resolution

The spatial resolution describes a system’s ability in order to distinguish tw o closel y spaced molecular probe concentrations and is important to detect and visualize subtle

molecular signals emitted from miniscule structures. PET spatial resolution is limited by the fact that one is trying to precisely deter mine the location of a positron-emitting radionuclide attached to the probe molecule indirectl y using the line drawn between the two annihilation photon hits in the detectors. Because this line results from tw o electronically deter mined interaction points, this process is called electronic collimation . The spatial resolution is typically measured b y imaging a point-lik e positron radioactive source and measuring its obser ved spread in the reconstructed images. The fundamental spatial resolution limit is dictated b y (1) the positron r ange effect, which is due to v ariations in direction and path length of all the possib le positron trajectories (F igure 4) created from a given point positron source; the e xtent of this resolution degrading effect depends upon the range of energies of the emitted positrons and the medium traversed by the positrons before the y annihilate; (2) the photon acollinearity effect, which is caused by the fact that since the positron and electron are not al ways at rest when they combine, the tw o annihilation photons are not al ways emitted 180° apart, and hence the line def ined by the two detector elements that w ere hit will not al ways pass through the point of the positron-electron annihilation; this acollinearity ef fect on spatial resolution is w orse for larger system diameters; (3) the size of the photon detector element (a.k.a. detector resolution or pixel size), which determines how precisely a system can localize the photon

4.5

Dd

Dd 5 2 mm

ui R

Dx r ui

Radial blurring (mm FWHM)

4

Dd 5 5 mm 3.5

Dd 5 10 mm

3 2.5 2 1.5 1 0.5 0 0

0.5

1

1.5

2

2.5

3

3.5

4

Radial coordinate (cm) Figure 3. A, Depiction of the radial resolution blurring Δx due to photon interaction depth variation in the detector crystal elements for coincident photons emitted from a point source located at radial coordinate r, entering obliquely into two isolated crystals. The resolution blurring is insignificant when the source is at the center (radial coordinate r = 0) and the photons enter normal (θi = 0) to the crystal elements, but away from the center can be shown to roughly vary linearly with r and Δd, the photon interaction depth resolution, and inversely with the system radius R.32 B, The estimated radial resolution degradation Δx (full-width-at-half-maximum [FWHM]) plotted as a function of radial position r of a point source for an 8 cm diameter positron emission tomography (PET) system for different photon interaction depth resolutions Δd.32 Very few PET systems incorporate photon interaction depth measuring capabilities, and thus Δd just equals the length of the crystal elements used, which for small animal PET is typically on the order of 10 mm.

Instrumentation and Methods to Combine Small Animal PET with Other Ima ging Modalities

hits. The size of the detector element used in PET has been gradually decreasing throughout recent y ears to impro ve spatial resolution. At present, typical clinical systems use 4 to 6 mm width detector elements and most small animal systems now use 1.5 to 2 mm width detector elements. Figure 5 shows the combined spatial resolution limit from these three effects as a function of detector pixel size for various system diameters ranging from small animal to clinical PET systems for an 18F point source.53 We see that in principle spatial resolution ma y be impro ved signif icantly by reducing the 511 k eV detector pix el size. This element size dominates spatial resolution for small diameter (< 20 cm) animal PET systems because the acolinearity effect on spatial resolution is minor for small detector ring diameters. Ho wever, de veloping 511 keV photon detector arrays with miniscule detector elements is challenging and typicall y results in perfor mance compromises in other impor tant system parameters. F or example, using a point 18F positron source, a 20 cm detector system diameter for small animal PET , and 1 mm scintillation crystal pixels, Figure 5 indicates that it is possible in principle to achie ve submillimeter full-width-athalf-maximum (FWHM) spatial resolution at the center of the system, pro vided there are enough counts in the acquired data (adequate 511 k eV photon sensiti vity) to reconstruct images at that desired spatial resolution without requiring signif icant smoothing.

103

However, it is very difficult to collect a high fraction of the a vailable light out of nar row (eg, < 2 mm width) and long (> 2 cm) scintillation cr ystals,54 and thus individual cr ystals may not be resolv ed in a detector flood histogram. Fur thermore, this light collection ef ficiency varies as a function of interaction location within the crystal, causing ener gy and ar rival time dispersion of the resulting detector signal, and so ener gy and time resolutions suf fer as a result. 54 Typically, to achie ve acceptable light collection with 1 mm cr ystal pix els, their length is limited to ≤ 10 mm (e g, see 35–43,54), but this significantly compromises the probability of absorbing 511 k eV photons, and hence limits the o verall photon sensitivity performance. Variations in photon interaction depths in cr ystals that can cause energy and arrival time dispersion can also further de grade spatial resolution if the PET detector design does not incor porate a means to measure photon interaction depth (see Figure 3). Energy and Coincidence Time Resolution

Energy resolution is the precision to which one can measure the incoming photon ener gy. Because photons that scatter lose energy, good ener gy resolution means one ma y use a narrow energy window to reduce scatter photon contamination in image data without signif icantly compromising

10 18F

18

POSITRONS

F (Emax 5 635 keV)

5

0

25

4 mm

10,000 EVENTS 2 10 2 10 2500

25 0 5 END POINT COORDINATE (mm) 100,000 EVENTS

EVENTS

2000

100 TRACKS

10

18

F

1500 1000

0.102 mm FWHM

500

1.03 mm FWTM

0

4 mm

22

21 0 1 2 END POINT COORDINATE (mm)

Figure 4. (Left) 100 18F positron tracks in water emanating from a point source. (Top right) Scatter plot of endpoints for 10,000 positron tracks in water. Annihilation photons are created at or very near the endpoints. (Bottom right) Histogram of x-coordinates of 18F positron endpoints shows a cusp-like function with only ~100 micron full-width-at-half-maximum (FWHM) (full-width-at-tenth-maximum [FWTM] is ~1 mm). Adapted from Levin CS and Hoffman EJ.53

104

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

System Diameter

SPATIAL RESOLUTION (mm)

6

FWHM 0 cm 10 cm 20 cm 80 cm

FWTM 0 cm 10 cm 20 cm 80 cm

18F

5 4 3 2 1 0 0

1 2 3 4 DETECTOR ELEMENT WIDTH (mm)

5

Figure 5. Spatial resolution limit (full-width-at-half-maximum [FWHM] and full-width-at-tenth-maximum [FWTM]) determined from three fundamental effects of positron range, photon acolinearity, and crystal element size for an 18F point source in water versus the latter for different system diameters. In principle, using an 18F point source placed at the system center, submillimeter spatial resolution is possible. Adapted from Levin CS and Hoffman EJ.53

photon sensitivity. A nar row energy window also helps to reduce the rate of random (unpaired) photon contamination because man y of these photons also under go scatter. The coincidence time r esolution determines how well one can decide w hether tw o coincident photons tr uly are detected simultaneously. Analogous to benef its of good energy resolution, good coincidence time resolution means one may use a nar row time windo w to reduce random e vents without compromising photon sensiti vity. Ener gy and coincidence time resolution are impro ved by using scintillation cr ystals that generate brighter and faster light pulses, low noise photodetectors, and by collecting a higher and constant fraction of the a vailable scintillation light into the photodetector to create larger, nonvarying, more robust electronic pulses. A typical value for PET ener gy resolution is 25% FWHM at 511 keV and 3 ns FWHM for coincidence time resolution. A figure of merit for SNR in PET images is provided by the noise-equivalent count rate (NECR), defined by T 2/(T + S + kR), where S and R are the photon scatter and random coincidence rates, respecti vely, and T is the tr ue, non-scattered, non-random photon coincidence rate (a.k.a. the signal). The factor k in front of R equals two when a separate measurement is made to estimate R, otherwise it is unity. Count Rate Performance

Each detector signal recorded in a PET system has a finite processing time. If too many photons hit the detectors in a

given time, the front-end photon detectors or subsequent acquisition electronics in the PET system can saturate due to piling up of more than one electronic detector pulse within the required electronic signal processing duration. Typically, the de gree of pile up is limited b y the photon detector signal processing time, w hich depends upon the decay time of the scintillation cr ystal, the ef fective integration time of the electronics, and the photon e vent rate seen by the detector. For example, suppose a 10 mCi (370 MBq) point source is placed at the center of a PET system with 10 detector modules providing 5% coincidence photon detection ef ficiency. Then, the a verage photon e vent rate per detector module is roughly 3.7 × 108 (radionuclide decays per second) × 2 (photons per event) × 0.05 (photon sensitivity) ÷ 10 (photon detector modules) = 3.7 × 106 counts per second. If each system detector module requires 1 µs of processing time per event, there could be significant pile up of e vents. For a gi ven system photon sensitivity, for the best count rate performance, the system should use scintillation cr ystals with fast decay time, fast processing electronics, and limited activity within the sensitive field-of-view (FoV).

Present Commercially Available Small Animal PET System Designs Present commercially available small animal PET system designs are technolo gy transfers from academic research system developments33 (see Figure 1), and nearl y all are variations of the same scintillation detector theme described. As will be seen, it has been dif ficult to obtain exceptional perfor mance in all parameters simultaneously; the majority of cur rently a vailable small animal PET system designs use trade-of fs betw een parameters, such as photon sensiti vity, spatial resolution, scintillation light collection ef ficiency, energy resolution, and coincidence time resolution. F or example, no system design to date can provide the following “wish list” of small animal PET system perfor mance values: ≥ 15% photon sensitivity, ≤ 1 mm FWHM spatial resolution that is unifor m throughout the sensitive FoV, > 90% scintillation light collection ef ficiency, ≤ 13% FWHM ener gy resolution at 511 keV, and ≤ 3 ns FWHM coincidence time resolution. To date, attempts to achie ve one of these perfor mance parameters, for e xample, 1 mm spatial resolution, has resulted in the signif icant compromise of one or more of the others. All of the current commercially available scintillation detector-based system designs ha ve nonunifor m spatial resolution that rapidly degrades with distance from the center, and only a few percent photon sensitivity using a narrow energy window (high energy threshold).

Instrumentation and Methods to Combine Small Animal PET with Other Ima ging Modalities

A technology development at University of California, Los Angeles in the late 1990s 35 was transfer red to what has become the Siemens microPET product line.46 The detector module design uses an optical f iber interface between a lutetium-oxyorthosilicate (LSO) scintillation crystal array and a multianode PSPMT to enab le a high inter -module packing fraction. The cur rent generation system, the In veon, uses b lock modules comprising 20 × 20 arrays of 1.5 × 1.5 × 10 mm 3 LSO crystals coupled through a tapered light guide into PSPMTs; these modules are ar ranged in a 16.1 cm diameter ring, with a 12 cm diameter bore and 10 cm transaxial and 12.7 cm axial FoV. The spatial resolution and photon sensiti vity are 1.4 mm FWHM and 10% (100 keV threshold), respectively, at the system center. The detector scheme developed at University of Pennsylvania37 went into the Philips Mosaic small animal PET system, 47 which uses just under 17,000 discrete 2 × 2 × 10 mm 3 gadolinium orthosilicate (GSO) cr ystals coupled through a single lar ge annular light dif fuser to a bank of standard PMTs. The system has a 21 cm diameter bore and a 12.8 cm transaxial and 11.6 cm axial FoV. Note that unlik e the other scintillation cr ystal designs, the Mosaic is not par titioned into indi vidual block modules. The reconstr ucted spatial resolution ranges betw een 2.7 mm FWHM at the center and 3.2 mm FWHM at a radial offset of 45 mm from the center with an absolute photon sensitivity of 0.65% (410 keV threshold).55 Work at National Institutes of Health, Depar tment of Nuclear Medicine 36 was translated to a Spanish compan y called Suinsa and subsequentl y became w hat is no w known as the General Electric (GE) eXplore Vista PET system.48 This detector concept uses a tw o-layer scintillation crystal array or phoswich to allow coarse estimation of photon depth of interaction (DOI). Each la yer is made from a dif ferent scintillation cr ystal type (lutetiumyttrium-oxyorthosilicate [LYSO] and GSO) with a different scintillation light deca y constant. Pulse shape discrimination methods are used to determine which array layer was hit by an incoming photon. The system has two rings of 6084 LYSO-GSO crystals arranged into modules, each containing 13 × 13 ar rays of 1.5 × 1.5 × 15 mm 3 LYSO-GSO pairs. The detector diameter is 11.8 cm, with a 6 cm transaxial and 4.6 cm axial FoV. The reconstructed spatial resolution is 1.6 mm FWHM at the center with a 4% absolute photon sensitivity (250 keV threshold).56 A research concept conceived at University of Texas38 has been incor porated into a small animal PET system built by Gamma Medica-Ideas49 (and currently distributed by GE). The system uses bismuth ger minate (BGO) scintillation cr ystals with a light sharing technolo gy that

105

allows them to be read out b y standard PMTs. The edge crystals of each module are trapezoidal to promote high inter-detector module cr ystal packing fraction (reduce gaps between modules). The system specif ications are an axial FoV of 11.8 cm, a useful transaxial F oV of 10 cm, and a center point source photon sensiti vity and spatial resolution of 10% (100 k eV threshold) and 1.5 mm, respectively. A University of Sherbrooke development39 was transferred to Advanced Molecular Imaging, Inc. under the name “LabPET” and more recentl y to Gamma MedicaIdeas.49 The basic detector design uses an array of two different scintillation cr ystal elements (L YSO and LGSO) with different decay times coupled to a singleAPD device. Using pulse shape discrimination enab les the identif ication of w hich of the tw o crystals is hit per module b y an incoming photon, using only one electronic readout channel. The resulting system has a 15.6 cm detector ring diameter with an 11 cm diameter aper ture, 10 cm useful transaxial FoV, and either a 3.75, 7.5 or 11.25 cm axial FoV with 1536, 3072 or 4608 APDs, respecti vely. The specified center point source photon sensiti vity is 2, 4, and 6%, respecti vely, for the three a vailable axial F oVs. The spatial resolution is 1.5 mm FWHM at the center . The ClearPET LYSO/lutetium-yttrium aluminum perovskite (LuYAP) phoswich scanner51 is a technology transfer from the ideas de veloped within the ClearPET g roup41 of the Cr ystal Clear Collaboration (CCC), a scintillation crystal research or ganization based at Or ganisation Européenne pour la Recherche Nucléaire (CERN), an international particle physics laboratory located in Geneva, Switzerland. The system has tw o adjustab le detector diameters, 13.5 and 28.5 cm, with an open gantry space of 12.5 and 22.0 cm, respecti vely. The phos wich detectors comprise two layers of 2 × 2 × 10 mm 3 crystals of LYSO and LuYAP coupled to PSPMTs. Due to signif icant gaps between the detectors, the system rotates around the subject to enable full angular sampling. At the system center , the specified spatial resolution is 1.5 mm and the absolute photon sensitivity is 3.8%. The Y AP-PET system50 is an of f-shoot from University of F errara and Pisa. 40 The system comprises four rotating heads spaced 15 cm apar t, each with an active area of 4 × 4 cm 2, containing a 20 × 20 ar ray of 2 × 2 × 30 mm3 optically isolated YAP crystals coupled to PSPMTs, forming a 4 cm useful transaxial and axial FoV. The reconstructed spatial resolution and absolute photon sensitivity are 1.8 mm FWHM and 1.7% (50 keV threshold) for a centered point source, respecti vely. The Oxford Positron Systems high-density avalanche chamber (HID AC) PET system 52 is a culmination of

106

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

years of high-resolution gas MWPC imaging system developments at CERN 44 that were modified and ref ined for small animal imaging. 57 In this position-sensitive gas ionization chamber , the annihilation photons are converted b y lead cathode plates into electrons, w hich are subsequently detected and localized b y collecting their ionization generated as the y drift and a valanche in the gas. The lead cathode plates are for med b y la yers of laminated lead containing interlea ved insulated sheets, mechanically drilled with ~200,000 holes, each 0.4 mm diameter, each of w hich acts as an independent detector element. The hole pitch, ~0.5 mm, sets the limit to the intrinsic resolution. The quad-HID AC consists of four large rotating heads, each with eight HIDAC detector layers with an active area of 17 × 28 cm2, with head separation of 17 cm. These layers provide information about the photon interaction depth. The transaxial and axial F oV are 17 and 28 cm, respecti vely. The reconstructed spatial resolution is 1.0 mm in all three-dimensions (3D). Due to this capability to precisely localize the interaction coordinates of incoming photons in 3D, the spatial resolution is uniform throughout the F oV (not just at the center [see Figure 3], w hich is another distinguishing feature of the HIDAC technology), and the absolute photon sensiti vity is 1.8% for an effective 200 keV threshold. However, due to signif icant dispersion of the signal in the lead la yers and detection process, the ener gy and coincidence time resolutions are dif ficult to assess. Two v ariants for this technology are commerciall y a vailable comprising 16 and 32-modules.58

MULTIMODALITY PET IMAGING Introduction Multimodal in vivo imaging enables the measurement of multiple, complementar y image contrast mechanisms to study anatomic, ph ysiologic, cellular , and/or molecular pathways of disease in li ving subjects. The remainder of this chapter focuses on a discussion of instr umentation and algorithms for combining PET with other imaging modalities to f acilitate biolo gical studies in small animals. Multimodality molecular imaging with PET has become an essential tool to study the molecular pathways of disease in li ving subjects, to aid in the disco very and testing of novel therapeutic approaches in animal models of human disease, and for the development of new molecular contrast agents. Multimodality imaging requires easy and con venient access to multiple imaging systems and robust registration

of images generated b y the indi vidual modalities. To address these concerns, industry and academic institutions are developing v arious approaches specif ically designed for imaging small animals that allows one to combine the power of PET with other imaging modalities, such as X-ray CT, MRI, and optical imaging. The focus of this chapter is on tr uly inte grated or “h ybrid” multimodal systems that have been developed for small animal imaging. This approach is more complex but has the advantage of a higher lik elihood of successful image re gistration because there is less chance of subject movement or physiological changes when multimodal studies are performed within the same hybrid system and there is a signif icantly shorter time span betw een multimodal studies, or the y may even be run simultaneously. The hybrid approach also provides a high le vel of con venience for biolo gical researchers because the multimodal studies are a vailable immediately without having to move the subject betw een systems in dif ferent rooms or buildings or requiring the scheduling of multiple imaging studies on dif ferent machines. Although the commerciall y a vailable multimodality hybrid clinical systems are either PET/CT or SPECT/CT scanners, the g reater system fle xibility enab led b y small animal imaging research has resulted in the de velopment of several high-resolution dual- and tri-modality systems, such as PET/CT,59–61 PET/SPECT/CT,62–64 PET/MRI,65–85 and PET/optical. 86–89 Such multimodality systems f acilitate a range of in vi vo strategies to obtain rich, cor relative information about the molecular basis of disease and enhance inter pretation and quantif ication capabilities of data from the individual modalities involved.

Software-Based Fusion of Multimodality Images Over the years, many techniques have been developed for multimodal clinical image re gistration.90,91 However, although there ha ve been a fe w incidental studies, 92–98 in general image registration algorithms used in human studies have not been well characterized for small animal imaging. Softw are fusion of data from tw o separate imaging modalities is possible with the help of anatomical or f iducial mark ers (e g, small, dense str uctures or radionuclide sources placed at the subject contours) that allo w spatial registration of the two image volumes, but such efforts are most successful for studies of or gans and tissues that do not move with time, such as the brain. 99 As is the case for clinical imaging re gistration, some techniques for small animal imaging have used external fiducial markers to aid

Instrumentation and Methods to Combine Small Animal PET with Other Ima ging Modalities

the re gistration process, w hereas other researchers ha ve attempted fully automated algorithms that do not in volve user interaction. 92–98 State-of-the-art image re gistration techniques enable automatic image re gistration through a rigid body transfor mation, for e xample, using mutual information-based criteria, ignoring or gan defor mation due to ef fects, such as cardiac and respirator y motion or spatial linearity differences between modalities. There has also been progress in the development of nonrigid registration algorithms that can compensate for defor mation perceived by different imaging modalities or re gister images from dif ferent subjects. 100 However, despite some progress, man y image re gistration prob lems par ticularly for small animal imaging remain unsolv ed, and this is likely to continue to be an acti ve f ield of research in the future.

Hardware-Based Approaches to Combine Small Animal PET with Other Modalities The reco gnized limitations and incon veniences of software-based image fusion led to the development of hardware approaches to f acilitate more accurate re gistration of image data from multiple modalities. Some g roups have simply developed special fixtures that can be rigidly and reproducibly mounted on the imaging beds of distinct small animal scanners (eg, PET and CT). For example, in the study b y Cho w and colleagues, 101 using such a movable f ixture and a 3D g rid phantom with 1288 lines, a spatial transfor mation matrix for re gistration w as derived using a 15-parameter perspective model, yielding an average registration error between PET and CT mouse bone scans of less than 0.335 mm. The reproducibility and robustness of the system also enab led the use of CT images for accurate attenuation cor rection of the PET data to increase quantitative accuracy.102 The other approach for multimodality imaging is to develop a system that combines more than one type of imaging technolo gy into one inte grated unit. Such a hybrid system allo ws simultaneous and/or sequential acquisitions with the dif ferent modalities, ideall y without compromising the performance of either system, and provides additional cor relative infor mation that cannot be obtained as easily using the two modalities separately. Combined modality instr umentation that is capab le of truly simultaneous acquisition of the distinct modalities allows multiple, time-cor related biolo gical measurements of the same disease state. Such a h ybrid system may also enable simultaneous imaging of multiple molecular tar gets if dif ferent molecular contrast agents/mechanisms can be used. F or imaging methods

107

that are time-consuming, such as PET and MRI, w hich can last ~30 to 45 minutes each, simultaneous imaging also provides the benef it of reduced imaging time compared to performing the two studies sequentially. The standard design of PET/CT systems is an example of an approach that enab les onl y sequential acquisitions with the two modalities. PET is used to measure the modified cellular or molecular characteristics of a diseased state, and CT is used in series to pro vide high-resolution visualization of the cor responding anatomy w here the diseased tissue (eg, cancer) resides, without compromising the PET measurement. Adding CT to PET has the additional benefit of enhancing PETs accuracy and throughput by facilitating a rapid, low-noise, accurate estimate of photon attenuation coefficients,103,104 a feature that in the pre-clinical arena is perhaps more impor tant for rats than for mice due to their larger size and hence increased photon attenuation. 105 In addition to pre-clinical PET/CT designs, 59–61 several investigators are investigating dual-modality systems that combine PET with other v arious imaging technologies, such as PET and SPECT,62–64,106 PET and MRI,65–85 and PET and optical imaging.86–89 Moreover, tri-modality pre-clinical systems inte grated in a single gantr y for PET/SPECT/CT have also been achieved.62–64 As one lear ns about multimodal system designs under investigation, impor tant questions should be k ept in mind: Does the integration of multiple systems provide new or more accurate infor mation than w as otherwise available with the distinct stand-alone systems? Does integration compromise either system’ s perfor mance compared to when they stand alone? That is, were any of the indi vidual system perfor mance specif ications compromised to f acilitate integration? Should inte gration of multiple modalities occur e ven at the cost of signif icant performance degradation/limitation of one or more of the modalities involved? After integration, is there an y measurable mutual interference betw een modalities that further degrades their performance during operation? In the previous section, we learned that an impor tant factor that strongly affects PET system performance is the photon sensiti vity. Thus, an y multimodal PET system design with a small axial FoV coverage, thin (short) detector cr ystals, or gaps betw een detector modules/elements will compromise the number of 511 k eV photons recorded, and therefore, image quality and quantitati ve accuracy. We also lear ned that the SNR and ener gy/time dispersion of the scintillation light signal measured by the PET system detectors are impor tant f actors af fecting nearly all perfor mance parameters to dif ferent de grees. Thus, any design that compromises this light signal will also compromise the PET system perfor mance.

108

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

PET/CT

Despite the popularity and widespread interest in clinical dual-modality PET/CT imaging, only a few small animal PET/CT prototypes ha ve been de veloped at academic institutions,39,59–61 although more are a vailable commercially.46–49 Excellent micro-CT images of live animals are being obtained using cone-beam X-ra y CT imaging and reconstruction.107 In X-ra y CT, the spatial resolution is limited by a convolution of the X-ray beam focal spot size produced b y the tube aper ture and the X-ra y detector intrinsic spatial resolution. Image SNR at a gi ven reconstructed resolution is deter mined by the photon statistics of the detected X-ray flux. However, it has also been recognized for high-resolution small animal CT that dose issues are critical. 108 The Uni versity of Califor nia at Da vis g roup de veloped a prototype microCT/microPET dual-modality small animal imaging system for combining anatomic and molecular imaging of the mouse. 59,60 The microPET detectors used 9 × 9 arrays of 3 × 3 × 20 mm3 LSO scintillator cr ystal elements coupled through a f iber-optic taper to a PSPMT . These 60 × 60 mm 2 flat-panel PET detectors were placed on opposite sides and rotated about the animal with the annihilation photons from the positron emission detected in electronic coincidence. The X-ray CT system used a < 75 micron microfocus X-ra y tube and a 6.6 × 5.5 cm 2 amorphous selenium detector array of 1024 × 832 pixels, each 66 microns, coupled to a complementary metal-o xide semiconductor (CMOS) flat-panel readout ar ray.109 The same g roup later de veloped a more adv anced microCT/microPET II system. The microPET II is a fullring tomograph comprising three rings of 30detector blocks made from 14 × 14 arrays of 1 × 1 × 12 mm3 LSO crystals coupled through a tapered fiber-optic bundle to PSPMTs. In total, the system has nearl y 18,000 cr ystal elements and 52 million response lines for med between all cr ystal elements. The bore diameter is 16.0 cm with axial and transaxial F oVs of 4.9 and 8.5 cm, respecti vely. The microCT system comprises a 50 kVp, 1.5 mA f ixed tungsten anode X-ray tube, with 70 µm focal spot size, and 5 × 5 cm2 position-sensitive photodetector comprising a 48 µm pix el CMOS array and a fast gadolinium oxysulfide (GOS) intensifying screen.60 The system is mounted on a flexible C-arm gantry design with adjustab le detector positioning and is integrated in series (on the back) with the microPET II scanner (Figure 6). 110,111 Although the inte gration of PET and CT is in series, and onl y sequential acquisitions are performed, there is no de gradation of either modality’s performance compared to their stand-alone operation.

The pre viously described LabPET scanner de veloped by the group from Sherbrooke61 and commercialized by Gamma Medica-Ideas, Inc. comprises detector modules made from tw o rows of four cr ystals, one of LYSO and the other of LGSO, each with dimensions of 2 × 2 × 10 mm 3. The system has been enhanced b y adding adv anced X-ra y CT capability for acquiring anatomical images using the PET detectors. That is, the X-rays from a micro-focus source are collected using the same detector cr ystals and electronics to enab le simultaneous PET/CT scanning (~1 mm reconstr ucted spatial resolution for both). The claimed advantages of truly simultaneous PET and CT acquisitions are (1) simplified and more accurate re gistration because the subject does not ha ve relative motion betw een the tw o modalities, (2) reduced footprint compared to a PET/CT with separate detection subsystems, (3) the FoV is more accessib le for inter ventions during imaging, and (4) the system w ould cost less because an expensive, separate high-resolution CT detector system and gantry are not required. Because PET already uses

microCT microPET II

Figure 6. The UC Davis high-resolution small animal positron emission tomography (PET)/computed tomography (CT) system uses the series approach to multimodality system integration with the PET in front and the CT in back. Reproduced with permission from Liang H et al.60

Instrumentation and Methods to Combine Small Animal PET with Other Ima ging Modalities

photon-counting electronics, the same operating mode is used for the CT acquisition. This can be achieved by sampling the analog signal waveform using high-speed analog-to-digital con verters and digital processing in field pro grammable gate ar rays. The possibility to count and ener gy discriminate indi vidual X-ra y photons in CT mode39,61 provides foreseen benefits over the standard integration mode used in most X-ra y CT system designs, such as (1) a lo wer relati ve contribution from electronic noise that can be discriminated via a low-energy signal threshold and (2) a more appropriate energy w eighting independent of the X-ra y photon energy transmitted through tissue compared to the proportional w eighting of inte grating detectors that are typically used in X-ray CT. These factors could imply a lower dose required to obtain a gi ven image quality . The parallel architecture and f ast digital processing electronics allo w high count rates (25% deadtime or count loss at an e vent detection rate of 1.5 × 106 cps) for both PET and CT modes, whereas the modularity of the system design allo ws one to e xtend the number of channels up to 10 4 or more.

PET/SPECT/CT

There have been efforts to combine three modalities (PET, SPECT, and CT) and record quasi-simultaneous, complementary information gathered from each. One such example is the FLEX Triumph™ system commerciall y available from Gamma Medica-Ideas, Inc. (Nor thridge, CA).63 Proof-of-principle examples of its tri-modal imaging capabilities are shown in Figure 7 for normal mice. It has been claimed that the LabPET APD-based detector module proposed in 64 is another good technolo gy for the design of compact tri-modality (PET/SPECT/CT) imaging systems. The YAP-(S)PET scanner50 can perform both PET and SPECT studies on small animals. 106 Operating the scanner in SPECT mode is possib le using the same detector conf iguration b y mounting a high-resolution parallel-hole collimator in front of each detector panel and acquiring a standard circular orbit acquisition of photons. Although these multimodality systems require separate acquisitions of the PET , SPECT, and/or CT, there are no mutual interference ef fects. At this point in time, it is unclear w hat impor tant biomedical applications will require both PET and SPECT measurements, but it will be likely involve the need for a measurement of two complementary molecular signatures of disease in a living subject that is onl y a vailable with a separate PET and SPECT tracer and image acquisition.

109

PET/MRI

PET data pro vide high molecular sensiti vity. Magnetic resonance (MR) data provide a rich variety of anatomical and ph ysiological contrast mechanisms, through a wide range of a vailable pulse sequences, to study disease processes without using ionizing radiation. Thus, there has been considerab le interest to combine the tw o technologies65–85 and the topic has been w ell reviewed.33,74,75,77 The majority of MR pulse sequences involve relatively long acquisitions compared to CT. Thus, developing a combined system with a PET system in series with an MR system, with sequential acquisitions similar to most PET/CT configurations, would be impractical due to long acquisition times.As a result, unlike most system designs that combine PET and CT , nearl y all designs to inte grate PET with MR under in vestigation enable truly simultaneous (temporally and spatially registered) acquisitions of both PET and MR data. In all designs to date, the PET system inser t resides outside of the radio frequenc y (RF) coils and inside the g radient coils as depicted in Figure 8. Aside from the wide v ariety of multiparameter , dynamic contrast mechanisms a vailable with MRI, combining PET with MR has a couple of signif icant benefits over PET/CT. MRI does not use ionizing radiation. Dose reduction is especiall y impor tant for patients that are undergoing multiple repeated studies for follo w-up over time. Also, in soft tissue, MR demonstrates better visualization of contrast differences (a.k.a. contrast resolution). There are se veral impor tant challenges that must be overcome in designing and operating a combined PET/MRI imaging system. As we saw in the previous sections, nearl y all small animal PET detector designs proposed to date use PMTs w hose perfor mance can be seriously affected in the presence of extremely strong magnetic f ields produced b y moder n MRI scanners. Fur thermore, an MRI scanner relies on rapidl y switching gradient magnetic fields and RF signals to produce the MR image. The presence of the magnetic f ield gradients and RF signals cer tainly could disr upt the perfor mance of a PMT based PET detector if the y w ere located within or e ven adjacent to the magnet of the MRI system. Similarl y, the operation of the MRI system relies on a very homogenous, uniform, and stab le magnetic f ield to produce the MR image. The introduction of radiation detectors, electronics, and other bulk materials can per turb the main magnetic field, RF pulses, and g radient f ields in a w ay that introduces artifacts in the MR image. Early studies to design MR-compatib le PET units were made b y the Uni versity of Minnesota. 112–114 In the

110

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

CT/PET/SPECT

Figure 7. (Left) Tri-modality computed tomography (CT)/positron emission tomography (PET)/single photon emission computed tomography (SPECT) image of a normal mouse acquired with a single, integrated CT-PET/SPECT system—the FLEX Pre-clinical Imaging System by Gamma Medica-Ideas49 that contains separate CT, PET, and SPECT subsystems. The normal mouse was injected initially with 18F sodium fluoride and imaged with both the CT and PET subsystems. Subsequently, it was injected with 99mTc-methylene diphosphonate (MDP) through a catheter (without moving the animal from the bed) and imaged immediately after injection. The mouse skeletal structure is displayed with surface rendering of the CT image (cream color). The PET image (shown in green) displays bone uptake and clearance through the bladder. The SPECT imaging (shown in orange) was completed within a short time period compared to the bone uptake and clearance, thus the image shows the subject’s vasculature, heart, kidneys, and liver (which has a high blood volume fraction). This image indicates that MDP is still in the blood stream and also shows an early phase of clearance through the kidneys. (Middle and Right) PET/CT images of normal mice acquired with a single, integrated CT-PET-SPECT system—the FLEX Triumph Pre-clinical Imaging System by Gamma Medica-Ideas equipped with CT and LabPET subsystems. The LabPET system uses a fundamentally different PET detector design than the standard FLEX PET subsystem (different scintillation crystals and avalanche photodiode [APD] photodetectors rather than photomultiplier tubes [PMTs]). The middle image is a PET-CT overlay of a mouse injected with fluorodeoxyglucose (FDG), showing myocardial metabolic activity. The right image is a mouse injected with 18F sodium fluoride displayed in overlaid volume rendering. The integrated design of the FLEX Triumph scanner facilitates co-registration of the two image volumes using a pre-calibrated spatial transformation that enables the PET image dynamic range to match that of the CT images. Courtesy of Koji Iwata, Gamma Medica-Ideas.49

following years University of Califor nia, Los Angeles (UCLA)65–67 developed the f irst system conf igured as a 5.6 cm diameter, single scintillation cr ystal ring inser t for a 1.5 T MR system. The single ring comprised 72 2 × 2 × 10 mm3 LSO crystals each coupled side ways to 4 m long optical f ibers that were read out b y a single PSPMT coupled to readout electronics, both located outside the system bore. By k eeping the radiation-sensiti ve elements of the detector within the MR system, w hile operating the photodetector and electronics a way from the magnetic f ield, the combined system could perfor m simultaneous PET/MR imaging without measurab le mutual interaction effects.66 A main drawback for this design is that coupling to fibers results in loss of a signif icant fraction of the light signal, affecting energy and time resolution perfor mance and de grading the ability to identify w hich cr ystal

absorbed a photon. In par ticular, the side ways manner in which the scintillation cr ystals were coupled to the f ibers in67 as well as light attenuation within the f ibers yielded low scintillation light collection ef ficiency and e vent energy/time dispersion, resulting in an energy resolution of 45% FWHM at 511 keV and coincidence time resolution of 26 ns FWHM. Another main drawback is that the single crystal ring and data acquisition design resulted in extremely poor photon sensiti vity (~10 −4). Ne vertheless, the indi vidual cr ystals w ere resolv ed suf ficiently b y the PSPMT to form a single image slice with 2.1 mm FWHM spatial resolution near the system center and produced the first simultaneously acquired PET and MR images. 67 Collaborators at Guys and St. Thomas Hospital, London placed the UCLA-built system inside of a 9.4 T nuclear magnetic resonance (NMR) spectrometer to

Instrumentation and Methods to Combine Small Animal PET with Other Ima ging Modalities

PET Insert Gradient Coils

RF Transmit and Receive Coil

Main Super Conducting Magnet

Figure 8. In nearly all integrated positron emission tomography (PET)/magnetic resonance (MR) system designs, the PET system insert is placed outside the radio frequency (RF) transmit/receive coils and inside the gradient coils. Courtesy of Peter Olcott, Stanford University.

study metabolism in an isolated , perfused rat heart model. 32P-NMR spectra from phosphor ylated glucose were acquired simultaneousl y with PET images of 18 F-fluorodeoxyglucose (FDG) uptak e in the myocardium68,69 and the tw o data w ere compared for consistency gi ven the kno wn biochemistr y of glucose metabolism. The group planned to extend this MR-compatible f iber-coupled PET scanner concept to de velop another single ring of 480 LSO crystals arranged in three layers (160 cr ystals per layer) for photon DOI measurement capability, with a ring diameter of 11.2 cm, and a 5 cm diameter useful FoV that is large enough to accommodate an animal within a stereotactic frame.70 The chosen fiber length of 3.25 m resulted in roughl y 70% light loss70 (other such long-f iber-coupled designs had up to ~90% light loss85). However, it was argued that the number of scintillation photons w as suf ficiently abo ve the noise le vel of the PSPMT to deter mine the cr ystal of interaction within the detector block, although it is clear that energy and time resolutions, and therefore contrast resolution and quantif ication capabilities w ould be diminished significantly with the long f iber coupling. Other subsequent approaches based on PMT -based PET detectors used either the same f iber-coupling-based principles79,80 or relied upon more comple x magnet designs, including a split magnet 83 or a f ield-cycled MRI85 and ha ve been re viewed else where.74,75,77 In the latter case, the PET and MR acquisitions are interlea ved between the MR f ield cycling, and thus the two data sets are not acquired simultaneously. New PET system designs have arisen in recent years due to the availability of new semiconductor photodetector technolo gies and other adv ances that for PET/MR

111

allowed one to a void using PMTs and reduce the length of or eliminate the need for optical f iber coupling. Se veral research g roups are cur rently investigating methods to integrate a PET system into an MRI scanner by designing detectors made from relatively nonmagnetic materials that can be placed within the magnetic field of an MRI or MR spectroscopy system. Here w e highlight a fe w PET systems conf igured with suitab le semiconductor photodetectors that are insensitive to magnetic f ields and can consequently be operated within an MR system for combined PET/MR imaging. Some g roups ha ve tested APDs within a high magnetic f ield and ha ve produced PET and MR images that appear to be free of se vere ar tifacts and distortion.71–78 Other g roups are studying Silicon PMTs (SiPMs),115–119 which are a relati vely new type of semiconductor photodetector (Figure 9) that show promise in the design of combined PET/MR scanners; in addition to being insensitive to magnetic f ields, these de vices have much lar ger signals than APDs, approaching that of PMTs (hence the name), and lower operating bias, which relaxes readout electronics requirements, and thus facilitates operation inside an MRI system. 116–119 The UCLA design w as improved upon at UC Da vis (in collaboration with Califor nia Institute of Technology [Caltech]) by using shor ter, bent optical f ibers and position-sensitive APDs (PSAPDs) (Figure 10) to replace the long optical f ibers and PSPMTs, respecti vely. The PET system inser t consists of one 60 mm diameter ring of detector modules, each comprising an 8 × 8 array of 1.43 × 1.43 × 6 mm 3 LSO cr ystals (1.5 mm pitch) coupled through a 10 cm long f iber-optic b undle that bends 90 degrees from the cr ystal rod ends do wn the axis and away from the sensiti ve region of the MR system to the PSAPDs. The PSAPDs are read out b y low noise charge sensitive preamplifiers that drive long coaxial cables to a

Figure 9. Compact silicon photomultiplier (SPM) array (left) configured into 4 × 4 pixels each with 3 × 3 mm2 area and comprising 3640 Geiger-mode avalanche photodiode cells (right) with a 35 micron active area and 42 micron pitch. This is a promising new photodetector technology for PET scintillation detector arrays operating inside a strong magnetic field. Courtesy of Joe O’Keeffe, SensL USA (Mountain View, CA).

112

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

data acquisition system residing outside the MR system. The useful axial and transaxial F oV are 12 and 35 mm, respectively. The entire inser t is encased in copper for electromagnetic shielding with an outer diameter of 11.8 cm, f itting tightly within the gradient coil set. The UC Davis/Caltech group have performed a number of tests of the PET system inser ted in the Bruker 7-T Biospec animal MRI system operating with spin-echo and g radient-echo pulse sequences that are commonl y used for small animal MRI studies. Since the scintillation crystals and f ibers are nonparamagnetic and nonconductive, and all front-end electronic components are located beyond the sensiti ve region of the MR system, this system design has v ery little effect on the MR perfor mance and vice versa.76 PET perfor mance compromises made in the UC Davis/Caltech PET system inser t design are the shor t (6 mm length) cr ystals and small (12 mm) axial F oV, which considerably compromise PET photon sensiti vity, and the relati vely poor scintillation light detection SNR and energy/time dispersion resulting from the light loss from f iber bending and attenuation do wn the f iber. The measured single cr ystal-fiber ener gy and coincidence time resolution before system constr uction w as > 24% FWHM at 511 keV and > 5 ns FWHM, respectively. Due to a position-dependent delay of crystal-fiber location on 8 mm A

C

Scintillation Light flash Corner contacts

(X,Y)

A, B, C, D

B

D

X

(A B) (C D) A B C D

Y

(A C) (B D) A B C D

Figure 10. Position-sensitive avalanche photodiode (PSAPD) is another promising photodetector technology that can operate inside a strong magnetic field. Light from a small crystal impinging on the PSAPD (as depicted in blue) creates charge that is collected by four corner contacts (A, B, C, D) on the back side of the device as well as a common signal contact on the top of the device. From these signals one may accurately estimate, for each incoming 511 keV photon event, the x-y coordinate of the center of the light flash (using the positioning logic shown) as well as the energy and arrival time of the event. The PSAPD is used in the small animal positron emission tomography (PET) insert described in reference76.

the PSAPD , after system constr uction the coincidence timing window used w as 40 ns, leading to an increased random coincidence rate. With the cur rent detector configuration and ring packing geometr y, it would be a challenge to e xtend the axial F oV fur ther. But with the general scintillator -fiber coupled PET detector scheme used, increasing the axial F oV w ould lik ely still not substantially alter MRI performance. The UC Davis/Caltech group has also be gun studying applications of combined PET/MR studies using dual-PET and MR molecular probes to tar get and image microphages involved in stenotic plaques in rat models of arterial injury.77 On the PET side, the probe molecule 64 Cu-labeled ligand designed to bind is based on a specifically to certain receptors on microphages. On the MR side, they have used polymer-based agents containing gadolinium w hich interact directl y with w ater protons to shor ten T1 relaxation times, resulting in an increase in signal intensity compared to nonlabeled neighboring tissues. This approach is capab le of high dynamic range of signals because the y produce positi ve contrast. They ha ve also produced nanopar ticle-based agents containing iron o xides that shor ten T2 relaxation times through a magnetic f ield ef fect, resulting in decreased signal intensity . This contrast mechanism is generally more sensiti ve, producing measurab le ef fects at lower MR probe concentrations. 77 The University of Tuebingen developed an MR compatible PET detector that does not use f iber-optic coupling.71,72 They later developed an improved system design (Figure 11) that uses 10 detector modules each comprising a 12 × 12 LSO scintillator ar ray (1.6 × 1.6 × 4.5 mm 3) coupled through a light diffuser to a 3 × 3 APD array, and custom charge sensitive pre-amplif ier electronics, located within a thin copper shielded housing.73–75 This design has an adv antage in ter ms of scintillation detector SNR because the crystals are coupled to the photodetector without using optical f ibers. The multiring PET scanner inser t has an axial and transaxial F oV of 19 and 40 mm, respectively, and generates 23 tomo graphic slices with a slice thickness of ~0.8 mm. The signals emanating from the 10 detector modules are read out and processed with dedicated PET electronics (Siemens Pre-clinical Solutions, Knoxville, TN). The housing was designed in such a w ay to limit potential interference with basic components of a 7 T ClinScan animal MRI scanner (Br uker BioSpin MRI, Ettlingen, Ger many) including the magnetic f ield, f ield gradients, and RF recei ver/transmitter electronics. The cylindrical PET shell is inser ted to f it inside the g radient coil set and outside the 35 mm diameter RF coil of the MRI scanner, which has a 30 cm bore diameter.

Instrumentation and Methods to Combine Small Animal PET with Other Ima ging Modalities

B

A PET Insert Gradient Set

RF-Coil

C 7 Tesla Magnet ClinScan

PET Detector Module

LSO crystal block Amplifier and electronics and APD array

Figure 11. (A) Depiction of the University of Tuebingen positron emission tomography (PET) insert for the 7T Bruker BioSpin magnetic resonance imaging (MRI) system, as well as pictures of (B) the insert by itself and (C) individual copper-shielded scintillation detector array module used to build the 19 mm axial and 40 mm transaxial field-of-view (FoV) insert. Reproduced with permission from Judenhofer MS et al.73

The sensiti ve F oV of the Tuebingen combined PET/MRI conf iguration is 19 mm in the axial direction (limited by the PET inser t) and 35 mm in the transaxial direction (limited b y the RF coil). This FoV is adequate to image an entire mouse brain/hear t, or e ven lar ge tumors on rodents. It has been sho wn that there is onl y minor mutual interference betw een PET and MRI w hen operated simultaneously even when using more demanding MR sequences lik e echo planar imaging (EPI) for functional magnetic resonance imaging (fMRI).72,73 Moreover, the g roup showed that NMR spectroscop y is 73 The feasible in parallel with PET data acquisition. group also in vestigated cancer imaging applications of simultaneous PET/MR imaging. In one study using a mouse model of colon carcinoma, 73 the g roup used a PET cell proliferation mark er, 18F-fluoro-thymidine, T1-weighted MR, and dynamic image acquisition to help differentiate acti ve malignant cells from necrosis and inflammation. The authors argued that this differentiation was improved by the temporally correlated PET and MR data acquired dynamically. The main perfor mance compromises of the Tuebingen PET system inser t, as with the UC Da vis system, is the relatively low photon sensitivity due to shor t crystals and small (< 2 cm) axial F oV. Also, with the cur rent detector design it is unclear how to expand the axial FoV while a voiding fur ther de gradation of the MR data because the number of electronic components, amount of shielding, and co-axial signal transmission cabling inside the sensitive MR region would increase.

113

A Brookhaven National Laboratory (BNL) group has developed an MR-compatible PET insert prototype based on the technology used for the most recent version of the Rat Conscious Animal PET (RatCAP), a complete 3D tomograph designed to image the brain of an awake rat.42 This system conf igures the PET system into a highl y compact ar rangement of LSO/APD ar rays with the help of custom-made, highl y inte grated electronics b uilt at BNL.81,82 The PET system comprises a 4 cm diameter detector ring containing 12 b lock detectors, each conf igured as a 4 × 8 array of 2.3 × 2.3 × 5 mm 3 LSO crystals read out with a matching array of APDs. As with the Tuebingen system, the PET detector modules are positioned just outside the RF pickup coil of the MRI scanner and inside the gradient coils. The plan is to configure the system to f it within a radial distance of ~3 to 4 cm outside the RF coil and arranged to have a fairly small number of cables e xiting from the bore of the MRI scanner . The team designed a special RF pickup coil comprising tw o orthogonal Helmholtz coils that f it inside the RatCAP and enab le compensation for residual magnetic f ield effects or eddy cur rents produced by the presence of the PET insert. As with the other designs, the cr ystals in the BNL design are shor t and the axial F oV is nar row, leading to poor photon sensitivity. There is clearl y still a need for inno vative, highperformance small animal PET system designs that can be inserted into MR systems. The ideal small animal PET insert for an MR system would have uncompromised specifications, similar to the ideal stand-alone PET system: a large axial FoV (eg, > 10 cm) and thick (e g, > 2 cm long) crystals to be ab le to image an entire rat with high photon sensitivity (eg, ≥ 15%), and superb spatial resolution (eg, ≤ 1 mm FWHM, unifor m throughout the sensitive F oV), energy resolution (e g, ≤ 13% FWHM at 511 k eV), and coincidence time resolution (e g, ≤ 3 ns FWHM). These performance specif ications are not y et possib le with current designs. It is clear that there is a need to develop an MR-compatible PET detector design that combines the superior scintillation detector SNR a vailable with direct scintillation crystal-photodetector coupling with the minimal ef fect on MR perfor mance ensured with f iber-optic signal transmission. Such a design is under development.120 PET/Optical

For completeness, w e briefly discuss attempts to inte grate PET with optical imaging systems that use light-emitting molecular probes, a subject of focus of a later chapter. Sensitive, cooled char ge-coupled de vice (CCD) cameras which detect emitted light from fluorescent/bioluminescent

114

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

probes within a living organism have shown their potential in tracking pro gression of disease in murine models. 2,121 The goal of simultaneousl y recording radionuclide and optical signals is being pursued b y several g roups86–89,121 and offers the possibility of measuring multiple molecular based processes concurrently on the same living subject. In the combined PET-optical (OPET) system under constr uction at UCLA, 87–89 the scintillation cr ystal ar ray plays the dual role of coupling the optical signal from bioluminescence emitted from the animal to the photodetector as w ell as channeling optical scintillations created from the annihilation photon interactions to the photodetectors. The PET component comprises a he xagonal conf iguration of six detector blocks with an inner system radius of 15.6 mm. Each detector consists of a 2D ar ray of 8 × 8 GSO scintillation crystals each with a 2 × 2 mm 2 cross-sectional area and varying lengths from ~8 mm in the ar ray center to ~10 mm near the array edge that create a concave shape for each array to couple directly to the rodent to collect the bioluminescence light as well as the PET 511 keV annihilation photons emitted from the animal. Unlik e all other scintillation detectors used in PET, the ends of the crystals of the OPET detectors that are in contact with the animal are open and not covered with reflector material to let the bioluminescence light into the detectors. Special electronics were built to integrate the optical signal emitted from the animal over a desired frame duration, but also r un at the same time in single pulse processing mode for the scintillation light produced b y the 511 k eV annihilation photons. The crystal ar rays are read out by multichannel PMTs. 87–89 Extensive detector studies and Monte Carlo simulations w ere perfor med to study the feasibility of the concept and assess the ef fect of v arious geometric parameters on the perfor mance characteristics of the system. 87,88 There are a fe w performance compromises of note for the design of the PET component of the OPET system. The first is signif icant detector SNR de gradation because at least one half of the scintillation signal is lost due to the absence of the top reflector on the cr ystal ar rays. Other compromises are the relati vely short (< 10 mm long) GSO crystals, lo w ef fective Z (59) and density (6.71 g/cc) of GSO, and narrow axial FoV. Regarding the bioluminescence imaging capabilities, perfor mance is compromised due to the use of the scintillation crystals as bioluminescence light guides into the PMTs and their relatively low spatial resolution compared to the superior SNR and high spatial resolution available with a cooled, lens-coupled CCD imager used in standard in vi vo bioluminescence imaging. The OPET system is reviewed in more depth in Chapter 9, “Fiber Optic Fluorescence Imaging” of this book.

CONCLUDING REMARKS To date, there are numerous small-animal PET system designs either commercially available or under investigation and this remains an acti ve field of research. To date, most small-animal PET system designs tend to compromise other performance parameters in favor of achieving high (eg, < 2 mm) spatial resolution at the system center. Furthermore, there are se veral hybrid designs that combine tw o or more small-animal imaging modalities, including PET/CT , PET/SPECT , PET/MRI, and PET/optical. To date, h ybrid designs to inte grate PET hardware with that of other modalities (e g, MRI) with truly simultaneous operation in mind lead to fur ther performance trade-offs due to compromises, such as f ibercoupling, short crystals, and narrow axial FoV inherent in the design as w ell as perfor mance de gradations due to inter-modality interference during operation. Discovering compelling applications that moti vate truly simultaneous, temporall y cor related imaging with PET and other modalities, as opposed to separate acquisitions and softw are fusion of image data, is also an acti ve field of research. F or e xample, the ideal applications for simultaneous PET/MR imaging are y et to be deter mined but will lik ely in volve dynamic acquisitions and e xploit temporal correlations between molecular information provided with PET and the v ariety of multiparameter , dynamic contrast mechanisms available with MRI. Assuming such compelling applications for simultaneous acquisitions arise, it is clear that fur ther ef fort is required to develop integrated system designs that do not compromise performance of the imaging modalities in either the basic PET design specif ications or during multimodal data acquisition. Thus, there is still room for inno vative instrumentation designs.

ACKNOWLEDGMENT The author w ould lik e to ackno wledge Dr . Virginia Spanoudaki, currently at Stanford University, for her help in reviewing this chapter.

REFERENCES 1. Gambhir SS. Molecular imaging of cancer with positron emission tomography. Nat Rev Cancer 2002;2:683–93. 2. Massoud TF, Gambhir SS. Molecular imaging in living subjects: seeing fundamental biolo gical processes in a ne w light. Genes De v 2003;17:545–80. 3. Haubner R, Weber WA, Beer AJ, et al. Noninvasive visualization of the acti vated αvβ3 integrin in cancer patients b y positron emission tomo graphy and [F-18]galacto-RGD . PLoS Med 2005;2:244–52.

Instrumentation and Methods to Combine Small Animal PET with Other Ima ging Modalities

4. Vavere AL, Welch MJ. Preparation, biodistribution, and small animal PET of Ti-45-transferrin. J Nucl Med 2005;46:683–90. 5. McQuade P, Martin KE, Castle TC, et al. Investigation into 64Cu-labeled bis(selenosemicarbazone) and bis(thiosemicarbazone) complexes as hypoxia imaging agents. Nucl Med Biol 2005;32:147–56. 6. Robinson MK, Doss M, Shaller C, et al. Quantitati ve immunopositron emission tomography imaging oil HER2-positive tumor xenografts with an iodine-124 labeled anti-HER2 diabody . Cancer Res 2005;65:1471–8. 7. Lyons SK. Advances in imaging mouse tumour models in vi vo. J Pathol 2005;205:194–205. 8. Waldherr C, Mellinghoff IK, Tran C, et al. Monitoring antiproliferative responses to kinase inhibitor therap y in mice with 3 ʹ′-deoxy3ʹ′-18F-fluorothymidine PET. J Nucl Med 2005;46:114–20. 9. Thakur ML, Aruva MR, Gariep y J, et al. PET imaging of onco gene overexpression using 64Cu-vasoactive intestinal peptide (VIP) analog: comparison with 99mTc-VIP analo g. J Nucl Med 2004; 45:1381–9. 10. Aliaga A, Rousseau J A, Ouellette R, et al. Breast cancer models to study the expression of estrogen receptors with small animal PET imaging. Nucl Med Biol 2004;31:761–70. 11. Barthel H, Wilson H, Collingridge DR, et al. In vivo evaluation of [F18]fluoroetanidazole as a ne w mark er for imaging tumour hypoxia with positron emission tomo graphy. Br J Cancer 2004; 90:2232–42. 12. Sundaresan G, Yazaki PJ, Shively JE, et al. 124I-labeled engineered anti-CEA minibodies and diabodies allo w high-contrast, antigen-specific small animal PET imaging of xenografts in athymic mice. J Nucl Med 2003;44:1962–9. 13. Solomon B , McAr thur GA, Cullinane C, et al. Applications of positron emission tomo graphy in the de velopment of molecular targeted cancer therapeutics. BioDrugs 2003;17:339–54. 14. Yang HH, Berger F, Tran C, et al. MicroPET imaging of prostate cancer in LNCAP-SR39TK-GFP mouse x enografts. Prostate 2003;55:39–47. 15. Ray P, Pimenta H, P aulmurugan R, et al. Nonin vasive quantitati ve imaging of protein-protein interactions in living subjects. Proc Natl Acad Sci U S A 2002;99:3105–10. 16. Luker GD, Sharma V, Pica CM, et al. Noninvasive imaging of proteinprotein interactions in li ving animals. Proc Natl Acad Sci U S A 2002;99:6961–6. 17. Adonai N, Nguyen KN, Walsh J, et al. Ex vivo cell labeling with 64Cupyruvaldehyde-bis (N4-methylthiosemicarbazone) for imaging cell trafficking in mice with positron-emission tomo graphy. Proc Natl Acad Sci U S A 2002;99:3030–5. 18. Wang J , Maurer L. P ositron emission tomo graphy: applications in drug discovery and drug development. Curr Top Med Chem 2005; 5:1053–75. 19. Frangioni JV, Hajjar RJ. In vivo tracking of stem cells for clinical trials in cardiovascular disease. Circulation 2004;110;3378–83. 20. Cao F, Lin S, Xie X, et al. In vivo visualization of embryonic stem cell survival, proliferation, and mig ration after cardiac deli very. Circulation 2006;113:1005–14. 21. Blasberg RG, Tjuvajev JG. Molecular -genetic imaging: cur rent and future perspectives. J Clin Invest 2003;111:1620–9. 22. Pither R. PET and the role of in vi vo molecular imaging in personalized medicine. Expert Rev Mol Diagn 2003;3:703–13. 23. Shah K, Jacobs A, Breakefield XO, Weissleder R. Molecular imaging of gene therapy for cancer. Gene Ther 2004;11:1175–87. 24. Luker GD. Special conference of the American Association for Cancer Research on molecular imaging in cancer: linking biolo gy, function, and clinical applications in vi vo. Cancer Res 2002; 62:2195–8. 25. Wu JC, Inubushi M, Sundaresan G, et al. Positron emission tomography imaging of cardiac repor ter gene e xpression in li ving rats. Circulation 2002;106:180–3.

115

26. Jacobs A, Voges J, Reszka R, et al. P ositron-emission tomography of vector-mediated gene e xpression in gene therap y for gliomas. Lancet 2001;358:727–9. 27. Inubushi M, Wu JC, Gambhir SS, et al. P ositron-emission tomography reporter gene e xpression imaging in rat m yocardium. Circulation 2003;107:326–32. 28. Hnatowich DJ. Observations on the role of nuclear medicine in molecular imaging. J Cell Biochem 2002;87:18–24. 29. Chakrabarti A, Zhang K, Aruva MR, et al. Radioh ybridization PET imaging of KRAS G12D mRNA expression in human pancreas cancer x enografts with [64Cu]DO3A-peptide nucleic acid-peptide nanoparticles. Cancer Biol Ther 2007;6:1–9. 30. Ingvar M, Eriksson L, Ro gers GA, et al. Rapid feasibility studies of tracers for positron emission tomography: high resolution PET in small animals with kinetic anal ysis. J Cereb Blood Flo w Metab 1991;11:926–31. 31. Levin CS. Primer on molecular imaging technolo gy. Eur J Nucl Med Mol Imaging 2005;32:S325–45. 32. Levin CS. New imaging technologies to enhance the molecular sensitivity of positron emission tomography. Proc IEEE 2008; 96:439–67. 33. Levin CS. Cur rent trends in pre-clinical PET system design. PET Clinics 2007;2:125–60. 34. Bloomfield PM, Rajeswaran S, Spinks TJ, et al. The design and physical characteristics of a small animal positron emission tomo graphy. Phys Med Biol 1995;40:1105–26. 35. Cherry SR, Shao Y, Silverman RW, et al. MicroPET: a high resolution PET scanner for imaging small animals. IEEE Trans Nucl Sci 1997;44(Pt 2):1161–6. 36. Seidel J, Vaquero JJ, Green MV. Resolution uniformity and sensitivity of the NIH Atlas small animal PET scanner: comparison to simulated LSO scanners without depth of interaction capability. IEEE Trans Nucl Sci 2003;50(Pt 2):1347–50. 37. Surti S, Kar p JS, P erkins AE, et al. Design e valuation of A-PET: a high sensiti vity animal PET camera. IEEE Trans Nucl Sci 2003;50(Pt 2):1357–63. 38. Uribe J, Wong W-H, Li H, et al. Signal characteristics of indi vidual crystals in a high resolution BGO detector design using PMT quadrant sharing. IEEE Trans Nucl Sci 2003;50:355–61. 40. Motta A, Damiani C, Del Guerra A, et al. Use of a fast EM algorithm for 3D image reconstr uction with the YAP-PET tomo graphy. Comput Med Imaging Graph 2002;26:293–302. 41. Ziemons K, Auffray E, Barbier R, et al. The ClearPET (TM) project: development of a 2nd generation high-perfor mance small animal PET scanner. Nucl Instrum Methods Phys Res A 2005;537:307–11. 42. Vaska P, Woody CL, Schl yer DJ, et al. RatCAP: miniaturized headmounted PET for conscious rodent brain imaging. IEEE Trans Nucl Sci 2004;51:2718–22. 43. Wu HY, Pa D, O’Sullivan JA, Tai YC. A feasibility study of a prototype PET inser t device to convert a general-pur pose animal PET scanner to higher resolution. J Nucl Med 2008;49:79–87. 44. Jeavons A, Kull K, Lindberg B, et al. A proportional chamber positron camera for medical imaging. Nucl Instr um Methods 1980; 176:89–97. 45. Spanoudaki VC, McElroy DP, Torres-Espallardo I, Zie gler SI. Ef fect of temperature on the performance of proportional APD-based modules for gamma ra y detection in positron emission tomo graphy. IEEE Trans Nucl Sci 2008;55:469–80. 46. Knoxville (TN): Siemens Pre-clinical Solutions. 47. Andover (MA): Philips Medical Systems. 48. Madrid, Spain: GE-Suinsa. 49. Northridge (CA): Gamma Medica-Ideas. 50. Pisa, Italy: Ingegneria dei Sistemi Elettronici. 51. Straubenhardt, Germany: Raytest GmbH. 52. Oxford, UK: Oxford Positron Systems. 53. Levin CS, Hoffman EJ. Calculation of positron range and its effect on the fundamental limit of positron emission tomo graphy system

116

54.

55.

56.

57.

58.

59.

60. 61.

62.

63.

64.

65.

66.

67. 68.

69. 70.

71.

72.

73.

74. 75.

76.

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

spatial resolution [cor rigendum: Ph ys Med Biol 2000;45:559]. Phys Med Biol 1999;44:781–99. Levin CS. Design of a high resolution and high sensitivity scintillation crystal array for PET with nearl y complete light collection. IEEE Trans Nucl Sci 2002;45:2236–43. Surti S, Kar p JS, Perkins AE, et al. Imaging perfor mance of A-PET: a small animal PET camera. IEEE Trans Med Imaging 2005;24:844–52. Wang Y, Seidel J, Tsui BM, et al. P erformance evaluation of the GE healthcare e xplore VISTA dual-ring small animal PET scanner . J Nucl Med 2006;47:1891–900. Jeavons AP, Chandler RA, Dettmar CAR. A 3D HIDAC-PET camera with sub-millimetre resolution for imaging small animals. IEEE Trans Nucl Sci 1999;46:468–73. Schafers KP, Reader AJ, Kriens M, et al. P erformance evaluation of the 32-module quadHID AC small animal PET scanner . J Nucl Med 2005;46:996–1004. Goertzen AL, Meadors AK, Silverman RW, Cherry SR. Simultaneous molecular and anatomical imaging of the mouse in vi vo. Ph ys Med Biol 2002;47:4315–28. Liang H, Yang Y, Yang K, et al. A microPET/CT system for in vi vo small animal imaging. Phys Med Biol 2007;52:3881–94. Bérard P, Riendeau J, Pepin C, et al. Investigation of the LabPETTM detector and electronics for photon-counting CT imaging. Nucl Instrum Methods Phys Res A 2007;571:114–7. Parnham KB , Cho wdhury S, Li J , et al. Second-generation, trimodality pre-clinical imaging system. IEEE Nucl Sci Symp Conf Rec 2006;3:1802–5. Patt B , P arnham K, Li J , et al. FLEX: tri-modality small animal tomography combining PET, SPECT and CT in a single modular gantry [abstract]. J Nucl Med 2005;46:207. Saoudi A, Lecomte R. A novel APD-based detector module for multimodality PET/SPECT/CT scanners. IEEE Trans Nucl Sci 1999; 46:479–84. Farahani K, Slates R, Shao Y, et al. Contemporaneous positron emission tomography and MR imaging at 1.5 T. J Magn Reson Imaging 1999;9:497–500. Slates R, Farahani K, Shao Y, et al. A study of artefacts in simultaneous PET and MR imaging using a prototype MR comptatible PET scanner. Phys Med Biol 1999;44:2015–27. Shao Y, Cherry SR, Farahani K, Meadors K. Simultaneous PET and MR imaging. Phys Med Biol 1997;42:1965–70. Garlick PB, Marsden PK, Cave AC, et al. PET and NMR dual acquisition (P ANDA): applications to isolated perfused rat hear ts. NMR Biomed 1997;10:138–42. Marsden PK, Strul D, Keevil SF, et al. Simultaneous PET and NMR. Br J Radiol 2002;75:S53–9. Mackewn JE, Str ul D, Hallett WA, et al. Design and de velopment of an MR-compatible PET scanner for imaging small animals. IEEE Trans Nucl Sci 2005;52:1376–80. Pichler BJ, Judenhofer MS, Catana C, et al. P erformance test of an LSO-APD detector in a 7-T MRI scanner for simultaneous PET/MRI. J Nucl Med 2006;47:639–47. Judenhofer MS, Catana C, Sw ann BK, et al. Simultaneous PET/MR images, acquired with a compact MRI compatible PET detector in a 7 Tesla magnet. Radiology 2007;244:807–14. Judenhofer MS, Wehrl HF , Ne wport DF , et al. Simultaneous PET/MRI: a ne w perspecti ve for functional and mor phological imaging. Nat Med 2008;14:459–65. Pichler BJ , Judenhofer MS, Wehrl HF . PET/MRI h ybrid imging: devices and initial results. Eur Radiol 2008;18:1077–86. Pichler BJ , Wehrl HF, K olb A, Judenhofer MS. P ositron emission tomography/magnetic resonance imaging: the ne xt generation of multimodality imaging? Semin Nucl Med 2008;38:199–208. Catana C, Wu Y, Judenhofer MS, et al. Simultaneous acquisition of multislice PET and MR images: initial results with a MR-compatib le PET scanner. J Nucl Med 2006;47:1968–76.

77. Cherry SR, Louie AY, Jacobs RE. The integration of positron emission tomo graphy with magnetic resonance imaging. Proc IEEE 2008;96:416–38. 78. Catana C, Procissi D, Wu Y, et al. Simultaneous in vivo positron emission tomo graphy and magnetic resonance imaging. Proc Natl Acad Sci U S A 2008;105:3705–10. 79. Raylman RR, Majewski S, Velan SS, et al. Simultaneous acquisition of magnetic resonance spectroscop y (MRS) data and positron emission tomography (PET) images with a prototype MR-compatib le, small animal PET imager. J Magn Reson 2007; 186:305–10. 80. Raylman RR, Majewski S, Lemieux SK, et al. Simultaneous MRI and PET imaging of a rat brain. Ph ys Med Biol 2006;51:6371–9. 81. Woody C, Schlyer D, Vaska P, et al. Preliminary studies of a simultaneous PET/MRI scanner based on the RatCAP small animal tomograph. Nucl Instrum Methods Phys Res A 2007;571:102–5. 82. Schlyer D, Rooney W, Woody C, et al. De velopment of a simultaneous PET/MRI scanner . IEEE Nucl Sci Symp Conf Rec 2004; 6:3419–21. 83. Handler WB, Gilbert KM, Peng H, Chronik B A. Simulation of scattering and attenuation of 511 k eV photons in a combined PET/field-cycled MRI system. Phys Med Biol 2006;51:2479–91. 84. Lucas AJ, Hawkes RC, Ansorge RE, et al. De velopment of a combined microPET -MR system. Technol Cancer Res Treat 2006;5:337–41. 85. Yamamoto S, Takamatsu S, Murayama H, Minato K. A block detector for a multislice, depth-of-interaction MR-compatib le PET. IEEE Trans Nucl Sci 2005;52:33–7. 86. Peter J, Semmler W. A modular design triple-modality SPECT -CTODT small animal imager [abstract]. Eur J Nuc Med Mol Imaging 2007;34:S158. 87. Rannou FR, K ohli V, Prout DL, Chatziioannou AF. In vestigation of OPET perfor mance using GA TE, a Geant4-based simulation software. IEEE Trans Nucl Sci 2004;51:2713–7. 88. Alexandrakis G, Rannou FR, Chatziioannou AF. Tomographic bioluminescence imaging b y use of a combined optical-PET (OPET) system: a computer simulation feasibility study . Phys Med Biol 2005;50:4225–41. 89. Vu NT, Silverman RW, Chatziioannou AF. Preliminar y perfor mance of optical PET (OPET) detectors for the detection of visib le light photons. Nucl Instrum Methods Phys Res A 2006;569:563–6. 90. Maintz JB, Viergever MA. A sur vey of medical image re gistration. Med Image Anal 1998;2:1–36. 91. Pluim JP, Maintz JB, Viergever MA. Mutual-information-based registration of medical images: a sur vey. IEEE Trans Med Imaging 2003;22:986–1004. 92. Cross DJ, Minoshima S, Nishimura S, et al. Three-dimensional stereotactic surface projection anal ysis of macaque brain PET : development and initial applications. J Nucl Med 2000;41:1879–87. 93. Shimada Y, Uemura K, Ardekani BA, et al. Application of PET-MRI registration techniques to cat brain imaging. J Neurosci Methods 2000;101:1–7. 94. Vaquero JJ, Desco M, P ascau J, et al. PET, CT, and MR image re gistration of the rat brain and skull. IEEE Trans Nucl Sci 2001; 48:1440–5. 95. Humm JL, Ballon D , Hu YC, et al. A stereotactic method for the three-dimensional registration of multi-modality biolo gic images in animals: NMR, PET, histology, and autoradiography. Med Phys 2003;30:2303–14. 96. Jan M-L, Chuang K-S, Chen G-W , et al. A three-dimensional re gistration method for automated fusion of micro PET -CT-SPECT whole-body images. IEEE Trans Med Imaging 2005;24:886–93. 97. Fei B, Wang H, Muzic RF Jr , et al. Defor mable and rigid re gistration of MRI and microPET images for photodynamic therapy of cancer in mice. Med Phys 2006;33:753–60. 98. Deroose CM, De A, Loening AM, et al. Multimodality imaging of tumor x enografts and metastases in mice with combined small animal PET , small animal CT , and bioluminescence imaging. J Nucl Med 2007;48:295–303.

Instrumentation and Methods to Combine Small Animal PET with Other Ima ging Modalities

99. Hill DLG, Batchelor PG, Holden M, Ha wkes D. Medical image re gistration. Phys Med Biol 2001;46:R1–45. 100. Slomka PJ. Software approach to mer ging molecular with anatomic information. J Nucl Med 2004;45:S36–45. 101. Chow PL, Stout DB, Komisopoulou E, Chatziioannou AF. A method of image re gistration for small animal, multi-modality imaging. Phys Med Biol 2006;51:379–90. 102. Chow PL, Rannou FR, Chatziioannou AF. Attenuation correction for small animal PET tomographs. Phys Med Biol 2005;50:1837–50. 103. Kinahan PE, Townsend DW, Beyer T, Sashin D. Attenuation correction for a combined 3D PET/CT scanner. Med Phys 1998;25:2046–53. 104. Townsend DW, Beyer T. A combined PET/CT scanner: the path to true image fusion. British Jour nal of Radiology 2002;75:S24–30. 105. Yang Y, Cherry SR. Observations regarding scatter fraction and NEC measurements for small animal PET. IEEE Trans Nucl Sci 2006; 53:127–32. 106. Del Guerra A, Bartoli A, Belcari N, et al. Performance evaluation of the fully engineered YAP-(S)PET scanner for small animal imaging. IEEE Trans Nucl Sci 2006;53:1078–83. 107. Kiessling F, Greschus S, Lichy MP, et al. Volumetric computed tomography (VCT): a ne w technolo gy for nonin vasive, high-resolution monitoring of tumor angiogenesis. Nat Med 2004; 10:1133–8. 108. Boone JM, Velazquez O, Cherry SR. Small animal X-ra y dose from micro-CT. Mol Imaging 2004;3:149–58. 109. Andre MP, Spivey BA, Martin PJ, et al. Inte grated CMOS-selenium X-ray detector for digital mammo graphy. Proc SPIE 1998; 3336:204–9. 110. Tai Y-C, Chatziioannou A, Yang Y, et al. MicroPET II: design, development and initial perfor mance of an improved microPET scanner for small animal imaging. Phys Med Biol 2003;48: 1519–37. 111. Yang Y, Tai YC, Siegel S, et al. Optimization and perfor mance evaluation of the microPET II scanner for in vi vo small animal imaging. Phys Med Biol 2004;49:2527–45.

117

112. Hammer BE, Christensen NL, Heil BG. Use of a magnetic f ield to increase the spatial resolution of positron emission tomo graphy. Med Phys 1994;21:1917–20. 113. Christensen NL, Hammer BE, Heil BG, Fetterly K. Positron emission tomography within a magnetic f ield using photomultiplier tubes and light-guides. Phys Med Biol 1995;40:691–7. 114. Raylman RR, Hammer BE, Christensen NL. Combined MRI-PET scanner: a Monte-Carlo e valuation of the impro vements in PET resolution due to the ef fects of a static homo geneous magnetic field. IEEE Trans Nucl Sci 1996;43:2406–12. 115. Dolgoshein B, Balagura V, Buzhan P, et al. Status repor t on silicon photomultiplier de velopment and its applications. Nucl Instr um Methods Phys Res A 2006;563:368–76. 116. Moehrs S, Del Guer ra A, Herber t DJ, Mandelkern MA. A detector head design for small animal PET with silicon photomultipliers (SiPM). Phys Med Biol 2006;51:1113–27. 117. Spanoudaki VC, Mann AB, Otte AN, et al. Use of single photon counting detector ar rays in combined PET/MR: characterization of L YSO-SiPM detector modules and comparison with a LSO-APD detector. J Instr 2007;2:P12002(1–16). 118. Herbert DJ, Saveliev V, Belcari N , et al. F irst results of scintillator readout with silicon photomultiplier . IEEE Trans Nucl Sci 2006; 53:389–94. 119. Otte AN, Barral J, Dolgoshein B, et al. A test of silicon photomultipliers as readout for PET . Nucl Instr um Methods Ph ys Res A 2005;545:705–15. 120. P.D. Olcott, H. Peng, and C.S. Levin. Novel Electro-Optical Coupling Technique for Magnetic Resonance–Compatib le P ositron Emission T omography Detectors. Molecular Ima ging, Vol 8, No 2 (March–April, 2009): pp. 74–86 121. Park JM, Gambhir SS. Multimodality radionuclide, fluorescence, and bioluminescence small animal imaging. Proc IEEE 2005; 93:771–83.

8 FUNCTIONAL IMAGING USING BIOLUMINESCENT MARKERS CHRISTOPHER H. CONTAG, PHD

OVERVIEW Biomedical tools based on optical methods are widely used for sensing and imaging in research and in the clinic; recent advances in optics ha ve led to a ne w generation of optical tools for biomedical imaging that enab le imaging of molecular and cellular processes in li ving subjects. The clinical utility of optical molecular imaging is conf ined to a narrow set of applications due to the limited penetration of light through mammalian tissues and the small number of approved contrast agents. Ho wever, optical imaging is ideally suited for the study of small laborator y animals. 1,2 As such, optical imaging has contributed considerab ly to the study of mammalian biolo gy and will continue to uncover biological mechanisms and accelerate drug development. In pre-clinical studies, the in vestigator has the opportunity to genetically manipulate transplanted cells, or those of the host, and therefore reporter genes can be incorporated into animal models with both specif icity and genetic control. This has led to an e xpansive set of tools based on optical reporter genes that can advance the study of animal models of human biolo gy and disease. Among the optical imaging tools used in small animals, in vi vo bioluminescence imaging (BLI) has had, and will continue to ha ve, a signif icant role in the f ield of molecular imaging.3–5 BLI is based on the use of optical reporter proteins called luciferases, and the genes that encode these light emitting enzymes are being de veloped as repor ter genes for use in mammalian subjects. 6–9 In addition, the proteins encoded by these genes can be conjugated directly to other targeting and sensing entities and used as repor ter proteins, or fusion proteins comprised of a luciferase and a targeting entity can be created geneticall y.10–13 The versatility of BLI presents a number of unique oppor tunities, and given the tremendous signal to noise ratios that can be 118

achieved, can pro vide sensiti ve in vi vo bioassa ys that reveal the location and magnitude of a wide range of biological processes. This is e videnced b y the ability to build informative animal models by incorporating reporter genes into target cell DNA or by creating unique repor ter conjugates through linking luciferases to tar geting molecules, such as antibodies for use as opticall y-based detectors.10,11,14 Creation of genetic constr ucts encoding fusion proteins comprised of coding sequences for the targeting antibody and that of a luciferase can be used as molecular probes in vivo.10,11 Understanding the properties of the v arious luciferases (T able 1), the parameters that control transmission of light through mammalian tissues, and the required cof actors for light emission are essential for selecting the appropriate repor ter for a gi ven application and for creating the most infor mative bioassay. A number of repor ter genes ha ve been de veloped for BLI and the proteins the y encode can be designed with specificity for a given biological process leading to markers that respond with optical signals to changes in the labeled process. These signals can be detected using lo wlight imaging systems that are external to the bodies of the study subjects. The repor ter genes used in BLI encode light-emitting enzymes, luciferases, and those that ha ve been used for BLI include genes from ter restrial and marine or ganisms that encode enzymes with emission peaks from 490 to 620 nm (seeTable 1). As with all modalities that operate in the visib le and near -infrared (NIR) region of the spectrum, the principles of tissue optics apply to the in vi vo detection of luciferase acti vity. Ho wever, there are some unique considerations including expression levels in target cells and the background emission from live animals that are unique to the in vi vo detection of these bioluminescent repor ters. The relati ve opacity of tissue permits only limited transmission of visib le light through

Functional Imaging Using Bioluminescent Markers

119

Table 1. CHARACTERISTICS OF LUCIFERASES THAT HAVE BEEN USED IN VIVO a

Luciferase

Species of Origin

Peak Emission [nm]

Substrate Used

Protein Size [kD]

Subcellular Localization

Energy Source

Cofactors

Citations

562/612

Luciferin

61

Cytoplasmic

ATP

O2b, Mg2+

Zhao et al.49

Pyrophorus plagiophthalamus (click beetle)

615

Luciferin

61

Cytoplasmic

ATP

O2

Wood et al.41,104,209 Zhao et al.49

CBLucGr68

Pyrophorus plagiophthalamus

543

Luciferin

61

Cytoplasmic

ATP

O2

Wood et al.41,104,209 Zhao et al.49

RLuc

Renilla reniformis (sea pansy)

480

Coelenterazine

36

Cytoplasmic Substrate

O2

Zhao et al.49

Rluc-mod

Renilla reniformis

547

Coelenterazine

36

Cytoplasmic Substrate

O2

Loening et al.48,110,111

Gluc

Gaussia princeps (copepod)

480

Coelenterazine

19.9

Extracellular Substrate

O2, Na+

Tannous et al.45 Wiles et al.210 Verhaegent and Christopoulos211

Aequorin (ALuc)

Aequorea victoria (jellyfish)

469

Coelenterazine

Cytoplasmic Substrate

O2, Ca2+

Shimomura212

Luxc

Photorhabdus luminescens (bacterium)

490

Decanal

Intracellular

O2

Fisher et al.213

Fluc

Photinus pyralis (firefly beetle)

CBLucred

77

FMN

nm = nanometers; kD = kilodaltons (kd). a Emission peak (λmax) measured at 37°C. b All luciferases characterized to date are oxygenases, however, some require cofactors, such as magnesium, calcium or sodium ions. c The lux genes are encoded on a five gene operon. The heterodimeric luciferase is encoded by two genes, Lux A and Lux B, which produce a 37 and 40 kd protein, respectively.

the tissues of mammals, and although largely attenuated by both absorption and scattering, light in the visible and NIR region of the spectr um has been successfull y used for imaging of biological process in living subjects.15,16 Bioluminescent signals are typically collected noninvasively using macro-optics pro viding a w hole-body image such that detectable signals from any tissue of the animal can be localized , however microscopic detection is also possib le. However, the relati vely weak signals of luciferase reactions and the inability to easil y counter stain tissues and retain enzymatic acti vity mak e microscopic detection of bioluminescent signals in tissue sections less than ideal. To o vercome this limitation bioluminescent repor ters are often combined. Bioluminescent repor ters are often combined with fluorescent reporter genes to create dual or multifunctional repor ters that can be detected with multiple methods for increased data per study and/or for v alidation of one repor ter with the other. To provide multiple methods of detection and for validation. Fluorescence is particularly well suited for

microscopic detection w here the collecting lens is a microscope that is placed on e xposed tissues, and fluorescence microscopes with bulk optics have been used to detect fluorescence in li ving animals; this method is called intravital microscopy.17,18 In addition, less invasive miniaturized microscopes ha ve been used in both mice and man to detect fluorescent signals, and these endomicroscopic tools are increasing in v ersatility and capability.19,20 Thus, bioluminescence and fluorescence are complementary modalities pro viding optimal macroscopic and microscopic detection. Macroscopic imaging with bioluminescence can provide information relative to when and w here to look with microscopic detection or with procedures that require anal yses after biopsy or necropsy. As such image guidance serves to refine animal models and optimize the data obtained from pre-clinical studies. The versatility of BLI has led to its use in di verse f ields, and a number of inno vations have increased the range of assa ys that can be perfor med in vivo. Recent technological advances include dual enzyme

120

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

assays for the study of tw o processes simultaneousl y,21 bioluminescence resonance ener gy transfer (BRET) for wavelength shifts and tagging cells,22 modified substrates for altered enzyme acti vity,23 and split luciferases for analyses of protein-protein interactions in li ve animals.24–27 As we are increasing the types of data that can be obtained in vi vo, BLI will continue to increase our understanding of mammalian biolo gy, ref ine preclinical studies, and accelerate the development of new therapeutic approaches. BLI has already become an essential tool for conducting predicti ve and infor mative studies of small animal models of human biolo gy and disease and responses to therap y, and with increased numbers of applications in industr y and academics its impact will continue to expand.

IMAGING RODENT MODELS OF HUMAN BIOLOGY AND DISEASE BLI is primaril y a preclinical imaging modality used mostly in laboratory rodents, and so in thinking about BLI one must f irst ask the question, “Wh y image a mouse?” There are two basic answers to this question. The first and perhaps the most ob vious answer is to use animal models for the de velopment of clinicall y relevant tools w here the tools themselves are f irst evaluated in animal models with the intent of translating the method or instr ument to the clinic. The second ans wer to the question, “Wh y image a mouse?” is the generation of ne w knowledge about mammalian systems and pathophysiology that can be translated to the clinic. This new infor mation can be used to ref ine clinical studies or to pro vide new insights that open ne w areas of clinical investigation. For the purpose of developing and testing of imaging instr umentation and imaging probes for clinical use, BLI has limited translational capability, with onl y v ery fe w potential niche applications in the clinic. Therefore, the v ast majority of preclinical BLI imaging studies are not aimed at building tools based on bioluminescence that translate to the clinic, but rather these studies are designed to v alidate potential therapeutic targets, test ne w compounds that tar get the basis of disease, and de velop deli very tools that can car ry estab lished or experimental compounds to the target site. The use of BLI to evaluate compounds and methods that can be translated to the clinic comprises the lar gest number of the studies using luciferases to report biological functions. In a g rowing number of studies, BLI is used to re veal basic features of mammalian biolo gy and biolo gical responses to insult. How imaging with bioluminescence can be used to test therapeutic approaches and to create opportunities for new clinical studies are the focus of this chapter .

Prior to the de velopment of imaging tools for small animals, animal studies lar gely involved serial sacrif ice to obtain temporal data. This meant that predeter mined time points needed to be selected and assa ys were performed on tissues from predeter mined locations. This effectively biases these studies b y linking the study design to a priori kno wledge and preconcei ved notions. As such this type of study has, in some respects, an inherent circularity—the e xperiment is being perfor med to learn about a bi ological process, b ut one needs to suf ficiently understand the process to appropriatel y select the times and tissues to study . Animal e xperimentation has used this approach for decades and although it has provided essential information, we now have better tools and can approach biolo gical questions with less bias. By applying molecular imaging tools to the study of animal models, we can remo ve some of the biases because the entire body can be imaged and the temporal resolution is within the rele vant range of seconds, minutes, or hours. Because imaging studies can remove many of these limitations and can serve to guide the investigator to specific times and tissues, e xperiments can be designed that reveal ne w features of disease processes and pro vide additional infor mation about a biolo gical process. 28,29 Because preclinical molecular imaging tools enab le realtime data acquisition using dynamic measures of biolo gical function, the animal e xperiments are more informative and the data sets are more complete. Moreover, given that imaging obviates serial sacrif ice, studies are not limited to assa ys on tissue samples obtained at necropsy where each time point is represented b y a different group of animals. Therefore, imaging reduces error in a given study, decreases the number of animals needed, and enab les the possibility of identifying outliers and individual variation. Prior to the widespread use of preclinical imaging tools, outliers w ere not studied and this important source infor mation w as not accessib le. The outliers may be due to e xperimental error, or may reveal some unique biological mechanism, and imaging enables making this distinction. By understanding e xperimental error and remo ving variability inherent in studies w here each time point is represented by a different group of animals, the statistics of animal studies are considerab ly improved. Real-time access to cellular and molecular infor mation in intact tissues and organs of animal models enables the researcher to obser ve mechanisms of action or cascading events within the animal that w ould not otherwise be detected using conventional methods. Imaging modalities that are already used clinicall y are being used in preclinical studies; however, BLI and other optical imaging tools w ere de veloped specif ically for the study of

Functional Imaging Using Bioluminescent Markers

small-animal models. 30–34 The use of visib le light of fers the advantages of rapid and high-throughput measures of biological function using relati vely low cost instr umentation yet with tremendous sensitivity. Because of their ease of use, optical imaging methods will continue to ha ve an impact on dr ug studies w here it is necessar y to e valuate disease process in models with intact biological pathways that interact with the potential therapies. The complexity of the re gulatory networks in disease processes needs to be studied in the context of intact organs and living tissues and cannot be readil y modeled in culture. Because the range of estab lished animal models that ha ve been w ell characterized is signif icant and optical imaging strategies can be superimposed on e xisting animal models, the use of imaging in preclinical studies is expanding rapidly, and new models that incor porate optical repor ter genes for imaging are being developed.

121

A

Light

B

ATP

D-Luciferin

ADP 1 Pi

Oxyluciferin Mg21 , O2 Luciferase

C

BASIC CONCEPTS AND METHODS Generating Signals and Sources of Noise BLI has been widel y adopted as a molecular imaging modality for preclinical studies o ver a wide range of disciplines largely because of its v ersatility, sensitivity, and accessibility. There are a number of ar ticles and re views describing the use of this approach to study mammalian biology, disease mechanisms, and therapeutic response.3,4,35–37 The light sources used in BLI are lightemitting enzymes, all of w hich are o xygenases and thus require oxygen in addition to a chemical substrate, generally referred to as luciferin (F igure 1), to produce optical signals. Because neither the enzymes nor most of the substrates produce signals alone, or in sera, backg round signals are minimal and a s witch occurs w hen the enzyme and substrate pairs come to gether to produce the signal, also ser ving to k eep the noise at a minimum. The luciferases from marine organisms use coelenterazine as a substrate and this high-ener gy molecule pro vides energy to the reaction such that additional cellular ener gy sources, such as adenosine triphoshate (A TP), are not needed. However, the autocatal ysis of coelenterazine can be a source of autoluminescence; ie, noise. The signals from these tw o enzymes are bright but not deepl y penetrating in mammalian tissues (see section “Enzyme Reporters”), so at the superf icial sites, the signal-to-noise ratios are still quite good. BLI is an e xtremely sensiti ve imaging modality in laboratory rodents due to the extraordinary signal-to-noise ratios that can be achieved,38 given the absence of noise in the reporter system and that naturally occurring sources of

HO

N

S

S

N

COOH

280.33 g/mol D-Luciferin: D-(-)-2-(6’-hydroxy-2’benzothiazolyl)thiazoline-4-carboxylic acid Figure 1. Firefly luciferase and its substrate. Bioluminescence reactions are the source of signal in bioluminescence imaging and the most commonly used luciferase is that from the North American firefly (A). This reaction (B) requires substrates and cofactors that link it to the cellular metabolism. The substrate for this reaction is D-luciferin (C) and the pharmacology of this molecule permits in vivo use with excellent signal-to-noise ratios.

light in mammalian tissues are infrequent39 or nonexistent. Signal-to-noise ratios approaching 10,000 can be obtained in BLI, and despite the signals from bioluminescent sources in the body being relati vely weak and the loss of signal due to absorbance b y mammalian tissues e xtreme, the absence of noise mak es these signals detectable, even using detectors that are external to the animal’s body. Unlike fluorescence, bioluminescence does not require an e xcitation light source and therefore e xcitation of naturall y occur ring fluors is absent in BLI. In contrast, autofluorescence is a signif icant source of noise in in vi vo fluorescence imaging, 38 thus e ven though fluorescent signals can be e xtremely intense when e xcited with a bright e xcitation source, the autofluorescence also scales with excitation intensity, and reduces signal to noise ratios. Autofluorescence is especially prob lematic at shor ter w avelengths of light (300–600 nm) in fluorescence imaging. At

122

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

longer wavelengths of light (> 600 nm) absorbance, autofluorescence decreases dramaticall y, and the scatter is also diminished but not as signif icantly. Because of the optical proper ties of mammalian tissue, optical imaging probes with longer w avelengths of emission are less attenuated than those with shor ter emission, and in all areas of optical imaging contrast agents and probes that function at the longer w avelengths are being sought.

Enzyme Reporters Luciferases used in BLI have been derived from a variety of different marine and terrestrial organisms (see Table 1) and have been e xtensively modif ied with adaptations to mak e them suitable as repor ter genes in mammalian cells. 6,9,40,41 These modifications include codon optimization for mammalian expression and elimination of sequences that tar get the luciferase to subcellular compar tments (peroxisomes42; or to the extracellular space).43–45 Increasing the wavelength of emission of the luciferases for use in BLI has onl y been moderately successful with a 615 nm peak for the longest emitting luciferases (both the click beetle [CBLuc] and firefly [FLuc] at 37°C). 9,46–48 All luciferases have broad emission spectra such that 60% of the emission for the longest emitting luciferases (FLuc and the red click beetle luciferase [CBLuc red]) is above 600 nm. 49 Transmission of light through mammalian tissue is most ef ficient at these longer wavelengths due to diminished absorbance of light by hemoglobin when wavelengths above 600 nm are used. This affects the signals that can be detected externally from internal bioluminescent sources (Figure 2). However, these broad emission peaks can present prob lems in scenarios where spectral resolution of multiple reporters would benefit the e xperiment; in such an e xperiment, the spectra are likely to overlap. Spectrally resolved imaging of luciferase emission enables localizing emitters of two different wavelengths in the body and this has been used in a limited number of studies. 50 Alternatively, because the chemistries are unique for some of the bioluminescent reporters, the use of luciferases with distinct substrates enables sequential imaging for localization of tw o markers.51 In this approach, the substrate that is cleared more rapidl y is used f irst followed by the second substrate. The broad emission spectra, ho wever, offers a solution to deter mining depth information by enabling a ratio of short to long wavelengths within a given emission spectra as a means of revealing depth of the source in the tissue. The use of light-emitting enzymes, that is, luciferases, and external imaging of the light transmitted through mammalian tissue was first demonstrated using bacteria labeled

through the e xpression of a bacterial repor ter gene, 32 and the versatility of this method has led to rapid adaptation to revealing tumor growth and response to therapy,52 stem cell engraftment and proliferation, 53 gene e xpression,33 and protein-protein interactions (See Chapter 47, “Molecular Imaging of Protein–Protein Interactions, ” b y Massoud et al.). 54–58 The enzyme from the Nor th American f irefly, FLuc, and its deri vatives have been used most commonl y. The emission of this luciferase is at 612 nm w hen assayed at 37°C making it among the longest emitting luciferases at mammalian body temperatures.49 Of the enzymes that have been examined for altered emission with changes in temperature, this w as the onl y one with a temperature shift where the λmax (emission peak) is at 560 nm at 22°C and 612 nm at 37°C—other luciferases appear to be temperature stab le.49 However, the enzymatic acti vity of man y luciferases is greater at 37°C than at room temperature. 49 BLI typicall y requires the use of geneticall y encoded markers of biolo gy, and this can be an adv antage because specif icity can be designed into the genetic constructs and the v ersatility of the repor ters is extreme given that an y number of genes can be tagged using a handful of a vailable bioluminescent repor ter genes. However, this is par t of the limitation to translation for BLI. Transfer of genetically engineered cells to humans can be justified for therapeutic genes, but use of bioluminescent repor ter genes is more dif ficult to justify given their limited use for deep tissue imaging. There are, however, bioluminescent fusions to antibodies that ma y ha ve clinical utility gi ven that the fusion protein is transfer red and not an engineered cell. Relative to other reporter gene strategies for imaging, BLI is a sensiti ve, rapid , relati vely ine xpensive approach for the study of laborator y animals and has high-throughput. Like other whole-body optical imaging approaches, it is a lo w-resolution imaging modality with limited areas of clinical translation. Luciferase reactions require oxygen and an ener gy source and therefore the signals are often tied to the metabolic activity of the tagged cell; this can be used to assess cellular function or tissue physiology,59 but it can also lead to v ariable data. Understanding these links to ph ysiology are essential for inter pretation of image data. Enzymes that use adenosine triphosphate (ATP) or other cellular cofactors as energy sources must be intracellular to produce signals—this includes the CBLuc and FLuc enzymes. Those enzymes that use coelenterazine as both the substrate and the ener gy source, Gaussia luciferase (GLuc) and Renilla luciferase (RLuc), are acti ve extracellularly and can therefore be used to tag antibodies and ligands for imaging.

Functional Imaging Using Bioluminescent Markers

A

B B

32

Open

500

520

24

540

2 1

560

580

600

1

620

8

C C

Lung

4

Liver

In Vitro

100

Skin

660

720

740

Lung

680

700

0

Open

Liver 5

100

4

80

3

60

3

2

40

2

1

20

1

0 400 450 500 550 600 650 700 750

0

0 400 450 500 550 600 650 700 750

60 40

CBGr68

4

5

100

4

80

4

60

3

60

3

40

2

40

2

20

1

20

1

0

0 400 450 500 550 600 650 700 750

0

0 400 450 500 550 600 650 700 750

100 80

Flu c 1

CBRe d

5

Molar Extinction Coefficient (cm2 1/M, 3 10000)

Normalized Light Intensity (A.U.)

640

5

h RLu c

80

20 0

Photons/sec (3 103)

28

S kin

123

Wavelength (nm) Figure 2. Transmission of bioluminescent light through murine tissues—the dependence of emission spectra and tissue depth on signal detection. Bioluminescent light emission was collected from four different sources in the skin, lung, or liver in the three animal models studied (A). The four luciferases were RLuc, FLuc, CBLuc,red and CBLuc.green68 A representative animal is shown for each model with the pseudocolor (Blue-Red) representing the bioluminescence data that is superimposed over a grayscale reference image showing the location and orientation of the animal. Spectral scanning was performed by collecting sequential bioluminescence images using an IVISTM200 loaded with 20 nm-band pass filters (filter wavelength indicated below each image); B, The open-filter images acquired before and after this spectral sequence are shown. These were used to calibrate the signal degradation during the scanning process. The entire sequence was acquired in less than five min to minimize signal deterioration. Spectra collected from skin, lung, and liver are compared with those from labeled cells in vitro (35°C); C, Photon fluxes are normalized to the values at 680 nm for beetle luciferases and 640 nm for Renilla luciferase (indicated by a green arrow). Hemoglobin (Hb) absorption curves (average values of oxy-Hb and deoxy-Hb) are plotted as background (shaded in gray). Reproduced with permission from Zhao H et al.49

The ability to study live cells in the context of living organs and live tissues in the body of fers new opportunities for examining pathologies—for example, because the cells are alive and biologically active, the molecular markers can be amplif ied through the e xpression of reporter genes 32,33,37,60–70; alter natively, amplif ication can be accomplished b y tar geting a probe to an 71–74 enzymatic acti vity that is intrinsic to the tissue. However, there are significant challenges to performing assays in vi vo, relative to standard histopatholo gy, that include the difficulty of getting a vital dye or molecular probe across biolo gical bar riers to the tar get tissue in concentrations that can produce a detectable signal, and

the inability to w ash of f unbound reagents for the purpose of reducing the noise in the image. These challenges are being met through interdisciplinar y approaches for code velopment of ne w chemical compounds, with phar macokinetic proper ties that are w ell suited for imaging, in conjunction with instruments and algorithms with impro ved sensiti vity and signif icant noise reduction. Because luciferases require a substrate to generate their optical signals, the phar macology of these substrates in mammals is relevant to interpretation of imaging data. The detection of optical signals in BLI is the focus of this section. Since enzyme acti vity is used as the readout for BLI it is essential that the

124

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

substrate be in e xcess, otherwise, the assa y will be measuring substrate le vels. At doses of 150 millig rams per kilogram body weight (mg/kg) of luciferin the substrate is generally in excess.

Detection Detection in BLI is typicall y planar and uses imaging systems that are sensiti ve to the w eak signals that escape the scattering and absorbing en vironment of the mammalian body. These systems are typically based on chargecoupled device (CCD) detectors fitted with optimal lenses and filters that operate in the visible to NIR regions of the 33 spectrum and collect the maximum amount of light. Because the detectors are sensiti ve and the signals are weak, any ambient light or noise generated b y the instr ument will reduce the image quality and sensiti vity. Therefore, the detectors are placed in light-tight chambers to exclude ambient light, and the materials in these chambers are selected for those with v ery little long-li ved fluorescence. Although most plastics ha ve a characteristic longlived fluorescence, appropriate materials for inside the chamber are limited. There are a variety of low-light imaging CCD architectures that include those that increase signal (amplified) or reduce noise (cooled); the uses of these cameras for BLI has been pre viously re viewed.10 The materials used for signal amplification can impose a spectral bias to ward the shor ter w avelengths, and although such cameras w ere used pre viously, most detectors currently being used for BLI preserve the spectral range of the CCD (roughl y 250–1100 nm). Back-thinned cooled CCDs are most widely used in the field, however, electromechanical cooling (EMC) detectors are gaining popularity due to f ast data collection and the ability to perfor m video rate imaging e ven if the signals are w eak. This is particularly impor tant for imaging using the calciumsensitive aquorin luciferase, ALuc, as an indicator of calcium fluxes in vivo. Devices used for BLI tend to be less e xpensive than those used in other imaging modalities, however, anatomic resolution is relati vely poor in w hole-body images and usually the reference images consist of a photograph of the subjects. When necessary a high-magnification lens can be directed at sites in the body w here labeled cells have been localized b y w hole-body imaging to produce highresolution images that complement the lo wer resolution images taken noninvasively of the w hole animal. The use of a high-magnif ication lens in li ving animals, intra vital microscopy, has been pub lished for a number of cell trafficking studies and tremendous insights ha ve been gained in this manner.75–80 The extreme attenuation of light, in the

visible and NIR regions of the spectrum, by human tissues will limit the translation of optical imaging modalities to specialized niches in the clinic, but since radioacti vity is not required and the instr uments are relatively lower cost, optical modalities are more accessib le and a vailable tools for the study of small-animal models.

Cellular and Molecular Biology as a Basis for in Vivo Optical Imaging Luciferase reporter genes ha ve been used in cell culture for several decades as assays for gene expression and for assessing the le vels of ATP, calcium, and other biolo gically relevant molecules. These assays were then used in thin and transparent organisms, and when initially used in larger animals had been assayed after necropsy or biopsy, in tissue homo genates. These studies ha ve pro vided a substantial amount of infor mation that can no w be used as w e transition from cells and tissue homo genates to imaging assa ys in li ving animals. The de velopment of luciferases for use in mammalian cell culture has led to well-developed luciferase repor ter genes that are expressed well in mammalian cells.44,45,49,81,82 In addition, large-scale sequencing ef forts and tremendous adv ances in high-density screens and gene e xpression studies have identified many key genetic elements in ph ysiology and disease that can be used to drive or control the expression of reporter genes as indicators of gene function. 83 These advances form the foundation on w hich in vivo BLI was built and continues to suppor t advancements in the f ield. Biological responses that comprise ph ysiology or pathophysiology typically consist of coordinated expression of multiple interrelated genes, proteins, and biological pathways, and the level of expression of a given gene and its protein product ma y be re gulated in this coordinated response depending on the interaction of multiple cell types. Because there are both resident and recr uited cells at the site of the biological response, differentiating between gene activation in the resident population versus an influx of cells already e xpressing a given gene is relevant to data inter pretation. Spatiotemporal patter ns of gene expression in dynamic environments are best e valuated through imaging, at least initiall y, as tissue str ucture and circulation can influence expression levels. The dynamic changes in e xpression can be monitored with imaging and then the infor mation can be used to guide the more labor -intensive assa ys on reco vered cells and tissues. Reconstr ucting the pathw ays and netw orks comprised of multiple single elements will require technological advances and a systems approach to biology be yond the capabilities of toda y’s imaging tools.

Functional Imaging Using Bioluminescent Markers

Such studies will require ef fective links betw een multiplexed ex vivo and in vi vo assays. Imaging approaches that can be used to monitor biochemical e vents in living cells, tissues, and in w hole organisms, are emer ging as an inte gral par t of systems biolo gy. Opticall y-based imaging methods will be at the forefront of these de velopments gi ven their ease of use, amenability to multiplexing, and their availability for biologists.

Reporter Genes Optimal repor ter genes for in vi vo studies ha ve se veral well-defined characteristics. They should encode w ellcharacterized gene products, proteins that can result in generation or accumulation of deepl y penetrating light emission with the potential for a high signal-to-noise ratio. The signal-to-noise ratio is dependent on le vels and location of repor ter gene e xpression and also on the optical properties of mammalian tissues.15 Once a reporter gene is expressed, signal detection is dependent on the absorbing and the scattering proper ties of the tissue, w hich ha ve a considerable influence on shor ter w avelengths of light.38,84,85 The versatility of BLI is due to the ability to engineer many different cell types and tag man y different genes with a small set of repor ter genes, but the variety of optical reporters is increasing and will lik ely contribute to increased numbers of application areas and uses of BLI in biomedical research and to the ability to multiple x these assays. The hardware for detection as described abo ve has been fairly well established with the promise of only incremental advances. However, we are only in the initial stages of probe development and advances in molecular contrast agents ha ve tremendous potential for di versifying and improving reporters used in BLI. By analogy to light microscopy where dyes are used to stain cells in tissues, adv ances in BLI will be in the area of new optical contrast that can mark cells and molecular processes in the body . Unlike histopathology, the inability to wash off unbound dye in vivo is a signif icant limitation. To address this issue, cells are typicall y labeled outside the body b y transfer of the repor ter gene prior to injection into an animal model and imaging either macroscopically or microscopically.18,86,87 Reporter genes encode reporter proteins that function as “labels” that can be traced noninvasively over time. The proteins interact with “repor ter probes,” applied substrates (luciferins) to generate a signal that can be initiall y localized from outside the body, and then these data can serve as a guide for selecting the appropriate tissues and times w here sampling will be most meaningful. Alternatively, genes can be transfer red to cells in vi vo, as in gene therap y

125

approaches, using viral and non viral gene transfer techniques. In vi vo gene transfer of luciferase genes serves as a model for de veloping new delivery tools and viral vectors for use in nucleic acid-based therapies, 88–91 and efficient gene transfer techniques are being used to deliver tar get genes for e valuation of RN A inhibition strategies for genetic control. 91–97 Because these reporter genes were initially developed for use in cell-based assa ys or for detection in e xcised tissues, after imaging their activity in live subjects, measurements of the reporter activity can also be made on e xcised tissue using the same optical signal that generated the in vivo image. As such the in vi vo and ex vivo measures are linked to each other and to cor relative cell culture assa ys (see section “Cor relative Cell Culture Assays”). In this manner, BLI becomes an iterati ve study of biolo gy using integrated assays that e xtend from biochemical assa ys to live animal models. Effective use of optical imaging therefore results in studies of animal models that are more informative and predicti ve than studies that onl y use conventional assays on excised tissues. For measurements of tumor burden or for assessing cell trafficking patterns, it is necessary to use genetic constructs consisting of a strong constitutively expressed promoter (such as those from the β actin, and ubiquitin c genes, or from vir uses, such as cytomegalovirus [CMV] and other sources) dri ving expression of a reporter gene, and these are integrated into the genome. Inte gration into the genome can be accomplished with viral v ectors, mammalian transposons or random integration with subsequent selection of cells with integrated repor ter genes. Inte gration circumv ents problems of loss of signal due to dilution from cell di vision as well as the confounding detection of signals that ha ve dissociated from the viab le tar get cell population. Despite these adv antages, these ectopicall y e xpressed repor ter genes can be silenced b y chromosomal modif ications. Additional de velopment in the area of optical repor ter genes will be motivated by the need for g reater sensitivity and specificity. Advances in understanding gene e xpression patterns as integrated sets using arrays of 30,000 elements can be used to indicate w hich promoters from w hich genes are best used to e xpress a repor ter designed mark a selected biological process. In this manner , the tar geting is at the level of transcription w here the cells are labeled b y targeting specif ic genetic re gulatory elements, that is, promoters, and linking them to sequences that encode reporter genes. Extensi ve use of this approach in cell biology has produced a wide ar ray of genetic elements from w hich to dra w from for molecular studies in animals. Furthermore, the transition from cell-based assa ys

126

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

to in vi vo assa ys has been enab led b y the adv ances in imaging instr umentation specif ically designed for small animals that are readil y a vailable. Repor ter genes ha ve also been developed that are re gulated posttranscriptionally and can be constitutively expressed without expressing an optical signal until another specif ic molecular event occurs (see section “Cor relative Cell Culture Assays”). In this case, the activation of the reporter is typically posttranslational.3,54–58,98,99

Correlative Cell Culture Assays Validation of in vi vo measurements of optical repor ters can be performed on excised tissues and in culture using the same repor ters that are used in vi vo. These sets of assays have shared roots in molecular and cell biolo gy. Correlative cell culture assays can be used as a means of developing an imaging strategy prior to in vivo use of the imaging approach and then can be used as a follo w-up assay for v alidation. The instr uments that detect optical signals from inside the animal can typicall y accommodate cell culture plates and thus even the detectors are the same. Imaging of cell culture cor relates can also be used to design probes and test their v alidity. The v alidation studies are conducted using biochemical measures on tissue lysates using well-established measures of enzyme activity, messenger RN A, and protein le vels.88,100 The data obtained in vivo has tremendous utility for selecting the times and tissues for anal yses that can be selected based on the image data or cell numbers, and this image guidance can be used to impro ve the data set b y conf ining the study to the rele vant times and tissues. The new advances in imaging are based on adv ances in cellular and molecular biolo gy, and the tools of these f ields are essential components of the f ield of molecular imaging and of BLI.

BIOLUMINESCENT REPORTERS AND THEIR SUBSTRATES FOR IN VIVO USE Luciferase enzymes ha ve been found in a wide range of organisms from several different genera (see Table 1), but there are essentially three basic biochemistries among the enzymes characterized to date. 6 Each of the biochemistries uses dif ferent substrates and conditions, and each chemistr y has been used in BLI. Genes encoding members of the three classes of luciferases ha ve been cloned and their chemistries characterized to a point where they can be routinel y used in the cor relative cell culture assays and in BLI. These include the luciferases from beetles (coleoptera), jell yfish and sea pansies

(cnidaria), and bacteria ( Vibrio spp. and Photorhabdus luminescens). All luciferases are o xygenases and require energy and a specific luciferin, for light production, often in the presence of cof actors.

Luciferase Enzymes The luciferases from f ireflies and related insects are single polypeptides related to the CoA ligase f amily of proteins101,102 and use a benzothiazole luciferin substrate to generate light in the presence of ATP, magnesium, and oxygen. The gene encoding FLuc is the most commonly used bioluminescent repor ter and has been codon optimized for expression in mammalian cells. In addition, mammalian transcription f actor binding sites have been removed with the intent of pre venting inadvertent control of e xpression, and man y of the cr yptic splice donor and acceptor sites ha ve been remo ved to maintain e xpression. These modif ications ha ve been largely conducted b y the v endors of these genes (Promega Cor p, Madison, WI) and ha ve produced reporter genes with high levels of expression in the target cells and tissues and that represent e xpression patterns of targeted genes. 103 Other beetle luciferases ha ve also been codon optimized, including CBLuc. 41,47,104 These proteins are broad spectrum emitters with a signif icant red component of the spectr um. Luciferases from the sea pansy ( Renilla reniformis; RLuc), 105–107 the jell yfish ( Aequorea aequorea),58,108,109 and marine copepod ( Gaussia princeps; GLuc) 45 have also been cloned , characterized, and used as in vi vo repor ter genes. All the enzymes from marine organisms that have been described today use coelenterazine as their substrate. These three coelenterazineusing enzymes dif fer in aspects that gi ve them unique characteristics for interrogating biology in vivo. Selection of the appropriate reporter gene for a given study requires some understanding of their characteristics, the bioa vailability of their substrates, and their unique biochemistries. The greatest diversity of bioluminescent proteins is in the enzymes derived from marine organisms and it is unfortunate that all of these are b lue emitters and use the less desirable in vivo substrate, coelenterazine. Since shor ter w avelengths of light are lar gely absorbed b y hemo globin, the b lue emission from the luciferase deri ved from marine or ganisms is absorbed in vivo and the sensiti vity of detection is dramatically reduced. Attempts have been made to shift the emission spectr um of RLuc to ward the red with mutations and substrate optimization, 48,110,111 but the shifts have been modest. The g reatest adv antage of these

Functional Imaging Using Bioluminescent Markers

mutations is lik ely its g reater stability in vi vo. The emission of nati ve RLuc peaks at a w avelength w here there is a signif icant dip in hemo globin absor ption (at 490 nm), and the red-shifted mutants emit light at a peak in the hemo globin absor ption (betw een 510 and 600 nm). The biodistribution of coelenterazine in mammals is not optimal for imaging as it has a relati vely short circulation time in vivo and some associated autoluminescence.49,112 The relatively short circulation time, however, mak es it a prefer red compound for the f irst reaction in dual enzyme assays, as there is little residual activity detectab le after 10 minutes. Codon-optimized versions of the Renilla enzyme ha ve ser ved as a light source for self-illuminating quantum dots (Qdots) (below).13 The luciferases from Renilla and Gaussia (RLuc and GLuc) are e xtremely bright and do not require cof actors from the host cell; this opens up the possibility of monitoring e xtracellular events and using fusion proteins (e g, antibody-luciferase fusions) as imaging agents in animals and possib ly humans. 10,11,14

127

within a fe w seconds to a minute after injection. D-luciferin can be injected IP or IV .112 Substrate derivatives and conjugates of fer ne w oppor tunities for imaging,23,119,120 and a v ariety of ne w deri vatives ha ve been described. It has also been reported that D-luciferin and coelenterazine are substrates for multidr ug resistance (MDR) proteins that pump xenobiotics out of cells.121–123 The two compounds are subject to different mechanisms with coelenterazine being a substrate for P gl ycoprotein121,122 and D-luciferin123 for alternate pathways. These observations have been used to de velop assays to assess the le vels of xenobiotic pumps on cells and to de vise methods of circumventing multidrug resistance in cancer.121 The differences in emission spectra and biochemistries of the luciferases offer the potential for multiple xing assays in cell culture and in vi vo. At present tw o, and possib ly more, biolo gical processes can be studied simultaneously.86,120 Combining bioluminescence with other enzyme activities offers another layer for in vivo bioassay development.

Substrates and Their Bioavailability In BLI, to generate a signal the substrate must be present in the same subcellular compar tment or tissue site and there are a number of cellular and tissue bar riers that the substrate must cross. These substrates are much lik e drugs that need to reach their tar get, and biodistribution of the substrate controls signal intensity in BLI. Pharmacokinetics of D-luciferin and coelenterazine, relative to tissues and or gans, is an impor tant consideration in BLI. Luciferin appears to access cells in most if not all tissues and crosses the blood-brain and placental bar riers.88,89,113 D-luciferin is cleared slo wing with the peak le vels in most tissues at 15 to 30 minutes after intraperitoneal (IP) injection; this is w hen mice should be imaged. After intravenous (IV) injections, the D-luciferin seems to peak at 1 min and be cleared within 5 min, but does appear to be more e venly distributed and higher concentrations appear to be present in the central nervous system.114–117 The optimal time from administration to data acquisition in BLI depends upon both the route of administration and the rate of clearance of the substrate in the tar get tissue. In contrast to D-luciferin, coelenterazine is subject to rapid clearance rates. D-luciferin is relati vely stab le in the body and has a relatively long circulation time. 33 In contrast, the substrate for RLuc, coelenterazine, is rapidl y cleared from the body and binds to serum proteins.112 Therefore, imaging protocols using coelenterazine generall y require IV injection and data acquisition must be complete

Multifunctional Reporter Genes Combinations of reporter genes that encode gene fusions that produce multifunctional single proteins, 121,122 or that incorporate genetic elements that enable multiple proteins to be made from a single RN A transcript (pol ycistronic message 53,103,123), ha ve been used to create multimodality imaging strategies. These are used for the purpose of pro viding additional data to a study in the form of v alidation or to link macroscopic and microscopic imaging modalities for a more complete anal ysis. Such approaches can be used to provide different types of information from a single repor ter. The ease of linking genes through genetic tools has led to the development of a number of these reporters that can be detected by two or more modalities. 100,121,124 Multireporter gene fusions were f irst described using luciferase and g reen fluorescent protein (GFP) for use in cells and flies,125 and similar fusions ha ve been used as a means of connecting BLI measurements in mice, to e x vi vo assa ys such as flo w cytometry, and fluorescence microscop y using GFP and related fluorescent proteins. 100 Triple gene fusions ha ve been used to link optical imaging and e x vi vo assa ys to positron emission tomography (PET) imaging.121,122,124 These multifunction repor ter genes can considerab ly increase the fle xibility of a repor ter for in vi vo assa ys and perhaps even more importantly offer validation measures that are not possib le using a single function reporter.100,126,127

128

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Gene Transfer A significant limitation to the use of repor ter genes is the transfer of these genes to cells in intact or gans of li ving mice with stab le inte gration and reliab le e xpression. Transformation of tumor cells with repor ter genes is generally straightforward and selection of transfor med cells with drug selection markers enables creation of stable cell lines. However, different cell types can require dif ferent methods and selection can tak e time and create cell v ariants with characteristics that differ from the parental line. In comparison, labeling primar y cells isolated from their host is more challenging, and transfer of genes to cells in intact or gans of li ving mice has remained a signif icant barrier despite adv ances in both non viral- and viralmediated gene transfer methods. One means of addressing this problem is to create transgenic mice that e xpress the reporter genes, single or dual function, with e xpression targeted to a specif ic cell type 128 or tissue. 129 These animals can be studied directl y or ser ve as cell and tissue donors for transplantation or cell trafficking studies.53 Transgenic donor mice that e xpress repor ter genes from strong constitutive promoters have been used for the study of trafficking and expansion of hematopoietic stem cells and immune cells 53,123 and the study of organ transplantation. By crossing these transgenic mice with transgenic lines of mice that ha ve been engineered to spontaneously develop malignancies under the control of doxycycline, the crossed lines can generate transplantable tumors. The onco genic potential of these transplanted cells can be controlled and the process of tumor g rowth, response to therap y, and relapse can be studied. 130 Studies of these transplanted tumors have revealed a persistent low-level signal after inacti vation of the onco gene that initiated the initial tumor growth.130 At the nadir when the disease burden w as minimal but detectab le, the labeled cells had a nor mal appearance by all measures, but after reactivation of the onco gene, these cells pro vided a source of relapse. The insight gained from this study was that in states of remission, the cells that persist ma y have stem cell characteristic giving rise to tissues with nor mal appearance in the absence of the onco gene or cancer when the onco gene is reacti vated. These insights w ere revealed b y the ability to monitor the entire disease course and use the images to guide tissue selection. In BLI, the repor ter genes are often inte grated into the mouse genome, and the signals are not lost during the course of the disease; this enab les long-ter m study of cells through each of the stages of malignanc y. The use of repor ter genes in BLI has made this modality ideall y suited for the de velopment of nucleic

acid-based therapies. By placing the reporter gene into the gene transfer vector or other gene delivery tool, the transfer of genetic material to the cells in tissues can be assessed.88,91,131–133 Conversely, with the repor ter gene in the cells as a transgene or temporaril y e xpressed gene, inhibitory nucleic acids, such as small inhibitor y RNAs (siRNA), can be deli vered and the reduction in signal measured as a reporter of effective delivery.92–94,134 In both of these scenarios, the inte grity of e xpression of the reporter gene is crucial to interpretation of the data. In all reporter gene imaging approaches, there is the potential of the promoter element to be silenced or otherwise modulated. This may be due to methylation of CpG dinucleotides in the promoter of to other epigenetic events.135,136 Promoters deri ved from vir uses are more likely to be silenced (e g, the immediate earl y promoter from CMV and that from simian vir us 40), than those derived from eukar yotic cells. Thus, in situations w here strong constituti ve e xpression is desired , the promoters from ubiquitin C and β actin are used. 53,123

BRET Bioluminescence is a biological source of light that can be manipulated and directed through genetic engineering and the gene products used for a v ariety of pur poses. One of these applications is BRET . BRET uses a bioluminescent luciferase that is geneticall y fused to a protein or peptide that can interact with a fluorescent moiety that can accept the photons from the luciferase and con vert them to a longer, lower energy signal.22 In the case of protein-protein interactions, the tw o fusion proteins bring the luciferase and the fluorescent protein close enough for resonance energy transfer to occur, leading to a signal s witch. BRET was first demonstrated in cell culture 22 and has since been used to assess protein-protein interactions in vi vo.137–140

Self-Illuminating Qdots One subset of BRET is the use of luciferases to e xcite Qdots. Qdots are particles of different sizes that fluoresce at dif ferent w avelengths using a common e xcitation wavelength, and this common w avelength matches the emission spectr um of RLuc. Thus, combining bioluminescence with Qdots has the adv antage of illuminating the dot at its surf ace to create a self-illuminating particle.13 Since the excitation light for Qdots is generally of short wavelength, by conjugating the excitation source to the fluor, the loss of e xcitation light due to absor ption by mammalian tissues is ob viated, and the distance that

Functional Imaging Using Bioluminescent Markers

the excitation light travels is signif icantly less than w hat would be needed for e xternal e xcitation.13 Such conjugates use BRET (also kno wn as chemiluminescence resonance ener gy transfer and luminescence resonance energy transfer) 22,141–143 to produce signals. In the study by So and colleagues, carboxylate-presenting Qdots were conjugated to a mutant for m of the luciferase from R reniformis. The long w avelength of emission from the conjugate and the colocalization of the e xcitation source with the fluorescence resulted in impro ved sensitivity in small-animal imaging and a high signal-to-noise ratio in comparison to Qdots requiring an e xternal e xcitation source.

DETERMINANTS OF DATA QUALITY Evaluation of BLI data requires an understanding of the determinants of data quality and signif icance. The sensitivity of detection for BLI is deter mined b y the brightness of the labeled cell (ie, photon flux from the source). Detection is then dependent on the depth of the source within the tissue and the optical proper ties of the tissues between the source and the detector—for example, liver is absorbing because it is highl y vascularized and therefore contains considerab le le vels of hemoglobin and bone scattering due to its str ucture. The extent of absorbance of the signal b y mammalian tissues is determined by the wavelength of the emission and the tissue composition. Therefore, grayscale reference images tak en under lo w-light illumination to reveal the position of the animal that are link ed to the optical data of v arious tissues can be used to impro ve data quality . Re gardless of the source intensity and depth, the ability to detect signals is dependent on the quantum ef ficiency and noise of the detector and the efficiency of the optics used to direct the signal to the detector. Because the source of the signal is a biochemical reaction that requires substrate, the intensity will also depend on the substrate availability at sites of reporter gene expression.

APPLICATIONS OF BLI: BLI IN THE DEVELOPMENT OF NEW THERAPEUTIC STRATEGIES A significant understanding of the molecular basis of disease has been generated o ver decades of reductionist science, and this has led to the identification of a number of molecular targets for therapy. As such the paradigm of directing therapies to specif ic molecular tar gets through

129

rational design, rather than targeting physiological states, such as accelerated cell di vision, is being applied to a wide range of diseases. The validation of molecular targets for such therapeutic advances requires testing in animal models, and BLI is being used to accelerate and refine this testing. An e xample of this type of study involves mouse models of cancer that are the result of a disregulated myc oncogene, and the effects of specifically inhibiting this molecule in v arious cancers. 130 In these models, cells with disre gulated myc expression are labeled and the effect of drugs that decrease myc expression are studied in li ving animals using luciferase as an indication of tumor burden. Traditionally, cancer therapies ha ve been tested using cell lines that represent frank malignancy and often end-stage disease. The sensitivity of BLI enab les the study of earl y disease and minimal residual disease, and therefore the effects of myc overexpression as an initiating event or as a cause of relapse, as well as decreasing myc as a therapeutic objective could be investigated.130 Although most patients that present with cancer in the oncolo gy clinic respond to established chemotherapies and radiotherapies, and ma y even demonstrate e xtended periods of remission, most patients will relapse. Thus, w e should be directing the development of ne w therapies to states of minimal disease and aim at controlling relapse as an impor tant therapeutic objecti ve. The sensiti vity of BLI enab les the study of minimal disease states and is pla ying a role in changing treatment paradigms. Imaging is opening windows into disease states that were previously inaccessible (Figures 3 and 4) and with advances in imaging w e will be ab le to study ne w disease mechanisms and develop therapies specif ically for these steps in disease pro gression. This includes targeting the small number of cancer cells that exist in the initial stages of disease or during states of minimal residual disease. 86,144–147 Elimination of these cells is being approached with ne w small molecule inhibitors,1,91,92,95,148 variations of e xisting chemotherapeutic agents, 146 and ne w combinations of therapies.86,146,149 In each of these approaches, BLI can assist in the de velopment and testing of these ne w tools and can be used as a general indicator of the e xtent of disease, that is, tumor burden in cancer, gene expression in genetic disease, or bacterial load in infection (see Figure 4), or of a specific biolo gical process, such as tumor hypoxia.59,150 Bioluminescent mark ers of h ypoxia have been developed by a number of groups and can be based either on transcriptional re gulation through tar geting hypoxia-inducible factors (HIF) 151 or on protein de gradation by directing a bioluminescent repor ter protein to

130

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

500

1000

450

800

400

600

350 400 300 200

7d

9d

16d

Figure 3. Sensitive imaging modalities provide access to new information about minimal residual disease in cancer models through in vivo bioluminescence imaging. Mice with labeled lymphoma cells were followed over time beginning prior to therapy (7 d) and after whole-body radiation and a bone marrow transplant (9 d and 16 d posttumor inoculation). 2 days after therapy a diminution in signal is observed, but the persistent minimal residual disease is apparent at 16 d—this is 11 d after treatment. The ability to monitor minimal residual disease enables the development of new therapies that can target the small number of remaining cells and thus offer the potential of preventing relapse. Adapted from Edinger M et al.100,126

genes to re veal biolo gical processes in vi vo, that is, controlling repor ter protein e xpression at the le vel of transcription4,33,153 or constitutively expressing a destabilized reporter protein that is stabilized by the targeted process.154,155 Therapies directed at gene e xpression include the development of nucleic acid-based therapies for either gain of function (gene therap y88,90,131,156–158) or loss of function (antisense RNA or RN Ai strategies91,93,95,134,159,160). These strategies ha ve been adv anced signif icantly b y appl ying preclinical imaging tools to the study of the tar get and the therapy. Through imaging, w e ha ve a better sense of promoter strength and silencing, 161,162 and duration of expression or effect on the target cells and tissue. 93–97 This will lead to improved therapies and refined clinical studies with more insightful study designs. The use of optical reporter genes in these studies can be a precursor to using reporter genes that confer detectable signatures on the molecular therapy in the clinic. These reporters would be those that can be assessed using nuclear medicine approaches. Thus, the information gleaned through imaging of laboratory animals will help direct the use of imaging in clinical studies.

APPLICATIONS OF BLI: ENABLING ADVANCES IN CLINICAL IMAGING MODALITIES

Figure 4. Imaging infection. Bacterial luciferases can be used to label bacteria and assess location of infection. In this example, Pseudomonas aeruginosa was labeled and used to infect the animals. Despite keeping the intranasal route of administration constant, both unilateral and bilateral infections were noted.

the proteosome for de gradation with an o xygen-dependent protein destabilization domain. 152 In the latter approach, the repor ter protein is stabilized in h ypoxic conditions and the signal is thus increased as the tumor vasculature is compromised and the pO 2 decreases. Development of such assa ys depends on understanding the oxygen sensitivity of each repor ter system 59 and on the ability to engineer a biological system using a sensitive reporter that can reveal subtle changes. These studies demonstrate tw o basic strate gies for using repor ter

By linking preclinical modalities that are based on clinical imaging, such as PET, single photon emission computed tomo graphy (SPECT), and magnetic resonance imaging (MRI), with imaging tools that are best suited for animal models, lik e BLI and fluorescence imaging, we can better develop strategies for imaging humans as one modality guides the other and can be used to v alidate new imaging tools. A g reat example of this is the use of gene fusions comprised of those encoding bioluminescent reporters and those that encode proteins that concentrate radiolabeled compounds, such as thymidine kinase and its substrates.121,122,124 These gene fusions enable monitoring of gene transfer for impro ved genebased therapies and for the de velopment of ne w cellbased therapies. Linking imaging modalities through mixed function repor ter genes creates po werful multimodality approaches for preclinical testing and better enables translation through validation studies.

Improved Delivery Tools The advances in small molecule therapies ha ve not been matched by a comparab le development of deli very tools

Functional Imaging Using Bioluminescent Markers

1d

4d Dorsal

Lateral

Ventral

131

7d

Lateral

Figure 5. Visualizing bone marrow transplantation. Transgenic donor mice that express firefly luciferase from strong constitutive promoters (eg, hybrid promoters from β-actin and cytomegalovirus) serve as sources of labeled cells for long-term study of transplantation and tissue regeneration. Single lateral views are shown for days 1 and 7, and two lateral, a dorsal, and a ventral view are shown for day 4. The multiple views demonstrate how the images of the same animal at the same time point differ with different perspectives. Adapted from Cao YA et al.123

d12

d17

d21

d31

Figure 6. Temporal analyses of hematopoietic stem cell (HSC) engraftment and expansion after transfer of a single labeled HSC. Transfer of a single HSC via intravenous injection results in hematopoietic reconstitution over a 30 day time course. Foci are detected at 6 to 12 d posttransfer and these early foci consist of approximately 2000 cells. These foci serve as a source of hematopoietic reconstitution and eventually signal is detected from the entire animal as the labeled cells repopulate the blood. Adapted from Cao YA et al.53

for directing these dr ugs to their therapeutic tar gets. The emerging nanotechnologies that are designed to address this problem as well as the more established technologies of liposomes and antibodies are advancing at an unprecedented rate due to the use of preclinical imaging tools that are based on optical repor ter genes. 159,163–165 The development of molecular transpor ters as methods of improved dr ug deli very ha ve adv anced through imaging.166 In addition, an increased understanding of immune cell and stem cell mig ration has been re vealed through imaging of laborator y animals, and this has created ne w opportunities for using cells to target specific pathologies and deliver therapies to the target tissues.5,53,86,123,124,167,168 To effect their function, immune cells must migrate to the

target tissue. It follo ws that these cells w ould be useful, therefore, to carry therapeutic constructs to sites of tissue destruction in autoimmune diseases or to deli ver oncolytic agents to malignant cells, and BLI has been used to e valuate and adv ance each of these strategies.86,132,133,144,169,170 The de velopment of such comple x combination biolo gical therapies benef its considerab ly from imaging, as the immune cell, the therapeutic, and/or the target can be labeled and imaged. 86,144 The complexity of de veloping cell-based therapies pales in comparison to the sophistication required for re generative medicine and stem cell therap y, and BLI is contributing to advances in this emerging area of research. BLI has become an essential par t of studies in regenerative medicine where bioluminescent reporter genes can be used to re veal the location and the number of transplanted stem cells, 53,63,123,168,171–177 the duration of g raft survival,123,168 and the maturation of a stem cell. 158,168,178 To date, this has been accomplished using constituti vely expressed reporter genes in stem cells (F igures 5 and 6); however, refined reporter genes with developmentally regulated promoters are be ginning to be used to assess the extent of stem cell dif ferentiation, and not just to assess cellular location and cell numbers. The ability to monitor engraftment and assess graft survival presents new opportunities for discovery that have not previously been possible. F or e xample, for an allo graft to sur vive immune suppression it is necessary to prevent rejection, and transfer of immunomodulator y genes or sets of genes into luciferase-labeled g rafts may prolong their sur vival. Use of BLI would enable assessing the best genes, and combinations of genes, for this pur pose. These approaches are beginning to emerge in the literature and comprise a functional genomics approach based on imaging where a large

132

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

number of genes can be screened for function in vi vo by conferring a g rowth advantage on the transplanted cells. Use of this approach will be realized in identif ication of genes that prolong engraftment of replacement tissues and those that accelerate cancer g rowth to be de veloped as new targets for therapy. The development of transgenic mice with strong constitutive promoters dri ving e xpression of the repor ter gene have been useful for transplantation studies53,123 and have also been found to be useful for de veloping ne w tools for small-molecule deli very.179 Since these animals were developed to express luciferase in every cell of their body, and luciferase acti vity depends on substrate a vailability, by controlling the delivery of a given substrate as a model small molecule, new tools can be developed that direct therapeutic compounds to target tissues. Controlled delivery has been demonstrated in skin 166 and in tissue window chambers. 179 Use of substrate deri vatives23 with different log p v alues should per mit the de velopment of improved delivery tools for a v ariety of small molecules with different pharmacological properties. Even though many of the optical imaging modalities are constrained in the clinic b y poor tissue penetration, they e xcel in the preclinical studies, and therefore are being widely used to direct clinical studies and adv ance many scientif ic f ields that depend on laborator y rodents as models. Since optical repor ters can be built into animals and linked to target genes and cells, the oppor tunities for de veloping ne w rob ust and infor mative models are significant. In addition, many of the optical tools can be superimposed on e xisting models without modif ication and this strengthens the study of w ell-developed approaches that are already pro viding new data, advancing our understanding of new targets, and accelerating the development of new therapies.

Using BLI to Improve Our Understanding of Mammalian Biology The second reason to image laborator y rodents and develop preclinical imaging tools is to improve our understanding of mammalian biolo gy, and here the end product is not a ne w dr ug, delivery tool, or imaging reagent, but new knowledge, and this is what is translated to the clinic. Transgenic reporter mice that are engineered to emit a bioluminescent signal in response to v arious chemical or physical stressors, or infectious insults can be used to guide 50,180–183 These the anal yses of the tissue responses. reporters can be designed b y tar geting transcriptional activity and using specif ic genetic re gulatory elements, promoters, to direct e xpression of the repor ter gene, or

through other tools of molecular biolo gy that are used in cell culture assa ys. These transgenic repor ter animals can be de veloped to respond to stimuli as di verse as malignancy, infection, chemicals—to xicology,184 physical stress—thermal, and physiologic changes. An alternative is to use strong constitutive promoters to drive the expression of reporter genes in transgenic animals, and this has led to the de velopment of labeled donor mice, as discussed above, that can be used as a source of labeled tissues and cells for transplantation into unlabeled recipients and the graft studied over time.53,123 These animals are particularly useful for cell trafficking studies and the study of the mammalian immune responses to infection and malignanc y. Immune cell function is based on sensing foreign proteins or cellular stress and recr uiting other immune cells to sites of insult b y either mig rating to a l ymphoid tissue, lik e dendritic cells, 185 and stimulating other immune cells to respond to the insult or secreting cytokines and chemokines that recruit effector cells to the target tissues. 100,186–189 As such, cell mig ration is a k ey feature of the mammalian immune response, and preclinical imaging is pro viding insights into these processes and increasing our knowledge of mechanisms and kinetics of immune cell function.5,167,190,191 As cell migration is also the foundation of metastatic disease and many of the cell signaling mechanisms used by immune cells are used by cancer cells (eg, the chemokine stem cell derived factor [SDF] and its cognate receptor CCR479), optical imaging is pro viding insights into mechanisms of disease progression in malignanc y and re vealing new targets for therapeutic intervention.192–195 In a study by Lin and colleagues,79 intravital microscop y w as used to re veal cell migration patter ns, and this study implicated SDF as a key f actor for metastatic disease and directed mig ration of cancer cells to specif ic regions of the bone. Coupling this microscopic technique with a macroscopic modality, such as BLI, creates a powerful combination because BLI is nonin vasive, and without injur y to the animal, can direct the in vestigator to specif ic times and tissues for placement of the bulk optics of the microscope objecti ve that is used for high-resolution imaging of cell mig ration patterns. Intravital microscopy is often a ter minal procedure that the animals do not reco ver from, and therefore, using a noninvasive, although low resolution, modality as a guide can lead to more insightful studies and can reduce wasted time and lost data. Advanced studies of oncogenesis and metastasis are enabled by imaging (see Figure 3), and although these processes ma y be thought of as disregulated mammalian de velopment, the use of imaging to study developmental biology will also lend insights into these disease processes.

Functional Imaging Using Bioluminescent Markers

The basis of development is, in par t, control of stem cells and their dif ferentiation and self-rene wal. Understanding self-renewal of stem cells or the alternative pathway of dif ferentiation and tissue re generation is the foundation of the f ields of stem cell biolo gy and tissue regeneration. Because these processes can onl y be studied in the context of the living body, imaging is inextricably link ed to the development of these areas of research, and BLI is refining models of regeneration and repair.53,63,168,171,172,174,176 It is apparent in studies that look at stem cell biolo gy that imaging pro vides a guide that directs tissue sampling to the appropriate sites and times such that the biochemical anal yses of specif ic molecules can be perfor med in a directed manner to pro vide more information. This is especiall y tr ue for assa ys that comprise genomics, proteomics, ph ysiomics, and gl ycomics because the number of multiple xed assa ys that can be performed can be se verely limited b y time and cost. Directed studies are thus imperati ve if we are to use our resources ef ficiently. The combination of imaging as a functional readout and high-throughput multiple xed assays has adv anced functional genomics w hereby w e can assess more than e xpression levels and can be gin to ascribe function to changes in gene expression. Although imaging is no w largely used to direct e x vivo assays of gene expression or protein levels in the areas of genomics and proteomics, it is becoming ob vious that w e can use imaging to de velop multiple xed in vi vo assa ys of gene expression and protein function in preclinical models. There are a limited number of unique bioluminescent proteins and chemistries that have been used in vivo, and therefore these assays will use one or two bioluminescent reporters45,49,51,120 but can incorporate multiple genes that alter the function of the target cell. Development of these techniques is the future of functional genomics as the y reveal the role that changes in expression have in specific biological processes and can link these changes to outcome—this cannot readil y be accomplished with standalone multiplexed assays. The benef its of preclinical imaging using bioluminescent, and other reporters, are obvious as is the role that these tools ha ve in testing compounds and deli very schemes for therapy. The use of these tools to study mammalian biolo gy and the transfer of this kno wledge will have a signif icant impact on ho w we approach the study of human biolo gy and in the management of disease. Imaging enab les the in vi vo study of cell biolo gy, and when integrated with thorough studies in culture and e x vivo, can re veal the nuances and subtleties of disease mechanisms and of therapeutic responses. In this w ay, visible animal models of human biolo gy and disease

133

comprise one of the most impor tant contributions of molecular imaging to human health as the y ser ve to accelerate and refine the analyses of mammalian biology and offer a rapid readout for the development of new therapies. The use of imaging in preclinical studies is apparent, but in applying these tools it is essential to ask which modality will most effectively and economically answer a given biological question with the greatest sensitivity and specificity. There is a wide range of choices and knowing what each modality can provide is crucial for developing an effective study design. As this chapter focuses on BLI, the ne xt section will attempt to ans wer the question, “Why perform imaging with bioluminescent markers?”

SUMMARY AND FUTURE OF BIOLUMINESCENCE REPORTERS IN LIVING ANIMALS The de velopment of ne w luciferase proteins and their respective substrates for use as molecular imaging tools will of fer po werful approaches for studying molecular changes in tissues and cells under physiological conditions where the conte xtual influences of intact or gan systems can be evaluated. BLI has already played a significant role in the study of animal models and , because it can be used to ref ine and accelerate these studies, has become a cornerstone technology in preclinical studies and the development of novel therapies. BLI and other imaging strate gies contribute to the de velopment of sophisticated animal models of human biolo gy and disease. The detection of internal sources of light that comprise BLI has largely been planar, and this too is evolving. The vast majority of studies using BLI represent the data as planar projections using pseudo-colored images to represent signal intensity, and these images are localized over g rayscale reference images of the subjects. Recent developments in three-dimensional (3D) reconstr uction are leading to instruments that generate 3D data sets from either multiple images from se veral vie ws, or from temporal196 or spectral data. 197,198 Reconstructing 3D images from bioluminescence data sets is not trivial199 and work in this area continues b y a number of g roups.200–202 All the advances in instr umentation for BLI are based on the same basic design that includes an imaging chamber that e xcludes ambient light and a sensiti ve CCD as a detector.15 Despite the simple design, there ha ve been a number of instr umentation advances for detecting bioluminescent signals in the body . A majority of these advances have been for the purpose of obtaining data that would permit the reconstr uction of 3D data sets. 197,203–205 These designs include a ring of detectors for obtaining

134

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

multiple vie ws, a stage for mo ving the animal coupled with a rotating mir ror for obtaining multiple vie ws, and improved spectral imaging using f ilters.197,206–208 Many of these advances are relatively new and are beginning to be applied to biological questions. We are at the v ery beginning of w hat promises to be a re volution in biolo gical investigation and in vi vo analyses of patholo gic changes. We ha ve seen the impact on studies in laborator y animals and its impact on clinical care is onl y beginning. As we approach the era of personalized medicine, the influence of the emerging tools of molecular imaging on biomedicine will lik ely be signif icant. in vi vo measures of cellular and molecular changes using imaging has re volutionized the study of laborator y animals and as these methods become inte grated into all f ields of biomedical research this will continue to grow. Perhaps the g reatest contributions to medicine of fered b y the field of molecular imaging will be increased understanding of the molecular basis of disease in animal models and the subsequent de velopment of ne w therapies that target these biological processes.

REFERENCES 1. Shah K, Weissleder R. Molecular optical imaging: applications leading to the de velopment of present da y therapeutics. NeuroRx 2005;2:215–25. 2. Gross S, Piwnica-Worms D. Spying on cancer: molecular imaging in vivo with genetically encoded reporters. Cancer Cell 2005;7:5–15. 3. Contag CH, Ross BD . It’s not just about anatom y: in vi vo bioluminescence imaging as an e yepiece into biolo gy. J Magn Reson Imaging 2002;16:378–87. 4. Contag CH, Bachmann MH. Advances in in vi vo bioluminescence imaging of gene e xpression. Annu Re v Biomed Eng 2002; 4:235–60. 5. Negrin RS, Contag CH. In vi vo imaging using bioluminescence: a tool for probing g raft-versus-host disease. Nat Re v Immunol 2006;6:484–90. 6. Hastings JW. Chemistries and colors of bioluminescent reactions: a review. Gene 1996;173:5–11. 7. Hastings JW , Morin JG. Photons for repor ting molecular e vents: green fluorescent protein and four luciferase systems. Methods Biochem Anal 2006;47:15–38. 8. Wilson T, Hastings JW. Bioluminescence. Annu Rev Cell De v Biol 1998;14:197–230. 9. Wood KV. Luc genes: introduction of colour into bioluminescence assays. J Biolumin Chemilumin 1990;5:107–14. 10. Venisnik KM, Olafsen T, Gambhir SS, Wu AM. Fusion of Gaussia luciferase to an engineered anti-carcinoembryonic antigen (CEA) antibody for in vi vo optical imaging. Mol Imaging Biol 2007;9:267–77. 11. Venisnik KM, Olafsen T, Loening AM, et al. Bifunctional antibodyRenilla luciferase fusion protein for in vi vo optical detection of tumors. Protein Eng Des Sel 2006;19:453–60. 12. So MK, Loening AM, Gambhir SS, Rao J. Creating self-illuminating quantum dot conjugates. Nat Protoc 2006;1:1160–4. 13. So MK, Xu C, Loening AM, et al. Self-illuminating quantum dot conjugates for in vivo imaging. Nat Biotechnol 2006;24:339–43.

14. Olafsen T, Cheung CW , Yazaki PJ, et al. Co valent disulf ide-linked anti-CEA diabody allo ws site-specif ic conjugation and radiolabeling for tumor tar geting applications. Protein Eng Des Sel 2004;17:21–7. 15. Rice BW, Cable MD, Nelson MB . In vi vo imaging of light-emitting probes. J Biomed Opt 2001;6:432–40. 16. Tromberg BJ, Shah N, Lanning R, et al. Non-invasive in vivo characterization of breast tumors using photon mig ration spectroscopy. Neoplasia 2000;2:26–40. 17. Jain RK, Munn LL, Fukumura D. Dissecting tumour pathophysiology using intravital microscopy. Nat Rev Cancer 2002;2:266–76. 18. Halin C, Rodrigo Mora J , Sumen C, v on Andrian UH. In vi vo imaging of l ymphocyte traf ficking. Annu Re v Cell De v Biol 2005; 21:581–603. 19. Hsiung PL, Hardy J, Friedland S, et al. Detection of colonic dysplasia in vi vo using a tar geted heptapeptide and confocal microendoscopy. Nat Med 2008;14:454–8. 20. Wang TD, Friedland S, Sahbaie P, et al. Functional imaging of colonic mucosa with a f ibered confocal microscope for real-time in vi vo pathology. Clin Gastroenterol Hepatol 2007;5:1300–5. 21. Wehrman TS, von Degenfeld G, Krutzik PO, et al. Luminescent imaging of β-galactosidase activity in living subjects using sequential reporter-enzyme luminescence. Nat Methods 2006;3:295–301. 22. Xu Y, Piston DW, Johnson CH. A bioluminescence resonance energy transfer (BRET) system: application to interacting circadian clock proteins. Proc Natl Acad Sci U S A 1999;96:151–6. 23. Shinde R, Perkins J, Contag CH. Luciferin derivatives for enhanced in vitro and in vi vo bioluminescence assa ys. Biochemistr y 2006; 45:11103–12. 24. Kanno A, Ozawa T, Umezawa Y. Intein-mediated reporter gene assay for detecting protein-protein interactions in li ving mammalian cells. Anal Chem 2006;78:556–60. 25. Ozawa T. Designing split repor ter proteins for anal ytical tools. Anal Chim Acta 2006;556:58–68. 26. Chan CT, Paulmurugan R, Gheysens OS, et al. Molecular imaging of the efficacy of heat shock protein 90 inhibitors in li ving subjects. Cancer Res 2008;68:216–26. 27. Villalobos V, Naik S, Piwnica-Worms D. Detection of protein-protein interactions in li ve cells and animals with split f irefly luciferase protein fragment complementation. Methods Mol Biol 2008;439:339–52. 28. Hardy J, Francis KP, DeBoer M, et al. Extracellular replication of Listeria monoc ytogenes in the murine gall b ladder. Science 2004; 303:851–3. 29. Hardy J, Margolis JJ, Contag CH. Induced biliar y excretion of Listeria monocytogenes. Infect Immun 2006;74:1819–27. 30. Germano G, Chen BC, Huang SC, et al. Use of the abdominal aor ta for arterial input function deter mination in hepatic and renal PET studies. J Nucl Med 1992;33:613–20. 31. Choi Y, Hawkins RA, Huang SC, et al. Parametric images of myocardial metabolic rate of glucose generated from dynamic cardiac PET and 2-[18F]fluoro-2-deo xy-d-glucose studies. J Nucl Med 1991;32:733–8. 32. Contag CH, Contag PR, Mullins JI, et al. Photonic detection of bacterial pathogens in living hosts. Mol Microbiol 1995;18:593–603. 33. Contag CH, Spilman SD, Contag PR, et al. Visualizing gene expression in li ving mammals using a bioluminescent repor ter. Photochem Photobiol 1997;66:523–31. 34. Tjuvajev JG, Stockhammer G, Desai R, et al. Imaging the e xpression of transfected genes in vivo. Cancer Res 1995;55:6126–32. 35. Sadikot RT, Blackwell TS. Bioluminescence imaging. Proc Am Thorac Soc 2005;2:537–40, 511–32. 36. Contag CH. In vivo pathology: seeing with molecular specif icity and cellular resolution in the li ving body . Annu Re v P athol 2007; 2:277–305. 37. Contag PR, Olomu IN , Stevenson DK, Contag CH. Bioluminescent indicators in living mammals. Nat Med 1998;4:245–7.

Functional Imaging Using Bioluminescent Markers

38. Troy T, Jekic-McMullen D , Sambucetti L, Rice B . Quantitati ve comparison of the sensitivity of detection of fluorescent and bioluminescent reporters in animal models. Mol Imaging 2004;3:9–23. 39. Steele RW. Clinical applications of chemiluminescence of g ranulocytes. Rev Infect Dis 1991;13:918–25. 40. Shapiro E, Lu C, Bane yx F. A set of multicolored Photinus p yralis luciferase mutants for in vi vo bioluminescence applications. Protein Eng Des Sel 2005;18:581–7. 41. Wood KV, Lam YA, Seliger HH, McElroy WD. Complementary DNA coding click beetle luciferases can elicit bioluminescence of different colors. Science 1989;244:700–2. 42. de Wet JR, Wood KV, DeLuca M, et al. Firefly luciferase gene: structure and expression in mammalian cells. Mol Cell Biol 1987;7:725–37. 43. Michelini E, Cevenini L, Mezzanotte L, et al. Combining intracellular and secreted bioluminescent repor ter proteins for multicolor cell-based assays. Photochem Photobiol Sci 2008;7:212–7. 44. Shao N, Bock R. A codon-optimized luciferase from Gaussia princeps facilitates the in vivo monitoring of gene expression in the model alga Chlamydomonas reinhardtii. Curr Genet 2008;53:381–8. 45. Tannous BA, Kim DE, Fernandez JL, et al. Codon-optimized Gaussia luciferase cDNA for mammalian gene expression in culture and in vivo. Mol Ther 2005;11:435–43. 46. Doyle TC, Burns SM, Contag CH. In vi vo bioluminescence imaging for integrated studies of infection. Cell Microbiol 2004;6:303–17. 47. Stolz U, Velez S, Wood KV, et al. Darwinian natural selection for orange bioluminescent color in a Jamaican click beetle. Proc Natl Acad Sci U S A 2003;100:14955–9. 48. Loening AM, Wu AM, Gambhir SS. Red-shifted Renilla renifor mis luciferase v ariants for imaging in li ving subjects. Nat Methods 2007;4:641–3. 49. Zhao H, Doyle TC, Coquoz O, et al. Emission spectra of bioluminescent reporters and interaction with mammalian tissue determine the sensitivity of detection in vivo. J Biomed Opt 2005;10:41210. 50. Kadurugamuwa JL, Modi K, Coquoz O, et al. Reduction of astrogliosis by early treatment of pneumococcal meningitis measured b y simultaneous imaging, in vivo, of the pathogen and host response. Infect Immun 2005;73:7836–43. 51. Bhaumik S, Le wis XZ, Gambhir SS. Optical imaging of Renilla luciferase, synthetic Renilla luciferase, and firefly luciferase reporter gene expression in living mice. J Biomed Opt 2004;9:578–86. 52. Sweeney TJ, Mailander V, Tucker AA, et al. Visualizing the kinetics of tumor-cell clearance in living animals. Proc Natl Acad Sci U S A 1999;96:12044–9. 53. Cao YA, Wagers AJ, Beilhack A, et al. Shifting foci of hematopoiesis during reconstitution from single stem cells. Proc NatlAcad Sci U S A 2004;101:221–6. 54. Paulmurugan R, Umeza wa Y, Gambhir SS. Nonin vasive imaging of protein-protein interactions in li ving subjects b y using repor ter protein complementation and reconstitution strate gies. Proc Natl Acad Sci U S A 2002;99:15608–13. 55. Ray P, Pimenta H, P aulmurugan R, et al. Nonin vasive quantitati ve imaging of protein-protein interactions in li ving subjects. Proc Natl Acad Sci U S A 2002;99:3105–10. 56. Paulmurugan R, Gambhir SS. Monitoring protein-protein interactions using split synthetic renilla luciferase protein-fragment-assisted complementation. Anal Chem 2003;75:1584–9. 57. Luker KE, Piwnica-Worms D. Optimizing luciferase protein fragment complementation for bioluminescent imaging of protein-protein interactions in li ve cells and animals. Methods Enzymol 2004; 385:349–60. 58. Wang Y, Wang G, O’Kane DJ , Szalay AA. A study of protein-protein interactions in li ving cells using luminescence resonance energy transfer (LRET) from Renilla luciferase to Aequorea GFP. Mol Gen Genet 2001;264:578–87. 59. Cecic I, Chan D A, Sutphin PD, et al. Oxygen sensiti vity of repor ter genes: implications for preclinical imaging of tumor hypoxia. Mol Imaging 2007;6:219–28.

135

60. Gambhir SS, Herschman HR, Cher ry SR, et al. Imaging transgene expression with radionuclide imaging technolo gies. Neoplasia 2000;2:118–38. 61. Chalfie M. Green fluorescent protein. Photochem Photobiol 1995; 62:651–6. 62. Chalfie M, Tu Y, Euskirchen G, et al. Green fluorescent protein as a marker for gene expression. Science 1994;263:802–5. 63. Tolar J, Osbor n M, Bell S, et al. Real-time in vi vo imaging of stem cells follo wing transgenesis b y transposition. Mol Ther 2005; 12:42–8. 64. Zhang M, Alicot EM, Chiu I, et al. Identif ication of the tar get selfantigens in reperfusion injury. J Exp Med 2006;203:141–52. 65. Siegel RM, Chan FK, Zacharias D A, et al. Measurement of molecular interactions in li ving cells b y fluorescence resonance ener gy transfer betw een v ariants of the g reen fluorescent protein. Sci STKE 2000;2000:PL1. 66. Tsien R Y. The g reen fluorescent protein. Annu Re v Biochem 1998;67:509–44. 67. Tsien RY, Miyawaki A. Seeing the machiner y of li ve cells. Science 1998;280:1954–5. 68. Cubitt AB, Heim R, Adams SR, et al. Understanding, impro ving and using g reen fluorescent proteins. Trends Biochem Sci 1995; 20:448–55. 69. Louie AY, Huber MM, Ahrens ET, et al. In vivo visualization of gene expression using magnetic resonance imaging. Nat Biotechnol 2000;18:321–5. 70. Moore A, Basilion JP, Chiocca EA, Weissleder R. Measuring transferrin receptor gene e xpression by NMR imaging. Biochim Biophys Acta 1998;1402:239–49. 71. Funovics M, Weissleder R, Tung CH. Protease sensors for bioimaging. Anal Bioanal Chem 2003;377:956–63. 72. Jaffer FA, Tung CH, Gerszten RE, Weissleder R. In vi vo imaging of thrombin acti vity in e xperimental thrombi with thrombin-sensitive near-infrared molecular probe. Arterioscler Thromb Vasc Biol 2002;22:1929–35. 73. Chen J, Tung CH, Mahmood U , et al. In vi vo imaging of proteol ytic activity in atherosclerosis. Circulation 2002;105:2766–71. 74. Weissleder R, Tung CH, Mahmood U, Bogdanov A Jr. In vivo imaging of tumors with protease-acti vated near -infrared fluorescent probes. Nat Biotechnol 1999;17:375–8. 75. von Andrian UH, Berger EM, Ramezani L, et al. In vi vo behavior of neutrophils from tw o patients with distinct inherited leuk ocyte adhesion deficiency syndromes. J Clin Invest 1993;91:2893–7. 76. Mazo IB , Gutier rez-Ramos JC, F renette PS, et al. Hematopoietic progenitor cell rolling in bone mar row micro vessels: parallel contributions b y endothelial selectins and v ascular cell adhesion molecule 1. J Exp Med 1998;188:465–74. 77. Dewhirst MW, Shan S, Cao Y, et al. Intravital fluorescence facilitates measurement of multiple ph ysiologic functions and gene e xpression in tumors of live animals. Dis Markers 2002;18:293–311. 78. Sumen C, Mempel TR, Mazo IB , v on Andrian UH. Intra vital microscopy: visualizing immunity in conte xt. Immunity 2004; 21:315–29. 79. Lin DA, Wei X, Wu JW, et al. In vi vo imaging of specialized bone marrow endothelial microdomains for tumour eng raftment. Nature 2005;435:969–73. 80. Germain RN, Castellino F, Chieppa M, et al. An extended vision for dynamic high-resolution intra vital immune imaging. Semin Immunol 2005;17:431–41. 81. Gooch VD, Mehra A, Lar rondo LF , et al. Full y codon-optimized luciferase uncovers novel temperature characteristics of the Neurospora clock. Eukaryot Cell 2008;7:28–37. 82. Patterson SS, Dionisi HM, Gupta RK, Sayler GS. Codon optimization of bacterial luciferase (lux) for e xpression in mammalian cells. J Ind Microbiol Biotechnol 2005;32:115–23. 83. Nielsen TO, West RB, Linn SC, et al. Molecular characterisation of soft tissue tumours: a gene expression study. Lancet 2002;359:1301–7.

136

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

84. Benaron D A, Cheong WF, Ste venson DK. Tissue optics. Science 1997;276:2002–3. 85. Jobsis FF. Noninvasive, infrared monitoring of cerebral and myocardial o xygen suf ficiency and circulator y parameters. Science 1977;198:1264–7. 86. Thorne SH, Negrin RS, Contag CH. Synergistic antitumor effects of immune cell-viral biotherapy. Science 2006;311:1780–4. 87. Sosnovik D, Weissleder R. Magnetic resonance and fluorescence based molecular imaging technologies. Prog Drug Res 2005;62:83–115. 88. Lipshutz GS, Gr uber CA, Cao Y, et al. In utero deli very of adenoassociated viral v ectors: intraperitoneal gene transfer produces long-term expression. Mol Ther 2001;3:284–92. 89. Lipshutz GS, Titre D, Brindle M, et al. Comparison of gene e xpression after intraperitoneal delivery of AAV2 or AAV5 in utero. Mol Ther 2003;8:90–8. 90. Lee CT, Lee YJ, Kwon SY, et al. In vi vo imaging of adeno virus transduction and enhanced therapeutic ef ficacy of combination therapy with conditionall y replicating adeno virus and adenovirus-p27. Cancer Res 2006;66:372–7. 91. McCaffrey A, Kay MA, Contag CH. Advancing molecular therapies through in vi vo bioluminescent imaging. Mol Imaging 2003; 2:75–86. 92. McCaffrey AP, Meuse L, Karimi M, et al. A potent and specific morpholino antisense inhibitor of hepatitis C translation in mice. Hepatology 2003;38:503–8. 93. Hickerson RP, Smith FJ, Reeves RE, et al. Single-nucleotide-specific siRNA targeting in a dominant-negative skin model. J Invest Dermatol 2008;128:594–605. 94. Ilves H, Kaspar RL, Wang Q, et al. Inhibition of hepatitis C IRESmediated gene expression by small hairpin RNAs in human hepatocytes and mice. Ann N Y Acad Sci 2006;1082:52–5. 95. Smith FJ, Hickerson RP, Sayers JM, et al. Development of therapeutic siRNAs for pachyonychia congenita. J In vest Dermatol 2008; 128:50–8. 96. Wang Q, Contag CH, Ilv es H, et al. Small hair pin RNAs efficiently inhibit hepatitis C IRES-mediated gene e xpression in human tissue culture cells and a mouse model. Mol Ther 2005;12:562–8. 97. Wang Q, Ilv es H, Chu P , et al. Deli very and inhibition of repor ter genes by small interfering RNAs in a mouse skin model. J In vest Dermatol 2007;127:2577–84. 98. Zhang GJ, Kaelin WG Jr. Bioluminescent imaging of ubiquitin ligase activity: measuring Cdk2 activity in vivo through changes in p27 turnover. Methods Enzymol 2005;399:530–49. 99. Laxman B, Hall DE, Bhojani MS, et al. Noninvasive real-time imaging of apoptosis. Proc Natl Acad Sci U S A 2002;99:16551–5. 100. Edinger M, Cao YA, Verneris MR, et al. Re vealing l ymphoma growth and the ef ficacy of immune cell therapies using in vivo bioluminescence imaging. Blood 2003;101:640–8. 101. Franks NP, Jenkins A, Conti E, et al. Str uctural basis for the inhibition of f irefly luciferase b y a general anesthetic. Bioph ys J 1998;75:2205–11. 102. Conti E, F ranks NP, Brick P. Cr ystal str ucture of f irefly luciferase throws light on a superf amily of aden ylate-forming enzymes. Structure 1996;4:287–98. 103. O’Connell-Rodwell CE, Shriver D, Simanovskii DM, et al. A genetic reporter of ther mal stress def ines ph ysiologic zones o ver a defined temperature range. FASEB J 2004;18:264–71. 104. Wood KV, Lam YA, McElroy WD, Seliger HH. Bioluminescent click beetles revisited. J Biolumin Chemilumin 1989;4:31–9. 105. Karkhanis YD, Cormier MJ. Isolation and properties of Renilla reniformis luciferase, a lo w molecular w eight ener gy con version enzyme. Biochemistry 1971;10:317–26. 106. Matthews JC, Hori K, Cor mier MJ. Purif ication and proper ties of Renilla reniformis luciferase. Biochemistry 1977;16:85–91. 107. Srikantha T, Klapach A, Lorenz WW, et al. The sea pansy Renilla reniformis luciferase serves as a sensitive bioluminescent reporter

108.

109.

110.

111.

112.

113.

114.

115.

116.

117. 118.

119.

120.

121. 122.

123.

124.

125.

126.

127.

128.

for differential gene e xpression in Candida albicans. J Bacteriol 1996;178:121–9. Shimomura O , Johnson FH. Chemical nature of bioluminescence systems in coelenterates. Proc Natl Acad Sci U S A 1975;72: 1546–9. Greer LF III, Szalay AA. Imaging of light emission from the expression of luciferases in li ving cells and or ganisms: a review. Luminescence 2002;17:43–74. Loening AM, F enn TD, Gambhir SS. Cr ystal str uctures of the luciferase and g reen fluorescent protein from Renilla renifor mis. J Mol Biol 2007;374:1017–28. Loening AM, F enn TD, Wu AM, Gambhir SS. Consensus guided mutagenesis of Renilla luciferase yields enhanced stability and light output. Protein Eng Des Sel 2006;19:391–400. Zhao H, Do yle TC, Wong RJ, et al. Characterization of coelenterazine analogs for measurements of Renilla luciferase acti vity in live cells and living animals. Mol Imaging 2004;3:43–54. Luo J, Ho P, Steinman L, Wyss-Coray T. Bioluminescence in vi vo imaging of autoimmune encephalom yelitis predicts disease. J Neuroinflammation 2008;5:6. Lin AH, Luo J , Mondshein LH, et al. Global anal ysis of Smad2/3dependent TGF-β signaling in li ving mice re veals prominent tissue-specific responses to injury. J Immunol 2005;175:547–54. Ando T, Hatsushika K, Wako M, et al. Orally administered TGF-β is biologically active in the intestinal mucosa and enhances oral tolerance. J Allergy Clin Immunol 2007;120:916–23. Luo J, Lin AH, Masliah E, Wyss-Coray T. Bioluminescence imaging of Smad signaling in living mice shows correlation with excitotoxic neurodegeneration. Proc Natl Acad Sci U S A 2006;103:18326–31. Tesseur I, Zou K, Berber E, et al. Highly sensitive and specific bioassay for measuring bioactive TGF-β. BMC Cell Biol 2006;7:15. Levi J, De A, Cheng Z, Gambhir SS. Bisdeo xycoelenterazine derivatives for impro vement of bioluminescence resonance ener gy transfer assays. J Am Chem Soc 2007;129:11900–1. Jones LR, Goun EA, Shinde R, et al. Releasable luciferin-transporter conjugates: tools for the real-time anal ysis of cellular uptak e and release. J Am Chem Soc 2006;128:6526–7. Bhaumik S, Gambhir SS. Optical imaging of Renilla luciferase reporter gene expression in living mice. Proc Natl Acad Sci U S A 2002;99:377–82. Ray P, De A, Min JJ, et al. Imaging tri-fusion multimodality reporter gene expression in living subjects. Cancer Res 2004;64:1323–30. Ray P , Tsien R, Gambhir SS. Constr uction and v alidation of improved triple fusion reporter gene vectors for molecular imaging of living subjects. Cancer Res 2007;67:3085–93. Cao YA, Bachmann MH, Beilhack A, et al. Molecular imaging using labeled donor tissues re veals patter ns of eng raftment, rejection, and sur vival in transplantation. Transplantation 2005;80:134–9. Ponomarev V, Doubrovin M, Serganova I, et al. A novel triple-modality reporter gene for whole-body fluorescent, bioluminescent, and nuclear noninvasive imaging. Eur J Nucl Med Mol Imaging 2004; 31:740–51. Day RN , Kawecki M, Ber ry D. Dual-function repor ter protein for analysis of gene e xpression in li ving cells. Biotechniques 1998;25:848–50, 852–4, 856. Edinger M, Hof fmann P, Contag CH, Ne grin RS. Ev aluation of effector cell f ate and function b y in vi vo bioluminescence imaging. Methods 2003;31:172–9. Mandl S, Mari C, Edinger M, et al. In vi vo dynamics of tumor cell death: multi-modality imaging identif ies key imaging times for assessing response to chemotherap y using an or thotopic mouse model of lymphoma. 2003. [Submitted] Yang Y, Contag CH, F elsher D, et al. The E47 transcription f actor negatively regulates CD5 e xpression during th ymocyte development. Proc Natl Acad Sci U S A 2004;101:3898–902.

Functional Imaging Using Bioluminescent Markers

129. Smith SJ , Zhang H, Cler mont AO, et al. In vi vo monitoring of pancreatic β-cells in a transgenic mouse model. Mol Imaging 2006;5:65–75. 130. Shachaf CM, K opelman AM, Arvanitis C, et al. MYC inacti vation uncovers pluripotent dif ferentiation and tumour dor mancy in hepatocellular cancer. Nature 2004;431:1112–7. 131. Rehemtulla A, Hall DE, Ste gman LD, et al. Molecular imaging of gene expression and efficacy following adenoviral-mediated brain tumor gene therapy. Mol Imaging 2002;1:43–55. 132. Slavin AJ, Tarner IH, Nakajima A, et al. Adoptive cellular gene therapy of autoimmune disease. Autoimmun Rev 2002;1:213–9. 133. Tarner IH, Nakajima A, Seroogy CM, et al. Retro viral gene therapy of collagen-induced ar thritis b y local deli very of IL-4. Clin Immunol 2002;105:304–14. 134. Eefting D, Grimbergen JM, de Vries MR, et al. Prolonged in vi vo gene silencing b y electroporation-mediated plasmid deli very of small interfering RNA. Hum Gene Ther 2007;18:861–9. 135. Chen ZY, Riu E, He CY , et al. Silencing of episomal transgene expression in li ver by plasmid bacterial backbone DN A is independent of CpG methylation. Mol Ther 2008;16:548–56. 136. Chen ZY, He CY, Meuse L, Ka y MA. Silencing of episomal transgene expression by plasmid bacterial DNA elements in vivo. Gene Ther 2004;11:856–64. 137. De A, Gambhir SS. Noninvasive imaging of protein-protein interactions from li ve cells and li ving subjects using bioluminescence resonance energy transfer. FASEB J 2005;19:2017–9. 138. Hoshino H, Nakajima Y, Ohmiya Y. Luciferase-YFP fusion tag with enhanced emission for single-cell luminescence imaging. Nat Methods 2007;4:637–9. 139. Massoud TF, Paulmurugan R, De A, et al. Reporter gene imaging of protein-protein interactions in living subjects. Curr Opin Biotechnol 2007;18:31–7. 140. Carriba P, Navarro G, Ciruela F, et al. Detection of heteromerization of more than tw o proteins by sequential BRET-FRET. Nat Methods 2008;5:727–33. 141. Arai R, Nakagawa H, Kitayama A, et al. Detection of protein-protein interaction b y bioluminescence resonance ener gy transfer from firefly luciferase to red fluorescent protein. J Biosci Bioeng 2002; 94:362–4. 142. Issad T, Boute N , P ernet K. The acti vity of the insulin receptor assessed by bioluminescence resonance ener gy transfer. Ann N Y Acad Sci 2002;973:120–3. 143. Boute N, Jockers R, Issad T. The use of resonance energy transfer in high-throughput screening: BRET v ersus FRET. Trends Pharmacol Sci 2002;23:351–4. 144. Thorne SH, Contag CH. Integrating the biological characteristics of oncolytic viruses and immune cells can optimize therapeutic benefits of cell-based delivery. Gene Ther 2008;15:753–8. 145. Dickson PV, Hamner B, Ng CY, et al. In vivo bioluminescence imaging for early detection and monitoring of disease progression in a murine model of neuroblastoma. J Pediatr Surg 2007;42:1172–9. 146. Nathan CO , Amirghahari N , Rong X, et al. Mammalian tar get of rapamycin inhibitors as possible adjuvant therapy for microscopic residual disease in head and neck squamous cell cancer . Cancer Res 2007;67:2160–8. 147. Wetterwald A, van der Pluijm G, Que I, et al. Optical imaging of cancer metastasis to bone mar row: a mouse model of minimal residual disease. Am J Pathol 2002;160:1143–53. 148. Flanagan JM, Truksa J, Peng H, et al. In vi vo imaging of hepcidin promoter stimulation by iron and inflammation. Blood Cells Mol Dis 2007;38:253–7. 149. Kaneda Y. Impro vements in gene therap y technolo gies. Mol Urol 2001;5:85–9. 150. Schmid T, Young MR. Lights on for lo w o xygen: a nonin vasive mouse model useful for sensing o xygen def iciency. Mol Inter v 2006;6:20–2.

137

151. Harada H, Kizaka-Kondoh S, Hiraoka M. Optical imaging of tumor hypoxia and evaluation of efficacy of a hypoxia-targeting drug in living animals. Mol Imaging 2005;4:182–93. 152. Safran M, Kim WY, O’Connell F, et al. Mouse model for nonin vasive imaging of HIF prolyl hydroxylase activity: assessment of an oral agent that stimulates er ythropoietin production. Proc Natl Acad Sci U S A 2006;103:105–10. 153. Contag CH, Ste venson DK. In vi vo patter ns of heme o xygenase-1 transcription [discussion]. J Perinatol 2001;21 Suppl 1:S119–27. 154. Stankunas K, Ba yle JH, Ha vranek JJ, et al. Rescue of de gradationprone mutants of the FK506-rapam ycin binding (FRB) protein with chemical ligands. Chembiochem 2007;8:1162–9. 155. Banaszynski LA, Selme yer M, Thorne SH, et al. In vi vo control of protein stability. Nat Med 2008. [In press] 156. Ciana P, Raviscioni M, Mussi P, et al. In vi vo imaging of transcriptionally active estrogen receptors. Nat Med 2003;9:82–6. 157. Luker GD, Bardill JP, Prior JL, et al. Nonin vasive bioluminescence imaging of her pes simplex vir us type 1 infection and therap y in living mice. J Virol 2002;76:12149–61. 158. Wu JC, Inubushi M, Sundaresan G, et al. Optical imaging of cardiac reporter gene e xpression in li ving rats. Circulation 2002; 105:1631–4. 159. Bartlett DW, Su H, Hildebrandt IJ, et al. Impact of tumor-specific targeting on the biodistribution and ef ficacy of siRNA nanoparticles measured by multimodality in vivo imaging. Proc Natl Acad Sci U S A 2007;104:15549–54. 160. Zhang GJ, Safran M, Wei W, et al. Bioluminescent imaging of Cdk2 inhibition in vivo. Nat Med 2004;10:643–8. 161. Moldt B, Yant SR, Andersen PR, et al. Cis-acting gene re gulatory activities in the terminal regions of sleeping beauty DNA transposon-based vectors. Hum Gene Ther 2007;18:1193–204. 162. Wilber A, F randsen JL, Wangensteen KJ , et al. Dynamic gene expression after systemic delivery of plasmid DNA as determined by in vi vo bioluminescence imaging. Hum Gene Ther 2005;16:1325–32. 163. Kim SI, Shin D , Choi TH, et al. Systemic and specif ic deli very of small interfering RNAs to the liver mediated by apolipoprotein A-I. Mol Ther 2007;15:1145–52. 164. Suzuki R, Takizawa T, Ne gishi Y, et al. Tumor specif ic ultrasound enhanced gene transfer in vi vo with no vel liposomal bubb les. J Control Release 2008;125:137–44. 165. Iyer M, Berenji M, Templeton NS, Gambhir SS. Nonin vasive imaging of cationic lipid-mediated delivery of optical and PET reporter genes in living mice. Mol Ther 2002;6:555–62. 166. Wender PA, Goun EA, Jones LR, et al. Real-time anal ysis of uptake and bioactivatable cleavage of luciferin-transpor ter conjugates in transgenic repor ter mice. Proc Natl Acad Sci U S A 2007; 104:10340–5. 167. Prins RM, Shu CJ, Radu CG, et al. Anti-tumor activity and trafficking of self, tumor -specific T cells against tumors located in the brain. Cancer Immunol Immunother 2008;57:1279–89. 168. Li Z, Wu JC, Sheikh AY, et al. Differentiation, survival, and function of embr yonic stem cell deri ved endothelial cells for ischemic heart disease. Circulation 2007;116:I46–54. 169. Costa GL, Sandora MR, Nakajima A, et al. Adoptive immunotherapy of experimental autoimmune encephalomyelitis via T cell delivery of the IL-12 p40 subunit. J Immunol 2001;167:2379–87. 170. Nakajima A, Seroo gy CM, Sandora MR, et al. Antigen-specific T cell-mediated gene therapy in collagen-induced arthritis. J Clin Invest 2001;107:1293–301. 171. Sheikh AY, Lin SA, Cao F, et al. Molecular imaging of bone marrow mononuclear cell homing and eng raftment in ischemic myocardium. Stem Cells 2007;25:2677–84. 172. Degano IR, Vilalta M, Bago JR, et al. Bioluminescence imaging of calvarial bone repair using bone mar row and adipose tissuederived mesenchymal stem cells. Biomaterials 2008;29:427–37.

138

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

173. BitMansour A, Bur ns SM, Traver D , et al. My eloid pro genitors protect against invasive aspergillosis and Pseudomonas aeruginosa infection following hematopoietic stem cell transplantation. Blood 2002;100:4660–7. 174. Chan KM, Raikw ar SP, Za vazava N . Strate gies for dif ferentiating embryonic stem cells (ESC) into insulin-producing cells and development of non-in vasive imaging techniques using bioluminescence. Immunol Res 2007;39:261–70. 175. Heckl S. Future contrast agents for molecular imaging in strok e. Curr Med Chem 2007;14:1713–28. 176. Okada S, Ishii K, Yamane J, et al. In vivo imaging of engrafted neural stem cells: its application in e valuating the optimal timing of transplantation for spinal cord injury. FASEB J 2005;19:1839–41. 177. Olivo C, Alblas J, Verweij V, et al. In vi vo bioluminescence imaging study to monitor ectopic bone for mation b y luciferase gene marked mesenchymal stem cells. J Or thop Res 2008;26:901–9. 178. Tanaka M, Swijnenburg RJ, Gunawan F, et al. In vivo visualization of cardiac allo graft rejection and traf ficking passenger leuk ocytes using bioluminescence imaging. Circulation 2005;112:I105–10. 179. Kim JB, Leucht P, Mor rell NT, et al. Visualizing in vi vo liposomal drug delivery in real-time. J Dr ug Target 2007;15:632–9. 180. Roberts ES, Malstrom SE, Dreher KL. In situ pulmonar y localization of air pollution par ticle-induced o xidative stress. J Toxicol Environ Health A 2007;70:1929–35. 181. Dohlen G, Odland HH, Carlsen H, et al. Antioxidant activity in the newborn brain: a luciferase mouse model. Neonatolo gy 2008; 93:125–31. 182. Su H, van Dam GM, Buis CI, et al. Spatiotemporal expression of heme oxygenase-1 detected b y in vi vo bioluminescence after hepatic ischemia in HO-1/Luc mice. Liver Transpl 2006;12:1634–9. 183. Wilmink GJ, Opalenik SR, Beckham JT, et al. Assessing laser-tissue damage with bioluminescent imaging. J Biomed Opt 2006; 11:041114. 184. Weir LR, Schenck E, Meakin J , et al. Biophotonic imaging in HO-1.luc transgenic mice: real-time demonstration of gender specific chlorofor m induced renal to xicity. Mutat Res 2005; 574:67–75. 185. Schimmelpfennig CH, Schulz S, Arber C, et al. Ex vi vo expanded dendritic cells home to T-cell zones of l ymphoid organs and survive in vi vo after allo geneic bone mar row transplantation. Am J Pathol 2005;167:1321–31. 186. Chan JK, Hamilton CA, Cheung MK, et al. Enhanced killing of primary o varian cancer b y retar geting autolo gous c ytokineinduced killer cells with bispecif ic antibodies: a preclinical study. Clin Cancer Res 2006;12:1859–67. 187. Kim D, Hung CF , Wu TC. Monitoring the traf ficking of adopti vely transferred antigen- specific CD8-positive T cells in vivo, using noninvasive luminescence imaging. Hum Gene Ther 2007;18:575–88. 188. Scheffold C, K ornacker M, Schef fold YC, et al. Visualization of effective tumor targeting by CD8+ natural killer T cells redirected with bispecif ic antibody F(ab’)(2)HER2xCD3. Cancer Res 2002;62:5785–91. 189. Zhang C, Lou J , Li N , et al. Donor CD8+ T cells mediate g raftversus-leukemia acti vity without clinical signs of g raft-versushost disease in recipients conditioned with anti-CD3 monoclonal antibody. J Immunol 2007;178:838–50. 190. Beilhack A, Schulz S, Baker J, et al. In vivo analyses of early events in acute g raft-versus-host disease re veal sequential inf iltration of T-cell subsets. Blood 2005;106:1113–22. 191. Lee MH, Lee WH, Van Y, et al. Image-guided analyses reveal that nonCD4 splenocytes contribute to CD4+ T cell-mediated inflammation leading to islet destr uction by altering their local function and not systemic trafficking patterns. Mol Imaging 2007;6:369–83.

192. Garcia T, Jackson A, Bachelier R, et al. A con venient clinicall y relevant model of human breast cancer bone metastasis. Clin Exp Metastasis 2008;25:33–42. 193. Cowey S, Szafran AA, Kappes J , et al. Breast cancer metastasis to bone: evaluation of bioluminescent imaging and microSPECT/CT for detecting bone metastasis in immunodef icient mice. Clin Exp Metastasis 2007;24:389–401. 194. Hensley H, P ollack A. In vi vo visualization of metastatic prostate cancer. Cancer Biol Ther 2003;2:661–2. 195. Rosol TJ, Tannehill-Gregg SH, LeRo y BE, et al. Animal models of bone metastasis. Cancer 2003;97:748–57. 196. Bloch S, Lesage F , McIntosh L, et al. Whole-body fluorescence lifetime imaging of a tumor -targeted near -infrared molecular probe in mice. J Biomed Opt 2005;10:054003. 197. Chaudhari AJ, Darvas F, Bading JR, et al. Hyperspectral and multispectral bioluminescence optical tomography for small animal imaging. Phys Med Biol 2005;50:5421–41. 198. Ntziachristos V, Ripoll J , Wang LV, Weissleder R. Looking and listening to light: the e volution of w hole-body photonic imaging. Nat Biotechnol 2005;23:313–20. 199. Burcin Unlu M, Gulsen G. Ef fects of the time dependence of a bioluminescent source on the tomo graphic reconstr uction. Appl Opt 2008;47:799–806. 200. Li S, Zhang Q, Jiang H. Two-dimensional bioluminescence tomo graphy: numerical simulations and phantom experiments. Appl Opt 2006;45:3390–4. 201. Dehghani H, Da vis SC, Jiang S, et al. Spectrall y resolved bioluminescence optical tomography. Opt Lett 2006;31:365–7. 202. Alexandrakis G, Rannou FR, Chatziioannou AF. Tomographic bioluminescence imaging b y use of a combined optical-PET (OPET) system: a computer simulation feasibility study . Phys Med Biol 2005;50:4225–41. 203. Dogdas B, Stout D, Chatziioannou AF, Leahy RM. Digimouse: a 3D whole body mouse atlas from CT and cryosection data. Phys Med Biol 2007;52:577–87. 204. Soloviev VY. Tomographic bioluminescence imaging with v arying boundary conditions. Appl Opt 2007;46:2778–84. 205. Ahn S, Chaudhari AJ, Dar vas F, et al. F ast iterati ve image reconstruction methods for full y 3D multispectral bioluminescence tomography. Phys Med Biol 2008;53:3921–42. 206. Kumar D, Cong WX, Wang G. Monte Carlo method for bioluminescence tomography. Indian J Exp Biol 2007;45:58–63. 207. Wang G, Cong W, Shen H, et al. Ov erview of bioluminescence tomography—a ne w molecular imaging modality . F ront Biosci 2008;13:1281–93. 208. Tonary AM, Pezacki JP. Simultaneous quantitati ve measurement of luciferase repor ter acti vity and cell number in tw o- and threedimensional cultures of hepatitis C virus replicons. Anal Biochem 2006;350:239–48. 209. Wood KV , Lam YA, McElro y WD. Introduction to beetle luciferases and their applications. J Biolumin Chemilumin 1989; 4:289–301. 210. Wiles S, Ferguson K, Stef anidou M, et al. Alternative luciferase for monitoring bacterial cells under adverse conditions. Appl Environ Microbiol 2005;71:3427–32. 211. Verhaegent M, Christopoulos TK. Recombinant Gaussia luciferase. Overexpression, purif ication, and analytical application of a bioluminescent repor ter for DN A h ybridization. Anal Chem 2002; 74:4378–85. 212. Shimomura O. A short story of aequorin. Biol Bull 1995;189:1–5. 213. Fisher AJ, Thompson TB, Thoden JB , et al. The 1.5-A resolution crystal str ucture of bacterial luciferase in lo w salt conditions. J Biol Chem 1996;271:21956–68.

9 OPTICAL MULTIMODALITY TECHNOLOGIES ARION F. CHATZIIOANNOU, PHD

In the pre vious chapters, v arious molecular imaging technologies and their combinations w ere introduced. As is commonplace with most technolo gical inno vations, new capabilities and features are often combined and inte grated with prior ar t, resulting in imaging systems that are f ar more capab le and v ersatile. We have already seen that optical bioluminescence imaging (BLI) has pro ven to be a tr uly e xceptional tool at the disposal of biological scientists (see Chapter 8, “Functional Imaging Using Bioluminescent Mark ers”).1,2 It combines a number of impor tant advantages, the most significant of which are (1) high sensitivity, (2) ease of operation, (3) low overall cost, and (4) an implementation that is an e xtension of assa ys that biolo gical researchers are already f amiliar with from the benchtop aspects of their w ork. When compared with radionuclide technolo gies, another k ey adv antage of the optical technologies is that the enzymatic reactions that produce visible light can continue as long as there is locall y a vailable ener gy, substrate, and enzyme, thereby producing thousands of photons per a vailable enzyme. In contrast, radionuclides produce a single emission from each parent isotope, based on a random decay process, meaning that there is no physical way of turning of f or quenching these emissions. Due to the unique combination of these adv antages, the penetration of in vi vo BLI into the instr umentation toolbox of molecular imaging scientists has been the f astest and highest compared with vir tually e very other in vi vo molecular imaging modality.3 On the flip side of these advantages is the limitation that visib le light propagation in li ving tissues is seriousl y hindered b y v ery strong scattering and strong absor ption, especially for wavelengths below 600 nm. 4 This fundamental ph ysical ef fect depends on the inherent optical proper ties of tissues, cells, and intracellular organelles and limits the application of BLI

to the superf icial tissues of lar ger animals, or the w hole body of v ery small mammals (mice). Fur thermore, this effect causes an une ven sensiti vity prof ile with strong preference for tar gets closer to the surf ace and tissues with low b lood content due to the absor ption of hemoglobin,5 and it precludes implementation in the general whole body sense in lar ger animals and cer tainly in humans.6 That does not mean that optical imaging is not applicable to humans, w here it can ha ve signif icant uses in endoscopic, intraoperati ve, and other similar conte xt applications. Another consequence of optical photon absorption and scattering is that the same physical effects also seriously affect image quantitation as well as spatial resolution, even within the small body of a mouse. 7,8 The adv antages listed abo ve are e xtremely compelling, and they have naturally stirred signif icant interest by biological researchers for in vivo BLI technologies. This de facto success has spa wned a ne w set of scientif ic endeavors, in which many research groups worldwide have embarked upon ef forts to provide (1) improved dedicated imaging instrumentation, (2) improved molecular imaging probes, and (3) methodolo gies for image reconstr uction algorithms, which will provide improved quantitation and spatial resolution (see Chapter 11, “Fluorescence Tomography” and Chapter 14, “Dif fuse Optical Tomography and Spectroscopy”). The goals of all of the abo ve efforts are two-fold: to first improve the qualitati ve and quantitati ve accuracies of routine optical imaging e xperiments and measurements, and second and perhaps most impor tant, to facilitate the translation of these measurements and the knowledge the y impar t to practical and useful clinical diagnostic applications in humans. 6 Achievement of these goals, as in essentiall y all aspects of molecular imaging, is a multidisciplinar y task that includes image reconstr uction algorithms, 8,9,10 photon propagation physics,11,12 advanced and sensitive 139

140

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

instrumentation technologies13, and the de velopment of novel molecular imaging probes. 14,15 One only needs to consider the significant leap of technologies from the in vitro realm of the cell culture flask and the petri dish, to the in vivo realm of small-animal imaging experiments, to the realm of human applications, to realize that multiple cr ucial v alidation e xperiments for both the technologies and the biolo gical conte xt are necessar y at each step. Fur thermore, in practice, radicall y dif ferent technologies have been optimized over the years and are used for each of these physical domains, ranging all the way from advanced microscopes, to clinical whole-body tomographic imaging systems (Figure 1). As stepping stones bridging this gap and f acilitating this translational validation, a series of technolo gies have been de veloped that span multiple molecular imaging modalities. These technolo gies include multimodality molecular imaging probes that are described in Chapter 29 “Multimodality Agents,” as w ell as multimodality imaging instr uments that include combinations of positron emission tomography (PET) and single photon emission computed tomo graphy (SPECT) systems with optical bioluminescence,16–19 combinations of optical and acoustic imaging detailed in Chapter 16 “Molecular Photoacoustic Tomography,” combinations of anatomic via X-rays and magnetic resonance imaging (MRI) with bioluminescence and fluorescence, 20–22 and e ven other optical imaging modalities such as optical transmission tomography.23–25 In this chapter, we will outline and describe the rationale and the goals behind the technical instr umentation development in multimodality PET/SPECT and X-ra y computed tomography (CT) with bioluminescence optical imaging and ho w the y f it together within the lar ger puzzle of molecular imaging. We will also discuss some of the technical hurdles that need to be overcome and how

A A

BB

Figure 1. A, Standard microscope, optimized for light collection from small specimens; B, Clinical imaging tomograph, optimized for whole-body surveys.

together with the work that describes the development of multimodality imaging probes in Chapter 29 “ Multimodality Agents,” these combinations should help us advance molecular imaging and bring us closer to the goals of the f ield.

MULTIMODALITY OPTICAL IMAGING GOALS One of the first issues that is considered within the context of the molecular imaging f ield is that the ultimate goal of an y of these de velopments is their application in human disease diagnosis and treatment. Some of the molecular imaging technologies have fundamental limitations when it comes to applications in humans as discussed abo ve. On one hand , optical imaging suf fers from poor tissue penetration. On the other hand , multiple high-sensiti vity molecular imaging modalities, whose response is independent of tar get depth, ha ve been established (PET/SPECT) and have been proven in the clinical environment.26,27 These modalities are based on imaging of radionuclide emissions, w hose fundamental infor mation car riers are man y orders of magnitude more energetic than those of optical photons in the visib le range (~10 5 eV vs 1–2 eV). Besides that significant dif ference, in se veral senses, in vi vo bioluminescence/fluorescence and PET/SPECT imaging are actually quite similar. They both are based on the measurement of photon signal emanating from a source inside the body and the y both depend on highl y sensitive photon detectors that are positioned outside the body. Furthermore, the generated signal at the source is, in principle, propor tional to the biolo gic process the y are examining. That is, these modalities are inherentl y dependent on direct signal measurements and not on signal contrast as in traditional MRI, 28 X-ray CT, and ultrasound applications. It is, therefore, these similarities that mak e it natural to consider that ne w optical molecular imaging probes and methodologies should be translated at their f inal stage into clinicall y a vailable radionuclide technologies. From the biolo gy side consequentl y, the o verall concept is to de velop and estab lish biological constr ucts that can be used in a similar f ashion to produce bioluminescence or fluorescence signal w hen used in small-animal preclinical models, or yield PET/SPECT signals, w hen used in the clinic with patients. Fur thermore, these dual probes can provide additional information when combined with intravital or local sur gical techniques, w here excised tissue can be visualized opticall y under a microscope. 29,30

Optical Multimodality Technologies

Because the optimization process of the de velopment of these molecular imaging methodolo gies is v ery time consuming and technicall y demanding, 31 the development of multimodality biological constructs is critical to f acilitate the ultimate translation of molecular imaging w ork based on bioluminescence and fluorescence on a biolo gist’s bench, to the clinic. Research from se veral g roups32–34 has yielded biological constr ucts acting as multimodality repor ter systems that can bind (or con vert and trap) either a PET/SPECT probe or produce emission of either bioluminescence or fluorescence photons from the same cells. Since the choice of these signals can be based on the experimental conte xt, the same biolo gical e xperiment can in essence be used and translated betw een in vi vo small-animal preclinical measurements and patients. This remarkab le technolo gical achie vement is an enabling technology that opens doors for translation of many molecular biolo gy e xperimental results based on optical bioluminescence and fluorescence into the clinical domain. Even under those circumstances, the tw o signals emanate as the result of dif ferent protein products, 32 and there is an underlying assumption that the two signals will be if not equal, at least proportional with each other.35 Because the measurements resulting from these biolo gic e xperiments should be directl y related to the le vel of produced photon signals, an impor tant step is the validation process for the underlying photon signal flux. This process naturall y entails the use of multimodality molecular imaging instruments that can simultaneously detect and quantify the signal from both optical and radionuclide emissions. These instruments should provide biologists with a tool to directl y quantify the signal under multiple dif ferent e xperimental conditions. A number of research g roups have developed or are in the process of de veloping such technologies. They are combinations of PET and SPECT systems with optical bioluminescence and fluorescence. With these imaging systems, one can simultaneously detect optical and radionuclide signals and examine different aspects of the complete imaging process that can be schematically described here in the following steps for imaging of gene e xpression: (1) delivery of the repor ter and therapeutic genes via the vector, (2) delivery of the imaging probe, (3) binding of the probe to its tar get, (4) clearance of the nonspecifically bound probe from its tar get, (5) signal generation at the source, and (6) signal propagation from the source through the tissues.

141

OPTICAL SIGNAL QUANTITATION A key technical parameter in this process is the quantif ication of optical signal at its source, deep within the body, which is something v ery different than the quantif ication of the signal emanating from the surf ace of a preclinical model. Due to the significant photon attenuation and scattering in tissues, absolute quantif ication and position information of optical signal at its source is a v ery difficult problem indeed and is still in its earl y stages. 36–38 In the de velopment process of these multimodality instr uments, as w ell as for the single modality imaging instr uments, it has been shown that the quantification of photon signal is highly dependent on (1) the exact optical properties of the tissues, (2) their spatial distribution, and (3) the wavelength of light at the source. Although (2) and (3) above can in principle be considered constants for each measurement in each animal, the e xact scattering and absorption coefficients of tissues could be affected by the physiologic state of the model at the time of the study , including b lood flow, fur ther complicating the measurements. Ev en so, as a result of the ef forts to pro vide answers to items (1) and (2) above, researchers have been led to the creation of multimodality optical imaging instruments that include strictly anatomic imaging instr uments such as X-ra y CT , MRI, and also systems that include optical transmission measurements. 9,21,25,39 The problem of robust reconstr uction of optical signal at the source, e ven with all this a priori infor mation is still exceedingly difficult and has y et to reliab ly replace semiquantitative estimates. It should be noted here that this is a significant research area in dif fuse optical tomo graphy (DOT) and that the concept of signal quantif ication at the source has multiple le vels of ans wers, ranging from the qualitative decision about a signal being increased or decreased “up or down,” to log order quantif ication, to the absolute quantif ication of the number of enzymatic reactions taking place in each second at the tar get site. Ev en though many researchers are stri ving for the latter def inition in tight conf idence intervals in an attempt to provide a truly quantitative explanation of the underlying biology,40,41 the f act is that biolo gical research has made signif icant progress by using simpler quantif ication schemes that lend themselves to easier data interpretation and higher throughput methodologies. A prime example of this approach is traditional BLI, which tends to serve at an earlier stage of the discovery process and feeds into more quantitati ve, but lower throughput and higher cost, imaging modalities.31 We will summarize here some of the technical aspects of these technically challenging and fascinating efforts.

142

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

PRINCIPLES OF OPTICAL AND RADIONUCLIDE IMAGING For the reader to understand the rationale behind the technologies used in these combinations, we will lay out here and summarize the principles of optical and radionuclide imaging. Although both optical and radionuclide imaging use electromagnetic radiation in the for m of photons for signal transport, the energy between the two is vastly different. This is a mixed blessing in the sense that although optical photons are easil y absorbed in v ery small distances in low density, low atomic number materials such as silicon (Si), the y are at the same time hea vily attenuated and scattered b y tissues. Therefore, it is onl y necessary to place a small amount of Si as in a typical charge-coupled de vice (CCD) camera pix el to e xactly determine the location of photon interaction. Advanced optical f ilters w hen used in the photon path can also selectively block or diffract different wavelengths, adding color information in the image, at some cost in sensiti vity. When standard optics are also placed in front of the CCD camera, the y ef fectively collimate the signal and thereby provide the direction of travel in a scattering free media such as air (Figure 2A). This information provides the point of origin of these photons from the last scattering position just under neath the tissue surf ace, in the form of an image of the surf ace distribution of photons. In contrast, photons emitted by radionuclides have a very low probability of interaction in most low atomic number Z materials typicall y in volved in tissues. To ef fectively stop these photons, specialized high-density and high atomic number scintillator materials are used such as LaBr, NaI, BGO, and LSO.42 Alternatively, a few systems are using solid-state semiconductor detector materials such as CdTe and TlBr.43 The scintillators act as energy modulators. They con vert the high-ener gy photons to

multiple low-energy photons in the visib le range, essentially the same photons that are used in bioluminescence and fluorescence. These photons can in tur n be detected with similar technolo gies as optical photons since the y have the same ener gy range. There is a k ey dif ference though, that a single high-energy photon from a radionuclide emission is now converted into a g roup of multiple photons numbering in the fe w hundreds to a fe w thousands, emitted typicall y within a fe w tens to a fe w hundreds of nanoseconds. 44 The number of these photons is directly related to the ener gy of the originating radionuclide emission. Therefore, PET and SPECT systems need to accumulate and measure the total amount of optical photons in this scintillation pulse within a small time window, to estimate the ener gy and ar rival time of the radionuclide emission and use this infor mation to estimate the coincidence time and ener gy for PET , and energy discrimination for SPECT . The ener gy infor mation for both PET and SPECT can be used to help in the rejection of scattered high-ener gy photon e vents in both the patient and in the detectors. Because these high-ener gy γ-ray photons are so penetrating, no conventional lens systems similar to those available for optical photons e xist, although attempts to create such systems are ongoing. 45 The principle of operation of these γ-ray optical lenses is based on small angle scattering without ener gy loss, and the best cur rent systems typically work at lower energies (< 40 keV) and with small scattering angles under typicall y low efficiencies.46 Therefore, for SPECT imaging, the direction of photon travel needs to be distinguished b y selective photon rejection accomplished through a hea vy absorber , called a collimator (Figure 2B). This collimator is typicall y made out of lead, or other high atomic number and density material, including e xotic options such as depleted uranium. 47 This device works effectively as a lens in optical systems

Scintillation detector

CCD

Collimator

PET/Optical detector

A A

B B

C

Figure 2. A, Traditional lens-based imaging system; B, Traditional single photon radionuclide detector module; and C, Combined positron emission tomography (PET)/optical detector.18

Optical Multimodality Technologies

and also typically rejects a very large fraction of the overall photons (10 −4 for a parallel hole collimator). In contrast, one of the k ey technical dif ferences betw een PET and SPECT is that this mechanical collimation is not necessary for PET. The direction of photon travel is determined electronically by nanosecond coincidence detection.48 This key difference pro vides a signif icant intrinsic adv antage to PET in ter ms of o verall system sensiti vity per unit of injected radionuclide activity.

SIMULTANEOUS OPTICAL AND RADIONUCLIDE IMAGING With these principles of image for mation in mind , the following methodologies of combining the tw o modalities become ob vious: (1) use tw o completel y separate imaging instr uments, inte grated perhaps in a single gantry and (2) use the same detector technolo gy and same physical detector, with the addition of an ener gy modulator at the front end to enab le detection of both photon signals. Although the f irst approach has merits in its simplicity and ease of implementation, it generates two separate data sets that require spatial co-registration and can possib ly have a higher cost. This methodology has been the choice of a number of research g roups and will be discussed belo w. The second approach is more technically challenging and less straightforw ard to implement, but it is also potentially lower in overall cost and should pro vide inherentl y spatiall y co-re gistered images. It has been the method of choice of a smaller set of research groups and will also be discussed belo w.

COMBINING BIOLUMINESCENCE WITH RADIONUCLIDES THROUGH SEPARATE INSTRUMENTS The first interest in the combination of radionuclide technologies with optical imaging was reported by Huber and colleagues.49 That initial system combined a single photon radionuclide detector with a parallel hole collimator for the detection and imaging of Tc-99m, with a highsensitivity CCD camera and an optical lens for the detection of bioluminescence. The radionuclide detector w as based on a cesium iodide (CsI) scintillator , coupled to a photodiode ar ray with specialized high-perfor mance electronics. The single nonrotating head radionuclide imaging system allowed for a significant amount of clear space that w ould accommodate the optical imaging system, and the tw o systems could operate simultaneousl y

143

and independentl y. A similar approach w as tak en b y another g roup,16 who combined a planar single photon detector with a fluorescence imaging system. The initial single photon radionuclide detector was based on Si technology that had rather lo w detection ef ficiency resulting in a lo w sensiti vity system. The detector w as later replaced by CdTe that works very well for single photon emitters up to about 140 k eV. The optical e xcitation source of this instr ument w as based on raster scanning illumination of the subject with a g reen pulsed laser through a scanning mir ror, and detection through a standard CCD camera and lens as reported in the first in vivo results.50 The fluorophore they used was intramuscularly injected hematoporphyrin that accumulated in the tumor xenografts. These early, projection-only systems were followed b y a commercial v endor system that inte grated bioluminescence, fluorescence, and X-ra y projection imaging in a single device, which later acquired the capability to image noncollimated radionuclide emissions with low spatial resolution. 51 Several other groups developed inte grations of tomo graphic systems; one of the earliest ones developed a tomographic SPECT and multiprojection optical detection system, 52 which used a pinhole collimator placed in the front of an ar ray of NaI scintillators, read out by large area flat panel photomultiplier tubes. The optical imaging system w as based on a lens coupled, cooled CCD camera, and a set of mir rors placed in the front of the pinhole collimator , providing the same view as the radionuclide image. A laser source was a vailable to pro vide raster scanning of the subject with excitation light for fluorescence imaging. In a later development, the same group developed the concept of a more advanced system that focused on PET and provided the additional capability of bioluminescence and fluorescence imaging 53 (Figure 3A). For the optical data detection, instead of a single lens and CCD , the g roup proposed the use of multiple lar ge area Complementar y Metal Oxide Semiconductor (CMOS) detectors and instead of con ventional lenses for visib le light collimation, it used a specially designed micro-lens array and an optical septum positioned in front of each of the Si-based detectors. The low atomic number of the CMOS detectors, together with the lo w mass of the micro-lens ar ray, allowed positioning of the optical detectors inside the PET gantry, with minimal attenuation of the PET photons based on simulations. Laser excitation was made possible with thin optical f ibers positioned betw een the scintillator detectors. The complete system concept w as demonstrated b y a combination of simulations and measurements on a clinical PET tomo graph.

144

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

AA

B B

Figure 3. A, Combined positron emission tomography (PET) and optical bioluminescence imaging (BLI) and fluorescence imaging system, with separate detectors53; B, Optical-PET (OPET), combined PET and bioluminescence imaging system using the same detectors.18

COMBINING BIOLUMINESCENCE WITH RADIONUCLIDES THROUGH ONE INSTRUMENT The more challenging approach of using the similarity in the nature of optical and radionuclide signals w as taken by another g roup18 that combined a PET and a BLI instrument, using the same detectors. The scintillators were used for the dual role of ener gy modulation and light guide, providing improved light collection from the subject (F igure 2C) w hile photomultiplier tubes w ere used as photodetectors. It should be noted here that the overall light collection of lens-based optical imaging systems is typically very low, significantly smaller than 1%, even for large aperture optics.54 In effect, optical imaging systems e xchange fle xibility and ease of operation for absolute detection sensitivity. Direct coupling of the light detectors to the source with optical f ibers and optical fiber bundles would allow a signif icant increase in their light collection efficiency, equivalent to one or two orders of magnitude. 55 In this approach for the combination of optical and PET detection, the optical light collection efficiency w as also considerab ly increased in a similar fashion as in f iber-optic coupled systems, at the e xpense of f ield-of-view flexibility and system throughput capability. Despite the improved geometry for light collection efficiency b y that approach, the use of photomultiplier tubes with their inherently low quantum efficiency counteracted somewhat this increase in sensitivity. With a single detector , the system could detect both optical and radionuclide emissions, and specialized electronics enabled separation betw een the visib le light photons emanating from bioluminescence, and the pulses of light emanating from the interaction of radionuclide emissions

in the scintillators.56 At a later development, other groups followed this approach and created similar detectors with additional features such as multiple la yers that pro vide depth of interaction capability for the PET signal, and dichroic mir rors that impro ved the collection ef ficiency of the scintillation light. 57

OPTICAL SOURCE RECONSTRUCTION As se veral g roups combined their ef forts to de velop these combined instr uments, the other aspect of the technology that in volves mathematical modeling of photon tracks, interactions, and signal reconstr uction demonstrated that to resolve the issue of 3D optical photon quantification, in the general sense and in heterogeneous li ving mouse tissues, a priori infor mation on optical properties of tissues in situ is required , together with anatomic infor mation of the spatial distrib ution of these tissues. 23,58–60 When in addition to that infor mation, extra knowledge from the w avelength of the emitted and detected photons is used , successful reconstructions for the location and magnitude of a source inside a mouse w as achieved.7,23 This result was demonstrated for both bioluminescence and fluorescence and uses the information from differential photon absorption and scattering as a function of w avelength. On the basis of these results, a number of g roups decided to fuse anatomic imaging modalities lik e X-ray CT and MRI, with their v olumetric high-spatial resolution, to gether with or gan se gmentation softw are and atlas and lookup tab le based results of tissue optical properties toward image reconstruction.20–22 Although these ef forts ha ve produced good results in simulation e xperiments, despite the signif icant

Optical Multimodality Technologies

mathematical challenges the y had to o vercome in this ill-posed prob lem,12 the f act is that e ven the best estimates of optical properties from the literature4,7 are based on measurements from e xcised or frozen tissue samples or from dif ferent species alto gether. Fur thermore, biologic motions and pock ets of air in the lungs, the intestines, and clear fluid in the stomach that mo ve during a study create prob lems in the modif ied dif fusion approximation or other models used for most solutions. Consequently, these atlas-based results are suboptimal w hen used in vi vo and methods that measure the optical proper ties in vi vo, in situ are being de veloped. Those include illumination with a series of optical f iber sources, measurements from multiple vie ws, and other simpler methods with a local f iber next to the e xit measurement point, w hich pro vide a good estimate of the local proper ties at the time of the e xperiment.24 These produce a better estimate of the source intensity with a more acceptable error tolerance.

FUTURE DIRECTIONS A signif icant wealth of kno wledge has resulted from the flurry of the scientific efforts described here. Technologies from optical multimodality probes to optical multimodality instr uments ha ve been de veloped and as this f ield evolves, a few new technological requirements are slo wly establishing themselves. One important and clear aspect is that a convergence of fields seems to be taking place, with the fields of fluorescence tomography (Chapter 11, “Fluorescence Tomography”), DOT (Chapter 14, “Diffuse Optical Tomography and Spectroscop y”), and multimodality BLI/fluorescence imaging coming closer to gether, to produce more quantitati ve and more accurate reconstr ucted results. That is certainly not a surprising or unexpected outcome, since all these modalities rel y heavily on accurate modeling of visib le light propagation of dif ferent w avelengths in tissues. Therefore, w hen the goal is to obtain quantitative infor mation about a ne w molecular imaging probe, instruments that can perform both optical transmission and optical emission measurements in a similar f ashion as a PET/SPECT and X-ra y CT system are a natural choice. This advancement will cer tainly benef it all three optical imaging modalities, bioluminescence, fluorescence, and DO T. The inclusion of additional imaging modalities such as PET and SPECT with their X-ray CT counter part will be a natural continuation that will enable the development and optimization of both optical and radionuclide probes. It will also allo w the ans wer of biological questions in a multiparameter space, on both the biodistribution of a probe and its tar get acti vity, and possibly its interactions in situ in vivo.

145

CONCLUSIONS In this very exciting f ield, many bright minds are producing outstanding scientif ic results and are solving dif ficult problems. It still remains to be seen though w hich, if any, of the methodolo gies will pro ve to be clearl y superior to others in terms of ease of use, accuracy of results, and cost. As we all strive for the quest of the ultimate multimodality instrument, we need to be reminded of the reason why this quest be gan, that is the enrichment of the infor mation obtained from preclinical molecular imaging studies and the translation of this infor mation into the clinical realm. Optical bioluminescence is attracti ve due to its simplicity , low cost, high throughput, lo w backg round, and ease of use. A methodology that would enable and simplify optical signal transmission measurements, in a similar f ashion as transmission scans did for PET and SPECT ,61 would greatly improve the accuracy of optical image reconstr uctions and the quantif ication of the signal. The integration of such a de vice with other imaging modalities such as PET and SPECT will be very useful indeed.

REFERENCES 1. Christopher H, Contag BDR. It’s not just about anatomy: in vivo bioluminescence imaging as an eyepiece into biology. J Magn Reson Imaging 2002;16:378–87. 2. Negrin RS, Contag CH. In vi vo imaging using bioluminescence: a tool for probing g raft-versus-host disease. Nat Re v Immunol 2006;6:484–90. 3. Cherry SR. Multimodality in vi vo imaging systems: twice the po wer or double the trouble? Annu Rev Biomed Eng 2006;8:35–62. 4. Cheong WF, Prahl SA, Welch AJ. A review of the optical-proper ties of biological tissues. IEEE J Quantum Electron 1990;26:2166–85. 5. Srinivasan S, Pogue BW, Jiang S, et al. Inter preting hemoglobin and water concentration, o xygen saturation, and scattering measured in vivo b y near-infrared breast tomo graphy. Proc Natl Acad Sci 2003;100:12349–54. 6. Culver J, Akers W, Achilefu S. Multimodality molecular imaging with combined optical and SPECT/PET modalities. J Nucl Med 2008;49:169–72. 7. Alexandrakis G, Rannou FR, Chatziioannou AF. Ef fect of optical property estimation accurac y on tomo graphic bioluminescence imaging: simulation of a combined optical-PET (OPET) system. Phys Med Biol 2006;51:2045–53. 8. Rice BW, Cable MD, Nelson MB . In vi vo imaging of light-emitting probes. J Biomed Opt 2001;6:432–40. 9. Dehghani H, Da vis SC, Jiang S, et al. Spectrall y resolv ed bioluminescence optical tomography. Opt Lett 2006;31:365–7. 10. Hebert T, Leah y R. A generalized EM algorithm for 3-D Ba yesian reconstruction from Poisson data using Gibbs priors. IEEE Trans Med Imaging 1989;8:194–202. 11. Arridge SR, Schw eiger M, Hiraoka M, Delp y DT. A f inite element approach for modeling photon transpor t in tissue. Med Ph ys 1993;20(2 Pt 1):299–309. 12. Boas DA, Brooks DH, Miller EL, et al. Imaging the body with diffuse optical tomography. Signal Processing Mag IEEE 2001;18:57–75. 13. Contag CH, Spilman SD, Contag PR, et al. Visualizing gene expression in li ving mammals using a bioluminescent repor ter. Photochem Photobiol 1997;66:523–31.

146

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

14. Herschman HR. Molecular imaging: looking at prob lems, seeing solutions. Science 2003;302:605–8. 15. Weissleder R, Ntziachristos V. Scaling down imaging: molecular mapping of cancer in mice. Nat Med 2003;9:123–8. 16. Celentano L, Laccetti P, Liuzzi R, et al. Preliminar y tests of a prototype system for optical and radionuclide imaging in small animals. Nucl Sci IEEE Trans 2003;50(5 Part 2):1693–701. 17. Peter J, Schulz RB, Semmler W. PET-MOT—a novel concept for simultaneous positron and optical tomography in small animals. In: 2005 IEEE, Nuclear Science Symposium Conference Record, 2005. 18. Prout DL, Silv erman R W, Chatziioannou A. Detector concept for OPET—A combined PET and optical Imaging system. IEEE Trans Nucl Sci 2004;51:752–6. 19. Tsyganov EN, Antich PP, Kulkarni PV, et al. Micro-SPECT combined with 3D optical imaging. In: 2004 IEEE, Nuclear Science Symposium Conference Record, 2004. 20. Allard M, Côté D, Davidson L, et al. Combined magnetic resonance and bioluminescence imaging of li ve mice. J Biomed Opt 2007; 12:1083–3668. 21. Joshi A, Rasmussen JC, Kw on S, et al. Multi-modality CT -PET-NIR fluorescence tomography. In: Biomedical Imaging: From Nano to Macro, 2008. ISBI 2008. 5th IEEE Inter national Symposium on 2008. 22. Yujie L, Jie T, Wenxiang C, Ge W. Experimental study on bioluminescence tomo graphy with multimodality fusion. J Biomed Imaging 2007;2007:9–9. 23. Chaudhari AJ, Dar vas F, Bading JR, et al. Hyperspectral and multispectral bioluminescence optical tomo graphy for small animal imaging. Phys Med Biol 2005;50:5421–41. 24. Comsa DC, Farrell TJ, Patterson MS. Quantitative fluorescence imaging of point-lik e sources in small animals. Ph ys Med Biol 2008; 53:5797–814. 25. Gulsen G, Xiong B, Birgul O, Nalcioglu O. Design and implementation of a multifrequency near-infrared diffuse optical tomography system. J Biomed Opt 2006;11:014020. 26. Phelps ME. PET: the merging of biology and imaging into molecular imaging. J Nucl Med 2000;41:661–81. 27. Sharma V, Luker GD, Piwnica-Worms D. Molecular imaging of gene expression and protein function in vi vo with PET and SPECT . J Magn Reson Imaging 2002;16:336–51. 28. Gleich B, Weizenecker J. Tomographic imaging using the nonlinear response of magnetic particles. Nature 2005;435:1214–7. 29. Blasberg RG. In vi vo molecular -genetic imaging: multi-modality nuclear and optical combinations. Nucl Med Biol 2003;30:879–88. 30. Cai W, Chen K, Li ZB , et al. Dual-function probe for PET and near-infrared fluorescence imaging of tumor v asculature. J Nucl Med 2007;48:1862–70. 31. Massoud TF, Gambhir SS. Molecular imaging in living subjects: seeing fundamental biolo gical processes in a ne w light. Genes De v 2003;17:545–80. 32. Ray P, De A, Min JJ, et al. Imaging tri-fusion multimodality repor ter gene expression in living subjects. AACR 2004;302:605–8. 33. Ponomarev V, Doubrovin M, Serganova I, et al. A novel triple-modality reporter gene for whole-body fluorescent, bioluminescent, and nuclear nonin vasive imaging. Eur J Nucl Med Mol Imaging 2004;31:740–51. 34. Kesarwala AH, Prior JL, Sun J, et al. Second-generation triple repor ter for bioluminescence, micro-positron emission tomography, and fluorescence imaging. Mol Imaging 2006;5:465–74. 35. Gambhir SS, Herschman HR, Cher ry SR, et al. Imaging transgene expression with radionuclide imaging technolo gies. Neoplasia 2000;2:118–38. 36. Hielscher AH. Optical tomo graphic imaging of small animals. Cur r Opin Biotechnol 2005;16:79–88. 37. Kuo C, Coquoz O , Troy TL, et al. Three-dimensional reconstruction of in vivo bioluminescent sources based on multispectral imaging. J Biomed Opt 2007;12:024007.

38. Ntziachristos V, Ripoll J, Wang LV, Weissleder R. Looking and listening to light: the e volution of w hole-body photonic imaging. Nat Biotechnol 2005;23:313–20. 39. Zavattini G, Vecchi S, Mitchell G, et al. A hyperspectral fluorescence system for 3D in vi vo optical imaging. Ph ys Med Biol 2006;51:2029–43. 40. Phelps ME. PET : molecular imaging and its biolo gical applications. New York: Springer; 2004. 41. Kim SJ, Doudet DJ, Studenov AR, et al. Quantitati ve micro positron emission tomography (PET) imaging for the in vivo determination of pancreatic islet graft survival. Nat Med 2006;12:1423–8. 42. van Eijk CWE. Inor ganic scintillators in medical imaging detectors. Nucl Inst Methods Phys Res A 2003;509:17–25. 43. Madsen MT . Recent adv ances in SPECT imaging. J Nucl Med 2007;48:661–73. 44. Knoll GF. Radiation detection and measurement. 3rd ed. Ne w York: Wiley and Sons; 2000. 45. Snigirev A, Kohn V, Snigireva I, Lengeler B . A compound refractive lens for focusing high-energy X-rays. Nature 1996;384:49–51. 46. MacDonald CA, Gibson WM. Applications and advances in polycapillary optics. X-Ray Spectrum 2003;32:258–68. 47. Meikle SR, K ench P, Kassiou M, Banati RB . Small animal SPECT and its place in the matrix of molecular imaging technolo gies. Phys Med Biol 2005;50:R45–61. 48. Phelps ME. Application of annihilation coincidence detection to transaxial reconstruction tomography. J Nucl Med 1975;16:210–24. 49. Huber JS, Sudar D, Moses WW. Conceptual design of a dual modality optical and radionuclide imaging camera. In: High Resolution Imaging In Small Animals. Rockville, MD: 2001. 50. Autiero M, Celentano L, Cozzolino R, et al. Experimental study on in vivo optical and radionuclide imaging in small animals. Nucl Sci IEEE Trans 2005;52:205–9. 51. Feke GD, Lee vy WM, Or ton S, et al. Har nessing multimodality to enhance quantif ication and reproducibility of in vi vo molecular imaging data. Nat Met 2008;5. 52. Peter J, Ruehle H, Stamm V, et al. De velopment and initial results of a dual-modality SPECT/optical small animal imager . IEEE Nuclear Science Symposium Conference Record; 2005. 53. Peter J, Unholtz D, Schulz RB, et al. De velopment and initial results of a tomo graphic dual-modality positron/optical small animal imager. Nucl Sci IEEE Trans 2007;54:1553–60. 54. Liu H, Karellas A, Harris LJ, D’Orsi CJ. Methods to calculate the lens efficiency in optically coupled CCD X-ray imaging systems. Med Phys 1994;21:1193–95. 55. Liu H, Karellas A, Har ris L, D’Orsi C. Optical proper ties of f iber tapers and their impact on the performance of a fiber optically coupled CCD X-ray imaging system. SPIE; 1993. 56. Prout DL, Silv erman RW, Chatziioannou A. Readout of the optical PET (OPET) detector. Nucl Sci IEEE Trans 2005;52:28–32. 57. Takahashi K, Inadama N , Murayama H, et al. Preliminar y study of a DOI-PET detector with optical imaging capability. IEEE Nuclear Science Symposium Conference Record; 2007. 58. Alexandrakis G, Rannou FR, Chatziioannou AF. Tomographic bioluminescence imaging b y use of a combined optical-PET (OPET) system: a computer simulation feasibility study . Phys Med Biol 2005;50:4225–41. 59. Guven M, Yazici B, Intes X, Chance B. Diffuse optical tomography with a priori anatomical infor mation. Ph ys Med Biol 2005; 50:2837–58. 60. Li A, Boverman G, Zhang Y, et al. Optimal linear in verse solution with multiple priors in dif fuse optical tomo graphy. Appl Opt 2005;44:1948–56. 61. Kinahan PE, Townsend DW, Beyer T, Sashin D. Attenuation correction for a combined 3D PET/CT scanner. Med Phys 1998;25:2046–53.

10 FIBER OPTIC FLUORESCENCE IMAGING RABI UPADHYAY, BS AND UMAR MAHMOOD, MD, PHD

The clinical endoscope has ser ved as an imaging tool since the se venteenth centur y for e xaminations of the canals and cavities of the human body. Even when compared to cross-sectional imaging modalities toda y, modern fiber-optic catheters intrinsically provide high spatial resolution images of anatomic aber rations ranging from subtle mucosal patterns to gross luminal narrowings. The images are a vailable in real time during acquisition, and the technology requires minimal device and maintenance costs. Hence, f iber-optic catheters pro vide a lo w bar rier of entry for clinical translation of fluorescence molecular imaging and much potential for rapid testing and de velopment of new probes and disease models. This chapter will begin with a brief overview of current endoscopy implementations and disease applications to describe the standard of care and the unmet clinical needs that moti vate better imaging methods. In our discussion of preclinical f iber-optic technolo gies, w e will begin with spectroscopic methods for molecular sensing and se gue to techniques that create tr ue spatial images. The imaging will be g rouped b y methods that use endogenous fluorophores, basic exogenous fluorophores, and sophisticated tar geted and acti vatable molecular probes. We will conclude with a brief discussion of the necessary instrumentation and current design constraints for fiber-optic fluorescence imaging.

CLINICAL STATE OF THE ART Direct visualization of colorectal cancer and precancerous adenomas by minimally invasive colonoscopy has been used over decades, is a mainstay of early detection toda y through screening pro grams,1 and per mits immediate treatments such as pol ypectomy w here applicable. Other widespread endoscopic applications include inspection of the upper gastrointestinal tract for Barrett’s esophagus, laparoscopy of the peritoneal cavity,

bronchoscopy of the respirator y tract, c ytoscopy of the urinary tract, and mediastinoscopy of the thorax. In addition, inter ventional cardiolo gists and radiolo gists use percutaneous coronar y angioscop y with f iber-optic catheters to deter mine plaque mor phology, guide stent placement, and repair aneur ysms. The high spatial resolution of f iber optics has enabled the earl y detection of v ery small anatomic changes and the differentiation of adjacent neoplastic and nonneoplastic lesions on or gan surf aces or within lumens. In complement, the high temporal resolution has enabled immediate inter vention such as the biopsy of neoplastic lesions leaving behind surrounding healthy tissue in situ. It is impor tant to note these primar y advantages as the introduction of molecular imaging into the clinic should onl y enhance traditional endoscop y and should not diminish cur rent functions. The resolution of most clinical endoscopes is a function of the number of individual imaging fibers that can be fit within the fiber bundle of the catheter. Typical numbers range from less than ten thousand f ibers within a 0.8 mm angioscope to several hundred thousand fibers within a 3 cm colonoscope. This number is fur ther limited w hen a w orking channel for biopsy forceps or fluid deli very consumes par t of the cross-sectional area a vailable for imaging and illumination f ibers. Recentl y, microchip cameras able to f it onto the tips of catheters ha ve largely supplanted traditional f iber-optic-based endoscopes thereby fur ther impro ving image resolution and clarity . These cameras ha ve also been engineered to f it entirely within a capsule. 2 Such systems are ingested b y the patient, image snapshots of the entire gastrointestinal tract during their passage, and are par ticularly helpful in the detection of small bowel pathology. However, during colonic e valuation with capsule endoscopy, the stochastic time and location of each image acquisition has been sho wn to miss a number of 147

148

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

prominent lesions. 3 In f act, e ven f iber-optic-based adenoma detection has ultimately been shown to be largely operator and time dependent,4 given the subtlety of detecting earl y lesions across a lar ge organ surf ace. In studies using back-to-back colonoscopies, more than 20% of colonic pol yps w ere missed. 5 Such anatomic imaging is e ven more lik ely to miss flat lesions, w hich may ha ve higher rates of dysplasia compared with polyps.6 Direct visualization of adenomas and colorectal cancer b y endoscopic methods remains the clinical standard, but engineering impro vements of the white light (WL) imaging paradigm, by themselves, are unlikely to resolve the missed lesion rate. This problem is further exacerbated, for instance, in ulcerati ve colitis in which dysplasia can develop in macroscopically normal appearing mucosa. Cur rent colonoscopic sur veillance in patients with ulcerative colitis relies on random biopsies throughout the colon, which is relatively insensitive and cumbersome. 7 In the case of intraperitoneal spread of cancer , intraoperati ve detection of small metastatic foci may be limited by the similar luminosity of tumors compared with adjacent nor mal tissue. These clinical needs ma y be addressed in par t b y using no vel approaches that combine ne w f iber-optic devices and fluorescence detection. As we co ver each molecular imaging technology, we can inquire into its usefulness compared with standard endoscopy. Is it practical to cover large organ surface areas with the technique or is the sampling as stochastic as random biopsies? Does the acquisition proceed in real time allowing for simultaneous inter vention? How deep into the mucosal surf ace can be imaged? What is the sensitivity and specificity for the disease application?

SCATTERING SPECTROSCOPY Some of the methods f irst developed to glean molecular information with fiber optics include spectroscopic measurements of photon scattering in epithelial tissue. Because the f iber optics that detect photon scattering events ha ve matured, the y ha ve been slo wly introduced into the working channels of endoscopes for human studies in the detection of v arious epithelial pathologies. Two types of photon scattering b y a molecule e xist: elastic and inelastic (or Raman) scattering. 8 Both Raman scattering and fluorescence alter the optical w avelength but through dif ferent mechanisms. In both cases, the excited molecule relaxes to an energy level of the ground state and emits a photon. In Raman scattering, a Stok es transition occurs when an interacting photon is less energetic after interaction and an Anti-Stokes transition

occurs w hen the interacting photon is more ener getic after interaction with the molecule. Inelastic scattering more typically shows specific chemical composition and molecular str ucture of tissue, w hereas elastic scattering more typically shows the size distribution of the scatterers. Both techniques are amenab le to in vi vo measurements as the photon flux and excitation wavelengths used are nondestructive to the tissue. 9 In practice, elastic light scattering spectroscop y (LSS) has been used to deter mine the size distribution of cell nuclei.10 The diameter of nondysplastic cell nuclei is typically 5 to 10 µm, w hereas dysplastic nuclei are often larger, up to 20 µm across. Epithelial cell nuclei can be modeled as transparent spheroids w hose refractive index is higher than that of the sur rounding c ytoplasm. The backscattered light characteristically varies depending on nuclear size and refracti ve inde x. F or a collection of nuclei of different sizes, the light-scattering signal is a superposition of these v ariations, enab ling the nuclear size distribution and refractive index to be determined from the spectrum of light backscattered from the nuclei. Once the nuclear size distribution and refractive index are known, quantitati ve measures of nuclear enlar gement, crowding, and hyperchromasia can be obtained. These are the same criteria used by pathologists to diagnose tissue biopsies for mucosal dysplasia. The potential of LSS has been tested in multiple patient studies 11,12 to diagnose dysplasia and carcinoma in situ in different human organs with different types of epithelium: columnar epithelia of the colon and esophagus, transitional epithelium of the urinar y bladder, and squamous epithelium of the oral ca vity. The technique delivers a weak pulse of WL through a f iber-optic bundle threaded through a standard endoscope. After pulsing a 1-mm2 tissue surface on the order of milliseconds at wavelengths of 350 to 650 nm, the fiber bundle collects the diffusely reflected light. The spectra consist of a lar ge background from submucosal tissue, on which is superimposed a small (2 to 3%) component secondary to scattering by cell nuclei in the mucosal la yer. Although the spectral analysis can be cumbersome, pre vious datasets can help create decision algorithms for immediate histologic classification based on nuclear enlargement and density. Beyond a certain threshold nuclear diameter and population, a sampled tissue location may be designated as dysplasia or carcinoma. Hence, LSS has shown a strong potential to detect epithelial pre-cancerous lesions in an objective manner. In the same vein as LSS, Raman (inelastic scattering) spectroscopy also relies on pre-def ined spectroscopic models of v arious patholo gies. It interrogates the vibrations of molecular bonds and pro vides a direct

Fiber Optic Fluorescence Imaging

method to quantify the chemical composition of biological tissue. The modeling approach is based on the assumption that the Raman spectr um of a mixture is a combination of the spectra of its components and that signal intensity and chemical concentration are linearl y related. The resulting fit coefficients yield the contribution of each basis spectrum to the macroscopic tissue spectr um thereby elucidating the chemical mor phological mak eup of the lesion. Spectroscopic models 13 usually f it macroscopic tissue spectra with a linear combination of basis spectra from Raman microscopy of components such as epithelial cell cytoplasm, the cell nucleus, f at, β-carotene, collagen, calcium hydroxyapatite, calcium oxalate dihydrate, cholesterol-like lipid deposits, and water. Tissue composition extracted through modeling is used as the basis of a diagnostic algorithm capable of differentiating between a normal and diseased state. Although initial Raman studies e xamining inter nal body tissues required long collection times in the range of 5 to 30 s, recent advances in catheter design have resulted in fle xible Raman f iber-optic catheters capab le of collecting spectra with large signal-to-noise in 1 second or less. This has, for e xample, enab led the technolo gy’s application to in vivo spectral pathology of human atherosclerosis and vulnerable plaque.14 The optical fiber may be advanced through a catheter during carotid endarterectomy and femoral bypass surgeries to obtain Raman spectra of endothelial tissue. 15 Representative f it coefficients of the major components from a pre viously de veloped morphological model described b y Buschman and colleagues16,17 (including collagen, elastin, cholesterol cr ystals, necrotic core, calcifications, adventitial fat, smooth muscle cells, and β-carotene cr ystals) can therefore be generated in situ. The model has previously shown accurate tissue characterization (as confirmed by histology of the sur gical biopsies) and achie ved a sensiti vity and specificity of 79 and 85%, respecti vely. Similar devices have been used in diagnosing breast cancers b y dif ferentiating benign and malignant lesions based on chemical composition.18,19 In this application, the fit coefficients for fat and collagen are the k ey parameters in the diagnostic algorithm, which classifies tissue samples according to their specif ic patholo gical diagnoses. In patient studies, the spectroscopic technique attained 94% sensitivity and 96% specif icity for distinguishing cancer from normal and benign tissues. The technique has particularly shown its ef fectiveness for i n vivo margin assessment during partial mastectomy breast surgery.19 Both spectroscop y techniques, LSS and Raman, clearly of fer an enhanced ability o ver traditional endoscopy to glean molecular infor mation from tissue

149

and to detect very subtle pathologies in their early stages. Although not always, the creation of quasi real-time data processing algorithms can enab le in situ measurements through the working channel of an endoscope. Ho wever, these methods are still inherentl y point measurements and do not generate tr ue spatial images. The problem of grossly undersampling the v ast surface area of epithelial tissues remains unsolved, and the questions remain as to what advantage spectroscopy provides over the histology of surgical biopsies? What is needed in the clinical realm, therefore, is a spatial imaging technology that approaches the sensiti vity/specificity of spectroscop y and that can quickly survey entire organ surfaces. Fluorescence generated from either endo genous or e xogenous fluorophores in epithelial tissue of fers exactly these benef its and will be discussed in the remainder of this chapter .

IMAGING ENDOGENOUS AUTOFLUORESCENCE As described above, the phenomena of fluorescence is a (Stokes) shift in emitted w avelength gi ven a par ticular incident wavelength that excites an electron in a molecule to a stationary state. A variety of endogenous substances in biological tissue show this property at different excitation wavelengths collectively giving tissue, which is commonly known as autofluorescence. Although almost all tissue components e xhibit autofluorescence at some le vel, the signal obser ved during fluorescence endoscop y is predominantl y generated b y molecules in the mucosa such as collagen, elastin, tr yptophan, nicotinamide adenine dinucleotide, fla vin adenine dinucleotide, and por phrins.20 Such molecules ma y differentially accumulate in areas of dysplasia, leading to autofluorescent characteristics that ma y help distinguish between nor mal and neoplastic tissue. By placing standard band-pass f ilter w heels within the e xcitation and emission light paths, imaging can easil y switch between normal WL endoscopy and autofluorescence endoscop y, which highlights the signal from these endo genous fluorophores.21–23 The wavelength ranges for autofluorescence typicall y f all into the lo wer re gion of the visib le light spectrum emphasizing blue and green fluorophores. However, this is completel y dependent on the fluorophores of interest, and sometimes, the range may reach higher than 1000 nm. There have been numerous studies of standard autofluorescence endoscopy coupled with WL endoscopy, but the main applications in humans ha ve been for the detection of Bar rett’s esophagus 24 and colon neoplasia. 25 In addition, the paradigm has been extended to diseases in

150

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

the breast, lung, and other or gans. The majority of these studies ha ve sho wn that autofluorescence-guided endoscopy improves the diagnostic yield for neoplasia in comparison with the conventional approach using WL only and four-quadrant biopsies 20 (Figure 1A, B). Ho wever, it has also been e xtensively sho wn that autofluorescence alone is not suitab le for replacing the standard four -quadrant biopsy protocol 24 and that autofluorescence detection is associated with a f alse positive rate as high as 51%. As the techniques for autofluorescence imaging have been ref ined, a specif ic type of autofluorescence imaging dubbed narrow band imaging (NBI) has shown promise in the detection of Bar rett’s esophagus. 26 NBI enhances the visualization of superficial mucosal structures b y nar rowing the band-pass ranges of the g reen and b lue components of the e xcitation light. This causes the relative intensity of the b lue emission spectrum to increase and impro ves the visualization of mucosal blood vessels (since the blue light excitation is highly absorbed by hemoglobin).27 NBI emphasizes features such as capillar y and cr ypt patter ns, and this technique has potential for diagnosing gastrointestinal diseases at an early stage. The results of NBI in human studies are similar to that of other autofluorescence techniques. Although NBI has some what enhanced conventional endoscopic detection of disease, the added benef it is incremental at best, and fur ther trials are required to deter mine the true advantage compared with conventional endoscopy. One of the factors hindering the extraction of quantitative biochemical information from measured tissue autofluorescence is the presence of potentiall y signif icant distortions introduced b y tissue scattering and absor ption. Although a number of methods have been proposed for the recovery of the intrinsic (undistor ted) tissue fluorescence, the y are not easil y implemented in a clinical setting or they have limited applicability in the 400- to 500-nm spectrum because of high hemoglobin absorption levels. One potential solution has been to combine information in simultaneously measured tissue autofluorescence and diffuse reflectance.28 Such a technique can e xtract intrinsic (undistor ted) tissue autofluorescence and isolate and quantify the spectral contributions of N AD(P)H and collagen. This can pro ve useful because the relati ve contribution of these tw o fluorophores to the intrinsic tissue autofluorescence ma y be modif ied during the de velopment of pre-cancerous lesions in tissues such as Bar ratt’s esophagus and the uterine cer vix.29 Thus, the y may ser ve in some cases as in vi vo biomarkers of pre-malignant change, without the need for tissue remo val.

A

B B

C C

Figure 1. Clinical example of autofluorescence endoscopy. A 64-year-old man with atrophic gastritis underwent endoscopic treatment for early gastric cancer on the posterior wall of the upper gastric body seen as a whitish 1-cm elevated lesion in the white light channel (A). Autofluorescence imaging in the 490 to 625 nm light band (B) reveals the tumor as purple and additionally detects a flat tumor extension on the distal side that is not clear in the WL image. Chromoendoscopy with 0.04% indigo carmine solution (C) confirms the extent of the tumor. Reproduced with permission from Uedo N et al.23

Fiber Optic Fluorescence Imaging

FLUORESCENCE LIFETIME TECHNIQUES Many of the fluorescence endoscopy techniques, including the a utofluorescence a pproaches d escribed a bove, a re steady state methods that dif ferentiate tissue based upon differential emission spectral prof iles. The autofluorescence techniques discussed above often have a reasonable sensitivity for the detection of earl y cancers but a lo w specificity and a high f alse positive rate. 30 An alter native approach is fluorescence lifetime imaging (FLIM). 31 The method is based on the measurement of the temporal decay in fluorescence intensity following excitation. Because the fluorescence lifetime is derived from relative intensity values, FLIM can provide useful information concerning fluorochrome localization in spite of differences in scattering and variation in fluorophore concentration. Wide-field microscop y and FLIM ha ve already been used in studies of tissue constituents, 32 cell cultures,33 and the human skin.34 FLIM can be perfor med in the frequenc y domain for w hich a high-frequency modulated laser beam e xcites the sample and the fluorescence lifetime is deter mined from the demodulation and phase shift of the fluorescence signal. FLIM can also be performed in the time domain for which the fluorescence decay is directl y measured after pulsed laser excitation. FLIM is not only sensitive to the type of fluorophores but also may depend on its environment. This functionality has been e xploited35 to quantify ph ysiological parameters including pH, [Ca 2+], and pO 2. Differences in the fluorescence lifetimes betw een normal and neoplastic tissue ha ve been sho wn in the colon,36 breast,37 and brain. 38 However, the instr umentation required for these studies has been generall y costly, bulky, and very sensitive to small changes in tissue f ixation and optical calibration. The difficulties mostly arise from the very sophisticated lasers and detectors required to e xcite and sense photons with nanosecond temporal resolution, w hich ha ve not been practical options for a clinical instrument. Until now, there have been only a few reports of high-speed wide-f ield FLIM, and these ha ve usuall y been restricted to a reduced number of pix els.39 The imaging paradigm has just recently been applied to fiberoptic catheters such that FLIM de vices may be threaded through standard endoscopes and potentiall y image tissue in vivo with high frame rates and reasonab le spatial resolutions.40 The design relies on a standard gated optical image intensif ier with a rapidl y s witchable dela y generator, and much pre vious w ork has been done on implementing an anal ytic rapid lifetime deter mination algorithm. Although the research has so f ar onl y been applied to the autofluorescence of healthy tissues (human

151

stomach and kidne y), FLIM endoscop y has much potential in the future to generate fluorescence lifetime “maps” through a catheter at a video rate so that the technology can become more clinically relevant.

IMAGING UNTARGETED EXOGENOUS FLUOROPHORES Administration of exogenous fluorescent dyes with a high quantum yield can achie ve a much stronger e xtrinsic fluorescence contrast in epithelial tissue compared with most autofluorescence techniques. Such compounds ma y be delivered intravenously before examination or intravitally through the working channel of the endoscope during an imaging session. Chromoendoscopy is the practice of spra ying a fluorescent dye onto the mucosa using a spra y catheter passed through a standard endoscope (see F igure 1C). Although various dyes have been tried , methylene blue is the most common agent, and it is primaril y used in the detection of intraepithelial neoplasia and colon cancer. It is an inexpensive, absor ptive stain that, in contrast to other substances such as indigo carmine, is taken up by the intestinal epithelium after its local application, resulting in a relatively stable staining pattern and the visualization of the opening of the glandular pits during chromoendoscopy. It has been shown in large human studies6 that chromoendoscopy impro ves earl y diagnosis of adenomas and earl y colorectal cancers. It allo ws prediction of the nature of mucosal lesions in the colorectum b y using the so-called pit patter n classif ication41 for mucosal staining patter ns to dif ferentiate betw een neoplastic and nonneoplastic changes. A randomized , controlled trial42 has shown that chromoendoscopy permits more accurate diagnosis of the extent and severity of the inflammator y activity in ulcerati ve colitis compared with con ventional colonoscopy. Limited success has also been repor ted43 using the technique for the detection of dysplasia in Bar rett’s esophagus. Despite all these positive preliminary results, the technique has yet to be rigorously proven as an advantage during endoscopy. Because meth ylene b lue is inherentl y a reducing agent, some studies ha ve sho wn that it can induce oxidative damage of DNA when photosensitized by WL during chromoendoscop y and therefore can accelerate carcinogenesis.44 Such risks need to be carefully balanced against the possib le benef its of improved early disease detection. One e xogenous fluorophore that has already been heavily tested for adv erse reactions 45 is indoc yanine green (ICG). It has been used for more than 30 years as

152

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

a water-soluble dye with a peak absor ption at ~800 nm that rapidly binds to blood proteins (primarily albumin) after intravenous injection. Its pre vious clinical applications ha ve been for deter mining cardiac output, hepatic function, and ophthalmic angiography.46 Given its low reactivity, intravenous ICG has been e xploited for fluorescent endoscop y of v ascular str uctures. Its near infrared absorption and emission has enabled ICG to w ell delineate esophageal v arices, where structural changes in the vascular wall are caused by portal hypertension, and this method has recentl y been extended to the detection of vascular lesions in the digestive tract.47 In an observational study of 30 patients with gastric tumors,48 fluorescence was positive in 8 of 10 cancers with submucosal in vasion and in 1 of 20 adenomas or intramucosal cancers. It is believed that the retention of ICG is cor related with the size of the submucosal v ascular bed, and near -infrared (NIR) endoscop y of ICG can enhance the visibility of deeper v essels within a gastric tumor. ICG has also been used for the detection of metastases to a sentinel l ymph node (the f irst lymph node that receives drainage from a cancer). Intraoperati vely injecting dy e at the site of a melanoma to identify sentinel nodes is a well-characterized technique,49 and it has been recently been paired with NIR laparoscopic imaging of ICG in gastric cancer50 and in lung cancer patients. 51 The findings so f ar suppor t the ef ficiency of sentinel node navigation using ICG for detecting clinicall y nodenegative cancers. Several other agents function similarl y to ICG as fluorescent markers of vasculature during perfusion. Fluorescein (and its deri vative fluorescein isothiocyanate, [FITC]) has long been used in the realms of microscopy and angiographic ophthalmology. However, it is important to note that its peak e xcitation (494 nm) and emission (521 nm) wavelengths are well below that of ICG, and therefore signal from fluorescein is significantly absorbed in vivo by surrounding tissue. In addition, both fluorescein and ICG are relati vely small molecules that quickl y leak through the v asculature into interstitial space. Given the parameters of optical w avelength and molecular size, agents de veloped to image and quantitate tumor v asculature ha ve been slo wly optimized.52 One simple approach has been to attach organic dyes to long circulating de xtranated nanoparticles.53 Other methods use high molecular w eight (around 250 kDa) pe gylated g raft copol ymers with indocyanine-type fluorophores optimized for nonquenching54 (Figure 2A, B). Because these imaging

agents are nonimmuno genic (due to pe gylation and isotonia), longitudinal endoscopic imaging studies are also possible. A

B

C

D

E

F

G

H

II

J

Figure 2. Preclinical mouse endoscopies depicting white light (WL) (left column) and near-infrared (NIR) (right column) imaging channels. In vivo images from peritoneum vasculature (A,B) illustrate the ability to quantify vascular leak of a fluorescent blood pool agent by comparing intravascular with adjacent extravascular signal. Colonoscopy shows the detection of orthotopic tumor implantations (C,D) with a modified NIR fluorescence agent containing a cyclic RPMC peptide motif that was derived from a library screen. Significantly lower signal is observed in a control experiment (E,F) with a scrambled peptide sequence. Quantitative real-time colonoscopy shows the detection of two colon tumors barely discernable in the WL channel with a protease-activatable NIR probe (G,H). Fluorescent signal from the lesions is seen to remain constant (I,J) as the tip of the catheter is advanced closer to the tissue. Adapted from Sheth RA et al.84; Kelly K et al.61; and Upadhyay R et al.80

Fiber Optic Fluorescence Imaging

IMAGING TARGETED/ACTIVATABLE MOLECULAR PROBES As f iber-optic fluorescence de vices are maturing, a diverse array of sophisticated optical probes is being synthesized to complement the technology. These probes are exogenously delivered and generate v ery large signal-tonoise ratios by either targeting or interacting with biological processes on a molecular le vel. Most are completing preclinical validation in mouse models, and a fe w have entered human imaging studies along with con ventional endoscopes modified to detect fluorescence. One such probe originated as an agent for photodynamic therapy. Specifically, the photosensitizer pre-cursor 5-aminolevulinic acid (5-ALA) is intracellularly converted to protoporphyrin IX (PpIX) a few hours after intravenous administration. Con ventional photodynamic therap y proceeds with laser excitation of the photosensitizer, generates a singlet state o xygen molecule, destr uctively reacts with any nearby biomolecules, and it may result in apoptosis or necrosis.55 However, it has been disco vered that at lo w doses, PpIX can also ser ve as an imaging agent because it fluoresces red under b lue illumination. 56 Because 5-ALA elicits synthesis and preferential accumulation of PpIX in epithelial neoplastic tissue, it has strong potential as a targeted/activatable imaging agent in addition to its cur rent clinical role as a therapeutic. Low doses of 5-ALA are already appro ved for use in patients, which has enab led human studies to assess the performance of fluorescent endoscopy of PpIX, for example, for detecting intraepithelial neoplasia and earl y cancers in Bar rett’s esophagus. 57,58 Briefly, the fluorescence detection was seen to achieve a similar performance compared with four-quadrant random biopsy (the current gold standard), but it resulted in signif icantly fewer biopsies. This has motivated the development of next generation photodynamic agents that minimize extraneous phototoxicity and improve delivery and efficacy.59,60 Several other tar geted probes in the preclinical pipeline that are specif ically designed for endoscopic (and not necessarily therapeutic) applications ha ve been demonstrated in mouse models and imaged with prototype endoscopic systems. The probe designs in volve attaching a fluorochrome to a targeting moiety such as a peptide sequence. One probe w as engineered to be specific to a colon cancer b y deri ving an af finity ligand from a phage librar y screen. The result w as a c yclic peptide motif that could be fluorescentl y labeled 61 and imaged with fluorescence endoscop y. Such librar yderived imaging agents can be easil y v alidated in vi vo and compared with control molecules to sho w

153

cancer-specific tar geting (F igure 2C– F). This endoscopic imaging paradigm can be e xtended to an y other targeted optical probe developed for intraoperative applications such as fluorescentl y labeled antibodies, 62 small molecules,63 and nanoparticles.64 Rather than targeting surface receptors and proteins, targeted probes can also be incor porated into biolo gical vectors. This has been demonstrated, for instance, with a herpes simple x viral v ector with cancer -selective infection and replication including a transgene for green fluorescent protein. 65 Fluorescence-aided minimally invasive endoscopy showed microscopic tumor deposits unreco gnized b y con ventional laparoscop y/thoracoscopy. The imaging vector model was also in vitro confirmed in 110 types of cancer cells from 16 different primar y or gans, and the vector was shown to infect tumors and metastases in both immunocompetent and immunodef icient mice. Despite such preclinical results, similar fluorescent proteins encoded in biolo gical vectors are unlik ely to translate clinicall y due to concer ns re garding long-ter m immunogenicity. Finally, there has been e xtensive work in the de velopment of a ne w class of optical imaging agents that change their fluorescent proper ties after tar get interaction.66 These smar t probes are initiall y opticall y silent, secondary to fluorochrome–fluorochrome quenching (ie, when the 2 fluorochromes are in close pro ximity on a backbone molecule, the y absorb the other lights) but become brightly fluorescent in areas of disease. One specific target is the increased protease expression present in neoplastic tissue, w hich mediates enzymatic clea vage of fluorochromes from a delivery backbone, resulting in signal amplif ication of up to several hundredfold. This imaging paradigm has pro ven particularly successful with endoscopic applications because high signal-tonoise imaging of protease o verexpression allo ws rapid screening of a lar ge surf ace area (F igure 2G, H). The probe has been e xploited in colonoscopies, 67–70 thoracoscopies,71 and laparoscopies, 72 and it is no w poised to enter initial clinical trials. In addition, the same model for an activatable probe has been translated to other types of enzyme o verexpression for application in cardio vascular73 and autoimmune diseases, 74 where f iber-optic fluorescence imaging is also feasible.

PROBE DESIGN CONSTRAINTS AND QUANTIFICATION Given the previous discussion of untargeted, targeted, and activatable e xogenous molecular probes, w e can be gin

154

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

to outline general constraints and considerations in the design of optical molecular imaging agents for fiber-optic detection. Optimizing the strength of the fluorescence and the ability to quantify the signal detected by the catheter are par ticularly impor tant to in vi vo applications. Of primar y impor tance to the signal strength is the optical wavelength of the tar get.75 Most fluorescent proteins, FITC conjugations, rhodamine conjugations, and methylene blue emit in the visible-light range where surrounding tissue markedly absorbs the signal. The excitation light is similarl y attenuated. The attenuation is conversely decreased relative to shorter wavelengths for NIR probes such as conjugations to c yanine dyes, ICG, and other synthetic NIR fluorochromes. Furthermore, the autofluorescence of endogenous fluorophores in the surrounding tissue tends to be the strongest in the visiblelight range. This illustrates another adv antage in the signal-to-noise ratio of NIR probes, and it helps e xplain the limited success of pre vious autofluorescence endoscopy studies in humans. 24 Complementary to signal strength, the schedule and endurance of the peak signal are especiall y impor tant to serial f iber-optic imaging of the same tissue location. Enzyme activatable probes generally reach their peak fluorescence 24 to 48 hours after injection, and the signal may endure several more days. Most blood pool imaging agents are designed to remain in the vasculature a specific amount of time, primarily depending on the size of the molecule. 52 The use of multiple molecular imaging agents necessitates optically distinct wavelengths that are each discer nable by the fiber-optic instrumentation. For enzyme activatable probes, the absolute value of fluorescence obtained is a function of the intensity of the incident light from the endoscope and the depth and size of the lesion producing the fluorescence. In addition, the amount of acti vated product produced b y the protease reflects not onl y the enzyme acti vity but also the delivery of the imaging probe substrate. This complicates the quantif ication of the enzyme of interest because both play a role in the f inal signal intensity. To address this issue, dual fluorochrome probes ha ve been synthesized that separate the tw o processes. 76 The design is based upon standard nanopar ticles with clea vable peptide spacers attached through a C-ter minal c ysteine and a fluorochrome attached to the N ter minus. In addition to the activatable/quenched fluorochrome, a second fluorochrome is attached directl y to the macromolecule car rier and is resistant to proteol ytic acti vation. Endoscopic i maging o f t his p article a t t wo d istinct

wavelengths of fers both pieces of infor mation. Signal strength of the acti vatable fluorochrome represents probe delivery and activation, whereas signal strength of the reporter fluorochrome only correlates to delivery. Hence, the ratio of the tw o signals can be used to nor malize for differences in the size and depth of a target lesion and differences in probe delivery. Although the above probe optimizations are sufficient to quantify signal with f ixed geometry intraoperative fluorescence imaging systems,77 catheter-based systems face an e xtra complication. F iber-optic catheters do not ha ve the advantage of a static-controlled imaging en vironment in which distances from the illumination source to the target tissue and from the target tissue to the charge-coupled device (CCD) are fixed, and the illumination intensity across the tar get remains unifor m over time. These distances become dynamic, essentiall y uncontrollab le v ariables with catheter-based systems, and the fluorescence emission of a par ticular location is no longer constant. Photon fluence decreases as the square of the distance between the tar get tissue and the catheter tip increases, resulting in a marked change in NIR photon counts as one approaches or retreats from the disease being investigated. Moreover, sharp angles of incidence between the catheter and the NIR signal source cause objects closer to the catheter to appear brighter than more distant objects within the same video frame. The fluorescence microscop y community has resolved similar concerns regarding the quantitative ability of their instr uments by imaging a unifor mly fluorescent reference sample to estab lish a baseline sample image and then dividing all subsequent images by the reference sample image. 78,79 A comparab le algorithm can also be applied to real-time catheter -based systems. 80 Rather than dividing by a constant reference image, however, each NIR frame can undergo pixel-wise division by a simultaneously acquired WL image. This allows for a dynamic frame-b y-frame nor malization that accounts for v ariations in signal intensity betw een and within individual frames due to changes in catheter position (Figure 2G–J and Figure 3).

INSTRUMENTATION DESIGN CONSTRAINTS The addition of fluorescent molecular imaging to current clinical endoscope instr umentation poses some necessary design constraints. One fundamental prerequisite is the need for a dichroic mir ror (beam splitter) within the

Fiber Optic Fluorescence Imaging

155

B

A Concentration 1 µM 5 µM 10 µM 20 µM

Distance to Surface

Raw NIR Counts/msec

30 25 20 15 10 5

Constant ROI

0

C

1

2

3 4 Distance (mm)

5

D

0.3

40 % Change

Corrected NIR Signal

60

0.2

0.1

20

Corrected Data

0 20 40 Raw NIR Data

60 0

1

2

3 4 Distance (mm)

5

80

1

2

3 4 Distance (mm)

5

Figure 3. Distance dependence of raw near-infrared (NIR) pixel values versus white light (WL) division corrected NIR values. A, Schematic of the experimental design shows the various concentrations of fluorescent dye used in the phantoms. ROI = region of interest. B, Distance dependence curves for raw NIR counts per ms for the concentrations show a large percentage change over the distance of a few mm. Error bars denote the standard of error. C, Distance dependence curves for the corrected NIR pixels for the same concentrations show an approximately flat relationship between corrected signal intensity and distance. The difference in corrected NIR signal intensity between all four concentrations is significant (p < .001), as determined with analysis of covariance. D, Percentage change from the mean values for raw and corrected NIR. The reduced absolute slope of the corrected data (compared with the slope of the raw data) shows the reduced dependence of the signal intensity on distance. Reproduced with permission from Upadhyay R et al.80

endoscopic light path to enab le simultaneous WL and NIR imaging 72 (Figure 4A). This allows the operator to visualize NIR molecular signals and co-re gister them with a natomic l andmarks ( as s een i n c onventional endoscopy). Currently, dichroic mir rors can onl y be placed after fiber-optic catheters in the optical train, and they are not compatib le with cer tain commercial endoscopes containing CCD chips on the tip of the catheter . This constraint produces a slight loss in spatial resolution (depending on the substituted f iber bundle), but it also allows for a modular design of two or more cameras that independently acquire WL and NIR images. Dif ferent excitation light bands (such as a x enon lamp and a NIR laser) are also usually combined with a dichroic before en tering t he f iber-optic i llumination b undle. Interference band-pass f ilters can be appropriatel y placed throughout the optical path to avoid cross-talk between imaging channels. 81

When multiple distinct w avelength probes or fluorochromes ha ve to be imaged , tw o instr umentation designs are possib le. One ar rangement69 places an initial dichroic to transmit WL to a CCD and reflect all NIR signal to a second dichroic and a full y reflective mir ror that separates the light into tw o independent images (lower and higher emission ranges) onto different portions of a single CCD chip. Although this method can image two molecular tar gets, it reduces the NIR spatial resolution, complicates image processing, and does not scale well be yond tw o NIR w avelengths. A more direct approach simply adds independent cameras for each NIR channel desired. Thus, e xposure times can be independently controlled , and digital images can be perfectl y aligned b y standard se gmentation82 and re gistration83 algorithms. Such re gistration is par ticularly impor tant to the image di vision algorithms (discussed abo ve) that can enable quantification of NIR signal.

156

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

Dichroic Mirror Emission Filter

NIR CCD

12 Mb/sec

Fiberoptic Catheter Input Relay Lenses Visible Light CCD

Real-time Processing

12 Mb/sec

Initialize both cameras’ hardware

B

Generate ROI via segmentation routine Align cameras via registration routine Generate histogram for ROI

Camera 1

Camera 2

Set camera exposure time

Set camera exposure time

Start camera

Start camera

Wait for a single image to be transfered to RAM

Wait for a single image to be transfered to RAM

Display 8-bit image

Display 8-bit image

Calculate new exposure time

Calculate new exposure time

Save 12-bit data

Save 12-bit data

Stop camera

Stop camera

Calculate pixel value corresponding to 95% of histogram area Compare pixel value to setpoint and adjust exposure time accordingly

Real-time Visualization

Data Storage: Raw Images and Parameters

Figure 4. Schematic of fluorescent endoscopy instrumentation optics and software pipeline. Emission light from a fiber-optic catheter is optically separated by a dichroic mirror into WL and NIR light beams (A). These are focused onto CCD cameras with additional NIR filtering through a band-pass filter. The 12-bit data streams reach the computer via 100-Mbit connections. Incoming data are managed completely in random access memory to achieve real-time latency (B). The top arrow diverges into two parallel threads executed simultaneously but independently to allow for different frame rates for the two cameras. Left and right arrows loop for each camera image. Expanded bubble shows histogram-based calculations. ROI = region of interest. Reproduced with permission from Upadhyay R et al.80

Fiber Optic Fluorescence Imaging

In k eeping wi th s everal o ther c linical m odalities (including computed tomo graphy, magnetic resonance, and ultrasound), the CCD cameras in f iber-optic instrumentation that image the various wavelengths require a lar ge dynamic range (generally 12 bit or higher). The detection of very weak NIR signals is generall y a limiting factor in such imaging. The WL image stream in con ventional endoscop y is usuall y ne ver count star ved, and the frames can be acquired at video rate (roughl y 30 frames per second). As NIR CCDs are engineered to be more sensitive, minimum e xposure times to acquire practical signal-to-noise ratios can be set e ven f aster, and the NIR imaging frequency can also slo wly approach a tr ue video rate. Regardless, independent control of each CCD enables individual e xposure times to be set based on the cur rent signal strength in each imaging channel. Autoexposure algorithms in fixed geometry modalities, such as microscopes, can be based on either the brightest pix el v alue or the mid-range of an entire histogram of pixel values from a previous image. The latter is especially desirab le in f iber-optic applications because strong illumination from a point source onto a wet epithelial tissue often produces direct (specular) reflection that artificially saturates a fe w pixels. For example, selecting the pixel value that occurs at 95% of the histo gram area avoids artifacts from specular reflection but is still indicative of the o verall image brightness. The autoe xposure algorithm can g radually bring this pix el value to a user defined set point such that there is sufficient overall signal without any substantial pixel saturation. However, if the cameras operate independently and shutter variable exposure times, the raw pixel values of different images are no longer comparab le betw een cameras or successi ve images. One solution is to normalize each image’s pixel values by its cor responding exposure time because there is a linear relationship between the tw o variables. This ratio of counts/second can be used as a constant, e xposure time-independent data value. The new value is also required by an image division algorithm in w hich pixels within the WL and NIR images need to ha ve the same units before an y mathematical operation can be perfor med. In the case of typical 12-bit image acquisition, the raw pixel values can range from 0 to 4096 (2 12). However, the new counts/second v alue has additional fle xibility in exposure time: data can take any value from 0 to 4096 di vided b y the range of possib le e xposure times. For e xample, if there are 2 20 possible e xposure times (typical for modern cameras), this results in 2 32 possible values, correlating to an effective 32-bit dynamic range. Thus, the counts/second v alue can also dramaticall y

157

increase the dynamic range and sensiti vity of the digital imaging. After initial images ha ve been acquired in each channel, future exposure times have been determined, and the di vision algorithm has been perfor med, the operator needs to be ab le to visualize the cur rent frames before acquisition can proceed. Because all the frames are acquired with a high bit depth and video displays are limited to an 8-bit dynamic range, all images must under go proper g radation (windo wing and le veling) before the y can be displayed. Typically, NIR frames are rescaled using a pseudo-coloring algorithm that colorizes the images based on each pixel’s percent change from a user -defined variable baseline. The WL image must under go a demosaicing algorithm to inter polate a complete color image from the 12-bit ra w data recei ved from the color -filtered image sensor. Ultimately, both routines can produce color WL (as seen in con ventional imaging) and pseudo-color NIR images for immediate displa y, and the ra w data can be saved with the original dynamic range. Given the demands of acquisition, autoe xposure, window leveling, and other image processing algorithms that function in real time to maintain a tr ue video rate, modern fiber-optic fluorescence imaging systems require significant computing power. Software frameworks have been developed84 that rely on multithreaded code that can distribute the computational load across multiple processors and computers to achieve real-time signal processing (Figure 4B). The softw are design remains impor tant because the ability to process more data must scale as additional NIR channels (and CCDs) are introduced to simultaneously visualize multiple optical probes.

REFERENCES 1. Wolff WI. Colonoscop y: histor y and de velopment. Am J Gastroenterol 1989;84:1017–25. 2. Lux G, Kn yrim K, Scheubel R, Classen M. Electronic endoscop y— fibres or chips? Z Gastroenterol 1986;24:337–43. 3. Schoofs N , De viere J , Van Gossum A. PillCam colon capsule endoscopy compared with colonoscopy for colorectal tumor diagnosis: a prospective pilot study. Endoscopy 2006;38:971–77. 4. Barclay RL, Vicari JJ, Doughty AS, et al. Colonoscopic withdra wal times and adenoma detection during screening colonoscop y. N Engl J Med 2006;355:2533–41. 5. van Rijn JC, Reitsma JB , Stoker J, et al. P olyp miss rate deter mined by tandem colonoscopy: a systematic review. Am J Gastroenterol 2006;101:343–50. 6. Rembacken BJ, Fujii T, Cair ns A, et al. Flat and depressed colonic neoplasms: a prospective study of 1000 colonoscopies in the UK. Lancet 2000;355:1211–14. 7. Bernstein CN, Weinstein WM, Levine DS, Shanahan F. Physicians’ perceptions of dysplasia and approaches to sur veillance colonoscopy in ulcerati ve colitis. Am J Gastroenterol 1995; 90:2106–14.

158

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

8. Wang L V, Wu H-I. Biomedical optics: principles and imaging. Hoboken (NJ): Wiley-Interscience; 2007. 9. Hanlon EB, Manoharan R, Koo TW, et al. Prospects for in vivo Raman spectroscopy. Phys Med Biol 2000;45:R1–59. 10. Wallace MB, Perelman LT, Backman V, et al. Endoscopic detection of dysplasia in patients with Bar rett’s esophagus using lightscattering spectroscopy. Gastroenterology 2000;119:677–82. 11. Backman V, Wallace MB , P erelman L T, et al. Detection of pre-invasive cancer cells. Nature 2000;406:35–6. 12. Georgakoudi I, Jacobson BC, Van Dam J , et al. Fluorescence, reflectance, and light-scattering spectroscop y for e valuating dysplasia in patients with Bar rett’s esophagus. Gastroenterolo gy 2001;120:1620–29. 13. Motz JT, Gandhi SJ, Scepanovic OR, et al. Real-time Raman system for in vivo disease diagnosis. J Biomed Opt 2005;10:031113. 14. Romer TJ, Brennan JF 3rd , Fitzmaurice M, et al. Histopatholo gy of human coronary atherosclerosis by quantifying its chemical composition with Raman spectroscopy. Circulation 1998;97:878–85. 15. Motz JT , F itzmaurice M, Miller A, et al. in vi vo Raman spectral pathology of human atherosclerosis and vulnerab le plaque. J Biomed Opt 2006;11:021003. 16. Buschman HP , Deinum G, Motz JT , et al. Raman microspectroscopy o f h uman c oronary a therosclerosis: b iochemical assessment of cellular and extracellular morphologic structures in situ. Cardiovasc Pathol 2001;10:69–82. 17. Buschman HP, Motz JT, Deinum G, et al. Diagnosis of human coronary atherosclerosis b y mor phology-based Raman spectroscop y. Cardiovasc Pathol 2001;10:59–68. 18. Haka AS, Shafer-Peltier KE, Fitzmaurice M, et al. Diagnosing breast cancer b y using Raman spectroscop y. Proc Natl Acad Sci USA 2005;102:12371–76. 19. Haka AS, Volynskaya Z, Gardecki J A, et al. in vi vo margin assessment during partial mastectomy breast surgery using raman spectroscopy. Cancer Res 2006;66:3317–22. 20. Borovicka J, Fischer J, Neuweiler J, et al. Autofluorescence endoscopy in sur veillance of Bar rett’s esophagus: a multicenter randomized trial on diagnostic efficacy. Endoscopy 2006;38:867–72. 21. Haringsma J, Tytgat GN. The value of fluorescence techniques in gastrointestinal endoscop y: better than the endoscopist’ s e ye? I: the European experience. Endoscopy 1998;30:416–18. 22. Uedo N, Iishi H, Tatsuta M, et al. A novel videoendoscopy system by using autofluorescence and reflectance imaging for diagnosis of esophagogastric cancers. Gastrointest Endosc 2005;62:521–28. 23. Uedo N, Iishi H, Ishihara R, et al. Novel autofluorescence videoendoscopy imaging system for diagnosis of cancers in the digestive tract. Digestive Endoscopy 2006;18:S131–6. 24. Curvers WL, Singh R, Song LM, et al. Endoscopic tri-modal imaging for detection of early neoplasia in Barrett’s oesophagus: a multicentre feasibility study using high-resolution endoscop y, autofluorescence imaging and nar row band imaging incor porated in one endoscopy system. Gut 2008;57:167–72. 25. Izuishi K, Tajiri H, Fujii T, et al. The histological basis of detection of adenoma and cancer in the colon b y autofluorescence endoscopic imaging. Endoscopy 1999;31:511–16. 26. Gono K, Obi T, Yamaguchi M, et al. Appearance of enhanced tissue features in narrow-band endoscopic imaging. J Biomed Opt 2004; 9:568–77. 27. Yoshida T, Inoue H, Usui S, et al. Nar row-band imaging system with magnifying endoscop y for superf icial esophageal lesions. Gastrointest Endosc 2004;59:288–95. 28. Georgakoudi I, Jacobson BC, Müller MG, et al. NAD(P)H and collagen as in vi vo quantitati ve fluorescent biomark ers of epithelial pre-cancerous changes. Cancer Res 2002;62:682–87. 29. Drezek R, Sokolov K, Utzinger U, et al. Understanding the contributions of N ADH and collagen to cer vical tissue fluorescence spectra: modeling, measurements, and implications. J Biomed Opt 2001;6:385–96.

30. Bard MP, Amelink A, Skurichina M, et al. Improving the specificity of fluorescence bronchoscopy for the analysis of neoplastic lesions of the bronchial tree b y combination with optical spectroscop y: preliminary communication. Lung cancer 2005;47:41–7. 31. Siegel J, Elson DS, Webb SE, et al. Studying biological tissue with fluorescence lifetime imaging: microscopy, endoscopy, and complex decay profiles. Appl Opt 2003;42:2995–3004. 32. Dowling K, Dayel MJ, Lever MJ, et al. Fluorescence lifetime imaging with picosecond resolution for biomedical applications. Opt Lett 1998;23:810–12. 33. Bastiaens PI, Squire A. Fluorescence lifetime imaging microscop y: spatial resolution of biochemical processes in the cell. Trends Cell Biol 1999;9:48–52. 34. Hartmann P, Mirtolouei R, Untersberger S, et al. Non-invasive imaging of tissue PO2 in malignant melanoma of the skin. Melanoma Res 2006;16:479–86. 35. Lakowicz JR. Principles of fluorescence spectroscop y. Ne w York: Springer; 2006. 36. Mycek MA, Schomacker KT, Nishioka NS. Colonic polyp differentiation using time-resolv ed autofluorescence spectroscop y. Gastrointest Endosc 1998;48:390–94. 37. Tadrous PJ, Siegel J, French PM, et al. Fluorescence lifetime imaging of unstained tissues: early results in human breast cancer. J Pathol 2003;199:309–17. 38. Marcu L, Jo JA, Butte PV, et al. Fluorescence lifetime spectroscopy of glioblastoma multiforme. Photochem Photobiol 2004;80:98–103. 39. Agronskaia AV, Tertoolen L, Gerritsen HC. Fast fluorescence lifetime imaging of calcium in living cells. J Biomed Opt 2004;9:1230–7. 40. Munro I, McGinty J, Galletly N, et al. Toward the clinical application of time-domain fluorescence lifetime imaging. J Biomed Opt 2005;10:051403. 41. Kudo S, Tamura S, Nakajima T, et al. Diagnosis of colorectal tumorous lesions b y magnifying endoscop y. Gastrointest Endosc 1996;44:8–14. 42. Kiesslich R, F ritsch J, Holtmann M, et al. Meth ylene blue-aided chromoendoscopy for the detection of intraepithelial neoplasia and colon cancer in ulcerative colitis. Gastroen terology 2003;124:880–8. 43. Sharma P, Weston AP, Topalovski M, et al. Magnification chromoendoscopy for the detection of intestinal metaplasia and dysplasia in Barrett’s oesophagus. Gut 2003;52:24–7. 44. Olliver JR, Wild CP, Sahay P, et al. Chromoendoscopy with methylene blue and associated DNA damage in Barrett’s oesophagus. Lancet 2003;362:373–4. 45. Hope-Ross M, Yannuzzi LA, Gragoudas ES, et al. Adverse reactions due to indocyanine green. Ophthalmology 1994;101:529–33. 46. Flower RW, Hochheimer BF. Indocyanine green dye fluorescence and infrared absor ption choroidal angio graphy perfor med simultaneously with fluorescein angio graphy. Johns Hopkins Med J 1976; 138:33–42. 47. Okamoto K, Mugur uma N , Kimura T, et al. A no vel diagnostic method for e valuation of v ascular lesions in the digesti ve tract using infrared fluorescence endoscopy. Endoscopy 2005;37:52–7. 48. Kimura T, Muguruma N, Ito S, et al. Infrared fluorescence endoscopy for the diagnosis of superf icial gastric tumors. Gastrointest Endosc 2007;66:37–43. 49. Morton DL, Wen DR, Wong JH, et al. Technical details of intraoperative l ymphatic mapping for earl y stage melanoma. Arch Sur g 1992;127:392–9. 50. Ishikawa K, Yasuda K, Shiromizu A, et al. Laparoscopic sentinel node navigation achieved by infrared ray electronic endoscopy system in patients with gastric cancer. Surg Endosc 2007;21:1131–34. 51. Ito N, Fukuta M, Tokushima T, et al. Sentinel node navigation surgery using indocyanine green in patients with lung cancer. Surg Today 2004;34:581–85. 52. Weissleder R, Bo gdanov A Jr , Tung CH, Weinmann HJ. Size optimization of synthetic g raft copolymers for in vi vo angiogenesis imaging. Bioconjug Chem 2001;12:213–9.

Fiber Optic Fluorescence Imaging

53. Montet X, Ntziachristos V, Grimm J , Weissleder R. Tomographic fluorescence mapping of tumor tar gets. Cancer Res 2005; 65:6330–6. 54. Montet X, Figueiredo JL, Alencar H, et al. Tomographic fluorescence imaging of tumor v ascular v olume in mice. Radiolo gy 2007; 242:751–8. 55. Peng Q, Ber g K, Moan J , et al. 5-Aminole vulinic acid-based photodynamic therap y: principles and e xperimental research. Photochem Photobiol 1997;65:235–51. 56. Stepp H, Sroka R, Baumgar tner R. Fluorescence endoscop y of gastrointestinal diseases: basic principles, techniques, and clinical experience. Endoscopy 1998;30:379–86. 57. Stepinac T, Felley C, Jornod P, et al. Endoscopic fluorescence detection of intraepithelial neoplasia in Bar rett’s esophagus after oral administration of aminolevulinic acid. Endoscopy 2003;35:663–68. 58. Endlicher E, Knuechel R, Hauser T, et al. Endoscopic fluorescence detection of low and high g rade dysplasia in Bar rett’s oesophagus using systemic or local 5-aminolae vulinic acid sensitisation. Gut 2001;48:314–19. 59. McCarthy JR, P erez JM, Br uckner C, Weissleder, R. P olymeric nanoparticle preparation that eradicates tumors. Nano Lett 2005;5:2552–6. 60. McCarthy JR, Weissleder R. Model systems for fluorescence and singlet o xygen quenching b y metallopor phyrins. ChemMedChem 2007;2:360–5. 61. Kelly K, Alencar H, Funo vics M, et al. Detection of in vasive colon cancer using a novel, targeted, library-derived fluorescent peptide. Cancer Res 2004;64:6247–51. 62. Gee MS, Upadhyay R, Bergquist H, et al. Multiparameter noninvasive assessment of treatment susceptibility , drug target inhibition and tumor response guides cancer treatment. Int J Cancer 2007; 121:2492–500. 63. Weissleder R, K elly K, Sun EY , et al. Cell-specif ic tar geting of nanoparticles by multivalent attachment of small molecules. Nat Biotechnol 2005;23:1418–23. 64. Sun EY, Josephson L, Weissleder R. “Clickable” nanoparticles for targeted imaging. Mol Imaging 2006;5:122–8. 65. Adusumilli PS, Stiles BM, Chan MK, et al. Real-time diagnostic imaging of tumors and metastases b y use of a replication-competent herpes vector to facilitate minimally invasive oncological surgery. Faseb J 2006;20:726–28. 66. Weissleder R, Tung CH, Mahmood U, Bogdanov A Jr, in vivo imaging of tumors with protease-acti vated near -infrared fluorescent probes. Nat Biotechnol 1999;17:375–8. 67. Alencar H, Funovics MA, Figueiredo J, et al. Colonic adenocarcinomas: near -infrared microcatheter imaging of smar t probes for early detection—study in mice. Radiology 2007;244:232–8.

159

68. Alencar H, King R, Funo vics M, et al. A no vel mouse model for segmental orthotopic colon cancer. Int J Cancer 2005;117:335–9. 69. Funovics MA, Alencar H, Montet X, et al. Simultaneous fluorescence imaging of protease e xpression and v ascularity during murine colonoscopy for colonic lesion characterization. Gastrointest Endosc 2006;64:589–97. 70. Funovics MA, Alencar H, Su HS, et al. Miniaturized multichannel near infrared endoscope for mouse imaging. Mol Imaging 2003; 2:350–7. 71. Figueiredo JL, Alencar H, Weissleder R, Mahmood U. Near infrared thoracoscopy of tumoral protease acti vity for improved detection of peripheral lung cancer. Int J Cancer 2006;118:2672–7. 72. Funovics MA, Weissleder R, Mahmood U . Catheter -based in vi vo imaging of enzyme activity and gene expression: feasibility study in mice. Radiology 2004;231:659–66. 73. Chen J, Tung CH, Allport JR, et al. Near-infrared fluorescent imaging of matrix metalloproteinase acti vity after m yocardial inf arction. Circulation 2005;111:1800–05. 74. Izmailova ES, P az N, Alencar H, et al. Use of molecular imaging to quantify response to IKK-2 inhibitor treatment in murine arthritis. Arthritis Rheum 2007;56:117–28. 75. Weissleder R, Ntziachristos V. Shedding light onto li ve molecular targets. Nat Med 2003;9:123–28. 76. Kircher MF, Weissleder R, Josephson L. A dual fluorochrome probe for imaging proteases. Bioconjug Chem 2004;15:242–48. 77. De Grand AM, F rangioni JV . An operational near -infrared fluorescence imaging system prototype for large animal surgery. Technol Cancer Res Treat 2003;2:553–62. 78. Model MA, Burkhardt JK. A standard for calibration and shading correction of a fluorescence microscope. Cytometry 2001;44:309–16. 79. Zwier JM, Van Rooij GJ , Hofstraat JW, Brak enhoff GJ. Image calibration in fluorescence microscopy. J Microsc 2004;216:15–24. 80. Upadhyay R, Sheth RA, Weissleder R, Mahmood U . Quantitati ve real-time catheter-based fluorescence molecular imaging in mice. Radiology 2007;245:523–31. 81. Mahmood U, Tung CH, Tang Y, Weissleder R. F easibility of in vi vo multichannel optical imaging of gene e xpression: e xperimental study in mice. Radiology 2002;224:446–51. 82. Yoo TS, Ackerman MJ, Lorensen WE, et al. Engineering and algorithm design for an image processing Api: a technical repor t on ITK— the Insight Toolkit. Stud Health Technol Inform 2002;85:586–92. 83. Woods RP, Grafton ST, Holmes CJ, et al. Automated image registration: I. General methods and intrasubject, intramodality v alidation. J Comput Assist Tomogr 1998;22:139–52. 84. Sheth RA, Upadhyay R, Weissleder R, Mahmood U. Real-time multichannel imaging frame work for endoscop y, catheters, and f ixed geometry intraoperative systems. Mol Imaging 2007;6:147–55.

11 FLUORESCENCE TOMOGRAPHY VASILIS NTZIACHRISTOS, MSC, PHD

From cell labeling to immunohistochemistr y and micro-array studies, fluorescence has been one of the most common sources of contrast in the biomedical laboratory. The development of fluorescence probes and of fluorescence proteins as reporter molecules for subcellular function has further solidified the versatility and wide application of fluorescence imaging in a v ariety of applications, for e xample, the study of protein function and interactions, gene transcription, and visualizing molecular pathways or cellular trafficking.1,2 In parallel, the associated photon sensing and imaging technolo gy has progressed over the y ears, to allo w visualization at many dif ferent scales, spanning from the nanometer scale to clinical imaging. 3,4 Fluorescence microscopy has been one of the common methods of fluorescence imaging of cellular mono-la yer assays or thin tissue sections, such as histolo gical slides. The de velopment of adv anced microscop y methods, for example, confocal and multiphoton microscopy, has further allowed imaging of thick er tissues. 5,6 These methods can effectively m inimize t he e ffect o f s cattering, o ffering images of high resolution compared with con ventional microscopy images. By sequentiall y imaging at dif ferent depths, three-dimensional images can be also generated. Depending on the tissue’s optical properties and the particular microscopic implementations, depths of up to 500 to 800 microns can be reached , especially when using multiphoton microscopy. These methods are reviewed in Chapter 13, “Intravital Microscopy.” To image deeper in tissue, it becomes essential to use tomographic techniques operating in meso and macroscopic scale. Tomography generally refers to the ability to obtain cross-sectional images from intact tissues and animal or human bodies, thus of fering a three-dimensional picture of fluorescence bio-distribution. In contrast to microscopic three-dimensional tissue-sectioning imaging, tomography and reconstruction imply the formulation of a 160

mathematical inverse problem, whose solution yields the tomographic images,7 in analogy to methods used in x-ray computed tomography (CT), single photon emission computed tomography (SPECT) or positron emission tomography (PET). Major technological approaches have been developed to achieve fluorescence tomography beyond the limits of intra vital microscopy, including the use of theoretical models of photon propagation in tissue.As a general rule, microscopy retains higher resolution and sensiti vity compared with tomographic methods developed for imaging deeper in tissue. Con versely, tomo graphy of fers the ability to offer noninvasive imaging of larger tissues allowing for in vivo imaging of insects, f ish, small animals, and even humans. The follo wing parag raphs detail the principles of in vivo fluorescence tomo graphy be yond the intra vital microscopy limit and showcase imaging applications.

TECHNOLOGY FOR FLUORESCENCE TOMOGRAPHY A common denominator of each tomography method is the ability to transmit a for m of ener gy to the sample of interest and detect changes in this transmitted energy due to its interaction with the sample from multiple depths or at dif ferent projections. The measurement of these changes are then digitized and used to gether with an inversion mathematic model to for m images. In the case of fluorescence, the transmitted energy is light that can be absorbed by the fluorochrome. Different substances can absorb light of different energy and emit fluorescence light at a lo wer energy. For imaging within only a few hundred microns to a few millimeters, light in the visib le can be used to e xcite common fluorescent proteins, such as the g reen fluorescent protein and its variants, and other spectrally shifted fluorescent proteins,

Fluorescence Tomography

such as the y ellow fluorescent protein, the red fluorescent protein, etc. In addition, it can be used to excite common or ganic fluorochromes, such as fluorescein, Texas Red, etc, or quantum dots. However, for penetration beyond a fe w millimeters, it becomes essential to use light in the near -infrared (NIR), that is, light of wavelength higher than 650 nm. This is because the light attenuation in tissue is signif icantly lo wer in the NIR, compared to the visible, due to the characteristic absorption spectr um of o xy- and deo xy-hemoglobin, w hich absorbs light mainly in the visible and not in the NIR. 8,9 Therefore, NIR light can penetrate much deeper in tissue compared with visib le light, up to se veral centimeters. One of the major challenges for tomo graphic imaging of tissues therefore is not the tissue absor ption but the tissue scattering. To achieve multiprojection illumination, the de velopment of appropriate imaging setups is required , which can direct a beam of light to the tissue surface. Figure 1 depicts a typical mode of imaging, that is, transillumination tomography w here the e xcitation light is projected from one side of the object and light that has propagated inside tissue is collected on the other side. This is a preferred mode of operation as it attains the most relaxed dynamic range requirements for the optoelectronic technology used while achieving vir tually symmetric v olume illumination coverage.

161

One of the impor tant aspects for fluorescence tomography is the use of accurate theoretical models that describe photon propagation in tissues. These models typically use some appro ximate solution to the transpor t equation.7 Tomographic de velopment generall y needs first to v alidate a suggested theoretical model against experimental measurements before using it in subsequent applications. The most common models used in optical tomographic imaging use solutions to the diffusion equation.10–13 Typically, these solutions w ork well for highl y scattering media of propagation dimension lar ger than 1 cm. F or smaller dimensions or moderatel y scattering media, more advanced models have been proposed. 14 The appearance of a typical optical tomo graphy problem takes the following form: N

φsc (r, rs) = ∑ W(r, rn, rs)O(rn), n=1

(1)

where W(r, rn, rs) represents a “w eight” that associates the effect of the optical proper ty O(rn) at position rn to a measurement at r owing to a source at rs. For a number of measurements, M, a system of linear equations is then obtained, resulting in a matrix equation: y = Wx,

(2)

where W is the weight matrix, x represents the distribution O(rn) of optical proper ty in each of the N voxels

Figure 1. Typical optical tomography arrangement in transillumination mode. Excitation light illuminates the object on one side, whereas light is collected on the opposite side of the object.

162

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

assumed, and y is the cor responding measurement vector. Inversion of this system of equations yields the unknown image x. Depending on the e xact for mulation of the problem, x can represent a two-dimensional or three-dimensional image. Ho wever, because photon propagation is fundamentally a volumetric phenomenon, x is typically a three-dimensional image. Computation times in optical tomography scale with the number of source-detector pairs used in the measurements. The use of CCD cameras readily offer millions of such pairs that need f irst to be processed (g rouped together) to reduce the size of v ector y. Even then inversion times may reach several minutes to hours, depending on the exact tomographic implementation, although typical commercial systems ha ve reconstruction times in the minute range.

MODES OF ILLUMINATION OPERATION There are three major modes of illumination that are typically considered in fluorescence tomo graphy15: continuous w ave (CW), time domain (TD), and frequenc y domain (FD) illumination. These categories relate to the technology used in generating light and result in different photon propagation characteristics. Constant wave methods use light of constant intensity and typically use CCD cameras for light detection. This approach requires low-cost implementations and offers high detection sensitivity and signal-to-noise ratio. They are well suited for fluorescent tomography although they cannot distinguish betw een absor ption and scattering, and are not prefer red for measurements of endo genous contrast (ie, absor ption or scattering). Compared with more adv anced illumination settings, CW methods have the limitation that they cannot minimize the effect of tissue scattering, that is, the resolution achieved is limited by the tissue optical proper ties. In addition, the y are not suitable for measuring fluorescence lifetime. Time domain methods use nar row photon pulses (typically < 10 ps). The detection in this case is achie ved with time-gated or time-resolv ed techniques, w hich record the ar rival of photons as a function of time, with time resolution on the order of picoseconds to hundreds of picoseconds. Time domain systems have the capability to independentl y disentangle the scattering and absor ption coefficients and to image fluorescence lifetime. Furthermore, b y using time-gating techniques, highl y scattered photons can be rejected , yielding images of improved resolution compared with CW systems. Compared with the CW method , TD systems yield lo wer

signal-to-noise ratio due to the low-duty cycles used due to pulsing, require more adv anced optical designs for optimal operation, and cost more. Finally, the frequenc y domain method uses light of modulated intensity and correspondingly uses demodulation methods to measure the amplitude and phase of the photon wave that is established into the object image. Modulation intensities span the 100 to 1,000 MHz range, and corresponding systems can operate at a single or multiple frequencies. Similar to the TD method, FD systems can measure lifetime, and by using high modulation frequencies, they can impro ve the image resolution over CW systems. A further advantage of the FD method is the ability to reject ambient, nonmodulated light. Practical implementation of this method is more challenging compared with the CW method due to the need of incor porating high frequenc y response optoelectronics and corresponding demodulation techniques. Again, these methods are more costly than CW.

TOMOGRAPHIC APPLICATIONS Fluorescence tomo graphy has been applied at dif ferent resolution scales, from the penetration limits of multiphoton microscopy (~500 microns) to human imaging. A particular class of fluorescence tomo graphy of tissues w as developed specif ically for molecular inter rogations of tissue in vi vo, ter med fluorescence molecular tomo graphy (FMT). Fluorescence molecular tomo graphy combines measurements at both emission and e xcitation wavelengths to quantify and to reconstruct fluorochromes of high molecular specif icity three-dimensionall y.15 The technique is used in conjunction with the systemic administration of fluorescence probes tar geting specif ic enzymes or other proteins. An associated technique, fluorescence protein tomo graphy (FPT) 16 is a tomo graphic method adapted to imaging fluorescence proteins by using appropriate spectral decomposition methods for reducing tissue autofluorescence ef fects w hen strong autofluorescence contributions are present. Figure 2 sho ws the application of FPT , operating in CW mode using an Ar+ laser to image mor phogenetic movements occur ring within the Drosophila pupae, in a study by Vinegoni and colleagues.17 In this case, the morphogenesis of the GFP-e xpressing wing discs and thorax was followed during the first few hours after preparation. The depicted time-lapse fluorescence imaging sequence shows the mor phogenesis of the wing imaginal discs and is in good ag reement with the cor responding histolo gic sections. In this case, a homemade system, similar to that

Fluorescence Tomography

A

B

C

D

E

F

163

Figure 2. Time-lapse imaging of Drosophila wing imaginal discs. The images are acquired from a single live specimen, at six different time points. Three different projections at 0, 50, and 180° with respect to the pupa’s dorsal view are shown. The reconstructions in the first column correspond to the sections indicated by the red lines in the third, fourth, and fifth columns. Comparison with DAPI-stained histological sections is shown in the second column. The histological images were acquired from morphologically matched areas of different staged pupae.

Figure 3. Imaging of lung inflammation. A, D, Photographs of a challenged and a control animal respectively. B, E, Fluorescence images (in color) for the experimental and control animals superimposed on the photographs of (A) and (D). C, F, H&E staining of lung sections obtained from the experimental and control animal 24 h after challenge. Inflammatory response of the challenged animal is seen as diffuse alveolar wall injury with noted thickening. Adatpted from Haller J et al.20 See also Chapter 67.

of Figure 1, was used. Laser light was beam expanded and directed t hrough a l ow n umerical a perture o bjective (Olympus, PlanN 10 × /0.25) to one side of the live pupa. Light propagation through the pupa body was captured by a CCD camera using a Leica Z16 APO apochromatically corrected zoom lens, using appropriate nar row band-pass interference filters (ex: 488 ± 5 nm, em: 513 ± 5 nm; the narrow width reducing autofluorescence ef fects). To improve image quality and reduce photon dif fusion, a polarization anal yzer w as placed in front of the microscope, oriented in parallel to the incident polarization light so as to reject highl y scattered photons, that is, photons that lose their polarization state. Theoretical modeling in this case w as based on the Fermi appro ximation inte grated over the ph ysical area seen by each pixel on the CCD camera used. To account for refractive index mismatch between inner volume of the pupae and the sur rounding air, the Fermi-based forward model w as fur ther cor rected b y calculating the correct angle of propagation in the medium using the Snell’s law of optical refraction. In version in this case was based on a back-projection algorithm. Fluorescence tomography of larger animals has been also shown, and imaging systems are now commercially available. Se veral other studies elucidate on specif ic

disease applications of this technolo gy.18,19,15 One specific example is shown in Figure 3, which depicts the results from imaging of inflamed mouse lungs, after administration of a cathepsin-acti vatable fluorescence probe (Prosense-680). In this case, imaging was also based on transillumination scanning using a limited angle projection scanner . The major dif ference of this approach, compared to the system shown on Figure 1, is that a focused laser beam is scanned on the animal surface when the animal is stationary, instead of the animal being rotated. Photographs of an animal challenged with intranasal instillation of O VA and a control animal treated with saline instillation only are shown in Figures 3A and 3D , respectively; see also Chapter 67 for more in-depth information on pulmonar y FMT imaging. Correspondingly, F igures 3B and 3E sho w fluorescence images obtained through the e xperimental and control animals, depicting a mark ed fluorescence increase from the experimental animal, congruent with the lung. Histologic analysis conf irmed parameters of an acute inflammatory response in the challenged lung (F igure 3C), featuring re gions of alv eolar w all thick ening and collapse with accompan ying edema and inflammator y cell penetration (obser ved at higher magnif ications; not shown), relative to that of healthy lung tissue (Figure 3F).

164

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

TOMOGRAPHIC CHARACTERISTICS Description of the de velopment of li ving organisms and of disease has lar gely relied on histolo gic sections or on other in vitro laboratory tests, allowing for measurements obtained from euthanized or e xcised tissues. Histolo gic analysis is challenging when monitoring quickly evolving phenomena or responses over time, for example, in monitoring disease e volution or dr ug ef fects. Although the GFP was imaged in the abo ve e xample, the technolo gy can be more generally applied to imaging red-shifted proteins or NIR fluorescence probes, in lar ger penetration scales. For example, it w as shown that the tomo graphic imaging of indoc yanine g reen accumulation is possib le through the human breast or small animals in vivo. Therefore, fluorescence tomo graphy becomes a cr ucial technology for studying the bio-distribution of new classes of fluorescent molecular probes or fluorescent p roteins i n l iving s ystems n oninvasively. Generally, a linear relation of reconstructed fluorochrome concentration and targeted molecule exist in biologically relevant concentrations, w hich allow for tr ue quantif ication, assuming that the nonlinearity associated with photon propagation in tissues is accounted for b y the appropriate theoretical model used.

REFERENCES 1. Giepmans BNG, Adams SR, Ellisman MH, Tsien RY. Review—the fluorescent toolbo x for assessing protein location and function. Science 2006;312:217–24. 2. Weissleder R, Ntziachristos V. Shedding light onto live molecular targets. Nat Med 2003;9:123–8. 3. Hell SW . Toward fluorescence nanoscop y. Nat Biotechnol 2003; 21:1347–55.

4. Ntziachristos V, Yodh AG, Schnall M, Chance B. Concurrent MRI and diffuse optical tomo graphy of breast after indoc yanine g reen enhancement. Proc Natl Acad Sci USA 2000;97:2767–72. 5. Jain RK. Nor malization of tumor v asculature: an emerging concept in antiangiogenic therapy. Science 2005;307:58–62. 6. Zipfel WR, Williams RM, Webb WW. Nonlinear magic: multiphoton microscopy in the biosciences. Nat Biotechnol 2003;21:1368–76. 7. Arridge SR. Optical tomography in medical imaging. Inverse Probl 1999;15:R41–R93. 8. Jobsis FF. Noninvasive, infrared monitoring of cerebral and m yocardial oxygen sufficiency and circulatory parameters. Science 1977; 198:1264–7. 9. Chance B. Optical method. Annu Rev Biophys Biophys Chem 1991; 20:1–28. 10. Patterson MS, Chance B , Wilson BC. Time resolv ed reflectance and transmittance for the nonin vasive measurement of tissue opticalproperties. Appl Opt 1989;28:2331–36. 11. Arridge SR, Hebden JC. Optical imaging in medicine II: modelling and reconstruction. Phys Med Biol 1997;42:841–53. 12. Yodh AG, Chance B. Spectroscopy and imaging with diffusing light. Phys Today 1995;48:34–40. 13. Chang JH, Graber HL, Barbour RL. Imaging of fluorescence in highly scattering media. IEEE Trans Biomed Eng 1997;44:810–22. 14. Klose AD, Netz U , Beuthan J , Hielscher AH. Optical tomo graphy using the time-independent equation of radiati ve transfer—part 1: forward model. J Quant Spectrosc RadiatTransf 2002;72:691–713. 15. Ntziachristos V, Ripoll J, Wang LHV, Weissleder R. Looking and listening to light: the e volution of w hole-body photonic imaging. Nat Biotechnol 2005;23:313–20. 16. Zacharakis G, Kambara H, Shih H, et al. Volumetric tomography of fluorescent proteins through small animals in-vi vo. Proc Natl Acad Sci USA 2005;102:18252–57. 17. Vinegoni C, Pitsouli C, Razansk y D , et al. In vi vo imaging of Drosophila melano gaster pupae with mesoscopic fluorescence tomography. Nat Methods 2007. [10.1038/ nmeth1149]. 18. Patwardhan SV, Bloch SR, Achilefu S, Culver JP. Time-dependent whole-body fluorescence tomo graphy of probe bio-distributions in mice. Opt Express 2005;13:2564–77. 19. Ntziachristos V, Tung C, Bremer C, Weissleder R. Fluorescence-mediated tomography resolves protease activity in vivo. Nat Med 2002; 8:757–60. 20. Haller J, De Kleine R, Hyde D, Ntziachristos V. Fluorescence tomography of inflammatory responses in the lung. J Appl Physiol 2007. [In press].

12 ENDOMICROSCOPY SEOK H. (ANDY) YUN, PHD AND CHARLES P. LIN, PHD

INTRODUCTION The aim of endoscopic microscop y is to enab le highresolution imaging of internal organs or tissue compartments at the cellular le vel through an optical probe that is introduced into the body in a minimall y in vasive manner.1 Endoscopic microscopy bridges an impor tant gap separating, on the one hand, traditional whole-body imaging modalities, such as magnetic resonance imaging and positron emission tomo graphy that lack the sensitivity and resolution (both spatial and temporal) to visualize single-cell dynamics in the body , and on the other hand, high-resolution optical imaging techniques, such as confocal and multiphoton microscopy, that lack the ability to image deep into tissue. By replacing the objective lens and other bulk y components of the standard microscope with a small-diameter probe, the ability of optical microscop y can be e xtended be yond easily accessib le surf aces, such as the skin or the cornea. F or small-animal molecular imaging research, cellular processes in inter nal or gans, including the brain, the gastrointestinal (GI) tract, the spleen, and the lymph nodes, can be visualized with much less tissue manipulation compared with traditional intra vital microscopy approaches.

TYPES OF ENDOMICROSCOPES Endomicroscopes Based on Fiber-Optic Bundles The most common type of endoscope uses a fle xible fiber-optic bundle to rela y images from the distal end of the bundle to the pro ximal end, where the image can be viewed b y e ye or captured b y a char ge-coupled de vice (CCD) or similar electronic cameras. The f ibers are arranged in a coherent manner such that the position of

each fiber within the bundle is maintained throughout the length of the bundle. In this w ay, an image is transmitted from one end of the bundle to the other end without getting scramb led. Depending on the imaging modality , a two-dimensional (2D) image is either transmitted in its entirety through all f ibers of the bundle at once, as in wide-field microscop y, or transmitted pix el-by-pixel sequentially through one f iber at a time, as in laser scanning microscop y. The dif ferent imaging modalities are discussed in section “Imaging Modalities.” The number of picture elements (pix els) within the image is determined by the number of fibers contained in the bundle. F or e xample, the Sumitomo IGN-08/30, a popular f iber-optic bundle, is 800 µm in diameter and contains 30,000 f ibers. Smaller bundles are a vailable with lower number of fibers (eg, IGN-028/06 is a 280 µm diameter bundle with 6,000 individual fibers). The ability to pack so man y f ibers into a small-diameter probe is remarkable, yet the numbers are lo w in comparison with the standard video resolution, such as the VGA for mat, which has 640 × 480 or a total of ~300,000 pix els. An issue related to the relati ve low pixel count is the low pixel density. To minimize cross talk between adjacent fibers, the interf iber spacing (center -to-center distance between neighboring f ibers) is al ways lar ger than the diameter of the light-transmitting cores. F or e xample, individual f ibers in the IGN-08/30 bundle mentioned above have a 2.4 µm core diameter, but the inter-fiber distance is about 4 µm. The dead space betw een adjacent cores reduces the light-coupling efficiency and also gives rise to a pixelated or honeycomb appearance in the image. The pix elation ar tifacts can be remo ved to some e xtent using image processing algorithms, such as lo w-pass f iltering or interpolation,2 but the underlying limit in resolution is not impro ved by such processing methods. If d is the inter-fiber spacing projected onto the sample (taking into account the demagnif ication f actor of the objected 165

166

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

lens), then according to sampling theor y, the highest spatial frequency that can be resolv ed is 1/2d. Therefore, to achieve a lateral resolution of 1 to 2 µm, the inter -fiber spacing will have to be demagnif ied by a factor of 4 to 8. While that is possib le, such resolution can onl y be achieved b y a signif icant reduction in light throughput 2 because of the relati vely high di vergence of the f iber output as dictated by the f iber numerical aperture (NA). Objective Lenses for Endomicroscopy

The output end of the f iber-optic bundle can be used in direct contact with the tissue sample (in w hich case the lateral resolution is determined by the inter-fiber spacing) or used with an objecti ve lens that (1) projects a (demagnified) image of the f iber bundle output surf ace onto the sample, (2) collects the remitted light from the sample, and (3) reimages it onto the f iber bundle. The physical size of the objecti ve lens, rather than the diameter of the fiber bundle, is often the limiting f actor in miniaturizing the probe. The NA of the objective lens, in a conventional microscope, deter mines the optical resolution and the light-gathering ef ficiency of the imaging system. In a fiber bundle endomicroscope, the resolution is usually not limited by the NA of the objective lens but by the (demagnified) inter-fiber distance at the sample plane. F or optimal coupling of the remitted light into the f iber-optic bundle, the lens should be designed for the specif ic fiberoptic bundle being used, such that the NA on the backside (fiber) of the lens matches the N A of the f iber. For fluorescence molecular imaging, chromatic aberration is of particular concern, especially if more than one fluorophore is to be used for multicolor imaging. Rouse and colleagues3 have designed a multielement lens with an NA of 0.46 and a maximum diameter of 3 mm. The lens has an achromatic range of 480 to 660 nm and is therefore suited for fluorescence molecular imaging applications. Karlson and colleagues4 and Chidley and colleagues5 used injection-molded plastic lenses with aspheric surfaces that can signif icantly reduce the manuf acturing cost. These lens assemblies are 7 mm in diameter and ha ve an NA of 1.0. However, they are designed for reflectance confocal microscopy at 1064 nm and ma y have not been tested for fluorescence imaging over a range of shorter wavelengths. Graded-index (GRIN) lens represents an attracti ve alternative to the con ventional spherical lenses. GRIN lens is made of a glass rod doped with metal ions, such as silver and thallium, to ha ve the inde x of refraction decreasing with radius. The “g raded” parabolic inde x profile causes the light ra ys to bend continuousl y within the lens and sta y focused on a spot, allo wing the real

image to be for med. The ion-exchange doping process 6,7 replaces the need for tightly controlled surface curvatures of miniature lenses. Commercial GRIN lenses are a vailable with NAs from 0.05 to 0.6, diameters from 0.35 to a few millimeters, and v arious pitches with relati vely low cost. An impor tant consideration is that the c ylindrical shape and planar end surf aces of a GRIN lens considerably facilitate miniature assembly. For their simplicity and a vailability, GRIN lenses ha ve been widel y used in v arious types of miniature endoscopic probes, despite some drawbacks, such as aber ration. Endomicroscopes with f iber-optic bundles coupled to GRIN objective lenses ha ve been described b y Knittel and colleagues8 and by Göbel and colleagues. 9

Endomicroscopes Based on Single Optical Fibers A second type of f iber-optic microscope uses a single optical f iber instead of a bundle of f ibers. This kind of microscope is inherently a scanning microscope because a single f iber cannot transmit an image. Consequentl y, the distal (output) end of the f iber has to be scanned in tw o dimensions to acquire an image, either b y mechanically vibrating10–12/rotating13 the f iber tip, the objecti ve lens together with the f iber tip, 14 or by using miniature beam scanners built into the endoscope.15 To achieve diffractionlimited focusing, single-mode f ibers are used w hose small-core diameters (3 to 7 µm) supports the propagation of a single mode of electromagnetic field within the fiber. One major advantage of the single fiber microscope is that images are free of the pixelation artifacts, unlike its fiberoptic bundle counter part. Ho wever, the need for distal scanning adds complexity and bulk to the probe. A tubular probe with a diameter as small as 2.4 mm has been realized that contains tw o pairs of piezoelectric actuators capable of resonantl y driving a cantilevered optical f iber in a 2D (spiral) scan patter n.12 Future advances in microfabrication processes, for e xample, using microelectromechanical systems technology,15,16 will undoubtedly lead to fur ther component miniaturization. F igure 1 sho ws a prototype side-looking optical probe assemb led on a micro-fabricated active silicon optical bench being de veloped for two-photon endomicroscopy.25 Spectrally Encoded Endomicroscopy

As an alternative to 2D beam scanning, a technique called spectrally encoded endoscopy (SEE)17 requires only onedimensional (1D) scanning to produce a 2D image. This

Endomicroscopy

is made possib le by a broadband light source and a dispersive element at the distal end of a single-mode f iber. Each w avelength component from the broadband light source is spatiall y separated b y the dispersi ve element and is brought to a dif ferent focus along a line at the sample b y the objecti ve lens (F igure 2). Backscattered light from different points along this line is brought back through the same dispersi ve element, w hich undoes the spatial separation, and delivered by the same single-mode optical fiber to a spectrometer. The spectrometer analyzes the amount of backscattering light at each location, encoded by its own wavelength. In this means, a 1D (line) image is formed, and physical scanning is only needed in the second dimension (in the direction per pendicular to the line) to obtain a 2D image. 18 Three-dimensional (3D) imaging is also possib le using optical interferometr y.19 A probe diameter of only 350 µm has been used to obtain the v olumetric images with about 400,000 resolv able points at 30 frames/second. 20 For fluorescence molecular imaging, the retur n photons do not ha ve the same w avelength, so an additional de gree of encoding is necessar y. In addition, the retur n (Stok es-shifted) photons do not retrace the same optical path through the dispersi ve

A

167

element, and consequentl y, it is not easil y coupled into the same optical fiber. Considerable reengineering is still needed to implement this interesting technology for fluorescence molecular imaging.

Rigid Endoscopes For small-animal imaging research, which does not need a flexible light guide o ver extended distances, the rigid endomicroscope is an attracti ve choice, allo wing minimally invasive imaging of inter nal or gans by inser ting the rigid probe tip through a small opening. Two types of rigid probes are a vailable. The GRIN endomicroscope22 is made by fusing together a short GRIN objective lens with a high N A (0.4 to 0.6) for tight beam focusing and a rod lens with lo w N A (0.1 to 0.2) for beam rela y.23 By replacing the f iber-optic b undle (see section “Endomicroscopes Based on Single Optical Fibers”) with the rod lens for beam rela y, the GRIN endomicroscope gains in light throughput and image quality (no pixelation artifact) but sacrifices mechanical flexibility. Typical length and diameter of the GRIN endomicroscopes are in the range of 10 to 25 mm and

B

Figure 1. Illustration (A) and image (B) of a prototype side-looking optical probe assembled on a microfabricated-active silicon optical bench. A double-clad photonic crystal fiber enters the device from the right. A microprism and a graded-index (GRIN) lens are visible in the center of the device. Arrows point to microfabricated-contoured thermomechanical actuators that move the prism and the GRIN lens.25

Figure 2. Spectrally encoded endoscope. Light from a broadband source that emerges from the single-mode fiber is spread along the wavelength axis by a dispersive element built into the probe tip so that different positions along this axis are encoded by a different wavelength (color). Tissue is scanned in the perpendicular direction to obtain a 2D image. Adapted from Yelin et al.20

168

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

0.3 to 1 mm, respectively. A second type of rigid probe, commonly called the stick lens, is a set of narrow diameter objective lenses developed by Olympus specifically for live-animal imaging pur poses.24 They are a vailable in 1.3 mm and 3.5 mm diameters, with N A of 0.5 and 0.7, and are cor rected for near-infrared wavelengths up to 1000 nm.

IMAGING MODALITIES Wide-Field Endomicroscopy In wide-f ield (epifluorescence) endomicroscop y, the entire field of view (FOV) is illuminated, and the image is acquired by a 2D array detector, such as a CCD. The widefield method does not pro vide optical sectioning since light from different tissue depths, both in and out of focus, contributes to image for mation and results in poor contrast. Whether a str ucture can be visualized depends on how bright the tar get is labeled and w hether it is strong enough to be visualized abo ve the backg round. Because no scanning is required, wide-field imaging has the advantages of simpler and less e xpensive instr umentation and faster full-frame acquisition. Both GRIN endoscope probes22,26 and f iber-optic bundles 27 have been used for wide-field endomicroscopy.

Endoscopic Confocal Microscopy In confocal microscop y, a point in the sample is illuminated b y a focused laser spot and remitted light from this point is detected through a confocal pinhole. The point is then scanned in two dimensions to build up an image. Conceptuall y, a straightforw ard way to convert the f iber bundle endomicroscope into a confocal system is to illuminate the indi vidual f ibers in the bundle one at a time and detect the remitted light coming back through the same f iber while rejecting light coming back through adjacent f ibers that car ry multipl y scattered photons and signals from out-of-focus objects. Although it is often stated that the illumination f iber acts as its own confocal pinhole in the detection path, a physical pinhole is in fact necessary to block light coming back through sur rounding f ibers that car ry unwanted photons. To obtain a 2D image, dif ferent fibers in the bundle are illuminated sequentiall y and each f iber detected confocall y. In the simplest implementation, a focused laser is scanned continuousl y over the entrance surf ace of the f iber bundle in a 2D (x–y) raster patter n. This raster scan patter n is rela yed to the distal end of the bundle and imaged onto the sample b y

the miniature objecti ve lens. Because the scan is continuous, the laser beam spends a considerab le portion of the scan time focused on the cladding instead of on the core of the f ibers, reducing the coupling ef ficiency, and increasing the backg round (nonconfocal) fluorescence le vel. Replacing the f iber bundle with a GRIN rod lens alle viates these prob lems and also removes the pixelation artifact, but it limits the imaging targets to str uctures that can be reached with a shor t rigid probe. Using the f iber bundle or the GRIN lens to relay the image has the adv antage that scanning is accomplished outside the endoscope, making it unnecessary to integrate a scan head into the probe. To increase the frame rate, an entire linear ar ray of fibers can be illuminated at once and detected by an array detector through a confocal slit aper ture.28 This arrangement is similar to a slit-scanning confocal microscope, which needs scanning only in the second dimension perpendicular to the long dim ension of the slit. With this approach, images can be obtained at video rate and above. The gain in speed and simplicity with the slit aperture design is balanced b y a compromise in its ability to reject of out of focus light. The rejection has a 1/z dependence as opposed to 1/z2 for the pinhole confocal system, where z is the distance away from the focal plane. 28 As described in section “Endomicroscopes Based on Single Optical Fibers,” a fiber-optic microscope based on a single-mode optical fiber requires a scanning mechanism at the output end of the fiber. The return light is descanned by the same scanner and coupled into the same optical f iber, which acts as a confocal pinhole. This type of endomicroscope is therefore, intrinsically, a scanning confocal microscope. In regular confocal microscopy, when detecting very weak fluorescence signals, it is customar y to increase the detection pinhole size, which increases the sensitivity while sacrificing the axial resolution. In f iber-optic confocal microscopy, light-collection ef ficiency can be impro ved using a doub le-clad f iber that has a center core, an inner cladding, and an outer cladding. The illumination light is carried in a single-spatial mode do wn the center core, whereas the return light is coupled into the multimode inner cladding, w hose diameter is much lar ger than the center core, thereby increasing the light-gathering efficiency.29

Endoscopic Two-Photon Microscopy Two-photon excitation fluorescence (TPF) microscop y has emerged as a powerful intravital imaging technique for biomedical research, par ticularly in small-animal models. Additionally, it has the potential to become a clinical tool for intraoperative tissue characterization and molecular

Endomicroscopy

diagnosis.30 Despite its use of near -infrared excitation that allows deeper tissue penetration and less photodamage, imaging depth in most tissues remain limited , typically to less than a fe w hundred microns from the surf ace. TPF endomicroscopy has been demonstrated with GRIN lenses as focusing optics in conjunction with either GRIN rela y lenses or with f iber-optic delivery.9,11,22 A special consideration for fiber-optic TPF endomicroscop y is the pulsebroadening ef fect w hen ultrashor t laser pulses are propagated through optical f ibers. This is undesirab le because the efficiency of two-photon excitation is reduced as the pulse duration increases. One cause of pulse broadening is g roup velocity dispersion, w hen slightly different “colors” within the laser pulse tra vel at dif ferent velocity through a medium, such as glass (because the inde x of refraction varies with wavelength). As a result, the different colors get out of step and the pulse broadens. Different colors exist in an ultrashor t laser pulse because of the uncertainty relation: the shor ter the pulse the broader its spectrum. F or e xample, a typical Titanium (T i):sapphire laser pulse used for tw o-photon microscop y has a pulse width of about 100 fsec and a spectral width of appro ximately 10 nm, centered at 800 nm (ie, a color spread o ver the 795 to 805 nm range). Group velocity dispersion can be compensated for by introducing a dif ferent delay length to each color (a technique called prechir ping) so that all the colors get back in step again. Another cause of pulse broadening is nonlinear self-phase modulation that occurs at high intensities and is more difficult to compensate. New hollow core or large mode area photonic cr ystal f ibers can deliver femtosecond laser pulses without self-phase modulation. Ho wever, these f ibers are not a vailable as bundles; therefore, it is necessar y to use single f iber with postfiber scanning for two-photon imaging.

Optical Coherence Endomicroscopy Optical coherence tomography (OCT) uses a low coherence (broadband) light source and interferometric techniques to perform depth-resolv ed imaging with resolutions ranging from 2 to 15 µm.31 With large penetration depths of about 2 mm in tissue, cross-sectional imaging capabilities, and relative ease of incor poration into f iber-optic catheters and endoscopes,32 OCT endomicroscop y is a promising candidate for comprehensi ve screening or intraoperati ve tissue characterizations. 33,34 Recent human studies ha ve demonstrated that OCT is capab le of identifying dysplasia in Barret’s esophagus and colonic adenomas and can identify all the histopatholo gic features of vulnerab le plaque.35–37 In the past fe w y ears, the imaging speed of OCT has been impro ved signif icantly using frequenc y

169

domain, instead of time domain, ranging. 33 A minimall y invasive procedure, such as catheterization, is used to deliver the OCT probe to the tar get or gan or system. Through rotation and longitudinal pullback of the inter nal portion of the probe, the OCT system records the full 3D microstructure of the tissue. Subsequent image processing can then be used to produce a for m of vir tual histolo gy, where arbitrar y cross-sectional vie ws of the tissue can be viewed to screen for disease. Gi ven the indi vidual v oxel dimensions of 15 µm × 15 µm × 10 µm, the systems obtain data at a rate of 60 mm 3/s permitting visualization of large tissue volumes with microscopic resolution. The coherent detection scheme used in OCT cannot directly detect fluorescence. Ho wever, several molecular contrast methods ha ve already been suggested for conventional OCT , such as Coherent anti-Stok es Raman shift, molecular probes, and second harmonic generation. Furthermore, functional techniques are on the horizon, including Doppler -OCT for quantitati ve assessment of blood flow38 and polarization-sensitive OCT for quantification of collagen content,39 providing further molecularand ph ysiologic-contrast mechanisms that ma y be used for the study of disease and primar y diagnosis. Recent developments in v arious contrast agents for OCT are reviewed in Boppart and colleagues. 40

IMPLEMENTATION AND APPLICATIONS System Description After taking into consideration the v arious design options discussed in the pre vious two sections, w e have chosen to implement a combined confocal and multiphoton endomicroscopy system based on a GRIN lens probe for in vi vo imaging of small animals. The system uses continuous wave lasers at 491 nm, 532 nm, and 635 nm for single-photon fluorescence e xcitation and a mode-lock ed tunab le Ti:Sapphire laser (780 to 920 nm) for multiphoton e xcitation, to provide broad spectral coverage to a wide range of molecular probes. A schematic of the system is sho wn in Figure 3. Flip mir rors and dichroic splitters are used to select and direct a laser beam to a raster beam scanner comprising a silv er-coated polygon scanner ( x-axis) and a galvanometer (y-axis). We designed the microscope to ha ve a FOV of 250 µm, when a 40 × 0.6 NA objective lens is used. The proximal end of the optical probe is placed appro ximately at the focal plane of the objecti ve lens. A confocal pinhole and a photomultiplier tube (PMT) are used for confocal fluorescence and reflectance imaging, w hereas another PMT-2 is used for tw o-photon fluorescence and second harmonic generation imaging. An 8-bit 3-ch frame

170

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A A

B B

C C

D D

Figure 3. Rigid graded-index (GRIN) microendoscope probes. A, A doublet probe with a high numerical aperture (NA) GRIN lens (NA 0.45–0.6, pitch 0.16–0.25) coupled to a low NA relay lens (NA 0.1–0.2, 3/4 pitch). B, A triplet with two high NA GRIN lenses (NA 0.45–0.6, pitch 0.16–0.25) sandwiching a low NA relay lens (NA 0.1–0.2, 3/4 pitch). C, Photograph of three probes with diameters of 1, 0.5, and 0.35 mm, respectively. Each probe is a compound triplet GRIN lens, as depicted in (B), consisting of a coupling lens, an objective lens, and a longer relay lens in between. Minor ticks on the scale bar represent 1 mm. D, Video-rate confocal and multiphoton imaging system. (A–C) adapted with permission from Jung et al.23 and (D) adapted from Kim et al.51

grabber digitizes the output of each PMT at 10 MS/s, acquiring 500 × 500 pixels/frame. The system acquires and displays images in real time at a frame rate of 30 Hz and can save them to hard disk simultaneousl y. There is a real-time display at a video rate ( > 15 frames/second) that considerably facilitates image-guided navigation of the probe within the animal.

Optical Probe and Resolution Figure 3 shows the principle of GRIN endoscopic probes. The scanning strategy shown in Figure 3A involves a doublet probe with a 3/4-pitch relay lens. The collimated laser beam (red ar rows) is focused onto the back f ace of the endoscope probe. The laser focus is scanned transv ersely (dashed arrow line). The scanning focus is rela yed to the image plane within the tissue sample (dashed ar row line). The two-photon-excited fluorescence (green arrows) generated at the sample focal spot returning through the endoscope probe is detected with a PMT via a dichroic mir ror. There are several variants of this strategy. The triplet probe shown in F igure 3B is comprised of an objecti ve lens, a relay lens, and a 1/4-pitch GRIN coupling lens with high NA matching that of a microscope objecti ve that couples the excitation laser light into the probe. We designed and f abricated se veral probes using 1-mm-diameter commercial GRIN lenses (NSG and Grintech) in the triplet str ucture. It comprises tw o high NA (0.45 to 0.6) GRIN lenses (pitch 0.16 to 0.25) and a

half-pitch low NA (0.1 to 0.2) relay lens in between. The probes w e used in the e xperiments described here are 15 mm long and have 0.45-imaging NA, a FOV diameter of 250 to 300 µm, and a working distance of 0 to 300 µm with water immersion. In two-photon excitation imaging at 800 nm, the measured resolution was 1.1 ± 0.08 µm in x or y and 13.4 ± 0.3 µm in z, defined as full width at half maximum. In confocal imaging, we measured the resolution as a function of the pinhole diameter . Pinhole sizes greater than the Airy disk are commonl y used for deeptissue imaging because lar ger pinhole sizes impro ve the photon collection ef ficiency at the e xpense of reduced axial resolution. Consistent with the theory, we measured the transverse resolution to be relatively insensitive to the pinhole size and found a strong dependence for both axial resolution and signal strength (Figure 4). We chose to use a pinhole with one Airy disk size (50 µm). This provided a resolution of 1.5 ± 0.08 µm in x or y and 12.4 ± 0.3 µm in z, compared with that of tw o-photon imaging. These values are approximately two times larger than theoretical diffraction limits of 0.45-NA optics, which is attributed to spatial aberration of the GRIN probe.

Method to Mitigate Chromatic Dispersion for Multicolor Confocal Imaging Multicolor imaging in confocal modality is challenging because the GRIN probe has signif icant chromatic aberration. We measured that the focal depth w as dif ferent

Endomicroscopy

A A

171

B B

Figure 4. Measure resolution (A) and signal strength (B) as a function of confocal pinhole size. Dotted lines in (A) denote the theoretical diffraction limits of a lens with numerical aperture 0.45. Adapted from Kim et al.51

between 491-nm and 635-nm e xcitation wavelengths by 35 to 90 µm, depending on specif ic GRIN lenses used. Such a large variation of focus, cor responding to several cell la yers, w ould be a signif icant prob lem fr ustrating multicolor-excitation cellular imaging. To mitigate this problem, we implemented a simple technique; w henever the excitation wavelength was changed, we adjusted the vertical position of the objecti ve lens b y a precalibrated distance (35 to 90 µm) accordingly so that the focal plane of the probe w as kept unchanged. Fur thermore, we optimized the pinhole position and the spectral width of a filter to minimize image b lur due to the w avelength differences between e xcitation and emission and within the emission band.

Images of Test Samples To test the imaging system, we acquired confocal images of a pine embryo at excitation wavelengths of 491 and 635 nm, respectively, using the chromatic aber ration compensation technique described earlier . A mer ged image (F igure 5) demonstrates co-re gistration of the red and g reen fluorescence images within a fe w microns o ver nearly the entire FOV. Image blurring near the FOV boundary is due to f ield curvature. Each image w as averaged over 30 frames (total acquisition time 1 sec) at the same sample position.

In Vivo Imaging of Intact Skin of Mice To demonstrate in vi vo tissue imaging, we conducted noninvasive confocal and multiphoton endoscop y in the ear skin of an anesthetized mouse. Using the GRIN probe, w e were able to visualize indi vidual dendritic cells e xpressing green fluorescent proteins (GFP) in epider mis and der mis (Figure 6A), blood plasma labeled with Rhodamine-B dyes conjugated with dextran after tail-vein injection (2,000,000

Figure 5. Two-color confocal image of a test sample (pine embryo) taken with the graded-index endoscope. Scale bar = 25 µm. Adapted from Kim et al.51

MW, 200 µg/200 µL) (Figure 6B), and collagen f ibers via endogenous SHG in the der mis (F igure 6C). Video-rate monitoring not onl y considerably facilitates image na vigation but also allo ws monitoring of f ast dynamic processes, such as cell traf ficking in the b lood stream. F igures 6D–F show the mo vie frames acquired b y confocal microendoscopy from a GFP+ Tie-2 mouse in which all the vasculature endothelial cells are GFP +, after tail-v ein injection of human ovarian cancer cells labeled with a membrane dy e, DiD (Invitrogen). Flowing cells could be obser ved clearly, and the cell count and v elocity could be measured. The maximum penetration depth was 50 to 100 µm from the surface of the skin, limited b y finite signal-to-noise ratio and the out of focus backg round.

In Vivo Imaging of the Mouse GI Tract A number of mouse models of inflammator y bowel disease and colon cancer are a vailable to study the mucosal immune system, colitis, and development of cancer in the gut. However, due to the small size of the GI or gans in mice, standard microscopy would require extensive surgical opening and , therefore, has not been widel y used. In the past fe w years, f iber-optic scanning endomicroscop y systems ha ve become commerciall y a vailable and ha ve been used for cellular and molecular imaging of the GI organs in mice and humans. 41,42 In addition, miniature wide-field endoscopes ha ve been demonstrated for fluorescence mouse colonoscop y.43,44 We ha ve recentl y

172

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

C

D

E

F

Figure 6. Images from the intact skin of an anesthetized mouse, taken with the graded-index endomicroscope. A, A confocal fluorescence image showing major histocompatibility complex (MHC) class-II+ Langerhans cells expressing green fluorescent proteins (GFP) in the epidermis. The MHC class II GFP mouse was kindly provided by Dr Marianne Boes at Harvard Medical School. Excitation 491 nm, emission 520/35 nm. B, A two-photon fluorescence image of blood vessels labeled with rhodamine dextran. Excitation 800 nm, emission 590/80 nm. C, Endogenous collagen second harmonic signal excited at 800 nm and detected at 400 nm. D–F, A sequence of frames showing ovarian cancer cells (red) in blood circulation, superimposed on a green fluorescence image of blood vessels in a GFP+ Tie-2 mouse. Scale bar = 50 µm. Adapted from Kim et al.51

studied the possibility of using GRIN microendoscope probes to obtain high-resolution fluorescence images. In one experiment, we used the 1-mm-diameter GRIN probe to image the small and lar ge intestines of a major histocompatibility comple x class-II GFP mouse via a minimally in vasive laparotom y procedure (F igure 7A). The dendritic-like cells in the lamina propria of the colon are clearly resolved (Figure 7B). The penetration depth of the confocal GRIN probe was sufficient to visualize the GFP+ epithelial dendritic cells through the entire colonic w all (Figure 7C). For noninvasive colonoscopy, we fabricated a side-looking GRIN probe with a 90º prism reflector attached to the distal end (F igure 7D). The colonoscope was introduced via rectal inser tion. The side-looking probe design a voids the need for insuf flation b y an air pump. Figure 7E shows the same GFP structure as seen in Figure 7F, now imaged from the inside of the colon.

In vivo Imaging of Mouse Model of Heart Transplant Coronary ar tery disease represents the major threat to patients w ho ha ve under gone hear t transplantation. Endomicroscopy can visualize the immune responses and disease progression in the mouse models of chronic or gan rejection,45 but cellular imaging of a transplanted hear t has not been possible due to heartbeat motion. To overcome this problem, we have developed a miniature tissue-holding suction tube and used it with the GRIN endoscope probe for

heart imaging. We modified a 15-gauge hypodermic needle (inner diameter 1.37 mm, outer diameter 1.83 mm) and connected it to a mini v acuum pump with a flexible plastic tube. A 1-mm-diameter GRIN probe w as inserted through the inner channel of the needle tube. The blunted distal tip of the needle is made to contact the tissue so that it can hold it by gentle suction. We found an optimum range of vacuum pressure in the neighborhood of 100 mm Hg or 13 kP a, which is sufficient to freeze the tissue mo vement but does not cause apparent adverse effects, such as tissue damage or ischemia. We have used this de vice to obtain images from the inter nal or gans of mice, including small and lar ge intestines, spleen, kidney, and transplanted heart. Figure 8 shows a sequence of images obtained from a coronar y ar tery in a syngeneic hear t transplantation mouse model w here the hear t is not rejected. The hear t was placed in the abdominal ca vity in an actin GFP + recipient mouse.46 The images show the recipient’s GFP + cells flowing in the coronar y ar tery, some of w hich are arrested in the endothelial w all and extravasate (circles). During the course of imaging for about 30 min, the tissue surface was held by suction; no adverse phenomena were detected and the blood flow appeared normal.

CHALLENGES AND OUTLOOK In summary, endoscopic microscopy is a promising technology for molecular imaging research because it provides unique ways to look into tissue at a resolution pre viously

Endomicroscopy

C

B

A

D

173

EE

Figure 7. A, In vivo imaging of the colon of a MHC class II green fluorescent proteins (GFP) mouse using the confocal graded-index (GRIN) endomicroscope. The probe was inserted through a small incision in the skin. B, Dendritic-like GFP+ cells in the lamina propria. C, GFP+ epithelial dendritic cells in the colonic wall. D, A GRIN endoscope with a side-looking prism. E, The same GFP+ epithelial dendritic cells imaged from the inside of the colon using the side-looking probe. Scale bar = 50 µm.

A

B B

Figure 8. A, Anatomy of transplanted heart in the abdominal cavity. B, A sequence of images showing the recipient’s green fluorescent proteins+ cells flowing in the coronary artery, some of which are arrested in the endothelial wall and extravasate (circles). Scale bar = 50 µm.

unattainable with traditional imaging modalities. Se veral commercial endomicroscopes are cur rently available providing a range of options from fiber-optic bundles (Mauna Kea) to single f iber deli very (OptiScan) and rigid stick lens probes (Ol ympus). As in other fluorescence-based techniques, rapid pro gress in the de velopment of fluorescent molecular sensors and repor ters will enab le more selective imaging of molecular targets and assessing cellular functions. In addition, ne w contrast mechanisms, such as Coherent anti-Stok es Raman scattering (CARS) microscopy, are also being studied for chemical imaging

of molecular species based on their vibrational signatures.47 Indeed, CARS microscopy through a 1.3 mm rigid probe (Olympus stick lens) has recentl y been repor ted.48 Fiber-optic transmission of picosecond laser pulses for CARS microscop y49 is potentiall y less demanding, because of lower peak power and narrower spectral width, compared with transmission of femtosecond laser pulses used for two-photon microscopy. Ongoing challenges include fur ther component optimization and miniaturization, using the latest microfabricated and nanof abrication technolo gy. In addition,

174

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

system-level inte gration will become more critical for practical users. Issues, such as FO V, na vigation, and motion of the li ve subject will confront da y-to-day practitioners and will ha ve to be addressed , likely in an application-specific manner.

FOV In general, FOV of a microscope decreases with increasing magnification and the N A of the objecti ve lens. The FOV also decreases as the size of the objecti ve lens is reduced while maintaining the NA and magnification. For example, the 0.5-N A microprobe ha ving a diameter of 1.3 mm, developed by Olympus, gives a FOV of 200 µm, whereas a standard objective lens with the same NA would allow at least three times larger FOV. As a result, sampling volume tends to be limited in small-diameter high-resolution endomicroscopy. A small FO V can cause sampling er rors in quantitative analysis or clinical diagnosis. In addition, it is not possib le to monitor spatiall y distant e vents at the same time, when separated more than the FOV. Therefore, FOV is an impor tant design parameter in de veloping or using endomicroscop y. In some applications, such as intraluminal imaging, rapid scanning of the probe o ver a wide area larger than the otherwise limited FO V may be a viable solution.18,33 The resulting large data set places a premium demand on the computation po wer to acquire, process, analyze, and display volumetric information.

Navigation The small FOV of a endomicroscope also renders it difficult to kno w e xactly w hat inter nal str ucture is being imaged. Knowledge of the anatomy alone may not be sufficient to na vigate the probe to the right tar get location, and a navigation strategy may need to be de veloped that combines imaging modalities capab le of visualizing different size scales, so that major anatomic landmarks can be used for initial cursory positioning of the probe, before zooming in to the precise tar get location. In the most demanding experiments, one may want to image the same cell or group of cells over time. Such experiments require repeated positioning accuracy to better than 100 µm. Fast scanning (ie, acquiring images at video rate or close to video rate) is v ery useful for rapidl y surveying the local tissue landscape 50 and orienting the probe. 51 Alternatively, the probe can be implanted or affixed to tissue, for example, with a head-mounted gear that allows long-term imaging of neurons in the brain, even in awake animals.11

Motion Physiological motion of the live subject, such as breathing and heartbeat, can be a serious prob lem in high-resolution endomicroscopy. A tissue mo vement of more than the instrumental resolution during image acquisition results in undesirable ar tifacts, such as b lur and distor tion. Image processing methods or time-gated data acquisition in synchronous with a periodic mo vement ma y be useful in reducing motion ar tifacts. Application of immobilizing drugs, gentle pressing of the tissue surf ace with the end of endoscopic probe or suck device, such as the one described in section “in vi vo Imaging of Mouse Model of Hear t Transplant” are all possible solutions to stabilize the tissue. Perhaps the most elegant solution, one that does not introduce any perturbation to the tissue, is to acquire images at very high frame rates such that during each frame the motion is ef fectively frozen. Image co-re gistration algorithms can then be used to eliminate motion-induced image shift from frame-to-frame. 50 Extension of this method to 3D image stabilization is in principle feasib le but remains technically challenging.

REFERENCES 1. Flusberg BA, Cocker ED, Piyawattanametha W, et al. F iber-optic fluorescence imaging. Nat Methods 2005;2:941–50. 2. Delaney PM, Harris MR. In: Pawley JB, editor. Handbook of biological confocal microscop y. Ne w York: Plenum Press; 1995. p. 515–23. 3. Rouse AR, Kano A, Udovich JA, et al. Design and demonstration of a miniature catheter for a confocal microendoscope. Appl Opt 2004;43:5763–71. 4. Carlson K, Chidley M, Sung KB, et al. In vivo fiber-optic confocal reflectance microscope with an injection-molded plastic miniature objective lens. Appl Opt 2005;44:1792–7. 5. Chidley MD, Carlson KD, Richards-Kortum RR, Descour MR. Design, assembly, and optical bench testing of a high-numerical-aper ture miniature injection-molded objecti ve for f iber-optic confocal reflectance microscopy. Appl Opt 2006;45:2545–54. 6. Ohmi S, Sakai H, Asahara Y, et al. Gradient-index rod lens made by a double ion-exchange process. Appl Opt 1988;27:496–99. 7. Messerschmidt B, Possner T, Goering R. Colorless g radient-index cylindrical lenses with high numerical aper tures produced b y silver-ion exchange. Appl Opt 1995;34:7825–30. 8. Knittel J, Schnieder L, Buess G, et al. Endoscope-compatib le confocal microscope using a g radient index-len system. Opt Commun 2001;188:267–73. 9. Göbel W, Kerr JND, Nimmerjahn A, Helmchen F. Miniaturized tw ophoton microscope based on a flexible coherent fiber bundle and a gradient-index lens objective. Opt Lett 2004;29:2521–23. 10. Helmchen F, Fee MS, Tank DW, Denk W. A miniature head-mounted two-photon microscope high-resolution brain imaging in freel y moving animals. Neuron 2001;31:903–12. 11. Flusberg BA, Jung JC, Cocker ED, et al. In vivo brain imaging using a por table 3.9 g ram tw o-photon fluorescence microendoscope. Opt Lett 2005;30:2272–74. 12. Myaing MT, MacDonald DJ, Li XD. Fiber-optic scanning two-photon fluorescence endoscope. Opt Lett 2006;31:1076–79.

Endomicroscopy

13. Bird D , Gu M. Two-photon fluorescence endoscop y with a micro-optic scanning head. Opt Lett 2003;28:1552–54. 14. Ota T, Fukuyama H, Ishihara Y, et al. In situ fluorescence imaging of organs through compact scanning head for confocal laser microscopy. J Biomed Opt 2005;10:024010. 15. Piyawattanametha W, Barretto RPJ, Ko TH, et al. Fast-scanning twophoton fluorescence imaging based on a microelectromechanical systems tw o- dimensional scanning mir ror. Opt Lett 2006; 31:2018–20. 16. Dickensheets DL, Kino GS. Micromachined scanning confocal optical microscope. Opt Lett 1996;21:764–66. 17. Tearney GJ , Webb RH, Bouma BE. Spectrall y encoded confocal microscopy. Opt Lett 1998;23:1152–54. 18. Yelin D, Boudoux C, Bounia BE, Tearney GJ. Lar ge area confocal microscopy. Opt Lett 2007;32:1102–04. 19. Yelin D , White WM, Motz JT , et al. Spectral-domain spectrall yencoded endoscopy. Opt Express 2007;15:2432–44. 20. Yelin D , Rizvi I, White WM, et al. Three-dimensional miniature endoscopy. Nature 2006;443:765–765. 21. Motz JT, Yelin D, Vakoc BJ, et al. Spectral- and frequenc y-encoded fluorescence imaging. Opt Lett 2005;30:2760–2. 22. Jung JC, Schnitzer MJ . Multiphoton endoscop y. Opt Lett 2003;28:902–4. 23. Jung JC, Mehta AD, Aksay E, et al. In vi vo mammalian brain imaging using one- and tw o-photon fluorescence microendoscop y. J Neurophysiol 2004;92:3121–33. 24. Alencar H, Mahmood U , Ka wano Y, et al. No vel multiw avelength microscopic scanner for mouse imaging. Neoplasia 2005;7:977–83. 25. Chen SC. [PhD disser tation]. Depar tment of Mechanical Engineering, Massachusetts Institute of Technology; 2007. 26. Monfared A, Ble vins NH, Cheung ELM, et al. In vi vo imaging of mammalian cochlear b lood flo w using fluorescence microendoscopy. Otol Neurotol 2006;27:144–152. 27. Hirano M, Yamashita Y, Miyaka wa A. In vi vo visualization of hippocampal cells and dynamics of Ca2+ concentration during anoxia: feasibility of a f iber-optic plate microscope system for in vivo experiments. Brain Res 1996;732:61–68. 28. Sabharwal YS, Rouse AR, Donaldson LT, et al. Slit-scanning confocal microendoscope for high-resolution in vivo imaging. Appl Opt 1999;38:7133–44. 29. Yelin D , Bouma BE, Yun SH, Tearney GJ . Doub le-clad f iber for endoscopy. Opt Lett 2004;29:2408–10. 30. Zipfel WR, Williams RM, Christie R, et al. Live tissue intrinsic emission microscop y using multiphoton-e xcited nati ve fluorescence and second harmonic generation. Proc Natl Acad Sci U S A 2003; 100:7075–80. 31. Fujimoto JG. Optical coherence tomo graphy for ultrahigh resolution in vivo imaging. Nat Biotechnol 2003;21:1361–67. 32. Tearney GJ, Brezinski ME, Bouma BE, et al. In vivo endoscopic optical biopsy with optical coherence tomo graphy. Science 1997; 276:2037–39. 33. Yun SH, Tearney GJ, Vakoc BJ, et al. Comprehensive volumetric optical microscopy in vivo. Nat Med 2006;12:1429–33.

175

34. Yaqoob Z, Wu J, McDowell EJ, et al. Methods and application areas of endoscopic optical coherence tomography. J Biomed Opt 2006; 11:063001. 35. Poneros JM, Brand S, Bouma BE, et al. Diagnosis of special ized intestinal metaplasia by optical coherence tomography. Gastroenterology 2001;120:7–12. 36. Westphal V, Rollins AM, Willis J, et al. Cor relation of endoscopic optical coherence tomo graphy with histolo gy in the lo wer-GI tract. Gastrointest Endosc 2005;61:537–46. 37. Jang IK, Tearney GJ, MacNeill BM, et al. In vi vo characterization of coronary atherosclerotic plaque using optical coherence tomography. Circulation 2005;111:1551–55. 38. Vakoc BJ , Shishk o M, Yun SH, et al. Comprehensi ve esophageal microscopy b y using optical frequenc y-domain imaging (with video). Gastrointest Endosc 2007;65:898–905. 39. Nadkarni SK, Pierce MC, P ark BH, et al. Measurement of collagen and smooth muscle cell content in atherosclerotic plaques using polarization-sensitive optical coherence tomo graphy. J Am Coll Cardiol 2007;49:1474–81. 40. Boppart SA, Oldenbur g AL, Xu CY, Marks DL. Optical probes and techniques for molecular contrast enhancement in coherence imaging. J Biomed Opt 2005;10:41208. 41. Goetz M, F ottner C, Schir rmacher E, et al. In-vi vo confocal real-time mini-microscopy in animal models of human inflammatory and neoplastic diseases. Endoscopy 2007;39:350–6. 42. Kiesslich R, Goetz M, Lammersdorf K, et al. Chromoscop y-guided endomicroscopy increases the diagnostic yield of intraepithelial neoplasia in ulcerative colitis. Gastroenterology 2007;132:874–82. 43. Becker C, Fantini MC, Neurath MF. High-resolution colonoscopy in live mice. Nat Protoc 2006;1:2900–04. 44. Alencar H, Funo vics MA, F igueiredo J, et al. Colonic adenocarcinomas: near-infrared microcatheter imaging of smart probes for early detection—Study in mice. Radiology 2007;244:232–8. 45. Benichou G, Akiyama Y, Roughan J, et al. Mechanisms of alloreco gnition. In: Wilkes D, Burlingham W, editors. Immunobiolo gy of organ transplantation. Ne w York: Kluv er Acad./Plenum Press; 2004. p. 107–37. 46. Hasegawa T, Visovatti SH, Hyman MC, et al. Heterotopic vascularized murine cardiac transplantation to study g raft arteriopathy. Nat Protoc 2007;2:471–80. 47. Evans CL, Potma EO, Pouris’haag M, et al. Chemical imaging of tissue in vivo with video-rate coherent anti-Stokes Raman scattering microsocpy. Proc Natl Acad Sci U S A 2005;102:16807–12. 48. Wang H, Huff TB, Fu Y, et al. Increasing the imaging depth of coherent anti-Stok es Raman scattering microscop y with a miniature microscope objective. Opt Lett 2007;32:2212–4. 49. Legare F , Ev ans CL, Ganikhano v F , Xie XS. Towards CARS endoscopy. Opt Express 2006;14:4427–32. 50. Veilleux I, Spencer J A, Biss DP , et al. In vi vo cell tracking with video rate multimodality laser scanning microscop y. IEEE J Select Top Quantum Electron 2008;14:10–18. 51. Kim P, Puoris’haag M, Cote D , et al. In vi vo confocal and multiphoton microendoscopy. J Biomed Opt 2008;13:010501.

13 INTRAVITAL MICROSCOPY THORSTEN R. MEMPEL, MD, PHD

The introduction of optical imaging, that is, the use of light (the visib le par t of the electromagnetic spectr um) as a source of image contrast, into the practice of molecular imaging has created an intersection with the field of intravital microscopy (IVM). Both disciplines are often refer red to as “in vivo imaging” but have developed from quite distinct conceptual origins. While the idea of molecular imaging has been de vised b y radiolo gists as an approach to improve the diagnosis of disease in human patients, IVM began as a means to satisfy the curiosity of natural scientists about the inner w orkings of v arious nonhuman subjects, such as bats and fro gs. Since the introduction of fluorescence to their w ork, intra vital microscopists ha ve been gathering e xperience with this tool and are now at a stage where improved imaging technolo gy will allow the dynamic visualization of molecular processes at subcellular resolution in vi vo. Molecular imaging and IVM will thus likely benef it from each other , and cross-fertilization should be culti vated by regular peeks into the neighbor’ s yard. In this chapter, some historical and technical aspects of IVM and its application to biomedical research will be reviewed in order to pro vide a resource and star ting point for further exploration for interested researchers rooted in molecular imaging.

WHY IVM? IVM is the obser vation of biolo gical processes within the physiological conte xt of a li ving specimen using a microscope. It provides dynamic infor mation about th e mechanics of multicellular organisms over a wide spatial and temporal range (less than a micrometer to more than a millimeter and m icroseconds to months). Its unique power lies in providing the ability to study the integrated function of cellular and subcellular systems subjected to comple x information inputs the y encounter in vivo. It compares to “seeing an animal in its natural 176

environment rather than in a zoo” (Mark Davis, personal communication). In multicellular organisms, biological processes are largely controlled at the single-cell level. Each individual cell is equipped with sensors for e xternal stimuli, a subcellular network of information pathways to transmit and integrate these inputs, and v arious w ays to respond b y altered behavior (eg, migration, cell division) or by communicating change back to the en vironment (expression of secreted or cell surf ace molecules). When multicellular systems are reconstituted in vitro, the external stimuli can be fairly well controlled (defined stimuli are varied while unknown and ar tifactual stimuli are k ept constant) to study specif ic responses at the single cell or population level. i n vivo, however, cells are subjected to a comple x and rapidl y changing spectr um of signals that v ary between dif ferent p hysiological and patho-ph ysiological states of an or ganism and betw een its dif ferent anatomic microenvironments. Except for some physical stimuli (eg, pressure, light, or temperature), these signals are received from other cells either through direct ph ysical contact, soluble factors (eg, cytokines, hormones) or via extracellular matrix components. These for ms of cell-cell communication therefore constitute another infor mation network at the supracellular le vel. The sum of these signals, which in their entirety are either not kno wn or cannot be simulated in vitro, to gether with a cell’ s genetic make-up, determines its function. Due to the dynamically changing nature of tissue environments and their spatial hetero geneity, the situation of two cells of the same lineage and at the same developmental stage, within the same tissue but separated b y a fe w micrometers, may be dramatically different, and their subsequent fates may be entirely distinct. Such hetero geneity is overlooked in e x vivo population measurements, w here similarity and dissimilarity of cells are deter mined based on a fe w selected def ining characteristics and w here the

Intravital Microscopy

spatial infor mation about otherwise tr uly similar cells is lost. In order to obtain an inte grated view of how cellular functions are controlled , one must def ine the rele vant forms of cellular infor mation exchange and then measure the biolo gical response of indi vidual cells i n vivo. The only w ay to achie ve such discriminator y potential is through single-cell measurements i n vivo, and cur rently, the only available methodology for this purpose is IVM. As a consequence of its unique capabilities, IVM has had major impact in multiple areas of biomedical research, first in microcirculator y, inflammation, and angio genesis research, later in cancer biolo gy, immunology, microbiology, de velopmental biolo gy, and neuroscience. F or the purpose of illustrating the principles of IVM, the focus of this chapter will be on w ork perfor med to in vestigate phenomena of inflammation and of the immune system, but other f ields will be included to highlight additional technological or conceptual aspects of IVM.

A SHORT HISTORY OF IVM IVM technolo gy is b y no means a recent de velopment. Some of the scientists using the earl y microscopes of the 17th century were intravital microscopists, among them Marcello Malpighi and Antonie v an Leeuw enhoek, the discoverer of bacteria. Inspired b y the privilege to be the first ones to redisco ver the w orld at a dif ferent spatial scale, they studied and recorded a g reat variety of specimen, ranging from unicellular or ganisms to insects to vertebrates. Their attention must ha ve been par ticularly drawn to the dynamic phenomena of the blood circulation. Malpighi studied small ar teries and v eins in the lungs of live frogs, but it was only in dried lung preparations where he discovered the existence of small channels connecting the ar teries and the v eins.1 It was Leeuwenhoek who, during his numerous intravital observations of blood flo w in the micro vessels of tadpoles, f ish, crabs, rooster combs, rabbit ears, and bat wings, was the first to demonstrate blood flow in these smallest vessels in living animals and thus conf irmed William Har vey’s theor y of the blood circulation, although he did not yet identify capillaries as distinct str uctural units. 2 He e ven constructed various specialized specimen containment apparatus, such as an aquatic chamber for eels (Figure 1), with which he demonstrated the microcirculation of their tailfin to the Russian czar Peter the Great. Leeuwenhoek used onl y a v ery simple microscope, consisting of a single high-magnif ication lens incor porated into a metal plate and scre ws to adjust the relati ve position of the specimen. Ho wever, the technolo gical

177

future w as with the compound microscope, as used b y Robert Hooke.3 Yet, it was only following improvements in glass manufacturing and the development of the achromatic lens in the 1830s that the microscopes produced were of a quality that allo wed English ph ysician and physiologist Marshall Hall to identify capillaries as distinct structural units4 and Hall as well as German pathologist and zoologist Rudolph Wagner to describe the cellular components of the b lood and their dif ferential behavior in the microcirculation of various amphibians and fish5 (see Figure 1). The British neurophysiologist Augustus Waller (the name-gi ver of Wallerian degeneration of peripheral ner ves), after unsuccessful attempts to image the human prepuce, also resorted to cold-blooded animals when first describing the extravasation of blood cells into tissues through inter-endothelial gaps in the frog tongue in vivo and estab lished the identity of the pus cor puscles with the leuk ocytes of the b lood.6 Julius Cohnheim later rediscovered Waller’s w ork but had also independentl y come to the same conclusion, w hich was in stark contrast to the beliefs of his academic teacher Rudolph Virchow, who thought that all extravascular cells were derived from tissue-resident precursors.7,8 Cohnheim thus laid a cornerstone to our cur rent understanding of inflammation through the use of IVM. A subsequent technological breakthrough was the superfusion of membranous tissues and or gan surfaces with temperature-controlled ph ysiological saline, which g reatly impro ved IVM in w arm-blooded animals, such as do gs, cats, rabbits, and guinea pigs, and indicates that ef forts to maintain obser ved tissues in a near-physiological state w ere already reco gnized as critical.9 IVM setups, although still quite bulk y, had already remarkab le similarity to the v arious contraptions used for the same pur pose today (see F igure 1). Using this approach, Bizzozero disco vered thrombocytes as the third cor puscular b lood component and carefully described their contrib ution to thrombosis in the mesentery of the rabbit and guinea pig. 10 One w ay b y w hich IVM studies in w arm-blooded animals w ere usuall y limited w as that the preparations were not stab le for more than a fe w hours, precluding studies of phenomena that occur over the range of days or weeks. This shor tcoming w as g reatly alle viated b y the development of chronic animal windo w chambers through Sandison under the guidance of Eliot and Eleanor Clark.11 The wound-healing processes occurring in these chambers allo wed researchers to monitor angiogenesis in endother ms.12 Glen Algire later adapted the chamber model to the dorsal skinfold of the mouse 13 and together with Harold Chalk ey star ted to implant tumors

178

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

into these chambers to study the angiogenic response of the host, thus initiating the f ield of angio genesis in cancer biology.14 It was probably ophthalmologists who f irst used fluorescence for their mesoscopic obser vations of the human eye, for example, after peroral administration of fluorescein sodium to patients, 15 but soon after the de velopment of the fluorescence microscope at the be ginning of the 20th century, intravital microscopists also took advantage of this new modality (see Figure 1).16 Ellinger and Hirt were aware of the pH dependence in the fluorescence of fluorescein during their studies on glomer ular filtration in the amphibian kidne y, making their studies also the f irst to mak e functional fluorescence-based measurements i n vivo.16 Beyond this, the introduction of fluorescence transfor med IVM in at least two more ways. First, the chemical properties of many of the subsequently used fluorescent intravital dyes allowed specific tagging of anatomic compartments or various tissue components or e ven subcellular compar tments, an approach later extended through their conjugation to other molecules with functional proper ties (such as dextrans or antibodies). Second , the possibility to obtain high quality optical images through epi-illumination extended the range of specimen suitab le for IVM obser vation from translucent tissues to solid or gans. As a consequence, researchers could advance from studying the basic principles of the microcirculation in tissues selected for their practical adaptability to IVM to studying the par ticularities of vascular systems in organs selected for their biological significance. Several other technolo gical de velopments, w hose impact on research was not specific to the practice of IVM, nevertheless generated significant advances in the field. The ability to record and document in an objective manner what an indi vidual researcher sa w under the microscope, using microphotography or microcinemaphotography, provided the oppor tunity for independent inter pretation of observations and their v alidation b y other researchers. While the f irst preser ved microcinemato graphic recordings of a di viding sea urchin embr yo date back to the first decade of the 20th century,17 motion picture recordings of intravital microscopic obser vation did not become popular until the 1940s.18 The recording of dynamic phenomena at re gularly spaced time inter vals also uncoupled the timing of the vie wing of indi vidual frames from the timing of the recording. This permitted the visualization of dynamic processes that are otherwise either too f ast or too slow for our visual perception through the use of slo w motion or time-lapse displa y and literally opened up the temporal dimension as the microscope had opened up the spatial dimension 300 years earlier.

As most other disciplines, IVM research also benefited from the increasing focus on inbred rodents as e xperimental subjects and the oppor tunity to study the molecular mechanisms of phenomena obser ved i n vivo by e xperimental manipulation through the use of b locking monoclonal antibodies or mutant mouse strains. An additional benef it of impro ved recording technolo gy has been the possibility of more in-depth, of f-line analysis of IVM recordings, especiall y due to the no w widespread use of digital storage mediums, w hich has in some areas allowed for some de gree of automation of the otherwise cumbersome and occasionall y bias-prone manual anal ysis of IVM data. As a consequence of improved means of data analysis and quantitation, IVM research, for instance in immunology, has recentl y again found g reater acceptance by researchers outside the circles of specialists. The last re volution to the practice of IVM has been the introduction of nonlinear optical imaging modalities to biological research, in particular multiphoton microscopy (MPM), in 1990 by the group of Watt Webb.19 As discussed belo w, multiphoton intra vital microscop y (MP-IVM) for the f irst time allowed researchers to mak e three-dimensionally resolv ed obser vations of dynamic processes deep within intact tissues in living organisms at microscopic resolution, w hile at the same time minimizing the unw anted effects of light-illumination. Neuroscientists were the first by a wide margin to make use of this powerful new technology for their studies,20,21 followed by developmental biologists,22 cancer biologists,23 physiologists,24 and immunologists.25 Currently, we are still in the ascending slope of the e xponential g rowth phase in se veral of these f ields with re gard to technolo gical development of IVM and its application to biolo gical questions.

SOURCES OF IMAGE CONTRAST In principle, an y for m of light-matter interaction can be used as a source of optical image contrast in IVM. I n practice, utilization of light absor ption, dif fraction, and reflection, as w ell as of fluorescence, has traditionall y dominated, but several new, especially nonlinear, modalities, have started to enter the f ield.

Bright-field Illumination and Related Conventional Techniques The most straightforw ard approach to obtain images of tissues by IVM is by trans- or epi-illumination through a conventional broad-spectrum light source, such as a halogen lamp. Optical contrast is hereb y generated either

Intravital Microscopy

A

B

D

C

E

179

180

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

through the differential absorption or through diffraction of light by various tissue components. Image for mation through absor ption requires the object of interest to contain signif icant concentrations of light-absorbing molecules and to be sur rounded by nonabsorbing tissue elements. Absorption leads to a decrease in amplitude of transmitted light, which becomes visible at the image plane through diminished signal intensity that can be detected by eye or recorded b y camera. Unstained biolo gical tissues are usually poorly absorptive, but one example of a strongly absorbing molecule is hemoglobin, which is contained in red b lood cells, thereb y enab ling the prominent visualization of blood vessels by IVM (Figure 2). Sources of light dif fraction, on the other hand , are abundant in tissues since diffraction is caused by virtually all inhomo geneities in refracti ve inde x, so-called phase gradients, provided by objects that are at a similar spatial scale as the w avelength of light used , such as cell membranes or g ranules. One approach to visualize dif fraction is to e xclude parts of the dif fracted light from the detection pathw ay in an asymmetric f ashion so that ne gative and positive interference of the diffracted light can lead to detectable net changes in light amplitude at the image plane. Differential interference contrast (DIC), oblique illumination, and Hof fman modulation contrast are techniques to achie ve this. Ho wever, even despite the added benefit of their usuall y e xcellent axial resolution, phase gradient techniques have not gained widespread popularity among intra vital microscopists. This ma y in par t be due to the difficulty of integrating the required equipment into IVM setups. DIC lenses, for e xample, are usuall y bulky and have a short working distance, whereas lenses used for IVM are ideally slender and have a long working distance, which facilitates access to surgically exposed tissues. Notable examples are, for instance, the work of Alan Groom who used oblique transillumination of the edges of solid organs, such as the li ver and spleen, with f iberoptic light sources, to obtain high-resolution mor phological images.26,27 More recently, reflected light ob lique transillumination has been achie ved through placement of a tilted mir ror under a translucent specimen and has

been used f or time-lapse i ntravital v isualization o f neutrophil transendothelial and e xtravascular mig ration (see Figure 2).28 A special for m of bright-f ield transillumination can be achieved by epi-illumination of scattering tissue with polarized light. Some of the incident light is scattered multiple times in the tissue and e xits the tissue in a direction opposite to that of the illuminating light. Those rays that are scattered more than 10 times are depolarized and are thus able to pass to a second polarizing filter in the detection light path with orientation or thogonal to the polarization of the illumination light. On their w ay out of the tissue, these depolarized ra ys are attenuated b y structures that absorb at a chosen illumination wavelength, for instance ~550 nm for the visualization of hemo globin, thus g enerating a “ quasi-transillumination” o r b ackillumination image of red b lood cells in b lood vessels on the wide-f ield detector. This type of contrast for mation has been proven to be an effective method to visualize the microcirculation in tissues w here transillumination and use of fluorescence are not possible29 and has even been used to build devices that allow for the visualization of the microcirculation in humans in clinical settings.30 A related approach to obtain nonfluorescent image contrast is confocal reflectance microscopy, which relies on the confocal detection of light reflected b y various tissue components, such as melanin, upon epi-illumination 31 and w hich for instance yields useful image contrast at depths of up to 350 µm in human skin in vivo when using infrared light.32

Fluorescence Despite their utility in generating a rich source of morphological infor mation, the visualization techniques listed above have one critical shor tcoming. The identity of morphological objects cannot be determined with certainty but must al ways be assumed based on inferences from existing knowledge of their shape, size, optical properties, or behavior. Also, molecular events cannot be directly studied due to the limits of optical resolution. This is why fluorescence, and the ability to fluorescentl y tag biolo gical

Figure 1. History of intravital microscopy (IVM). A, Timeline of some important advances made in IVM-based research (grey panels) or technical advances relevant to IVM investigation (white panels). B, Original drawing of one of the first IVM setups, the eel chamber used around 1690 by Antonie Leeuwenhoek (left panel) and a redrawn illustration of the eel microcirculation as observed by van Leeuwenhoek (right panel). C, Original illustrations to IVM studies of the microcirculation of the frog interdigital web published in 1839 by Rudolph Wagner. Top: an individual web at 3 × resolution. Bottom: a postcapillary venule of the interdigital web at 350 × resolution (adapted from Wagner R5). D, An IVM setup from an 1878 publication by Thoma, designed to irrigate the membranous tissues of warm-blooded animals with temperature-controlled saline buffer (Adapted from Thoma R9). E, Intravital micrographs of the kidney from two different eras. Left, from the first publication using fluorescence in IVM in the toad in 192916; right, from a recent publication in 2002 using MP-IVM in the mouse.24 In both cases, epithelial nuclei were stained with an intravenously-injected intravital dye, Trypaflavin on the left, Hoechst 33342 on the right.

Intravital Microscopy

181

A

Conventional bright-field transillumination

B

C

Figure 2. Sources of image contrast. A, Enhanced visualization of extravasated leukocytes through use of phase gradients to generate image contrast. Left, a bright-field transmitted light intravital micrograph of the mouse cremaster muscle, where darker image elements result mostly from light absorption, eg, by hemoglobin in blood vessels. Right, image of a similar preparation obtained by reflected light oblique transillumination, where extravasated leukocytes are more clearly discernable due to their diffractive properties (Adapted from Mempel TR et al 28). B, From left to right, after intravenous injection, FITC-dextran (green) highlights the lumen of blood microvessels in the murine skin, while Rhodamine 6G (red) labels some intravascular leukocytes that roll along the vessel wall, as well as some endothelial cells and adventitial cells of the tissue surrounding the blood vessels. In a mouse lymph node, T and B cell areas are delineated by T (red) and B cells (green) harvested from a donor mouse and injected intravenously after fluorescent labeling with organic fluorochromes. Blood vessels are visualized through injection of a mixture of green- and red-labeled dextrans (adapted from von Andrian UH and Mempel TR116). In mice, which express major histocompatibility class II molecules fused with EGFP, endogenous dendritic cells (green) in lymph nodes can be detected by MP-IVM interacting with adoptively transferred T cells (red). In reactive lymph nodes, macrophages (yellow-white) stand out by virtue of their autofluorescence. Images were obtained by confocal or MP-IVM. Second harmonic signals from collagen fibers are shown in blue. C, Fluorescent labeling with molecular specificity. A fluorescently tagged monoclonal antibody (MECA-79) specific for a carbohydrate epitope modification was injected intravenously to identify the vascular beds in lymph nodes in which endothelial cells express the enzymatic activity required to generate this modification (adapted from M’Rini C et al 49).

182

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

structures and molecules with a high de gree of specificity, has adv anced the possibilities of intravital obser vation tremendously. Given the exponentially growing availability of fluorescent probes for the identif ication of cellular and subcellular structures, it can be assumed that this modality will also for some time in the future be dominating our efforts to understand biology through in vivo imaging, be it IVM or molecular imaging by optical means. Fluorescence is generated b y molecules containing functional chemical g roups that ef ficiently absorb electromagnetic energy through interaction with photons and can, after some of this energy is lost through nonradiative processes, again release most of this ener gy by emitting secondary photons. Due to the nonradiati ve ener gy losses, the emitted photons are, at least with single-photon e xcitation techniques, such as con ventional widefield or confocal fluorescence microscopy, of a longer wavelength than the e xcitation light and can be used to obtain spatial information about the fluorophor. In multiphoton excitation techniques, where the combined energy of se veral lo wer ener gy photons is absorbed b y the fluorophor, the emission is of a shor ter wavelength than the excitation light. Besides location, fluorescence encodes infor mation in the form of intensity (number of emitted photons per unit time), wavelength (of excitation and emitted light), and the lifetime of fluorescence, which makes it suitable for multiplexed for ms of in vestigations. The variety of fluorescent probes applicab le to IVM is enor mous and consists of organic chemical compounds, fluorescent proteins (FPs), and semiconductor nanocrystals as will be discussed below.

Nonlinear Imaging Modalities The success of multiphoton e xcitation fluorescence microscopy has brought high-ener gy, puls ed infrared lasers to man y biol ogical laboratories, including those engaged in the practice of IVM. Ho wever, the high laser peak po wers achie ved with these instr uments (se veral hundred kilowatts) not only permit multiphoton excitation but also make several other nonlinear optical contrast techniques, such as second-har monic generation (SHG), third-harmonic generation (THG) and coherent antiStokes Raman scattering (CARS) microscopy possible. The principle of SHG is dif ficult to comprehend on an intuitive basis but can be lik ened to the phenomenon of har monic overtones produced b y vibrating strings in musical instruments (Irving Bigio, personal communication). It is the exact frequency doubling that occurs when high-energy light strik es matter with cer tain str uctural features such as re gularity, noncentrosymmetr y, and

nanometer scale. 33 These conditions are fulf illed for a variety of biological macromolecules. One of these, which is abundant in tissues and has strong SHG properties, is collagen. 34 Intravital microscopists of several disciplines ha ve star ted to mak e use of this endo genous source of contrast to obtain a str uctural conte xt for the observation of dynamic cellular events35,36 or to study the dynamic alterations of collagen str ucture itself, for instance during anticancer therapy.37 THG is frequency tripling of light through interaction with refractive index inhomogeneities as found at cellular membranes. It has not been used in IVM studies y et, but Debarre and colleagues 38 have used it together with SHG and multiphoton-excited autofluorescence to image fresh rat liver explants. Curiously, they found that the strongest signals were generated by lipid bodies contained in hepatocytes. Unfortunately, the laser powers required to induce THG are even higher than those used in multiphoton excitation or SHG, w hich ma y mak e it impractical for the continuous recordings in IVM due to the in this case undesirab le simultaneous tw o- and three-photon absorption of endogenous or exogenous fluorophores.38 CARS microscop y, f inally, allo ws one to appl y the spectroscopic specif icity of Raman imaging in order to visualize the chemical constituents of a tissue by a nonlinear optical process. 39,40 Recently, this modality has been applied to IVM of the mouse ear and has enab led the distinction of various lipid species that dominate different fatcontaining compar tments of this or gan.41 CARS imaging requires significant hardware upgrades of today’s standard multiphoton microscopes, but the f act that spectroscopic imaging of protein or DN A appears to be within reach 41 may mak e such an in vestment w orthwhile for intra vital microscopists with related biological interests. It should be mentioned that all three of the listed nonlinear nonfluorescence optical processes generate signals in the forward direction of the illumination light beam. What makes them applicable to imaging in epi-detection mode, as required for IVM studies in solid organs, is that the forw ard-generated signal is par tially re versed in direction by multiple scattering events in turbid biological tissues. Since signal for mation, in principle (as in MPM), occurs onl y in the objecti ve’s focal point, these backscattered photons can be used for image for mation.

FLUORESCENT PROBES The property of fluorescence is commonl y linked to the presence of multiple aromatic g roups in a molecule, allowing for the for mation of lar ge conjugated electron systems. The number of carbon-carbon doub le bonds is

Intravital Microscopy

inversely correlated with the energy required to excite the molecule and thus positively correlated with its excitation and emission w avelength. This proper ty is shared b y a large number of or ganic compounds and proteins (see Chapters 27, “Optical Imaging Agents” and Chapter 48, “Fluorescence Readouts of Biochemistr y in Li ve Cells and Organisms”). A distinct mechanism provides fluorescence to semiconductor cr ystals,42 which w ere recentl y added to the palette of probes used for IVM.

Intravital Stains Many membrane-permeable fluorescent dyes have chemical properties that cause their accumulation in specific subcellular compar tments. Among these are acridine orange, w hich accumulates in cell nuclei, and the mitochondrial stain rhodamine 6G (see F igure 2). Both of these dyes have been used to label blood leukocytes in situ by simple intra venous bolus injection of the reagent into live animals and subsequent visualization of leuk ocyte behavior in the microcirculation by fluorescence IVM.43,44 The convenience of use of b lue fluorescent dyes in MPM has also repopularized another highl y dif fusible nuclear stain, Hoechst 33342, as an intra vital dy e to label nuclei of tissue-resident cells. 24 Propidium iodide, a nuclear dy e that onl y permeates membranes of apoptotic or necrotic cells, has been injected intra venously together with Hoechst 33342 into mice to monitor the viability of tubular epithelial cells b y MP-IVM in the kidney after ischemia and reperfusion injur y.24 Recently, probes have emerged that even allow for the specif ic visualization of tissues af fected by certain disease processes. Brad Hyman’s group has for instance validated various reagents that after intra venous injection selectively target plaques in the brains of mice suffering from a condition mimicking Alzheimer’s disease in humans. 45 Miller and colleagues ha ve used the fluorescein-based chemical 5-(and-6)-carbo xyfluorescein diacetate, succinimidyl ester (CFDA, SE, commonly simply termed CFSE) to tag antigen-presenting dendritic cells (DCs) in mice in vivo b y subcutaneousl y injecting a mixture of a model antigen, CFSE, and aluminum h ydroxide to for m a long-lasting local depot. The normally nonfluorescent, cellpermeable CFSE is g radually released from this depot and after local uptake by DC together with the antigen rendered fluorescent and char ged and thus cell imper meable by the action of intracellular esterases, in addition to binding to cellular amine groups. Migratory DC that eventually reach the draining lymph node (LN) can thus be imaged by MPM of explanted LNs. 46 This elegant approach ma y have been

183

inspired b y t he pr actice o f i ntravital m icroscopists i n neurobiology to load neurons of the central nervous (CNS) system with fluorescent dy es in vi vo through direct intracellular injection via micropipettes. 21

Functionalized Fluorochromes A second modality of assigning specificity to a fluorescent label for IVM is to attach it to a particle or macromolecule with certain biological properties or to use it to label cell lines or ex vivo purified primary cell populations for subsequent injection into animals. Classical e xamples of macromolecules tagged with fluorochromes to track their behavior after application to a live specimen are dextrans of various molecular weights and albumin to highlight either b lood plasma or l ymph compartments after intra venous or interstitial injection (see F igure 2). 47,48 Intravenous injection of conjugated molecules or unconjugated dyes can also be used to quantitatively measure vascular permeability.47 An e xtension of this principle is to use the binding specificity of antibodies to tar get fluorochromes to specific antigenic epitopes in li ve animals. Ulrich v on Andrian’s group has visualized carbohydrate moieties displayed on the luminal surf ace of specialized microvessels in lymph nodes that mediate the tethering of lymphocytes to the endothelium, which initiates their recruitment from the bloodstream into the tissue (see Figure 2). By simultaneous visualization of the b lood-borne lymphocytes, they could directly correlate the molecular endothelial staining with the biolo gical beha vior of immune cells. 49 Sipkins and colleagues 50 similarly characterized the recr uitment sites of leukemic cells to the bone mar row. One caveat of this approach is that the Fc portion of antibodies also possesses considerab le binding af finity to its receptors. Hauser and colleagues 51 therefore generated F ab fragments of an anti-CD21 antibody through papain digestion to label follicular DCs in l ymph nodes draining the subcutaneous injection site of their reagent. A modality of molecular imaging is the use of fluorochromes conjugated to molecules that confer them with specificity for tissues in dif ferent disease states either by virtue of their specific retention or by their alteration and activation through enzymatic or other molecular activities. It seems lik ely that, on the one hand , intravital microscopists will star t to use no vel tools generated in this f ield. One the other hand , molecular imaging might benefit from contributions b y IVM to their preclinical validation. In analogy to molecules, cells are also suitab le vehicles for fluorescence, allowing for the visualization of their

184

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

behavior by IVM (see F igure 2). In f act, the bur geoning field of immunoimaging b y IVM often relies on adopti ve transfer of fluorescentl y labeled immune cells, as will be discussed later in this chapter. Cell labeling mostl y relies on intracellular retention proper ties of dy es, as described earlier for CFSE, which makes this dye a useful reagent for the purpose of long-term tracking of adoptively transferred cells52; today, a wide spectrum of dyes covering the entire visible spectrum are available. Lastly, packaging fluorescence in for m of par ticles, such as microspheres or beads a vailable in sizes from a few tens of nanometers to tens of micrometers, pro vides a mean to target them to cells with phagocytic properties, such as macrophages of the spleen. 53 The same par ticle characteristics ha ve also hampered the use of quantum dots, nanometer -size semiconductor cr ystals with v ery favorable fluorescent proper ties,42 as mark ers for specific targeting reagents, and their use in IVM has been limited.54–56

Fluorescent Proteins The green fluorescent protein (GFP) was discovered more than 30 years ago,57–59 cloned in 1992,60 and first used as a marker for gene expression in 1994.61 Since then, GFP and its multicolored variants have transformed all areas of biological research b y ser ving as intracellular fluorescent reporters for visualization-based specimen anal ysis after introduction of recombinant DN A. It has also changed IVM by allowing for the nonin vasive fluorescent tagging of cell lineages or subcellular compartments (see Figure 2) or for reporters of transcriptional activity in whole animals through transgenesis or gene-tar geting approaches and through ballistic (“gene gun”), viral, electroporation, 62 or other ways of exogenous gene-delivery in vivo. Early versions of FPs, such as GFP derived from Aequorea Victoria or DsRed from Discosoma sp, were limited in their utility due to moderate fluorescence intensity, slow maturation, or the tendency to multimerize in cells. The latter proper ty for instance prohibited the transgenic expression of DsRed in rodents due to toxicity during embr yogenesis w hile slo w maturation (and thus acquisition of fluorescence) lo wers its utility as a transcriptional repor ter. Extensive mutagenesis of the wildtype proteins, ho wever, has b y no w not onl y lar gely overcome these limitations but has also g reatly extended the palette of available colors.63,64 In addition to their use to nonin vasively label endogenous cells in li ving e xperimental specimen, FPs are also adv antageous for the long-ter m tracking of rapidly proliferating, adoptively transferred tumor cells 65

or immune cells,66 as well as bacteria67,68 or other cellular pathogens b y IVM. Fluorescence le vels are maintained throughout cell di vision, w hereas or ganic cell track er dyes w ould be diluted in each generation of daughter cells, quickly rendering them nondetectable by IVM. Besides ser ving as stab ly expressed mark ers to tag cells for identif ication, FPs with shor t maturation times will be useful as in vivo reporters of transcriptional activity. Lastl y, the a vailability of s everal FP pairs (e g, BFPCFP, CFP-YFP , or GFP- RFP) with susceptibility for Forster resonance ener gy transfer (FRET) upon spatial proximity opens up the oppor tunity to use their fusions with probes for various molecular activities as molecular sensors (see Miyawaki63,69 for reviews).

Autofluorescence Some biological tissues and cells stand out in their content of v arious organic molecules with autofluorescent properties, such as serotonin, fla vins, or retinol, w hich can therefore be a useful source of contrast in fluorescence microscop y.70 The redo x sensiti vity of the coenzyme nicotinamide adenine dinucleotide (NAD+/NADH) has allo wed neurobiolo gists to study metabolic activity in the brain of living mice at cellular resolution.71 Cellular N AD+ along with N ADP+ can also be used to visualize macrophages b y MPM in explanted tumor tissue72 or in lymph nodes in vivo (see Figure 2). Since the N AD+/NADP+ redox state is also variable in these cells, their inducible changes in autofluorescence73 might in the future also provide a means to monitor their cellular activation status by IVM.

IVM INSTRUMENTATION Bright-field Microscopy Conventional IVM in its simplest for m uses brightfield transillumination of the specimen through a condenser on either an in verted or , more commonl y, an upright microscope. A con ventional halo gen lamp serves as a source of light that is focused through a collecting lens into the front aper ture of the condenser . This requires special long working distance condensers but allo ws for even illumination of the microscopic field for optimum image contrast and resolution (so-called Köhler illumination). The transmitted light is collected through long w orking distance, usuall y water-immersion, objecti ve-lenses, and detected on a charge-coupled device (CCD) camera for recording on video tape or , more and more frequentl y, on digital

Intravital Microscopy

storage media. Although primary digital data recording makes subsequent transfor mation and displa y of the data much more con venient, the often impracticab ly large data sets resulting from several minutes of recording at video-rate (25 or 30 images per second) still today make analog technology worthwhile in some situations. Although bright-f ield IVM generates a w ealth of morphological infor mation, its utility is limited to translucent tissues, and quantitati ve data can onl y be obtained in a few situations where the object under study is unambiguousl y identif ied b y mor phological criteria. This is for instance the case for leuk ocytes that tether , roll, and adhere to the luminal side of blood v essel endothelia and become easil y discer nable from other blood components as dif fractive spheres. 74 In other cases, the visualization of suf ficient mor phological detail can occasionall y be achie ved also outside b lood vessels through contrast-enhancing optic techniques 26,28 or through additional use of color as a source of contrast.75,76

Wide-field Fluorescence Microscopy For the reasons detailed earlier , the range of applications of con ventional IVM can be signif icantly e xtended through the use of fluorescence. Due to the lo w quantum yield of most fluorochromes, this requires the use of more powerful light sources, such as mercur y or x enon lamps. Dichroic mir rors and optical f ilters nar row do wn the illumination light to a spectr um suitable for e xcitation of the fluorochromes of interest w hile excluding light of the wavelengths of the e xpected flu orescence emission. Unlike with most bright-f ield techniques, the specimen is illuminated through the same objecti ve lens that also collects the emitted signals. Emitted light is typicall y f iltered to exclude excitation light and to select for specif ic fluorescence emission and then detected b y a monochrome camera. Color cameras allow for the simultaneous recording of the emission of multiple fluorochromes, bu t they ha ve ne ver gained popularity in fluorescence IVM because the y are either less sensiti ve, slo wer, or simpl y more expensive than monochrome cameras. The high light intensities required in fluorescence IVM unfortunately also command consideration of phototoxic effects in every experimental investigation, especially if continuous obser vation is emplo yed.77 One approach to reduce the risk of inducing ar tifacts through phototoxicity is to use highl y sensitive, for instance, silicone-intensified target tube or, more recently, back-illuminated electron-multipl ying CCD (EM-CCD) cameras

185

in combination with schemes to limit light exposure, such as stroboscopic illumination. Video-triggered xenon arc stroboscopes can pro vide light flashes of one microsecond duration at rates of 30/s, w hich reduces light e xposure of the specimen by a factor of 3 × 104 compared with continuous illumination.74,78 The strength of wide-f ield e xcitation fluores cence microscopy compared with the techniques discussed below is its high speed at full frame, allowing for data acquisition at video-rate or faster. Its disadvantages are the impracticality of multichannel recordings and its poor axial resolution. Nonconfocal optical sectioning techniques such as deconvolution79 or str uctured illumination 80 to impro ve axial resolution have not yet found entr y into the practice of IVM.

Laser-Scanning Microscopy Molecular and cellular mo vement in li ving specimen occurs in all three dimensions (3Ds), and acquisition of two-dimensional image data is therefore limiting in the study of solid tissues. Various modalities of laser-scanning microscop y can o vercome this prob lem. Lasers provide near -monochromatic, coherent light that can be focused to excite fluorescence in a diffraction limited spot within the focal plane of the objecti ve lens. The emitted light is usually detected by photomultiplier tubes (PMTs), and the spatial infor mation of the signal is preserved through synchronization of image re gistration with point-by-point scanning of the sample. Images are digitally rendered from the individual pixel data acquired during the laser scan. During single-photon excitation, sample fluorescence is also generated within the cones of illumination light abo ve and belo w the focal plane. To achieve confocal detection, this fraction of the fluorescence emission, along with signals fla wed b y sampleinduced light scattering in the e xcitation or in the emission path, is excluded by use of a pinhole aperture in an intermediate image plane before it reaches the d etector. During multiphoton e xcitation, as discussed belo w, extraction of full spatial infor mation in 3D is already achieved through selecti ve excitation of fluorescence in the focal plane, w hich obviates the need for descanning of the emitted light or use of a confocal pinhole aperture. Due to the sequential re gistration of image information, laser -scanning de vices are intrinsicall y inferior to wide-field excitation technology in terms of acquisition speed. Using the most frequently used galvanometer-driven xy mir ror scan-head, recording of a single full-frame (for e xample, 512 × 512 pixels) at a pixel dwell time of a fe w microseconds requires close

186

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

to a second. Various ways exist to improve scan speed, such as the use of rotating polygonic81 or resonant scan mirrors,82,83 acousto-optic deflectors, 84 or lens ar rays to scan the specimen with multiple beams. 85,86 All of these allow confocal image acquisition at video-rate or faster. Confocal microscop y has been implemented in IVM,44,87–89 but its use has not gained widespread popularity. This may be related to the limitation of optical penetration depth (typicall y less than 100 µm), w hich is greatly improved in MPM. Confocal microscopy may, however, remain the method of choice in tissues w here high concentrations of pigments or other sources of autofluorescence make multiphoton excitation problematic, such as the skin.90–92

MPM The near-simultaneous interaction of two or more lowenergy photons with a fluorophor can induce its excited state, and subsequentl y fluorescence emission, similar to w hat is achie ved with one indi vidual high-ener gy photon. This phenomenon is put to use in MPM w here conditions conducive to multiphoton-fluorophor interactions are generated b y condensing the infrared light from femtosecond-pulsed lasers in the focal point of high-numerical aperture objectives.19,93 Since photon density f alls of f quadraticall y with distance from the focal point and the efficiency of multiphoton excitation depends on the square of the photon density (for twophoton e xcitation), fluorescence f alls of f as a quar tic function outside the focal point of the microscope objective. F or practical pur poses, all e xcited fluorescence therefore originates from the focal point and can be used for image for mation, even the fraction of scattered emission, w hich w ould be e xcluded in confocal microscopy and w hich increases with increasing depth in the sample. The latter fact accounts for the high signal-to-noise ratios achie ved in MPM, especiall y in greater tissue depths. In addition, limiting fluorescence excitation to the focal point prevents significant energy absorption outside the focal plane and thus limits photo-bleaching and phototo xic ef fects. Fur thermore, the low scattering coefficient of the infrared e xcitation light typicall y used allo ws for focusing of the laser beam deep in turbid tissues. As a result, imaging depths off up to 1 mm ha ve been achie ved in the mouse brain.94 Finally, the broader and frequently overlapping two-photon cross sections of most fluorochromes (compared with their single-photon absorption spectra) make it possib le to simultaneousl y visualize multiple

fluorescent labels. This includes b lue fluorophores whose single-photon excitation otherwise requires special ultra violet light-transmissi ve optics that are not conducive to simultaneous imaging of red and f ar-red light. All of these qualities to gether mak e MPM the currently most po werful technolo gy for threedimensionally resolv ed fluorescence IVM studies in solid tissues of li ving specimen, and rapidl y ongoing technological developments guarantee that the range of possibilities will increase in the future.

Signal Detection Minimalization of light exposure should be a goal of all IVM methodology, and optimization of light detection is one way to contribute to this goal. For both CCD cameras and PMTs, ne w technolo gy is a vailable to mak e light detection more sensiti ve. Back-illuminated EM-CCD will lik ely ha ve utility for wide-f ield fluorescence IVM techniques, w hile Gallium arsenide phosphide (GaAsP)-based PMTs, w hich are more than twice as quantum-efficient as conventional bialkali or multialkali photocathodes, show promise for confocal and MPM. 95 The use of the GaAsP detectors however is still limited by their low damage threshold and their small photosensitive area (nondescanned detection of scattered emission benefits from larger detection areas). As a general r ule, CCD cameras are used in IVM to record bright-field and fluorescence wide-field microscopy images, while PMTs serve to acquire data in laser-scanning microscopes. Ho wever, a fe w e xceptions appl y. F or instance, when the specimen is scanned b y multiple beamlets generated through lens ar rays (Bewersdorf and Br uist, 1998) or a series of 50:50 beam-di viders,96,97 the spatial image information from the multitude of focal points must be preser ved through wide-f ield detection b y CCD cameras. This approach has the adv antage of faster frame rates and of the higher quantum ef ficiency of CCDs compared with PMTs. Multifocal illumination is thus an alternative to resonant scanners or AODs to achie ve video-rate image acquisition using multiphoton e xcitation85,86,96 for IVM. 97 However, the requirement for wide-f ield detection eliminates the ability to collect scattered emission light, which is the key advantage of multiphoton excitation for deep tissue imaging. Since the benef it of deeper tissue penetration of infrared light and of restricting phototo xicity and photobleaching to the focal plane remain, multifocal MPM ma y nevertheless be useful for IVM applications requiring video-rate. The gain in speed may for instance prove beneficial in the anal ysis of fluorescence lifetimes. 98 Low data sampling efficiency makes this latter imaging modality,

Intravital Microscopy

which otherwise has high potential for utility in IVM investigations, dramatically slower than fluorescence intensity measurements. Multifocal multiphoton e xcitation combined with CCD-based detection may offset this shortcoming and make the fluorescence lifetime-based observation of the dynamic events typically encountered in vivo practical, while the PMT-based detection of fluorescent lifetimes currently remains a challenge.

IVM MODELS The list of animals that have been subjected to IVM investigation is long. Cold-b looded species, such as fro g, toad, and eel, are cur rently more of historic interest, and lar ge warm-blooded v ertebrates, such as cats, do gs, rabbits, rooster, or turtles, have also lost popularity in comparison to small rodent species, which today dominate the work of most intra vital microscopists (T able 1). 74 On the other hand, a few new species have been adopted for use in IVM. Although the ph ysiology of fr uit flies, zebra f ish, or the nematode caenorhabditis ele gans is more dif ferent from ours than that of rodents, their fast-paced ontogenesis, ease of genetic manipulation, and f avorable optical characteristics have also made them attracti ve subjects for the IVMbased studies of de velopmental biologists.20,99–102 For the purpose of this chapter , several general aspects of animal preparation for IVM will be reviewed and a fe w selected rodent IVM models will be discussed in order to illustrate individual technical challenges in their design. Every obser vation by IVM requires a violation of the subject’s physical integrity, which harbors the risk of confounding the results of the in vestigation and antagonizes the ideal of IVM, the obser vation of biological processes in a tr uly physiological environment. Even the most noninvasive approaches, such as IVM of the eye or the skin, demand either some means of specimen immobilization, such as anesthesia, or optical coupling of an objective lens to the tissue with immersion medium, w hich may change the ph ysiological tissue temperature, or simpl y illumination at light energies that are not nor mally encountered at this site. The approach to tackle this problem is to take any available measures to minimize in vasiveness and to k eep in mind the v arious possibilities of inducing ar tifacts, putting controls in place wherever possible.

Anesthesia Some IVM models allow, with limitations, imaging in the awake animal. Examples are the dorsal skinfold chamber in hamsters, which can be calmed by positioning them in

187

tubes that mimic their natural ca vernous habitats, thus immobilizing them sufficiently for video-rate imaging.103 Studies in neurobiolo gy occasionally demand conscious animals, and MP-IVM in the brain of nonanesthetized freely moving rat has been achie ved through f iber-optic coupling of the microscope to a chronic brain windo w preparation.104 Generally, ho wever, IVM studies are conducted under general anesthesia of the animal for the pur pose of immobilization and to per mit surgical exteriorization of the tissue of interest. Yet, all available anesthetic reagents have not onl y neurolo gical but also cardio vascular side effects,105 and the choice of anesthetic is dictated b y the selected animal species and the anticipated minimal interference of its side effect with the experimental observations. Especiall y, studies in w hich micro vascular dynamics play a role should ideally use some form of cardiovascular monitoring to guide the narcotic re gimen. Finally, anesthetic agents may also exert uncharacterized effects o n v arious o ther p hysiological f unctions, f or instance on inflammatory processes.106,107

Surgery Except for studies of the skin and e ye, and future endoscopic approaches, IVM requires sur gical preparation of tissues to mak e them accessib le to the microscope optics, either through acute preparations or through the installation of chronic animal windows. The principle of minimal invasiveness hereby dictates the use of the least traumatic and, if feasible, aseptic technique. Membranous tissues and surfaces of inner or gans should be constantl y kept moist with appropriate buf fers that are free of contamination, for example, with bacterial to xins. The rodent cremaster muscle preparation, for instance, requires a buffer with defined pH and ionic strength since otherwise muscle fasciculations antagonize imaging efforts. The exposed brain, on the other hand, should be ir rigated with ar tificial cerebrospinal fluid for best results. Dr ying of tissues rapidl y causes microvascular dysfunction and often times ir reversible deterioration of optical transparenc y. The main concer ns with chronic animal windo w preparations are infection and e xcessive wound healing reactions, w hich both will confound not only microvascular studies but also the investigation of tumor biology or of immunological phenomena. Despite utmost care, some de gree of local tissue trauma during specimen preparation is una voidable, and immediate effects, such as acute local or e ven systemic inflammation, need to be tak en into consideration w hen interpreting the results of IVM studies. The opened cremaster muscle model, for instance, requires a comparably

188

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 1. SELECTED ORGANS/TISSUES USED FOR INTRAVITAL MICROSCOPY STUDIES IN MICE Organ/Tissue

Field of Study

Microscopy Technique

Reference

Microcirculation; leukocyte migration

BF, WF, MP

Mazo and colleagues,115 Cavanagh and colleagues155

Microcirculation

WF

Carvalho-Tavares and colleagues156

Brain parenchyma

Neuronal differentiation; neuronal activity; leukocyte migration

MP

Svoboda and colleagues,21 Davalos and colleagues157

Spinal cord meninges

Microcirculation

WF

Vajkoczy and colleagues158

Microcirculation; leukocyte migration

BF, OT, WF

Mempel and colleagues,28 Hickey and colleagues,76 Baez108

Microcirculation

BF, OP, WF

Algire13

Microcirculation; tumor/tissue growth

BF, OP, WF, MP

Algire and Chalkey14

Microcirculation; leukocyte migration

RE, WF

Becker and colleagues159

Microcirculation

WF, CF

Miyamoto and colleagues160

Kidney

Microcirculation, physiological function

BF, MP

Dunn and colleagues,24 Buhrle and colleagues161

Knee joint (fat body)

Microcirculation

WF

Veihelmann and colleagues162

Liver

Microcirculation; leukocyte migration; host-pathogen interaction

BF, OT, WF, CF

Geissmann and colleagues,113 Frevert and colleagues,152 McCuskey163

Lung

Microcirculation

WF

Tabuchi and colleagues112

Microcirculation; leukocyte migration

WF, MP

von Andrian78

Popliteal LN

Microcirculation; leukocyte migration

MP

Mempel and colleagues35

Mesenteric LN

Microcirculation

WF, CF

Grayson and colleagues89

Mesentery

Microcirculation; leukocyte migration; lymph flow

BF, WF

Bienvenu and colleagues,75 Dixon and colleagues,144 Atherton and Born164

Pancreas

Microcirculation

BF, OT

Covell,165 McCuskey and Chapman166

Peyer’s patch

Microcirculation

WF

Bjerknes and colleagues167

Skin

Microcirculation; leukocyte migration

WF, CF

Kissenpfennig and colleagues,90 Nishibu and colleagues,91 Eriksson and colleagues168

Skin transplants

Microcirculation

MP

Chakraverty and colleagues169

Small intestine

Microcirculation, host-pathogen interaction

WF

Chieppa and colleagues,67 Massberg and colleagues170

Large intestine

Microcirculation

WF

Soriano and colleagues171

Spleen

Microcirculation

BF, OT, WF, CF

Grayson and colleagues,88 McCuskey and colleagues172

Tail skin

Lymph flow

WF

Leu and colleagues142

Bone marrow CNS Brain meninges

Cremaster muscle

Dorsal skinfold chamber Striated muscle Tumor/tissue implants Eye Iris Retina

Lymph node Inguinal LN

BF = bright-field; CF = confocal fluorescence; MP = multiphoton fluorescence; OPS = orthogonal polarization spectroscopy; OT = oblique transillumination; RE = reflectance; WF = wide-field fluorescence.

Intravital Microscopy

invasive procedure and considerable tissue trauma, which correlates with the skills and experience of the investigator, is ine vitable.108 In f act, for the no vice in the use of this model, it is impor tant to note that the baseline of measured parameters of inflammation, such as leukocyteendothelium interactions or micro vascular per meability, will change with increasing sur gical prof iciency and decreasing duration of the procedure. To mitigate the effects of investigator variability on the results of studies on inflammation, researchers can standardize the de gree of inflammation by preinjecting the scrotum with inflammatory mediators, such as TNFα or IL-1β, and beginning the recording of micro vascular parameters at def ined time points after this stimulus. 109

Immobilization Under the microscope, tissue mo vements o ver a fe w micrometers, caused by respiratory excursion of the chest cavity, cardiac motion, or the pulse of a local ar terial blood vessel, translate into imaging ar tifacts that can in the best case mak e visual anal ysis of the recording cumbersome and in the w orst case pre vent useful image acquisition. Video-rate imaging is hereby generally much less profoundl y af fected than 3D laser -scanning techniques, in which collection of an individual frame in the form of a stack of optical sections usuall y tak es on the order of a few seconds to minutes. The lung and the hear t pose the g reatest challenges in terms of overcoming specimen motion. At least for videorate studies of the hear t b y IVM, electrocardio gramtriggered image acquisition has been successfull y used to circumvent the necessity of immobilization.110 Lung movements, on the other hand, might be too ir regular for a similar approach, but electrom yography of intercostals muscle or the diaphragm could pro vide a useful synchronization trigger pulse. So f ar, pulmonar y physiologists have relied on the immobilizing ef fect of inter mittent inter ruption of mechanical v entilation or of adhering the obser ved lung area to a chest windo w using the ne gative pressures physiologically present in the intrapleural space. 111,112 Other or gans and tissues are also af fected b y respiratory and cardiac mo vements to a de gree inversely proportional to their distance from the chest cavity. In the case of li ver, spleen, and kidne ys, these mo vements can be par tially alle viated b y using microscopes with an inverted geometry so that the weight of the animal resting on the organ contributes to immobilization.24,88,113 Organs more distal to the tr unk are more suitab le for high-resolution 3D microscopy but still require efforts to eliminate residual micrometer-scale movements for optimal results.

189

The popliteal lymph node situated in the back of the knee can for instance be shielded from respirator y motion through the application of mechanical f ixtures to bon y pivots of the spine and the hip, 35,74 while imaging of the brain or the skull bone mar row benef its from the use of stereotactic holders.114,115

Tissue Homeostasis Apart from the inflammator y reaction to the sur gical preparation, attention should also be paid to other aspects of tissue homeostasis, depending on the biolo gical question addressed and the organ under study. Control of baseline macrocirculator y and microcirculator y parameters such as blood flow velocity is a prerequisite for an y studies on vascular biology or leukocyte-endothelium interactions, but also e xtravascular events noticeably depend on intact tissue perfusion. The interstitial mig ration of lymphocytes in l ymph nodes ceases within seconds w hen arterial blood flow is interrupted (TRM, unpublished observations). Although imaging studies in explanted lymphoid organs superfused with oxygenated buffer have so f ar yielded similar results with re gard to l ymphocyte migration and interaction as intra vital studies, it is conceivable that at some le vel intact blood flow, lymph flow, and inner vation ha ve an impact on immunolo gical processes as evidenced by alterations in the microanatomy of lymph nodes deprived of afferent lymph flow (reviewed in von Andrian and Mempel 116). Most biolo gical processes are to v arying de grees temperature dependent. F or instance, Miller and colleagues117 have found that the migration speed of naive T cells in explanted lymph nodes is maximal in the range of 36ºC but steepl y decreases at belo w 32ºC or above 42ºC. Possible causes of subphysiological temperatures in vi vo can be an anesthesia-related decrease in core body temperature or peripheral hypoperfusion. More significantly, the use of water immersion turns the objective lens into a heat sink for the tissue under obser vation. Aside from k eeping the animal core temperature in the physiological range, local temperature should therefore ideally be controlled through the use of objective heaters, heat lamps, or local sources of heat in direct contact with the specimen, such as ther mal putty or heating coils.

Pitfalls Beyond f ailure of the sur gical preparation or the anesthetic regimen, there are a fe w less ob vious f actors that can negatively affect the experimental outcome or lead to the unnoticed introduction of ar tifacts.

190

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Effects of the, in most cases unnatural, light exposure of the tissue under study are potentiated through the presence of light absorbers, such as natural fluorochromes in some tissues, genetically encoded probes, or exogenously introduced organic fluorochromes. Well-perfused tissues efficiently dissipate heat through convection, but fluorescence e xcitation also leads to increased generation of reactive oxygen species in biolo gical tissues, and ef fects of continuous light exposure on cellular behavior are well documented.77 Some fluorescent dy es also ha ve intrinsic to xicity, Rhodamine 6G, for instance, is an inhibitor of o xidative phosphorylation,118 in addition to being mutagenic at higher doses, 119 and even some FPs can be to xic to cells due to their agg regation tendencies. FPs can also be problematic in long-ter m adopti ve transfer studies of tumor cells or immune cells due their potential immuno genicity. Some inbred mouse strains seem to be more af fected by this than others. 120 As with any scientific experiments, care must also be applied to the approaches and methodolo gies used to extract data from image material and ho w to process, analyze, and interpret the results. Some considerations are reviewed by Mempel and colleagues. 74

APPLICATIONS OF IVM Intravital microscopic study has been used in numerous fields of biomedical research. The following is a selection of areas of study , w here IVM in vestigations ha ve contributed to our understanding of biological processes, and which will serve to illustrate the range of possibilities of this methodology.

Cellular Migration and Interaction One of the classical domains of IVM is the study of the adhesive interactions between cells in the bloodstream and the endothelium of microvascular beds that are conducive to cellular recr uitment to tissues. This process usuall y occurs as a sequence of molecular interactions, and the specificity of recr uitment of the appropriate cell type to the right tissue at the right time is f acilitated through the regulated expression of signaling and adhesion receptors and their ligands on both blood and endothelial cells.121,122 Sequential molecular interactions are reflected b y different behavioral patterns of blood cells, such as initial tethering to the vessel wall, rolling, firm adhesion, spreading, crawling, and endothelial diapedesis. These steps are most characteristic for cells of the hematopoietic system, which

migrate to lymphoid and peripheral tissues as pro genitors cells115 or as mature dif ferentiated cells of the immune system.74 Similar or abbreviated recruitment cascades are observed for platelets in studies that in vestigate their role in atherosclerosis 123 or thrombus for mation,124 as well as for tumor cells in settings that mimic aspects of metastasis formation.125 While the bioph ysical conditions within blood v essels, such as the shear rate, can be par tially recreated in in vitro flow chamber systems, the special differentiation state of any endothelium is rapidly lost in culture in absence of the b lood-borne and tissue-deri ved factors. In situ study b y IVM is thus essential and has made critical contributions to the clarif ication of the molecular specif icities. The pace of cellular recr uitment to tissues is such that imaging at video-rate (or be yond) is usually most appropriate to resolv e the details of the process. The earliest studies that demonstrated the significance of specific molecular events instead of merely mechanical factors for leuk ocyte e xtravasation in vestigated the behavior of endo genous neutrophil g ranulocytes of the experimental animal using bright-f ield IVM in rodent mesenteries subjected to inflammator y s timuli.126,127 Fluorescence microscop y later allo wed for the specif ic identification of adoptively transferred, purified leukocyte subpopulations that could be either manipulated ex vivo or obtained from genetically altered donor animals in the microcirculation of recipient animals. 128 Tissues that are most amenable to this type of in vestigation are characterized b y v essel geometries that allo w visualization of sufficiently long vessel segments in the focal plane of the microscope. The poor axial resolution of con ventional bright-field and wide-f ield fluorescence microscop y is in fact of benefit in this context because it extends the depth of the microscopic field and thus the number of leukocytes passing a vessel that can be recorded. Once cells have arrested and begin to migrate on the luminal endothelium, their speed of locomotion decreases by two to three orders of magnitude, while their movement directionality transitions from nearl y onedimensional trajectories to 3D migration patterns.92,129 Intravital observation of the slow movements of any kind of cell during intralumenal crawling or during interstitial migration (which typically occurs at instantaneous velocities in the range of 1–30 µm/min) following diapedesis is more practical with time-lapse recordings and g reatly facilitated b y the 3D resolving po wer of confocal 92 or multiphoton microscopes.25 Although the high mig ratory speeds of leuk ocytes were anticipated based on in vitro studies and earlier observations of neutrophils in inflamed rodent mesentery

Intravital Microscopy

and cremaster muscle, the relentless and seemingl y random motility of l ymphocytes in secondar y lymphoid organs surprised most immunologists, but helped explain how rare antigen specif ic T and B cells can f ind the specialized cells presenting their co gnate antigen in the v ast volumes of secondar y lymphoid organs for the initiation of an immune response. 25,35,46 During their e xtravascular e xcursions, leuk ocytes physically interact and e xchange signals not onl y with other immune cells, but also with man y other tissue residents. In the case of ef fector CD8 T cells, cellular encounters with antigen-presenting tar get cells, such as tumor cells or virall y infected cells of the parench yma, triggers c ytolytic function (F igure 3). Other interactions induce developmental or behavioral modifications, such as the acti vation, proliferation, and ef fector dif ferentiation of T cells upon encounter with antigen-presenting DCs35,46,130,131 or the migration of tumor cells upon interaction with tumor-associated macrophages.132 The description of the in vi vo dynamics and r ules of cellular engagement by IVM ha ve contributed a g reat deal to our understanding of these processes.

Cellular Signaling The dynamics of cellular migration and interaction in the aforementioned studies are the net effect of a multitude of intercellular and intracellular molecular processes, and the ability to monitor these in vi vo and to correlate them to behavioral patterns will improve our ability to inter rogate the functionality of cellular networks. Neurobiologists, owing to their ref ined methods of loading environmentally sensitive fluorescent dy es into

191

individual neurons through microinjection into intact CNS tissue, were the first to measure Ca2+-fluxes in vivo as a readout of membrane depolarization during neuronal activity, triggered by a physiological stimulus, such as w hisker mo vement in the rat. 21 It took until 2006 before immunolo gists w ere ab le to measure the Ca 2+ response in B cells 133 upon their encounter of co gnate antigen on DCs in l ymph nodes in vi vo. Organic Ca 2+indicator dy es are plagued b y poor cellular retention. Genetically encoded , FP-based repor ters will in the future enable studies of Ca 2+-fluxes and of specif ic signaling pathw ays using FRET or FLIM techniques b y IVM. Ca2+ levels have already been indirectly monitored in muscle cells through IVM by using the Ca2+-sensitive proteolytic activity calpain on a FP-based FRET-probe as an indirect readout. 62

Physiologic and Pathophysiologic Functions The use of fluorescence greatly enhances the ability to not only describe but also quantitatively measure physiological processes b y IVM. Classical e xamples are studies on glomerular filtration or tubular secretion of free or conjugated fluorescent dy es in the kidne y16,24,134 or of v ascular permeability using fluorescently tagged macromolecules. 47 The same principle w as also used e xtensively to monitor transvascular and interstitial molecular transport in tumor tissue in order to refine the delivery of anticancer drugs (for a review, see Jain and colleagues135). The tumor microenvironment has also been inter rogated for hetero geneities in oxygen pressure through a phosphorescence-based IVM method.136,137 Malignant transformation of epithelial cells is B

A

Figure 3. Multiphoton intravital microscopy (MP-IVM) studies of cellular effector activity in the immune system. A, To measure the cytotoxic activity of tumor-reactive CD8+ T cells (green) in a tumor-draining lymph node at the single-cell level by MP-IVM, B cells coated with a tumor-expressed antigenic peptide were injected into the bloodstream, from where they would rapidly migrate to the draining lymph node and serve as surrogate target cells for T cells. B cells were labeled ex vivo with cytoplasmatic (Celltracker Orange, red) and nuclear organic fluorescent dyes (Hoechst 33342, blue) and changes in their fluorescent properties (loss of red signal and gain in blue signal) could be used to monitor loss of B cell structural integrity during cytotoxic T cell-induced apoptosis, along with loss of cellular function reflected by cessation of cellular motility. Yellow dots indicate B cell path. B, Time-resolved measurement of changes in B cell motility and structural integrity (“red/blue ratio”) upon encounter with a cytotoxic T cell (grey-shaded area) allows for determination of a T cell’s cytotoxic capacity. Modified from Mempel TR et al.66

192

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

accompanied by changes in their metabolic prof ile that are reflected in the fraction of free versus protein-bound NAD+. Free NAD+ has a shorter fluorescence lifetime than its protein-bound form, and MP-IVM has been used to detect precancerous lesions in hamsters. 138,139 IVM studies on microvascular perfusion have assessed parameters such as blood vessel diameter, blood flow velocity, or functional capillary density in a large number of disease states af fecting microcirculatory function. 140 Also the regulation of ar terial contractility has been subjected to imaging studies cor related with electroph ysiologic recordings to monitor the role of gap junctions in endothelial cellsmooth muscle cell communication in vivo.141 The lymphatic system has not been explored to such great detail as the blood vascular system, and in absence (until recently) of reliable histological markers, most of our knowledge of the anatomy of initial lymphatics was obtained from in vivo microlymphangiographic studies, initially in humans48 and later in mice,142 where the surprising f inding was made that tumors lack a functional lymphatic vasculature.143 High-speed video microscopy (at 500 frames per second) has recently allowed for precise determination of peak flow velocities in contractile segments of microlymphatic vessels.144

Cellular Growth and Differentiation While developmental processes occur at a f ast pace in lower animals, such as zebra fish, and can thus be studied by conventional time-lapse recordings, dif ferentiation processes in adult mammals oftentimes occur at timescales that require observation over days to weeks. The investigation of vessel formation in warm-blooded animals was the initial incentive for the development of animal chamber techniques, and the oppor tunity to study angiogenesis in the mouse, especially in response to solid tumor growth, has been the strength of the dorsal skinfold chamber model 13,14,135,145 which allo ws for the longitudinal obser vation of the same micro vascular bed over the course of weeks. Neuronal plasticity is another slo w process, and Karel Svoboda’s g roup has made use of a thinned skull preparation to monitor the changes in dendritic spines in nerve cells of the optical cortex under conditions of visual deprivation over the course of a month. 146

Host-Pathogen Interface Identifying and understanding the strate gies of pathogens use to enter a host, evade its immune defense mechanisms, and sometimes establish more or less peaceful, long-ter m

coexistence requires approaches to track the patho gens in vivo in relevant disease models. Methods to fluorescentl y tag pathogens have started to allow intravital visualization of their interaction with the host at dif ferent stages of infection (for a recent re view, see Mansson and colleagues147 and Velazquez and colleagues 148). Samel and colleagues 149 studied the kinetics of translocation of GFP-transfected , li ve E. coli bacteria through the intestinal w all in a model of bo wel obstruction, while Chieppa and colleagues67 investigated the role of epithelial DCs in the uptake of attenuated salmonella from the gut lumen. Other studies relied on the intravenous or subcutaneous injection of attenuated or inactivated patho gens to in vestigate their interaction with vascular endothelium as a putati vely relevant step in the development of bacterial sepsis 150 or with macrophages lining the floor of the subcapsular sinus in skin-draining lymph nodes as a threshold event in viral infection.151 Uta Frevert has pro vided breath-taking intra vital footage of plasmodium sporozoites transmitted through the skin of rodents by the bite of infectious mosquitoes, then using Kupffer cells as gates to exit the liver sinusoids, and eventually entering hepatocytes.152 The additional detail in our kno wledge of the lifestyle of pathogens from studies lik e this will in the future accelerate the development of therapeutic strategies to interfere with their initial entr y or their persistence. On the other hand, since our immune system has likely e volved primaril y to fend of f patho gens or to establish equilibria with commensals at our epithelial surfaces, its study in the context of infections will teach much about its function.

OUTLOOK Many advances in imaging technolo gy that will increase the range of applications of IVM are already in place and need to be adapted to the special challenges of imaging in live animals. One cur rent limitation of IVM is still in man y cases imaging depth in turbid tissues. Impro vement is in sight through the use of longer w avelength e xcitation light (beyond the range of Titanium:sapphire lasers mostly used in MP-IVM today) and of red and infrared fluorochromes. Better beam conditioning through the introduction of ne gative g roup-velocity dispersion, adapti ve w ave front sensing to correct for wave front distortion through refractive inde x inhomo geneities in biolo gical tissues, 153 and higher peak po wers achie ved with re generative amplifiers94 are also de velopments with potential benef it, but theoretically predicted limits of optical penetration at about

Intravital Microscopy

1 mm ma y remain in place. The use of needle-shaped optical lenses, so-called g radient index-lenses, which can be introduced into solid or gans without e xcessive tissue trauma ma y tur n out to be v aluable additions to IVM setups.95 Their combination with f iber-optical technology (see Chapter 14, “Dif fuse Optical Tomography and Spectroscopy”) might turn out to be a powerful tool for the minimally invasive optical access to internal organs and could also f ind application in tissues diagnosis in humans as instruments to obtain “optical biopsies” from patients. The utility of fluorescence imaging is par tly due to the possibilities of multiple xed data acquisition, and the use of spectral detectors instead of a limited number of PMTs may enhance the breadth of infor mation that can be simultaneously extracted from an IVM specimen. Finally, w e are witnessing the de velopment of more and more ref ined fluorescent probes for the inter rogation of molecular events, along with tools to influence processes under observation in vivo, for example, through photo-activated release of biologically active molecules.154 Apart from all technolo gical possibilities, ho wever, the ultimate limit to scientif ic investigation will remain the imaginati veness of the in vestigator, and it will tak e time until researchers in their respecti ve f ields will be able to ask all the biologically relevant questions that can usefully be addressed given the recent advances in intravital imaging technology.

REFERENCES 1. Malpighi M. De Pulmonibus. Bologna, Italy: Observationes Anatomicae; 1661. 2. Leeuwenhoek A. Collected letters. Edited and annotated b y a c ommittee of Dutch scientists. Amsterdam, The Netherlands: Swets and Zeitlinger; 1939–1999. 3. Hooke R. Micrographia: or some physiological descriptions of minute bodies made b y magnifying glasses with obser vations and inquiries thereupon. London: Jo. Martyn and Ja. Allestry, printers to the Royal Society; 1665. 4. Hall M. Critical and e xperimental assa y on the circulation of the blood. London: Sherwood; 1831. 5. Wagner R. Erläuter ungstafeln zur Ph ysiologie und Entwicklungsgeschichte. Leipzig, Germany: Leopold Voss; 1839. 6. Waller A. Microscopic examination of some of the principal tissues of the animal frame of the li ving frog, toad etc. Microscopic observations on the perforation of the capillaries b y the cor puseles of the b lood, and on the origin of mucus and pus-globules. Lond Edinb Dubl Phil Mag J Sci 1846;29:271–87, 379–405. 7. Cohnheim J . Über Entzündung und Eiter ung. Virchows Arch 1867;40:1–79. 8. Cohnheim J. Lectures on general patholo gy: a handbook for practitioners and students. London: The New Sydenham Society; 1889. 9. Thoma R. Über entzündliche Stör ungen des Kapillarkreislaufs bei Warmblütern. Virchows Arch 1878;74:360–93. 10. Bizzozero J . Ueber einen neuen F ormbestandteil des Blutes und dessen Rolle bei der Thrombose und der Blutgerinnung. Virchows Arch 1882;90:261–332.

193

11. Sandison JC. A ne w method for the microscopic study of li ving growing tissues b y the introduction of a transparent chamber in the rabbit’s ear. Anat Rec 1924;28:281–7. 12. Clark ER, Clark EL. Microscopic observations on the growth of blood capillaries in the living mammal. Am J Anat 1939;64:251–301. 13. Algire GH. An adaptation of the transparent chamber-technique to the mouse. J Natl Cancer Inst 1943;4:1–11. 14. Algire GH, Chalkey HW. Vascular reactions of normal and malignant tissues in vivo. I. Vascular reactions of mice to wounds and to normal and neoplastic transplants. J Natl Cancer Inst 1945;6:73–85. 15. Thiel R. Untersuchungen zum Flüssigkeitswechsel am lebenden Menschenauge. Graefes Arch Clin Exp Ophthalmol 1924;113:329–46. 16. Ellinger P, Hirt A. Mikroskopische Untersuchungen an lebenden Or ganen. I. Mitteilung. Methodik: Intra vitalmikroskopie. Zeitschrift für die gesamte Anatomie 1929;90:791–802. 17. Ries J. Kinematographie der Befruchtung und Zellteilung. Archiv für Mikroskopische Anatomie 1909;74:1–34. 18. Fulton GP, Jackson RG, Lutz BR. Cinephotomicroscop y of nor mal blood circulation in the cheek pouch of the hamster . Science 1947;105:361–2. 19. Denk W, Strickler JH, Webb WW. Two-photon laser scanning fluorescence microscopy. Science 1990;248:73–6. 20. Potter SM, Wang CM, Gar rity PA, Fraser SE. Intra vital imaging of green fluorescent protein using tw o-photon laser -scanning microscopy. Gene 1996;173:25–31. 21. Svoboda K, Denk W, Kleinfeld D, Tank DW. In vivo dendritic calcium dynamics in neocor tical p yramidal neurons. Nature 1997; 385:161–5. 22. Squirrell JM, Wokosin DL, White JG, Bavister, BD. Long-term twophoton fluorescence imaging of mammalian embr yos without compromising viability. Nat Biotechnol 1999;17:763–7. 23. Brown E, Campbell RB, Tsuzuki Y, et al. Measurement of physiological parameters in tumors in vi vo using MPLSM. Proc Soc Photo Opt Instrum Eng 2001;4262:134–45. 24. Dunn KW, Sandoval RM, K elly KJ, et al. Functional studies of the kidney of living animals using multicolor two-photon microscopy. Am J Physiol Cell Physiol 2002;283:C905–16. 25. Miller MJ, Wei SH, Cahalan MD, Parker I. Autonomous T cell trafficking examined in vivo with intravital two-photon microscopy. Proc Natl Acad Sci USA 2003;100:2604–9. 26. MacDonald IC, Ragan DM, Schmidt EE, Groom AC. Kinetics of red blood cell passage through interendothelial slits into v enous sinuses in rat spleen, anal yzed by in vivo microscopy. Microvasc Res 1987;33:118–34. 27. Schmidt EE, MacDonald IC, Groom AC. Interactions of leuk ocytes with vessel walls and with other b lood cells, studied by high-resolution intra vital video microscop y of spleen. Micro vasc Res 1990;40:99–117. 28. Mempel TR, Moser C, Hutter J , et al. Visualization of leuk ocyte transendothelial and interstitial mig ration using reflected light oblique transillumination in intra vital video microscop y. J Vasc Res 2003;40:435–41. 29. Slaaf DW, Tangelder GJ, Reneman RS, et al. A versatile incident illuminator for intra vital microscop y. Int J Microcirc Clin Exp 1987;6:391–7. 30. Groner W, Winkelman JW, Harris AG, et al. Or thogonal polarization spectral imaging: a new method for study of the microcirculation. Nat Med 1999;5:1209–12. 31. Rajadhyaksha M, Grossman M, Estero witz D, et al. In vi vo confocal scanning laser microscop y of human skin: melanin pro vides strong contrast. J Invest Dermatol 1995;104:946–52. 32. Rajadhyaksha M, Gonzalez S, Za vislan JM, et al. In vi vo confocal scanning laser microscop y of human skin II: adv ances in instr umentation and comparison with histolo gy. J In vest Der matol 1999;113:293–303. 33. Ping-Chin C, Sun CK. In: Pawley JB, editor. Handbook of biological confocal microscopy. New York City: Springer; 2006. p. 703–21.

194

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

34. Zoumi A, Yeh A, Tromberg BJ. Imaging cells and extra cellular matrix in vi vo b y using second-har monic generation and tw o-photon excited fluorescence. Proc Natl Acad Sci USA 2002;99:11014–9. 35. Mempel TR, Henrickson SE, von Andrian UH. T-cell priming by dendritic cells in lymph nodes occurs in three distinct phases. Nature 2004;427:154–9. 36. Sahai E, Wyckoff J, Philippar U, et al. Simultaneous imaging of GFP, CFP and collagen in tumors in vi vo using multiphoton microscopy. BMC Biotechnol 2005;5:14. 37. Brown E, McKee T, diTomaso E, et al. Dynamic imaging of collagen and its modulation in tumors in vivo using second-harmonic generation. Nat Med 2003;9:796–800. 38. Debarre D, Supatto W, Pena AM, et al. Imaging lipid bodies in cells and tissues using third-har monic generation microscop y. Nat Methods 2006;3:47–53. 39. Maker PD, Terhune RW. Study of optical ef fects due to an induced polarization third order in the electric f ield strength. Ph ys Re v 1965;137:A801–18. 40. Cheng J-X, Xie XS. Coherent anti-stok es raman scattering microscopy: instr umentation, theor y, and applications. J Ph ys Chem B 2004;108:827–40. 41. Evans CL, Potma EO, Puoris'haag M, et al. Chemical imaging of tissue in vivo with video-rate coherent anti-stok es raman scattering microscopy. Proc Natl Acad Sci USA 2005;102:16807–12. 42. Bruchez M Jr, Moronne M, Gin P, et al. Semiconductor nanocr ystals as fluorescent biological labels. Science 1998;281:2013–6. 43. Goldberg M, Seraf in D, Klitzman B . Quantif ication of neutrophil adhesion to sk eletal muscle v enules following ischemia-reperfusion. J Reconstr Microsurg 1990;6:267–70. 44. Villringer A, Dirnagl U, Them A, et al. Imaging of leuk ocytes within the rat brain cortex in vivo. Microvasc Res 1991;42:305–15. 45. Bacskai BJ, Hickey GA, Skoch J, et al. Four-dimensional multiphoton imaging of brain entry, amyloid binding, and clearance of an amyloid-beta ligand in transgenic mice. Proc Natl Acad Sci USA 2003;100:12462–7. 46. Miller MJ, Safrina O, Parker I, Cahalan MD. Imaging the single cell dynamics of CD4+ T cell acti vation by dendritic cells in l ymph nodes. J Exp Med 2004;200:847–56. 47. Svensjö E. Bradykinin and prostaglandin E1, E2 and F2alphainduced macromolecular leakage in the hamster cheek pouch. Prostaglandins Med 1978;1:397–410. 48. Bollinger A, Jager K, Sgier F, Seglias J. Fluorescence microlymphography. Circulation 1981;64:1195–200. 49. M’Rini C, Cheng G, Schw eitzer C, et al. A no vel endothelial L-selectin ligand acti vity in l ymph node medulla that is re gulated b y a(1,3)-fucosyltransferase-IV . J Exp Med 2003; 198:1301–12. 50. Sipkins DA, Wei X, Wu JW, et al. In vivo imaging of specialized bone marrow endothelial microdomains for tumour eng raftment. Nature 2005;435:969–73. 51. Hauser AE, Junt T, Mempel TR, et al. Def inition of ger minal-center B cell mig ration in vi vo reveals predominant intrazonal circulation patterns. Immunity 2007;26:655–67. 52. Lyons AB, Parish CR. Determination of lymphocyte division by flow cytometry. J Immunol Methods 1994;171:131–7. 53. Hoffmeister KM, F elbinger TW, Falet H, et al. The clearance mechanism of chilled blood platelets. Cell 2003;112:87–97. 54. Larson DR, Zipfel WR, Williams RM, et al. Water-soluble quantum dots for multiphoton fluorescence imaging in vi vo. Science 2003;300:1434–6. 55. Stroh M, Zimmer JP, Duda DG, et al. Quantum dots spectrally distinguish multiple species within the tumor milieu in vi vo. Nat Med 2005;11:678–82. 56. Rieger S, K ulkarni RP, Darc y D, et al. Quantum dots are po werful multipurpose vital labeling agents in zebrafish embryos. Dev Dyn 2005;234:670–81.

57. Shimomura O , Johnson FH, Saiga Y. Extraction, purif ication and properties of aequorin, a bioluminescent protein from the luminous h ydromedusan, Aequorea. J Cell Comp Ph ysiol 1962; 59:223–39. 58. Morin JG, Hastings JW. Energy transfer in a bioluminescent system. J Cell Physiol 1971;77:313–8. 59. Morise H, Shimomura O, Johnson FH, Winant J. Intermolecular energy transfer in the bioluminescent system of Aequorea. Biochemistry 1974;13:2656–62. 60. Prasher DC, Eckenrode VK, Ward WW, et al. Primary structure of the Aequorea victoria g reen-fluorescent protein. Gene 1992; 111:229–33. 61. Chalfie M, Tu Y, Euskirchen G, et al. Green fluorescent protein as a marker for gene expression. Science 1994;263:802–5. 62. Stockholm D, Bar toli M, Sillon G, et al. Imaging calpain protease activity by multiphoton FRET in li ving mice. J Mol Biol 2005; 346:215–22. 63. Miyawaki A. Inno vations in the imaging of brain functions using fluorescent proteins. Neuron 2005;48:189–99. 64. Shaner NC, Steinbach PA, Tsien RY. A guide to choosing fluorescent proteins. Nat Methods 2005;2:905–9. 65. Farina KL, Wyckoff JB, Rivera J, et al. Cell motility of tumor cells visualized in living intact primary tumors using green fluorescent protein. Cancer Res 1998;58:2528–32. 66. Mempel TR, Pittet MJ , Khazaie K, et al. Re gulatory T cells reversibly suppress c ytotoxic T cell function independent of effector differentiation. Immunity 2006;25:129–41. 67. Chieppa M, Rescigno M, Huang AY, Germain RN. Dynamic imaging of dendritic cell extension into the small bowel lumen in response to epithelial cell TLR engagement. J Exp Med 2006;203:2841–52. 68. Mansson LE, Melican K, Boek el J, et al. Real-time studies of the progression of bacterial infections and immediate tissue responses in live animals. Cell Microbiol 2007;9:413–24. 69. Miyawaki A. Visualization of the spatial and temporal dynam ics of intracellular signaling. Dev Cell 2003;4:295–305. 70. Zipfel WR, Williams RM, Christie R, et al. Li ve tissue intrinsic emission microscop y using multiphoton-e xcited nati ve fluorescence and second harmonic generation. Proc Natl Acad Sci USA 2003; 100:7075–80. 71. Takano T, Tian GF, Peng W, et al. Cortical spreading depression causes and coincides with tissue hypoxia. Nat Neurosci 2007;10:754–62. 72. Mrass P, Takano H, Ng LG, et al. Random mig ration precedes stable target cell interactions of tumor -infiltrating T cells. J Exp Med 2006;203:2749–61. 73. Kable E P, K iemer AK. N on-invasive l ive-cell m easurement o f changes in macrophage N AD(P)H b y tw o-photon microscop y. Immunol Lett 2005;96:33–8. 74. Mempel TR, Scimone ML, Mora JR, von Andrian UH. In vivo imaging of leuk ocyte traf ficking in b lood v essels and tissues. Cur r Opin Immunol 2004;16:406–17. 75. Bienvenu K, Harris N, Granger DN. Modulation of leukocyte migration in mesenteric interstitium. Am J Physiol 1994;267:H1573–7. 76. Hickey MJ, Forster M, Mitchell D, et al. L-selectin facilitates emigration and e xtravascular locomotion of leuk ocytes during acute inflammatory responses in vivo. J Immunol 2000;165:7164–70. 77. Saetzler RK, Jallo J , Lehr HA, et al. Intra vital fluorescence microscopy: impact of light-induced phototoxicity on adhesion of fluorescently labeled leuk ocytes. J Histochem Cytochem 1997; 45:505–13. 78. von Andrian UH. Intravital microscopy of the peripheral lymph node microcirculation in mice. Microcirc 1996;3:287–300. 79. Agard DA, Sedat JW . Three-dimensional architecture of a pol ytene nucleus. Nature 1983;302:676–81. 80. Neil MAA, Juskaitis R, Wilson T. Method of obtaining optical sectioning b y using str uctured light in a con ventional microscope. Opt Lett 1997;22:1905–7.

Intravital Microscopy

81. White JG, Amos WB, F ordham M. An e valuation of confocal versus conventional imaging of biological structures by fluorescence light microscopy. J Cell Biol 1987;105:41–8. 82. Tsien RY, Bacskai BJ . In: P awley JB, editor. Handbook of confocal microscopy. New York: Plenum; 1995. p. 459–78. 83. Callamaras N, Parker I. Constr uction of a confocal micro scope for real-time x-y and x-z imaging. Cell Calcium 1999;26:271–9. 84. Lechleiter JD , Lin DT , Sienear t I. Multi-photon laser scanning microscopy using an acoustic optical deflector . Bioph ys J 2002; 83:2292–9. 85. Bewersdorf J, Pick R, Hell SW. Multifocal multiphoton microscopy. Opt Lett 1998;23:655–7. 86. Buist AH, M¸ller M, Squier J , Brak enhoff GJ . Real time tw ophoton absor ption microscop y using mutli point e xcitation. J Microsc 1998;192:217–26. 87. Villringer A, Haberl RL, Dirnagl U, et al. Confocal laser microscopy to study microcirculation on the rat brain surf ace in vi vo. Brain Res 1989;504:159–60. 88. Grayson MH, Chaplin DD, Karl IE, Hotchkiss RS. Confocal fluorescent intravital microscopy of the murine spleen. J Immunol Methods 2001;256:55–63. 89. Grayson MH, Hotchkiss RS, Karl IE, et al. Intra vital microscop y comparing T lymphocyte trafficking to the spleen and the mesenteric l ymph node. Am J Ph ysiol Hear t Circ Ph ysiol 2003; 284:H2213–26. 90. Kissenpfennig A, Henri S, Dubois B, et al. Dynamics and function of Langerhans cells in vi vo: dermal dendritic cells colonize lymph node areas distinct from slo wer mig rating Langerhans cells. Immunity 2005;22:643–54. 91. Nishibu A, Ward BR, Jester JV, et al. Behavioral responses of epidermal Langerhans cells in situ to local patholo gical stimuli. J Invest Dermatol 2006;126:787–96. 92. Auffray C, Fogg D, Garfa M, et al. Monitoring of b lood vessels and tissues b y a population of monoc ytes with patrolling beha vior. Science 2007;317:666–70. 93. Zipfel WR, Williams RM, Webb WW. Nonlinear magic: multiphoton microscopy in the biosciences. Nat Biotechnol 2003;21:1369–77. 94. Theer P, Hasan MT, Denk W. Two-photon imaging to a depth of 1000 microm in living brains by use of a Ti: Al2O3 regenerative amplifier. Opt Lett 2003;28:1022–4. 95. Levene MJ, Dombeck DA, Kasischke KA, et al. In vi vo multiphoton microscopy of deep brain tissue. J Neurophysiol 2004;91:1908–12. 96. Nielsen T, Fricke M, Hellweg D, Andresen P. High efficiency beam splitter for multifocal multiphoton microscop y. J Microsc 2001; 201:368–76. 97. Niesner R, Andresen V, Neumann J , et al. The power of single and multibeam two-photon microscopy for high-resolution and highspeed deep tissue and intra vital imaging. Bioph ys J 2007; 93:2519–29. 98. Gerritsen HC, Draaijer A, van den Heuv el DJ, Agronskaia AV. In: Pawley JB, editor. Handbook of biological confocal microscopy. New York City: Springer; 2006. p. 516–34. 99. Williams D W, Truman JW. Mechanisms of dendritic elabora tion of sensory neurons in Drosophila: insights from in vi vo time lapse. J Neurosci 2004;24:1541–50. 100. Herbomel P, Thisse B, Thisse C. Onto geny and beha viour of earl y macrophages in the zebraf ish embr yo. De velopment 1999; 126:3735–45. 101. Lawson ND, Weinstein BM. In vivo imaging of embr yonic vascular development using transgenic zebraf ish. De v Biol 2002; 248:307–18. 102. Orozco JT, Wedaman KP, Signor D , et al. Mo vement of motor and cargo along cilia. Nature 1999;398:674. 103. Endrich B, Asaishi K, Gotz A, Messmer K. Technical report––a new chamber technique for micro vascular studies in unanesthetized hamsters. Res Exp Med (Berl) 1980;177:125–34.

195

104. Helmchen F , F ee MS, Tank D W, Denk W. A miniature head-mounted two-photon microscope. High-resolution brain imaging in freely moving animals. Neuron 2001;31:903–12. 105. Janssen BJ, De Celle T, Debets JJ, et al. Effects of anesthetics on systemic hemodynamics in mice. Am J Ph ysiol Hear t Circ Ph ysiol 2004;287:H1618–24. 106. Brookes ZL, Reill y CS, La wton BK, Bro wn NJ. Intravenous anesthesia inhibits leuk ocyte-endothelial interactions and e xpression of CD11b after hemor rhage. Shock 2006;25:492–9. 107. Lee HT , Kim M, Kim N , et al. Isoflurane protects against renal ischemia and reperfusion injury and modulates leukocyte infiltration in mice. Am J Physiol Renal Physiol 2007;293:F713–22. 108. Baez S. An open cremaster muscle preparation for the study of blood vessels by in vivo microscopy. Microvasc Res 1973;5:384–94. 109. Ley K, Bullard DC, Arbonés ML, et al. Sequential contrib ution of L- and P-selectin to leuk ocyte rolling in vi vo. J Exp Med 1995; 181:669–75. 110. Chilian WM, La yne SM. Coronar y micro vascular responses to reductions in perfusion pressure. Evidence for persistent ar teriolar v asomotor tone during coronar y h ypoperfusion. Circ Res 1990;66:1227–38. 111. Kuebler WM, Parthasarathi K, Linder t J, Bhattachar ya J. Real-time lung microscopy. J Appl Physiol 2007;102:1255–64. 112. Tabuchi A, Mertens M, Kuppe H, et al. Intravital microscopy of the murine pulmonary microcirculation. J Appl Physiol 2007. 113. Geissmann F, Cameron TO, Sidobre S, et al. Intra vascular immune surveillance b y CXCR6+ NKT cells patrolling li ver sinusoids. PLoS Biol 2005;3:e113. 114. Melder RJ, Salehi HA, Jain RK. Interaction of activated natural killer cells with nor mal and tumor v essels in cranial windo ws in mice. Microvasc Res 1995;50:35–44. 115. Mazo IB , Gutier rez-Ramos J-C, F renette PS, et al. Hematopoietic progenitor cell rolling in bone marrow microvessels: parallel contributions by endothelial selectins and VCAM-1. J Exp Med 1998; 188:465–74. 116. von Andrian UH, Mempel TR. Homing and cellular traffic in lymph nodes. Nat Rev Immunol 2003;3:867–78. 117. Miller MJ, Wei SH, P arker I, Cahalan MD . Two-photon imaging of lymphocyte motility and antigen response in intact l ymph node. Science 2002;296:1869–73. 118. Gear ARL. Rhodamine 6G. J Biol Chem 1974;249:3628–37. 119. Umeda M. Experimental carcino genesis: rat sarcoma produced b y the injection of rhodamine 6G. Gan 1954;45:391–3. 120. Stripecke R, Carmen Villacres M, Skelton D, et al. Immune response to green fluorescent protein: implications for gene therap y. Gene Ther 1999;6:1305–12. 121. Springer TA. Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell 1994;76:301–14. 122. Butcher EC, Pick er LJ. Lymphocyte homing and homeostasis. Science 1996;272:60–6. 123. Massberg S, Gawaz M, Gruner S, et al. A crucial role of glycoprotein VI for platelet recruitment to the injured arterial wall in vivo. J Exp Med 2003;197:41–9. 124. Denis C, Methia N , Frenette PS, et al. A mouse model of se vere von Willebrand disease: defects in hemostasis and thrombosis. Proc Natl Acad Sci USA 1998;95:9524–9. 125. Ludwig RJ, Boehme B , Podda M, et al. Endothelial P-selectin as a target of heparin action in e xperimental melanoma lung metastasis. Cancer Res 2004;64:2743–50. 126. von Andrian UH, Chambers JD, McEvoy LM, et al. Twostep model of leukocyte-endothelial cell interaction in inflammation: distinct roles for LECAM-1 and the leukocyte β2 integrins in vivo. Proc Natl Acad Sci USA 1991;88:7538–42. 127. Ley K, Gaehtgens P. Endothelial, not hemodynamic, differences are responsible for preferential leuk ocyte rolling in rat mesenteric venules. Circ Res 1991;69:1034–41.

196

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

128. Stockton BM, Cheng G, Manjunath N , et al. Ne gative re gulation of T cell homing by CD43. Immunity 1998;8:373–81. 129. Phillipson M, Heit B, Colarusso P, et al. Intraluminal cra wling of neutrophils to emig ration sites: a molecularl y distinct process from adhesion in the recruitment cascade. J Exp Med 2006;203:2569–75. 130. Shakhar G, Lindquist RL, Sk okos D, et al. Stab le T cell-dendritic cell interactions precede the development of both tolerance and immunity in vivo. Nat Immunol 2005;6:707–17. 131. Garcia Z, Pradelli E, Celli S, et al. Competition for antigen determines the stability of T cell-dendritic cell interactions during clonal expansion. Proc Natl Acad Sci USA 2007;104:4553–8. 132. Wyckoff JB , Wang Y, Lin EY , et al. Direct visualization of macrophage-assisted tumor cell intravasation in mammary tumors. Cancer Res 2007;67:2649–56. 133. Qi H, Egen JG, Huang AY, Germain RN. Extrafollicular activation of lymph node B cells b y antigen-bearing dendritic cells. Science 2006;312:1672–6. 134. Steinhausen M, Muller P, Parekh N. Renal test dyes IV. Intravital fluorescence microscop y and microphotometr y of the tub ularly secreted dye sulfonefluorescein. Pflugers Arch 1976;364:83–9. 135. Jain RK, Munn LL, Fukumura D. Dissecting tumour pathophysiology using intravital microscopy. Nat Rev Cancer 2002;2:266–76. 136. Torres F ilho IP, Intaglietta M. Micro vessel PO2 measurements b y phosphorescence decay method. Am J Physiol 1993;265:H1434–8. 137. Torres Filho IP, Leunig M, Yuan F, et al. Nonin vasive measurement of microvascular and interstitial oxygen profiles in a human tumor in SCID mice. Proc Natl Acad Sci USA 1994;91:2081–5. 138. Skala MC, Riching KM, Bird DK, et al. In vivo multiphoton fluorescence lifetime imaging of protein-bound and free nicotinamide adenine dinucleotide in normal and precancerous epithelia. J Biomed Opt 2007;12:024014. 139. Skala MC, Riching KM, Gendron-Fitzpatrick A, et al. In vivo multiphoton microscopy of N ADH and FAD redox states, fluorescence lifetimes, and cellular mor phology in precancerous epithelia. Proc Natl Acad Sci USA 2007;104:19494–9. 140. Messmer K, Krombach F. Microcirculation research in experimental surgery. Chirurg 1998;69:333–8. 141. de Wit C, Hoepfl B, Wolfle SE. Endothelial mediators and communication through vascular gap junctions. Biol Chem 2006;387:3–9. 142. Leu AJ, Berk DA, Yuan F, Jain RK. Flo w velocity in the superf icial lymphatic netw ork of the mouse tail. Am J Physiol 1994; 267:H1507–13. 143. Leu AJ, Berk D A, Lymboussaki A, et al. Absence of functional lymphatics within a murine sarcoma: a molecular and functional evaluation. Cancer Res 2000;60:4324–7. 144. Dixon JB , Za wieja DC, Gashe v AA, Cote GL. Measuring microlymphatic flo w using f ast video microscop y. J Biomed Opt 2005;10:064016. 145. Brown EB, Campbell RB , Tsuzuki Y, et al. In vi vo measure ment of gene e xpression, angio genesis and ph ysiological function in tumors using multiphoton laser scanning microscop y. Nat Med 2001;7:864–8. 146. Grutzendler J, Kasthuri N, Gan WB. Long-term dendritic spine stability in the adult cor tex. Nature 2002;420:812–6. 147. Mansson LE, Melican K, Molitoris B A, Richter-Dahlfors A. Progression of bacterial infections studied in real time––no vel perspectives provided b y multiphoton microscop y. Cell Microbiol 2007;9:2334–43. 148. Velazquez P, Waite JC, Dustin ML. Dynamics of host defense: the view at the front lines. Nat Immunol 2007;8:1153–7. 149. Samel S, K eese M, Kleczka M, et al. Microscop y of bacterial translocation during small bo wel obstr uction and ischemia in vivo––a new animal model. BMC Surg 2002;2:6. 150. Laschke MW, Kerdudou S, Her rmann M, Menger MD . Intravital fluorescence microscopy: a novel tool for the study of the interaction of Staphylococcus aureus with the micro vascular endothelium in vivo. J Infect Dis 2005;191:435–43.

151. Junt T, Moseman EA, Iannacone M, et al. Subcapsular sinus macrophages in l ymph nodes clear l ymph-borne vir uses and present them to antiviral B cells. Nature 2007;450:110–4. 152. Frevert U, Engelmann S, Zougbede S, et al. Intravital observation of Plasmodium berghei sporozoite infection of the li ver. PLoS Biol 2005;3:e192. 153. Rueckel M, Mack-Bucher JA, Denk W. Adaptive wavefront correction in tw o-photon microscopy using coherence-gated w avefront sensing. Proc Natl Acad Sci USA 2006;103:17137–42. 154. Mulligan SJ, MacVicar BA. Calcium transients in astroc yte endfeet cause cerebrovascular constrictions. Nature 2004;431:195–9. 155. Cavanagh LL, Bonasio R, Mazo IB , et al. Activation of bone marrow-resident memory T cells by circulating, antigen-bearing dendritic cells. Nat Immunol 2005;6:1029–37. 156. Carvalho-Tavares J , Hick ey MJ , Hutchison J , et al. A role for platelets and endothelial selectins in tumor necrosis factor-alphainduced leuk ocyte recr uitment in the brain micro vasculature. Circ Res 2000;87:1141–8. 157. Davalos D , Gr utzendler J , Yang G, et al. ATP mediates rapid microglial response to local brain injur y in vivo. Nat Neurosci 2005;8:752–8. 158. Vajkoczy P, Laschinger M, Engelhardt B. Alpha4-integrin-VCAM-1 binding mediates G protein-independent capture of encephalitogenic T cell blasts to CNS white matter microvessels. J Clin Invest 2001;108:557–65. 159. Becker MD, Nobiling R, Planck SR, Rosenbaum JT . Digital videoimaging of leukocyte migration in the iris: intravital microscopy in a ph ysiological model during the onset of endotoxin-induced uveitis. J Immunol Methods 2000;240:23–37. 160. Miyamoto K, Ogura Y, Hamada M, et al. In vi vo neutralization of P-selectin inhibits leuk ocyte-endothelial interactions in retinal microcirculation during ocular inflammation. Micro vasc Res 1998;55:230–40. 161. Buhrle CP, Hackenthal E, Helmchen U, et al. The hydronephrotic kidney of the mouse as a tool for intra vital microscopy and in vitro electrophysiological studies of renin-containing cells. Lab Invest 1986;54:462–72. 162. Veihelmann A, Szczesny G, Nolte D , et al. A novel model for the study of synovial microcirculation in the mouse knee joint in vivo. Res Exp Med (Berl) 1998;198:43–54. 163. McCuskey RS. A dynamic and static study of hepatic ar terioles and hepatic sphincters. Am J Anat 1966;119:455–77. 164. Atherton A, Born GVR. Quantitati ve investigation of the adhesi veness of circulating polymorphonuclear leucocytes to blood vessel walls. J Physiol 1972;222:447–74. 165. Covell WP. A microscopic study of pancreatic secretion in the living animal. Anat Rec 1928;40:213–23. 166. McCuskey RS, Chapman TM. Microscopy of the li ving pancreas in situ. Am J Anat 1969;126:395–407. 167. Bjerknes M, Cheng H, Otta way CA. Dynamics of l ymphocyteendothelial interactions in vivo. Science 1986;231:402–5. 168. Eriksson E, Boykin JV, Pittman RN. Method for in vivo microscopy of the cutaneous microcirculation of the hairless mouse ear . Microvasc Res 1980;19:374–9. 169. Chakraverty R, Cote D, Buchli J, et al. An inflammatory checkpoint regulates recr uitment of g raft-versus-host reacti ve T cells to peripheral tissues. J Exp Med 2006;203:2021–31. 170. Massberg S, Enders G, Leiderer R, et al. Platelet-endothelial cell interactions during ischemia/reperfusion: the role of P-selectin. Blood 1998;92:507–15. 171. Soriano A, Salas A, Sans M, et al. VCAM-1, but not ICAM-1 or MAdCAM-1, immunob lockade ameliorates DSSinduced colitis in mice. Lab Invest 2000;80:1541–51. 172. McCuskey RS, Meinek e HA, Townsend SF. Studies of the hemopoietic microen vironment. I. Changes in the micro vascular system and stroma during er ythropoietic regeneration and suppression in the spleens of CF1 mice. Blood 1972;39:697–712.

14 DIFFUSE OPTICAL TOMOGRAPHY SPECTROSCOPY

AND

DAVID R. BUSCH, MS AND BRITTON CHANCE, PHD, SCD (CANTAB), MD (HON)

Many for ms of molecular imaging rel y on optical techniques, often using some form of absorption, scattering, fluorescence, phosphorescence, or bioluminescence; some of these techniques are described else where in this book. Absorption and fluorescent spectroscop y are ubiquitous bench-top tools; dif fuse optics e xtends these techniques deep into scattering media, allowing quantitative measurements of chromophores se veral centimeters into biological tissues. These diffuse measurements are distinct from surface or near surface measurements, trading spatial resolution for the ability to quantify chromophore concentrations centimeters into tissue. If one is only interested in surface or near surface information, other techniques are more appropriate. Optical measurements of cells in suspension or on the surface of tissue are easil y accomplished with a microscope or camera; optical coherence tomo graphy pro vides high-resolution imaging of features less than a millimeter into the tissue. However, w hen absorbing features or fluorophores are embedded in fe w millimeters of tissue, quantif ication of chromophore concentrations requires modeling of the scattered photon path lengths; in cer tain limits, these path lengths can be described through the dif fusion equation. This chapter addresses dif fuse optics: the re gime where scattering dominates and light ener gy ef fectively diffuses away from a source. Dif fuse optics in biomedicine seeks to measure the concentrations of intrinsic chromophores (primaril y hemo globin, o xyhemoglobin, water, and f at) and contrast agents, along with scattering parameters in deep tissue (~cm), using sources and detectors on the surf ace of the skin. F igure 1 shows low (most light scattering forward) and high (light directionality lost within a few mm) scattering regimes.

The sensiti vity of dif fuse optical measurements to hemoglobin provides a f ast and nonionizing technique to monitor muscle function, brain acti vation, various types of hematoma, sub-surf ace wound healing, angio genesis, and even cognition. Work with the sole absorption/fluorescence contrast agent (indocyanine green [ICG], described below) approved for humans suggests that there is signif icant potential for both scientif ically interesting and clinicall y useful w ork with optical molecular imaging agents currently under development, especially those targeting particular disease states. These targeted agents will increase the contrast between targeted and healthy tissue, improving the ability of diffuse optics to detect the tissue of interest. This chapter focuses on applications of dif fuse optics to clinical prob lems, especiall y that of breast cancer. Clinically, advances in detection, diagnosis, and therapy monitoring for breast cancer offer much potential benef it to both suf ferers and, due to the high incidence of breast cancer ,1 society. Breast cancer has attracted much attention from the dif fuse optics community because, in addition to the clinical need, the human breast is some what easier to probe with dif fuse light than other tissues. The deformability and low optical absor ption of the breast allo w researchers to use a variety of geometries and mak e measurements through thicker tissue than is possib le with other or gans. This clinical need and reduced e xperimental difficulty may allow breast cancer to be one of the f irst clinical applications of molecular imaging agents in dif fuse optics. Leff et al.,2 in a recent review of clinical applications of diffuse optics to breast cancer , suggested that ~85% of cancers repor ted in these w orks* are detectab le using intrinsic optical contrast; Chance’ s study of o ver

*

See cited papers for inclusion criteria. 197

198

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 1. An example of scattering: low scattering (left) and high scattering (right). The high scattering example approaches the diffusive regime with an almost spherical distribution of light energy. Images were made by filling vials with water and two concentrations of scattering agent, then illuminating with a HeNe laser from the left of the figure.

100 women3 suggests sensitivity and specif icity above 90%, again with intrinsic contrasts. As Ntziachristos discusses fluorescence tomo graphy else where in this book, w e will focus mostl y on absorption tomo graphy, mentioning a fe w results especially important to clinical work. Similarly, we will not e xamine endoscopic or interoperati ve applications of dif fuse optics nor photodynamic therap y (PDT). Also, Wang discusses the related field of photoacoustic tomography else where in this book. Table 1 lists the acronyms used in this chapter.

NEAR-INFRARED DIFFUSION REGIME IN TISSUE Light propagation in a scattering, absorbing media can be described b y the dif fusion appro ximation to the radiative transport equation (RTE) under certain limits.† In tissue, these requirements are often satisf ied b y †

light appro ximately betw een 650 and 950 nm, the near-infrared (NIR) re gion of the spectr um. Other situations may require the full RTE or more careful approximations.‡ Modeling light propagation in tissue with the diffusion equation was pioneered by Patterson et al.6 in muscle; a recent tutorial b y Jacques and P ogue7 describes anal ytical, per turbative, and numerical approaches to dif fuse light transpor t and pro vides a useful introduction to the f ield. Additional references cited in this study pro vide justif ication for use of the diffusion equation for describing light transpor t and detail the associated limitations. The dif fusion re gime is def ined as the ratio of the reduced scattering and absorption coefficients ( µ sʹ′ / µ a, see definition below). Jacques and Pogue7 suggest µ ʹ′s /µ a > 20 as a good limit for use of the diffusion equation to describe light transpor t. These limits appl y to measurements of many tissues with light in the NIR: for e xample, human breast tissue ( µ a ∼ 0.05cm −1 , µ ʹ′s ∼ 10 cm −1 ) has µ ʹ′s / µ a ∼ 200 at ~800 nm. Use of the dif fusion approximation also requires measurements made se veral photon mean free paths (each ∼ 1 / µ ʹ′s ) away from the source and effectively isotropic scattering. The NIR w avelengths used in dif fuse optics are safe for frequent and long-ter m measurements as the NIR photons are nonionizing, unlik e X-ra ys. Tissue damage mechanisms are considered to be e xclusively thermal b y both American National Standards Institute8 (ANSI) and the United States F ood and Dr ug Administration 9 (FDA). Skin and e ye maximum permissible exposures (MPE) both vary by wavelength but are generally well above the power levels required to obtain useful dif fuse optical signal o ver se veral centimeters of source-detector separation; man y instruments operate at about the same power as a common laser pointer. These low power requirements allow design of instr uments for continuous optical monitoring of tissue without ph ysiological damage. This is also tr ue, within limits, for magnetic resonance imaging (MRI) and ultrasound , but dif fuse optical devices are much less e xpensive than MRI, do not require exclusion of patients with metal implants, and are a vailable at the bedside. Ultrasound pro vides the same adv antages but pro vides str uctural tissue information, whereas diffuse optics provides hemodynamic functional information. Fluorescence tomo graphy, typicall y using e xogenous agents in the NIR, introduces both additional

See, for example, the P1 approximation in Case 4 applied to the problem of neutron transport. ‡ See Klose and Hielscher 5 for an overview of optical tomography in the context of the RTE.

Diffuse Optical Tomography and Spectroscopy

Table 1. ACRONYMS AND SYMBOLS ASL

Arterial spin labeling

ANSI

American National Standards Institute

BOLD

Blood oxygenation level dependent (fMRI signal)

c

Speed of light in the relevant medium

CCD

Charge coupled device

CSF

Cerebral spinal fluid

CW

D0

Continuous wave (e.g., FD with ω = 0) c Diffusion coefficient: 3µsʹ′ [ r ] Homogeneous diffusion coefficient

δ[x]

Kronecker delta function of [x]

Δx

Change in variable x

DCS

Diffuse correlation spectroscopy

DOS

Diffuse optical spectroscopy

DOS

Diffuse optical tomography

DP

Differential path length (see DPF)

DPF

Differential path length factor

DPDW

Diffuse photon density wave

DWS

Diffusing wave spectroscopy, synonym for DCS

FD

Frequency domain

FDA

Food and Drug Administration (United States)

FDG

Fluorodeoxyglucose, an F18 PET agent

FEM

Finite element method

fDOS

Functional DOS

fMRI

Functional MRI

fNIRS

Functional NIRS

g

Anisotropy, mean cosine of scattering angle

γ

Tikhonov regularization constant. Typically denoted λ

Γ

Tikhonov regularization matrix

Gd

Gadolinium, MRI contrast agent

Gd-DTPA

Gadolinium Diethylenetriamine Penta-acetic Acid; commonly used Gd chelate for MRI contrast

Hb

Hemoglobin

HbO2

Oxyhemoglobin

Hbt

Total [Hb] = Hb + HbO2

H2O

Water

ICG

Indocyanine green

IRF

Instrument response function, used in TD measurements

λ

In this work, wavelength in nm

k0

Complex wavenumber of DPDW

LD

Laser diode

MPE

Maximum permissible exposure

MRI

Magnetic resonance imaging

µa

Absorption coefficient

µs

Scattering coefficient

D[ r ]

(Continued)

199

200

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 1. (Continued) µsʹ′

Reduced scattering coefficient

Nd

Number of Detectors

NIR

Near Infrared (~650–950 nm)

NIRS

Near Infrared spectroscopy; synonym for DOS

Ns

Number of Sources

ω

Angular frequency

OI

Optical index, composite parameter to increase breast cancer contrast developed at the University of Pennsylvania

PD

Photodiode

PDT

Photodynamic therapy

PET

Positron emission tomography

PMT

Photomultiplier tube

ψ

Photon fluence

rd

Detector Position

rs

Source Position

RTE

Radiative transport equation

Sat.

Oxygen saturation = HbO2/Hbt

SNL

Sentinel lymph node

SVD

Singular value decomposition

TCSPC

Time-correlated single photon counting, used for TD measurements

TD

Time domain

TIBS

Transillumination breast spectroscopy

TOI

Tissue optical index, composite parameter to increase breast cancer contrast developed at University of California at Irvine

TRS

Time-resolved spectroscopy

information (agent distribution) and a complication, as the positions of fluorescent emitters are not known. However, the contrast betw een health y and labeled tissues can be signif icantly impro ved if the agent concentration in tar geted tissue is suf ficiently elevated compared with health y tissue (the tar get to background ratio). Autofluorescence in the NIR is quite low, fur ther reducing backg round fluorescence. These fluorescent imaging techniques ha ve been hugel y useful in small animal imaging but are presentl y not in common use for human studies due to a paucity of FDA-approved fluorescent agents. The exception, ICG, has been used in humans for decades and is described in Section “Contrast Agents for Optical Tomography.” Ntziachristos focuses on fluorescence imaging elsewhere in the book; see Weissleder10 for a comparison of diffuse optical tomo graphy (DOT) with other fluorescence modalities. *

PHOTON DIFFUSION EQUATION Experiments in this dif fusion regime require understanding of the mathematical model used to describe light transport; this section pro vides an introduction to the photon dif fusion equation, brok en do wn b y the w ay in which the light source is modulated in time. If w e consider photons to be par ticles tra versing a scattering medium, w e can describe the probability of scattering as a photon travels some distance by considering an ef fective scattering cross section and a number density of the scatterers. * This probability can be converted into an a verage path length before scattering, the mean free path ( ls). In the dif fusion equation, w e are interested in the scattering coef ficient ( µ s = ls−1 ) : the average number of scattering events per distance. If w e assume that the angular dependence of the scattering can be described solely by the angle between

In tissue, there is a distrib ution of scatterer cross sections and densities; w e will simply look at the effective average quantities.

Diffuse Optical Tomography and Spectroscopy

the incident and scattered photon paths ( θ), we can use the a verage cosine of this angle to characterize the anisotropy of this medium. We def ine an anisotrop y g = and, from this, define the reduced scattering coef ficient µsʹ′= µs (1 − g ) . The reduced scattering coefficient can be thought of as the in verse distance between ef fectively isotropic scattering e vents: 100 individual e vents with high forw ard scattering ma y have occurred, but we can lump them all together into a single, effective, scattering e vent in w hich the photon loses all memor y of its initial direction. The reduced scattering coef ficient allo ws us to use the dif fusion equation, which assumes isotropic scattering, in tissue, even though the anisotrop y of tissue ranges from ~0.7 to 0.9911 (e.g., tissue is strongly forward scattering). The distance betw een these ef fectively isotropic e vents is sometimes denoted ls* = ( µ sʹ′ )−1 and is ~1 mm in breast tissue. Similarly, w e def ine an absor ption coef ficient µa = la −1 , where la is the mean free path before absor ption. F or a single absorbing species, this def inition is equivalent to µa = C ⋅ ∈, where ∈ is the e xtinction coefficient and C is the concentration of the species. † Scattering and absorbing cross sections v ary with wavelength and therefore both µ ʹ′s and µa are wavelength dependent. The idea of calculating the probability of absorption or scattering as a function of distance tra veled o ver many individual steps is ter med a “random w alk”; this is the concept behind Monte Carlo calculations in many fields. Indeed, the diffusion equation describes the large N limit of man y disparate processes, w hich can be described as a random w alk. Chandrasekhar 12 has a classic treatment; Feynman et al.13 provided an insightful explanation of the transition betw een random walks and the diffusion equation. Ishimaru14 described scattering of w aves in random media; F ishkin and Gratton 15 and Haskell et al.16 provided useful solutions to the diffusion equation in the conte xt of dif fuse optics. Note that the accurac y of solutions w hen applied to e xperiments depends significantly on how the boundary of the medium is modeled. Hask ell et al .16 compared se veral techniques, but ongoing w ork on this topic is signif icant. Ishimaru wrote the classic text on the problem; see also the review by Jacques and Pogue.7 Arridge et al.17 provided a useful explanation of the diffusion equation and a tabulation of anal ytic solutions †

201

in various geometries. Boas et al.18 assembled a valuable review of dif fuse optics applied to medical imaging. Gibson et al19 provided a mathematically focused review on applications of diffuse optics.

Photon Diffusion Equation in the Time Domain r The dif fusion equation for photon density at a point and time t due to an isotropic point source at the origin S [ r , t ] = S0 δ[t ]δ[ r ]‡ is shown as follows6,§:

⎛ ∂ ⎞ ( [ ] ) µ [ ] − ∇ ⋅ D r ∇ + c r a ⎜⎝ ∂t ⎟⎠ Ψ d [ r , t ] = cS0 δ[t ]δ[ r ] ,

(1)

where w e ha ve def ined a dif fusion constant ⎡ cm 2 ⎤ ⎡ cm ⎤ is the speed of light in c , c D[ r ] = ⎢ ⎥ ⎢ s ⎥ 3µ ʹ′s [ r ] ⎣ s ⎦ ⎣ ⎦ the medium, µa is the absor ption coef ficient, and ⎡ W ⎤ is the dif fuse photon fluence rate. Note ψ d ⎢ 2 ⎥ ⎣ cm ⎦ that there are several conventions for the definition of the diffusion coefficient. We will use D = c as it has the 3µsʹ′ standard units (length squared per time) for this coef ficient, and se veral authors 20,21 have suggested that this formulation for D is required if one is to preser ve terms of the same order in the deri vation of Equation 1 from the RTE. The time domain (TD) solution to Equation 1 in a inf inite homo geneous dif fusing medium ( µa [r ] = µa0 , D[ r ] = D0 ) is as follows: Ψ d [r , t ] =

cS0 −3/ 2 3/2

( 4 πD0t )

e

−

r2 − µ 0 ct 4 D0t a

,

(2)

where r = r . Equation 2 describes photon propagation away from a sub-nanosecond pulsed source in an inf inite medium; Figure 2 provides a schematic of this process in a semi-infinite medium. The pulse broadening in the TD of a brief pulse of light passing through a scattering medium

See Section “Physiologic Information from DOT and Spectroscopy” for an explanation relating this description of absor ption to the Beer-Lambert Law. ‡ δ [x] denotes the Kronecker delta function for x. § Patterson cites Chandrasekhar,12 who uses somewhat different notation.

202

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

(∇

2

+k

2 0

)

cS0 Ψ AC = − δ[ r ], D0

(6)

cµa − ιω and other v ariables are D as in Section “Photon Dif fusion Equation in the Time Domain.” The solution in an infinite homogeneous media is as follows: where k = k − ιk = 0 R I

S0 Ψ AC [ r ] = 4πD0

k |r|

el I e × . |r| Damping exponential − k R |r|

(8)

Spherical wave

Figure 2. A schematic of a time domain measurement in the remission geometry. A narrow input pulse (intensity is plotted versus time, black down arrow; red schematic) is introduced into the tissue. Photons scatter off the scattering centers (blue) and are absorbed by the absorbing centers (red). Some photons will travel short distances before escaping the tissue (magenta); others will be absorbed close to the input site (cyan). Most photons will travel considerable distance from the input site (green); a few of these will be detected (black) at a site several mean free paths away from the input (black arrow up). As photons will travel many different paths to arrive at the detector, the input pulse is broadened by the time it reaches the detector.

is perhaps more intuitively obvious than the phase shift one observes in the more commonl y used frequenc y domain (FD) measurement discussed below; the TD is also convenient for Monte Carlo simulations. Most impor tantly, TD measurements allow absolute quantif ication of scattering and absorption coefficients.

Photon Diffusion Equation in the FD Writing Equation 1 in the FD brings out the close analogy to near f ield optics. Consider a source modulated at frequency ω, placed at the origin. There will be a constant (DC) and oscillating (AC) part:‡ S [ r , ω ] = δ[ r ]S0 e −ιωt + δ[ r ]SDC .

(3)

Examining onl y the oscillating par t of the detected wave: ∇( D[ r ]∇Ψ AC ) + (ιω − cµ a [ r ]) Ψ AC = − cS0δ [ r ].

(4)

In a homo geneous medium, D[r] = D 0 and the oscillating part of the photon fluence rate satisfies: ∇ 2 ΨAC +

‡

ιω − cµa cS Ψ AC = − 0 δ[ r ], D0 D0

This notation is adapted from F ishkin.15

(5)

Note that Equation 6 is the Helmholtz Equation. The µ sʹ′ = 10 cm –1, µa = 0.05 cm –1, (~150 cm for ω = 2π70 MHz) wavelength of these diffuse photon density w aves (DPD W) is much lar ger than the typical source-detector separation (~2–10 cm). Such sourcedetector separations are limited by signal strength, as the signal has an e xponential attenuation ( kR) with distance. Thus, dif fuse optics tak es place in the near f ield. It is important to remember that w e are considering photon density, not the indi vidual photon, w hen we discuss the wavelength or frequency of DPDWs. The DC signal (discussed belo w) is often discarded in FD experiments as it has a higher potential to be contaminated b y e xterior light sources or electronic noise. Lock-in amplifiers allow efficient elimination of detected signal, which is not frequency matched to the source. TD measurements can be Fourier transformed into many different frequenc y measurements; thus measurements at many frequencies (s wept frequenc y) pro vide the same absolute information on absorption and scattering. Single frequency measurements require multiple sourcedetector separations to obtain absolute optical properties.

Photon Diffusion Equation for Continuous Sources The continuous source (Continuous Wave [CW] ω = 0) solution is the constant (DC) por tion of the abo ve FD solution; it also describes light transport away from a steady state source. cµa cSDC δ[ r ], Ψ DC = − ∇ Ψ DC − D0 D0

(9)

cSDC ΨDC = − δ[ r ], D0

(10)

2

(∇

2

2 CW

−k

)

Diffuse Optical Tomography and Spectroscopy

SDC Ψ DC ⎡⎣ r ⎤⎦ = − 4π D0

e

− k DC r

,

(11)

Damping exponential

cµa . Note that this equation describes an D exponential decay as distance from the source increases. Modulation frequencies of ~2 kHz are often used to reduce noise contamination or allo w simultaneous detection of illumination from multiples sources. 22 However, the wavelengths of these ~2 kHz DPDWs are so long that CW solutions are more appropriate than FD. where kDC =

Resolution Limit in DOT Resolution in dif fuse optics depends on the size of the target and several other properties: the contrast relative to the background, depth into the medium, optode spacing, and distance from the optode. In deep tissue (se veral centimeters), DOT can resolve absorption targets ~0.5 to 1 cm. 23 In transverse measurements of a dif fusing slab, Culver, with singular v alue anal ysis,24 and Mark el and Schotland,25 with inversion formulae, derived similar resolution limits. A recently published article using Mark el and Schotland’s formulae experimentally found a resolution of 8 mm for bar targets in the center of a 6 cm slab 26 with 4:1 tar get to backg round contrast and optode spacing of 2 mm. This is similar to the contrast found in Leff ’s review (summarized in Table 2) betw een malignant lesions and healthy, post menopausal, breast tissue. Torricelli et al.27,28 have recently proposed using time resolved, zero source-detector separation measurements to improve resolution. Boas et al.29 reviewed techniques to impro ve spatial resolution in functional brain measurements. Zef f et al . at the Washington Uni versity30 approached the functional brain mapping prob lem b y signif icantly increasing the optode density and impro ving the e xperimental reconstruction.

PHYSIOLOGIC INFORMATION FROM DOT AND SPECTROSCOPY Much work has been focused on using dif fuse optics to obtain functional infor mation about tissue through local concentrations of hemo globin (Hb), o xyhemoglobin (HbO2), f at, and w ater. Changes in total hemo globin concentration (Hbt = HbO2 + Hb) can identify angiogenesis or suggest temporal changes in blood volume; changes in *

203

oxygen saturation (Sat. = HbO2/Hbt) provide information on o xygen metabolism. The f at/H2O ratio pro vides information on tissue composition. Dif fuse optics is sensitive to micromolar concentrations in deep tissues; both absolute and relative measurements of chromophore concentration are used. Measurements ma y be taken relative to health y tissue, an ar tificial tissue reference “phantom,” or to some baseline. Scattering ( µ ʹ′s ) changes may provide information on cellular density and structure (e.g., more, smaller cells will ha ve more or ganelles ~1µm~NIR wavelengths). Relati ve measurements allo w less e xpensive instr umentation and simpler anal ysis techniques but require a careful choice of nor malization. Heterogeneities in both nor mal and especially cancerous tissues give a wide range of optical proper ties (see standard de viation of cancerous measurements in Table 2). These data can be e xpressed g raphically in tw o dimensions with a nomogram; see Section “Cluster Analysis to Identify Malignant Breast Tumors.” Absorption diffuse optical spectroscopy (DOS) can be thought of as an e xtension of the f amiliar Beer Lambert la w,* relating incident ( I0) to detected ( I) intensities after a beam of light passes through an absorbing solution: I = I 0 ⋅ 10−∈⋅C⋅d .

(12)

The dif fuse optics community uses base e and µa = C × ∈e, for a chromophore at concentration C with molar absor ption coef ficient of ∈. This convention matches with that typicall y used in discussions of the RTE and random walks. Note that many authors tabulate the extinction coefficients in base 10 (∈10), not the base e (∈e = lo g [e] · ∈10) used here. With this substitution, Equation 12 becomes as follows: I = I0 ⋅ e

− µ ad

.

(13)

In the presence of scattering, the path length ( d) is not simply the width of the cuv ette used (as in standard bench-top spectroscop y of clear liquid samples), but rather must be measured or calculated from known tissue properties. The simplest cor rection is to include a “differential path length factor” (DPF)34,35 which, when multiplied by the distance betw een the source and detector , approximates the mean distance a photon tra vels. This distance is also sometimes referred to as the “differential path length” (DP) and can be measured from the a verage tissue transit time of a brief (sub nanosecond) light

Originally, Bouguer31 and Beer,32 but also described in many modern texts, for example, Born and Wolf .33

204

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 2. OPTICALLY DETECTABLE PHYSIOLOGIC CONTRASTS IN BREAST CANCER Hbt [l mol/L]

Contrast

Sat.[%]

Contrast

Healthy, premenopausal

34 ± 9

1.9

75 ± 2

0.88

Healthy, postmenopausal

14 ± 0

4.6

80 ± 4

0.83

Healthy, not specified

21 ± 6

3.1

68 ± 5

0.97

Malignant lesion

65 ± 34

1

66 ± 24

1

Summary of currently published (before 2006) data by Leff et al.,2 given as mean ± standard deviation. Contrasts are calculated as Ylesion /Yx. Pre(post)menopausal data is calculated from three studies with 16(13),111 15(13),53 and 4(19) subjects.112 Data with menopausal status not specified was averaged over 17 studies with several hundred subjects in total. Note that malignant lesions appear to be much more heterogeneous than healthy tissue; see text for discussion.

pulse;36 Patterson et al .,6 quantified this path length in 1989. F igure 2 sho ws the schematic of the process. Arridge et al.,17 reported a summary of these formulae in their review; for the inf inite medium, the a verage tissue transit time in terms of the optical properties of the tissue and spatial position † is as follows: | r |2 1 t = . 2 D+ | r | µ ⋅ c ⋅ D a

(14)

µ sʹ′ [ λ, r ] = A[ r ]λ − b[ r ] .

Again extending the familiar Beer-Lambert law, we can consider multiple chromophores. The absor ption coefficient, µa [λ], can be decomposed into the sum of the products of the concentrations ( Ci) and e xtinction coefficients (∈i [λ]) of the chromophores in the tissue of interest. In most tissues, the primar y absorbers are oxyhemoglobin (HbO 2), hemo globin (Hb), f at, and w ater. Equation 15 breaks the total, w avelength-dependent, absorption do wn into the concentrations ( Cx) of these chromophores and their extinction coefficients (∈x [λ]), N chromo

µ a [λ] =

∑

i =1

Ci ∈i [ λ ]

(15)

+ CW at × ∈W at ⎡⎣ λ⎤⎤⎦ + CF at × ∈F at ⎡⎣ λ ⎤⎦ ,

(16)

where w e ha ve assumed that there are no other chromophores absorbing at the w avelengths of interest. Figure 3 sho ws an e xample of this decomposition in breast tissue. Note that Arridge uses D = γ

*

(17)

Here, A is the “scattering pref actor,” and b is the “scattering power”; both are position dependent. The scattering po wer depends on both the relati ve indices of refraction betw een scatterers and the bulk medium, as well as the size and number density of these par ticles. Nilsson et al.,38 successfully applied this appro ximation, and it has been widel y used since 39,40 (and many others). The scattering power (b) is often assumed to be constant to reduce the number of f itting parameters.

MEASUREMENT TECHNIQUES Temporal Data Type

CHbO × ∈HbO ⎡⎣ λ ⎤⎦ + CHb × ∈Hb ⎡⎣ λ ⎤⎦ 2 2

†

Dr. Scott Prahl, at the Oregon Medical Laser Center, provides a tabulation of the e xtinction coef ficients of hemoglobin, water, porcine f at, and se veral other useful chromophores* from several sources. Mourant et al .,37 suggested appro ximating tissue scattering with a power law, after the dependence of the scattering efficiency factor on λ from a simplif ied version of Mie scattering theor y:

2

=

Diffuse optical data is collected from pulsed (TD), modulated (FD), steady state (CW , ω = 0), or spatiall y modulated41 light sources. Detectors include photomultiplier tubes (PMTs), avalanche photodiodes (APDs), photodiodes (PDs), streak cameras, and charge coupled devices (CCDs). TD or time-resolv ed spectroscopic (TRS) techniques rely on introducing a brief pulse of light into a medium and measuring the transit time of man y photons through this

c 3( µ + µ ʹ′ ) a s

Available at http://www.omlc.ogi.edu/spectra/hemoglobin/summary.html, downloaded December 2006.

Diffuse Optical Tomography and Spectroscopy

0.07

Hbt : 23.4µ M Sat: 63% H2O: 15% Fat: 51% = 15

0.06

µ a [cm2 1]

0.05 0.04 0.03 0.02 0.01 0

700

750 λ [nm]

HbO2

Hb

Combined

Fat and Water

800 H2O

850 Fat

Measurements

Figure 3. Time domain measurements of absorption (black diamonds) at six wavelengths (690, 750, 780, 800, 830, 838 nm) and fits to spectra of major chromophores (HbO2, Hb, fat, H2O) in breast tissue for a 27-year-old premenopausal women, transmission geometry (5.2 cm thickness), homogeneous slab solution. Both µa and µ sʹ′ are in cm–1; fat and water are reported in percentage of tissue Hbt = Hb + HbO2; Sat. = HbO2/Hbt. D. R. Busch, unpublished data.

medium. Together with the tissue inde x of refraction, the mean of this transit time gi ves the a verage photon path length, which is related to the absor ption and scattering of the tissue;6,17,36 see Equation 14. Moder n techniques use a streak camera 42,43 or time-cor related single photon counting (TCSPC) 44,45 devices. The TCSPC technique relies on detectors (APDs or PMTs) operating in Geiger mode and a low photon flux: on a verage, each channel recei ves less than one photon per laser c ycle; lasers are pulsed at ~2 to 70 MHz. The time of each photon arrival is recorded and a histogram built up o ver an inte gration period; the results are fit to Equation 1 to obtain absolute values for µa[λ] and µ sʹ′ [λ]. TD systems also allow rejection of unwanted photons (e.g., those due to reflections in the system) through timegating. Torricelli et al.,27 extended this concept b y placing their source and detector directl y adjacent to one another and enab ling their shor t rise-time detector onl y after photons e xperiencing specular reflection from the surface or only a few scattering events had arrived. The phase of a modulated measurement encodes similar infor mation to the mean tissue transit time. These FD measurements are the F ourier analo gs of the TD measurements, typically using frequencies 30 to 200 MHz. 46–48 However, for practical reasons, most researchers measure one or a fe w frequencies instead of sweeping the frequenc y through a range. This frequency restriction reduces the infor mation content of the FD

205

measurements; those researchers 39 who use s wept frequency de vices ha ve much more infor mation per measurement. However, single frequenc y lock-in electronics are relati vely ine xpensive, and these instr uments can be more economical than the TD or s wept FD instr uments. Swept FD systems are cur rently limited to less than ~1 GHz b y the practical speed of modulation of source laser diodes, w hereas the frequenc y component of the TD signal extends to se veral GHz. Gibson et al .,19 have discussed this issue briefly in their review. In the ω = 0 (steady state or CW) case, it is dif ficult to separate changes in absorption and scattering, except in a few special cases. 49 However, CW techniques can measure changes in the absor ption in space (e.g., hematomas) and time ( e.g., exercising muscle) and can be v ery inexpensive (e.g., ~$50 in par ts for a 1 w avelength, 1 sourcedetector system). In practical e xperiments, TD and swept FD devices with a single source-detector separation can recover absolute optical properties, whereas other devices generally measure optical proper ties relati ve to some reference phantom or use multidistance measurements (single ω FD) to obtain absolute values.

Experimental Geometry In addition to the type of source modulation, a v ariety of data collection geometries are used. Hand-held probes with one or a few source-detector pairs offer simplicity and easy placement wherever measurements are desired. Probes for the muscle and brain are generall y taped or otherwise f astened to the body, not hand held, but are similar in concept, as both function in the remission geometr y (sometimes termed as the reflection geometr y). Hand-held probes can be used for cancer detection 3,39 and therapy monitoring50,51 and can be combined with multimodality measurements (see Section “Multimodal Imaging”). The most basic of these hand-held measurements use a homo geneous, semi-infinite solution to the diffusion equation; more sophisticated instr uments allow limited imaging (e.g., distinct la yers); multimodal instr uments can use str uctural information from the other modality (e.g., ultrasound) to image tissue. Li et al .,52 recently implemented a la yered model and applied the technique to phantom studies. Layered techniques impro ve the quantif ication of optical proper ties deeper into a medium, w hich is important to those researchers using naturally la yered (e.g., scalp/skull/brain) organs. Spatial infor mation for hand-held de vices is mostl y provided b y mo ving the probe betw een sequential measurements. Thus, hand-held devices produce a map of local

206

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

optical properties, with practical resolution deter mined by the number of dif ferent positions used; with a suf ficiently large number of positions, this could be used for limited tomography. Note that most surf ace remission measurements are dependent on the source-detector separation; this is par t of the reason authors cite widely varying depth sensiti vity (e.g., Chance 3 and Cerussi,53 1–4 cm). Additionally, the most probable photon penetration depth for a remission geometry device depends on the modulation frequenc y of the source. In a phantom study of quantif ication possible with CW remission measurements in tar get position, diameter , and concentration, Kepshire et al .,54 found accurac y to f all off signif icantly with a depth of onl y 5 mm. Full tomo graphic measurements are made with many fixed or a few scanning sources and detectors, although Ge et al.,55 have recently developed a two part hand-held breast FD tomographic probe. Much functional brain w ork with dif fuse optics uses devices with too fe w source-detector pairs to perfor m tomography without outside spatial infor mation (e.g., MRI). The human brain is too highl y absorbing for transillumination; adult brain tomography is therefore done with a 2D ar ray of sources and detectors placed on the outside of the skull. These ar rays can, for e xample, detect regional acti vation of the prefrontal cor tex during mental tasks.56 The smaller size allo ws the entirety of neonatal brains to be measured.57 Zeff et al.,30 have extended this 2D array geometr y to a “high-density” ar ray of sources and detectors, signif icantly improving image resolution in the adult brain. In the breast, c ylindric or cone geometries ha ve been used by several research58–60 groups, the Phillips “Optical Mammoscope,”61,62 and the breast v ersion of the NIRx DYNOT system. 63 Slab geometr y (mild compression of the breast betw een tw o flat plates) is cur rently used b y several groups for breast imaging. Scanning slab geometry, in w hich a fe w sources and detectors are raster scanned in tandem across the compression plates, has been used b y academic g roups64–66 and in the ART “SoftScan.”43 However, these measurements are all on or about an axis through the slab (i.e., the source and detector are almost directly across from one another), allowing few ob lique measurements. Therefore, it is dif ficult to ascertain the depth of an y lesion detected; alter natively, the transverse resolution is much greater than the throughslab resolution. These measurements are sometimes known as “transillumination” but are distinct from the planar transillumination pioneered b y Culter 67,68 in 1929. Fixed position slab geometry, in which an array of optical fibers delivers light to the breast, has been used primaril y by Culver et al.,69 and Choe et al.,40 and commercially in

the DOBI “Comfor tScan™.” Detection can be either b y an ar ray of optical f ibers leading to detectors, detectors placed directly against the tissue surf ace, or by lens-coupled CCD cameras. Re views of diffuse optics applied to breast cancer have been published by several authors.70–72

COMMERCIAL DIFFUSE OPTICAL INSTRUMENTS A number of commercial v entures ha ve attempted to apply diffuse optics to biomedical problems, mostly concentrating on functional brain monitoring and cancer in the breast. A few also look at muscle function and brain injury. Others provide instrumentation widely used in the diffuse optics community and ha ve star ted bundling instrumentation kits for general use. An incomplete list of some of the major companies is included in Table 3, along with the organ of focus.

DIFFUSE CORRELATION SPECTROSCOPY Diffuse cor relation spectroscopy (DCS) uses the detected autocorrelation function of long-coherence (lo w bandwidth) laser illumination of tissue to measure the speed of moving scatterers. Boas 79 described ho w one could construct a dif fusion equation for this cor relation and thence access blood flow in microvasculature. A full description of this technique is be yond the scope of this w ork; se veral authors80–84 have studied biomedical applications and v alidations of this technique against other modalities. Note that this technique is also refer red to as “Dif fusing Wave Spectroscopy”85 (DWS).

DOT DOT localizes and quantif ies hetero geneities in a v olume based on surf ace measurements. Calculation of a simulated measurement v ector from kno wn inputs and a spatial distribution of optical parameters (i.e., µ a [ λ, r ] and µ ʹ′s [λ, r ]) is ter med “the forw ard prob lem” and is computationall y simple. The inverse problem, calculating the distribution of optical properties from the measurement v ector, is generall y ill-posed (more volume elements than measurements) and is much more computationally complex. A variety of techniques have been developed to address this problem. For small variations, the inverse problem can be treated as a per turbation: a small or slightl y absorbing heterogeneity will slightly perturb the measurement vector from the homogeneous case, to which solutions are known or can be calculated. The commonl y used Bor n

Diffuse Optical Tomography and Spectroscopy

207

Table 3. SAMPLING OF COMMERCIAL DIFFUSE OPTICS DEVICES Company/System Name

Focus

Data

ART SoftScan®

Clinical breast43

TD

Becker and Hickl

Nonspecific

TD

DOBI Medical International

Clinical breast

CW

Hamamatsu

Brain,73 breast74

TD

Hitachi

Brain

CW

Imaging Diagnostic Systems CTLM®

Clinical breast

CW

Infrascan

Clinical brain injury75,76

CW

ISS

Brain, breast, muscle

FD

NIRx Medical Technologies

Breast,63 brain77

CW

Picoquant

Nonspecific

TD

Somanetics InVos®

Clinical brain

CW

Spectros FirstScan Lori ix®

Clinical breast

CW/FD

TechEn

Brain and breast

CW

ViOptics

Clinical breast78

CW

CW = continuous wave; FD = frequency domain; TD = time domain.

approximation can be applied to this situation 0 δ µa [ r ] − µa 0 is the a verage backg round (e.g., 1 10 6)40 frequently using CCD detectors. Deliolanis et al.,108 have applied this to small animal fluorescence imaging. Culv er †

et al .,69 and Choe et al .,40 applied CCD detection to 3D tomography of absor ption proper ties in the breast; Corlu et al.,106 extended this work to perfor m 3D reconstr uction of ICG fluorescence in breast cancer . The resulting data sets, especiall y w hen scaled up to the human brain or breast, require signif icant computational resources to reconstruct tomographically with f inite element modeling or finite difference methods. Konecky et al., applied much more ef ficient inversion algorithms de veloped b y Schotland, Mark el, and Mital 97,109,110 to phantoms 26 and, in unpublished data, to health y human breasts. ¶ Large data sets can also be acquired by using devices with high frame rates. Schmitz et al.,63 report a system with 32 sources and 31 detector positions on each breast capab le of generating tomographic data at 2 to 75 Hz (depending on the number of sources used).

INTRINSIC CONTRASTS Devices and techniques that obtain clinically useful information without use of a contrast agent a void both complication and potential aller gic reactions and allo w frequent repetition of the measurement. Dif fuse optical intrinsic contrasts are primaril y hemo globin concentration (Hb t) and saturation (Sat.). Measurements of these intrinsic mark ers have been applied to muscle function, brain function and injury, and breast cancer.

Intrinsic Contrast in Breast Cancer Leff et al .,2 at Uni versity Colle ge London ha ve recentl y published an exhaustive review* of diffuse optics applied to breast cancer , summarizing the e xperimental conditions and results of some 34 studies, each having at least 5 subjects; the results of this tabulation are summarized in Table 2. Note the signif icant contrast betw een the means of the malignant lesions and health y tissue, but also the signif icant standard de viation, especiall y in the malignant lesions. Some of the variance in each chromophore isprobably due to g rouping all malignant cancers to gether, as cancers are kno wn to be hetero geneous both microscopically and across populations. Each of the intrinsic parameters measured by diffuse optics has a distribution function; this complicates de veloping diagnostic metrics. See Sections “Indices for Breast Cancer Detection in Dif fuse

http://www.web4.cs.ucl.ac.uk/research/vis/toast/. ‡ Time-resolved Optical Absorption and Scattering Tomography. Despite the name, now focused on FD solutions. ¶ Private communication, S. Konecky. * Leff only considered studies published before August 1, 2006.

Diffuse Optical Tomography and Spectroscopy

Optics,” “Cluster Analysis to Identify Malignant Breast Tumors,” and “Multidimensional Image Analysis” for a discussion of how several research g roups have attempted to resolve this difficulty. In a recent study (not included in Leff), Choe et al .,40 found a 1.25-2 X contrast betw een healthy tissue and 41 biopsy-conf irmed malignant lesions in 3D tomo graphic reconstr uctions. Images from tw o of the subjects in this study are shown in Figure 5. Several researchers ha ve sought to cor relate the tissue microstr ucture to DO T f indings. P ogue et al .,113 initiated comparison between DOT-reconstructed Hbt and blood-vessel density in tumors in a small study; Zhu et al.,114 have followed up on this research. Scattering in tissues is dominated b y str uctures appro ximately the same size as the light wavelength (~ 1 µm) . The cell proliferation of some cancer types has led some researchers to consider using µ sʹ′ as a diagnostic parameter; Li brok e this do wn into scatterer size and v olume fraction in search of an indicator of malignanc y.115 See Section “Dynamic and Ph ysical Contrasts” for intrinsic contrasts modulated b y pressure and inhaled gasses.

Intrinsic Contrast in Human Muscle Optical spectroscop y has been applied to ph ysiologic problems in in vi vo muscle at least since Jobsis obtained NIR measurements of the m yocardium.116 A decade later , Chance et al .,117 examined sk eletal muscle with TD instr umentation. Indeed, the application of the dif fusion equation to these TD measurements in muscle b y P atterson et al ., 6 allowed quantitative in vi vo measurements and signif icantly stimulated the development of the f ield. Diffuse optics of fers much potential as a monitor of muscle function as measurements can be tak en during exercise118,119 and in harsh environments. However, muscle contains a signif icant amount of m yoglobin, complicating analysis of the NIR spectra as m yoglobin absor ption is almost identical to hemoglobin in this region. A number of reviews are a vailable on this topic; see Boushel et al .,120 and Wolf et al.,121 and references therein.

Intrinsic Contrast in Brain: Function and Injury NIR light can penetrate the scalp, skull, and cerebral-spinal fluid (CSF) to interrogate the underlying brain, as shown by

211

Jobsis in 1977. 116 In infants,57 the head is small enough for the entire brain to be imaged, but the absor ption properties ( µ a ∼ 0.1 cm −1 ) of the adult brain preclude a source-detector separation beyond a few centimeters and therefore NIR techniques can only sample the top ~1 cm of the adult brain. Epidural or subdural hematomas and hemor rhagic strok e can cause b leeding at and belo w the surf ace of the brain, leading to an easil y detectab le DOS signal, up to ~2 cm from the surf ace.75 To see the magnitude of this signal, assume a fully deoxygenated bleed directly under the skull, the subject has 150[g/L] Hb, and 800 nm light. Fur ther assuming the bleed is simply whole blood, we can use tabulated extinction coefficients for Hb, and f ind µ a ∼ 4 cm −1; a 40 X contrast to the health y µa. This absorption will produce a readily detectable change in signal w hen a probe is moved across the skull o ver healthy ( µ a ∼ 0.1 cm −1 ) to the injured brain. However, the detected signal will be very low, quite possib ly belo w the noise le vel, with µ a ∼ 4 cm −1, making quantification of the bleed difficult. The key caveat of this technique is that DOS can indicate if there is a bleed within ~2 cm of the surf ace on an adult brain, but cannot indicate if there is not a bleed deep in the brain. Despite this limitation, researches into DOS applications on head trauma are underw ay in a number of g roups. Infrascan is acquiring data similar to Gopinath et al.,122 to commercialize a small, hand-held instr ument for this purpose. In addition to use of DOS in emer gency situations, DOS has the potential to monitor brain blood volume (Hbt), oxygenation (Sat.), and flow (using DCS, see references in Section “Dif fuse Cor relation Spectroscop y”) in unstab le patients. For example, Kirkpatrick et al.,123 used DOS for long-term monitoring (14 patients with a total of 886 h) in the neurointensive care unit, cor relating the results to se veral other modalities. Unfortunately, such monitoring is difficult to bring to reliable clinical realization: Buchner found high f ailure rate in clinical setting for v ery ill patients 124 with a CW monitoring de vice. Ne vertheless, the INV OS Cerebral Oximeter b y Somanetics ® has been used for a number of clinical studies, * and is commerciall y available in the United States for monitoring changes in regional blood o xygen saturation during sur gery and anesthesia, whereas Zhang et al.,76 Hueber et al.,127 and Zhou 128 et al. had success using FD in piglet studies. Al-Rawi provided a review of DOS applications to brain injur y.129 Measurement of brain function with DOS/DO T offers a much lo wer potential signal le vel than cranial bleeding or ph ysiologically unstab le subjects: one is looking for small changes in flo w, oxygenation level, or

*See Austin et al.125 and Casati et al.126 as well as the bibliography maintained at http://www.somanetics.com/ clinical_invos_bibliography.htm.

212

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Malignant: Invasive Ductal Carcinoma

Benign: Fibroadenoma

MRI axial slice

MRI axial slice

rTHC

rTHC

rTHC rTHC

0.8

0.9

1

1.1

1.2

1.3

0.8

0.9

1

rStO2

1.1

1.2

1.3

rStO2

rStO2 rStO2

0.8

0.9

1

rµ′ s

1.1

0.8

0.9

1

1.1

rµ′ s

rµs ′ rµs ′ 0.6

0.8

1

1.2

1.4

1.6

0.6

0.8

1

Optical Index

1.2

1.4

1.6

Optical Index

OI OI

0.6

0.8

1

1.2

1.4

1.6

1.8

2

0.6

6 cm

17 cm

1

1.2

1.4

1.6

1.8

2

Region of Interest

Region of Interest

8 cm

0.8

9 cm 6.5 cm 17 cm

Figure 5. DOT reconstructions of breast cancer. In descending rows: nonsimultaneous Gd-DTPA MRI, relative total hemoglobin (rTHC = rHbt), relative oxygen saturation (rSat.), relative µsʹ′ at 780 nm ( rµsʹ′ ), a composite optical index, and a 3D rendering of the tumor volume. Image courtesy of R. Choe, data described in Choe et al.40; note the significant difference between the benign and malignant lesions.

Diffuse Optical Tomography and Spectroscopy

blood vessel dilation (e.g., increases in local Hbt). Nevertheless, there is e xtensive literature describing dif fuse optical functional measurements of the brain; see Wolf et al.,121 and Strangman et al .,130 and references therein for detailed descriptions. Boas et al .,29 have re viewed some of the constraints, limitations, and opportunities for optimization of diffuse optics measurements in the brain. Hillman has e xamined dif fuse optics and related techniques applied to brain imaging in animals and humans.131 Several companies are attempting to commercialize devices to monitor brain functional or trauma. See Table 3 for a partial list.

MULTIMODAL IMAGING Multimodal imaging with dif fuse optics ma y focus on validation of the dif fuse optical measurement, comparison of dif fuse optics with a similar measurement using another technique, or synthysis of different types of information into a composite. A validation study might compare DO T with MRI imaging for detection of breast cancer. A comparison study might bring to gether functional MRI (fMRI) b lood o xygenation le vel dependent (BOLD) imaging and functional DOS/DO T, as both signals depend on neurovascular coupling. A synthesis study could use the tissue str uctural infor mation from X-ra y mammography to constrain an optical tomo graphic reconstruction showing Hbt and Sat. distributions. Diffuse optics can relati vely easil y be combined with se veral standard clinical techniques, such as MRI,132,133 ultrasound,134,135 magnetoencephalography,136,137 and X-ray mammography,138,139 as fiber optics can be inte grated into clinical de vices with little disturbance to other instrumentation. For example, fiber optics can carry light more than 10 meters to and from the tissue of interest, allo wing optoelectronic por tions of diffuse optical instr uments to be placed outside of the shielding about an MRI device. This ease of inte gration has inspired man y researchers to conduct multimodality studies. This section will focus on applications to breast cancer . Readers interested in small animal applications should see review by Ntziachristos et al.,140 and references therein. Those interested in measurements of the human brain should see re views b y Strangman et al .,130 and Steinbrink et al.,141 Multimodal dif fuse optical imaging with optical contrast agents will be discussed in Section “Multimodal Imaging with Contrast Agents.” Various g roups ha ve focused on combining the functional infor mation (e.g., o xygen metabolism

213

and angiogenesis) from optical imaging with high resolution str uctural infor mation pro vided b y other modalities, especiall y MRI, ultrasound , and X-ra y mammography. Simultaneous multimodality imaging has ob vious adv antages, especiall y in a defor mable organ like the breast. Ho wever, practical e xperimental concerns sometimes require sequential imaging and subsequent coregistration based on f iducial markers. Zhu et al.,134 demonstrated that ultrasound is a natural co-modality for hand-held DOS, as both are ine xpensive, easil y inte grated into a hand-held probe, and nonionizing.135,142,143 This last point allo ws ultrasound and DOS to be used for frequentl y repeated measurements (e.g., to monitor the course of treatment 144) without fear of causing damage from ionizing radiation. Ultrasound pro vides str uctural infor mation, allo wing reconstruction of the tissue volume to improve quantification. Ho wever, unlik e ultrasound that produces a series of images requiring human inter pretation, a DOS monitoring device can be taped or otherwise secured to a subject and set to aler t operators if some preset change, for e xample, in b lood o xygen saturation, occurs. Hand-held de vices ha ve also been cor related with MRI using sequential measurements MRI. 51 The MGH g roup under Da vid Boas has combined X-ray tomosynthesis with a tw o w avelength FD system139,145 and later incor porated an eight w avelength FD system 146 by ISS Inc. in their studies of the ef fects of compression on breast ph ysiology; see Section “Dynamic and Physical Contrasts.” Ntziachristos45 collected TD data simultaneousl y with MRI, using contrast agents for both modalities; see Section “Multimodal Imaging with ContrastAgents.” The Dartmouth optical breast cancer g roup headed b y Brian Pogue conducted a study of 11 health y subjects, with Brooksb y et al. 133,147 performing FD-DO T constrained with spatial tissue type distribution from Gd-enhanced MRI. Car penter et al. 148 extended this study into a cancerous subject with the addition of gadolinium contrast agent (Gd-DTP A), appl ying a regularization scheme allowing gradual changes inside a tissue type and abrupt changes on the boundaries between types. This instrument has 16 sources and 15 detectors, is operated at 100 MHz, and uses a circular geometr y. Azar149 has developed sophisticated defor mation software to compare optical measurements with 3D images from other imaging modalities tak en nonconcur rently; in this work, Gd-DTPA enhanced MRI. These fused data sets offer infor mation from or comparison of both modalities, when simultaneous data acquisition is impractical. Positron emission tomo graphy (PET) using fluorodeo xyglucose

214

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

(FDG) is a clinical imaging modality w hich shows cellular glucose uptake and therefore of fers a measurement of gl ycolytic cellular metabolism; combining PET and DO T offers the opportunity to compare this with blood oxygenation saturation. K onecky et al. , ha ve applied the data set fusion techniques to sho w cor relation between optical parameters, especially total hemoglobin and scattering with FDG uptake measured with whole body PET.150 Saturation measured b y optical methods had little cor relation with FDG uptake, but the study w as fairly small ( N = 9). They also extended their study with an e xperimental breast-only PET scanner to show colocalization of tumors in three subjects. Work appl ying this technique to hand-held optical measurements and MRI is ongoing in collaboration with University of California Irvine.

CONTRAST AGENTS FOR OPTICAL TOMOGRAPHY Contrast agents are used in medical imaging to create a detectable difference between tissue of interest and the surrounding background. In diffuse optics, contrasts can be absor ptive or fluorescent, but fluorescent agents are potentially detectab le at much lo wer concentrations. 151 Unfortunately, onl y a single dif fuse optical contrast agent is appro ved for clinical use in the United States. This will change as additional agents are de veloped and shepherded through the re gulatory process, but cur rent clinical work focuses almost e xclusively on ICG (a vailable in the United States from Akorn Specialty Pharmaceutical Compan y, http://www .akorn.com/), as this FDA-approved agent absorbs and fluoresces in the NIR. It was f irst applied to dif fuse optical imaging of breast cancer b y Nioka. 60,152 Many animal studies ha ve been conducted with diffuse optical contrast agents; one of the most useful studies from the standpoint of human measurements w as Cuccia et al .’s combination of optical contrasts (ICG and meth ylene b lue) with MR contrast (Gd-DTPA) in rat imaging. 153 Ntziachristos e xamines fluorescence tomo graphy elsewhere in this book. Sevick-Muraca and Rasmussen154 have also recently published a primer on molecular imaging, comparing optics with nuclear imaging and genetic techniques. Various other commercial agents ha ve been developed for animal use, some with conjugating moieties, which can be attached to various ligands known to target specif ic disease states. Work in animals with these tar geted agents, for e xample, that b y Re ynolds et al .,155 has demonstrated detection of ~1 cm lesions with micromolar concentrations of the agents in canine

mammary tumors. Klohs et al.,156 have provided a recent addition to a substantial body of re views on the topic. Unfortunately, none of these agents are commerciall y available for humans.

Introduction to ICG ICG w as introduced b y F ox et al .,157,158 at the Ma yo Clinic in 1957 for measurement of b lood dilution curves independent of b lood oxygen saturation. Since then, ICG has been used clinicall y159 primarily for hepatic,160–162 ocular,163,164 and cardiac 165,166 measurements (these citations are e xamples of e xtensive literature; for a re view of uses, see F rangioni 167). Landsman et al .,168 produced a series of commonl y used spectra for ICG. These spectra depend on both the solvent (water or blood plasma) and the concentration (due to agg regation). ICG agg regation and selfabsorption of emitted photons can be quite impor tant: for example, Brasch et al.,169 found significant nonlinearities in the in vi vo ICG signal with increasing dose on a rat model. This complicates quantif ication at higher concentrations. ICG is not an ideal NIR contrast agent in terms of fluorescent yield or absor ption cross section; but FD A approval, its long history of safe use in humans,170 and the lack of alternate fluorophores or chromophores in the NIR approved for human use mak e ICG the cur rent primar y contrast agent for dif fuse optical studies in humans. Preclinical ICG derivatives are reviewed by Tung171 and optical agents more generall y are reviewed by Licha et al.,172 and elsewhere in this book by Hilderbrand. Sevick-Muraca et al .,151 explored the limits of ICG imaging and also compared CW and FD (100 MHz) fluorescent imaging on a phantom with a 100 µL target of 100 fM ICG at depths from 1 to 7 cm. Their results showed obvious superiority of FD in depth sensitivity: CW only had an obvious signal at 1 cm depth, but FD images showed a target down to 7 cm. Sevick-Muraca also provides a re view of ICG and the de velopments in targeted contrast agents. Diffuse optics with ICG contrast has been applied to breast cancer ,45 brain monitoring, 173 and, recently, the flow of lymph in human limbs. 154

Applications of ICG to Breast Cancer ICG is quickl y cleared from the b lood stream b y the 174 liver (half-life in the breast ~3–5 min ), changing concentration quite rapidly in the f irst several minutes.

Diffuse Optical Tomography and Spectroscopy

It is primaril y a b lood pooling agent, binding to albumin w hen injected intra vascularly. Se veral researchers106,132 have found ICG concentrations in tumors are higher than that in sur rounding tissue. Tumors are known to produce additional microvasculature (angiogenesis); this is one of the sources of intrinsic DO T contrast. These quickl y g rown v essels are known to be more per meable, leading to the h ypothesized origin of the ICG concentration contrast: this heightened per meability allo ws ICG or e ven ICGbound albumin to e xit the v asculature and produces a transient build up in the tumors. This extravascular ICG apparently exchanges with vascular ICG f airly quickly (~10–20 min174), and all ICG is e ventually removed by the liver. Several researchers have attempted to deter mine if the kinetics of this process ma y be diagnostic for malignancy. Intes et al., 174 used a CW stand-alone instrument described by Nioka et al., 152 for monitoring ICG kinetics in breast cancer and obser ved substantially dif ferent kinetic prof iles betw een three subjects with dif ferent types of cancer . In a lar ger study, Rinneberg et al. ,44 using a TD scanning system, did not observe a dif ference in ICG kinetics betw een types of lesions. Additionall y, Sevick-Muraca et al., 175 have examined the use of ICG to map sentinel l ymph nodes (SNL) in patients with breast cancer and found subcutaneous injections of 10 to 100 µg ICG allo wed imaging of l ymph drainage paths and the SNLs, comparing the technique with lymphoscintigraphy. This w ork suggests that ICG ma y help identify those SNLs to remove for pathologic examination and which lymph nodes can be safely left intact. Corlu et al.,106 applied their 45 source CCD detection (thousands of source-detector pairs) instr ument to imaging ICG deposition in cancers and pub lished the f irst tomographic breast reconstr uctions using ICG fluorescence. An example of their images is sho wn in Figure 6. This image w as collected w ell do wn on the tail of the kinetics curve (~6 min post injection) to reduce the effect of the rapidly changing drug concentration on the reconstruction, of course reducing the ICG signal.

Applications of ICG to Human Brain Function and Injury ICG has been used to measure cerebral b lood flow,176,177 including changes due to ischemic strok e.178 Keller et al also applied this technique to subarachnoid hemorrhage.179 *

Manuscript in preparation, private communication G. Faris, Oct 2008.

215

Liebert et al., 173 reported use of ICG fluorescence in the brain in a proof of concept study.

Multimodal Imaging with Contrast Agents A few researchers have used both structural information from other modalities and injected optical contrast agents. Ntziachristos et al., 45,132 pioneered the simultaneous acquisition of dif fuse optical data with MRI. These researchers used a TD instrument to obtain simultaneous optical measurements with ICG contrast and Gd-DTPA-enhanced MRI using 24 sources, 8 detectors, 830 nm system with a 5 10 cm g rid. Ntziachristos produced core gistered optical and Gd-subtraction images, allowing comparison of the spatial distrib ution of uptake as well as monitoring of the ICG kinetics. The relative sensitivities of the techniques may not be immediately apparent from the e xperimental protocol: Ntziachristos describes injections of 0.1 mmol/kg of Gd-DTPA and 0.25 mg/kg of ICG (per kilo gram bodyweight). ICG has a molecular w eight of 775 g/mol, so this cor responds to an injection of ~0.3 µmol/kg for ICG, a factor of ~300 less in concentration.

Dynamic and Physical Contrasts In addition to the breast cancer contrasts discussed above, researchers ha ve also e xamined the possibility of using external modulations of intrinsic contrast. The Valsalva maneuv er (e xhaling against a closed glottis) raises pressure in the chest ca vity and possib ly restricts venous return from the breasts of a woman lying in the prone position, thus increasing contrast in tumors.63 External compression of the malleab le breast can force out blood, reducing Hb t, or even restrict blood flow; several groups78,146,180,181 are currently working on compression-related projects. Restricted flow impairs the delivery of contrast agents, making this w ork of interest to researchers in other modalities. Additionally, both DOBI and ViOptix78 are developing commercial de vices using pressure modulation (see Section “Commercial Dif fuse Optical Instruments”). Vascular dynamics can be modif ied through changes to inhaled gases. A carbo gen (CO 2 and O 2 mixture) inhaled contrast agent has been applied to animal models by Kotz et al.,182 and to the palms of human volunteers by Dixit et al.,183 Applications of this technique to breast cancer are ongoing. 184*

216

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Total Hemoglobin Concentration

5

1.3 1

Blood Oxygen Saturation

Hbt

4.5

Sat. µ s ‘[786]

4

ICG Fluoro

1.2

3.5

µs ′ (786 nm)

1.5 1

Fluorescence

Contrast

1

3 2.5 2 1.5

4.5

1 1

0.5 y 5 5 cm

0

1

2

3

4

5

6

7

cm

Figure 6. ICG fluorescence contrast in 3D tomographic imaging of the human breast; a single craniocaudal slice is shown for a 52-year-old postmenopausal women. ~1.5 cm diameter retroareolar invasive carcinoma was identified by a multimodality (Gd-DTPA MRI, ultrasound, PET, mammogram) study. Left: images of Hbt, Sat., µʹ′s, and ICG fluorescence. Right: line plots, corresponding to the magenta line in images on the left, showing the lesion contrast for each chromophore. Adapted from Corlu A et al.106

DOT/SPECTROSCOPY DATA ANALYSIS Diffuse optical instruments differ in the quantity of spatial, spectral, and temporal information provided; this is represented graphically in Figure 7. Spatially, an instr ument with fe w source-detector pairs collects re gional (a few cm3 of tissue) information, whereas an instrument with hundreds or thousands of source-detector pairs can yield a 3D tomo graphic reconstr uction. Spectrall y, an instrument with onl y a single w avelength can monitor changes in absor ption at that w avelength and therefore track the kinetics of a contrast agent, but it cannot separate two simultaneously changing chromophores, both of which absorb at the w avelength used. A multi-w avelength or e ven broadband instr ument can distinguish between several chromophores. Instruments with temporal information (TD or FD) can separate absor ption and scattering, allowing absolute quantif ication of the chromophore concentrations by solving Equation 15 for C t. The infor mation content of dif fuse optical measurements therefore has dimensions of frequenc y (TD data has many frequencies), source-detector pairs, and w avelength. The most basic instr ument consists of a single sourcedetector combination and one w avelength, with a continuous ( ω = 0) light source, v ery similar to the 2 wavelength system described by Chance et al.,185 and later commercialized as the “R UNMAN.” The instr ument with the most complete data set is TD (or swept frequency FD), †

with many source and detector † positions and wavelengths. In the “wavelength” and “frequency” dimensions, the most complete data set so f ar reported is the 32 w avelength, TD instrumentation de veloped b y the g roup at P olitecnico di Milano for scanning mammo graphy186 and brain.187 The source-detector dimension is maximized b y Choe et al., 40 (4 104 source-detector pairs) for human studies and 10 7 pairs b y K onecky et al., 26 in phantom studies. The cost, complexity, and measurement time roughly correlated with each dimension and therefore the richness or completeness of the data set. For example, a pulsed super-continuum light source cur rently costs ~$70 k w hile constant output LED might be available for $1.

Clinical Instrument Design However, the question a clinical instrument designer asks is not “How do I make an instrument to collect the richest data set?”, but rather “What infor mation is required to address the clinical question of interest and w hich instrument will provide that infor mation most economicall y in time, complexity, and cost, with both clinician and patient acceptance?” The instr uments described in this chapter dif fer in the clinical need addressed and therefore in function. We have attempted to group these instruments spatially: local or 3D tomo graphic and functionall y: contrast or physiologic characterization (T able 4). The f irst criterion is deter mined b y ho w man y measurements the

“Sources and Detectors” are sometimes jointly referred to as “optodes”.

Diffuse Optical Tomography and Spectroscopy

Wavelengths

SD Pairs Frequencies Figure 7. Information content of diffuse optical techniques. More wavelengths, source-detector pairs, and frequencies increase the completeness of the data set. Time domain measurements can be Fourier transformed into many frequencies; wavelength data can be transformed into chromophore concentrations. Cost, instrument complexity, and measurement time scale with each dimension.

instrument can mak e on the tissue and , therefore, ho w well data from the instr ument can be reconstr ucted to provide spatial distributions of concentrations. The second criterion is deter mined b y the number of w avelengths available. Measurements at a single w avelength can only provide information on concentration of a single chromophore, but there are se veral major chromophores in tissue. One might choose the isobestic point of Hb and HbO 2 for a single w avelength instr ument to track Hb t over time, but this instr ument could not provide information on oxygen metabolism through blood o xygenation, as one requires at least one w avelength per chromophore reco vered. Ho wever, this instrument might serve well to detect spatial changes in Hbt from a subdural hematoma. The clinical requirements will determine the instrument chosen for the task; the most complete (and comple x) instruments are those which require characterize ph ysiologic infor mation in 3D volumes.

Clinical Questions If we consider the clinical questions diffuse optics attempts to answer, we find that most questions reduce to diagnostic capability: separating one group from another. For example, we might separate out those with head injuries causing bleeding near the surf ace of the brain, abnor mal muscle function, particular patterns of functional brain acti vation, or malignant tumors. In reality, this separation may or may not exist in the multidimensional space probed b y diffuse optics, with each dimension a value or dynamical change in a chromophore or scattering coef ficient. In practice, each instrument probes a subset of the dimensions in the possible information space; see the following sections for several techniques attempting to distill a diagnostic criterion from the space probed.

217

The instr ument designer attempts to choose that subset of dimensions containing the infor mation vital to making the separation between groups. For example, one can simply compute the distance betw een two groups in any one of these dimensions and look for statisticall y significant separations. Functional brain activation studies might look onl y at local increases in Hb t. Alter natively, one can constr uct a v ector from some physiologically sensib le combination of these parameters (see Section “Indices for Breast Cancer Detection in Diffuse Optics”) or from some statistical procedure (see Section “Cluster Analysis to Identify Malignant Breast Tumors”), then look for separations in this composite parameter. These composites are dependent on some type of per -subject nor malization to reduce the lar ge inter-subject variability. DOS/DOT has a strong potential to be complimentary to e xisting clinical modalities: the tomo graphic reconstructions from DOT will not match the 3D images of tissue str ucture from MRI, X-ra y imaging, or ultrasound; however, DOT can provide functional information on tissue metabolism, does not use ionizing radiation, requires relati vely ine xpensive instr umentation, is suitable for long-term monitoring, and has the potential to be very sensiti ve to contrast agents. Thus, DO T and the structural modalities are probing dif ferent dimensions of the infor mation space, allo wing ne w choices of separation metrics. In the language of radiolo gists, the information from dif fuse optics ma y be ab le to impro ve the overall sensitivity and specificity of the clinical exam.

Indices for Breast Cancer Detection in Diffuse Optics As discussed abo ve, clinical dif fuse optical measurements frequentl y attempt to separate one g roup from another. In breast cancer studies, these groups are often malignant tumors from benign lesions or healthy tissue. Several research g roups ha ve de veloped indices that combine se veral of the parameters measured through DOT/DOS to enhance the contrast betw een malignant lesions and nor mal tissue. Researchers at the Uni versity of Califor nia at Ir vine ha ve de veloped and clinically tested a “tissue optical index” (TOI).188,189 Choe et al.,40 have sho wn the ability to distinguish 10 benign from 41 malignant lesions (all biopsy-verified) using a CW/FD tomographic device. Lesion and normal region optical properties were averaged and the lesion contrast compared to sur rounding tissue using nor malized single optical parameters (Hbt, HbO2, µ ʹ′s ) and a combined “optical index.” Choe found her tumor to bulk contrast

218

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 4. INSTRUMENTATION AND DATA TYPES IN DIFFUSE OPTICAL SPECTROSCOPY/DIFFUSE OPTICAL TOMOGRAPHY Local→Few SD

3D→Many SD

Contrast (1–2 λ)

Localization, Monitoring, Pharmacokinetics

3D Pharmacokinetics

Physiologic Characterization (> 3λ)

Local Physiology, Physiology Monitoring

3D distribution of Physiologic Chromophores

SD = source-detector pairs; λ = wavelengths. Many of the characterization instruments can function as contrast instruments by limiting the wavelengths used. This table does not show the temporal dimension of instrument type: TD or FD instruments can fall in any of the above categories.

increased from a maximum of ~1.5 using r s to ~1.8 using the optical inde x; the UCI g roup reports a tumor contrast of ~3 using their TOI. Kukreti et al.,190 have developed “double differential spectroscopy” in their search for the signatures of malignant tumors. This technique looks for the dif ference between spectra measured on a lesion, spectra of health y tissue measured on the same subject, and a linear combination of kno wn chromophore spectra. The “double differential spectra” output from their algorithm contains spectral features specif ic to malignanc y. Lilge’ s group191–193 has de veloped “transillumination breast spectroscopy” (TIBS) in an attempt to deter mine the risk of future cancer development in women with high breast density.

Cluster Analysis to Identify Malignant Breast Tumors Another technique to separate lesion types relies on plotting suitab ly nor malized data and looking for clusters corresponding to lesion types. Chance et al., 3 used eight detectors, one source, and three wavelength hand-held probes to measure the contrast between a tumor -bearing re gion and the contralateral breast. The voxel probed b y each source-detector pair w as about 1 cm 3. The researchers took the source-detector pair of greatest contrast and created a 2D “nomo gram” of ΔHbt and ΔSat. (Figure 8). This protocol was applied to 72 “cancer-free” and 44 biopsy-verified malignant lesions, and the nomogram was segmented into tw o sections, roughl y corresponding to higher Hb t and lower Sat. versus the remaining parameter space (including the region around the origin corresponding to small dif ferences between the reference and test breasts). Cancers fell almost e xclusively into the high Hb t/low Sat. quadrant. This data pro vided high sensitivity, specif icity, and Area Under the Recei ver

Operating Characteristic curve (AUC; 0.96, 0.93, and 0.95, respectively), but these metrics were calculated on the same data set used to separate the tw o re gions (e.g., the training and test sets were identical). The distribution of cancer and nor mal subjects in this plot suggests an optical index cor responding roughly to a line betw een the lower left and upper right hand corner may be useful in separating the cancerous and healthy groups; ongoing research is focused on expanding this to 3D.

Multidimensional Image Analysis Neither of the techniques described in the pre vious sections uses (or gi ves) spatial infor mation on the lesion, beyond the placement of the probe. As cancers are known to be heterogeneous194,195 in gene expression and metabolism,196 this poses ob vious prob lems. Tomographic reconstr uctions for each of se veral chromophores in 3D are dif ficult to visualize, and detecting multiparameter signatures of malignant lesions across these dimensions is e ven more dif ficult. Additionally, DOT reconstr uctions are frequentl y plagued b y image artifacts, especially when only a single chromophore is visualized. Song et al., 96,197,198 have demonstrated statistical image processing tools to separate lesions from background tissue. Se veral research g roups have sought to combine three or more optical parameters to fur ther identify optical signatures of v arious lesion types. Wang et al.,199 use a support vector machine analysis of the refractive index, µa, and µ ʹ′s to identify and separate malignant from benign lesions, with a sensiti vity of ~82% and a specif icity of ~92%. Busch et al., 200 have recently presented a lo gistic regression model on Hbt, Sat., and µ sʹ′ to calculate a probability of malignanc y for each voxel in 3D reconstr uctions of subjects drawn from a study b y Choe, 40 focusing on applications to

Diffuse Optical Tomography and Spectroscopy

REFERENCES

Oxygenation desaturation (%)

2 60 2 50 I

2 40 2 30 2 20

L L

2 10

L L

L L L

0

L

L L L

L

L

L

L

L L L L L L

L

L

L L

L L

10

II

20 30 22

219

0

2 4 6 Blood volume increment (µM)

8

10 JZ 18

Figure 8. DOS measurements of changes in Hbt and Sat., for biopsy-verified malignant lesions, relative to contralateral, cancer-free, breast. Cancers are marked with a “•” and cancer-free subjects are marked with a “x .” This study was conducted at two institutions: the Hospital of the University of Pennsylvania and University of Leipzig (marked with an “L”). Adapted from Chance B et al.3

computer-aided detection of malignancies in dif optical images.

fuse

CONCLUSION Diffuse optics provides micromolar sensitivity to intrinsic contrasts; nanomolar sensiti vity to contrast agents; access to functional/metabolic infor mation; penetration se veral centimeters into human brain, breast, and muscle tissue; and the ability to safel y mak e long-ter m measurements without tissue damage or high cost. Fur thermore, targeted optical probes for molecular imaging are under acti ve development in vitro and in animal studies, some of which are described elsewhere in this book. These targeted probes will increase the tar get-to-background ratio be yond that provided by the cur rent blood pooling agent and thereb y increase the spatial sensitivity of diffuse optics. In this chapter, we have discussed the state of the ar t in dif fuse optics, focusing on biomedical applications. The f ield is e xpanding in scope and mo ving toward significant commercialization for specif ic applications. We anticipate diffuse optical devices for cancer detection and therapy monitoring, as w ell as brain injur y, will become widely available in the next decade.

ACKNOWLEDGMENT The authors wish to thank Re gine Choe and Turgut Durduran for helpful comments during the preparation of this manuscript.

1. Jemal A, Sie gel R, Ward E, et al. Cancer statistics, 2007. CA Cancer J Clin 2007;57:43–66. 2. Leff DR, Warren OJ, Enfield LC, et al. Diffuse optical imaging of the healthy and diseased breast: a systematic re view. Breast Cancer Res Treat 2008;108:9–22. 3. Chance B, Nioka S, Zhang J, et al. Breast Cancer detection based on incremental biochemical and ph ysiological proper ties of breast cancers a six-year, two-site study. Acad Radiol 2005;12:925–33. 4. Case KM, Zw eifel PF . Linear Transport Theory. Addison-Wesley, ch-8; 1967. Library Congress 67–17256. 5. Klose AD, Hielscher AH. Optical tomography with the equation of radiative transfer. Int J Numer Methods Heat Fluid Flo w 2008;18:443. 6. Patterson MS, Chance B , Wilson BC. Time resolv ed reflectance and transmittance for the nonin vasive measurement of tissue optical properties. Appl Opt 1989;28:2331–6. 7. Jacques SL, Pogue BW. Tutorial on diffuse light transpor t. J Biomed Opt 2008;13. 8. Laser Institute of America. American National Standard for the Safe Use of Lasers: ANSI Z136.1-2000. Orlando (FL): Laser Institute of America. 9. Food and Dr ug Administration. Code of F ederal Regulations, Title 21, Volume 8. US. Go vernment Printing Offce via GPO Access: 2003. 21CFR1040.10. 10. Weissleder R, Ntziachristos V. Shedding light onto live molecular targets. Nat Med 2003;9:123–8. 11. Cheong WF, Prahl SA, Welch AJ. A review of the optical proper ties of biological tissues. IEEE J Quantum Electron 1990;26:2166–85. 12. Chandrasekhar S. Stochastic prob lems in ph ysics and astronom y. Rev Mod Phys 1943;15:1–89. 13. Feynman RP, Leighton RB , Sands ML. The F eynman Lectures on Physics: Commemorative Issue. Reading (MA): Addison-Wesley; 1965. 14. Ishimaru A. Wave Propagation and Scattering in Random Media, v olume I and II of IEEE/OUP series on electromagnetic w ave theory. A classic reissue ed. IEEE Press; 1978. 15. Fishkin JB, Gratton E. Propagation of photon-density waves in strongly scattering media containing an absorbing semi-inf inite plane bounded by a straight edge. J Opt Soc Am A 1993;10:127–40. 16. Haskell RC, Svaasand LO, Tsay TT, et al. Boundary-conditions for the diffusion equation in radiati ve-transfer. J Opt Soc Am 1994; 11:2727–41. 17. Arridge SR, Cope M, Delp y DT. The theoretical basis for the determination of optical pathlengths in tissue: temporal and frequenc y analysis. Phys Med Biol 1992;37:1531–60. 18. Boas DA, Brooks DH, Miller EL, et al. Imaging the body with diffuse optical tomography. IEEE Signal Process Mag 2001;18:57–75. 19. Gibson AP, Hebden JC, Arridge SR. Recent advances in diffuse optical imaging. Phys Med Biol 2005;50:R1–43. 20. Furutsu K, Yamada Y. Diffusion approximation for a dissipative random medium and the applications. Phys Rev E 1994;50:3634–40. 21. Durduran T, Yodh AG, Chance B , Boas D A. Does the photon-dif fusion coeffcient depend on absorption? J Opt Soc Am A Opt Image Sci Vis 1997;14:3358–65. 22. Piao D, Jiang S, Srini vasan S, et al. Video-rate near-infrared optical tomography using spectrall y encoded parallel light deli very. Opt Lett 2005;30:2593–5. 23. Hebden J , Arridge S, Delp y D . Optical imaging in medicine: I. Experimental techniques. Phys Med Biol 1997;42:825–40. 24. Culver JP, Ntziachristos V, Holbok e MJ , Yodh AG. Optimization of optode ar rangements for dif fuse optical tomo graphy: a singularvalue analysis. Opt Lett 2001;26:701–3. 25. Markel VA, Schotland JC. Scanning paraxial optical tomography. Opt Lett 2002;27:1123–5.

220

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

26. Konecky SD, Panasyuk GY, Lee K, et al. Imaging complex structures with diffuse light. Opt Express 2008;16:5048–60. 27. Torricelli A, Pifferi A, Spinelli L, et al. Time-resolved reflectance at null source-detector separation: improving contrast and resolution in dif fuse optical imaging. Ph ys Re v Lett 2005; 95:078101. 28. Spinelli L, Mar telli F, Del Bianco S, et al. Absorption and scattering perturbations in homogeneous and layered diffusive media probed by time-resolv ed reflectance at null source-detector separation. Phys Rev E Stat Nonlin Soft Matter Phys 2006;74(2 Pt 1):021919. 29. Boas D A, Dale AM, F ranceschini MA. Dif fuse optical imaging of brain acti vation: approaches to optimizing image sensiti vity, resolution, and accuracy. Neuroimage 2004;23(Suppl 1): S275–88. 30. Zeff BW, White BR, Dehghani H, et al. Retinotopic mapping of adult human visual cor tex with high-density dif fuse optical tomo graphy. Proc Natl Acad Sci 2007;104:12169–74. 31. Bouguer P. Essai d’Optique sur la g radation de la Lumiere. P aris: C. Jombert; 164: tavv. ft; in 12.; VII. 2.2. a, 1729. 32. Beer A. Einleitung in die Elektrostatik, die Lehre v om Magnetismus und die Elektrodynamik. 1854. 33. Born M, Wolf E. Principles of Optics. 7th (e xpanded) ed. New York: Pergamon Press; 1999. 34. Delpy DT, Cope M, v an der Zee P, et al. Estimation of optical pathlength through tissue from direct time of flight measurement. Phys Med Biol 1988;33:1433–42. 35. Cope M, Delp y DT. System for long-ter m measurement of cerebral blood and tissue oxygenation on newborn infants by near infra-red transillumination. Med Biol Eng Comput 1988;26:289–94. 36. Chance B, Leigh JS, Miyak e H, et al. Comparison of time-resolv ed and -unresolved measurements of deoxyhemoglobin in brain. Proc Natl Acad Sci U S A 1988;85:4971–5. 37. Mourant JR, Fuselier T, Boyer J, et al. Predictions and measurements of scattering and absor ption over broad wavelength ranges in tissue phantoms. Appl Opt 1997;36:949–957. 38. Nilsson AMK, Sturesson C, Liu DL, Andersson-Engels S. Changes in spectral shape of tissue optical proper ties in conjunction with laser-induced thermotherapy. Appl Opt 1998;37:1256–1267. 39. Bevilacqua F, Berger AJ, Cerussi AE, et al. Broadband absorption spectroscopy in turbid media b y combined frequenc y-domain and steady-state methods. Appl Opt 2000;39:6498–507. 40. Choe R, K onecky SD , Corlu A, et al. Dif ferentiation of benign and malignant breast tumors by in-vivo three-dimensional parallel-plate diffuse optical tomography. J Biomed Opt (submitted), Vol 14, 2009. 41. Cuccia DJ, Bevilacqua F, Durkin AJ, Tromberg BJ. Modulated imaging: quantitative analysis and tomography of turbid media in the spatialfrequency domain. Opt Lett 2005;30:1354–6. 42. Selb J, Joseph DK, Boas D A. Time-gated optical system for depthresolved functional brain imaging. J Biomed Opt 2006;11:044008. 43. Intes X, Djeziri S, Ichalalene Z, et al. Time-domain optical mammography softscan: initial results. Acad Radiol 2005;12:934–47. 44. Rinneberg H, Grosenick D, Moesta KT, et al. Detection and characterization of breast tumours b y time-domain scanning optical mammography. Optoelectron Rev 2008;16:147–62. 45. Ntziachristos V, Yodh AG, Schnall M, Chance B . Concur rent MRI and diffuse optical tomography of breast after indocyanine green enhancement. Proc Natl Acad Sci U S A 2000;97:2767–72. VasilisNPNAS99, NtziachristosYSC00. 46. Franceschini MA, Moesta KT , Fantini S, et al. F requency-domain techniques enhance optical mammo graphy: initial clinical results. Proc Natl Acad Sci U S A 1997;94:6468–73. 47. Li XD, Durduran T, Chance B, et al. Diffraction tomography for biochemical imaging with dif fuse-photon density w aves. Opt Lett 1997;22:573–5. 48. Pogue B, Testorf M, McBride T, et al. Instrumentation and design of a frequency-domain diffuse optical tomography imager for breast cancer detection. Opt Express 1997;1:391–403.

49. Corlu A, Choe R, Durduran T, et al. Diffuse optical tomography with spectral constraints and wavelength optimization. Appl Opt 2005; 44:2082–93. 50. Sunar U, Quon H, Durduran T, et al. Non-invasive diffuse optical measurement of b lood flo w and b lood o xygenation for monitoring radiation therapy in patients with head and neck tumors: a pilot study. J Biomed Opt 2006;11:064021. 51. Shah N, Gibbs J , Wolverton D, et al. Combined dif fuse optical spectroscopy and contrast-enhanced magnetic resonance imaging for monitoring breast cancer neoadjuv ant chemotherapy: a case study. J Biomed Opt 2005;10:051503. 52. Li A, Kwong R, Cer ussi A, et al. Method for reco vering quantitative broadband diffuse optical spectra from la yered media. Appl Opt 2007;46:4828–33. 53. Cerussi AE, Jakubowski D, Shah N, et al. Spectroscop y enhances the information content of optical mammo graphy. J Biomed Opt 2002;7:60–71. 54. Kepshire DS, Davis SC, Dehghani H, et al. Subsurface diffuse optical tomography can localize absorber and fluorescent objects but recovered image sensiti vity is nonlinear with depth. Appl Opt 2007;46:1669–78. 55. Ge J , Zhu B , Re galado S, Goda varty A. Three-dimensional fluorescence-enhanced optical tomography using a hand-held probe based imaging system. Med Phys 2008;35:3354–63. 56. Chance B, Nioka S, Sadi S, Li C. Oxygenation and blood concentration changes in human subject prefrontal acti vation by anag ram solutions. Adv Exp Med Biol 2003;510:397–401. 57. Gibson AP, Austin T, Ev erdell NL, et al. Three-dimensional w holehead optical tomography of passive motor evoked responses in the neonate. Neuroimage 2006;30:521–8. 58. Enfield LC, Gibson AP, Everdell NL, et al. Three-dimensional timeresolved optical mammography of the uncompressed breast. Appl Opt 2007;46:3628–38. 59. Poplack SP, Tosteson TD, Wells WA, et al. Electromagnetic breast imaging: results of a pilot study in w omen with abnor mal mammograms. Radiology 2007;243:350–9. 60. Nioka S, Yung Y, Shnall M, et al. Optical imaging of breast tumor by means of continuous w aves. Adv Exp Med Biol 1997;411:227–32. 61. Hoogenraad JH, van der Mark MB, Colak SB, et al. First results from the Philips optical mammoscope. Photon Propagation in Tissues III, 1998. 62. Nielsen T, Brendel B, Ziegler R, et al. Linear image reconstruction for a dif fuse optical mammo graphy system in a noncompressed geometry using scattering fluid. Appl Opt 2009;48:D1–13. 63. Schmitz CH, Klemer DP, Hardin R, et al. Design and implementation of dynamic near-infrared optical tomographic imaging instrumentation for simultaneous dual-breast measurements. Appl Opt 2005;44: 2140–53. 64. Durduran T, Choe R, Culv er JP, et al. Bulk optical proper ties of healthy female breast tissue. Phys Med Biol 2002;47:2847–61. 65. Grosenick D , Wabnitz H, Rinneber g HH, et al. De velopment of a time-domain optical mammograph and f irst in vivo applications. Appl Opt 1999;38:2927–43. 66. Spinelli L, Torricelli A, Pif feri A, et al. Characterization of female breast lesions from multi-wavelength time-resolved optical mammography. Phys Med Biol 2005;50:2489–502. 67. Cutler M. Transillumination of the breast. Ann Surg 1931;93:223–34. 68. Cutler M. Transillumination as an aid in the diagnosis of breast lesions. Surg Gynecol Obstet 1929;48:721–9. 69. Culver JP, Choe R, Holboke MJ, et al. Three-dimensional diffuse optical tomography in the parallel plane transmission geometr y: evaluation of a h ybrid frequency domain/continuous w ave clinical system for breast imaging. Med Phys 2003;30:235–47. 70. Nioka S, Chance B . NIR spectroscopic detection of breast cancer . Technol Cancer Res Treat 2005;4:497–512.

Diffuse Optical Tomography and Spectroscopy

71. Hebden JC, Delp y DT. Diagnostic imaging with light. Br J Radiol 1997;70 Spec No:S206–14. 72. Ntziachristos V, Chance B . Probing ph ysiology and molecular function using optical imaging applications to breast cancer . Breast Cancer Res 2001;3:41–6. 73. Ueda Y, Yamanaka T, Yamashita D , et al. Reflectance dif fuse optical tomography: its application to human brain mapping. Jpn J Appl Phys 2005;44:L1203–6. 74. Ueda Y, Ohta K, Oda M, et al. Three-dimensional imaging of a tissuelike phantom b y dif fusion optical tomo graphy. Appl Opt 2001; 40:6349–55. 75. Gopinath SP , Rober tson CS, Grossman RG, Chance B . Near-infrared spectroscopic localization of intracranial hematomas. J Neurosurg 1993;79:43–7. 76. Zhang Q, Ma H, Nioka S, Chance B . Study of near infrared technology for intracranial hematoma detection. J Biomed Opt 2000; 5:206–13. 77. Schmitz CH, Locker M, Hielscher AH, Barbour RL. Instr umentation for f ast functional optical tomo graphy. Re v Sci Instr um 2002; 73:429–39. 78. Xu RX, Qiang B , Mao JJ, Povoski SP. Development of a hand-held near-infrared imager for dynamic characterization of in vivo biological tissue systems. Appl Opt 2007;46:7442–51. 79. Boas D A. Dif fuse Photon Probes of Str uctural and Dynamical Properties of Turbid Media: Theory and Biomedical Applications. PhD thesis. University of Pennsylvania; 1996. 80. Boas D A, Campbell LE, Yodh AG. Scattering and imaging with diffusing temporal f ield cor relations. Ph ys Re v Lett 1995; 75:1855–8. 81. Stephen MJ. Temporal fluctuations in w ave propagation in random media. Phys Rev B 1988;37:1–5. 82. MacKintosh FC, John S. Dif fusing-wave spectroscopy and multiple scattering of light in cor related random media. Ph ys Re v B 1989;40:2383–406. 83. Zhou C. In-vi vo Optical Imaging and Spectroscop y of Cerebral Hemodynamics. PhD thesis. University of Pennsylvania Philadelphia, Pa USA; 2007. 84. Cheung C, Culver JP, Takahashi K, et al. In vivo cerebrovascular measurement combining diffuse near-infrared absorption and cor relation spectroscopies. Phys Med Biol 2001;46:2053–65. 85. Pine DJ, Weitz DA, Chaikin PM, Herbolzheimer E. Dif fusing wave spectroscopy. Phys Rev Lett 1988;60:1134–7. 86. Yao Y, Wang Y, Pei Y, et al. F requency-domain optical imaging of absorption and scattering distributions by a born iterative method. J Opt Soc Am A 1997;14:325–42. 87. O’Leary MA, Boas D A, Chance B, Yodh AG. Experimental images of hetero geneous turbid media b y frequenc y-domain dif fusingphoton tomography. Opt Lett 1995;20:426–8. 88. Kak AC, Slane y M. IEEE Engineering in Medicine, and Biolo gy Society. Principles of Computerized Tomographic Imaging. Ne w York: IEEE Press; 1988. 89. Arridge SR, Hebden JC. Optical imaging in medicine II modelling and reconstruction. Phys Med Biol 1997;42:841–53. 90. Gaudette RJ, Brooks DH, DiMarzio CA, et al. A comparison study of linear reconstr uction techniques for dif fuse optical tomo graphic imaging of absorption co-effcient. Phys Med Biol 2000; 45:1051–70. 91. Boas D. A fundamental limitation of linearized algorithms for diffuse optical tomography. Opt Express 1997;1:404–13. 92. Hielscher AH, Bluestone AY, Abdoulaev GS, et al. Near-infrared diffuse optical tomography. Dis Markers 2002;18:313–37. 93. Arridge SR, Schw eiger M, Hiraoka M, Delp y DT. A f inite element approach to modelling photon transpor t in tissue. Med Ph ys 1993;20:299–309. 94. Schweiger M, Arridge SR, Delp y DT. Application of the f inite element method for the forw ard and in verse prob lem in optical tomography. J Math Imaging Vis 1993;3:263–83.

221

95. Arridge SR. Optical tomo graphy in medical imaging. In verse Probl 1999;15:R41–93. 96. Pogue BW, Davis SC, Song X, et al. Image analysis methods for diffuse optical tomography. J Biomed Opt 2006;11:33001. 97. Markel VA, Schotland JC. In verse prob lem in optical dif fusion tomography. I. Fourier-Laplace inversion formulas. J Opt Soc Am A 2001;18:1336–47. 98. Schotland JC, Haselgrove JC, Leigh JS. Photon hitting density. Appl Opt 1993;32. 99. Hubbell JH, Seltzer SM. Tables of x-ray mass attenuation coeffcients and mass energy-absorption coeffcients 1 keV to 20 MeV for elements to 92 and 48 additional substances of dosimetric interest. Gaithersburg (MD): Technical repor t, Ionizing Radiation Di vision, Ph ysics Laborator y, National Institute of Standards and Technology; 1995. 100. Zwillinger D, editor. Standard Mathematical Tables and F ormulae. 30th edition, ch-2.5. CRC Press; 1996. 101. Penrose R. A generalized inverse for matrices. Proc Camb Philol Soc 1955;51:406–13. 102. Tikhonov AN. On the stability of in verse problems. Dokl Akademii Nauk izvestija Poossiiskoi [Proceedings of the Russian Academy of Sciences], 1943;39:195–8. 103. Contini D, Martelli F, Zaccanti G. Photon migration through a turbid slab described b y a model based on dif fusion appro ximation. I. Theory. Appl Opt 1997;36:4587–99. 104. Pogue BW, Patterson MS, Jiang H, Paulsen KD. Initial assessment of a simple system for frequency domain diffuse optical tomography. Phys Med Biol 1995;40:1709–29. 105. Paulsen KD, Jiang H. Spatiall y varying optical proper ty reconstruction using a finite element diffusion equation approximation. Med Phys 1995;22:691–701. 106. Corlu A, Choe R, Durduran T, et al. Three-dimensional in vivo fluorescence diffuse optical tomo graphy of breast cancer in humans. Opt Express 2007;15:6696–716. 107. Walker SA, F antini S, Gratton E. Image reconstr uction by back projection from frequenc y-domain optical measurements in highl y scattering media. Appl Opt 1997;36:170–174. 108. Deliolanis N, Lasser T, Hyde D, et al. Free-space fluorescence molecular tomo graphy utilizing 360 geometr y projections. Opt Lett 2007;32:382–4. 109. Markel VA, Schotland JC. In verse prob lem in optical dif fusion tomography. II. Role of boundar y conditions. J Opt Soc Am A 2002;19:558–66. 110. Markel VA, O’Sullivan JA, Schotland JC. Inverse problem in optical diffusion tomography. IV. Nonlinear inversion formulas. J Opt Soc Am A 2003;20:903–12. 111. Shah N, Cerussi AE, Jakubowski D, et al. Spatial v ariations in optical and physiological properties of healthy breast tissue. J Biomed Opt 2004;9:534–40. 112. Poplack SP, Paulsen KD, Hartov A, et al. Electromagnetic breast imaging: average tissue property values in women with negative clinical findings. Radiology 2004;231:571–80. 113. Pogue BW, Poplack SP, McBride TO, et al. Quantitati ve hemoglobin tomography with dif fuse near-infrared spectroscopy: pilot results in the breast. Radiology 2001;218:261. 114. Zhu Q, K urtzma SH, He gde P, et al. Utilizing optical tomo graphy with ultrasound localization to image hetero geneous hemo globin distribution in lar ge breast cancers. Neoplasia 2005; 7:263–70. 115. Li C, Grobm yer SR, Massol N , et al. Nonin vasive in vi vo tomographic optical imaging of cellular mor phology in the breast: possible con vergence of microscopic patholo gy and pomacroscopic radiology. Med Phys 2008;35:2493–501. 116. Jobsis FF. Noninvasive, infrared monitoring of cerebral and myocardial oxygen suffciency and circulatory parameters. Science 1977; 198:1264–7.

222

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

117. Chance B , Nioka S, K ent J , et al. Time-resolved spectroscop y of hemoglobin and myoglobin in resting and ischemic muscle. Anal Biochem 1988;174:698–707. 118. Chance B, Dait MT, Zhang C, et al. Reco very from e xercise-induced desaturation in the quadriceps muscles of elite competiti ve rowers. Am J Physiol 1992;262(3 Pt 1):C766–75. 119. Belardinelli R, Barsto w TJ, P orszasz J , Wasserman K. Changes in skeletal muscle oxygenation during incremental exercise measured with near infrared spectroscop y. Eur J Appl Ph ysiol 1995; 70:487–92. 120. Boushel R, Langber g H, Olesen J , et al. Monitoring tissue o xygen availability with near infrared spectroscop y (NIRS) in health and disease. Scand J Med Sci Spor ts 2001;11:213. 121. Wolf M, F errari M, Quaresima V. Pro gress of near -infrared spectroscopy and topo graphy for brain and muscle clinical applications. J Biomed Opt 2007;12:062104. 122. Gopinath SP, Rober tson CS, Contant CF , et al. Earl y detection of delayed traumatic intracranial hematomas using near -infrared spectroscopy. J Neurosurg 1995;83:438–44. 123. Kirkpatrick PJ, Smielewski P, Czosnyka M, et al. Near-infrared spectroscopy use in patients with head injur y. J Neurosurg 1995;83:963–70. 124. Buchner K, Meixensberger J, Dings J, Roosen K. Near -infrared spectroscopy–not useful to monitor cerebral o xygenation after se vere brain injury. Zentralbl Neurochir 2000;61:69–73. 125. Austin EH, Edmonds HL, Auden SM, et al. Benef it of neurophysiologic monitoring for pediatric cardiac sur gery. J Thorac Cardiovasc Surg 1997;114:707–15, 717; discussion 715–6. 126. Casati A, Fanelli G, Pietropaoli P , et al. Monitoring cerebral o xygen saturation in elderly patients undergoing general abdominal surgery: a prospective cohort study. Eur J Anaesthesiol 2007;24:59–65. 127. Hueber DM, Franceschini MA, Ma HY, et al. Non-invasive and quantitative near-infrared haemoglobin spectrometry in the piglet brain during h ypoxic stress, using a frequenc y-domain multidistance instrument. Physics in Medicine and Biology, 2001;49:41–62. 128. C. Zhou., S. Euck er, T. Durduran, G. Q . Yu, S. F riess, R. Ichord , S.Margulies, and A. G. Yodh. Diffuse optical monitoring of hemodynamics in piglet brain with head trauma injury. J. Neurotrauma., In review, 2009. 129. Al-Rawi PG. Near infrared spectroscopy in brain injury: today’s perspective. Acta Neurochir Suppl 2005;95:453. 130. Strangman G, Boas DA, Sutton JP. Non-invasive neuroimaging using near-infrared light. Biol Psychiatry 2002;52:679–93. 131. Hillman EM. Optical brain imaging in vivo: techniques and applications from animal to man. J Biomed Opt 2007;12:051402. 132. Ntziachristos V, Ma XH, Chance B . Time-correlated single photon counting imager for simultaneous magnetic resonance and near infrared mammography. Rev Sci Instrum 1998;69:4221–33. 133. Brooksby B , Jiang S, Dehghani H, et al. Combining near -infrared tomography and magnetic resonance imaging to study in vi vo breast tissue: implementation of a laplacian-type re gularization to incorporate magnetic resonance str ucture. J Biomed Opt 2005; 10:051504. 134. Zhu Q, Durduran T, Ntziachristos V, et al. Imager that combines near-infrared dif fusive light and ultrasound. Opt Lett 1999; 24:1050–2. 135. Holboke MJ, Tromberg BJ, Li X, et al. Three-dimensional dif fuse optical mammo graphy with ultrasound localization in a human subject. J Biomed Opt 2000;5:237–47. 136. Ohnishi M, Masaki S, Honda K, et al. Ev oked hemodynamic response of the auditory cortex measured by magnetoencephalography and near -infrared spectroscop y. J Acoust Soc Am 1996; 100:2630. 137. Sander TH, Liebert A, Mackert BM, et al. Dc-magnetoencephalo graphy and time-resolv ed near-infrared spectroscopy combined to

138.

139.

140. 141.

142.

143.

144.

145.

146.

147.

148.

149.

150.

151.

152.

153.

154.

155.

156.

study neuronal and vascular brain responses. Physiol Meas 2007; 28:651–64. Li A, Miller EL, Kilmer ME, et al. Tomographic optical breast imaging guided by three-dimensional mammography. Appl Opt 2003; 42:5181–90. Zhang Q, Br ukilacchio TJ, Li A, et al. Core gistered tomo graphic X-ray and optical breast imaging: initial results. J Biomed Opt 2005;10:024033. Ntziachristos V, Culver JP, Rice BW. Small-animal optical imaging. J Biomed Opt 2008;13:011001. Steinbrink J, Villringer A, Kempf F, et al. Illuminating the bold signal: combined fMRI-fNIRS studies. Magn Reson Imaging 2006;24:495–505. Zhu Q, Cronin EB, Currier AA, et al. Benign v ersus malignant breast masses: optical dif ferentiation with us-guided optical imaging reconstruction. Radiology 2005;237:57–66. Jiang H, Li C, P earlstone D , F ajardo LL. Ultrasound-guided microwave imaging of breast cancer: tissue phantom and pilot clinical experiments. Med Phys 2005;32:2528–35. Zhou C, Choe R, Shah N , et al. Diffuse optical monitoring of b lood flow and o xygenation in humanbreast cancer during earl y stages of neoadjuv ant chemotherap y. J Biomed Opt [in press], 2007;v12:051903. Boverman G, Fang Q, Car p SA, et al. Spatio-temporal imaging of the hemo globin in the compressed breast with diffuse optical tomography. Phys Med Biol 2007;52:3619–41. Carp SA, Kauffman T, Fang Q, et al. Compression-induced changes in the ph ysiological state of the breast as obser ved through frequency domain photon mig ration measurements. J Biomed Opt 2006;11:064016. Brooksby B, Pogue BW, Jiang S, et al. Imaging breast adipose and fibroglandu-lar tissue molecular signatures by using hybrid MRIguided near-infrared spectral tomography. Proc Natl Acad Sci U S A 2006;103:8828–33. Carpenter CM, P ogue BW, Jiang S, et al. Image-guided optical spectroscopy pro vides molecular -specific infor mation in vi vo: MRI-guided spectroscop y of breast cancer hemo globin, w ater, and scatterer size. Opt Lett 2007;32:933–5. Azar FS, Lee K, Khamene A, et al. Standardized platform for coregistration of nonconcurrent diffuse optical and magnetic resonance breast images obtained in dif ferent geometries. J Biomed Opt 2007;12:051902. Konecky SD, Choe R, Corlu A, et al. Comparison of diffuse optical tomography of human breast with w hole-body and breastonly positron emission tomo graphy. Med Ph ys 2008;35: 446:10.11181/1.2826560. Sevick-Muraca EM, Houston JP , Gurf inkel M. Fluorescenceenhanced, near infrared diagnostic imaging with contrast agents. Curr Opin Chem Biol 2002;6:642–50. Nioka S, Colak SB, Li X, et al. Breast tumor images of hemodynamic information using a contrast agent with backprojection and FFT enhancement. In: Advances in Optical Imaging and Photon Mitration (AOIPM), Optical Society of America; 1998. p. 266–270. Cuccia DJ, Bevilacqua F, Durkin AJ, et al. In vivo quantification of optical contrast agent dynamics in rat tumors b y use of dif fuse optical spectroscopy with magnetic resonance imaging core gistration. Appl Opt 2003;42:2940–50. Sevick-Muraca EM, Rasmussen JC. Molecular imaging with optics: primer and case for near -infrared fluorescence techniques in personalized medicine. J Biomed Opt 2008;13:041303. Reynolds JS, Troy TL, Ma yer RH, et al. Imaging of spontaneous canine mammar y tumors using fluorescent contrast agents. Photochem Photobiol 1999;70:87–94. Klohs J, Wunder A, Licha K. Near-infrared fluorescent probes for imaging vascular pathophysiology. Basic Res Cardiol 2008; 103:144–51.

Diffuse Optical Tomography and Spectroscopy

157. Fox IJ, Brooker LGS, Heseltine D W, et al. A tricarbocyanine dye for continuous recording of dilution curves in whole blood independent of v ariations in b lood o xygen saturation. Ma yo Clin Proc 1957;32:478–84. 158. Fox IJ , Wood EH. Indoc yanine g reen: ph ysical and ph ysiological properties. Mayo Clin Proc 1960;35:732–44. 159. Cherrick GR, Stein SW, Leevy CM, Davidson CS. Indocyanine green: observations on its ph ysical proper ties, plasma deca y, and hepatic extraction. J Clin Invest 1960;39:592–600. 160. Svensson CK, Edw ards DJ, Mauriello PM, et al. Ef fect of food on hepatic blood flow: implications in the “food ef fect” phenomenon. Clin Pharmacol Ther 1983;34:316–23. 161. Edwards DJ, Babiak LM, Beckman HB . The effect of a single oral dose of ethanol on hepatic b lood flow in man. Eur J Clin Pharmacol 1987;32:481–4. 162. Akman A, Kadayifilar S, Oto S, Aydin P. Indocyanine green angiographic features of traumatic choroidal r uptures. Ey e 1998; 12(Pt 4):646–50. 163. Baumal CR, Reichel E, Duk er JS, et al. In-doc yanine g reen hyperfluorescence associated with serous retinal pigment epithelial detachment in age-related macular de generation. Ophthalmology 1997;104:761–9. 164. Bischoff PM, Flower RW. Doc Ophthalmol 1985;60:235–91. 165. Imai T, Takahashi K, Fukura H, Morishita Y. Measurement of cardiac output b y pulse dy e densitometr y using indoc yanine g reen: a comparison with the ther modilution method. Anesthesiology 1997;87:816–22. 166. He YL, Tanigami H, Ueyama H, et al. Measurement of b lood volume using indocyanine green measured with pulse-spectrophotometry: its reproducibility and reliability . Crit Care Med 1998;26:1446. 167. Frangioni JV. In vivo near-infrared fluorescence imaging. Curr Opin Chem Biol 2003;7:626–34. 168. Landsman ML, Kw ant G, Mook GA, Zijlstra WG. Light-absorbing properties, stability , and spectral stabilization of indoc yanine green. J Appl Physiol 1976;40:575–83. 169. Brasch R, F ournier-Dujardin L, Lucidi V, et al. Toward the goal of signal quantif ication in contrast-enhanced optical imaging. Acad Radiol 2005;12:73–4. 170. Hope-Ross M, Yannuzzi LA, Gragoudas ES, et al. Adverse reactions due to indo-cyanine green. Ophthalmology 1994;101:529–33. 171. Tung C-H. Fluorescent peptide probes for in vi vo diagnostic imaging. Biopolymers 2004;76:391–403. 172. Licha K, Schir ner M, Henr y G. Optical agents. Handb Exp Pharmacol 2008;185:203–22. 173. Liebert A, Wabnitz H, Obrig H, et al. Non-in vasive detection of fluorescence from e xogenous chromophores in the adult human brain. Neuroimage 2006;31:600–8. 174. Intes X, Ripoll J , Chen Y, et al. In vi vo continuous-wave optical breast imaging enhanced with indoc yanine g reen. Med Ph ys 2003;30:1039–47. 175. Sevick-Muraca EM, Shar ma R, Rasmussen JC, et al. Imaging of lymph flow in breast cancer patients after microdose administration of a near -infrared fluorophore: feasibility study . Radiolo gy 2008;246:734–41. 176. Patel J, Marks K, Roberts I, et al. Measurement of cerebral blood flow in newborn infants using nearinfrared spectroscopy with indocyanine green. Pediatr Res 1998;43:34–9. 177. Gora F, Shinde S, El well CE, et al. Nonin vasive measurement of cerebral blood flow in adults using near-infrared spectroscopy and indo-cyanine green: a pilot study. J Neurosurg Anesthesiol 2002; 14:218–22. 178. Liebert A, Wabnitz H, Steinbrink J , et al. Bed-side assessment of cerebral perfusion in strok e patients basedon optical monitoring

179.

180.

181.

182. 183.

184.

185.

186.

187.

188. 189.

190.

191.

192.

193.

194. 195. 196.

197.

223

of a dy e bolus b y time-resolved diffuse reflectance. Neuroimage 2005;24:426–35. Keller E, Nadler A, Imhof HG, et al. Ne w methods for monitoring cerebral o xygenation and hemodynamics in patients with subarachnoid hemor rhage. Acta Neurochir Suppl 2002;82: 87–92. Jiang S, Pogue BW, Paulsen KD, et al. In vivo near-infrared spectral detection of pressure-induced changes in breast tissue. Opt Lett 2003;28:1212–4. Busch DR, Choe R, Durduran T, et al. Measurement of microvascular blood flow in the human breast during compression with diffuse cor relation spectroscopy. In Photonics West BIOS 2009. SPIE; 2009:7174–75. Kotz KT, Dixit SS, Gibbs AD, et al. Inspirator y contrast for in vi vo optical imaging. Opt Express 2008;16:19–31. Dixit, S. S.; Kim, H.; Visser, B. & Faris, G. Development of a Transillumination Infrared Modality for Dif ferential Vasoactive Optical Imaging, Applied Optics, 2009;100949. Conference (dixitBioMed2008) Dixit, S.; Kim, H.; Visser, B.; Comstock, C. & F aris, G. Hypero xic/Hypercapnic Gas Inhalation as a Route to Increase Contrast from Tumor Tissue in Near -Infrared Imaging of Breast Tissue OSA BioMed, 2008. Chance B, Luo Q, Nioka S, et al. Optical in vestigations of physiology. A study of intrinsic and extrinsic biomedical contrast. Philos Trans R Soc B Biol Sci 1997;352:707–16. Article (Bassi2006) Bassi, A.; Spinelli, L.; D'Andrea, C.; Giusto, A.; Swartling, J.; Pifferi, A.; Torricelli, A. & Cubeddu, R. F easibility of w hite-light time-resolv ed optical mammo graphy. J Biomed Opt 2006;11:054035. Comelli D , Bassi A, Pif feri A, et al. In vi vo time-resolv ed reflectance spectroscopy of the human forehead. Appl Opt 2007; 46:1717–25. Shah N. The role of diffuse optical spectroscopy in the clinical management of breast cancer. Dis Markers 2004;19:95–105. Cerussi A, Shah N , Hsiang D , et al. In vi vo absor ption, scattering, and physiologic proper ties of 58 malignant breast tumors determined by broadband dif fuse optical spectroscop y. J Biomed Opt 2006;11:044005. Kukreti S, Cer ussi A, Tromberg B , Gratton E. Intrinsic tumor biomarkers revealed by novel double-differential spectroscopic analysis of near -infrared spectra. J Biomed Opt 2007; 12:020509. Blyschak K, Simick M, Jong R, Lilge L. Classif ication of breast tissue density by optical transillumination spectroscopy: optical and physiological ef fects go verning predicti ve v alue. Med Ph ys 2004; 31:1398–414. Simick MK, Jong R, Wilson B, Lilge L. Non-ionizing near -infrared radiation transillumination spectroscop y for breast tissue density and assessment of breast cancer risk. J Biomed Opt 2004; 9:794–803. Blackmore KM, Knight JA, Jong R, Lilge L. Assessing breast tissue density by transillumination breast spectroscop y (tibs): an intermediate indicator of cancer risk. Br J Radiol 2007;80:545–56. Fidler IJ. Tumor heterogeneity and the biology of cancer invasion and metastasis. Cancer Res 1978;38:2651–60. Reya T, Mor rison SJ, Clarke MF, Weissman IL. Stem cells, cancer , and cancer stem cells. Nature 2001;414:105–11. Chen Y, Zheng G, Zhang ZH, et al. Metabolism-enhanced tumor localization by fluorescence imaging: in vivo animal studies. Opt Lett 2003;28:2070–2. Song X, P ogue BW , Tosteson TD, et al. Statistical anal ysis of nonlinearly reconstr ucted near -infrared tomo graphic images. II. Experimental inter pretation. IEEE Trans Med Imaging 2002; 21:764–72.

224

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

198. Song X, Pogue BW, Jiang S, et al. Automated region detection based on the contrast-to-noise ratio in near -infrared tomography. Appl Opt 2004;43:1053–62. 199. Wang JZ, Liang X, Zhang Q, et al. Automated breast cancer classification using near -infrared optical tomo graphic images. J Biomed Opt 2008;13:044001.

200. Busch DR, Choe R, Durduran T, et al. Tissue-type image segmentation in optical mammo graphy with population-deri ved probability functions: a step towards optical computer aided diagnosis. In: Photonics West BIOS 2009. SPIE; 2009;7174–30.

15 ULTRASOUND F. STUART FOSTER, PHD, KEVIN CHEUNG, MD, AND EMMANUEL CHERIN, PHD

Ultrasound is one of the most widel y a vailable and pervasive technolo gies used in clinical imaging toda y. Ultrasound images are essentiall y maps of tissue elastic properties derived from the same pulse-echo principles as solar near -surface acti ve-region rendering. The simplicity, ease of use, speed , and safety of ultrasound ha ve led to a signif icant role in diagnosis, treatment assessment, follow-up, and guidance of therap y.1 Conventional ultrasound images (B-scans) primarily report tissue structure, but functional imaging is possib le b y measuring b lood flow or, more precisely, blood velocity using the Doppler principle. Although late to enter the molecular imaging f ield, ultrasound’s competiti ve adv antages mak e it an attractive alter native to other modalities in this arena. Ultrasound imaging is nonin vasive, high resolution, inherently real time, and e xaminations are routinel y completed in a matter of minutes. Additionally, ultrasound imaging does not require dedicated facilities, ionizing radiation, or radioactive contrast agents.2 The idea that biocolloids consisting of microb ubbles (MBs) or nanoparticles could be coupled to ligands for specif ic endothelial cell surf ace proteins and imaged to repor t molecular targets was developed in the late 1990s in the laboratories of Lanza, 3–5 Unger,6,7 and Lindner.2,8 However, until recentl y, it has suf fered from a paucity of molecular imaging contrast agents and fe w strate gies for tar geting and quantifying useful biomark ers. Researchers over the last se veral years have shown that MB ultrasound contrast agents can be selecti vely targeted to intra vascular molecular mark ers of v arious disease processes, including inflammation,9–12 thrombosis,7,13,14 and tumor angio genesis.15–19 Detection sensitivities on the order of a single MB are possib le.20 The resolution of clinical scanners still remains too low to resolv e some critical anatomical features and microcirculation in mice. The de velopment of

noninvasive, high-frequenc y microultrasound imaging platforms by Foster and colleagues 21 has enabled 30 to 150 µm resolution to be achieved in the mouse creating the oppor tunity for v ery high-resolution preclinical molecular imaging. Because MBs are conf ined to the v ascular compartment, they are true intravascular tracers as opposed to diffusible tracers a vailable with other modalities. As a consequence, the reper toire of tar gets available for MB contrast agents are limited to those e xpressed on the surfaces of endothelial cells. F ortunately, interesting opportunities abound to measure interactions with tar gets that are of fundamental rele vance to angiogenesis, apoptosis, inflammation, and thrombus formation. The interaction of ultrasonic energy with MB populations is comple x. At lo w pressures (< ~50 kilo P ascals [kPa]), MBs oscillate linearl y in synchron y with the insonifying pressure w ave. At inter mediate pressures between about 50 and 500 kP a, MBs under go nonlinear oscillations that can be used to separate them from surrounding tissue. Above about 500 kPa, bubbles can be disrupted thereby eliminating their capacity to retur n echoes for imaging pur poses. All three of these pressure ranges are used in molecular imaging and therapeutic targeting in bioresearch toda y. F or an e xcellent re view of bubb le physics, targeting and therapeutic aspects of MBs, refer to the study by Ferrara and colleagues.22 The purpose of this chapter is to e xamine the de velopment of ultrasound instrumentation and its use for molecular imaging.

ULTRASOUND FUNDAMENTALS All ultrasound imaging systems rel y on the propagation of mechanical (ultrasonic) w aves from a transducer into tissue and the subsequent interactions of these waves with mechanical discontinuities in the tissue. At each discontinuity, a small reflected w ave (scatter), called an echo, is 225

226

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

created and its ener gy is detected b y the transmitting transducer. The range, z, of the scattering str ucture is determined by the time-of-flight of the ultrasound to and from the target via the simple equation:

ct z = , 2

(1)

where c is the speed of sound. The propagation of ultrasound in a medium implies a traveling wave for which the wavelength, frequency, and speed of sound are go verned by the wave equation:

∂2 p 2 2 c = ∇ p, 2 ∂t

(2)

where p represents acoustic pressure, c is the speed of sound and ∇ represents the second-order spatial deri vative (Laplacian) operating on pressure. On the basis of this equation, it can be shown as follows:

c λ= . f

(3)

Assuming linearity, equation 2 has har monic solutions of the form:

{

}

p( r , t ) = Re P( r ) ⋅ e jωt ,

(4)

in w hich P represents a comple x spatiall y dependent solution to the dif fraction inte gral for specif ic aper ture boundary conditions. A complete examination of diffraction is be yond the scope of this chapter , but P(r) can be thought of as the spatial distribution of pressure arising from the transmission of ultrasound from a par ticular transducer aper ture. It, therefore, tak es into account the effects of focusing and propagation. Equations 2 and 3 assume linearity , as discussed abo ve, but w hen beams with high peak pressures are transmitted into tissue, nonlinear effects can become impor tant and can, in f act, be exploited to impro ve image quality . Nonlinearity arises from the f act that the speed of sound e xpressed in equation 2 is itself a function of pressure and acoustic v elocity. Acoustic velocity, v, in this case is distinct from speed of sound in that it represents the actual v elocity of the molecules arising from the acoustic pressure. A more accurate nonlinear representation of speed of sound in equation 2 is given as follows:

c = ( c0 + βv )2 ,

(5)

where

B β = 1+ 2A

(6)

and B/A represents the ratio of the second- and first-order terms in the e xpansion of the e xpression for pressure. B/A is often refer red to as the “nonlinear parameter” in ultrasound and ranges in v alue from about 5 for w ater to about 12 for f atty tissues. Nonlinear propagation has a significant and cumulati ve ef fect on the propagation of ultrasound at rele vant po wer le vels. Essentiall y, the regions of compression tra vel slightl y f aster than the regions of raref action causing the w ave to shar pen up, ultimately leading to a shock front. An example of nonlinear pulse wave distortion is shown in Figure 1. The low pressure 50 kP a pulse of F igure 1A has a smooth Gaussian shape with no evidence of distortion. However, a similarly shaped pulse propagating to the focus with a higher peak pressure of 3.5 MPa results in significant distortion that shar pens the peaks and b lunts the troughs of the pulse pressure as sho wn in F igure 1B. This leads to the generation of harmonics in the pulse spectrum. Modern ultrasound instr uments routinel y e xploit nonlinear propagation to perfor m “tissue har monic imaging” to improve the quality of ultrasound images.

Resolution The ultimate imaging perfor mance of an y ultrasound scanner is deter mined by the frequenc y, geometr y of its transducer, and tissue proper ties in accordance with the laws of dif fraction. These issues are re viewed by Foster and colleagues 23 who described the trade-of fs betw een resolution, penetration, and system dynamic range. The resolution of the image is deter mined by the beam distribution (lateral direction) and the pulse bandwidth (axial direction). Simple expressions for lateral resolution (Rlat), axial resolution ( Rax), and depth of f ield (DOF) for a spherical radiator at the focus are:

focal length = λ( f − number ), Rlat = λ diameter

(7)

1 c , Rax = 2 BW

(8)

DOF = 7λ( f − number )2 ,

(9)

− where λ is the a verage w avelength, c is the speed of sound, and BW is the bandwidth of the transducer . The

Ultrasound

A

B B

227

axial resolutions of 60 µm and 25 µm, respectively, as shown in Figure 2. Table 1 sho ws a numerical comparison of resolution for several relevant frequencies. Note that fixing the bandwidth at 50% results in axial resolution appro ximately twice that of lateral resolution, but that this is onl y tr ue over a comparatively narrow DOF. This leads to one of the central issues in ultrasound w hich is ho w to s weep the focus over the entire f ield of view in an ef fective manner. Generally this problem is solved by the use of a linear array of transducers in w hich the delay patterns of transmit and receive are manipulated to achie ve a focus that dynamically s weeps from near to f ar. Dynamic focusing and phased linear arrays are firmly entrenched as the dominant technology for clinical ultrasound imaging. Unfor tunately, technological barriers restrict the use of linear arrays to frequencies less than about 15 MHz. Consequentl y, microultrasound imaging in the 20 to 60 MHz range is perfor med with single element focused transducers with their consequent limitations.

Reflectivity and Signal Strength

Figure 1. Ultrasound pulses from a spherical source converging to a tight beam at the focus of a 50 MHz 3 mm diameter f/2 transducer. A, A low pressure beam reaches the focus with peak positive and negative pressures of approximately 50 kPa. The pulse is symmetric and Gaussian. B, A high pressure source, however, shows significant distortion and asymmetry with peak positive pressure of 3.5 MPa and peak negative pressure of 2 MPa. This distortion is manifest as harmonic content in the spectrum of the pulse and must be carefully considered relative to nonlinear distortion caused by microbubble contrast agents.

ratio of the focal length to the diameter is often referred to as the f-number . Equation 7 is v alid onl y for a focused transducer in the focal zone or for an unfocused transducer at axial distances beyond the far-field transition. Figure 2 shows plots of equation 7 for transducers with f-numbers ranging from 1 to 5. Depending on the choice of frequency and f-number, the resolution can vary over several orders of magnitude. Thus, a typical clinical system with a center frequenc y of 5 MHz, an f-number of 2, and 50% BW (50% of center frequenc y) will have lateral resolution of 0.6 mm and axial resolution of 300 µm. By comparison, a microultrasound scanner with similar geometr y but with a center frequency of 50 MHz will have lateral and

Signal strength or brightness in ultrasound images is determined by the reflectivity of the tissues being imaged. From a theoretical point of vie w, prediction of the reflected signal requires solution of the inhomogeneous wave equation and detailed knowledge of the mechanical properties of tissue. In particular, the acoustic impedance, Z, defined as the product of density and speed of sound:

Z = ρ0 c

(10)

Figure 2. Lateral resolution (Eq. 7) as a function of frequency for f-numbers ranging from f/1 to f/5.

228

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 1. COMPARISON OF RESOLUTION AND DEPTH OF FIELD FOR A SPHERICALLY FOCUSED TRANSDUCER WITH AN F-NUMBER OF 2 Frequency (MHz)

Axial Resolution (mm)

Lateral Resolution (mm)

Depth of Field (mm)

5

300

600

8.4

25

60

120

1.68

50

30

60

0.84

is critical in deter mining reflecti vity. In general, tw o regimes are impor tant: (1) reflection at surf aces with − dimensions >> λ and (2) situations w here the reflector − dimension is 20 MHz) microultrasound was first demonstrated by Sherar and colleagues34 to image the str ucture of tumor spheroids and later emer ged in clinical imaging of the e ye and skin.23,33,35–37 Unfortunately, current technology limitations restrict the use of transducer ar rays to less than about 15 MHz so that microultrasound requires mechanical actuation of highly focused single element transducers as shown in Figure 6. The principles of the block diagram of Figure 5 apply, but the scanhead (see F igure 6, lo wer) is acti vated mechanically to sweep the beam of the imaging transducer across the tissue to generate images. The beamformer is, of course, replaced with single channel processing. Operation at higher frequencies results in less penetration of ultrasound because attenuation increases signif icantly with frequency. However, at 30 MHz, a f ield of view of approximately 15 × 15 mm is possib le, making it w ell suited for applications in the mouse. Microultrasound for mouse imaging w as introduced in 1995 b y Turnbull and colleagues38 in a study of early mouse embryonic brain development. In 2002, F oster and colleagues 21 introduced a practical mouse imaging system that has seen widespread commercial distribution (V isualSonics Inc, Toronto, Canada).

Figure 5. Block diagram of an ultrasound imaging system. Dynamic delay patterns are used by the beamformer to focus the ultrasound on transmit and receive. Specific pulse sequences are used to optimize detection of nonlinear phenomena associated with targeted microbubbles used in molecular imaging.

MOLECULAR IMAGING WITH ULTRASOUND Molecular Imaging with Diagnostic Frequency Ultrasound As described previously, MB contrast agents or nanopar ticles are the cor nerstones of tar geted molecular imaging with ultrasound. The concept of tar geted acousticall y active agents arose in the early 1990s. The concept of molecular imaging with ultrasound is illustrated in F igure 7. Essentially, particles or MBs in the micron size range have ligands embedded in their shells that target endothelial cell surface proteins and receptors. The incoming and scattered ultrasound waves are sho wn schematically in F igure 7A. These particles contact and bind their target (Figure 7B) as part of the circulation process and become a vailable for detection using PIAM or similar processing. By 1996, Lanza and colleagues had be gun investigations of submicron “nano” par ticles. Their agent w as comprised of a biotinylated, lipid-coated , perfluorocarbon emulsion that had low inherent echogenicity unless bound to a surface.3,4 Thrombi exposed to antif ibrin-targeted contrast e xhibited increased echo genicity delineating their surf ace with the blood pool. The degree of enhancement with liquid perfluorocarbon agents w as relati vely modest in comparison with enhancement possible with a gaseous agent. In 1998, Unger and colleagues in vestigated a 2 to 3 µm MB consisting of a lipid shell enclosing a gaseous perfluorobutane core. The targeting strategy for a thrombus was comprised of a peptide specif ic for glycoprotein IIb/IIIa on acti vated platelets, covalently bound to dipalmitoyl glycerol succinate which in tur n was linked to the bubb le using a pol yethylene gl ycol spacer . The ne w results propelled the field forward. Jonathan Lindner and Alexander Klibanov made seminal contributions that, o ver the past fe w years, have been par ticularly influential. Lindner’s work focused first on using the intrinsic proper ties of the shell to tar get inflammation. In 2000, his g roup repor ted on acti vated leukocyte adhesion and MB inter nalization dri ven b y β2-integrin- and complement-mediated binding to activated leukocytes. Attachment of specific ligands to the surfaces of MBs w as fur ther ref ined b y Leong-P oi and Ellegala to enable targeting of relevant angiogenesis markers such as the inte grin αvβ3.39–41 Villanueva and colleagues42 investigated other peptide motifs associated with angiogenesis19 and contributed to the development of molecular echocardio graphy. Excellent re views of this w ork are provided by Kaufman 43 and Lindner.44 An example of molecular imaging of tumor angio genesis with tar geted MBs taken from the w ork of Elle gala and colleagues 39 is

Ultrasound

231

Figure 6. Low- (5 MHz) and high-(30 MHz) frequency ultrasound imaging systems, scanheads, and images. Increasing the frequency by a factor of 6 scales resolution (see Table 1) into the microimaging range but at the cost of penetration. The use of frequencies greater than 20 MHz is well suited to small-animal studies. Images courtesy of GE Healthcare and VisualSonics Inc.

shown in Figure 8. The tumor in this e xample was created by intracerebral implantation of human glioma cells in an athymic rat. 1 × 105 U87MG cells were implanted into the cerebral hemisphere at a depth of 4 to 6 mm. A jugular vein w as cannulated for administration of 1 × 108 αvβ3targeted contrast MBs. Imaging was performed in the midcoronal plane using an HDI 5000, Philips Ultrasound system with a transmission frequenc y of 3.3 MHz and pulse inversion processing as described earlier (see section “Ultrasound Pulse Sequencing”). Examining Figure 2 indicates that this approach w ould provide resolution on the order of 1 mm. The gross features of molecular expression are therefore visib le, but the detailed mor phology of expression of αvβ3 requires the use of signif icantly higher frequencies as will be discussed later. The quantification of the molecular signal in Figure 8 was performed as follows: 10 minutes postinjection image frames representati ve of bound bubb les and the fe w remaining circulating MBs were recorded. The image plane w as then flooded with high-power ultrasound to disr upt and eliminate all the MBs. Subsequent recording of images then contained only circulating bubbles flowing back into the image plane. By subtracting postinjection from preinjection, an indication of the bound “molecular” signal w as determined, normalized, and displayed as a color overlay in Figure 8. Binding and adherence of tar geted MBs are signif icant issues in the hostile en vironment of the micro and macro circulations. The chemical and ph ysical strategies

designed to o vercome these prob lems are discussed b y Ferrara in Chapter 28 “Ultrasound Contrast Agents.” In particular, the de velopment of leuk ocyte mimetic MBs 45 46,47 and the use of ultrasound radiation force appear particularly promising approaches to ensure that MBs find their tar gets with high af finity. Reduction of the immunogenicity of tar geted MBs is also an impor tant issue. Borden and colleagues48 has recently reported on a method of bur ying the ligand in a pol ymeric overbrush and transiently revealing it by ultrasound radiation force to reduce its immunogenicity.

Molecular Imaging with High-Frequency Microultrasound The requirement for increased resolution in preclini cal imaging studies has led to the de velopment of microultrasound imaging technolo gies that operate in the 20 to 60 MHz range for mice. These approaches pro vide 30 to 150 micron resolution as described earlier (see section “Ultrasound Instruments for Molecular Imaging”) and are sho wn in Figures 10 and 11. Some of the molecular imaging strategies de veloped for lo wer frequencies ha ve been recentl y adapted for use in the higher frequenc y range. Typically, a high-frequency microultrasound imaging system (Vevo 770; VisualSonics Inc) is used in these studies. Cine loops of approximately 800 B-mode images of tumors under study

232

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 7. The concept of molecular imaging with targeted microbubbles (MBs). MBs freely circulate in the vascular space (above) and are excellent red blood cell tracers. Altering the intrinsic properties of the shell material or adding ligands via linkers (see Chapter “Chemistry of Molecular Imaging: An Overview”) enables MB contrast agents to contact and bind endothelial cell surface markers of interest. Bound MBs oscillate in the acoustic field returning strong echoes to the imaging transducer for display.

are initially acquired at the center frequenc y of the chosen scanhead at a frame rate of 15 Hz.Tumors are imaged at the widest cross-section with the scanhead held f ixed in an integrated small-animal rail system. MBs bearing a ligand for the chosen tar get, such as an antiVEGFR-2 antibody or an isotype control antibody , are administered via a tail v ein or jugular cannula. MBs are usually administered as a bolus of 5 × 107 particles in ~100 µL followed by a 20 µL saline flush. MB w ash-in within the tumor is v erified b y imaging at lo w po wer (10%) immediatel y follo wing administration. Targeted MBs are allowed to accumulate for approximately 4 minutes after w hich imaging is resumed at a po wer of 50%. A high-po wer destr uctive pulse sequence with a center frequency of 10 MHz is then applied after approximately 200 frames are imaged; the destr uctive sequence ser ves to destroy MBs within the beam ele vation. Immediately after the destr uction sequence, imaging is recommenced at a lo wer power of 50%, and residual circulating MBs are obser ved to replenish the beam. The time sequence for this imaging protocol is illustrated schematicall y in Figure 12. F or the f irst 6 seconds, reference images are acquired in the absence of an y contrast agent. The MB contrast agent is then injected and allo wed to wash in to

the tissue of interest. The wash-in period allows dynamic observation of the hemodynamics of b lood flo w. This process is illustrated in the tw o images of F igure 9 in which a human melanoma x enograft in a nude mouse flank is being observed. The still frame taken at the peak of wash-in (Figure 10A) shows a cloud of individual MBs inside the tumor . By inte grating the intensity due to the MBs over time (25 seconds in this case) using a maximum intensity persistence algorithm, the tracks of the MBs trace out the microcirculatory network in the image plane as sho wn in F igure 10B. After 4 minutes, the v ast majority of circulating MBs ha ve been eliminated , but MBs targeted to an endothelial tar get (VEGFR-2 in this case) have stochastically bound to their targets creating a stationary component of the MB signal. By subtraction of the earlier reference image frames, the image from adherent and residual circulating MBs is computed.This image data are superimposed as shades of g reen on the grayscale image and, at a time of approximately 300 seconds, provides a graphic indication of molecular binding. High-frequency “molecular” images of VEGFR-2 in melanoma xenografts are shown for four different tumors in F igure 11. Note that in contrast to lo wer frequenc y molecular imaging, the molecular labeling appears to be

Ultrasound

233

Figure 8. Molecular imaging of tumor angiogenesis with targeted microbubbles (MBs). From upper left counter clockwise: immunohistology of tumor microvessels demonstrating αvβ3-integrin staining of tumor neovessels. Confocal microscopy demonstrating retention of αvβ3-integrin targeted MB (red) in tumor neovascular network stained with green fluorescent lectin. Parametric CEU imaging of microvascular blood velocity (MBV) showing high velocity (oranges and reds) of the outer angiogenic region and very slow flow velocity at the center of the tumor (greens). Coronal plane CEU images of a rat brain with a glioblastoma tumor with αvβ3-targeted MBs demonstrating enhancement in the region of the primary tumor (T), as well as in a very small periventricular metastasis (M). Reproduced with permission from Ellegala DB et al.39

confined to discrete v ascular channels that are readil y visible in shades of g reen. The high resolution of microultrasound molecular imaging is unique among the molecular imaging techniques described in this book. The f inal quantification of the VEGFR-2 signal is accomplished by subtracting the small component of circulating MBs. This is accomplished by momentarily turning up the acoustic intensity in the ultrasound beam to disr upt all of the MBs in the imaging plane. Circulating MBs quickl y reenter the image plane via the microcirculation and are subtracted o ver a re gion of interest. Quantif ication of molecular tar gets (VEGFR-2) with microultrasound w as f irst demonstrated by Rychak and colleagues49 in melanoma models using biotinylated MBs (Targeson Inc.) coupled to a biotin ylated VEGFR-2 antibody (A vas 12a1; eBioscience, San Die go, CA) via strepta vidin coupling. Lyshchik and colleagues investigated the use of a ne w tar geted MB in w hich the strepta vidin w as directly link ed to the MB . This agent (MicroMark er; VisualSonics Inc.) w as used to successfull y image VEGFR-2 e xpression in mouse models of breast cancer.50 In par ticular, the y sho wed dif ferences in VEGFR-2 e xpression patter ns betw een a highl y in vasive metastatic (4T1) and a nonmetastatic (67NR) breast cancer tumor . Willmann and colleagues 51 recently

provided additional v alidating e vidence for tar geted imaging of VEGFR-2 using the MicroMark er agent in mouse angiosarcoma and rat malignant glioma.

CONCLUSIONS Targeted ultrasound MB contrast agents offer a promising means of molecular imaging. Because single MB detection is possib le, the sensiti vity of ultrasound molecular imaging approaches or exceeds that of nuclear or optical method in cer tain situations. The combination of high sensitivity with high resolution is a unique feature of ultrasound molecular imaging. Taking advantage of this strength will require focused ef forts to impro ve binding strategies and in par ticular to de velop nona vidin-biotin approaches for human imaging. New methods are needed to optimize the ratio of specif ic to nonspecif ic binding and to reduce the immunogenicity of the agents. Also, the development of quantitati ve methods that are independent of tissue-specif ic attenuation and beam characteristics are impor tant to de velop to help adv ance the f ield. Targeted molecular imaging with ultrasound is at a comparatively earl y stage of de velopment, but the future potential of this technology for rapid, inexpensive studies of molecular e xpression is w ell w orth the ef fort to develop.

234

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 9. Imaging sequence for molecular imaging. Reference images are recorded for 10 seconds followed by 40 seconds of microbubble (MB) wash-in. After a 4 minute wait during which freely circulating MBs are largely eliminated, imaging resumes for 10 seconds to record the cumulative contrast signal. Subtraction of these images from the reference data set reveals the bound MBs. These are overlaid, typically in green, on the reference image sequence. The ultrasound power is then sharply increased to disrupt the MBs (arrows) in the plane of view. Subsequent imaging reveals only the circulating component of MBs. The “molecular signal” is quantified by subtracting the latter from the former over a region of interest.

A A

B B

Figure 10. A, Still image taken at 10 seconds and maximum intensity persistence (MIP) image at 35 seconds following the injection of a microbubble (MB) contrast agent. Individual MBs are clearly visible at 10 seconds (arrow). Integrating over time in the MIP image reveals the paths taken by MBs as they traverse the microcirculation in the plane of observation. B, This tumor exhibits a very uniform filling pattern with intercapillary spacing of approximately 100 µm.

Figure 11. Molecular images of VEGFR-2 expression in four human melanoma (MeWo) xenografts. Grayscale shows the structure of these 5 to 8 mm diameter tumors while the green overlay shows regions of molecular expression of VEGFR-2. Reproduced with permission from Rychak J et al.49

Ultrasound

ACKNOWLEDGMENTS Support for this w ork was provided by the National Cancer Institute of Canada with funds from the Terry F ox Foundation, the Canadian Institutes for Health Research, the Ontario Research and De velopment Challenge Fund , and VisualSonics Inc. F .S. F oster disclosed a f inancial interest in VisualSonics.

REFERENCES 1. Rumack CM, Wilson SR, Charboneau JW . Diagnostic ultrasound. New York: Mosby; 1998. 2. Lindner JR. Molecular imaging with contrast ultrasound and targeted microbubbles. J Nucl Cardiol 2004;11:215–21. 3. Lanza GM, Wallace KD, Fischer SE, et al. High-frequency ultrasonic detection of thrombi with a tar geted contrast system. Ultrasound Med Biol 1997;23:863–70. 4. Lanza GM, Wallace KD, Scott MJ, et al. A novel site-targeted ultrasonic contrast agent with broad biomedical application. Circulation 1996;94:3334–40. 5. Lanza GM, Wickline SA. Targeted ultrasonic contrast agents for molecular imaging and therapy. Prog Cardiovasc Dis 2001;44:13–31. 6. Unger E, Metzger P III, Kr upinski E, et al. The use of a thrombusspecific ultrasound contrast agent to detect thrombus in arteriovenous fistulae. Invest Radiol 2000;35:86–9. 7. Unger EC, McCreery TP, Sweitzer RH, et al. In vitro studies of a new thrombus-specific ultrasound contrast agent. Am J Cardiol 1998;81:58G–61G. 8. Lindner JR, Song J , Xu F, et al. Nonin vasive ultrasound imaging of inflammation using microb ubbles tar geted to acti vated leuk ocytes. Circulation 2000;102:2745–50. 9. Bachmann C, Klibano v AL, Olson TS, et al. Targeting mucosal addressin cellular adhesion molecule (MAdCAM)-1 to nonin vasively image e xperimental Crohn’ s disease. Gastroenterolo gy 2006;130:8–16. 10. Lindner JR. Assessment of inflammation with contrast ultrasound. Prog Cardiovasc Dis 2001;44:111–20. 11. Linker RA, Reinhardt M, Bendszus M, et al. In vi vo molecular imaging of adhesion molecules in e xperimental autoimmune encephalomyelitis (EAE). JAutoimmun 2005;25:199–205. 12. Reinhardt M, Hauf f P, Briel A, et al. Sensiti ve par ticle acoustic quantification (SPAQ): a new ultrasound-based approach for the quantification of ultrasound contrast media in high concentrations. Invest Radiol 2005;40:2–7. 13. Hamilton A, Huang SL, Warnick D, et al. Left v entricular thrombus enhancement after intra venous injection of echo genic immunoliposomes: studies in a ne w e xperimental model. Circulation 2002;105:2772–8. 14. Schumann P A, Christiansen JP , Quigle y RM, et al. Targetedmicrobubble binding selectively to GPIIb IIIa receptors of platelet thrombi. Invest Radiol 2002;37:587–93. 15. Korpanty G, Carbon JG, Gra yburn PA, et al. Monitoring response to anticancer therap y by targeting microbubbles to tumor v asculature. Clin Cancer Res 2007;13:323–30. 16. Korpanty G, Grayburn PA, Shohet RV, Brekken RA. Targeting vascular endothelium with a vidin microbubbles. Ultrasound Med Biol 2005;31:1279–83. 17. Villanueva FS. Molecular images of neo vascularization: ar t for ar t’s sake or form with a function? Circulation 2005;111:3188–91. 18. Weller GE, Villanueva FS, Tom EM, Wagner WR. Targeted ultrasound contrast agents: in vitro assessment of endothelial dysfunction and multi-targeting to ICAM-1 and sial yl Lewisx. Biotechnol Bioeng 2005;92:780–8.

235

19. Weller GE, Wong MK, Modzelewski RA, et al. Ultrasonic imaging of tumor angiogenesis using contrast microb ubbles targeted via the tumor-binding peptide ar ginine-arginine-leucine. Cancer Res 2005;65:533–9. 20. Klibanov AL, Rasche PT, Hughes MS, et al. Detection of indi vidual microbubbles of ultrasound contrast agents: imaging of free-floating and targeted bubbles. Invest Radiol 2004;39:187–95. 21. Foster FS, Zhang MY, Zhou YQ, et al. A new ultrasound instr ument for in vi vo microimaging of mice. Ultrasound Med Biol 2002;28:1165–72. 22. Ferrara K, P ollard R, Borden M. Ultrasound microbubb le contrast agents: fundamentals and application to gene and dr ug delivery. Annu Rev Biomed Eng 2007;9:415–47. 23. Foster FS, Pavlin CJ, Harasiewicz KA, et al. Advances in ultrasound biomicroscopy. Ultrasound Med Biol 2000;26:1–27. 24. Hope Simpson D, Chin C, Bur ns PN. Pulse inversion doppler: a new method for detecting nonlinear echoes from microbubble contrast agents. IEEE Trans Ultrason F erroelectr F req Control 1999;46:372–82. 25. Burns PN , P owers JE, Hope Simpson D , et al. Har monic imaging: principles and preliminary results. Angiology 1996;47:63–73. 26. Burns PN, Hope-Simpson D. Pulse inversion doppler ultrasonic diagnostic imaging. United States Patent 6,095,980. 2000. 27. Haider B, Chiao R Y. High order nonlinear ultrasonic imaging. Proceedings of the IEEE Ultrasonics Symposium 1999;2:1527–31. 28. Eckersley RJ, Chin CT, Bur ns PN. Optimising phase and amplitude modulation schemes for imaging microbubb le contrast agents at low acoustic power. Ultrasound Med Biol 2005;31:213–9. 29. Chomas J, Dayton P, May D, Ferrara K. Nondestructive subharmonic imaging. IEEE Trans Ultrason F erroelectr F req Control 2002;49:883–92. 30. Forsberg F, Liu JB , Shi WT, et al. In vi vo pressure estimation using subharmonic contrast microbubb le signals: proof of concept. IEEE Trans Ultrason Ferroelectr Freq Control 2005;52:581–3. 31. Forsberg F, Shi WT, Goldberg BB. Subharmonic imaging of contrast agents. Ultrasonics 2000;38:93–8. 32. Shankar PM, Krishna PD , Newhouse VL. Subharmonic backscattering from ultrasound contrast agents. J Acoust Soc Am 1999; 106:2104–10. 33. Shi WT, F orsberg F , Hall AL, et al. Subhar monic imaging with microbubble contrast agents: initial results. Ultrason Imaging 1999;21:79–94. 34. Sherar MD, Noss MB, Foster FS. Ultrasound backscatter microscopy images the inter nal str ucture of li ving tumour spheroids. Nature 1987;330:493–5. 35. el Gammal S, Auer T, Popp C, et al. Psoriasis vulgaris in 50 MHz Bscan ultrasound—characteristic features of stratum cor neum, epidermis and dermis. Acta Derm Venereol Suppl 1994;186:173–6. 36. Pavlin CJ, Sherar MD, Foster FS. Subsurface ultrasound microscopic imaging of the intact eye. Ophthalmology 1990;97:244–50. 37. Turnbull DH, Stark oski BG, Harasie wicz KA, et al. A 40-100 MHz B-scan ultrasound backscatter microscope for skin imaging. Ultrasound Med Biol 1995;21:79–88. 38. Turnbull DH, Bloomf ield TS, Baldwin HS, et al. Ultrasound backscatter microscope anal ysis of earl y mouse embr yonic brain de velopment. Proc Natl Acad Sci U S A 1995; 92:2239–43. 39. Ellegala DB, Leong-Poi H, Carpenter JE, et al. Imaging tumor angiogenesis with contrast ultrasound and microb ubbles tar geted to αvβ3. Circulation 2003;108:336–41. 40. Leong-Poi H, Christiansen J, Heppner P, et al. Assessment of endogenous and therapeutic arteriogenesis by contrast ultrasound molecular imaging of inte grin e xpression. Circulation 2005; 111:3248–54. 41. Leong-Poi H, Christiansen J, Klibanov AL, et al. Noninvasive assessment of angiogenesis by ultrasound and microbubbles targeted to αv-integrins. Circulation 2003;107:455–60.

236

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

42. Villanueva FS, Lu E, Bowry S, et al. My ocardial ischemic memory imaging with molecular echocardio graphy. Circulation 2007; 115:345–52. 43. Kaufmann BA, Lindner JR. Molecular imaging with targeted contrast ultrasound. Curr Opin Biotechnol 2007;18:11–6. 44. Lindner JR. Microbubb les in medical imaging: cur rent applications and future directions. Nat Rev Drug Discov 2004;3:527–32. 45. Rychak JJ, Lindner JR, Le y K, Klibano v AL. Deformable gas-f illed microbubbles targeted to P-selectin. J Control Release 2006. 46. Zhao S, Borden M, Bloch SH, et al. Radiation-force assisted targeting f acilitates ultrasonic molecular imaging. Mol Imaging 2004;3:135–48. 47. Zhao S, Kr use DE, Ferrara KW, Dayton PA. Acoustic response from adherent tar geted contrast agents. J Acoust Soc Am 2006; 120:EL63–9.

48. Borden MA, Zhang H, Gillies RJ , et al. A stimulus-responsi ve contrast agent for ultrasound molecular imaging. Biomaterials 2008;29:597–606. 49. Rychak JJ, Graba J, White C, et al. Micro-ultrasound molecular imaging of VEGFR-2 in a mouse model of tumor angio genesis. Mol Imaging 2007;6:289–96. 50. Lyshchik A, Fleischer AC, Huamani J, et al. Molecular imaging of vascular endothelial growth factor receptor 2 expression using targeted contrast-enhanced high-frequenc y ultrasono graphy. J Ultrasound Med 2007;26:1575–86. 51. Willmann JK, P aulmurugan R, Chen K, et al. Ultrasonic imaging of tumor angiogenesis with contrast microbubbles targeted to vascular endothelial g rowth f actor receptor 2 in mice. Radiolo gy 2008;246:508–18.

16 MOLECULAR PHOTOACOUSTIC TOMOGRAPHY LIHONG V. WANG, PHD

The f ield of photoacoustic tomo graphy (PAT) has g rown considerably in the past fe w years. Several commercially available high-resolution pure optical imaging modalities, including confocal microscop y, tw o-photon microscop y, and optical coherence tomography, have been broadly used in biomedicine. These technologies, however, are based on ballistic (ie, unscattered or singly backscattered) and quasiballistic photons. As a result, they cannot provide penetration beyond ~1 to 2 mm into scattering biolo gical tissue. PAT, which combines high ultrasonic resolution and strong optical contrast in a single modality, fills this void. PAT refers to imaging based on the photoacoustic effect. Although the photoacoustic ef fect as a ph ysical phenomenon was first reported on by Alexander Graham Bell in 1880, 1 PAT as an imaging technolo gy was developed only after the adv ent of lasers, ultrasonic transducers, and computers. 2–30 The moti vation for P AT is to combine optical-absor ption contrast with ultrasonic spatial resolution for deep imaging in the optical quasidiffusive or diffusive regime (ie, multiple scattered until a nearly isotropic angular light distrib ution is reached), which is > 1 mm in most biolo gical tissues. A comprehensive review on biomedical photoacoustics and imaging is a vailable else where.31 In this chapter , w e co ver the principles of photoacoustic imaging and the applications in molecular imaging using a tar geted optical contrast agent or a reporter gene.

IMAGING PRINCIPLE In PAT, the tissue is usuall y ir radiated by a shor t-pulsed laser beam to produce ther mal and acoustic impulse responses. Locally absorbed light is con verted into heat, which is fur ther converted to a pressure rise via ther moelastic expansion of the tissue. The initial pressure rise— determined b y the local optical ener gy deposition (also called specif ic or v olumetric optical absor ption in J/m 3)

and other thermal and mechanical properties—propagates in the tissue as an ultrasonic w ave that is refer red to as a photoacoustic w ave. The photoacoustic w ave is detected b y ultrasonic transducers, producing electric signals. The electric signals are then amplif ied, digitized, and transferred to a computer. PAT has tw o major for ms of implementation. One is based on a scanning focused ultrasonic transducer. Acoustic focusing in combination with time-resolv ed detection of photoacoustic w aves produces tomo graphic images directly. Dark-field confocal photoacoustic microscopy32,33 belongs to this cate gory, where the light illumination pattern on the tissue surface is donut-shaped with a dark core. The other is based on an ar ray of unfocused ultrasonic transducers. This method, also referred to as photoacoustic computed tomography, requires the use of a reconstr uction algorithm to produce a tomographic image. The image contrast of P AT is based on optical absorption in the photoacoustic excitation phase. Optical absorption is associated with endogenous molecules such as oxygenated and deo xygenated hemoglobin as w ell as melanin. Concentrations of multiple chromophores whose spectra of absor ption coefficient are different can be quantified by varying the wavelength of the irradiating laser. Such quantif ication of o xygenated and deo xygenated hemoglobin, for example, can provide functional imaging of the total concentration and o xygen saturation of hemoglobin. Optical absorption is also associated with exogenous contrast agents such as or ganic dy es and nanoparticles. When these contrast agents tar get biomarkers, PAT can be used to pro vide molecular imaging. Furthermore, optical absor ption is associated with some gene expression products. Therefore, PAT can image gene expressions. The spatial resolution of P AT is deter mined b y the ultrasonic parameters in the photoacoustic detection phase provided the laser pulse is sufficiently short. In dark-field confocal photoacoustic microscop y, the lateral resolution 237

238

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

is deter mined by the center acoustic w avelength and the numerical aper ture of the ultrasonic detector , w hereas the axial resolution is in versely related to the bandwidth of the ultrasonic transducer.32 In photoacoustic computed tomography, the spatial resolution is related to the bandwidth, the aperture of each element, and the density of elements of the ultrasonic transducer ar ray.22 As a salient feature, PAT images are devoid of speckle artifacts, which are conspicuous in both ultrasono graphy and optical coherence tomography images.

EXPERIMENTAL SETUP The e xperimental setup of in vi vo P AT is sho wn in Figure 1. A tunable Ti:sapphire nanosecond pulsed laser (LT-2211A; Lotis T II, Minsk, Belar us) pumped b y a Q-switched Nd:YAG laser (LS-2137/2; Lotis T II, Minsk, Belarus) is used to pro vide laser pulses with a pulse repetition frequency of 10 Hz. The laser beam is routed by prisms, e xpanded by a conca ve lens (abo ve the light diffuser, not sho wn), homo genized b y a light dif fuser, and then deli vered to the object such as the head of an animal. The incident energy density of the laser beam on the surface of the object is controlled to be ~20 mJ/cm2, which is the ANSI safety limit in the visib le spectral region.34 The energy of each laser pulse is detected b y a

photodiode (PD) (DET110; ThorLabs, Inc.) and recorded by an oscilloscope (TDS5054; Tektronix, Inc.) to compensate for shot-to-shot laser energy fluctuations. After a thin layer of ultrasound coupling gel is applied to the surface of the object, the object is positioned belo w a hole in the bottom of the water tank where the hole was sealed with a piece of polyethylene membrane. Two ultrasonic transducers (V323/2.25 MHz and XMS-310/10 MHz, Panametrics) immersed in the water tank detect the photoacoustic signals. The active areas of the 2.25-MHz and 10-MHz transducers are 6 mm and 2 mm in diameter , respectively. The corresponding nominal bandwidths are 66% and 80%, respecti vely. A computer-controlled step motor rotates the two transducers along a complete circle with 120 equall y spaced steps to recei ve the photoacoustic signals (the use of more steps pro vides g reater signal-to-noise ratio at the e xpense of acquisition time). These signals are low-pass filtered and amplified by two ultrasonic receivers (5072PR; P anametrics). The amplified signals are then digitized and a veraged 20 times at each scan position with a data acquisition card (CS14100, Gage, Inc.) operating at a 50-MHz sampling rate and 14-bit resolution, where averaging improves the signal-to-noise ratio at the cost of time. The digital signals are used to reconstr uct an image of the specif ic optical absorption in the imaging (x–y) plane.

Figure 1. Experimental setup of PAT for in vivo imaging. Ti:Sa, Ti:sapphire. PD, photodiode. The laser beam is routed by prisms, expanded by a concave lens (above the light diffuser, not shown), homogenized by a light diffuser, and then delivered to the object. The energy of each laser pulse is detected by a photodiode and recorded by an oscilloscope to compensate for shot-to-shot laser energy fluctuations. Two ultrasonic transducers immersed in the water tank detect the photoacoustic signals. A computer-controlled step motor rotates the two transducers along a complete circle to receive the photoacoustic signals. These signals are low-pass filtered and amplified by two ultrasonic receivers. The amplified signals are then digitized and averaged at each scan position with a data acquisition card. The digital signals are used to reconstruct an image of the specific optical absorption in the imaging (x–y) plane. Reproduced with permission from Li M et al (Figure 1).39

Molecular Photoacoustic Tomography

INITIAL PHOTOACOUSTIC PRESSURE Upon short laser pulse excitation, local fractional volume expansion dV/V of the heated tissue at position r can be expressed as35: dV/V = –κ p(r) + βT(r).

(1)

Here, κ denotes the isother mal compressibility (~5 × 10−10 Pa−1 for water or soft tissue); β denotes the thermal expansion coefficient (~4 × 10−4 K−1 for muscle); p and T denote the changes in pressure (Pa) and temperature (K), respectively. The isother mal compressibility κ can be expressed as: κ =

Cp . 2 ρv sCV

(2)

Here, ρ denotes the mass density (~1000 kg/m 3 for water and soft tissue); vs denotes the speed of sound (~1480 m/s in water); Cp and CV (~4000 J/(kg K) for muscle) denote the specif ic heat capacities at constant pressure and v olume, respectively. It is impor tant to distinguish betw een Cp and CV for gases but not for tissue. For a sufficiently short laser pulse, the fractional volume expansion is negligible and the local pressure rise p0 immediately after the laser excitation can be derived from Eq. (1)36: βT(r) p0(r) = . κ

(3)

It can be estimated that each mK temperature rise yields an 8-mbar (or 800 Pa) pressure rise. If we assume that all absorbed optical ener gy is con verted into heat, meaning that nonthermal relaxation such as fluorescence is ne gligible, the temperature rise generated b y the shor t laser pulse is: T=

Ae , ρCV

(4)

where Ae denotes the specif ic optical absorption (J/m3). From Eqs. (3) and (4), we have: p0 =

β κρCV Ae.

(5)

RECONSTRUCTION ALGORITHM In reconstr uction-based photoacoustic computed tomography, photoacoustic data recei ved by unfocused ultrasonic transducers, ideall y point transducers, are used to produce an image by image reconstruction. The

239

initial photoacoustic pressure e xcited b y a laser pulse δ(t) equals p0(r) [Eq. (5)]. The acoustic pressure distribution p(r0, t) at position r 0 and time t, initiated b y source p0(r), is measured around the tissue b y unfocused ultrasonic transducers. Spherical, cylindrical, and planar detection conf igurations are considered , w here the detection surf ace S0 encloses the source p0(r) (cylindrical and planar detection surf aces are considered to meet at inf inity). For brevity, we simply provide the f inal back-projection formula37: 1 p0(r) = Ω0

cosθ 0 [2p(r0,t) − 2t∂p(r0,t) / ∂t] |r − r |2 dS0, 0 S0

(6)

where Ω0 = 2π for the planar geometr y, Ω0 = 4π for the spherical or c ylindrical geometr y, and θ0 denotes the angle between the detection surface normal and the vector pointing to the reconstruction point r. When a circle, rather than a full spherical surf ace, is usuall y scanned, the reconstr uction algorithm gi ven b y Eq. (6) is onl y approximately applicable.38

MOLECULAR PAT OF CELL RECEPTOR (INTEGRIN) Simultaneous molecular and functional P AT w as demonstrated in li ve mice using the system sho wn in Figure 1. 39 One million human U87 gliob lastoma tumor cells w ere implanted stereotacticall y into young adult immunocompromised nude mice (20 g, Harlan, Co.) in the head ~3 mm belo w the scalp surface. One week post inoculation, 20 nmol of IRDye800-c(KRGDf)40 was administered through the tail vein while mannitol w as used to per meabilize the blood brain bar rier. Approximately 20 hours later, PAT imaging was conducted. During the data acquisition for PAT, the mouse was under full anesthesia. The 2.25-MHz transducer w as adjusted in height to image the tumor at appro ximately the depth of the tumor inoculation and w as scanned with a radius of 4.5 cm. The 10-MHz transducer w as adjusted to image the brain cor tex at appro ximately the depth of the scalp surf ace and w as scanned with a radius of 3.5 cm. F our laser w avelengths—764, 784, 804, and 824 nm—w ere used to measure the spectral distribution of optical absor ption. F rom the four photoacoustic images, the concentrations of the three dominant optical absorbers—o xyhemoglobin, deo xyhemoglobin, and IRDy e800-c (KRGDf)—were quantified. Then, the o xygen saturation of hemo globin w as

240

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

computed. The scanning time for this e xperiment was about one hour but can be signif icantly improved (see Discussion). Figure 2A sho ws an in vi vo pseudocolor molecular PAT image of IRDy e800-c(KRGDf). Inte grin αvβ3 is known to overexpress in U87 gliob lastoma tumors while IRDye800-c(KRGDf) has been demonstrated in vitro to bind to αvβ3 integrin of U87 tumor cells. The molecular image is overlaid on a str uctural PAT image of the brain cortex acquired at the 804 nm w avelength. The 10-MHz transducer provided the str uctural image of hemo globin with a 60-µm in-plane resolution, whereas the 2.25-MHz transducer provided the molecular image with a 312- µm in-plane resolution. Owing to the compromise betw een spatial resolution and sensiti vity, the higher frequenc y transducer provides better resolution but poorer sensiti vity. Therefore, the higher frequenc y was used to achie ve high spatial resolution for structural and functional imaging, and the lower frequency was used to obtain high sensitivity for molecular imaging. As can be seen in the image, the tar geted contrast agent w as concentrated in a small re gion, w hich w as histologically conf irmed to be the tumor area. Figure 2B shows an in vivo pseudocolor functional PAT image of the o xygen saturation of hemo globin. A focal re gion sho ws lo wer o xygen saturation of hemoglobin than the sur rounding area, w hich indicates hypoxia characteristic of tumor due to h ypermetabolism. The hypoxic region is colocated with but slightl y wider than the uptak e area of IRDy e800-c(KRGDf). The focal re gion of both the functional and molecular contrasts w as cor roborated to be the tumor re gion through postmor tem histolo gic e xamination and fluorescence imaging.

A

MOLECULAR PAT OF REPORTER GENE (LACZ ) Molecular PAT of reporter gene lacZ was demonstrated in live rats.41 The lacZ gene is one of the most widel y used reporter genes. It encodes β-galactosidase, an Escherichia coli enzyme responsible for lactose metabolism. The sensiti ve chromogenic assa y 5-bromo-4-chloro-3-indol yl-β-D-galactoside (X-gal), an opticall y transparent lactose-lik e substrate that serves as a colorimetric assay for β-galactosidase, yields a stable dark b lue product. The b lue product increases optical absorption and ser ves as a contrast agent for P AT. The PAT system used in this experiment is similar to the one described in Figure 1 except for the laser . A dye laser (ND6000, Continuum) operating with DCM dy e, pumped by a Q-switched Nd:YAG laser , emitted light at 650 nm. The illumination was measured to be ~5 mJ/cm 2 near the skin. The ultrasonic transducer (V323-SU, Panametrics) with a center frequenc y of 2.25 MHz was used. Sprague-Dawley rats were inoculated with 5 million 9L/lacZ gliosarcoma tumor cells. Tumors that had grown to be visible were imaged with PAT before and after local subcutaneous injection of 25 µL of X-gal solution (20 mg/mL, F ermentas). The same amount of X-gal w as locally injected in a nontumor location on the contralateral side of the head as a control. Local injection of X-gal was adopted because systemic deli very through the tail vein appeared to be inefficient. We expect this constraint to be o vercome with future de velopment of absor ptionbased reporter gene systems. Figure 3A sho ws the P AT image of the animal acquired prior to X-gal injection. This PAT image does not reveal the tumor. Figure 3B shows the PAT image of the animal acquired following the X-gal assay. This PAT

B

Figure 2. A, In vivo molecular PAT image of tail-vein injected IRDye800-c(KRGDf) in a nude mouse brain with a U87 glioblastoma xenograft. B, In vivo functional PAT image of hemoglobin in the same nude mouse brain. Four laser wavelengths—764, 784, 804, and 824 nm—were used to measure the spectral distribution of optical absorption due to oxyhemoglobin, deoxyhemoglobin, and IRDye800c(KRGDf). The arrow indicates the hypoxic region. Reproduced with permission from Li M et al (Figures 2 and 3).39

Molecular Photoacoustic Tomography

A A

C C

B B

D D

Figure 3. Molecular photoacoustic images (A) before and (B) after the injection of X-gal. C, Photograph of the rat’s head prior to the second scan. The two arrows indicate the positions of injection of X-gal. D, Photograph of the underside of the rat’s scalp in the rectangular region in (C), excised after sacrificing the animal. The scale bars represent 5 mm. Reproduced with permission from Li L et al (Figure 2).41

image sho ws the geneticall y tagged tumor clearl y, indicating concentrated b lue product at the tumor . The contralateral side does not sho w signif icant change in optical absorption. Figure 3C shows a photograph of the rat head taken after X-gal injection and just before PAT imaging. Although the photograph shows a small bump, it does not reveal the size of the tumor and the distrib ution of lacZ expression as the P AT image does. The region enclosed in the dashed rectangle in F igure 3C was dissected and photo graphed from the underside. The photograph (Figure 3D) shows the dark b lue X-gal stained tumor, w hich is appro ximately a mir ror image of the tumor re gion sho wn in F igure 3B . The small discrepancy between the PAT image and the photograph is most lik ely due to tissue distor tion during e xcision and different light illuminations from opposite sides of the tissue.

DISCUSSION The detection sensiti vity to contrast agent is a k ey parameter in molecular imaging. The molar sensitivity of PAT to indocyanine green in tissue phantoms w as reported to be on the order of nM 42 to µM,28 depending on the depth of the contrast agent. A sensitivity of µM of indocyanine green (fmol in each resolution cell) in b lood w as also

241

estimated.43 In a study b y Li and colleagues, 39 5 µM of IRDye800-c(KRGDf) in a 250- µm inner -diameter tube covered by 2-mm thick chicken tissue was imaged with a signal-to-noise ratio of ~26 dB . In a study b y Li and colleagues,41 5.4 µM of blue product in a 250- µm innerdiameter tube covered by 5-mm thick chicken tissue was imaged with a signal-to-noise ratio of ~20 dB . PAT is competitive in sensitivity compared with other existing molecular imaging modalities. In comparison to the nM–pM sensitivity of positron emission tomo graphy (PET) at mm spatial resolution, the P AT has poorer sensitivity but better resolution. Although PET uses ionizing radiation, P AT uses nonionizing light and ultrasound. Even at its cur rent level, the sensiti vity of PAT is much greater than that of magnetic resonance imaging (MRI). Furthermore, the sensiti vity of P AT can be impro ved potentially b y increasing the local optical fluence, increasing the optical absorption cross section of the contrast agent by using, for e xample, nanoparticles, increasing the photoacoustic con version ef ficiency, increasing signal a veraging, and decreasing the spatial resolution. Although image for mation in PAT is based on ultrasonic detection, the contrast is deri ved from optical proper ties. In comparison, pure ultrasound imaging measures mechanical contrast. Ultrasound molecular imaging is based on microbubbles that are conf ined to intravascular space only. It is worth mentioning that X-ray imaging has not demonstrated molecular imaging yet. The PAT system reported here has long data acquisition times. Because of the lo w laser pulse repetition frequency, slo w w avelength tuning, and time-consuming mechanical scanning, this system takes about one hour if four w avelengths, 20-times signal a veraging, and 120 scanning steps are used. The acquisition time, ho wever, can be shortened significantly by improving these speedrelated f actors. By using an ultrasound ar ray alone, the acquisition speed can be impro ved b y one order of magnitude. F aster laser pulse repetition frequenc y and wavelength tuning can improve the data acquisition speed further. The author’s laboratory has recently achieved 50 frames per second B-scan rate. 44 There are tw o impor tant depth limits for optical imaging. The 1-mm depth limit is cor related with the optical transpor t mean free path, representing the depth of the quasi-ballistic re gime in biolo gical tissue. 35 We refer to this bar rier as the soft limit for high-resolution optical imaging. Confocal microscop y, tw o-photon microscopy, and optical coherence tomo graphy are limited b y this bar rier because the y depend on ballistic or quasi-ballistic photons. Photoacoustic imaging is ab le to break through this barrier because any absorbed photons,

242

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

either unscattered or scattered, can produce photoacoustic signals as long as the photon e xcitation is relax ed thermally. Although dif fuse optical tomo graphy also o vercomes this soft limit, its spatial resolution is poor . Another depth limit e xists at ~50 mm, w hich equals roughly 10 times the 1/ e optical penetration depth or 43 dB one-way light intensity decay. We refer to this limit as the hard limit for optical imaging. Both P AT and diffuse optical tomo graphy have reached this limit. 28,45 Beyond this limit, e ven diffuse photons are fe w. However, if w e image tissue such as the human breast from both sides, a 10 cm thickness can be potentially imaged, which is adequate for many biomedical applications. The v arious optical imaging modalities complement each other in both penetration and contrast. In PAT, an y fractional change in the optical absor ption coefficient translates into an equal amount of fractional change in the photoacoustic signal, w hich means a relative sensiti vity of unity . Confocal microscop y, tw ophoton microscopy, and optical coherence tomo graphy have lo wer sensiti vities to optical absor ption. F or example, for a typical pix el length (1 µm) and a nominal hemo globin absor ption coef ficient of 4, 300, and 3000 cm −1 at 800, 560 (Q-band), and 420 nm (Soret band), respectively, the relative sensitivities of confocal microscopy to optical absorption are 8 × 10−4, 6 × 10−2, 6 × 10−1, respectively. Unless the Soret band is chosen, the sensitivity of confocal microscop y is signif icantly lower than that of PAT. However, operation at the Soret band limits the penetration se verely because the penetration depth is ~3 µm in blood. While PAT is sensitive to optical absorption, it is insensitive to optical scattering or fluorescence. Therefore, the v arious modalities will coexist. An approach to overcome the optical depth limit is to adopt microwaves for photoacoustic e xcitation.5,6 In this case, the technology is referred to as microwave-induced acoustic (ther moacoustic) imaging. The laser in P AT is replaced with a micro wave generator , w hich transmits microwave pulses into the tissue to be imaged. Microwave absor ption produces heat and subsequentl y ultrasonic w aves. The ultrasound detection and image formation are similar to those in PAT. Ultrasound suffers signif icant attenuation and phase distortion in thick bone. As a result, P AT through thick skull is challenging. Fortunately, unlike pulse-echo ultrasound imaging, P AT in volves onl y one-w ay ultrasound attenuation through the skull. As shown in Figures 2 and 3, ultrasound penetration through small animal skulls is feasible, and the associated distor tions are relati vely small. The author’ s g roup also demonstrated one-w ay

ultrasound penetration through Rhesus monk ey skulls. 46 Extension of PAT to brain imaging through thick er skull is an active area of research. A primary task is to cor rect for the phase distortion due to the skull. Ultrasound sustains strong reflection from gas–liquid or gas–solid interfaces because of the strong mismatch of acoustic impedances. Therefore, ultrasound signals cannot penetrate through gas ca vities or the lung ef ficiently. Ultrasound technolo gies ha ve not o vercome this challenge yet. For the same ph ysical reason, ultrasound coupling gel is applied to the tissue surf ace in ultrasound imaging to a void air ca vities betw een the ultrasound transducer and the tissue. At the stage of galle y proof, I would like to point the readers to a couple of impor tant developments in molecular photoacoustic imaging. One was based on single-w alled carbon nanotubes, 47 and the other on a reporter gene.48 In summary, molecular imaging of tar geted contrast agents or gene expression products along with functional imaging of hemo globin has been demonstrated in small animals b y PAT. The optical w ave pro vides the image contrast, w hereas the laser -induced acoustic w ave provides the spatial resolution. Because multiple optical scattering can be tolerated , PAT can image be yond the soft limit (~1 mm) for high-resolution optical imaging. PAT represents a ne w paradigm for molecular and functional imaging. Translation of P AT to the clinic is expected to follo w the a vailability of molecular contrast agents for human use.

ACKNOWLEDGMENT This work was sponsored by National Institutes of Health grants R01 EB000712 and R01 NS46214.

REFERENCES 1. Bell AG. On the production and reproduction of sound by light. Am J Sci 1880;20:305–24. 2. Karabutov AA, P odymova NB , Letokho v VS. Time-resolved laser optoacoustic tomography of inhomogeneous media. Appl Phys B 1996;63:545–63. 3. Hoelen CGA, de Mul FFM, Pongers R, Dekker A. Three-dimensional photoacoustic imaging of b lood v essels in tissue. Opt Lett 1998;23:648–50. 4. Esenaliev RO, Karabutov AA, Oraevsky AA. Sensitivity of laser optoacoustic imaging in detection of small deepl y embedded tumors. IEEE J Sel Top Quant Electron 1999;5:981–8. 5. Wang LHV, Zhao X, Sun H, K u G. Micro wave-induced acoustic imaging of biological tissues. Rev Sci Instrum 1999;70:3744–8. 6. Kruger RA, Reineck e DR, Kr uger GA. Thermoacoustic computed tomography—technical considerations. Med Ph ys 1999; 26:1832–7. 7. Karabutov AA, Sa vateeva EV , P odymova NB , Orae vsky AA. Backward mode detection of laser -induced wide-band ultrasonic

Molecular Photoacoustic Tomography

8. 9.

10.

11.

12. 13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25. 26.

27.

transients with optoacoustic transducer . J Appl Ph ys 2000; 87:2003–14. Ku G, Wang LHV. Scanning microwave-induced thermoacoustic tomography: signal, resolution, and contrast. Med Ph ys 2001; 28:4–10. Kostli KP, Frauchiger D, Niederhauser JJ, et al. Optoacoustic imaging using a three-dimensional reconstr uction algorithm. IEEE J Sel Top Quant Electron 2001;7:918–23. Kostli KP, Frenz M, Weber HP, et al. Optoacoustic tomography: timegated measurement of pressure distributions and image reconstruction. Appl Opt 2001;40:3800–9. Xu MH, K u G, Wang LHV . Micro wave-induced ther moacoustic tomography using multi-sector scanning. Med Ph ys 2001;28: 1958–63. Xu Y, Wang LHV. Signal processing in scanning ther moacoustic tomography in biological tissues. Med Phys 2001;28:1519–24. Paltauf G, Viator JA, Prahl SA, Jacques SL. Iterati ve reconstruction algorithm for optoacoustic imaging. J Acoust Soc Am 2002;112:1536–44. Xu MH, Wang LHV. Time-domain reconstruction for thermoacoustic tomography in a spherical geometr y. IEEE Trans Med Imaging 2002;21:814–22. Xu Y, Feng DZ, Wang LHV. Exact frequenc y-domain reconstruction for thermoacoustic tomography—I: planar geometry. IEEE Trans Med Imaging 2002;21:823–8. Xu Y, Xu MH, Wang LHV. Exact frequenc y-domain reconstr uction for ther moacoustic tomography—II: cylindrical geometr y. IEEE Trans Med Imaging 2002;21:829–33. Andreev VG, Karabutov AA, Oraevsky AA. Detection of ultra wideband ultrasound pulses in optoacoustic tomo graphy. IEEE Trans Ultrason Ferroelectr Freq Control 2003;50:1383–90. Finch D, P atch SK, Rak esh. Deter mining a function from its mean values o ver a f amily of spheres. SIAM J Math Anal 2003;35:1213–40. Karabutov AA, Savateeva EV, Oraevsky AA. Optoacoustic tomo graphy: ne w modality of laser diagnostic systems. Laser Ph ys 2003;13:711–23. Kostli KP, Beard PC. Two-dimensional photoacoustic imaging by use of Fourier-transform image reconstruction and a detector with an anisotropic response. Appl Opt 2003;42:1899–8. Wang XD , P ang YJ, K u G, et al. Three-dimensional laser -induced photoacoustic tomography of mouse brain with the skin and skull intact. Opt Lett 2003;28:1739–41. Xu MH, Wang LHV. Analytic explanation of spatial resolution related to bandwidth and detector aper ture size in ther moacoustic or photoacoustic reconstruction. Phys Rev E 2003;67:056605.1–056605.15. Copland J A, Eghtedari M, P opov VL, et al. Bioconjugated gold nanoparticles as a molecular based contrast agent: Implications for imaging of deep tumors using optoacoustic tomo graphy. Mol Imaging Biol 2004;6:341–9. Haltmeier M, Scherzer O , Burgholzer P, Paltauf G. Thermoacoustic computed tomography with lar ge planar recei vers. Inverse Probl 2004;20:1663–73. Ku G, Wang XD, Stoica G, Wang LHV. Multiple-bandwidth photoacoustic tomography. Phys Med Biol 2004;49:1329–38. Xu Y, Wang LHV. Time re versal and its application to tomo graphy with dif fracting sources. Ph ys Re v Lett 2004;92:033902. 1–033902.4. Cox BT, Beard PC. Fast calculation of pulsed photoacoustic f ields in fluids using k-space methods. J Acoust Soc Am 2005;117: 3616–27.

243

28. Ku G, Wang LHV. Deeply penetrating photoacoustic tomo graphy in biological tissues enhanced with an optical contrast agent. Opt Lett 2005;30:507–9. 29. Ku G, Wang XD, Xie XY, et al. Imaging of tumor angiogenesis in rat brains in vi vo by photoacoustic tomo graphy. Appl Opt 2005;44: 770–5. 30. Zhang J, Anastasio MA, P an XC, Wang LHV. Weighted expectation maximization reconstr uction algorithms for ther moacoustic tomography. IEEE Trans Med Imaging 2005;24:817–20. 31. Xu MH, Wang LHV. Photoacoustic imaging in biomedicine. Re v Sci Instrum 2006;77:041101.1–041101.22. 32. Maslov K, Stoica G, Wang LHV. In vivo dark-field reflection-mode photoacoustic microscopy. Opt Lett 2005;30:625–7. 33. Zhang HF, Maslov K, Stoica G, Wang LHV. Functional photoacoustic microscopy for high-resolution and nonin vasive in vivo imaging. Nat Biotechnol 2006;24:848–51. 34. Laser Institute of America. American national standard for the safe use of lasers: ANSI standard Z136.1-2000. Ne w York: American National Standards Institute, Inc.; 2000. 35. Wang LHV, Wu H-I. Biomedical optics: principles and imaging. Hoboken (NJ): Wiley; 2007. 36. Gusev VE, Karabutov AA. Laser optoacoustics. New York: American Institute of Physics; 1993. 37. Xu MH, Wang LHV. Universal back-projection algorithm for photoacoustic computed tomo graphy. Ph ys Re v E 2005;71:016706. 1–016706.7. 38. Xu Y, Wang LHV, Ambartsoumian G, Kuchment P. Reconstructions in limited-view ther moacoustic tomo graphy. Med Ph ys 2004;31: 724–33. 39. Li M, Oh J , Xie X, et al. Simultaneous molecular and h ypoxia imaging of brain tumors in vi vo using spectroscopic photoacoustic tomography. Proc IEEE 2008;96:481–9. 40. Wang W, Ke S, Wu Q, et al. Near-infrared optical imaging of integrin αvβ3 in human tumor xenografts. Mol Imaging 2004;3:343–51. 41. Li L, Zemp RJ, Lungu G, et al. Photoacoustic imaging of lacZ gene expression in vivo. J Biomed Opt 2007;12:020504.1–020504.3. 42. Kruger RA, Kiser W, Reineck e DR, et al. Thermoacoustic optical molecular imaging of small animals. Mol Imaging 2003;2:113–23. 43. Wang X, K u G, Wegiel MA, et al. Non-in vasive photoacoustic angiography of animal brains in vivo with NIR light and an optical contrast agent. Opt Lett 2004;29:730–2. 44. Zemp RJ, Song L, Bitton R, et al. Realtime photoacoustic microscopy in vivo with a 30-MHz ultrasound ar ray transducer. Opt Express 2008. [In press] 45. Ku G, Fornage BD, Jin X, et al. Thermoacoustic and photoacoustic tomography of thick biolo gical tissues to ward breast imaging. Technol Cancer Res Treat 2005;4:559–65. 46. Xu Y, Wang LHV. Rhesus monkey brain imaging through intact skull with ther moacoustic tomo graphy. IEEE Trans Ultrason F erroelectr Freq Control 2006;53:542–8. 47. De La Zerda A, Za valeta C, K eren S, Vaithilingam S, Bodapati S, Liu Z, Levi J, Smith B R, Ma T J, Oralkan O, Cheng Z, Chen X Y, Dai H J, Khuri-Yakub B T and Gambhir S S. Carbon nanotubes as photoacoustic molecular imaging agents in li ving mice Nature Nanotechnol 2008;3:557–62. 48. Razansky D, Distel M, Vinegoni C, Ma R, P errimon N, Köster R W and Ntziachristos V. Multispectral opto-acoustic tomo graphy of deep-seated fluorescent proteins in vi vo Nature Photonics 2009;3:412–7.

17 OPTICAL PROJECTION TOMOGRAPHY JAMES SHARPE, PHD

Although it had forer unners in nonbiomedical imaging, 1 the development of optical projection tomo graphy (OPT) into a v ersatile form of biomicroscop y and the in vention of the name were both f irst reported in 2002. 2 In the subsequent 5 years, the f ield has seen steady g rowth both in technical improvements and development of applications. It has had most impact in the f ield of developmental biology, both because of the optical suitability of embr yos for the technique and because the comple xity of 3D tissue morphology during organogenesis is particularly difficult to understand without good 3D models. The range of species successfully imaged ex vivo now includes human, mouse, chick, zebraf ish, Drosophila, and Arabidopsis. However, another impor tant area of impact is the e x vivo imaging of adult mouse organs, particularly in the case of pancreatic imaging for research in diabetes. 3 For this particular example, it is now clear that a very accurate quantification of Islets of Langerhans can be made from a single OPT scan of an adult mouse pancreas that is fluorescently labeled using antibodies against insulin (described more in sections “What Types of Specimens Can Be Imaged?” and “What Types of Research Is OPT Useful For?”). The most recent e xciting de velopment is 4D OPT imaging of li ve specimens—both g rowing plant roots and de veloping mouse or gans in in vitro culture (section “Dynamic Molecular Imaging Using 4D OPT”). OPT was developed in response to the reco gnition of an “imaging gap” in the spectrum of existing techniques.4 At one end of the spectr um are optical sectioning approaches, such as confocal laser scanning microscop y5 and multiphoton microscop y.6 These are e xcellent for imaging small biolo gical samples, such as cells and tissues, and ha ve become a ubiquitous approach for man y areas of biomedical research. The main limits of confocal are (1) size—high-quality images are usuall y onl y obtained up to a depth of fe w hundred microns and (2) dependence on a fluorescent signal—the v ery sharply 244

focused plane that allo ws confocal to generate highquality 3D voxel stacks is achie ved through the principle of fluorescent e xcitation. The f act that onl y one point within the specimen emits light at an y given moment is essential to the technique, and this is onl y possible when the specimen has fluorescent dy es suitab ly distributed throughout it. Although a “transmission mode” is sometimes available on a confocal microscope, this is not ab le to create genuine 3D v oxel stacks—the lack of localized fluorescent e xcitation in this mode reduces the system back to a standard bright f ield microscope. Single plane illumination microscopy (SPIM) is a ne wer technique for optical sectioning7 and is able to image much larger specimens than confocal, but it is still limited to fluorescent signals. Not all 3D optical imaging techniques are limited to fluorescent signals—optical coherence tomo graphy (OCT) uses a v ery dif ferent principle to generate 3D images of nonfluorescent samples—interferometr y and a coherent light source. 8 In contrast to OCT and confocal, which are each restricted to one mode (either fluorescent or nonfluorescent imaging), OPT has the significant advantage that it can produce 3D images from both transmission and fluorescent modes. At the other end of the size spectr um are techniques such as magnetic resonance imaging (MRI) and computed tomography (CT), which have traditionally been aimed at imaging the w hole or ganisms, but w hich ha ve been increasingly adapted to wards smaller specimens such as embryos.9–11 Their major limitation is lack of tar getable contrast agents—in par ticular their inability to monitor gene e xpression le vels within the conte xt of the w hole organism. CT images depend on spatial differences in the absorption of X-rays; hence, compounds with heavy metals (such as osmium tetro xide) are sometimes used to increase the contrast of anatomical str uctures.11 However, a targetable v ersion of hea vy metal labeling suitab le for recording the distribution gene expression patterns has not

Optical Projection Tomography

been developed. In principle, MRI has more options available, for example, it has been shown that dynamic changes in e xpression of a transgenic repor ter constr uct can be observed over space and time within a Xenopus embryo.12 The repor ter gene encoded for the β-galactosidase enzyme and the specimen w ere injected with a substrate with a caged high-contrast agent. Ho wever, this technology has not become widespread because of both the cost and technical difficulties involved. Instead, both MRI and CT are increasingl y used in combination with molecular imaging technologies such as positron emission tomo graphy (PET) and single photon emission computed tomography (SPECT)—the for mer pro vides the higher resolution anatomical imaging, whereas the latter provides lower resolution data on molecular distributions. Interestingly, the question of w hether specimens are imaged alive or ex vivo makes little difference for both the two ends of the scale—microscopic and macroscopic specimens—but has a much bigger impact on the inter mediate mesoscopic specimens typical for OPT . As described above, small microscopic specimens are usually imaged by optical techniques (such as confocal), and because of their small size, light passes through them easily whether or not a clearing agent is used. The larger macroscopic specimens are typicall y imaged b y the nonoptical techniques (MRI and CT), and since these do not depend on the transport of photons, it also mak es little dif ference whether the specimen is ali ve or not. By contrast, for a biolo gical tissue a few millimeters thick, the question of w hether clearing solvents can be used (ie, whether the imaging can be done ex vivo or instead requires anal ysis of living tissue) has a significant impact on their potential for optical imaging. The intrinsic photon scattering from a li ving or gan will cause significantly lower spatial resolution than if the same specimen was f ixed and cleared. Ne vertheless, in cer tain experiments, the temporal dimension is impor tant enough that high spatial resolution is less important than capturing the dynamics of the system. This balance e xists for all optical imaging techniques, and time-lapse e xamples of OPT are described in section “Dynamic Molecular Imaging Using 4D OPT.” The rest of this chapter will gi ve an o verview of OPT—both the principles of the technique (section “Principles of Optical Projection Tomography”), applications (sections “Cur rent Applications of OPT” and “Dynamic Molecular Imaging Using 4D OPT”), and future directions (section “Future Prospects”). There are two features of OPT that are occasionall y overlooked, so these will be repeated at v arious points through the te xt, and briefl y mentioned here: (1) In addition to fluorescent imaging, OPT is capab le of creating genuine 3D v oxel data from

245

staining techniques that produce colored (nonfluorescent) precipitates. This is especially useful for gene e xpression analysis at the RNA level, as the most common technique for visualization of in situ hybridization results is using BCIP and NBT to produce a pur ple precipitate. (2) As predicted a fe w years ago, optimized algorithms are no w able to reconstr uct a 512 3 voxel 3D dataset within a fe w minutes on a standard PC, thus, making it a very efficient 3D imaging technolo gy (compared with the original reports, in w hich such a reconstr uction took se veral hours). This speedup is mostl y due to algorithmic improvements and only slightly to faster computers.

PRINCIPLES OF OPTICAL PROJECTION TOMOGRAPHY In this chapter , I will gi ve an o verview of the standard OPT setup, the consequences of using light for computed tomography (rather than, for example, X-rays or electron beams), and the basics of the reconstr uction and visualization approaches.

Data Capture The basic setup of an OPT scanner is sho wn in Figure 1. The apparatus consists of a rotary stage for supporting and carefully positioning the specimen, lenses to magnify and focus the image, and a CCD chip to record the ra w projection images. Not shown in the f igure are the two types of illumination: for transmission OPT , a w hite light diffuser is placed on the opposite side of the specimen from the imaging lenses (ie, on the right side of the f igure) and for fluorescence imaging, an ar rangement including mercury source, f ilter w heels, and a focusing lens is also included—fluorescence illumination is usuall y pro vided on the same side as the imaging lenses. The rotary stage is usually driven by a stepper motor , and the specimen is often supported within a cylinder of agarose (although this is purely for mechanical support, and in certain cases, the specimen is glued directly to the imaging mount). If e x vi vo anal ysis is possib le, the specimen (and agarose support) is suspended in an index matching liquid to (1) reduce the scattering of light and (2) reduce heterogeneities of refracti ve inde x throughout the specimen. This means that light passes through the specimen in approximately straight lines, and a standard backprojection algorithm can generate relati vely high-resolution images. The liquid most often used is B ABB or Mur ray’s Clear (a mixture of benzyl alcohol and benzyl benzoate). 2 Inside the OPT scanning de vice, the specimen is

246

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

C

D

E

F

G

Figure 1. OPT Scanner. The basic features of an OPT scanner are schematized (A) and shown in plan view (B). The specimen is usually supported in a cylinder of agarose (imaging support) although this is not essential and certain specimens can be glued directly to the mount of the rotary stage. The specimen is rotated about a vertical axis and suspended in an imaging chamber (not shown) that maintains it in suitable imaging fluid (for example, BABB for most ex vivo imaging). All information from one section through the specimen (red ellipse) is captured by one row of pixels on the CCD (red line). The blue lines in (A, B, and C) represent the light path for one projection. In plan view (B), the usual focusing scheme can be seen: the depth of focus is adjusted to cover the front half of the specimen (from the front of the specimen itself, up to the axis of rotation— red dot). Although projections through the specimen are double-inverted cones (rather than straight cylinders), they may project through the specimen in parallel paths (C) depending on the optics used. Columns (D, E, and F) illustrate different information during the scanning process: D, the projection images captured by the system, E, the cumulative orientations of projection data captured, F, the gradually calculated backprojection that results in a virtual section through the specimen. The three rows from top to bottom display this information after projections have been captured for 0° (the start of the process), 90°, and 180°. G, A photo illustrating the implementation of a standard commercial OPT scanner.

Optical Projection Tomography

maintained within the liquid, rotated to a series of angular positions (usually less than 1° apart), and an image is captured at each orientation. The apparatus is carefull y aligned to ensure that the axis of rotation is per pendicular to the optical axis so that projection data per taining to each plane is collected b y a linear ro w of pix els on the CCD of the camera (Figure 1, red line). The f act that lenses are used to focus an image means that OPT suf fers one technical dra wback compared with its X-ray equivalent: as with all optical imaging for ming systems, there is onl y a limited depth of focus, and this usually cannot encompass the entire specimen. In other words, although the shape of the sampled projections in CT can be appro ximated to long, thin cylinders w hose cross-sections are roughl y constant from one end to the other , in OPT , the cross-sections vary—they are small near the focal plane and become larger fur ther a way from it. Reducing the numerical aperture of the system to increase the depth of focus and force these acceptance cones to be more parallel ha ve limited success—the raw images soon loose their shar pness because of the reduced resolving po wer of the system. The most con venient scheme therefore uses the following compromise—the focal plane is positioned halfway between the axis of rotation and the edge of the specimen closest to the objecti ve lens (Figure 1B). This means that every image contains both focused data from the “front-half ” of the specimen (the half closest to the lens) and out-of-focus data from the “back-half ” of the specimen. Most OPT results to date ha ve been achieved using this raw data in a standard f iltered backprojection algorithm. However, an improved algorithm has recently been de veloped that considers this issue (see section “Computed Tomography Reconstruction”). In an idealized parallel beam v ersion of projection tomography, raw data is onl y required from orientations covering 180°. This is because an image from one direction should be an e xact mirror image of a vie w from the opposite direction, and therefore contain exactly the same information. In X-ray CT, this is not the case because the beam geometr y is usually f an shaped or cone shaped (ie, the “perspective” from one angle is not the same as that from the opposite vie w). In OPT, this is also not the case, although for dif ferent reasons: (1) The focusing scheme described above means that two opposite images will actually be focused on dif ferent regions of the specimen. (2) Even if the focal plane is positioned on the axis of rotation (such that the same region is in focus from any angle), the residual scattering and dif fraction of light passing through the tissue means that “deeper” tissue will always be less resolved than tissue closer to the lens. This

247

is particularly true for larger more opaque tissues. 13 OPT is therefore routinel y perfor med b y capturing images from a full 360° rotation. The current OPT scanner design can take any even number of images for the full rotation, but the standard approach is to take 400 images (one image every 0.9°). A good quality CCD camera is required, and the resulting images (12-bit or higher) must not be subjected to lossy compression before being reconstructed, so the volume of data generated can also become an issue (for transfer , storage, etc) “2 × 2 binning” was a useful feature of CCD cameras for the f irst few years of OPT because although the resolution w as reduced (ra w projection images of approximately 5122), the imaging sensitivity was quadrupled and the amount of data for processing and storage was more manageable. However, the advent of a commercial scanner (in 2006) with more efficient algorithms has made it routine to capture full-resolution projection data with 10242 pixel dimensions.

Tomography with Lenses Although OPT displa ys intuiti ve similarities with X-ray CT, it is in f act closer to SPECT , and I will therefore describe emission OPT (eOPT) f irst in w hich the light to be detected is emitted from within the specimen (rather than transmission mode in w hich the light source is behind the sample). Any tomo graphic technique requires information about the paths taken by the detected rays to correctly reconstruct a 3D image. Photon detectors by themselves only measure the intensity of a signal, not the direction it w as traveling before detection, and each tomographic technology therefore requires a strate gy for determining this direction. In con ventional transmission tomography (such as CT or electron tomo graphy), this is simple because the position of the source is kno wn, and the trajectories therefore lie on straight paths betw een it and the detectors (F igure 2A). In emission tomo graphy, however, the source is distributed throughout the specimen, so for each photon detected , there is no single unambiguous path along w hich it must ha ve traveled. In PET, the problem is solved by detecting a par ticular type of γ-ray, which is always produced as a pair of high-energy photons traveling in opposite directions. By placing detectors on both sides of the specimen and re gistering when two photons are detected within a few nanoseconds of each other, the system def ines a straight line along which the original emission event must have occurred.14 SPECT, by contrast, does not produce pairs of photons—just single photons at a time. Its detectors are therefore f itted with collimators—slabs of lead containing nar row slits. 14

248

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

C

D

Figure 2. Virtual histology and nonfluorescent gene expression analysis. A transgenic E12.5 mouse embryo carrying a “knocked-in” LacZ reporter showing the expression pattern of the Pkd1 gene. The embryo was whole-mount stained with the X-Gal substrate, which produces a blue precipitate wherever the reporter gene is active. This light-absorbing stain was captured and reconstructed through transmission OPT (or brightfield OPT) and is false-colored here in green. The anatomy of the specimen was captured using fluorescence OPT, by choosing a wavelength that picks up autofluorescence from the fixed tissue. Virtual sagital sections can be viewed through the entire specimen (A), and the quality of signal localization deep within the embryo (B) can be appreciated in closeups (C, D) despite the fact that this signal is not fluorescent. (See Yoder and colleagues36 for more information about this specimen).

Only photons traveling at the same angle as the slit will pass through to the detector , so the photon trajector y is implicitly known. The main drawback of this approach is that high propor tions of the photons ha ve to be screened out b y the lead and cannot therefore contribute to the image. In effect, OPT uses a similar approach to SPECT . It uses the acceptance cone for each pix el (def ined by the lenses of the system) as a collimator, screening out photons that are tra veling at the wrong angle. Fluorescent molecules are e xcited throughout the specimen, and emitted photons are detected b y the pix els of the CCD only if the y emer ge from the specimen at the cor rect positions and angles. Usuall y, the e xcitation light is directed onto the specimen from the same side as the imaging lenses although in very transparent cases, this is not necessary. At f irst glance, transmission OPT (tOPT) appears superficially to be more lik e CT than SPECT . Light is shone into the “back” of the specimen, directl y towards

the objective lens, and the image formed records the attenuation of this beam. Ho wever, in reality, even specimens maintained in an index matching liquid, such as Mur ray’s Clear (or B ABB), still cause dif fraction, refraction, and scattering of photons as they pass through. For this reason, OPT imaging cannot be achieved using the same principle as CT (ie, calculating the photon trajectories through knowledge of the relati ve positions of source and detector). Without lenses, a single beam of light loses its narrow specific path after passing through the specimen and becomes a b lur on the detection side. Instead , as mentioned above for eOPT, the lenses act in a similar way to the collimators in SPECT, only sampling rays that emerge from the specimen at specif ic positions and angles. In other words, rather than measuring a quantitati ve shadow (which is indeed what is captured in CT), instead the data recorded on the CCD is a focused image of the specimen. The effective projections through the sample (the centers of the cones of acceptance) do not con verge on a single point behind the sample (the y may in f act be parallel to

Optical Projection Tomography

each other), so as a result, the optimal lamp for OPT is not a point source, but a wide-f ield illumination. Also important is the f act that a widened laser beam is not a suitab le illumination because of the resulting speckles from interference of the coherent light. Instead , a dif fuse source appears to be the optimal illumination (although more collinear beams provide higher definition results). For this reason, transmission OPT is sometimes refer red to as brightfield OPT because of its similarity with con ventional 2D brightf ield microscopy. The choice of wavelength used for OPT is impor tant in obtaining a good image. As specimens become more opaque or more heavily stained (more absorbing), shorter visible w avelengths penetrate less ef ficiently. Ho wever, shifting up towards the infrared end of the spectr um has proven to signif icantly improve the image quality.15 This is particularly useful for specimens stained with colored precipitates (which are imaged using transmission mode) and for lar ger adult tissues that tend to be more opaque because of the more differentiated state of the cells.

Computed Tomography Reconstruction Although OPT data collection exhibits significant differences compared with X-ra y CT, the algorithm for reconstructing this data into a 3D v oxel stack—f iltered backprojection16—is very similar to the CT case. Minor modifications to the approach include cor recting for the gradual fading of a fluorescent signal during the scanning procedure and cor recting for the small random intensity fluctuations caused by the mercury light source. 17 The onl y signif icant modif ication e xplored so f ar relates to the limited depth of f ield mentioned abo ve. In conventional CT, there is no focal plane, so the onl y difference between two images captured 180° apart is due to the f an beam or cone beam ar rangement of the projections.13 In OPT, the fact that there is a focal plane, and that this is not centered on the axis of rotation (see section “Data Capture”) means that a second dif ference e xists between the 180° rotated images—in each case, a dif ferent region of the specimen will be in focus. In principle, therefore, considering the opposing vie ws as identical mirror image v ersions of each other means that outof-focus information can slightly degrade the f inal image quality. Until recently, this potential problem has not been taken into account in the reconstr uction process, and in fact, this simplification still results in high-quality images (all images refer red to in this chapter ha ve been reconstructed in this w ay). However, recentl y Walls and colleagues18 explored an impro ved approach that uses a frequency–distance relationship (FDR) to perfor m the

249

frequency f iltering of ra w projection data before it is backprojected into the reconstr uction.18 This ef fectively excludes out-of-focus data and improves the resolution of the point spread function for in-focus data. It is also w orth reiterating that the use of optimal algorithms means that the reconstr uction of a 512 3 dataset cur rently tak es about 5 minutes on a single PC (for a single channel), w hich is appro ximately equal to the time it tak es to perfor m the scan. As such, w e now routinely capture and process higher resolution OPT scans—a 1024 3 voxel scan is typicall y 4 times slo wer to scan and takes about 45 minutes to reconstruct.

Visualization The 3D v oxel datasets from OPT are similar to those produced by other techniques such as MRI, and as such, similar visualization techniques can be used. Virtual sections through the data can be vie wed at an y orientation, and the specimen can also be rendered as a 3D object—either as isosurf aces, v olume renderings, or a combination of both. The major difference between OPT data and MRI or CT data is that multiple channels (different wavelengths) can be captured , so e very voxel has multiple 16-bit data points associated with it. Interestingly, within the world of scientif ic 3D visualization, there appears to ha ve been a “rendering gap” analo gous to the “imaging gap” mentioned in the introduction. Although e xpensive dedicated hardw are can cer tainly cope with interacti ve speed rendering of the lar ge datasets produced by OPT, affordable desktop computers are only just star ting to catch up. Estab lished commercial software has tended to focus on tw o impor tant but distinct preexisting imaging markets: on the one hand is the software for large volume datasets such as MRI and CT. These algorithms allow a standard PC to rapidly volume render lar ge datasets, but onl y for single channels. On the other hand is softw are for multichannel 3D confocal datasets. Although packages lik e these cope v ery well with man y channels in the same space, the lar ge datasets from OPT scanning (e g, 3 channels × 10243 voxels) can still be prob lematic. One option for highquality rendering is to use “off-line” precalculation of images (we currently use the package VTK). In this case, an interactive phase allo ws the user to def ine rendering parameters (such as colors, transparenc y v alues, and angles), and the softw are then renders the images or movies o vernight. Ho wever, the most recent de velopments mean that f irst waves of genuine interactive algorithms are appearing that are tailored to tak e advantage of the latest generation of af fordable computer g raphics

250

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

cards. Softw are such as Bioptonics Viewer is ab le to interactively display three different color channels, each at 10243 resolution and with channel-specific interactive clipping ().

CURRENT APPLICATIONS OF OPT OPT has e xciting potential as a molecular imaging approach to track dynamic gene e xpression in 3D o ver time, for or ganogenesis using in vitro or gan culture techniques. The f irst developments in this direction will be described in section “Dynamic Molecular Imaging Using 4D OPT.” However, to date, OPT has mostl y been used on ex vivo specimens, so in order to give a complete overview of the technolo gy, this section will summarize what has been achieved so far.

What Types of Specimens Can Be Imaged? As explained in section “Principles of Optical Projection Tomography,” b y def inition, OPT is limited to samples that allo w photons to tra verse them in substantiall y straight paths. (This is inherent in the use of the word projection within the name of the technique.) Once a sample gets beyond this nondiffuse or low-diffuse regime, deterministic paths give way to probabilistic functions, and we enter the domain of dif fuse optical tomo graphy (DOT)19 and fluorescence mediated tomo graphy (FMT) 20—see other chapters in this te xtbook. Sample size has often therefore been cited as the main criterion for successful OPT imaging, and indeed with a constant probability of photon scattering per unit pathlength, the o verall degree of scattering increases e xponentially with tissue depth, and an upper size limit therefore e xists for each specimen type. However, two other issues are in f act also important: the intrinsic optical proper ties of the tissue and w hether the specimen can be treated with a suitab le clearing agent. The first factor is yet another reason why developmental biolo gy has benef ited strongl y from OPT— embryonic tissues generally contain very undifferentiated cells that tend to be optically more transparent than adult differentiated tissue (presumab ly because of the abundance of more specialized str uctural proteins, extracellular matrix, collagens, etc). An embryo of a given diameter may therefore allo w higher resolution imaging than an adult tissue sample of the same size. The second f actor generally has a dramatic impact—all samples will yield higher resolution results if the y can be cleared before scanning. A certain type and size of specimen may not be

imageable at all in aqueous conditions but may, however, provide v ery useful data once treated with Mur ray’s Clear.2 This has implications about the types of e xperiments that the technique can be useful for (see section “What Types of Research Is OPT Useful For?”). In ter ms of size constraints, it should also be noted that the lo wer size limit for OPT is the same as for an y other form of light microscopy, that is, the spatial resolution is limited by the usual considerations for diffractionlimited optics, 21 rather than the sample size per se . Accordingly, successful OPT imaging has been demonstrated on a very small specimen—a single cell nucleus.22 Although it was already possible to capture 3D images of single cells before the adv ent of OPT (using standard confocal microscopy), one potential advantage of OPT in this case is that the cell did not ha ve to be fluorescentl y labeled—the imaging was performed by tOPT and so was using intrinsic optical contrast within the nucleus instead of fluorescence. This opens the possibility of a new imaging modality for microscopy, potentially capturing infor mation about the cell not accessib le to nor mal fluorescence imaging.22 An impor tant general point is that autofluorescence can be a potential prob lem for imaging more dif ferentiated adult tissues. Not onl y are such tissues less transparent but the y also tend to displa y much stronger nonspecific background fluorescence. For this reason, the development of protocols for bleaching autofluorescence3 or for minimizing its appearance in the f irst place 23 has become impor tant for projects that aim to fluorescentl y image specimens such as a w hole brain or pancreas. Mammalian Organs and Embryos

Most ex vivo OPT imaging so f ar has been perfor med on either the embr yos or the adult soft or gans of mice and other laborator y rodents. Mouse embr yos are routinely imaged at E9.5 up to E12.5, 24 and subregions scanned for older specimens, for example, the developing limb 25 or brain, 26 at E14.5. The Edinburgh Mouse Atlas Project (EMAP) has also perfor med purel y anatomical scanning (of autofluorescence) for a fe w whole embr yos up to an age of ~E15. 24 Younger and smaller embr yos can also be imaged b y OPT , but because of the e xistence of pre vious techniques (such as confocal) that w ere already ab le to capture these smaller specimens, it has not been considered as an important area for OPT itself. In addition to imaging embr yos, the last couple of years have seen imaging success with a number of whole

Optical Projection Tomography

organs tak en from the adult mouse. Results ha ve no w been pub lished for brain, 3,27 pancreas,3 kidney,28 and lungs.29 This opens up e xciting new applications such as preclinical disease research, and accordingl y, the w hole pancreas imaging w as perfor med quantitatively to compare the mass of β-cell tissue in nor mal versus diabetic specimens, using the NOD mouse model3 (section 3.3.5). A related type of soft tissue, b ut with a v ery dif ferent application, is human biopsy material. Although still at the e xploratory phase, earl y tests look encouraging (section “Future Prospects”). Other Suitable Specimens

In addition to imaging of mammalian samples, a wide variety of other standard laborator y models ha ve been successfully anal yzed b y OPT , including chick, 30,31 zebrafish,32 Drosophila,33 and plants. 34 In the case of Drosophila, successful imaging to pinpoint neurobiolo gical lesions in the brain of the adult fly was possible without any special bleaching of the chitinous exoskeleton.

251

mapping projects.35 The other very common colored stain used in genetic e xperiments is the repor ter LacZ that turns the substrate X-Gal into a b lue precipitate. 36 Less commonly explored for OPT scanning is D AB (because unlike WMISH, fluorescent labeling is usuall y possib le for immunohistochemistry). Regarding emission OPT , only fluorescent labeling has been e xplored so f ar (rather than, for e xample, bioluminescence) and again the range of successful fluorophores is lar ge and essentiall y the same as an y other fluorescent microscop y technique. Especiall y good are the Alexa dyes and CY3,2 however, FITC and Texas Red have also been used. The endo genous fluorescence of GFP has indeed been recorded in a couple of studies, 3,33 although for the reasons mentioned abo ve, it is not routinely used for e x vi vo imaging. Ho wever, perfor ming normal immunohistochemistry using an antibody against the GFP protein easily solves this problem.25

What Types of Research Is OPT Useful For? Roles of OPT

What Signals/Contrast Agents Can Be Imaged? The simple answer to this question is that any absorbing or fluorescent dye can be imaged using either the transmission mode or emission mode. In general, this pro vides a wealth of staining possibilities—the only caveats are that ex vivo imaging will invariably wish to take advantage of clearing agents such as BABB, and this means that the dye must remain (1) stably incor porated in the tissue and (2) stably detectab le, upon deh ydration. Notab le e xceptions to this are: (1) lipophilic dyes, such as DiI, that integrate into the lipid membranes of cells and therefore wash out of the tissue due to disruption of the membranes when the specimen is deh ydrated and (2) fluorescent proteins, such as GFP, whose fluorescence tends to decrease over a few hours in alcohol. The latter case is not so problematic as the distribution of GFP protein can easil y be recovered by whole-mount antibody labeling. 25 As mentioned pre viously, one major adv antage of OPT is that it can create 3D v oxel data from nonfluorescent signals. The biggest practical adv antage of this is seen within the f ield of gene e xpression analysis at the RNA le vel because the most common protocols for whole-mount in situ hybridization (WMISH) result in the production of a pur ple precipitate (using the substrates BCIP and NBT) rather than a fluorescent dye. This is still the most reliab le approach for lar ge gene e xpression

Before listing the specif ic research areas that OPT is involved in, it is useful to summarize the types of data or the imaging “roles” that the technique can pro vide. Virtual Histology. Although one of its clear adv antages is the ability to produce complete 3D models, for a number of research projects, it is simpl y the ability to rapidly create 2D “vir tual histology” that mak es it attractive. An embryo can be imaged and reconstr ucted in minutes, providing a complete set of vir tual sections all the way through the specimen (F igure 2), whereas cutting the equivalent series of real paraf fin sections, mounting them on glass slides and digitally photographing each one takes days. A potential disadvantage compared to real sections is resolution—real sections can be imaged at submicron resolution, whereas OPT of a mouse embr yo will provide resolution of approximately 5 to 10 microns. 2 However, as illustrated in F igure 2, the achie vable resolution is generally considered ideal for histolo gy level or anatom y level studies (as opposed to cellular level studies). In addition to being fast, another advantage of OPT in this context is that virtual sections can subsequentl y be vie wed at an y angle through the specimen, whereas for real sections, a particular cutting angle must be chosen, w hich cannot later be changed. This advantage can be very important to histologists and anatomists because cor rectly identifying tissues depends on the angle of the section through the specimen (eg, whether the section is an accurate sagital or transverse plane, rather than a slightl y ob lique angle). Of course,

252

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

these 2D vir tual sections are useful for e xtracting both histological and anatomical infor mation, as w ell as gene expression information. 3D Morphology. Capturing the 3D shapes of complete undistorted specimens represents one of the k ey roles of OPT imaging. As will be described in the subsequent examples, seeing the tr ue 3D shapes of anatomical str uctures— especially the ability to interacti ve explore these str uctures on a computer—is an in valuable tool for v arious types of research, including de velopmental biology, mutant phenotyping, genetic screens, disease research, and neurobiology. This last e xample has not y et been full y de veloped but holds particular promise as an exciting application for OPT because of the intrinsic 3D complexity of nerve tracts in the mammalian brain. OPT imaging has already managed to obtain full v olumetric datasets for 1-month-old mouse brains, and protocols that allo w penetration of antibodies into the core of the organ have also been developed.3 Manual inspection of 3D mor phology (using interactive computer g raphics) has been a po werful tool for discovering many new mutant phenotypes2,37–39 and even new aspects of wildtype mor phology.40 However, this data is also the perfect substrate for mor phometrics—quantifying aspects of mor phology, such as natural v ariation of an organ shape or measuring the de gree of abnor mality in mutant or diseased organisms. One clear example of this is the quantification of the distributions of insulin-expressing

A

D

tissue in a mouse model of diabetes (described in more detail in section 3.3.5). Gene Expression. The mapping of gene e xpression patterns in 3D throughout an or gan or embr yo can be performed either at the RNA level (by WMISH) or at the protein le vel (using specif ic antibodies). The for mer has the advantage that an RN A probe can be easil y generated for any gene in the genome, w hereas raising specif ic antibodies is e xpensive, time-consuming, and not possib le for many proteins of interest. However, immunohistochemistry has the adv antage that it can easil y be perfor med fluorescently, which allows double or triple labeling e xperiments, whereas RN A in situs are mostl y limited to producing a light-absorbing precipitate at the site of e xpression, which limits accurate 3D imaging to one gene per specimen. Imaging gene e xpression patter ns has tw o distinct roles. Clearl y, for genes of unkno wn acti vity, a 3D expression anal ysis is an ideal method to deter mine where the gene is active, and therefore what its biological functions may be. However, it is equall y common to use a gene expression pattern (especially at the protein level) as a “molecular mark er” to highlight a par ticular subset of the 3D anatom y or mor phology of a specimen. F or example, in F igure 3, an antibody against the protein neurofilament (NF160) has been used (g reen signal), but not for the pur pose of e xploring an unkno wn gene expression patter n. On the contrar y, it is precisel y

B

E

C

F

Figure 3. 3D embryology and anatomical markers. Different views of the same E10.5 mouse embryo. A, One of the 400 raw projection images, showing 3 different fluorescent signals: antibody labeling for neurofilament in green, HNF3β in blue, and autofluorescence in red. After the filtered backprojection algorithm has reconstructed a full 3D voxel dataset, virtual sections can be explored (B and C) displaying the same colors as (A). The same voxel dataset can also be explored interactively as a 3D isosurface model (D, E, and F) in which the isosurface representing the outline of the embryo has been rendered in grey, and a second high-threshold isosurface has been generated from the autofluorescent channel to highlight the atria of the developing heart (pink). (See Sharpe and colleagues2 for more details of this embryo).

Optical Projection Tomography

because w e understand e xactly w here this gene is expressed that the antibody can be used for a purel y anatomical reason—to highlight one step in the comple x 3D sequence of neural de velopment. Other e xamples include the diabetic pancreas imaging described in section 3.3.5 and the use of an anti-PECAM antibody to show the de veloping v asculature during mouse embryogenesis.40,41 Pinpointing Labeled Cells. OPT usuall y does not achieve single cell spatial resolution (apart from the rare cases where individual cells are especiall y large such as the zebraf ish notochord 32 and cer tain plant cells). 34 However, if one or a few labeled cells are surrounded by unlabeled cells, the y can be detected and their position within the context of the sample can be determined. Two published e xamples of clonal anal ysis illustrate this application very well. In Wilkie and colleagues, 26 a clonal anal ysis was performed within the de veloping mouse brain. A transgenic constr uct w as used that contained a ubiquitously expressed but nonfunctional v ersion of the popular reporter gene LacZ. The loss-of-function mutation in the gene reverts to normal activity at a known low frequency during embr yo de velopment. This produces rare clonal labeling events in which all subsequent LacZexpressing cells are descended from the single cell in which the spontaneous re version e vent occur red. Although neither the timing nor the location of the clonal labeling can be controlled , by screening through enough embr yo specimens can be found w hich display clones in the or gan of interest. In this example, OPT scans of the resulting embr yos displa yed clones in the developing neocor tex. Analysis of the 3D data sho wed clear evidence for widespread tangential mo vements of the neural pro genitors. This highlighted tw o of OPT’ s strengths—the ability to capture the whole organ and the fact that the labeling w as not fluorescent. Staining the specimen with X-Gal produces a b lue precipitate w herever the repor ter gene is e xpressed. Comparison of results from 2D serial sections pro ved that it w as v ery difficult to arrive at the same conclusions without a genuine 3D visualization. 26 The second e xample concer ns mouse limb de velopment. In this case, a more controllab le v ersion of clonal LacZ labeling was used in w hich the timing and frequency of clonal events could be adjusted by exploring injections of different concentrations of tamo xifen.42 The clonal analysis proved for the f irst time the e xistence of a dorso-v entral lineage–restricted compar tmental boundar y within the mesenchyme of the limb bud. OPT again allo wed 3D datasets to be created from nonfluorescent samples, w hich

253

helped pro ve that the ne w boundaries did indeed for m a 2D curved surface r unning between the dorsal and v entral surfaces of the limb bud. Atlases of Development

The comprehensiveness of OPT imaging, along with its true three-dimensional geometr y, mak es it ideal for creating digital atlases of development, and so far atlases for three dif ferent v ertebrate species ha ve emplo yed OPT as the k ey source of anatomical data—the mouse, human, and zebraf ish. These datasets are primaril y intended for the academic community as a resource to aid research into de velopmental biology. In addition to anatomical infor mation, the y also aim to become 3D databases of gene e xpression anal ysis and studies that specifically use OPT to gather medium-throughput data on spatial patter ns of gene e xpression are therefore also underway.35 The EMAP has developed into the most detailed and comprehensive digital atlas for mouse de velopment.24 Until the adv ent of OPT , the digital data w ere gathered only from ph ysically cut paraf fin sections of mouse embryos. Digital photographs were taken of every single section, and software was developed to allow the vir tual sections to be digitall y stack ed on top of each other to reconstruct the full 3D anatom y of each embr yo. Apart from being a painstaking task, lasting man y w eeks for each embr yo, a fundamental prob lem e xists for this approach. If each section is aligned as accuratel y as possible to its adjacent sections, the tr ue geometr y of the embryo will become distorted.43 This is because perfectly aligning tw o features of tw o adjacent sections assumes that the y for m par t of a str ucture that passes at right angles through the plane of the sections. F or e xample, tube-like structures (such as the neural tube or alimentary canal) will tend to become straightened along an axis perpendicular to the sections. If in reality the tube passes through the plane at a 45°angle, then alignment of the 2D sections should in fact show a displacement to accommodate that f act. This problem is made e ven worse by the fact that paraffin sections tend to “warp” during processing (ie, to displa y nonunifor m distor tions within the 2D plane of the section). Remo ving these distor tions also involves warping the digital images back to their original shape. However, because of the same prob lems as mentioned above, it is impossib le to kno w a priori w hat the correct shape is. This highlights the main reason w hy the EMAP adopted OPT in its model-making process. Real paraffin sections are still cut because of the higher spatial

254

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

resolution compared with OPT, but before sectioning a 3D OPT, a scan of each embr yo is made, which can be used as an accurate 3D geometric template. The 3D stack containing hundreds of 2D sections can then be warped back into the 3D template to recreate the correct shape of the w hole sample. The atlas cur rently contains 3D models of 17 mouse embryos starting with Theiler Stages 7 (~E5) up to Theiler Stage 23 (~E15).24 A human atlas of embr yo de velopment has also adopted OPT—the Electronic Atlas of the De veloping Human Brain (EADHB), 44 based in Ne wcastle, UK, which contains 3D models from Carnegie Stage 12 up to Stage 23. In this case, the project took advantage both of the accurate geometr y (as described abo ve) and of the nondestructive nature of OPT . The specimens for the human atlas are a very precious and carefully controlled resource, and the project wished to create 3D representations of the embryos while still keeping the specimens for fur ther anal ysis. Indeed , the y pro ved that after performing an OPT scan, these embr yos could subsequently be sectioned as nor mal and used for gene expression analysis, both at the RNA level (using in situ hybridization)45 and protein le vel (using immunohistochemistry). By using alter nate sections for dif ferent genes, the y w ere ab le to maximise the infor mation provided by each embryo. More recently, a 3D atlas of zebraf ish development has been constr ucted based entirel y on OPT images. FishNet,32 created by the Victor Chang Cardiac Research Institute in Sydne y, Australia, contains 18 g rey-scale datasets covering the period from a 24-hour lar va up to the adult f ish (17 mm in length). As with the pre vious atlases, it can be accessed b y an interacti ve w ebsite that allows the user to vie w the vir tual sections in an y of the three orthogonal orientations. A subset of sections in each orientation ha ve been manuall y annotated to aid with identif ication of anatomical features. Web sites for OPT-based atlases of development are: , , . Genetic Screens and Phenotyping Mutants

There is increasing realization that a standardized, accurate, sensiti ve, ef ficient, and economical platfor m for phenotyping mutant mouse embr yos w ould pro vide a significant benef it to the mouse genetics community . For this reason, the technologies that have been recently explored for 3D phenotyping of e x vivo f ixed embryos include micro computed tomo graphy (microCT), micro magnetic resonance imaging (microMRI), and optical

coherence tomo graphy (OCT). MicroCT has been proposed as an economical high-throughput 3D screen for large-scale screens. 11 It requires preparing the samples with hea vy metal compounds (osmium tetro xide) and then subjecting them to X-ray CT imaging, which is optimized for small specimens (12 hours for a high 8 µm3 resolution scan or 2 hours for a lo wer 27 µm3 resolution scan). It has been repor ted specif ically for highlighting brain defects (in a transgenic mouse expressing the Pax3:Fkhr oncogene)11 but could in principle be used for man y de velopmental abnor malities. MicroMRI has been proposed as a similar platfor m and specifically for use in hear t phenotyping. 10 It tends to generate images with lo wer spatial resolution (40 µm3 for higher throughput and 25 µm3 for single-embr yo scans) but has been successfull y used to identify v entricular septal defects, double outlet right ventricle, and hypoplasia of the pulmonary artery and thymus in E15.5 embryos that w ere homozygous null for the Ptdsr gene.10 Finally, OCT has more recentl y been proposed also for hear t phenotyping, but this has onl y been performed on hearts that are physically dissected out of the embryo.46 MicroCT, microMRI, and OCT all display one particular limitation in common for the pur pose of phenotyping—their inability to image molecular labels such as antibody staining, in situ h ybridization, or repor ter gene e xpression. Ev en the last of these (OCT), although it is an optical technique, is unab le to tak e advantage of fluorescent labels. The most useful information for a mor phometric phenotyping system to extract from a specimen is accurate 3D data on the morphology of your organ of interest. The lack of molecular imaging for microCT , microMRI, and OCT means that the shape of the or gan can onl y be e xtracted by either (1) ph ysically dissecting the or gan a way from the rest of the embr yo (as w as done for OCT), 46 which is time-consuming and can introduce ar tefactual distortions to the shape or (2) scanning the whole embryo, but then manuall y def ining w hich pixels/voxels cor respond the hear t (as done for microMRI),10 which is also time-consuming and can introduce manual er rors. By contrast, it is more ef ficient, accurate, and objecti ve to use a molecular label (for example, organ-specific antibody) to def ine automatically and consistentl y w hich v oxels belong to the studied organ. Any technique that can perfor m fluorescence imaging will therefore allo w a “molecular dissection” of the str ucture out of the 3D embr yo. For this reason, OPT is an e xcellent tool for phenotyping mouse embryos.47,48 Although it has not yet

Optical Projection Tomography

been used in a high- and medium-throughput screen, it has been used as a phenotyping tool in a number of studies, and these point to its possib le use for screens in the future. In 2002, OPT was able to discover a new phenotype in the de velopment of the stomach of Bapx1 mutants.2 As suggested above, it was the ability to use a fluorescently labeled molecular mark er for the gut epithelium w hich made the abnor mal anatom y of the mutant stomach ob vious once it w as visib le as a 3D model. OPT also helped characterize the hear t-defect phenotype of the Baf60c knockout39 in which a molecular label was also used as an anatomical marker to avoid the problems mentioned above. In this case, the nonfluorescent imaging capabilities of OPT w ere highlighted again because the molecular marker used was visualized by WMISH producing the typical pur ple precipitate of BCIP/NBT. Indeed, heart defects have proven a popular phenotype for OPT anal ysis,49,50 complementing other studies that ha ve characterized defects in limbs, 25 lungs,29 and pancreas.37 Analysis of Disease Models

Preclinical disease research sho ws promise as an important area for OPT imaging. So f ar, only one disease area has tak en adv antage of the technique, but it ser ves to illustrate the potential for other f ields—in particular any diseases w hose patholo gy af fects specif ic inter nal soft organs. OPT has recentl y been used to study w hole adult mouse pancreases in a common mouse model of diabetes—the NOD strain. 3 This application is closel y related to the phenotyping mentioned in the pre vious section because it in volves a molecular mark er to extract infor mation about a par ticular tissue subset of the organ studied. Antibodies against insulin were used to label all the β-cell tissue within the Islets of Langerhans (F igure 4). A single OPT scan pro vides enough information to produce a detailed quantif ication of the size distribution of Islets within the entire pancreas. Currently, the only alternative approach is the painstaking sectioning and stereolo gy of man y ph ysically cut sections from each specimen. The speed and con venience of OPT has allo wed quantitati ve comparisons between signif icant numbers of nor mal and diabetic mice in a reasonable time. It also provides comprehensive data on the spatial distribution of Islets with respect to the pancreas as a w hole, w hich is gi ving additional clues about the nature of insulitis during the onset of diabetes.3

255

DYNAMIC MOLECULAR IMAGING USING 4D OPT Spatial Resolution versus Temporal Dynamics OPT is able to generate high-resolution images because light passes through the sample in f airly straight lines—in other w ords, it operates in the nondif fuse or moderately scattering regime. This is why the majority of applications e xplored so f ar in volve e x vi vo samples: it allo ws us to use an or ganic clearing agent such as Mur ray’s Clear (or BABB2) to dramatically reduce the scattering of a tissue, w hich might otherwise be much diffused. Nevertheless, tracking the dynamics of li ving tissue often yields impor tant information not obtainable from a series of static f ixed samples. Much w ork o ver recent y ears has been de voted to related tomo graphic approaches (such as FMT) that aim to generate 3D images of entire li ving adult mice. 20 In these cases, photons must penetrate through 1 to 2 cm of living heterogeneous tissue, so the signals emer ging are necessarily weak and highl y scattered and can onl y be reconstr ucted into f airly lo w-resolution images. Nevertheless, this data has been sho wn to pro vide biologically useful infor mation, highlighting that a useful compromise al ways e xists betw een our desire for high spatial resolution and the desire for longitudinal studies that capture the dynamics of a single indi vidual over time. 20 The number of scattering e vents a photon will be subjected to depends on both the opacity of the tissue and the thickness of tissue to be tra versed. If useful information can be obtained using FMT from scattered photons traveling through centimeters of living tissue, it therefore follo ws that imaging through millimeters of living tissue should provide higher resolution and could be useful for cer tain e xperiments. This is despite the fact that this resolution will ne vertheless be signif icantly lower than when the sample is f ixed and cleared (by w hich scattering is reduced to almost zero). F or these reasons, it has been w orth e xtending OPT technology into the f ield of time-lapse imaging—it f ills an optical imaging gap here betw een optical sectioning techniques on the one hand (such as confocal microscopy and SPIM) and dif fuse optical tomo graphy approaches on the other (such as DO T and FMT). The goal therefore is to image small li ving specimens 1 to 2 mm across.

256

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

C

Figure 4. OPT imaging of an entire adult mouse pancreas. By developing whole-mount immunohistochemistry for the whole pancreas using antibodies against the insulin protein, it has been possible to label all the β-cells within the Islets of Langerhans and to quantify this information from a single OPT scan. A, The shape of the pancreas determined from autofluorescence. B, A part of the pancreas tissue has been virtually clipped to show the antibody-labeled Islets inside (red). C, Quantification of this data is then possible, identifying the Islets within a wide range of size categories (from small Islets of 8 to 16,000 µm3, up to large Islets of 2 to 4,000,000 µm3). Grey bars represent the number of Islets in each category, and red bars represent the total β-cell volume composed of Islets from each size category. Reproduced from Nature Methods.3

Dynamic Imaging of Plant Growth Although many parts of a growing plant are opaque due to chlorophyll and other compounds, the roots of man y plants are semitransparent and are 1 to 2 mm across. This structure therefore provided an ideal opportunity to try a proof-of-concept experiment for in vivo time-lapse OPT, and results from the g rowing root of an Arabidopsis plant were the f irst successful pub lished example.34 Model systems that g row at room temperature also display the adv antage that the y can b e imaged using a standard OPT scanner (as opposed to mammalian or gan culture for example, which requires a modified device—

see next section). A germinating seed can be embedded in agarose and introduced into the scanner as an y normal f ixed specimen. The agarose is strong enough to hold the specimen in the cor rect positions for tomographic imaging, w hile also being soft enough for the root to push its w ay through during e xtension. Results achieved so f ar w ere perfor med on a nonfluorescent wildtype plant, highlighting again one of the strengths of OPT—that it can perfor m genuine 3D reconstr uctions using the transmission (brightfield) mode. At each timepoint of interest, the specimen w as rotated through a full 360°, capturing the usual 400 raw projection images. The transparenc y of t he root allo wed v ery

Optical Projection Tomography

high-quality reconstr uctions i n which man y of the individual cells can be clearly seen. 34 This proof of concept was achieved using only transmitted light but allows detailed tracking of tissue mo vements o ver time for an indi vidual specimen (and e ven some cell mo vements). It also suggests that monitoring dynamic gene expression patterns using transgenic plant strains carrying a GFP repor ter gene should be possib le in future studies.

research. Or gan culture is f ar more complicated for these mammalian species, and research has therefore been pursued to mer ge OPT technolo gy with in vitro organ culture techniques (F igure 5). Onl y one e xample of this has been pub lished so far; however, it represents an impor tant con vergence point of OPT with other molecular imaging techniques. 51 The organ chosen was the developing limb bud of the mouse embryo since it is visually accessible (unlike the heart for example, which is concealed by other tissues) and is about 1 mm in size, therefore allowing reasonable transmission of photons. Imaging is perfor med within a chamber heated to 37°. Pre vious standard protocols for culturing limb buds involved dissecting a por tion of the tr unk and positioning this at the air–liquid interf ace.52 For time-lapse OPT, the specimen must be full y submer ged to a void

Tracking Global Tissue Movements of Mouse Limb Bud Development Although a range of model systems exist that grow happily at room temperature (e g, zebraf ish and Xenopus), rodents are the primar y research models for biomedical

A

C

G

257

B

D

H

E

F

I

J

Figure 5. 4D time-lapse optical projection tomography. The layout (A) and design features (B) of the time lapse OPT apparatus are shown. Tissue to be cultured is pinned to the mount while in the dissecting dish and protected during transport into the imaging chamber by the plastic capsule. Three liquids are placed into the imaging chamber: perfluorodecalin to supply oxygen (white), medium (yellow), and mineral oil to prevent evaporation (orange). Experiments using fluorescent microspheres to track ectodermal movements are shown in C–J. A raw image at the beginning of culture (C) shows the distribution of microspheres over the limb bud—2 highlighted with blue arrowheads. After 6 hours, the bud has grown and the positions of microspheres changed accordingly (D). Superimposing the first and last timepoint illustrates the movements made by each microsphere (E), and this can be represented as a series of 3D vectors (F, arrows). Similar data was obtained for younger limb buds (G, H), and results from different experiments can be aligned in 3D (pink and blue arrows) confirming the repeatability of the experiments (I). Finally, radial basis function interpolation was performed to extract a more uniform description of tissue movements, which showed a double-inverted vortex flow in which tissue rotates around two almost fixed points (J, blue asterisks). For more details, see Boot and colleagues.51

258

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

refractive distortions from the surface of the limb, and so a new method was introduced for supplying extra oxygen to the sample—a la yer of perfluorodecalin (or “ar tificial blood”)53 saturated with o xygen, w hich is g radually released during the e xperiment. As with pre vious OPT experiments, the axis of rotation was vertical; however, a number of instr umentation changes w ere necessar y to allow the attachment and alignment of this rather delicate living specimen (Figure 5B). In particular, embedding the sample in agarose is unlik ely to yield good results as it would: (1) increase the handling time between dissection and starting the time-lapse imaging, (2) act as a physical restraint against normal growth, and (3) restrict access to oxygen and nutrients within the medium. As a result, living tissue samples w ere pinned directl y to the apparatus though a region of the tissue not required for the imaging. As this process makes it very difficult to correctly control the angle of the sample with respect to the axis of rotation, a ne w micromanipulator w as designed allo wing careful adjustments to be made after the specimen has already been transferred into the culture chamber. The goal of the experiment was to build up a dynamic picture of normal limb growth by tracking the tissue-level movements of ectoder m; however, the limb bud contains no natural landmarks to act as reference points. Therefore, before placing the sample inside the apparatus, fluorescent microspheres w ere distributed around the ectoder m. The specimen w as imaged from 200 angles (e very 1.8°) at 15-minute inter vals in both fluorescent and transmission modes—the for mer to track the ar tificial landmarks and the latter to measure the o verall shape of the or gan. From this data, both the shape changes and the tissue direction could be e xtracted (see F igure 5C–J), pro viding no vel information on this process and discovering for the f irst time that a twisting motion is i nvolved in normal limb development (Extensive tests w ere perfor med to conf irm that the in vitro g rowth is occur ring at nor mal speeds compared with in utero development).

cells organize itself over a matter of hours into a comple x shape with a specif ic spatial arrangement of gene expression. It is therefore an ideal subject for experiments of 4D OPT both because the patter n changes signif icantly during the time-window available for organ culture protocols and because the biolo gical process is complicated enough that it benef its from the adv antages of a comprehensive 4D description. A transgenic mouse line w as created that e xpresses the GFP protein under control of the Scleraxis promoter,54 and limb bud culture from these mice was performed in the same time-lapse OPT device as described above.51 The resulting OPT anal ysis describes ho w the GFP expression dynamicall y changes its 3D spatial patter n over time (Figure 6). At the beginning of the culture, the GFP signal is seen on the dorsal and v entral sides of the autopod, restricted to the center of the limb bud (specif ically the medial re gions w here tendon specif ication is

A

B

C

D

E

F

G

H

I

Dynamic Gene Expression Imaging in 4D Although tracking surf ace movements is v ery useful, as with other for ms of molecular imaging, the ultimate goal for 4D OPT is to monitor changing gene expression activities throughout the 3D volume of a living tissue while an interesting biolo gical process occurs. So f ar, a single example of this has been achieved—using the same model system as above, the developing mouse limb bud. Organogenesis is a highl y dynamic process in volving spatial pattern for mation. A f airly simple, homolo gous ball of

Figure 6. OPT for dynamic gene expression. Dynamic gene expression in a cultured limb bud from a transgenic mouse embryo carrying the GFP reporter gene under control of the Scleraxis promoter (Scx-GFP). Columns show a time-course of the experiment (A–C, at 0, 13, and 19 hours, respectively). The first row (A–C) shows raw projection images from one angle. The second and third rows (D–I) show 3D renderings of the gene expression pattern at the same 3 timepoints as (A–C). The emergence of new domains of GFP expression can be seen, in particular for digits 2 and 3 (white arrowheads) in the middle column. Reproduced from Boot and colleagues.51

Optical Projection Tomography

starting, that is, away from the anterior, posterior, or distal edges—see Figure 6A, D, G). During 19 hours of culture, the GFP domain increases in size and changes shape (Figure 6B, E, H and C, F , I). To illustrate the changing 3D shape of this dynamic domain, a single threshold isosurface was chosen to create the surface rendering, which captures the spatio-temporal dynamics of the process—in particular showing how the emer gence of digits 4 and 3 clearly precedes digit 2. These experiments represent the first direct 4D obser vation of dynamic spatial patter ning of the mesench ymal tissue in a v ertebrate limb. Assessment of the image quality w as performed by comparing the vir tual sections with real ph ysically cut sections, 51 and this emphasized that the obser ved spatial changes of the expression domain appear to in volve gene acti vation in the digital domains. The ability to monitor growth and gene expression in 3D o ver time is a technical step forward, which should become in valuable for a full understanding of or ganogenesis and ser ve as a quantitati ve basis for computational modeling of or gan development.

259

across and consist of a fe w hundred cells) may be missed during con ventional e xamination but could be more reliably found b y rapidly browsing through the complete set of vir tual OPT sections. (2) 3D geometr y. A less certain, but possible advantage for certain types of biopsy could be the ability to examine the true shape of the tissue. In par ticular, polyps, which are routinel y extracted from patients with a risk of bowel cancer, are known to display certain “phenotypic” changes during the transition from benign to malignant (such as the dif ferences between the villous v ersus squamous appearance). It is possib le that such mor phological changes could be more accuratel y tracked with good 3D data although it must be emphasized that this idea is still speculati ve, not least because biopsies ha ve simpl y not been in vestigated in this w ay before—cur rently patholo gists w ould ha ve no standard reference for judging what they saw.

FUTURE PROSPECTS Because of the range of applications and pub lications described in previous sections, it is clear that OPT should probably not be considered a “ne w” technique an ymore. Indeed, in ter ms of applications, a substantial body of work has been repor ted for a number of dif ferent f ields (especially its use for general embryology and phenotyping), and for man y other possib le applications, at least one proof of concept has been repor ted: databases of unknown gene e xpression patter ns,35 preclinical disease research,3 plant imaging,34 and 4D time-lapse imaging of growing embryonic organs in culture. 51 However, proof of principle for another field of application that w as originally suggested for OPT has not y et been confirmed. Analysis of human biopsies appears to be a possible future direction for OPT scanning (F igure 7). There are two main advantages this technique could ha ve over conventional approaches, both of which reflect OPT’s normal strengths: (1) Comprehensi veness. Standard protocols for assessing e xtracted l ymph nodes from cancer patients in volve embedding each node in paraf fin w ax, bisecting each one, e xamining the inter nal cut surf aces using traditional histolo gy, and sometimes e xtending this analysis to sections cut roughl y every 1 to 2 mm through the tissue. The goal is to determine whether any secondary metastases have developed in each node, and the patient’s prognosis will be hea vily influenced b y the result. The potential advantage of OPT is that a cer tain category of micrometastasis (those w hich are roughl y 50 to 200 µm

Figure 7. Potential future application—biopsy analysis. A virtual section through an OPT scan of a human liver biopsy displaying a metastatic adenocarcinoma (top region of the section). The biopsy was approximately 2 mm across. The top row shows the same section four times, imaged under different channels (transmission OPT, followed by three fluorescent channels: GFP1, Texas Red, and CY3). This highlights how different tissue types intrinsically fluoresce with different wavelength spectra—and also that imaging the absorption of white light can act as a useful extra source of spatial information. The channels can be merged (bottom row) using various combinations of pseudocolour to highlight different aspects of the sample. Image courtesy of Dr. Sarah Wedden of MRC Technology, Dr. Katie Robertson of Clinical Pathology, University of Dundee, and the Tayside Tissue Bank Dundee.

260

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

The other future considerations for OPT are technical. Although its approach for e xtracting 3D information is v ery dif ferent from other techniques (computed tomography rather than optical sectioning), 5–7 in fluorescence mode, it relies on the same principles re garding wavelengths as many other forms of microscopy. It therefore may see a number of improvements taken from these other f ields, including fur ther impro ved reconstr uction algorithms, multispectral linear unmixing, 55 (either for more automated extraction of different tissue types from the autofluorescence of unstained tissue—for e xample, the biopsies mentioned above—or to image more molecular dy es within a single specimen), and time-resolv ed imaging56 (for fluorescence lifetime imaging), in addition to more straightforward improvements due to basic optimization of the cur rent state of the ar t. Thus, although OPT has come a long w ay in the last 5 years and is no longer a “ne w” approach, it is not y et a “mature” f ield that has e xplored most of its options—it can be hoped that the next 5 years will see as many developments, both in principles and practice, as the last 5 years.

REFERENCES 1. Doran SJ, Krstajic´ N. The history and principles of optical computed tomography for scanning 3-D radiation dosimeters. J Ph ys Conf Ser 2006;56:45–57. 2. Sharpe J, Ahlgren U, Perry P, et al. Optical projection tomo graphy as a tool for 3D microscop y and gene e xpression studies. Science 2002;296:541–5. 3. Alanentalo T, Asayesh A, Morrison H, et al. Tomographic molecular imaging and 3D quantif ication within adult mouse or gans. Nat Methods 4:31–3. 4. Sharpe J. Optical projection tomo graphy as a ne w tool for studying embryo anatomy. J Anat 2003;202:175–81. 5. Kulesa PM, F raser SE. Confocal imaging of li ving cells in intact embryos, in protocols in confocal microscope. In: Paddock S, editor. Methods in molecular biology. Vol. 122. Humana Press; 1998. 6. Potter SM, F raser SE, Pine J . The g reatly reduced photodamage of 2-photon microscop y enab les e xtended 3-dimensional time-lapse imaging of living neurons. Scanning 1996;18:147. 7. Huisken J, Swoger J, Del Bene F, et al. Optical sectioning deep inside live embryos by selective plane illumination microscopy. Science 2004;305:1007–9. 8. Huang D, Sw anson E, Lin C, et al. Optical coherence tomo graphy. Science 1991;254:1178–81. 9. Dhenain D, Ruffins S, Jacobs RE. Three-dimensional digital mouse atlas using high-resolution MRI. Dev Biol 2001;232:458–70. 10. Schneider JE, Böse J, Bamforth S, et al. Identification of cardiac malformations in mice lacking Ptdsr using a no vel high-throughput magnetic resonance imaging technique. BMC De v Biol 2004;4:16. 11. Johnson JT, Hansen MS, Wu I, et al. Virtual histology of transgenic mouse embr yos for high-throughput phenotyping. PLoS Genet 2006;2:e61. 12. Louie AY, Huber MM, Ahrens ET, et al. In vivo visualization of gene expression using magnetic resonance imaging. Nat Biotechnol 2000;18:321–5.

13. Sharpe J . Optical projection tomo graphy. Annu Re v Biomed Eng 2004;6:209–28. 14. Massoud TF, Gambhir SS. Molecular imaging in living subjects: seeing fundamental biolo gical processes in a ne w light. Genes De v 2003;17:545–80. 15. Sharpe J. Optical projection tomo graphy: imaging 3D or gan shapes and gene e xpression patter ns in w hole v ertebrate embr yos. In: Yuste, Konnerth, editors. Imaging in Neuroscience and De velopment. CHSL Press; 2005. 16. Kak AC, Slaney M. Principles of computerized tomographic imaging. IEEE Press; 1988. 17. Walls JR, Sled JG, Shar pe J, Henkelman RM. Correction of artefacts in optical projection tomography. Phys Med Biol 50:1–21. 18. Walls JR, Sled JG, Shar pe J, Henk elman RM. Resolution impro vement in emission optical projection tomo graphy. Phys Med Biol 2007;52:2775–90. 19. Gibson AP, Hebden JC, Arridge SR. Recent advances in diffuse optical imaging. Phys Med Biol 2005;50:R1–R43. 20. Ntziachristos V, Ripoll J, Wang LV, Weissleder R. Looking and listening to light: the e volution of w hole-body photonic imaging. Nat Biotechnol 2005;23:313–20. 21. Jonkman JEN , Sw oger J , Kress H, et al. Resolution in optical microscopy. Methods Enzymol 2003;360:416–46. 22. Fauver M, Seibel EJ , Rahn JR, et al. Three-dimensional imaging of single isolated cell nuclei using optical projection tomo graphy. Opt Express 2005;13:4210–23. 23. Sakhalkar HS, Dewhirst M, Oliver T, et al. Functional imaging in bulk tissue specimens using optical emission tomo graphy: fluorescence preser vation during optical clearing. Ph ys Med Biol 2007;52:2035–54. 24. Baldock RA, Bard JBL, Bur ger A, et al. EMAP and EMA GE: a framework for understanding spatiall y or ganized data. Neuroinformatics 2003;1:309–25. 25. DeLaurier A, Schw eitzer R, Lo gan M. Pitx 1 deter mines the morphology of muscle, tendon, and bones of the hindlimb . Dev Biol 2006;299:22–34. 26. Wilkie A, Jordan SA, Shar pe J, et al. Widespread tangential dispersion and e xtensive cell death during earl y neuro genesis in the mouse neocortex. Dev Biol 2004;267:109–18. 27. Hajihosseini MK, Langhe S, Lana-Elola E, et al. Localization and af te of Fgf10-e xpressing cells in the adult mouse brain implicate Fgf10 in control of neuro genesis. Mol Cell Neurosci 2008;37:857–68. 28. Davies JA, Armstrong J. The anatomy of organogenesis: novel solutions to old problems. Prog Histochem Cytochem 2006;40:165–76. 29. Langhe S, Car raro G, Warburton D, et al. Le vels of mesench ymal FGFR2 signaling modulate smooth muscle pro genitor cell commitment in the lung. Dev Biol 2006;299:52–62. 30. Fisher ME, Clelland AK, Bain A, et al. Integrating technologies for comparing 3D gene e xpression domains in the de veloping chick limb . Dev Biol 2008. [In press] 31. Tickle C. The contribution of chicken embryology to the understanding of vertebrate limb development. Mech Dev 2004;121:1019–29. 32. Bryson-Richardson RJ , Ber ger S, Schilling TF, et al. F ishNet: an online database of zebraf ish anatomy. BMC Biol 2007;5:34. 33. McGurk L, Morrison H, Keegan LP, et al. Three-dimensional imaging of drosophila melanogaster. PLoS ONE 2007;2:e834. 34. Lee K, Avondo J, Morrison H, et al. Visualizing plant development and gene e xpression in three dimensions using optical projection tomography. Plant Cell 2006;18:2145–56. 35. Summerhurst K, Stark M, Shar pe J, et al. 3D representation of Wnt and F rizzled gene e xpression patter ns in the mouse embr yo at embryonic day 11.5 (Ts19). Gene Expr P atterns 2008;8:331–48. 36. Yoder BK, Mulro y S, Eustace H, et al. Molecular patho genesis of autosomal dominant pol ycystic kidney disease. Exper t Rev Mol Med 2006;8:1–22.

Optical Projection Tomography

37. Hecksher-Sørensen J, Watson RP, Lettice LA, et al. The splanchnic mesodermal plate directs spleen and pancreatic laterality , and is regulated by Bapx1/Nkx3.2. Development 2004;131:4665–75. 38. Asayesh A, Sharpe J, Watson RP, et al. Spleen v ersus pancreas: strict control of or gan inter relationship re vealed b y analyses of Bapx1–/– mice. Genes Dev 2006;20:2208–13. 39. Lickert H, Takeuchi JK, Von Both I, et al. Baf60c is essential for function of B AF chromatin remodelling comple xes in hear t development. Nature 2004;432:107–12. 40. Walls JR, Coultas L, Rossant J , Henkelman RM. Three-dimensional analysis of earl y embr yonic mouse v ascular development. PLoS ONE 2008.[Submitted] 41. Coultas L, Cha wengsaksophak K, Rossant J . Endothelial cells and VEGF in vascular development. Nature 2005;438:937–45. 42. Arques CG, Doohan R, Shar pe J , Torres M. Cell tracing re veals a dorsoventral lineage restriction plane in the mouse limb b ud mesenchyme. Development 2007;134:3173–722. 43. Hecksher-Sorensen J, Shar pe J. 3D confocal reconstr uction of gene expression in mouse. Mech Dev 2001;100:59–63. 44. Kerwin J, Scott M, Shar pe J, et al. 3-dimensional modelling of earl y human brain de velopment using optical projection tomo graphy. BMC Neuroscience 2004;5:27. 45. Sarma S, K erwin J, Puelles L, et al. 3D modelling, gene e xpression mapping and post-mapping image anal ysis in the de veloping human brain. Brain Res Bull 2005;66:449–53. 46. Jenkins M, Patel P, Deng H, et al. Phenotyping transgenic embr yonic murine hear ts using optical coherence tomo graphy. Appl Opt 2007;46:1776–81. 47. Ruijter JM, Souf an AT, Hagoort J, Moorman AF. Molecular imaging of the embr yonic hear t: fables and f acts on 3D imaging of gene

48. 49.

50.

51.

52.

53.

54.

55.

56.

261

expression patter ns. Bir th Defects Res C Embr yo Today 2004; 72:224–40. Dickinson ME. Multimodal imaging of mouse de velopment: tools for the postgenomic era. Dev Dyn 2006;235:2386–400. Hart AW, Morgan JE, Schneider J , et al. Cardiac malfor mations and midline sk eletal defects in mice lacking f ilamin A. Hum Mol Genet 2006;15:2457–67. Risebro CA, Smar t N , Dupays L, et al. Hand1 re gulates cardiomyocyte proliferation v ersus differentiation in the de veloping heart. Development 2006;133:4595–606. Boot M, Westerberg H, Sanz-Esquer ro J , et al. In vitro w holeorgan imaging: quantitative 4D analysis of growth and dynamic gene e xpression in mouse limb buds. Nat Methods 2008; 5:609–612. Zuniga A, Haramis AP, McMahon AP, Zeller R. Signal rela y by BMP antagonism controls the SHH/FGF4 feedback loop in vertebrate limb buds. Nature 1999;401:598–602. Rinaldi A. A b loodless re volution––A g rowing interest in ar tificial blood substitutes has resulted in ne w products that could soon improve transfusion medicine. EMBO Rep 2005;6:705–8. Pryce B A, Brent AE, Murchison ND , et al. Generation of transgenic tendon repor ters, ScxGFP and ScxAP, utilizing re gulatory elements of the Scleraxis gene. De v Dyn 2007;236: 1677–82. Dickinson ME, Bearman G, Tilie S, et al. Multi-spectral imaging and linear unmixing add a w hole ne w dimension to laser scanning flourescence microscopy. Biotechniques 2001;31:1274–78. Requejo-Isidro J, McGinty J, Munro I, et al. High-speed wide-field time-gated endoscopic fluorescence-lifetime imaging. Opt Lett 2004;29:2249–51.

18 POTENTIAL ROLES FOR RETROSPECTIVE REGISTRATION IN MOLECULAR IMAGING CHARLES R. MEYER, PHD, HYUNJIN PARK, PHD, BING MA, PHD, BOKLYE KIM, PHD, AND PEYTON H. BLAND, PHD

CHALLENGES There are numerous challenges encountered in molecular imaging that ma y be assisted b y retrospecti ve re gistration, that is, re gistration implemented after imaging has been perfor med, both in isomodality and multimodality imaging. The following sections describe in more detail just two of the many possible challenges.

MAPPING HISTOLOGY INTO IN VIVO IMAGING Current and previous literature is replete with hypotheses of cellular level mechanisms whose corroboration is proffered by microscopic photos of one or more stained histolo gy slide(s) obtained from e xperiments with ar rows drawn to the portions of the slide that are consistent with the outcome of the stated h ypothesis. Often the in vestigators include a macroscopic image from one of many modalities, for example, computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomo graphy (PET), or single photon emission computed tomography (SPECT), that also demonstrates the hypothesized local effect occurring at sufficiently many tissue sites so as to be visible in the macroscopic image. In a practice to w hich we all ha ve become much too accustomed the investigator “finds” cellular level evidence that suppor ts his/her hypothesis without knowing that the mark ed region on the histolo gy slide is ph ysically consistent with the location of the visualized ef fect on the macroscopic image. Without any intentional duplicity the investigator is ne vertheless strongl y biased to f ind “the” matching region or set of re gions on the microscopic slide that are responsible for the h ypothesized effect being visualized on the macroscopic images. An alternative nonbiased 262

approach would use registration to map histology back to in vivo, macroscopic imaging in a manner consistent with all correlated spatial features common to the in vivo image and the tissue sample.

EARLY DETECTION OF THERAPEUTIC RESPONSE Another area within w hich molecular imaging ma y be extremely useful is that of detecting early tumor change in response to radiation and/or chemotherap y. In the cur rent paradigm, patients are initiall y treated according to outcomes dra wn from Phase III lar ge population response statistics. Cur rently, assessment of the indi vidual’s response usuall y is measured b y radiological response evaluation criteria in solid tumors (RECIST) or World Health Organization measurements of size change made typically at 10 weeks for brain tumors following therapeutic initiation at which time the size change ef fect is larger than the measurement noise. 1,2 Software tools cur rently under development, such as semiautomatic se gmentation or registration tools, may support more accurate estimates of tumor volume change. Even if volume change could be reliably measured earlier, such changes ma y yield anomalous results, that is, the tumor may have an increased volume due to initial edema e ven though the f inal outcome will be tumor cell death and compar tment shrinkage. Early results suggest that molecular imaging techniques will have sufficient signal-to-noise ratios to reliably predict the outcome to therapy within a 1 to 3 week window follo wing therapeutic initiation. 3,4 Such capabilities will suppor t indi viduating therap y due to the ability to rapidl y s witch therapies until an ef ficacious therapy is found. F or this to occur , techniques for

Potential Roles for Retrospective Registration in Molecular Ima ging

measuring functional tumor change must be de veloped and refined such that their signal, that is, the effect size, is signif icantly higher than their noise. Typically, changes are measured b y observing differences in estimated parameters obtained between exams, also known as interval exams. Diffusion-weighted MRI acquisitions support calculation of dif fusion estimates, that is, the apparent diffusion coefficient (ADC), and for computation of perfusion parameters the modality of choice is typically MRI due to its lo w patient risk, e xcept in patients with renal insuf ficiency.5 Because an isotropic quantitative ADC image volume can be computed from a single scan obtained with interlaced dif fusion g radients of different strengths without incurring patient risk due to the use of contrast agents, dif fusion MRI is an ideal inter val e xam modality with repeatab le results across scanner manufacturers. Typically, techniques for measuring changes in computed diffusion and perfusion parameters ha ve been that of estimating a mean change within the v olume of interest drawn by a radiologist using the T1-weighted, post-gad acquisition sequence to def ine the e xtent of the tumor . But the simplistic mean ma y not of fer a high signal-tonoise ratio, so we need to see earl y dif ferences. Due to tumor hetero geneities some tumor compar tments ma y respond to therapy with massive cellular edema and eventual rupture leading to more unbound w ater with longer free diffusion paths and thus higher ADC values, whereas other cellular dense areas ma y actuall y become more packed due to the pressure of adjoining, e xpanding compartments resulting in reduced ADC values. In such scenarios, the mean is relati vely insensiti ve to all of the balanced but opposite reactions. After observing such effects and the f ailure of v olumetric means to discer n early tumor diffusion changes, Mof fitt and Ross e xamined the voxelbased changes and chose a tw o-dimensional (2D) scatter plot analysis method, where the locus of each tumor voxel is based on both its earl y (abscissa) and late (ordinate) ADC v alues. The method of deciding v oxel pairing between the early and the late dif fusion scans was based on volumetric registration of the dif fusion data sets. On the basis of a noise anal ysis involving the contralateral brain, thresholds on the 2D scatter plot for calling tr ue change were set at an α = 5% and those v oxels above, below, and within those limits w ere colored in a threedimensional (3D) tumor overlay for visualization. In their publication, the authors named the v oxel-based change analysis “functional Dif fusion Mapping (fDM). ”3 This method suppor ts simultaneous measurement of both increased and decreased parameters, and based on the preliminary recei ver operating characteristic anal yses,

263

the resulting area under the cur ve is g reater for the fDM than for the spatial mean. It seems likely that the sensitivity and the specif icity of the inter val perfusion data will benef it from this anal ysis method as w ell. The generalization of this v oxel-based change detection method to diffusion and perfusion is called response parametric mapping.

REGISTRATION’S ASSISTANCE IN SUCH CHALLENGES Theory and Design of Registration Algorithms Before we describe what might be accomplished using registration, it is insightful to study the “how’s” and “why’s” of registration at a lo wer, w orking le vel of e xpertise. If w e reject (and we do as biased) the older , manual methods of identifying homolo gous points (e xcept as appro ximations for initializing a re gistration algorithm), then w e are left only with using the mutual infor mation content of the data sets we wish to re gister independent of the cost/objecti ve function and method, for e xample, intensity-based, feature matching, w e implement. Here, it is impor tant to understand that if w e are re gistering two data sets automaticall y (other than approximately aligning their initial poses manually) it is the mutual infor mation between the two data sets that is impor tant. If the infor mation content of one of the two image v olumes to be re gistered is v ery high and the other is low, the resulting accuracy of the registration is limited by the lower information data set; mutual infor mation can be no greater than the lower information content of the two data sets and typically is less. Although registration is typically dichotomized into feature-based 6–10 or intensitybased11–28 methods, we will discuss only the intensity-based registration method; the interested reader can access the supplied references for each as desired. The typical components of automatic, intensity-based re gistration method consist of the following: 1. an objective function, that is, something to be maximized, for e xample, mutual infor mation, or a cost function, that is, something to be minimized , for example, sum of square error (SSE), to measure how well the data sets are re gistered; 2. a defor mation inter polant, that is, a function that mathematically describes the defor mations between the control points changed by the optimizer; 3. an optimizer, that is, algorithm that moves the control points or changes the coef ficients of the defor ming interpolant; and

264

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

4. a re gularizer that smoothes defor mations and thus biases the inter polant against “folding,” that is, ha ving a one-to-man y mapping (see F igure 1 for a 2D example of folding).

data sets to be registered. For coregistration, that is, the registration of a pair of data v olumes, mutual infor mation is most easily described as that term which must be subtracted from the sum of the tw o indi vidual a verage infor mation terms to mak e up for their cor relation to yield the a verage joint information of the pair, that is,

RE 1, Objective Functions

H ( a, b ) = H ( a ) + H (b ) − I ( a, b ),

For completeness we will mention that there are man y objective and cost functions. F or isomodality registration prob lems typical cost functions include SSE, sum of absolute er ror (SAE), trimmed (ie, robust) versions of SSE and SAE, and cross-cor relation. But problems often arise using these cost functions w here something in the f ield of vie w of the scan is dif ferent, for example, the oral contrast in one scan is in different loci than in the other scan or the phase of the contrast injection during acquisition is dif ferent betw een the two scans. It is for these reasons that w e advocate the use of the mutual information as the objective function for almost all re gistrations. Such dif ferences as those just listed ha ve little noticeab le ef fect on the results of re gistration using the mutual infor mation objective function. One of the most robust of objective functions is mutual information and its deri vatives, for e xample, nor malized mutual information, and it is capable of registration even in the presence of nonmonotonic intensity relationships in the

where H(a, b) is the a verage joint infor mation of a pair of random variables/images, a and b, H(a) and H(b) are the a verage information of each v ariable/image, and I(a, b) is the mutual infor mation shared between images a and b.

Reference frame coordinates BE

Since the a verage infor mation of the v ariable a, also called its entropy, is def ined as

H ( a ) = − ∫ p( a ) ln( a ) da, where

p( a ) =

∫ p(a, b) db

p(b ) =

∫ p(a, b) da

and

0.095828

Homologous frame coordinates

1

1.2

0.9 1 0.8 0.8

0.7

0.6

0.5

y

y

0.6

0.4

0.4 0.3

0.2

0.2 0 0.1 0

0

0.2

0.4

0.6 x

0.8

1

0.2 0.2

0 0.2 0.4 0.6 0.8 1 Move / Add / Delete / Relax/1-ref / 2-hol / Force

1.2

Figure 1. The two-dimensional folded deformation of the “floating” image computed using ten homologous control point pairs where one control point (left central) has been displaced to the right of the right central control point. Here, the mapping in the folded region is one-to-three, not one-to-one.

Potential Roles for Retrospective Registration in Molecular Ima ging

are the mar ginal probabilities of the joint probability p(a, b), then it is easy to deri ve that

I ( a, b ) =

∫∫

p( a, b ) ln

p( a, b ) da db, p( a ) p(b )

which is the usual formulation for calculating the mutual information. Note that as mutual infor mation increases, the estimate of the average joint information, that is, joint entropy, decreases because the information common, that is, mutual, to both marginals cannot be double counted in the joint measure. When tw o images that describe the same scene, but potentiall y come from dif ferent sensors, are in register on some common coordinate scheme, their mutual infor mation is g reatest because the tw o images are highl y cor related in the statistical sense. If this is thought of as clustering on a 2D axis as a function of the intensities for the two images at each common voxel, the clusters formed are as tight as they can be given the noise in each image. As the images are slo wly misre gistered, the clusters rapidly widen/blur and the joint density function be gins to approach a unifor m distribution that has higher joint entrop y because the infor mation required to describe it is lar ger, that is, the mutual infor mation has dropped. The reverse of this description occurs in re gistration of tw o statisticall y cor related images w hich are out of re gister, that is, the algorithm iterati vely searches for the geometric transfor m between the two images that minimizes joint infor mation and simultaneousl y maximizes mutual information. Here, mutual infor mation can be used as the objecti ve to be maximized in the re gistration process where the act of registration is written as

⎧⎪ T = arg maxT ⎨ I ( a, T (b ) ⎪⎩ =

∫∫

)

p ( a, T (b )

)

⎫⎪ p ( a, T (b ) ln da db ⎬ , p( a ) p (T (b ) ⎪⎭

)

)

that is, f ind the transform T that maps image b onto image a such that the argument within the braces {} is maximized; in the same manner , the joint information can be used as a cost to be minimized. One last comment about information theoretics: note that mutual infor mation as for mulated abo ve can be viewed as the K ullback-Leibler (KL) di vergence or directed distance betw een the joint density function and the product of its marginals.29 The previous statement can be seen from the def inition of the KL-directed distance which states that the average distance from distribution p1 to p2 is defined as

∫

265

⎛ p1 ( a ) ⎞ p1 ( a ) ln ⎜ da. ⎟ ⎝ p2 ( a ) ⎠

Recall that, p(a, b) = p(a)p(b), the product of its marginals, only when the distributions of a and b are independent. Thus, optimizing mutual infor mation can also be seen as dri ving the KL distance from p(a, b) to its independent form, p(a)p(b), as far as possible, thus maximizing correlation. The KL-directed distance has utility in at least one other way. If we are registering the same anatomy between different imaging modalities repeatedl y, instead of treating each re gistration as an independent process, w e consider minimizing (sic) the KL-directed distance between the joint distribution generated b y the re gistration process and its expected distribution,30 that is,

)

⎧⎪ p ( a, T (b ) T = arg min T ⎨ ∫∫∫ p ( a, T (b ) ln E p ( a, Ti (b ) ⎪⎩

)

{

)}

⎫⎪ da db ⎬ . ⎪⎭

Initially, w e for m the e xpected joint distrib ution, E p a, Ti (b ) , by a veraging se veral joint probability distributions obtained from N independent re gistrations, w here i = [1: N]. Re gistration b y minimizing KL divergence is v ery efficient, but if the images to be re gistered have other features that do not match the expected distribution, the results may be very different than expected. Obviously, implementations must be discrete rather than continuous as described b y the pre vious equations and therein lies a considerable deal of magic and functionality. To compute mutual infor mation it is necessary to:

{ (

)}

a. build the discrete joint probability density function so that its constr uction represents the statistics of the underlying distribution; or b. use other approximations, such as entropic graphs, that build tree g raphs instead of histo grams. This approach is cer tainly ideal for multidimensional applications, such as vector intensities, vector features, or simultaneous joint optimization of multiple images, 31,32 and ma y ha ve man y other uses as well. To achieve (a), the discrete histo gram must be carefull y built using the appropriate number of bins 33 or if man y bins are used , appropriate histo gram smoothing must be implemented with Parzen windowing.34

266

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

RE 2, Deformation Interpolants

Among the e xisting re gistration packages, defor mation interpolants are quite v aried and include the follo wing methods of inter polation: pol ynomials, F ourier e xpansions, thin plate splines (TPS), cubic B-splines, radial basis functions, etc. Here, the general notion is to reduce the de grees of freedom (DOF) of the defor mation that must be computed by defining the defor mation at only a few loci and inter polating the remainder . Cubic B-splines, popularized b y Rueck ert35,36 as free-for m deformations and se veral radial basis functions (multiquadric and Wu’s)37 enjoy only local e xtent and thus are typically f aster to compute than functions with global extent. Other interpolants have global extent and include linear (rigid body through full af fine) and nonlinear transforms. One popular nonlinear inter polant is also a radial basis function, that is, the TPS, which has intrinsic regularization attributes38 that we will discuss in detail in section “RE 4, Re gularizers.” Note that 3D inter polants that require more than 12 DOF are nonlinear w arping interpolants; those with 12 DOF or fe wer are linear and members of the af fine class; see Table 1. Also note that the user typically picks the DOF of the interpolant needed for a specif ic job.

RE 3, Optimizers

There are multiple optimizer algorithms from w hich to choose. We can dichotomize the description of optimizers into those that do not require par tial derivatives and those that require partial derivatives. Those in the category of not requiring derivatives include the Nelder-Mead simplex and Powell’s one-dimensional search algorithms. 39–41 These approaches tend to tak e longer to con verge but of fer less sensitivity to noise in the objective/cost functions. Depending on conditioning, the computation of mutual information Table 1. DEGREES OF FREEDOM (DOF) FOR LINEAR TRANSFORMS 3D Linear Transforms

DOF Total

Component DOF Translation Rotation Scaling

Shear

Rigid body

6

3

3

—

—

Isotropic scaling

7

3

3

1

—

Anisotropic scaling

9

3

3

3

—

Full affine

12

3

3

3

3

can be noisy; Maes 42 has examined this area thoroughly in an excellent publication. In contrast, optimizers that require derivatives are much faster for low noise objective functions in that they require fewer iterations to find the solution, but frequently will not con verge to the global optimum under noisy conditions. Just a shor t list of these optimizers include steepest descent, conjugate g radient, and quasiNewton algorithms, such as Fletcher -Powell and Levenberg-Marquardt.41 A major f actor contributing to the computational load for these algorithms is estimating the derivatives; w hen the objecti ve function is not described analytically, the f inite difference approximation to deri vatives requires the time to e xperimentally examine registration outcomes of these f inite steps to estimate these derivatives. Recentl y, stochastic appro ximation of these derivatives has helped reduce the computational load further. These techniques randoml y update onl y a fe w of the required derivatives at each step with little or no concomitant increase in the number of total iterations required thus reducing the computational time signif icantly.43 RE 4, Regularizers

Regularization, as its name implies, forces the computed deformations in the re gistration process to be smooth, a necessary, but not sufficient condition for good registrations. Ideall y, the mappings w ould also be one-to-one (not folded), and consistent, that is, if point A in data set 1 maps to point B in data set 2, then point B in data set 2 maps e xactly back to point A in data set 1. 44 Further, the mappings from A to B and B to A should be everywhere differentiable up to some stated order; this condition is referred to as a diffeomorphism.20,24,45–52 For example, TPS are e verywhere continuousl y dif ferentiable and are intrinsically smooth, that is, have the least bending energy (integral of all second-order derivatives) of all inter polants that pass through the homolo gous points, but are strictl y consistent onl y at the control points. F igure 2 sho ws a defor mation inter polant computed for f ive homolo gous point pairs; note that e ven though the TPS has the least bending ener gy, that is, the smoothest of all interpolants, it still can be driven to fold as sho wn in F igure 1. Cubic B-splines can be made smooth by the addition of a bending energy penalty, that is, the inte gral of all second-order par tial derivatives of the inter polant, to the objecti ve function. Additionally, diffeomorphisms can be implemented b y limiting the deformation at each le vel of the scale. 24,50 Intrinsic diffeomorphisms arise in nature (eg, viscous flow) and thus more computationall y intensi ve algorithms implement

Potential Roles for Retrospective Registration in Molecular Ima ging

267

varying k, the number of scalars comprising the model vector X’s coef ficients, that is, the model’ s DOF . If k is too small, then X is a poor model and

{

–2log p ( y | X

Figure 2. A two-dimensional thin plate spline (TPS) warping generated by five homologous control points where only the middle point is displaced in the floating “image.” The loci of points constructing the left reference mesh is mapped to the right “floating image” after first computing the coefficients of the TPS model using the loci of both sets of five homologous control point pairs.

solutions to the Navier-Stokes flow equation or integrate velocity vectors.25,49,53,54

Accuracy Limitations The issues of noise, infor mation content, and DOF of the applied model’s interpolant are routinely handled in statistical modeling of all kinds of data sets. F or instance, w e know, in general, that if w e try to f it data with a statistical model having very few DOF, the f it is robust to noise, but yields an unsatisf actory approximation to the data. Lik ewise, if we use a model having very high DOF, the performance of the model in noise is unstab le, that is, other data sets drawn from the same noisy e xperiment do not f it as well. Tools for making this trade-off between DOF and fitting error have been available since Akaike55 published the first article dealing with this topic. After the Akaike Information Criteria, other criteria that follo wed of fered the desirable innovations such as decreased bias, for e xample, the often used Ba yesian Infor mation Criteria (BIC). 56 Given a user -constructed model X and the data v ector y consisting of n observations, recall that the less biased BIC is expressed as

{

}

BIC = − 2 log p( y | X ) + k log( n), where k = DOF is the number of independent v ariables in the model v ector X. The BIC is typicall y used for choosing model comple xity and is minimized b y

)} is very large despite the use of small

k. If k is picked large enough such that the model y well fits the observations of the process being modeled, then the BIC is near its minimum; increasing k further increases the complexity of the model without improving the f it and causes the BIC to increase again. From the preceding discussion, w e kno w ho w to pick the DOF of a model for an gi ven infor mation content. But as the local infor mation content (ie, “density”) of the data y increases/decreases, the probability of y at the ideal solution X0 is higher/lower (more/less peaked), and thus k, the model’ s DOF must increase/decrease, respecti vely, to k eep BIC minimized. The analo gy with r egistration is str aightforward. The higher DOF desired for the w arping inter polant (ie, the geometric model of the underl ying w arping) can be suppor ted onl y w hen the mutual information between the tw o regions being mapped is suf ficiently high (this statement is tr ue even when mutual information is not the criterion being optimized , e g, minimizing SSE). Examples are abundant. F irstly, consider the case in 3D w here the DOF required for modeling tr ue rigid body re gistration are onl y six, that is, three rotation and three translation parameters. Depending on the size of the volumes being mapped under this model and the contrast-to-noise of the modalities for the specif ied voxel size, the mutual infor mation for a lar ge volume can be quite high and the resulting re gistration accuracy can easil y be one to tw o orders of magnitude smaller than a v oxel at the centroids of the re gions (rotation er rors increase positional er ror at increasing radii from the centroid). No w consider w arping a v olume w here def ining the spatial w arping inter polant requires higher DOFs than linear re gistration. Gi ven that mutual infor mation can change locall y (def ined at any given level of scale on the reference mapping), the analogy becomes one where the local DOF density used in def ining the local warping should track (ie, change) with the local mutual infor mation density to k eep the equivalent of BIC minimized and the model of the warping optimized. In w arping the f ailure to k eep the DOF of the w arping near the minimum of the BIC without other maneuv ers, such as increasing or decreasing regularization penalties or other equi valent tactics, r uns the risk of computing locall y undesired , for e xample, folded , or o ver-regularized mappings, respectively, where both outcomes are inaccurate.

268

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

EXAMPLES OF RESPONSES TO CHALLENGES Mapping Histology into in Vivo Imaging Mapping, (ie, registering) histology into in vi vo imaging has been attempted b y man y in vestigatorss.57–65 Firstly, methods that require human marking of homologous features for de facto registration are biased. Assumptions that evenly spaced points on radial boundaries are homologous are flawed; yes, the boundaries should be in alignment, but not b y forcing specif ic, ar tificially placed , nondescriptive points on the boundar y to match other equally biased (in the sense that the y cannot be cor rect) nondescriptive points. F inally, methods that do not f irst map each histolo gy section back onto their o wn photograph of the tissue b lock’s f ace (ie, b lock f ace photo) taken before the tissue sample w as sectioned are also seriously flawed. This important step removes random 2D deformations associated with cutting the section from the block that cannot later be compensated b y 3D defor mations. Most importantly, many methods make the remarkable assumption that the plane of sectioning in histolo gy can be found as a plane in the in vi vo image v olume. Given the man y defor mations, the tissue sample undergoes in its e xtraction from the in vi vo state, it is v ery unlikely that the plane of in vi vo imaging still represents a plane in the tissue block; yes, a curved 2D manifold, but a not plane. Methods that map (2D w arp) histology sections back to their tissue b lock geometr y and then map (3D w arp) the tissue b lock geometr y into their in vi vo image volume have the potential to be f aithful registration methods. 58,66 Although some re gistration methods require the preplacement of e xtrinsic f iducial markers in the tissues of interest, in this study , we demonstrate the use of one that depends onl y on the mutual information content of each of the data sets. Human Prostate

In cur rent medical practice, most patholo gists are too busy to gather and e xamine a lar ge number of thin sections. Instead most will g rossly slice the or gan into multiple larger pieces and then prepare one or tw o 10 to 20 µm thin sections from each lar ger piece for staining and microscopy. Thus, methodologies such as, 58,66 which require uniform sectioning of a whole organ for mapping into in vi vo imaging, must be modif ied to allo w sparse sampling. Additionally, w hile patholo gists onl y need standard 1ʹ′ × 3ʹ′ slides, larger specimens may require special handling, such as w hole mount, 2 ʹ′ × 3ʹ′ slides which may require commercial processing.

Further, in line with the preceding discussion of accuracy, w e ma y need an inter mediate mapping stage to register histological data faithfully into in vivo image volumes. Recall that the in vi vo images of some anatomies are information limited; CT essentiall y has no soft tissue contrast obser vation capabilities unless associated with dynamic contrast studies involving X-ray contrast agents. While MRI is kno wn for its good soft tissue contrast, it typically has lo wer resolution than CT , is susceptib le to motion ar tifacts, and often has penetration prob lems in people with larger body habitus. Thus, the use of an intermediate MRI target between the in vivo image volume and histology ma y be useful in se veral w ays. The use of an ex vi vo specimen for the inter mediate MRI tar get is attractive because the specimen can be imaged immediately upon resection without imaging problems associated with attenuation of overlying tissue or motion. The sample can be preser ved in buf fered for malin w hile imaging. Although the geometr y of the e x vi vo specimen is deformed from its in vi vo state, it has not y et undergone additional defor mations associated with the sectioning process; thus, in terms of bending energy the ex vivo specimen lies somewhere between the in vivo volume and that of histolo gy. The consequence is that lo wer DOF are required to map between the lower information content in vivo MRI v olume and the high infor mation content ex vivo MRI v olume, than would be required to map the in vivo MRI directly to the geometry of the high information content histology specimen. Additionally, where high mutual information is required to map between the ex vivo MRI volume and histology due to the high DOF deformations potentially required, the information content of both data sets is sufficiently high to support such requirements. Figure 3A sho ws the mapping of six histolo gy slides in block face geometry back into the ex vivo 3T-MRI. Figure 3B shows the second histolo gy slice mapped into b lock face geometry, while Figure 3C shows overlays of slice 2’s histology mapped to the 3T-MRI in vivo T2-weighted and diffusion-weighted images with 11C-choline PET also mapped to the in vi vo 3T-MRI diffusion study. Complete details of this re gistration are a vailable in the study b y Park and colleagues. 67 Rodent Brain

Consider the process of e xtracting a rat brain from its cranial vault previously seeded several weeks earlier with a 9L glioma. Upon opening the cranium, the highl y pressured brain pushes the tumor outw ard in the process of stress reduction. Then, the brain is pried loose from its cranial base and placed in a test tube of buf fered for malin while it is imaged at 7T; these ex vivo images serve as the

Potential Roles for Retrospective Registration in Molecular Ima ging

A

B

C

Figure 3. A, The mapping of six sparse histology sections into the ex vivo prostate as seen on three orthogonal planes (origin of the planes is shown at the center of the three axes). B, Histology slide 2 of the previously visualized six slides with the boundaries of the prostatic tumors outlined by the pathologist. C, First row: • column 1: T2-weighted magnetic resonance imaging (MRI). • column 2: computed apparent diffusion coefficient (ADC) parametric MRI image mapped onto T2-weighted MRI. • column 3: 11C-choline positron emission tomography mapped onto T2-weighted MRI. The second row shows the registration of histology slide 2 in Figure 3B mapped into the geometry of the in vivo diffusion MRI shown in row one; the histology slide has been color coded in a green hue. The third row shows the registration of the histology slide mapped into the geometry of the in vivo diffusion MRI as shown in row one: the registration is depicted using the checkerboard display where alternating blocks are intensity coded from alternating image sources.

269

intermediate image volume for registration between the in vivo MRI images and the histology’s block face geometry. After ex vivo imaging, the specimen is placed on a glass slide and pack ed with dr y ice. Upon cr yomicrotome sectioning, the tissue block will retain the geometr y in which it was f irst frozen on the glass slide. Clearl y, the brain’ s frozen tissue block has undergone significant deformations from its in vi vo status. A planar histolo gical section through the tumor cut after the ex vivo specimen expanded most assuredly will not be planar in the in vivo image volume. In F igure 4, we see that the re gistered planar histology section passes through the tumor in a v ery nonplanar fashion. F igure 5 sho ws the accurac y of the mapping in delineating a hematoma within the tumor . F igure 5A shows the histology slide restored to the block face geometry with an outline traced around a hematoma, F igure 5B shows the same outline mapped onto the MRI dif fusion scan showing higher dif fusion within the hematoma, and Figure 5C sho ws the o verlay of A and B in check erboard block without the outline. The complete methodolo gy is described in Meyer and colleagues. 68 The pre vious demonstration lies at the e xtreme of what can be accomplished with mutual infor mation and sparse data, that is, mapping one histolo gy slice back to its in vivo image volume. In general, it is preferable to use more than one slice due to the increased infor mation obtained from simultaneously registering multiple linked slices, each contributing additional mutual information to a sparsely sampled v olume registration process and providing constraints on the loci of adjacent slices imposed by minimum bending energy.

Figure 4. The curve of the two-dimensional histology manifold mapped back into the in vivo MRI of the brain and the tumor. Note that after opening the cranium the brain decompressed by displacing the tumor peripherally, and it was in that expanded geometry that a plane was cut through the frozen tissue block. Reprinted with permission from Principles and Practice of PET and PET/CT, RL Wahl, 2nd Ed., LWW Inc. 2009; p. 114.

270

A A

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

B B

C C

Figure 5. The histology slide (A) shown in the geometry of the block face volume as mapped onto the in vivo apparent diffusion coefficient diffusion magnetic resonance imaging (B). The same region of the hematoma is outlined in both (A) and (B). The registration of (A) and (B) is shown in checkerboard in (C). Reprinted with permission from Molecular Imaging 2006;5:22.

Registering Brain Tumors across Interval Exams In addition to local growth, brain tumors also invade adjacent tissue via w hite matter tracts. The invasion process can be visualized on Flair -weighted MRI image v olume when the nor mally dark tracts become h yperintense similar to the tumor body pre vious to in vasion. This hypointense to h yperintense intensity change can pose a problem for registration where the investigator has chosen a cost function, such as SSE, because the project of registering Flair -weighted images betw een inter val e xams appears to be a simple isomodality re gistration problem. With this cost function, the mapping of the ne wly hyperintense white matter tracts would be mapped back onto the original, hyperintense tumor body, which in this case is a mistake because the in vasion has af fected ne w material outside of the geometry of the previous tumor. Avoiding this problem is precisely the major benefit of using mutual infor mation as an objecti ve function e ven within the same modality. If a subset of the brain changes grayscale, but not geometr y (on the Flair MR sequence in Figure 6B w hite matter inf iltration is seen as a ne wly enhancing re gion), then this intensity-changed material simply leads to another peak in the joint density histo gram used to compute MI w hen the v olume is cor rectly re gistered, and MI is maximized as sho wn in Figure 6. As seen in Figures 6A and 6B, the MRI Flair v olumes shown were acquired 1 month apart for this patient harboring an aggressive gliob lastoma multifor me (GBM) tumor . Due to the high (93) DOF w arping in the re gion of the tumor to map the later image volume of the tumor at time 1 (T 1) onto the earlier image volume at time 0 (T0), the size change of the GBM is not evident from the images, but it can be quantified by integrating the resulting Jacobian. Note that the registration of T1 onto T0 correctly places the enhanced white matter tracts of T1 onto the unenhanced white matter tracts in T0 as desired , the g ray matter folds and the inter nal structure of the lesion is preser ved as well.

A

B

C

Figure 6. Demonstrates the value of using mutual information as an objective function even in registering the same (iso) modalities. The change in white matter tract intensities pre/post invasion would have duped cost functions, such as sum of square error. A, shows a slice through the patient’s tumor in the early T1-weighted magnetic resonance imaging examination. The arrows depict loci of gray matter cortical folds in the brain anterior to the tumor. B, shows the same slice as seen in (A) 1 month later after the cancer has infiltrated the white matter tracts medial and anterior to the tumor. Note the presence of the same gray matter cortical folds. C, demonstrates the registration of the later exam in (B) onto the earlier exam in (A) by showing (A) to the split’s right, and (B) to its left.

The preceding discussion assumes that the time course of the lesion’s growth is sampled sufficiently often to suppor t f inding tr uly homolo gous features in time adjacent data sets. Most warping algorithms can map two disparate volumes onto the same geometry, but if the volumes do not contain enough distinct, unique features, the resulting re gistration is simpl y imposed b y the implicit regularization of the inter polant and has almost nothing to do with the temporal sequence of changes that led to the final morphological difference. Since aggressive clinical tumors change rapidly, to attach physical meaning to following lesion v oxels o ver time w e must image the patient suf ficiently often to model local voxel g rowth, regression, and/or displacement, that is, the warping must be dri ven b y mutuall y present, homolo gous features in the data volumes.

Early Detection of Therapeutic Response in Tumors Recall in the pre vious discussion that the benef it of voxel-by-voxel analysis of inter val change appears to be both increased sensitivity and specificity for tumors with heterogeneous response to therapy. In our experience, the most often offered reason by others why such procedures are theoretically “nice” b ut “not practical” relates to the general perception that accurate re gistration betw een deformable soft tissues is not possib le. Despite the agile capabilities of warping registration, there is another reason that makes such concerns relatively moot, that is, we are primarily interested in detecting early change where the ter m early is measured relati ve to visualization of morphological change. If w e w ait long enough, as in

Potential Roles for Retrospective Registration in Molecular Ima ging

measuring radiological response at 10 weeks, we can see the mor phological change and the ter m early no longer applies. Thus, the ter m early implies that the inter val between exams will be suf ficiently shor t such that morphological changes are relati vely small. In this chapter , we will address re gistration for a defor mable soft tissue organ, the female, human breast.

REGISTRATION OF INTERVAL BREAST EXAMS To demonstrate that (a) breast re gistration is relati vely straightforward and (b) automatic re gistration is easil y within reach, w e have included the images of F igure 7, which shows the results of the inter val registration of a normal breast using e xams separated b y 2 da ys; no special repositioning techniques, such as skin marks, w ere used. The registration, or lack thereof, is sho wn using a checkerboard technique w here alter nating squares come from the reference and resulting registration of the second interval examination. For this demonstration, we are using a diffusion acquisition with a 7-channel breast coil from Philips 3.0T, that is, the b0 acquisition, w here the dif fusion gradients are tur ned off; SENSE = 2 w as chosen to reduce artifacts. Voxel size is 0.85 × 0.85 × 2 mm 3, and the information content of the scan is remarkab ly high. Our registration software’s user interface shows all of the control points in the reference data v olume and the first N points in the “floating” v olume to be re gistered. The f irst N control points ma y be placed b y the user to

271

initialize the poses of the tw o volumes to be within the capture range of the automatic algorithm or the y may be placed fully automatically within a user-defined cropped region that appro ximately cor responds to a similarl y defined cropped re gion on the reference image v olume. The optimizer moves the control points in the “floating” image as necessary to maximize the mutual infor mation between the tw o data sets. After each mo vement of points, the algorithm recomputes the new interpolant and reconstructs the floating volume so that the mutual information can be computed between the reference and the “floating” volumes. The w arping re gistration sho wn in F igure 7 w as achieved by providing an automatically produced matrix of 3 × 3 × 3 control point “handles” in the reference data set based on volume bounds obtained by the user’s cropping down to the same breast in both inter val exams. If both breasts are in volved, it is more ef ficient to re gister them as separate pairs after the associated cropping and then combine them b y addition of the separate control point sets if desired , followed by more optimizing iterations using all control points so that interactions between relatively close control points may be appropriately compensated. The algorithm w as star ted b y automaticall y providing five approximate homologous control points in the second, that is, “floating, ” inter val e xam that cor responds to the position of the f irst 5 control points in the reference set of 27, again based on the resultant cropping. Although the reference set of points remain f ixed, an

Figure 7. Bilateral breast registration (repeat examination) shown at multiple levels using the checkerboard matrix presentation of alternate data sets.

272

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

optimizer moves the control points in the second set to warp the second inter val exam so that it will match the reference. A cr ude warp was optimized using these f ive point pairings on a decimated version of the data set, and then this solution w as used to instantiate the f irst 10 points from the reference to repeat the process. After repeating the process for 15, 20, and f inally all 27 points on the full, nondecimated data set, the f inal solution was computed. Although each point represents three DOF, the final solution represents the use of 81 DOF in the w arping. The choice of decimation and DOF used at each stage is controlled b y a schedule that is selected and can be edited by the user.22 TPS are used to interpolate warping deformations between control point locations.38 Since the control points are also distance sorted, that is, the first N points are separated b y lar ger distances than the remaining 27-N points for N = [2:25], using a schedule of progressively increasing control points results in a scalebased set of solutions w here the lar ger scale prob lem is solved before the smaller. We performed the following simulation experiment to quantify our expected-capture range for interval breast MR registration. A breast MR, that is, the b0-w eighted diffusion image (D WI), cropped left breast v olume shown in Figure 7, was deformed in a known way, used as the second interval scan, and then we try to register the deformed MR with the original MR to reco ver the defor mation. The left breast data sho wn in F igure 8A is defor med using a perturbed uniform grid of 3 × 3 × 2 = 18 control points using TPS as the geometric inter polant. Zero mean Gaussian noise of varying standard deviation is added to the location of control points to realize a kno wn, random deformation. A total of 65 kno wn defor mations are generated using standard de viation v alues betw een 3 and 15 resulting in mean deformations between 2 and 24 mm. We then register the original MR with the homologous deformed MR using 4 × 4 × 2 = 32 control points using TPS for a total of 65 registrations. Note that we have intentionally used a different configuration of control points for registration than w e used for the original defor mation. Contrary to man y held beliefs otherwise, e ven though we used TPS for the defor mation, it is not possib le to exactly recover the deformation using TPS with a different configuration of control points. We used TPS to perform the random deformations because we have code readily available to compute the Jacobian of the defor mation. We w ere concerned that in generating the random defor mations we had inadvertently caused the defor mation to fold , that is, become a one-to-many mapping; if that happened, we had no chance to reco ver the defor mation by registration. By checking the resulting random defor mations using the

A

B

C

Figure 8. A, Three orthogonal rendering planes cutting through the original b0-diffusion breast volume. B, Same planes cutting through the breast with known applied deformation. C, Shows grid associated with the deformation of volume A into B (mean deformation of 17.6 mm).

Jacobian, we were able to see that the defor mations were indeed sometimes folded for mean er ror defor mations of 10 mm or g reater. Thus, w e used the ne gative Jacobian criteria to discard such defor mations and randomly regenerated others that satisf ied the criteria of a nonne gative Jacobian everywhere in the deformed data set. We initialized the re gistration process using identically the same f ive control point loci for all homolo gous, deformed data sets. We also used a multilevel registration

Potential Roles for Retrospective Registration in Molecular Ima ging

approach, where we systematically increase the number of control points in stages such that w e solve the long range problem f irst and then iterati vely work at a smaller scale by using more points. Each ne w stage be gins with more control points placed in the homologous data set based on the best optimization at the lar ger, previous scale. The displacement error is computed between the known deformation and the reco vered deformation for 65 random deformations as shown in Figure 9. Figure 9 shows that up to an original mean defor mation approaching 18 mm, the registration algorithm results in subvoxel displacement error Mean deformation before and after registration

Registered mean deformation [mm]

14 12 10 8 6 4 2 0

0

5

10 15 20 Original mean deformation [mm]

25

Figure 9. Shows the final residual deformation (error) associated with fully automatic registration of originally deformed breast. Note that the error is on the order of 1 mm just before capture rate degrades. The slope of the slowly increasing residual error is due to increased smoothing of the grossly deformed breast data sets, that is, the larger spacing between original voxels after gross deformation and the resulting degradation of the trilinear interpolant’s performance under such conditions leads to less high frequency information to support registration.

273

(eg, ~1 mm or less); above this 18 mm threshold, the rate of capture becomes problematic. Thus, we believe that almost all mean deformations up to the limit of 18 mm mean deformation should be within the e xisting capture range of our algorithm yielding registrations under 1 mm average error. Although a mean deformation of 18 mm might sound small, in reality the local defor mations can be quite high and the scan might be severely deformed locally. Figures 8B and 8C sho w such an e xample w here the defor med MR has a mean defor mation of 17.62 mm, w here the maximum local deformation is 46.51 mm! Also note that the nonzero slope of the reco vered mean defor mation in Figure 9 is due to the smoothing artifact of trilinear interpolation in computing the resulting initiall y defor med data set. The lar ger the defor mation, the more distance there is between original pixels in the defor med data set, which increases the point spread function of the trilinear interpolation and increases the smoothing artifact, that is, reduced information content due to mechanical deformation of the breast does not occur and is an ar tifact of this postprocessing experiment. Figure 10 shows the automatic registration of another set of “coffee-break” exams, that is, the two pretreatment exams follo wing manual cropping. The data are from another consenting patient par ticipating in the IRB approved study using the 7-channel, sense-capable breast coil on the 3T Philips research magnet. These registered data sets come from tw o coronal b0-D WI acquisitions separated by 15 minutes during which time the patient is removed from the magnet’s table, and then is retur ned to be repositioned and rescanned. The registration is shown using α blending of the color combination of the two data sets by presenting the reference data set in a b lue-green hue and the registered data set in g rayscale. While Figure 10 was generated to show results typical of automatic re gistration of the w hole breast, in the

Figure 10. Four of 104 slices of a co-registered coronal b0-weighted diffusion image (DWI) b0 baseline (grayscale) and repeat b0 baseline (aqua/green hue) exams mapped onto T1-FatSat axial scan geometry.

274

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

following section we show automatic registration of just the lesion in the same patient for the prechemotherap y and postchemotherapy exams. In Figure 12, we show three triads of regularly spaced control points automaticall y generated based on the manually, tightly cropped volume bounds (35 × 35 × 18 voxels in this case only) containing principally just the primar y breast lesion in the pretherapeutic e xam. The w arping results are sho wn in F igure 12, w here the coronal posttherapeutic b0-DWI lesion has been geometrically mapped (w arped) onto the coronal pretherapeutic

Figure 11. Three sets of colored triads (blue, white, blue) of control points in the cropped volume of the breast lesion displayed via three orthogonal cut planes.

b0-DWI reference lesion using just the nine control points from the three triads. The two registered tumors are shown using the α-blended color combination of the two data sets, where in this case the pretherapeutic scan sho wn in the yellow hue is more visib le for larger amplitudes. Misregistration is observed when the two “hues” (yellow in this case and grayscale) appear separatel y with spatial of fsets. Individual voxels are visible in this 35 × 35 coronal view. Because the number of v oxels that the algorithm m ust register is now very small, the warping registration computation times drop propor tionally anywhere from 1 to 2 hours to 1 to 2 minutes. Additionally, the registration of the lesion is typically more accurate than that achie ved by registration of the w hole breast, w hich is e xactly w hat is needed because fDM is computed for lesions! The schedule for this registration used the f irst 3 control points to rigidly align the tw o lesions, then with increased accurac y used five control points to g rossly w arp the tw o lesions, and finally used all nine control points to ref ine the w arping. The use of such schedules enables the algorithm to robustly and quickl y capture the solution al ways w orking from larger to smaller scales. Such schedules are profor ma for a given control point configuration and density, which in turn is driven by the information content of the image sets.Thus, once a schedule is de vised for the data from a par ticular scanner and sequence, it can be reused for other similar tasks with little, if an y, revision. In some cases, w here the warping between the tw o data sets is e xcessive it ma y be necessary to star t the schedule with the lo west DOF warping solution, that is, five control points in 3D, instead of initially using the rigid body geometr y model to suppor t

A A

B B

C C

D D

E E

FF

G G

H H

II

JJ

Figure 12. Series (left-to-right, top-to-bottom) of coronal slices through the posttherapeutic primary breast lesion (shown in grayscale) warped onto its pretherapeutic scan (in yellow hue) using the nine reference control points shown in Figure 11.

Potential Roles for Retrospective Registration in Molecular Ima ging

capture of the large scale warp as was done for the capture range e xperiment presented previously in Figure 9. But since the mechanical elastic modulus of most cancers69 is typically an order of magnitude g reater than the reminder of the breast, rigid re gistration for the initial element of the schedule is likely to be all that is ever needed in terms of capture range for re gistering just the breast’s primary lesion.

17.

18.

19. 20.

REFERENCES 21. 1. Therasse P, Arbuck S, Eisenhauer E, et al. New guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst 2000;92:205–16. 2. Gehan E, Tefft M. Will there be resistance to the RECIST? J Natl Cancer Inst 2000;92:179–81. 3. Moffat B, Chenevert T, Lawrence T, et al. Functional diffusion map: a noninvasive MRI biomark er for earl y stratif ication of clinical brain tumor response. Proc Natl Acad Sci USA 2005;102:5524–9. 4. Hamstra DA, Galban CJ, Meyer CR, et al. The functional diffusion map (fDM): an earl y imaging biomark er for o verall sur vival in high-grade glioma. J Clin Oncol 2008. [in press] 5. FDA. Infor mation on gadolinium-containing contrast agents. 2008. (Center for Dr ug Ev aluation and Research (http://www .fda.gov/ cder/drug/infopage/gcca/default.htm)). 6. Chui H, Win L, Schultz R, et al. A unified non-rigid feature registration method for brain mapping. Med Image Anal 2003;7:113–30. 7. Clatz O, Delingette H, Talos IF, et al. Robust nonrigid re gistration to capture brain shift from intraoperati ve MRI. IEEE Trans Med Imaging 2005;24:1417–27. 8. Thirion J.-P. Non-rigid matching using demons. In: Proceedings of International Conference on Computer Vision and Pattern Recoginition (CVPR’96). San F rancisco, CA: IEEE Computer Society; 1996;517081:245–51. 9. Thompson P, Toga AW. A surface-based technique for w arping three dimensional images of the brain. IEEE Trans Med Imaging 1996;15:402–17. 10. Wu G, Qi F, Shen D. Learning-based deformable registration of MR brain images. IEEE Trans Med Imaging 2006;25:1145–57. 11. Arsigny V, Pennec X, Ayache N. Polyrigid and polyaffine transformations: a ne w class of dif feomorphisms for locall y rigid or af fine registration. In: Proceedings of Medical Image Computing and Computer-Assisted Inter vention—MICCAI 2003. Springer Berlin/Heidelberg: Lecture Notes in Computer Science; 2003; 2879:829–37. 12. Ashburner J, Friston K. Multimodal image core gistration and par titioning—a unified framework. NeuroImage 1997;6:209–17. 13. Bro-Nielsen M, Gramk ow C. F ast fluid re gistration of medical images. In: Proceedings of VBC ‘96. Hambur g, Ger many, Springer-Verlag: Lecture Notes in Computer Science; 1996;1131: 267–76. 14. Christensen GE, Rabbitt RD, Miller MI. Deformable templates using large defor mation kinetics. IEEE Trans Image Process 1996;5: 1435–47. 15. Collignon A, Vandermeulen D , Suetens P , Marchal G. 3D multimodality medical image registration using feature space clustering. In: Proceedings of CVRMed. Nice, FR, Springer -Verlag: Lecture Notes in Computer Science; 1995;905:195–204. 16. Collins DL, P eters TM, Ev ans AC. An automated 3D nonlinear image defor mation procedure for deter mination of g ross morphometric v ariability in human brain. In: Proceedings of

22.

23.

24.

25.

26.

27.

28. 29. 30.

31. 32.

33. 34. 35.

36.

37.

275

Visualization in Biomedical Computing. Rochester , MN: SPIE; 1994;2359:180–90. D’Agostino E, Maes F , Vandermeulen D , Suetens P . A viscous fluid model for multimodal non-rigid image re gistration using mutual information. Med Image Anal 2003;7:565–75. Hawkes DJ, Hill DLG, Bracey ECML. Multimodal data fusion to combine anatomical and ph ysiological infor mation in the head and heart. In: Cardiovascular nuclear medicine and MRI: quantitation and clinical applications. Norwell, MA: Kluwer; 1992. Hill DLG, Batchelor PG, Holden M, Hawkes DJ. Medical image registration. Phys Med Biol 2001;46:R1–45. Joshi S, Da vis B, Jomier M, Gerig G. Unbiased dif feomorphic atlas construction for computational anatom y. NeuroImage 2004;23: S151–60. Maes F, Vandermeulen D, Suetens P. Medical image re gistration using mutual information. Proc IEEE 2003;91:1699–722. Meyer CR, Boes JL, Kim B , et al. Demonstration of accurac y and clinical v ersatility of mutual infor mation for automatic multimodality image fusion using af fine and thin-plate spline w arped geometric deformations. Med Image Anal 1997;1:195–206. Pluim J, Maintz J , Viergever M. Mutual-infor mation-based registration of medical images: a sur vey. IEEE Trans Med Imaging 2003;22:986–1004. Rueckert D, Aljabar P, Heckemann R, et al. Dif feomorphic registration using B-splines. In: Proceedings of MICCAI. Springer -Verlag: Lecture Notes in Computer Science; 2006;4191:702–9. Studholme C, Derapaca C, Iordano va B, Cardenas V. Defor mationbased mapping of v olume change from serial brain MRI in the presence of local tissue contrast change. IEEE Trans Med Imaging 2006;25:626–639. Viola P, Wells WM. Alignment by maximization of mutual information. In: Proceedings of 5th Int’l. Conf. on Computer Vision. MIT, IEEE Press; 1995;95CH35744:16–23. Wells W, Viola P, Kikinis R. Multimodal volume registration by maximization of mutual infor mation. Proceedings of Medical Robotics and Computer Assisted Sur gery. Philadelphia: Wiley; 1995; 2:55–62. Woods R. 2006. Automated image re gistration (AIR). http://bishopw.loni.ucla.edu/air5/. Kullback S, Leib ler RA. On infor mation and suf ficiency. Ann Math Stat 1951;22:79–86. Soman S, Chung A, Grimson E, Wells W III. Rigid re gistration of echoplanar and con ventional MRI b y minimizing K ullbackLeibler distance. In: Proceedings of Workshop in Biomedical Image Registration (WBIR). Springer-Verlag: Lecture Notes in Computer Science; 2003;2717:181–90. Neemuchwala H, Hero AO, eds. Entropic g raphs for re gistration. Boca Raton: Taylor and Francis Group; 2005. Neemuchwala H, Hero AO, Carson PL, Me yer CR. Local feature matching using entropic g raphs. In: Proceedings of Intl. Symp. Biomed. Imag. (ISBI). Arlington, Va: IEEE; 2004;704–7. Izenman AJ. Recent de velopments in nonparametric density estimation. J Am Stat Assoc 1991;86:205–24. Parzen E. On estimation of a probability density function and mode. Ann Math Stat 1962;33:1065–76. Rueckert D, Hayes C, Studholme C, et al. Non-rigid re gistration of breast MR images using mutual infor mation. In: Proceedings of Medical Image Computing and Computer -Assisted Inter vention. Springer Berlin/Heidelberg; 1998;1496:1144–52. Rueckert D, Sonoda LI, Ha yes C, et al. Nonrigid re gistration using free-form defor mations: application to breast MR images. IEEE Trans Med Imaging 1999;18:712–21. Rohde G, Aldroubi A, Dawant B. The adaptive bases algorithm for intensity-based nonrigid image re gistration. IEEE Trans Med Imaging 2003;22:1470–9.

276

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

38. Bookstein FL. Principal w arps: thin-plate splines and the decomposition of defor mations. IEEE Trans P attern Anal Mach Intell 1989;11:567–85. 39. Nelder JA, Mead R. A simplex method for function minimization. J Comput 1965;7:308–13. 40. Powell MJD. An efficient method for finding the minimum of a function of se veral variables without using deri vatives. Br Comput J 1964;7:155–62. 41. Press WH, Flanner y BP, Teukolsky SA, Vetterling WT. Numerical recipes in C: the art of scientific computing. In: Cambridge: Cambridge University Press; 1988. p. 305–137. 42. Maes F, Vandermeulen D, Suetens P. Comparative evaluation of multiresolution optimization strategies for multimodality image registration by maximization of mutual infor mation. Med Image Anal 1999;3:373–86. 43. Spall JC. Stochastic search and optimization. Wiley; 2003. 44. Christensen G, Johnson H. Consistent image registration. IEEE Trans Med Imaging 2001;20:568–82. 45. Arsigny V, Commo wick O , P ennec X, Ayache N . A lo g-euclidean framework for statistics on dif feomorphisms. In: Proceedings of Medical Image Computing and Computer -Assisted Inter vention, MICCAI 2006. Springer Berlin/Heidelber g: Lecture Notes in Computer Science; 2006;4190:924–31. 46. Avants BB, Epstein CL, Grossman M, Gee JC. Symmetric dif feomorphic image registration with cross-correlation: evaluating automated labeling of elderl y and neurode generative brain. Med Image Anal 2008;12:26–41. 47. Camion V, Younes L. Geodesic inter polating splines. In: Proceedings of Energy Minimization Methods in Computer Vision and Pattern Recognition. Lecture Notes in Computer Science; 2001; 2134:513–27. 48. Marsland S, Twining CJ. Constructing diffeomorphic representations for the g roupwise anal ysis of nonrigid re gistrations of medical images. IEEE Trans Med Imaging 2004;23:1006–20. 49. Miller MI, Younes L. Group actions, homeomor phisms, and matching: a general framework. Int J Comput Vis 2001;41:61–84. 50. Narayanan R, Fessler JA, Park H, Meyer CR. Diffeomorphic nonlinear transformations: a local parametric approach for image re gistration. In : Proceedings of Infor mation Processing in Medical Imaging (IPMI). Lecture Notes in Computer Science; 2005; 3565:174–85. 51. Pennec X. Intrinsic statistics on Riemannian manifolds: basic tools for geometric measurements. J Math Imaging Vis 2006;25: 127–54. 52. Vercauteren T, P ennec X, P erchant A, Ayache N . Nonparametric diffeomorphic image re gistration with the Demons algorithm. In: Proceedings of MICCAI. Springer-Verlag, Lecture Notes in Computer Science; 2007;4792:319–26. 53. Arsigny V, Pennec X, Ayache N. Polyrigid and polyaffine transformations: a no vel geometrical tool to deal with non-rigid deformations—applications to the re gistration of histolo gical slides. Med Image Anal 2005;9:507–23.

54. Joshi SC, Miller MI. Landmark matching via lar ge defor mation diffeomorphisms. IEEE Trans Image Process 2000;9:1357–70. 55. Akaike H. Infor mation theor y and an e xtension of the maximum likelihood principle. In: Proceedings of Second Inter national Symposium on Information Theory. Budapest; 1973;267–81. 56. Schwarz G. Estimating the dimension of a model. Ann Stat 1978; 6:461–4. 57. Breen M, Lancaster T, Lazebnik R, et al. Three-dimensional method for comparing in vi vo inter ventional MR images of ther mally ablated tissue with tissue response. J Magn Reson Imaging 2003;18:90–102. 58. Dauguet J, Delzescaux T, Condé F , et al. Three-dimensional reconstruction of stained histological slices and 3D non-linear registration with in vi vo MRI for w hole baboon brain. J Neurosci Methods 2007;164:191–204. 59. Lazebnik R, Lancaster T, Breen M, et al. Volume registration using needle paths and point landmarks for e valuation of interventional MRI treatments. IEEE Trans Med Imaging 2003;22:653–60. 60. Lee Z, Sodee DB, Resnick M, MacLennan GT. Multimodal and threedimensional imaging of prostate cancer . Comput Med Imaging Graph 2005;29:477–86. 61. Rouvie`re O, Reynolds C, Hulshizer T, et al. MR histological correlation: a method for cutting specimens along the imaging plane in animal or e x vi vo e xperiments. J Magn Reson Imaging 2006;23:60–9. 62. Wilson D, Breen M, Lazebnik R, et al. Radiofrequenc y thermal ablation: 3D MR-histology correlation for localization of cell death in MR lesion images. In: Proceedings of Inter nat Symp Biomed Imaging. Arlington, VA: IEEE; 2004;1537–40. 63. Zarow C, Kim T-S, Singh M, Chui H. A standardized method for brain-cutting suitable for both stereology and MRI-brain coregistration. J Neuorsci Methods 2004;139:209–15. 64. Zhan Y, Ou Y, F eldman M, et al. Re gistering histolo gic and MR images of prostate for image-based cancer detection. Acad Radiol 2007;14:1367–81. 65. Zimmer C, Weissleder R, O’Connor D, et al. Cerebral iron oxide distribution: in vi vo mapping with MR imaging. Radiolo gy 1995;196:521–7. 66. Kim B, Boes JL, F rey KA, Meyer CR. Mutual infor mation for automated unw arping of rat brain autoradio graphs. Neuroimage 1997;5:31–40. 67. Park H, Pier t M, Kahn A, et al. Re gistration methodology for histological sections and in vi vo imaging of human prostate. Acad Radiol 2008. [in press] 68. Meyer C, Moffat B, Kuszpit K, et al. A methodology for registration of a histological slide and in vivo MRI volume based on optimizing mutual information. Mol Imaging 2006;5:16–23. 69. Samani A, Zubo vits J , Ple wes D . Elastic moduli of nor mal and pathological human breast tissues: an in version-techniquebased in vestigation of 169 samples. Ph ys Med Biol 2007;52: 1565–76.

19 CHEMISTRY OF MOLECULAR IMAGING: AN OVERVIEW SILVIO AIME, PHD, GIOVANNI BATTISTA GIOVENZANA, PHD, AND ENZO TERRENO, PHD

The last tw o decades of the tw entieth century witnessed enormous advances in our understanding of the pro gression of diseases at the molecular le vel thanks to the outstanding achievements in molecular and cellular biolo gy. The new information that has been gained on the f actors determining the cause and progression of the most important diseases has led to the de velopment of inno vative pharmaceutical treatments. Meanwhile, a number of in vitro techniques are no w available to biologists for assessing, at the molecular level, the occurrence of abnormal gene expression that accompanies the de velopment of a patholo gical state. In this scenario, the new f ield of molecular imaging has emer ged to seek for the in vivo visualization of molecular e vents occurring at the cellular le vel. The de velopment of approaches that visualize molecules, which are the “signature” of a gi ven disease, represents an outstanding breakthrough in the diagnostic modalities cur rently pro vided clinically. Therefore, molecular imaging (MI) is the noninvasive assessment, characterization, and quantif ication of gene and protein function, protein–protein interaction, and profiling of signal transduction pathways in animal models of human disease and in patients to gain fur ther insight into the molecular pathology of a specif ic disease.1 At present, in vivo diagnostic systems basically assess the structure and function of human organs. Therefore, for important diseases such as cancer and cardio vascular pathologies and also diseases of the central ner vous system, only the late symptoms are detected. It is expected that the adv ances in genomics and proteomics will ha ve a tremendous impact on human health care of the future. However, adv ances in molecular biolo gy are already redefining diseases in ter ms of molecular abnor malities. With this kno wledge, ne w generations of diagnostic

imaging agents can be defined that aim at the detection of these molecular processes in vivo. Any molecular imaging procedure requires an imaging probe that is specif ic for a gi ven molecular e vent. Therefore, chemistry plays a vital role in the development of this re volutionary methodology. In f act, by exploiting the outstanding possibilities offered by modern synthetic organic and coordination chemistry and the efficient procedures provided by conjugation chemistry, it is possible to tackle the most challenging prob lems to give molecular imaging the opportunity to express all its potential. The goal is, therefore, to fur ther the development of innovative imaging probes through the pursuit of innovations in a number of dif ferent areas, ranging from the design of imaging units endowed with enhanced sensitivity to the control of the structural and electronic determinants responsib le for the molecular reco gnition of the target molecule. Currently, the diagnostic imaging mark et is dominated b y state-of-the-ar t imaging modalities, such as X-ray, computed tomo graphy (CT), magnetic resonance imaging (MRI), ultrasound , and emission imaging (positron emission tomography, PET; single photon emission computed tomography, SPECT). Often, state-of-theart contrast agents and imaging probes, apar t from radiopharmaceuticals used for PET and SPECT, are nonspecific as the y enhance onl y the str ucture and macroscopic functions. Although some results on such target-specific agents ha ve been repor ted, still much has to be done to de velop systems for clinical molecular imaging. The basis for designing imaging probes for a gi ven application is dictated b y the chosen imaging modality , which in tur n is dependent upon the concentration and localization prof ile (v ascular, e xtracellular matrix, cell 277

278

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

membrane, intracellular, near or at the cell nucleus) of the target molecule. The de velopment of high-af finity ligands and their conjugation to the targeting vector is one of the k ey steps for pursuing ef ficient molecular imaging probes. As targets there are a number of possibilities: • • • • •

Growth factors (eg, VEGF, F6F, integrins) Membrane receptors stimulated by growth factors Intracellular targets (enzymes, steroid receptors) Transporters of nutrients and pseudo-nutrients Marker associated with changes of the e xtracellular matrix (eg, matrix metalloproteases) • Marker associated with the malign for mation of the cell membrane matrix (eg, prolin, cholin) • Marker of apoptosis • Marker of vulnerab le atherosclerosis plaques (eg, integrins, LDL) After the identif ication of lead tar get candidates, imaging modality-specific probes have to be optimized to be sensitive repor ters of the in vivo local distribution and quantitati ve e xpression of their related molecular targets. Several biological and chemical f actors will ha ve to be impro ved to positi vely influence the ef fect of the agent. They include: • Intravascular half-life (should be suf ficiently long to allow the agents to accumulate at the tar get site), b y controlling the molecular size, lipophilicity , char ge, and other pharmacological parameters • Use of multivalency (to increase binding affinity) • Depending on the imaging modality , the capacity to carry a high pa yload of probe molecules and the use of contrast materials with strong contrast effects • The ability to accept dif ferent ligands and imaging probes • Aspects of biodistribution, excretion, and toxicity

technique, and one that is also safe and flexible. However, both f avorable biodistribution and radionuclide characteristics are desirab le with receptor imaging. Radionuclear probes (for PET and SPECT) displa y an e xcellent sensitivity (nanomolar scale), but the attainab le resolution of the technique is rather poor (although an important development in “preclinical” nuclear medicine is the use of pinhole SPECT , w hich displays an almost unlimited resolution). In PET studies, the most straightforward route to design an ef ficient imaging repor ter is to directl y label small endogenous molecules with F-18 or C-11 isotopes, such as in the case of [ 18F]fluoro-16 α-fluoroestradiol ( 18F-FES, estrogen receptor ligand) 2 or in [ 18F]fluoroethyl-L-tyrosine (18F-FET) and [11C]-methyl-L-methionine (11C-MET, amino acid transporter tracers), respectively. Although most of the w orks are devoted to the use of light elements such as C-11, O-15, and F-18, metalpositron emitters such as Zr -89, Cu-64 (produced in lar ge amounts and high-specific activity with “small biomedical cyclotrons”), and Ga-68 (new Ge-68/Ga-68 generators are available) are v ery promising systems. One question is whether an analo gue of [ 18F]fluoro-deoxyglucose (FDG) based on Ga-68 can be de veloped. It w ould be a breakthrough for PET imaging to de velop a “generator produced” positron emitter, very efficiently, at any PET center that does not ha ve its o wn c yclotron. The a vailability of “generator produced” Ga-68 at lo w cost will stimulate the development of chelator-coupled small molecules such as peptides for dif ferent specif icities. This may open a ne w generation of kit-for mulated PET radiophar maceuticals similar to the ones in daily nuclear medicine practice based on the Mo-99/Tc-99m generator. Along with the long halflife of the generator , which can be used for more than a year, Ga-68–based radiophar maceuticals ma y become a very cost-ef fective alter native to c yclotron-based tracers. If the rather short physical half-life of Ga-68 of 68 minutes is not adequate to the molecular imaging of longer lasting biological processes, a similar PET radionuclide generator based on Ti-44/Sc-44 (T 1/2 = 4.8 hours) might be developed.

IMAGING PROBES FOR NUCLEAR MEDICINE

MRI PROBES

Although nuclear imaging agents can be detected at picomolar concentration le vel, often the specif icity of these probes is not satisf actory. It is a challenge to de velop probes that ha ve a high af finity to bind the molecular “signature” of a gi ven disease. New chemistry is needed to ensure optimal coupling betw een ligand and mark er molecules. Receptor imaging is an e xtremely po werful

MRI contrast agents mak e use of local changes in the relaxation times of tissue water. The limited sensitivity of MRI requires amplif ication, that is, the use of high relaxivity centers and many centers per target. Efforts still have to be devoted to seeking real breakthroughs in the search of high relaxivity systems. 3 Although the optimization of the T 1 shortening ef fect in ne w Gd(III) chelates is an

Chemistry of Molecular Imaging: An Overview

important goal, impro ved preparations of ultra small particles of iron o xide (USPIO) ha ve been e xtensively used in MR-molecular imaging investigations.4 Recently, a new class of MRI contrast agents has been proposed, the so-called chemical exchange saturation transfer (CEST) agents.5 These systems contain at least one pool of e xchangeable protons, w hich upon ir radiating at their absorption frequenc y, transfer saturated magnetization to the bulky water signal. The use of lanthanide or transition metal paramagnetic chelates has been shown to be particularly benef icial because the paramagnetic ion induces a large shift of resonance on the nuclei sur rounding it that in turn allows the exploitation of larger exchange rates which, in turn, result in larger magnetization transfer effects. More work appears necessar y to endo w paramagnetic CEST agents with the sensiti vity required b y MRI e xperiments. Such agents might be par ticularly useful also in cellular labeling procedures.

OPTICAL IMAGING PROBES Optical imaging probes possess high sensiti vity but the technique has the drawback of limited penetration of light through biological tissues. Improvements can be foreseen with the use of time-resolv ed fluorescence-based approaches. From the vie wpoint of the imaging repor ter design, chemistry has had the key role in the rapid development of optical imaging procedures as far as concerns: (1) the identification of fluorophores with optimal separation between excitation and emission w avelength and impro ved resistance to autob leaching phenomenon, (2) the procedures leading to “quenching” and “de-quenching” of fluorescence upon the control of the str uctural characteristics, (3) the implementation of the use of quantum dots as tools for enhancing sensitivity and specificity in optical imaging detection.

TARGETED VERSUS NOT-TARGETED PROBES The biodistribution of an imaging probe is the result of a complex interplay among a number of parameters including hydrophilic/hydrophobic characteristics, size, char ge and its distribution, the nature of residues on the outer surface, etc. Small-sized, highl y hydrophilic systems distribute in vessels and in the e xtravascular space, thus acting as reporters of or gan perfusion. Conversely, small-sized systems bearing hydrophobic/lipophilic groups may passively diffuse through membranes and accumulate in lipid-rich

279

regions. Fur thermore, imaging repor ters that beha ve as organic anions or or ganic cations ma y ef ficiently enter cells bearing suitable transporters6 known as organic anion transporter proteins (O ATP) or or ganic cation transpor ter proteins (OCTP), respecti vely. In general, the route to endow an imaging reporter with a targeting capability is to conjugate a proper v ector on its outer surf ace. The conjugation needs to be carried out in mild conditions to prevent any change to the indi vidual moieties. As vectors a number of systems ha ve been considered including or ganic synthons, peptides, antibodies and their fragments, etc. Herein, the basics of bioconjugation techniques are summarized.

Bioconjugation Routes The conjugation of a molecular imaging probe to a specific vector relies on the for mation of a linkage betw een the two components (Figure 1). Some requisites are mandatory for this connection: • The linkage must be stab le under the conditions used for the molecular imaging procedure; these conditions are generally mild as the analysis is usually performed in li ving or ganisms. Ne vertheless, it should resist to hydrolytic, o xidizing, and reducing conditions of a nearly neutral aqueous medium containing se veral anions, metal cations, small molecules, and enzymes. Even if not indef initely stable, the linkage should be stable at least for the w hole duration of the molecular imaging procedure. • The conjugation step should be quantitative and selective to reduce or even avoid a tedious purif ication step for the isolation of the conjugate. Atom-economic reactions,7,8 with few or no byproducts, are particularly useful for this pur pose. • The reaction should be f ast and easy to perfor m. • The linkage should be perfor med in mild conditions, possibly those used in the molecular imaging procedure, to avoid vector and/or probe degradation.

Figure 1. Conjugation between a generic imaging probe and a given biological target.

280

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

The strate gies for conjugation of the MI probe to the desired vector rely on the presence of both components of reactive functional g roups. It is possib le to indi viduate four main strategies followed in the conjugation depending on the nature of the abo ve-mentioned pair of functional g roups. The reacti ve g roups present on v ectors vary according to the nature of the v ector itself. Vectors are ob viously dependent on the specif ic molecular imaging application; typical v ectors used are peptides, proteins, antibodies, (oligo)nucleic acids, carboh ydrates, lipids, and small- or medium-sized or ganic molecules. Peptides and proteins ha ve an amino acidic backbone in w hich se veral dif ferent g roups are helpful for conjugation: • The N-terminal amine • The C-terminal carboxylate • A range of side chain moieties including primar y amines (Lys, Orn), carboxylic/carboxylate (Asp, Glu), hydroxyl (Ser , Thr, Hyp), thiol (Cys), phenol (T yr), heterocyclic (His, Trp) residues. Carbohydrates are more dif ficult to conjugate as their structure is comple x, and the a vailable functional g roups are restricted to (emi)acetalized aldeh ydes or ketones, seldom amines and carbo xylates and an ubiquitous bunch of hydroxyl groups. The latter react only in strong conditions and their number and similarity reduce the possibility of selective conjugations. Oligo- and polynucleotides may be connected to MI probes through the hydroxyl groups of the (deoxy)ribose moiety or through activated positions of the nitrogen base or even through the phosphate g roups. Other small- or medium-sized molecules used as vectors are connected to the MI probe taking adv antage of the functional g roups (avoiding those fundamental for their activity) of the nati ve molecule or of ad hoc functionalized derivatives. Even if the chemistr y in volved in the conjugation is quite simple, a judicious choice of the cor rect protocol is O C

needed to optimize the conjugation step. Here, we illustrate a brief summary of the main protocols for conjugation of a MI probe to a generic vector.

Direct Coupling

This strategy (Scheme 1) is based on the for mation of a direct chemical bond betw een the v ector and the MI probe. In this approach, the functional g roups present on the vector and on the MI probe ma y be complementar y, thus allowing a direct reaction; this is possib le whenever a proper functionalized MI probe is a vailable; unfor tunately, few of these compounds are mark eted. When a vailable MI probes are not prone to direct reaction, one of the functional groups should be activated by a suitab le combination of reagents to react promptl y with the other one. This possibility e xpands the ar ray of useful functional g roups amenab le to conjugation and , accordingly, the number of the a vailable MI probes. Carboxylic groups are frequently available as conjugation handles on MI probes. They can be used to link the MI probe to the desired vector provided that the latter contains

+

−

Figure 2. Chemicals typically used for activating carboxylic groups to the reaction with amines.

Activating agent

OH

O C X

O C X

“Activated form”

NH2

O C N H

Scheme 2.

Conjugated imaging probe

Scheme 1.

OH

O C O

Functionalized imaging probe

Vector

Chemistry of Molecular Imaging: An Overview

amines or hydroxyl groups; the link occurs through the formation of an amide or an ester moiety , respectively. Both derivatives ma y be clea ved b y competing h ydrolytic processes, possibly activated by enzymes such as proteases and lipases. Amides are usually more stable than esters. Carboxylic acids do not react as such with amino or hydroxyl groups; they must be converted in an “activated form” by means of an activating agent (Scheme 2). The acti vating agent of choice for perfor ming this conjugation in aqueous media is usually the water-soluble EDC9 (Figure 2); inclusion of NHS 10 or sulfo-NHS 11 in the reaction mixture leads to faster and cleaner formation of the amides, and with these additi ves, the acti vated form of the acid may be isolated and stored. Examples of available MI probes with free carboxylic acids are repor ted in F igure 3. Macroc yclic DO TAtriesters12,13 (R = t-butyl, benzyl, methyl) are bound to the desired v ector, and then ester g roups are remo ved. The conjugated ligand ma y be coordinated with the proper metal ion (Gd 3+ or other lanthanides for MRI, 64Cu2+ for PET, 111In3+ for SPECT), depending on the MI application and rel ying on the lar ge stability constant of the complexes formed by this macrocyclic ligand. The acyclic ligand dieth yleneetriaminepentaacetic acid , sold in its activated anhydride form DTPAA,14–16 has been used for the preparation of se veral conjugates b y direct reaction with the v ectors. Fluorescent probes holding remote free carboxylic groups are available; an example is the cyanine shown in Figure 3. Another pi votal g roup for MI probe conjugation is the amino group, usually a primary one. The nucleophilic properties of this nitrogen function may be exploited in a number of reactions. The amine reacts with activated carboxylic and phosphonic acids, isothioc yanate and epo xide-tailored vectors, as summarized in Scheme 3.

281

Examples of MI probes equipped with primar y amino groups are the ligands DTP A-Lys17 and the commerciall y available cyanine dye NIR-5c (Figure 4). The above-mentioned isothiocyanate group represents a useful acti vated for m of the amine. Isothioc yanates are reasonably air and w ater tolerant, require no additional reagent to react with an amino-tailored molecule and do not yield b yproducts. Valuable isothioc yanate deri vatives of selected MI probes, e xemplified b y the archetypal FITC18–20 or by the ligand ITC-DTPA21–24 and its isomers (Figure 5), are commercially available and used to tag vectors in buffered aqueous solution in countless applications. Finally, free thiol g roups (-SH), frequently present in peptides and proteins, ma y be tar geted by suitably functionalized MI probes, provided that the latter have reactive and selecti ve moieties, such as maleimides. Maleimidofunctionalized probes react cleanly with free thiols giving the corresponding Michael adducts (Scheme 4). Examples of this kind of probes are the maleimidoligand MDTPA,25 the commerciall y a vailable fluorescent probes fluorescein diacetate 5-maleimide and Atto-610 maleimide, and the PET agent 18F-benzamido ethylmaleimide (Figure 6). These e xamples represent a brief o verview of the commonest functionalized MI probes. Other functional groups may be included in MI probes to help conjugation with specif ic v ectors, and the y will be described in the chapters dealing with specif ic applications. O

OH

H N

N

O

O

H

X

NH2 O NCS

H N

H N O

O P OH Activating agent

OH H OH O N P O

Scheme 3. −

+

Figure 3. Some available ligands for preparing imaging probes endowed with free carboxylic acid.

Figure 4. groups.

Some examples of MI probes containing primary amino

282

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 5.

Commercially available isothiocyanate derivatives of selected MI probes.

O

O S

SH N O

N O

Scheme 4.

Figure 6.

Examples of maleimido-based ligands to be used for conjugating free thiol groups.

Vector Functionalization

Although recent advances in the f ield led to a proliferation of functionalized molecules, the number and the v ariety of the commercially available tailored MI probes are still f ar away from optimal. Once the v ector is identif ied, the chemist should select the imaging repor ter depending on the technique in volved and on the site of conjugation offered by the vector itself. Nevertheless, the length of the synthesis of most MI probes limits the access to a wide array of functionalized deri vatives. As a consequence, the majority of functionalized MI probes on the mark et typically endow one of the three functional g roups (carboxyl, amino, and maleimido) repor ted in the preceding section. Clearly, the limit of this approach is represented b y vectors lacking the specif ic functionalities to bind the a vailable MI probe. Of course, these v ectors may be conjugated as

long as their a vailable functional g roups are modified, turning them into a ne w chemical moiety. Modification is achieved by reaction with a heterobifunctional reagent, acting as a linker and at the same time as a spacer between the vector and the probe. Conjugation is achie ved in a second step in w hich commercial functionalized MI probes react with the new “complementarized” vector. We report a brief survey of classical modif ications of functional g roups present on vectors; the reaction conditions are mild and compatible with biomolecules (Scheme 5). Vectors presenting free amines (primary or secondary) are easily modif ied. As reported in Scheme 6, amines are reacted with a dicarbo xylic anhydride obtaining a product showing a free carboxylic group. Succinic26,27 and glutaric anhydride28 are the commonest, being cheap and satisf actorily reactive, although other anhydrides may be useful for conjugation purposes.

Chemistry of Molecular Imaging: An Overview

Amines ma y be used to introduce thiol g roups, b y reaction with suitab le reagents. Thiobutyrolactone and Traut’s reagent 29,30 acylate the amine leaving at the other end of a 4 carbon chain a free thiol g roup (Scheme 6). Carboxylic acids are common functional g roups in peptides (either as C-terminal or deriving from aspartic or glutamic residues), but even in carbohydrate and in small nonbiogenic v ectors. Carbo xylic acids are modif ied b y formation of an amide with a proper amine (Scheme 7). Amidation with excess of a α, ω-diamine leads to the formation of a deri vative endowing a free primar y amine. Amidation is perfor med by means of a condensing agent, usually the w ater-soluble EDC or acting on pre-acti vated forms of carbo xylic acids. 31,32 α, ω-Diamines endo wed with different chain lengths are available, thus allowing the choice of the spacing betw een the original and the ne w functional g roups. If the diamine is c ystamine, the obtained amide ma y be clea ved at the disulf ide bond

Scheme 5.

X Scheme 6.

Scheme 7.

283

generating a free thiol g roup; this protocol translates a carboxylic into a thiol group.33,34 Modification of thiol and hydroxyl groups is possible, although less common. Thiols are derivatized taking adv antage of the nucleophilicity of the corresponding anion, for med in basic solutions; treatment of thiols in these conditions with iodoacetic acid leads to displacement of the halo gen by the sulfur and to the introduction of a carboxylic group (Scheme 8). 35 A related alkylation may be carried out on a hydroxyl group with chloroacetic acid , but for this nucleophilic substitution to occur, strongly, basic conditions are needed to deprotonate significantly the hydroxyl group. This reaction is actually limited to carbohydrate modification. Obviously, con version of an a vailable g roup into a different one with better conjugation proper ties is not limited to these e xamples; selecti ve and high yield functional g roup interconversions and ne w and ef ficient reagents are continuously added to the existing array.

284

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Coupling Mediated by a Homobifunctional Linker

The third conjugation strate gy is applied w henever the selected vector and the available tailored MI probe share the same type of functional group. Direct conjugation is impossible and modification of one of the functional g roups may not be con venient as both par tners already ha ve suitab le reactive groups. In this case, homobifunctional linkers may be used. Homobifunctional linkers own two identical reactive groups, linked and spaced by a suitable bivalent moiety. As sho wn in Scheme 9, a homobifunctional link er with properly chosen functional groups (complementary to those shown b y the v ector and the MI probe) reacts with both components linking them. The conjugation reaction is seldom made in a single step, to avoid homodimers formation. Usually, one of the components is reacted f irst with the homobifunctional linker and then, after an e ventual purification step, the second component is added to complete the conjugation protocol (Scheme 10). The most common application of this strate gy applies to MI probes and v ectors embodying amines, conjugated b y se veral a vailable homobifunctional linkers (Scheme 11). The list includes: (1) bis-acylating bis-N-hydroxysuccinimido esters of alkanedicarbo xylic acids (bis-NHS-esters) or (2) the more w ater-soluble sulfonic counter part (sulfo-bis-NHS-esters), 36 both available in variable chain length offering a customized spacing; (3) diethyl squarate, a small and rigid link er;

(4) bis-imidates, 37 leading to conjugates holding positively char ged amidine g roups, and (5) ar yl-based diisothiocyanates,38 leading to strong thiourea linkages.

Coupling Mediated by a Heterobifunctional Linker

The last strate gy is designed for MI probes and v ectors tailored with dif ferent noncomplementar y functional groups, that is, when, as in the preceding strategy, the direct coupling is impossib le. Functional g roup modif ication is possible but often inconvenient: modification and conjugation require two different steps while a less time-consuming one pot conjugation is clearl y preferable. The one-pot conjugation ma y be perfor med using heterobifunctional reagents. Such chemicals ha ve tw o dif ferent and noncomplementary reactive functional g roups, connected b y a spacing substr ucture. As sho wn in Scheme 12, the dif ferent functional g roups and their dissimilar reacti vities allow a self-sorting reaction in w hich the three components are directly coupled in an ordinate arrangement; no alternative coupling is possib le, and this reduces strongl y byproduct formation, allowing easier isolation of the target conjugate. Several combinations of functionalities ma y be envisaged for heterobifunctional link ers, but the number and variety of commerciall y a vailable ones is quite limited. Their applications are exemplified in Scheme 13; typically,

Scheme 8.

Scheme 9.

1st step Scheme 10.

2nd step

Chemistry of Molecular Imaging: An Overview

NH2

285

NH2

O O

O

O

O N (CH2)n

N O O

OH N O

O O

O

O (CH2)n

Bis-NHS Alkanedioates

NH

HN

O HO3S

O

O (CH2)n

N O

O O

SO3H

SO3H

Bis-Sulfo-NHS Alkanedioates

O

O

OCH2CH3 Diethyl squarate

(CH2)n

CH3CH2OH

O

N H

N H

H2N

O CH3

H3CO

NH.HCI Bis-Imidates (Dihydrochlorides)

S C N Aryl

O

OCH2CH3

O

HCI.HN

O

O N

OH N O

NH

CH3OH, Cl

N C S

Diisothiocyanates

S

(CH2)n

NH2 HN

S

NH Aryl N H HN NH

Scheme 11.

Scheme 12.

amino-tailored MI probes are conjugated with thiol groups of peptides/proteins using one of the accessib le heterobifunctional linkers. NHS ω-maleimidoalkanoates under go Michael addition of thiols to the strong acceptor site represented by the maleimide moiety; ac ylation of the amino groups on the second component is perfor med b y the activated NHS ester g roup leading to the required conjugate.39–41 Similar reacti vity is sho wn b y the simpler NHSiodoacetate39,42,43 in w hich the thiol-directed reaction

involves the strong nucleophilicity of the sulphur containing g roup in a S N2 substitution of the halo gen atom. Chemistry of conjugation is not restricted to the examples sho wn in these pages; se veral reactions and reagents are cur rently used to link biolo gical and synthetic molecules for countless applications. Nevertheless, the conjugation chemist’ s ar mory continuousl y needs new and improved tools for f aster and cleaner reactions, with few or no byproducts and selective linking at predetermined molecular sites.

286

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Scheme 13.

Peptide Synthesis The reason why peptides are under intense scrutiny is basically related to the fact that the highly specific, but biologically long-lived monoclonal antibodies (mAbs) in b lood when used in nuclear medicine applications, often result in images with lo w signal-to-noise ratio. The lar ge size of mAbs (e g, 145 kDa for IgG) slo ws their dif fusion into tissues, thus hindering their utility as imaging probes, because of the hepatic metabolism of mAbs. The imaging requirement for lo w nonspecif ic signal from backg round necessitates rapid clearance and e xcretion of the probe from the blood to urine via the kidneys. Peptides appear to provide good alter native solutions to the dra wbacks brought about by the macromolecular system. Thus, small antibody fragments are preferred over intact antibody, and, in turn, small peptides are the prefer red solution. Optimization of endo genous ligands is also another valuable strategy. Along this line, systems lik e Ga-68 or F-18 labeled octreotide derivative for imaging of neuroendocrine tumors ha ve been de veloped.44–47 Several other regulatory peptides whose receptors are o verexpressed in several tumors ha ve been considered (e g, bombesin, neurotensin). Cyclic Arg-Gly-Asp-containing (RGD) systems have been subject of intense study because of their high affinity towards the αvβ3 integrin.48,49 The peptide-containing reporters act as agonists with respect to the natural ligands and often inter nalize into cells by receptor-mediated endocytosis. Thus, they yield

to high tar get-to-background ratio thanks to their accumulation in endosomal vesicles. Recently, it has been shown that e xcellent targeting is achieved also with systems showing an antagonist behavior.50 The synthesis of peptides is carried out either in solution or by using dedicated resin at solid state. Traditionally, the chemical synthesis of peptides was carried out in solution, adding from time to time an amino acid to the pre vious, until the completion of the peptide chain. This approach could be used to synthesize only small peptides in acceptable yields. To extend the method to the practical synthesis of lar ge peptides, a process of b lock synthesis followed by fragment condensation w as developed. In this process, the desired peptide is ideally broken down into con venient sized b locks of 5 to 10 residues. These polypeptide blocks are synthesized by conventional methods and obtained in protected form. The final step of the process would be the selecti ve removal of the N- and C-terminal protecting groups followed by a coupling step to afford the desired peptide (Scheme 14). However, the application of this methodolo gy w as found rather dif ficult. Besides requiring a considerab le experience to be car ried out ef ficiently, it is e xtremely demanding both in ter ms of time and costs. Moreo ver, it shares with the traditional peptide synthesis approach the need for intermediate stages of isolation and purification. However, it retains usefulness in lar ge-scale production of peptides for industrial pur poses.

Chemistry of Molecular Imaging: An Overview

P1

P2

A1 A6

P1

Remove protecting group P2 and activate

A7 A12

287

P2

Remove protecting group P1

O P1

A1 A6

H2N

X

A7 A12

P2

Couple O P1

A1 A6

N H

A7 A12

P2

Scheme 14.

A remarkab le pro gress has been brought b y the synthetic methodology developed in the 1960s b y Bruce Merrifield, who introduced the use of a pol ystyrene support in the synthesis of peptides. 51 This technique w as called solid-phase peptide synthesis (SPPS). In this approach, the N-protected amino acid attached via its free carboxyl group to an insoluble polymeric material, originally polystyrene. To better anchor the amino acid to the resin, the latter could be acti vated b y using selected spacer groups called linkers. The linker was then used to anchor the f irst amino acid by covalent bond. The pur pose of the link er is to protect the C-terminal during the process of chain e xtension and to provide a cleavable linkage between the synthetic peptide chain and the solid suppor t. In this sense, most link ers release the peptide in the form of free C-terminal acids or amides upon treatment with acids (e g, trifluoroacetic acid); other kinds of link ers promote the clea vage with nucleophiles. The major inno vation introduced b y Mer rifield, thanks to the use of a solid support, was the possibility of using a lar ge excess of each amino acid during the coupling step. This allows high yields in the coupling reaction while a simple w ashing step is suf ficient to remove the excess of reagent. In this sense, the Merrifield technique really provides a solution to the prob lems suf fered from the traditional approach to peptides synthesis. In 1984, Mer rifield received the Nobel Prize in recognition of his contribution in the peptide synthesis f ield. The general principle of SPPS is the repetition of deprotection cycle and coupling: the amino g roup of the first residue anchored on the resin is deprotected and

reacts with the free carbo xyl g roup of a ne w amino acid N-α protected. The solid-phase synthesis is usuall y car ried out as follows (Scheme 15): 1. loading of the C-terminal amino acid to the resin 2. deprotection: remo val of the N-ter minal protecting group of the amino acid bound to the resin 3. activation of the next amino acid carboxyl group 4. coupling reaction followed by washing out of excess reagents 5. repetition of synthetic steps 2 to 4 or cleavage of full sequence peptide off the resin The tw o N-ter minal protecting g roups commonl y used for solid-phase synthesis are t-butyloxycarbonyl (Boc) and 9-fluorenylmethyloxycarbonyl (Fmoc, Figure 7). An important advance was made by the introduction of the Fmoc protecting g roup as it is remo vable under very mild conditions such as treatment with an or ganic base. For this reason, the Fmoc approach has pre vailed over the Boc approach, w hich requires the use of strong acids for its cleavage (Scheme 16). Amino acids with alcohols or carboxylic acids in the side chain (Asp, Glu, Ser , Thr, Tyr) can be protected either as benzyl/t-butyl ethers or as benzyl/ t-butyl esters. The amino g roup of L ys, guanidino g roup of Arg, thiol group of Cys, or imidazole of His may require other special protecting g roups such as Boc, Pbf (2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl), and Trt (triphenylmethyl). The choice of side chain– protecting agents will depend on the protecting g roup used for the N-ter minal, because it is essential that the

288

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Deprotection

Coupling Solid support

Linker Amino acid N-α-protection Side chain protections Cleavage

Scheme 15.

Figure 7.

N-terminal protecting groups commonly used for solid-phase synthesis.

Scheme 16.

Chemistry of Molecular Imaging: An Overview

side chains remain protected w hen N-ter minal amino acids are deprotected for chain elongation, to prevent side reactions. The basis of formation of the peptide bond, occurring during the coupling reactions, is to con vert the carboxylic acid function of one amino acid to promote the nucleophilic attack by the amino group of the second amino acid. The most impor tant activation methods of carbo xylic g roup in Fmoc SPPS in volve the use of acti vator agents such as dic yclohexylcarbodiimide (DCC) or benzotriazole he xafluorophosphate salts (HBTU, PyBOP, HATU, Figure 8). In the original method de veloped b y Mer rifield, chloromethyl g roups w ere introduced into the pol ystyrene resin, and the C-terminal amino acid was attached to for m an ester linkage. This method w as not ideal because the conditions required to attach the C-ter minal amino acid on the suppor t and to release the completed peptide w ere harsh. An impor tant pro gress has been brought b y the use of 4-alk oxybenzylalcohol link er on Wang resins. It is ideall y suited to Fmoc synthesis because the linkage to the resin is not af fected b y the

OH

N

Fmoc

O

2

H N

DIPEA

Fmoc

basic conditions necessary to release the N-terminal protecting group. A range of Wang resins are available commercially, each with a dif ferent Fmoc-protected amino acid already attached. Other impor tant resins for SPPS are chloro/ aminomethyl polystyrene, Tentagel, PEGA, and No vagel (Table 1). Important characteristics of the solid suppor t are the small par ticle size with lo w cross-linking, w hich allo ws rapid diffusion of reagents and the comparab le polarities between resin and peptide backbone. Apart from PEGA resin, all resins should be swollen before use: underivatized polystyrene resins s well only in dichloromethane, all the others are s wollen in DMF. Resins are fragile and should be handled carefully. There are some possible problems in SPPS like aggregation, when peptide chains for m secondary structures or aggregates either with other peptide chains or with the polymer support, racemization during coupling, and side reactions, rare if the amino acids are protected and the correct procedures are follo wed. Moreo ver, if the coupling reactions do not go to completion at an y step, the f inal

O

N

H N

O

N

R

R

2

P O 1 N

PF6

N N

N

PyBOP

2 Fmoc

H N

O

N N

R

O N N N

N

(2 OBt)

O N N NH2

P

N1

NH2

HN Fmoc

H N

O R

1

Figure 8.

N O5 P N N

289

1 (2 OBt)

Formation of the peptide bond in Fmoc SPPS involving the use of PyBOP as activator.

O O R Nu

:

290

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 1. BASIC CHARACTERISTICS OF THE RESINS COMMONLY USED FOR SSPS PROCEDURES Resin

Composition

Functional group

Chloro/aminomethyl polystyrene

1% divinylbenzene cross-linked polystyrene

CH2Cl, CH2NH2, benzhydrylamino

Tentagel

Polystyrene-polyethylene glycol graft polymer

CH2OH, CH2NH2

PEGA

Bis 2-acrylamidoprop-1-ylPEG,2-acrylamidoprop-1yl[2-aminoprop-1-yl]PEG dimethyacrylamide co-polymer

CH2NH2

Novagel

(4'-O-methylpolyethylene glycoxycarbonylamino-methyl) polystyrene

CH2NH2

product will contain deletion byproducts that may be quite difficult to remo ve. F or this reason, it is impor tant to ensure the completion of each coupling step (the coupling reaction can be repeated several times if necessary). Notwithstanding these prob lems, the solid-phase approach of fers g reat adv antages, such as (1) the possibility to remove excess reagents by simple washing, (2) high yields, (3) synthesis of long sequences, and (4) opportunity of full automation of the process. There are se veral automated peptide synthesizers available commerciall y, some of them equipped with a single-mode microwave reaction vessel (microwave peptide synthesizer, Figure 9). The synthesis protocol for the automated process is: • • • • •

swelling of the resin; 1st Fmoc amino acid attachment; deprotection Fmoc group with piperidine in DMF; washing with DMF; Fmoc amino acid coupling (acti vators/DIPEA/ DMF); • washing with DMF; • deprotection Fmoc, and so on; • After the last coupling, the resin has to be washed and dried in vacuo before cleavage. The clea vage from the resin in volves the use of cleavage cocktails; if Cys is absent, one ma y use TFA (trifluoroacetic acid) and TIS (triisopropylsilane) 95:5. During the clea vage, highl y reacti ve cationic species are generated that can react with amino acids containing electron-rich functional g roups such as Tyr, Trp, Met, and Cys. To quench these ions, nucleophilic reagents (water, TIS, etc) are added to TFA. With Trp being par t of the sequence, a fur ther step with a 5% aqueous solution of AcOH is required to obtain the fully deprotected peptide. The clea vage cocktail depends on the amino acids present in the peptide and on the resin used for the synthesis.

Figure 9. Example of an automated microwave solid-phase peptide synthesizer.

The final peptide is usually obtained by precipitation with cold diethyl ether and dried in v acuo. After that, it is necessar y to purify the f inal product from b yproducts (deletion peptides due to incomplete coupling or side products arising from modif ications of amino acids side chains during the deprotection process). The usual w ay to purify the peptide is b y preparati ve reverse-phase HPLC with a range of dif ferent mobile phases. The characterization of the f inal peptide is made by MS, especially by the MALDI-TOF technique. The gi ven peptide has then to be conjugated to the imaging repor ter to yield the designed probe. Most

Chemistry of Molecular Imaging: An Overview

conveniently, the coupling can be car ried out on the resin by using suitab ly functionalized imaging repor ters. F requently, a spacer is introduced between the peptide and the imaging repor ter to a void possib le interactions that ma y affect the recognition/binding capabilities of the peptide.

AGENTS AND PROCEDURES FOR CELLULAR LABELING For this application, the ideal imaging modality could be the one that pro vides a sensiti ve in vivo tracking, assessment of the f ate of the administered cells, and assessment of their therapeutic ef fects. No single modality of fers all these options at the same time. Thus, the choice of the imaging probe for cellular labeling is strictly related to the specific application one is dealing with. Optical imaging, including bioluminescence and fluorescence intra vital microscopy, of fers tools for monitoring the location, growth, and functional state of light-emitting cells, but it has the dra wback of being unab le to reconstr uct anatomical structure as well as being limited in deep tissue penetration. Ho wever, it has the adv antage of allo wing the simultaneous detection of se veral targets in the same image. Nuclear probes pro vide e xcellent sensiti vity but are obviously not particularly useful for long-term studies. MRI offers a lot of potential in this respect. It offers excellent resolution and high penetration depth, and probes may be properly designed to report about tissue physiology and metabolism in addition to the simple lighting-up in tracking experiments. In general, se veral pathw ays ma y be en visaged for cellular labeling: 1. Endosomal Uptak e b y Pinoc ytosys. This procedure consists of the entrapment of por tions of e xtracellular fluid through the en vagination of the cellular membrane that leads to the for mation of endocellular v esicles. Thus, to entrap into endosomes high amounts of imaging agent, it is necessar y to add it to the cell culture medium in w hich the cells are incubated for several hours. For instance, it is possib le to internalize Gd-HPDO3A (a commercial MRI agent) up to 10 10 units per cells when the cells are incubated overnight in the presence of 25 to 50 mM concentration of the agent.52 This procedure has been sho wn to be effective in the labeling of stem cells, tumor cells,53 and also pancreatic islets.54 Cells labeled in this way may be detected up to several weeks after their in vivo administration. 2. Cytoplasmatic Uptak e b y Electroporation. The method consists of the application of high-v oltage

291

pulses to the cell culture to yield re versible pore formation in the cellular membrane. Thus, a single pulse may be sufficient to internalize a high member of contrast agent molecules. The electroporation method has been applied to label cells either with iron o xide par ticles55 or h ydrophilic Gd–containing chelates.52 Interestingly, in the latter case, the relaxivity of the paramagnetic chelates ma y be signif icantly higher than that attainab le with the same agents b y pinoc ytosis. It has been sho wn that the additional bar rier of the endosome represents a “quenching” f actor that is, ob viously, not present when the agent is distributed in c ytoplasmatic compartment as it occurs in the case of the electroporation treatment. 3. Receptor-mediated Endocytosis. This procedure is usually quite efficient. It depends on the number of available receptors and their binding af finity and on their rec ycling time. Therefore, one needs to select the receptor/transpor ter of choice and synthesize the imaging probe containing the proper vector for the selected tar get. Though the v ector may be highl y similar to the natural ligand , the internalization process may present significant differences. The tar geted receptor/transpor ter ma y migrate on the cell membrane to reach clathrin-rich regions w here the in vagination of the membrane takes place with the for mation of endosomes that incorporate the receptor protein in their membrane. Release of the imaging probe in the endosomial cavity and recycling of the receptor/transpor ter are the important steps accounting for the efficiency of the method. 4. Internalization b y Phagoc ytosys. The method is applicable to cells that are ab le to phagoc yte particles. Of course, this methodology may be highly efficient because in a single step, it is possib le to internalize a huge number of imaging repor ters contained in a single par ticle. It is lar gely used for the entrapment of iron o xide par ticles, quantum dots, and any kind of lipid-based particles. We have shown that par ticles made of Gd-chelates containing a cleavable linker between the chelate and the insolubilizing moieties may act as a responsi ve MRI agent provided that the link er has been designed to be cleaved only by the enzyme whose activity has to be assessed.56 5. Labeling at the Outer Cell Surf ace. Recently, it has been sho wn that cells are ma y be instantaneousl y labeled when charged imaging reporters (ie, micelles or liposomes) are anchored to the cell surf ace

292

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

through a proper macromolecular link er (e g, a cationic polypeptide in the case of negatively charged particles, F igure 10). 57 For the less sensiti ve MRI technique, a suf ficient number of imaging repor ters can be loaded b y interfering with onl y 10 to 15% of the cellular surf ace, thus maintaining the o verall characteristics of the labeled cells.

profile of the imaging probe. The f irst point is par ticularly rele vant for MRI-based protocols because of the intrinsic poor sensitivity of the technique or in the case of the highly sensitive quantum dots that require the design of nanoparticles, whereas the other tw o reasons are also important for the other imaging techniques. Chemistry offers a number of nanosized systems that may be e xploited to car ry a high number of imaging reporters (Figure 11). Nanoscale devices are typically smaller than several hundred nanometers and are comparab le to the size of large biolo gical tar get molecules such as enzymes, receptors, and antibodies. Their size is signif icantly smaller than human cells so that these nanosystems offer unprecedented interactions with biomolecules either on the surf ace or inside cells, w hich ma y re volutionize disease diagnosis and treatment. F or this reason, in the following sections, par ticular emphasis will be gi ven to the recent adv ances in the design of nanosized MI probes.

MULTISYSTEMS IN IMAGING The necessity to design multiimaging probes (multimers, multiscope, multi valency, etc) to be used in molecular imaging protocols is dictated b y several reasons including: (1) the increase in the sensiti vity of the imaging reporter through the deli very of a higher number of probes to the biolo gical target of interest, (2) the possibility to modulate the contrastographic response to make it dependent on specif ic characteristics of the tar get (eg, pH, enzymatic acti vity), and (3) the modulation of the binding characteristics and the phar macokinetic

-

+

-

-

A

B

Figure 10. Labeling the outer surface of cells (Neuro 2A) by anchoring a cationic polypeptide (polyarginine) able to bind negatively charged micelles made of a Gd-based MRI probe. A, Confocal microscopy image obtained by including an amphiphilic fluorescent dye to the paramagnetic micelles. B, T1w MR image at 7 T of cellular pellets labeled (right) and not-labeled (left) with the imaging probe. Adapted with permission from Nicolle GM et al.,59 and Gianolio E et al.60

Chemistry of Molecular Imaging: An Overview

Auto-assembled Nanoparticles The most important families of nanosized auto-assembled systems include micelles, liposomes, and nanoemulsions. In MRI, paramagnetic Gd-containing micelles ha ve often been considered as contrast agents58 and to this end, several amphiphilic Gd-comple xes ha ve been synthesized. These molecules spontaneously form aggregates in which the h ydrophobic por tions are oriented within the core of the par ticle, w hereas the h ydrophilic moieties (containing the Gd-probe) are e xposed to the solv ent. Depending on the chemical str ucture of the amphiphile (length, number , and distribution of the h ydrophobic groups in the molecule) and the nature of the hydrophilic heads, the resulting micelles can ha ve different sizes and shapes (spherical, cylindric, ellipsoidal). The obser ved relaxivities are basicall y deter mined by the number of w ater molecules (q ) in the inner h ydration sphere of the paramagnetic metal ion. Usuall y, the systems with a labile single coordinated water have relaxivity values (0.47 T, 25°C) in the range 18 to 23 (mM Gd × s)−1, whereas the cor responding inter val for q = 2 systems is 29 to 30 (mMGd × s)−1. Despite the size of the agg regates, the relati vely low relaxivity v alues (e xpressed per paramagnetic centre) shown by the micelles primarily depends on the nonoptimal

1

10

dynamic re gime of these systems, w here, usuall y, the Gd-containing cage can rotate f aster and independentl y from the motion of the w hole par ticle. In addition, it has been shown that also the direct dipolar interaction betw een nearby Gd(III) centers can reduce the ef ficacy of such systems, but this drawback can be removed by diluting the paramagnetic ion with a diamagnetic analog like Y(III).59,60 Paramagnetic micelles have been examined in several MR-targeted molecular imaging e xperiments.61–63 An illustrative example was published by Fayad and coworkers who used Gd-containing immunomicelles for detecting atherosclerosis. 64 The selected cellular tar get was the macrophage scavenger receptors (MSR), a macrophagespecific cell-surf ace protein, signif icantly overexpressed on atherosclerotic macrophages and foam cells. In vivo MRI e xperiments sho wed that 24 hours post-injection immunomicelles yielded a 79% increase in signal intensity of the atherosclerotic aor tas in ApoE−/− mice compared with onl y 34% using untar geted micelles, w hereas no enhancement was detected using Gd-DTPA. Confocal laser scanning microscopy showed colocalization between fluorescent immunomicelles and macrophages in plaque. Another rele vant and inno vative use of micelle-lik e systems in the MI f ield w as pub lished b y Mulder and colleagues who prepared a tar geted and paramagneticall y coated solub le quantum dot for multimodal optical/MRI

Nano-ruler

Proteins (Apoferritin Virus capsides,...)

Dendrimers

293

1000 nm

100

Water

Nanovesicles (e.g. Liposomes) Micronized solid metal-based particles

Micelles Lipid

Nanoemulsions (Per Fluoro Carbon nanop., Solid Lipid Nanoparticles)

Solid Nanoparticles (Metal Oxides, Hollow nanoshells,...) Figure 11. Schematic representation of nanosized carriers relevant to molecular imaging applications. Adapted with permission from Aime S et al.3

294

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

visualization of tumor neoangio genesis. The QD nanocrystal w as entrapped in the h ydrophobic core of a micelle coated with pe gylated phospholipids, par tly conjugating with the RGD peptide (a w ell-known c yclic peptide targeting the αvβ3 integrins upregulated in neoangiogenic endothelium), and amphiphilic Gd-containing probe (Figure 12).65 Analogous systems w ere also tested for the multimodal visualization of apoptosis in vitr o by using Annexin A5 as targeting vector.66,67 Liposomes are nanosized vesicles in which an aqueous core is usuall y encapsulated b y unilamellar phospholipidic bilayer. As soon as they were discovered in the middle of the 1960s, the great relevance of such systems in the biomedical f ield as model systems for studying biological membranes and as drug carriers, was immediately recognized primarily for their biocompatibility and chemical v ersatility. Liposomes ha ve been e xtensively investigated for MRI applications because, in addition to display a high pa yload of contrast agents, the y can be successfully used for impro ving the contrasto graphic efficiency of the transpor ted agent, as well as for generating new contrast mechanisms. Most of the works carried out in vivo dealt with liposomes incorporating amphiphilic Gd(III) complexes. The millimolar relaxivity of these agents is enhanced thanks to the slow tumbling motion of the paramagnetic complex

A

Probe

embedded in the bila yer. In addition, the total number of probes that are car ried b y a single liposome can be very high (up to10 5), thus resulting in a v ery ef ficient nanoparticles. Besides the rotational dynamics of the comple x, the relaxivity of liposomes incor porating Gd(III) comple xes is strongly influenced b y the w ater exchange rate, kex, of the metal-coordinated w ater. This parameter has to be f inely tuned (the optimal values are in the range 108 to 109 s−1)68 to maximize the o verall relaxi vity of the system. The tw o routes that ha ve been follo wed so f ar for conjugating the lipophilic chains to the coordination cage of the metal complex involve: (1) the for mation of amide bonds transforming the carbo xylic groups of the ac yclic ligand DTPA and (2) linkages that minimally alter the chemical characteristics of the donor atoms of macrocyclic (DOTA-based) ligands. The first approach was the most used, primarily for the relative easiness in the ligand synthesis, but it suf fers from two major dra wbacks: (1) the ther modynamic and kinetic stabilities of DTPA amides could be too low for the safety of in vivo applications,54 and (2) the kex values for this coordination cage is much lo wer than optimal ( ≤ 106 s−1), thus reducing the relaxi vity attainab le with these nanoprobes. Conversely, macroc yclic DO TA-like str uctures displa y a much higher chemical stability and w ater exchange values closer to the optimum, e ven w hen one of the carbo xylate groups is transfor med in amide for conjugating

RGD peptide

Paramagnetic lipid PEGylated lipid Quantum dot

B

In vivo tumor vasculature imaging IVM

MRI

Figure 12. A, Schematic view of a water-soluble QD-micelle loaded with the RGD peptide and a paramagnetic MRI probe. B, Intravital fluorescence microscopy (left) showed tumor vasculature at the cellular level, whereas MRI (right) visualizes tumor neovessels at anatomical level. Yellow arrows indicate the endothelium. Adapted with permission from Mulder WJ et al.83

Chemistry of Molecular Imaging: An Overview

the lipophilic tail/s. Typical relaxivity values (at 0.47 T) for Gd-DTPA-bisamides incorporated in liposomes range from 8 to 12 (mM Gd × s)−1 depending on temperature and liposome for mulation,69,70 whereas similar nano vesicles incorporating amphiphilic DO TA-based compounds displa y relaxivity values from 25 to 45 (mMGd × s)−1.71,72 Liposomes incor porating Gd(III) comple xes ha ve been successfully used in many in vivo MRI-targeted molecular imaging e xperiments including the visualization of tumors,73,74 detection of atherosclerotic plaques, 75 lymph nodes,76 inflammation sites, 77 and visualization of myocardium infarcted areas.78 The w ell-established use of liposomes as dr ug delivery systems prompted researches aimed at imaging the dr ug deli very process. An e xample w as published by Viglianti and colleagues, 79 who prepared thermosensitive liposomes containing in their aqueous cavity the antitumoral dr ug do xorubicin along with paramagnetic Mn(II) ion as T1-agent. The lo w w ater permeability of the liposome bila yer mak es silent the encapsulated paramagnetic probe, but when the system is brought at temperature higher than the gel-to-liquid transition phase temperature of the phospholipids (> 40°C), the relaxi vity of the system increases o wing to the release of Mn(II) ions. In such a way, the indirect visualization of the do xorubicin release w as possib le. This system w as also tested in vivo on rats bearing a xenografted fibrosarcoma. Other impor tant applications of paramagnetic liposomes are in the f ield of responsi ve MRI agents, w here they were investigated as temperature80,81 and pH probes.82 As already anticipated for the micelles, liposomes appear excellent candidates for designing probes containing imaging reporters that can be visualized with different imaging modalities. Most of the w orks in this f ield have been carried out b y combining MRI and optical imaging probes.83 Besides micelles and liposomes, other routes based on self-assembling ha ve been approached. Lanza, Wickline, and coworkers at Washington State University have widely explored the MRI visualization of tar geting molecules b y using lipidic microemulsions containing several thousands of Gd(III) chelates. 84 Recently, they showed that targets at picomolar concentrations on a single la yer of cells can be visualized b y their imaging probe containing 96,400 Gd(III) units.85 Another approach that has been followed to accumulate a high number of imaging repor ters at the tar geting site relies on the formation of avidin/biotin supramolecular adducts. The exploitation of the outstanding binding affinity between biotin and a vidin (KA ~1015) appears to

295

be increasingly used in MR-molecular imaging applications. Bhujwalla and coworkers have recently applied this recognition path for the visualization of HER-2/C-neu receptors (a member of the epidermal growth factor family hyperexpressed in multiple cancers).86 They addressed the extracellular domain of the receptors b y means of a biotinylated monoclonal antibody (mAb). After clearance of the unbound mAb, Gd-labeled a vidin is administered and binds, with a high af finity, to the biotin ylated mAb. The expression le vel of the receptor w as estimated at 7 × 105 receptors/cell, and the a veraged number of Gd-DTPA units per a vidin molecule w as 12.5. The method has been successfully applied in an experimental mouse model of breast carcinoma. Another example based on the reco gnition properties of the biotin/a vidin pair has been repor ted by Kobayashi and Brechbiel, 87 who investigated the uptak e of a macromolecular construct comprised of avidin and a biotinylated dendrimer bearing 254 Gd-DTP A chelates into SHIN3 cells (a cell line obtained originall y from human o varian cancer). The internalization process is driven by the avidin molecule, a gl ycoprotein that reco gnizes β-D-galactose receptors, w hich are present in either nor mal hepatocytes o r cancer cells (especiall y o varian and colorectal adenocarcinoma cells). We ha ve fur ther de veloped this approach b y using mono- and bis-biotinylated Gd-chelates. By proper addition of avidin and the two chelates, it is possible to build up, at the a vidin receptor sites, multila yered str uctures containing a number of Gd(III) ions suf ficient for the MRI detection (Figure 13). 88

Probing Multiple Molecular Pathways Simultaneously X-ray and MRI contrast agents, as w ell as radio-labeled tracers, have been developed to act as repor ters of a single, physiological, or molecular e vent. However, the adv ent of molecular imaging poses the challenging question of visualizing multiple molecular pathw ays in the same anatomical region to elucidate possible casual relationships between them. Clearly, the biolo gists’ view is strongl y affected by the armory of tools that immunohistochemistry can provide and certainly optical imaging is the most adequate modality for the in vivo translation of this pool of infor mation. In fact, a number of fluorescent dyes, quantum dots, and even bioluminescent proteins emitting at dif ferent w avelengths are already a vailable for the setup of molecular imaging protocols aimed at visualizing dif ferent epitopes and sequencing cellular events.

296

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

BIOTIN

STREPTAVIDIN

BIOTIN

H 2O C3d Gd

Gd Gd

Gd Gd

Gd

H2O

Gd

Gd-loaded Apoferritin

Biotin-ASKKPKRNIKA C3d-Bio 5

Biotin-ASKKPKRNIKA Biotin-ASKKPKRNIKA

H N N H H N

Lys Lys

CONH2

Lys Biotin-ASKKPKRNIKA

N H

Figure 13. Routes to the formation of multilayered structures by exploiting the strong interaction between mono- and bis-biotinylated Gd-based probes and avidin. Adapted with permission from Crich SG et al.88

In principle, SPECT detection of different molecular probes is possib le pro viding that the y are labeled with radioisotopes that emit gamma rays at different energies. Although the possibility of attaining quantitati ve images of different radiotracer distributions has not yet been fully explored, it is worth mentioning as likely it will be further developed in the for thcoming years. Figure 14 repor ts a typical dual probe SPECT image. 89 Besides 99mTc/111In, other radioisotope pairs (e g, 99mTc/123I or 99mTc/125I) are under intense scrutiny. Much activity is currently present in the MRI f ield for the de velopment of probes that detect simultaneously molecules and molecular e vents in the same region. In f act, several new research lines are g rowing very f ast in the f ield of frequenc y-encoding contrast agents. The possibility of exploiting this parameter led to the development of agents based on peculiar proper ties of given heteronuclei (e g, h yperpolarized nob le gases atoms90 or 13C-containing molecules, 91 as well as

19

F-based chemicals. 92 This approach has the adv antage of dealing with images characterized b y a zero background signal, but the attainment of sufficiently high sensitivity remains a quite challenging task. Therefore, the possibility of designing frequenc y-encoded MRI protocols based on the detection of the 1H-water signal is still a very fascinating perspective. The route to a new class of agents that conjugates frequency-encoding and water signal detection has its roots in the well-established magnetization transfer (MT) MRI procedure, which relies on the transfer of saturated magnetization from tissutal mobile protons (w ater and labile protons from proteins) upon RF irradiation of their semisolid like broad NMR absor ption.93 The use of e xogenous probes, the so-called CEST agents, whose exchangeable proton resonances are characterized by a shar per absor ption signal, introduces the possibility of selecti ve RF ir radiation.5,94 Thus, one ma y design ne w protocols in w hich the contrast in a MR image is generated “at will” onl y if the proper frequenc y

Chemistry of Molecular Imaging: An Overview

OH

HO O P

P O O

O

99mTc-MDP

Tc O

O

O P

P O

HO

2 2

N OOC

OH

COO2

In31

OOC

297

N

N CO-D-Phe-Cys-Phe

COO2 Thr(ol)

S

D-Trp

S

Lys

Cys-Thr

111In-Octreoscan

Figure 14. Dual-isotope SPECT image. The animal was injected with 99mTc-MDP (a bone imaging probe shown in a full spectrum color scale) followed by the administration of 111In-Octreoscan (a somatostatin analogue shown in hot iron color scale). Images were acquired for 28 minutes using a single detector with seven pinholes. Adapted with permission from Meikle SR et al.89

Figure 15. MR images of a phantom consisting of four capillaries containing: unlabeled cells (A), cells labeled with a Tb(III)-based PARACEST probe (B) and with an Eu(III)-based PARACEST probe (C), and cellular pellets obtained by mixing (1:1) B- and C- labeled cells. Middle: proton image. Right: CEST-MR difference image upon irradiation of the specific frequency of Eu(III)-based probe. Left: CEST-MR difference image upon irradiation of the specific frequency of Tb(III)-based probe. The contrast was arbitrarily colored in red and green, respectively. Adapted with permission from Aime S et al.95

corresponding to the e xchangeable protons of the added contrast agent is ir radiated. The possibility of detecting more than one agent in the same region is shown in Figure 15, where sets of the same cell line ha ve been labeled with tw o CEST agents differing onl y for the coordinated lanthanide ion. 95 The exchangeable protons of the coordinated water molecules resonate at quite dif ferent frequencies for w hich it is straightforward to sho w their presence in the phantom upon their specif ic irradiation. Because CEST agents w ere disco vered in 2000, 94 it was realized early that one of the main limiting f actors for

the in vivo application of such systems w ould have been represented by their relatively low sensitivity in comparison to the con ventional Gd- and iron-based agents. Therefore, many efforts have been devoted to overcome this drawback. Among the physico-chemical parameters controlling the ef ficiency of the saturation transfer (ST), the exchange rate of the mobile protons of the agent, kex, received much attention because either it can be modulated b y introducing suitab le changes in the chemical structure of the probe or this parameter is usuall y dependent on v ariables, such as temperature and pH, that can make CEST agents potential responsive agents.96

298

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Another w ay to impro ve ST can be attained b y increasing the number or equivalent or pseudo-equivalent (ie, saturable by a single B2 pulse) mobile protons belonging to the CEST agent. A breakthrough in the development of highly sensitive paramagnetic CEST agents w as achieved by using lar ger nanosystems lik e liposomes. Such biocompatib le nanovesicles are optimal CEST platfor ms because their aqueous core contains an extraordinarily high number (107 to 10 9 depending on the liposome size) of w ater protons exchanging with the bulk solvent. The water protons inside and outside the ca vity resonate at the same frequenc y (Δω = 0), but the entrapment of a h ydrophilic paramagnetic shift reagent in the liposomal ca vity produces a shift of the encapsulated w ater resonance, thus allo wing the basic condition for a CEST probe ( Δω > kex) to be fulfilled.97 The best shift reagents ha ve been found among paramagnetic metal complexes based on Ln(III) ions containing highl y shifted and f ast e xchanging coordinated water molecule. The effect on the chemical shift of w ater protons induced by a Ln(III) complex is mainly dependent

on the paramagnetism of the metal ion (e xpressed by the value of Bleane y’s constant) and on the str ucture of the metal complex that determines the orientation of the metalbound water protons with respect to the magnetic axis of the comple x. F or instance, in axiall y symmetric macrocyclic Ln(III) comple xes (e g, [Ln-DO TMA]−), the coordinated water molecule is aligned along the magnetic axis, and consequently, the observed shift is very large.3 In addition to the intrinsic efficiency of the SR, the chemical shift of the intraliposomal w ater protons is directly cor related to the concentration of the SR entrapped in the liposome. F igure 16 (top) repor ts the 1H-NMR spectr um of a suspension of liposomes entrapping [TmDO TMA]− at the concentration of 0.12 M. The signal at about 4 ppm do wnfield from bulk water corresponds to the intraliposomal w ater protons. Accordingly, the Z-spectr um of the liposome suspension (F igure 16, bottom) sho ws a CEST peak at the same frequency. The very large number of CEST-active water protons mak es such agents (that w e dubbed

2

COO2

OOC N

N Tm31

N 2

N COO2

OOC

100

80

60 Iwat /Iwat0/% 40

20

0 8

6

4

2

0 22 d/ppm

24

26

28

Figure 16. NMR (at 14.1 T) and Z-spectra (at 7 T) of a suspension of LIPOCEST agent encapsulating Tm-DOTMA (shown in the inlet) as shift reagent. The signal detected at 3.1 ppm from bulk water corresponds to the intraliposomal water protons. Adapted with permission from Aime S et al.97

Chemistry of Molecular Imaging: An Overview

LIPOCEST) e xtremely sensiti ve. F igure 17 repor ts a CEST dif ference image of a phantom of capillaries containing different liposome concentrations (indicated in nanomolar scale in the bottom left scheme) to sho w the dependence of the ST e xtent on the LIPOCEST concentration. The image indicates that a CEST contrast is still detectab le in a solution containing appro ximately 100 pM of agent!

The Multivalence Approach The search for enhanced tar geting capabilities that provide higher concentration at the site of interest is a neverending objecti ve in molecular medicine. Therefore, the

180

299

in-depth knowledge of the recognition pathway that leads to binding remains a central task in the design of the “magic bullet” that is the dream of phar macologists. Once a given binding motif has been identified, a route to further enhance the tar geting capabilities of the probe is to design a multimeric deri vative of the selected motif that is ab le to pro vide multiple connections to its tar get molecule. Thus, in this context, multivalency simply consists of clustering v ector molecules to enhance the o verall binding to the magnetic repor ter. It is a quite general procedure that may be applied any time one has identified the binding motif for a gi ven epitope. F or instance, w e have applied this procedure to visualize endothelial tumor cells in a MR image targeting overexpressed neural cell adhesion molecules (NCAM). 98 The e xtracellular

360

720

90

22.5

1440

45

2880

Figure 17. CEST-MR images of a phantom consisting of eight capillaries containing different concentrations of the LIPOCEST probe displayed in Figure 16. Top left: image after saturation at 3.1 ppm (from bulk water). Top right: image after saturation at −3.1 ppm. Bottom right: difference between image at 3.1 and image at −3.1. Bottom left: concentrations (picomolar scale) of the LIPOCEST probe in the capillaries. Adapted with permission from Aime S et al.97

300

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

portion of NCAM consists of f ive Ig and two fibronectin type III modules. A ligand (C3m) for NCAM Ig1 has been identified from a combinatorial librar y of synthetic peptides.99 Furthermore, the characteristics of the binding scheme between C3m and NCAM Ig1 ha ve been elucidated by NMR spectroscop y. As it is kno wn that multimeric for ms of peptide ligands identif ied b y means of ohage display peptide libraries have a higher potency for receptor acti vation than monomeric for ms,100,101 a dendrimer composed of four monomers coupled to a l ysine backbone has been prepared (C3d). It binds NCAM with a dissociation constant ( Kd) of 10 −5 M. Interestingly, the dendrimeric str ucture possib ly f acilitates NCAM Ig1 binding making C3d less sensiti ve to single amino acid substimulation in the individual peptide chains than C3m. On this basis the MR visualization of tumor angiogenesis has been pursued by a two-step procedure (Figure 13) that consists of: (1) tar geting NCAM with a biotin ylated derivate of a C3d-peptide, and (2) deli very of a streptavidin/ gadolinium (Gd)-loaded apofer ritin 1:1 adduct of the biotinylated sites. The remarkable relaxation enhancement (r1 ≈ 80 mM −1 s−1) of the Gd-loaded apofer ritin system allowed the visualization of the tumor endothelial cells when organized in microvessels connected to mouse vasculature. Further illustrati ve e xamples of the multi valency approach in designing probes for molecular imaging applications are reported in ref 104.

Final Remarks The molecular imaging approach of fers a g reat potential for earlier detection and characterization of diseases and evaluation of treatment. However, more research is necessary to bring these ideas to clinical applications and a key aspect relates to the de velopment of high-specif icity, high-sensitivity imaging probes for the different detection modalities. Apart from its use in earl y diagnosis, the molecular imaging approach will ha ve also a major impact on the development of ne w phar maceuticals. The re gulatory agencies indicate that the use of “sur rogate or biomarkers” can accelerate dr ug-approval procedures. Molecular imaging can be considered as such a biomark er because it has g reat potential to mak e better predictions on the effectiveness and toxicity of dr ugs. Therefore, the development of this f ield will not onl y enable early diagnosis but will also signif icantly increase the a vailability and “speed to market” of new drugs.

Chemistry of fers a lot for the de velopment of the currently a vailable probes for the dif ferent imaging modalities as well as provides solutions for dual imaging modalities probes. Ho wever, real breakthroughs in the field may come out onl y from an impro ved integration among chemists, biolo gists, and imaging technolo gists. Molecular imaging is f ar from being a mature science, and each procedure still needs a continuous syner gic optimization of all the components to pass from the “proof-of-concepts” stage to become a robust protocol.

REFERENCES 1. Blasberg R. Imaging update: ne w windows, new views. Clin Cancer Res 2007;13:3444–8. 2. McGuire AH, Dehdashti F, Siegel BA. Positron tomographic assessment of 16 alpha-[18F] fluoro-17 beta-estradiol uptake in metastatic breast carcinoma. J Nucl Med 1991;32:1526–31. 3. Aime S, Geninatti Crich S, Gianolio E, et al. High sensiti vity lanthanide(III) based probes for MR-medical imaging. Coord Chem Rev 2006;250:1562. 4. Corot C, Rober t P, Idee JM, P ort M. Recent adv ances in iron o xide nanocrystal technology for medical imaging. Adv Drug Deliv Rev 2006;58:1471. 5. Zhou J, van Zijl PCM. Chemical e xchange saturation transfer imaging and spectroscop y. Pro g Nucl Magn Reson Spectrosc 2006;48:109–36. 6. Aime S, Cabella C, Colombatto S, et al. Insights into the use of paramagnetic Gd(III) comple xes in MR-molecular imaging in vestigations. J Magn Reson Imaging 2002;16:394–406. 7. Trost BM. The atom econom y: a search for synthetic ef ficiency. Science 1991;254:1471–7. 8. Trost BM. On inventing reactions for atom econom y. Acc Chem Res 2002;35:695–705. 9. Pearson AJ, Roush WR, editors. Handbook of reagents for organic synthesis, acti vating agents and protecting g roups. John Wiley & Sons; 2001. p. 186–8 and references cited therein. 10. Klausner YS, Bodansky M. Coupling reagents in peptide synthesis. Synthesis 1972;453–63. 11. Staros JV . N-h ydroxysulfosuccinimide acti ve esters: bis (N-hydroxysulfosuccinimide) esters of two dicarboxylic acids are hydrophilic, membrane-imper meant, protein cross-link ers. Biochemistry 1982;21:3950–5. 12. Anelli PL, Lattuada L, Gabellini M, Recanati P . DOTA tris(phenylmethyl) ester: a ne w useful synthon for the synthesis of DO TA monoamides containing acid-labile bonds. Bioconjug Chem 2001;12:1081–4. 13. Jaakkola L, Ylikoski A, Ho vinen J. Simple synthesis of a building block for solid-phase labeling of oligonucleotides with 1,4,7,10tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA). Bioconjug Chem 2006;17:1105–7. 14. Konings MS, Dow WC, Love DB, et al. Gadolinium complexation by a ne w dieth ylenetriaminepentaacetic acid-amide ligand. Amide oxygen coordination. Inorg Chem 1990;29:1488–91. 15. Maisano F, Gozzini L, De Haen C. Coupling of DTP A to proteins: a critical analysis of the c yclic dianhydride method in the case of insulin modification. Bioconjug Chem 1992;3:212–7. 16. Hnatowich DJ, Layne WW, Childs RL, et al. Radioacti ve labeling of antibody: a simple and ef ficient method. Science 1983;220:613–5.

Chemistry of Molecular Imaging: An Overview

17. Anelli PL, F edeli F , Gazzotti O , et al. L-Glutamic acid and L-lysine as useful building b locks for the preparation of bifunctional DTPA-like ligands. Bioconjug Chem 1999;10:137–40. 18. Tung C-H, Mahmood U, Bredow S, Weissleder R. In vivo imaging of proteolytic enzyme acti vity using a no vel molecular repor ter. Cancer Res 2000;60:4953–8. 19. Yu KN, Lee S-M, Han JY , et al. Multiple x targeting, tracking, and imaging of apoptosis b y fluorescent surf ace enhanced Raman spectroscopic dots. Bioconjug Chem 2007;18:1155–62. 20. Kelly KA, Waterman P, Weissleder R. In vivo imaging of molecularly targeted phage. Neoplasia 2006;8:1011–8. 21. Brechbiel MW , Ganso w O A, Atcher R W, et al. Synthesis of 1-(p-isothiocyanatobenzyl) derivatives of DTPA and EDTA. Antibody labeling and tumor -imaging studies. Inor g Chem 1986;25:2772–81. 22. Keana JFW , Mann JS. Chelating ligands functionalized for f acile attachment to biomolecules. A convenient route to 4-isothiocyanatobenzyl derivatives of diethylenetriaminepentaacetic acid and ethylenediaminetetraacetic acid. J Org Chem 1990;55:2868–71. 23. Deal KA, Motekaitis RJ , Mar tell AE, Welch MJ. Ev aluation of the stability and animal biodistribution of gadolinium(III) benzylamine-derivatized dieth ylenetriaminepentaacetic acid. J Med Chem 1996;39:3096–106. 24. Corson DT, Meares CF . Ef ficient multig ram synthesis of the bifunctional chelating agent ( S)-1-p-isothiocyanatobenzyl-diethylenetetraminepentaacetic acid. Bioconjug Chem 2000;11:292–9. 25. Arano Y, Uezono T, Akizawa H, et al. Reassessment of diethyl enetriaminepentaacetic acid (DTP A) as a chelating agent for indium111 labeling of pol ypeptides using a ne wly synthesized monoreactive DTPA derivative. J Med Chem 1996;39:3451–60. 26. Shi X, Thomas TP, Myc LA, et al. Synthesis, characterization, and intracellular uptak e of carbo xyl-terminated pol y(amidoamine) dendrimer-stabilized iron o xide nanopar ticles. Ph ys Chem Chem Phys 2007;9:5712–20. 27. Wang W, McMurray JS, Wu Q, et al. Convenient solid-phase synthesis of dieth ylenetriaminepenta-acetic acid (DTP A)-conjugated cyclic RGD peptide analo gues. Cancer Biother Radiophar m 2005;20:547–56. 28. Aime S, Dastrù W, Gobetto R, Viale A. para-Hydrogenation of unsaturated moieties on pol y(lysine) derived substrates for the de velopment of novel hyperpolarized MRI contrast agents. Org Biomol Chem 2005;3:3948–54. 29. Gløgård C, Stensr ud G, Aime S. Novel radical-responsive MRI contrast agent based on paramagnetic liposomes. Magn Reson Chem 2003;41:585–8. 30. Ozker SK, Hellman RS, Krasno w AZ. Preparation of 99mTc labeled substance P (SP). Appl Radiat Isot 2002;57:729–32. 31. Andrè JP, Geraldes CF, Martins JA, et al. Lanthanide(III) comple xes of DOTA-glycoconjugates: a potential ne w class of lectin-mediated medical imaging agents. Chem Eur J 2004;10:5804–16. 32. Essien H, Lai JY , Hwang KJ. Synthesis of dieth ylenetriaminepentaacetic acid conjugated inulin and utility for cellular uptak e of liposomes. J Med Chem 1988;31:898–901. 33. Ke T, Jeong E-K, Wang X, et al. RGD tar geted pol y (L-glutamic acid)-cystamine-(Gd-DO3A) conjugate for detecting angiogenesis biomarker αvβ3 integrin with MR T1 mapping. Int J Nanomedicine 2007;2:191–9. 34. Wang X, Feng Y, Ke T, et al. Phar macokinetics and rissue retention of (Gd-DTPA)-cystamine copolymers, a biodegradable macromolecular magnetic resonance imaging contrast agent. Phar m Res 2005;22:596–602. 35. Cole RD, Stein WH, Moore S. Cysteine content of human hemo globin. J Biol Chem 1958;233:1359–63. 36. Hirai M, Minematsu H, K ondo N, et al. Accumulation of liposome with Sialyl Lewis X to inflammation and tumor region: application

37.

38.

39.

40.

41.

42.

43.

44.

45. 46.

47.

48.

49.

50.

51. 52.

53.

54.

55. 56.

301

to in vivo bio-imaging. Biochem Bioph ys Res Commun 2007;353:553–8. O’Keefe ET, Mordick T, Bell JE. Bovine galactosyltransferase: interaction with alpha-lactalbumin and the role of alpha-lactalbumin in lactose synthase. Biochemistry 1980;19:4962–6. Belinka B A Jr , Coughlin DJ , Alvarez VL, Wood R. Preparation of metal-binding tar geted pol ypeptide constr ucts for diagnosis and therapy. Int Pat. WO950913, 2005. Nampalli S, McDougall MG, La vrenov K, et al. Utility of thiolcross-linked fluorescent dy e labeled ter minators for DN A sequencing. Bioconjug Chem 2002;13:468–73. Cai W, Shin D-W, Chen K, et al. P eptide-labeled near-infrared quantum dots for imaging tumor v asculature in li ving subjects. Nano Lett 2006;6:669–76. Li C, Winnard PT Jr , Takagi T, et al. Multimodal image-guided enzyme/prodrug cancer therap y. J Am Chem Soc 2006;128:15072–3. Josephson L, Kircher MF , Mahmood U , et al. Near -infrared fluorescent nanopar ticles as combined MR/optical imaging probes. Bioconjug Chem 2002;13:554–60. Zhao M, Kircher MF , Josephson L, Weissleder R. Dif ferential conjugation of tat peptide to super paramagnetic nanopar ticles and its ef fect on cellular uptak e. Bioconjug Chem 2002;13: 840–4. Hofmann M, Maeck e H, Bor ner R, et al. Biokinetics and imaging with the somatostatin receptor PET radioligand (68)Ga-DO TATOC: preliminary data. Eur J Nucl Med 2001;28:1751–7. Maecke HR, Hofmann M, Haberk orn U . (68)Ga-labeled peptides in tumor imaging. Eur J Nucl Med 2005;46:172–88. Wester HJ , Schottelius M, Scheidhauer K, et al. PET imaging of somatostatin receptors: design, synthesis and pre-clinical e valuation of a novel 18F-labelled, carbohydrated analogue of octreotide. Eur J Nucl Med Mol Imaging 2003;30:117–22. Meisetschlager G, Poethko T, Stahl A, et al. Gluc-Lys([18F]FP)-TOCA PET in patients with SSTR-positi ve tumors: biodistribution and diagnostic evaluation compared with [111In]DTP A-octreotide. J Nucl Med 2006;47:566–73. Haubner R, Wester HJ , Weber WA. Nonin vasive imaging of αvβ3 integrin e xpression using 18F-labeled RGD-containing gl ycopeptide and positron emission tomo graphy. Cancer Res 2001;61:1781–5. Beer AJ, Haubner R, Sarbia M. Positron emission tomography using [18F]Galacto-RGD identifies the level of integrin αvβ3 expression in man. Clin Cancer Res 2006;12:3942–9. Ginj M, Zhang H, Waser B. Radiolabeled somatostatin receptor antagonists are preferable to agonists for in vivo peptide receptor targeting of tumors. Proc Natl Acad Sci USA 2006;103:16436–41. Merrifield BR. Solid phase peptide synthesis. I. The synthesis of a tetrapeptide. J Am Chem Soc 1963;85:2149–54. Terreno E, Crich SG, Belf iore S, et al. Ef fect of the intracellular localization of a Gd-based imaging probe on the relaxation enhancement of water protons. Magn Reson Med 2006;55:491–7. Crich SG, Biancone L, Cantaluppi V, et al. Impro ved route for the visualization of stem cells labeled with a Gd-/Eu-Chelate as dual (MRI and fluorescence) agent. Magn Reson Med 2004;51: 938–44. Cabella C, Crich SG, Cor pillo D , et al. Cellular labeling with Gd(III) chelates: only high thermodynamic stabilities prevent the cells acting as ‘sponges’ of Gd 3+ ions. Contr Media Mol Imaging 2006;1:23. Tai JH, Foster P, Rosales A, et al. Imaging islets labeled with magnetic nanoparticles at 1.5 Tesla. Diabetes 2006;55:2931–8. Himmelreich U, Aime S, Hieronymus T, et al. A responsive MRI contrast agent to monitor functional cell status. Neuroimage 2006;32:1142.

302

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

57. Gianolio E, Giovenzana GB, Ciampa A, et al. A novel method of cellular labeling: anchoring MR-imaging repor ter particles on the outer cell surface. ChemMedChem 2008;3:60–2. 58. Mulder WJM, Strijk ers GJ , v an Tilborg GAF , et al. Lipid-based nanoparticles for contrast-enhanced MRI and molecular imaging. NMR Biomed 2006;19:142–64. 59. Nicolle GM, Helm L, Merbach AE. 8S paramagnetic centres in molecular assemblies: possible effect of their pro ximity on the w ater proton relaxivity. Magn Reson Chem 2003;41:794–9. 60. Gianolio E, Gio venzana GB , Longo D , et al. Relaxometric and modelling studies of the binding of a lipophilic Gd-AAZT A complex to fatted and defatted human serum albumin. Chem Eur J 2007;13:5785–97. 61. Accardo A, Tesauro D, Roscigno P, et al. Ph ysicochemical properties of mix ed micellar agg regates containing CCK peptides and Gd complexes designed as tumor specif ic contrast agents in MRI. J Am Chem Soc 2004;126:3097–107. 62. Accardo A, Tesauro D, Morelli G, et al. High-relaxi vity supramolecular aggregates containing peptides and Gd complexes as contrast agents in MRI. J Biol Inor g Chem 2007;12:267–76. 63. Raatschen HJ, Swain R, Shames DM, et al. MRI tumor characterization using Gd-Gl yMe-DOTA-perfluorooctyl-mannose-conjugate (Gadofluorine MTM), a protein-a vid contrast agent. Contrast Media Mol Imaging 2006;1:113–20. 64. Amirbekian V, Lipinski MJ , Brile y-Saebo KC, et al. Detecting and assessing macrophages in vivo to evaluate atherosclerosis noninvasively using molecular MRI. Proc Natl Acad Sci U S A 2007;104:961–6. 65. Mulder WJ, Koole R, Brandwijk RJ, et al. Quantum dots with a paramagnetic coating as a bimodal molecular imaging. Nano Lett 2006;6:1–6. 66. Van Tilborg GA, Mulder WJ, Chin PT , et al. Annexin A5-conjugated quantum dots with a paramagnetic lipidic coating for the multimodal detection of apoptotic cells. Bioconjug Chem 2006;17:865–8. 67. Prinzen L, Miser us RJ, Dirksen A, et al. Optical and magnetic resonance imaging of cell death and platelet activation using Annexin a5-functionalized quantum dots. Nano Lett 2007;7:93–100. 68. Aime S, Botta M, F asano M, Terreno E. Prototropic and w aterexchange processes in aqueous solutions of Gd(III) chelates. Acc Chem Res 1999;32:941–9. 69. Bertini I, Bianchini F , Calorini L, et al. P ersistent contrast enhancement b y stericall y stabilized paramagnetic liposomes in murine melanoma. Magn Reson Med 2004;52:669–72. 70. Strijkers GJ, Mulder WJM, Heeswijk RB, et al. Relaxi vity of liposomal paramagnetic MRI contrast agents. MA GMA 2005;18:186–92. 71. Gløgard C, Stensr ud G, Ho vland R, et al. Liposomes as car riers of amphiphilic gadolinium chelates: the effect of membrane composition on incorporation efficacy and in vitro relaxivity. Int J Pharm 2002;233:131–40. 72. Tei L, Crich SG, Bussolati B , et al. (2007). Proc. Joint Annual Meeting ISMRM-ESMRMB, Berlin, Germany 256. 73. Sipkins D A, Cheresh D A, Kazemi MR, et al. Detection of tumor angiogenesis in vivo by αvβ3-targeted magnetic resonance imaging. Nat Med 1998;4:623–6. 74. Mulder WJ, Strijkers GJ, Habets JW, et al. MR molecular imaging and fluorescence microscop y for identif ication of acti vated tumor endothelium using a bimodal lipidic nanopar ticle. F ASEB J 2005;19:2008–10. 75. Mulder WJM, Douma K, Koning GA, et al. Liposome-enhanced MRI of neointimal lesions in the ApoE-KO mouse. Magn Reson Med 2006;55:1170–4. 76. Trubetskoy VS, Cannillo JA, Milshtein A, et al. Controlled delivery of Gd-containing liposomes to l ymph nodes: surf ace modif ication

77.

78.

79.

80. 81.

82.

83.

84.

85.

86.

87.

88.

89.

90. 91. 92.

93. 94.

95.

96.

97.

may enhance MRI contrast proper ties. Magn Reson Imaging 1995;13:31–7. Sipkins D A, Gijbels K, Tropper FD , et al. ICAM-1 e xpression in autoimmune encephalitis visualized using magnetic resonance imaging. J Neuroimmunol 2000;104:1–9. Chu WJ, Simor T, Elgavish GA. In vivo myocardial tissue kinetics of Gd(ABE-DTTA), a tissue-persistent contrast agent. NMR Biomed 1997;10:87–92. Viglianti BL, Abraham SA, Michelich CR, et al. In vivo monitoring of tissue phar macokinetics of liposome/dr ug using MRI: illustration of tar geted deli very. Magn Reson Med 2004;51:1153–62. Kong G, De whirst MW. Re view Hyper thermia and liposomes. Int J Hyperthermia 1999;15:345–70. Frich L, Bjø rnerud A, Fossheim S, et al. Experimental application of thermosensitive paramagnetic liposomes for monitoring magnetic resonance imaging guided ther mal ab lation. Magn Reson Med 2004;52:1302–9. Løkling KE, Skurtveit R, Dyrstad K, et al. Tuning the MR properties of b lood-stable pH-responsi ve paramagnetic liposomes. Int J Pharm 2004;274:75–83. Mulder WJ, Griffionen AW, Strijkers GJ, et al. Magnetic and fluorescent nanopar ticles for multimodality imaging. Nanomedicine 2007;2:307–24. Tran TD, Caruthers SD, Hughes M, et al. Clinical applications of perfluorocarbon nanopar ticles for molecular imaging and tar geted therapeutics. Int J Nanomedicine 2007;2:515–26. Morawski AM, Winter PM, Crowder KC, et al. Targeted nanoparticles for quantitative imaging of sparse molecular epitopes with MRI. Magn Reson Med 2004;51:480–6. Artemov D , Mori N , Ra vi R, Bhujw alla ZM. Magnetic resonance molecular imaging of the HER-2/neu receptor . Cancer Res 2003;63:2723–7. Kobayashi H, Kawamoto S, Saga T, et al. Avidin-dendrimer-(1B4MGd)(254): a tumor targeting therapeutic agent for gadolinium neutron capture therapy on intraperitoneal disseminated tumor which can be monitored by MRI. Bioconjug Chem 2001;12:587–93. Crich SG, Barge A, Battistini E, et al. Magnetic resonance imaging visualization of tar geted cells b y the inter nalization of supramolecular adducts for med betw een a vidin and biotinylated Gd 3+ chelates. J Biol Inorg Chem 2005;10:78–86. Meikle SR, K ench P, Kassiou M, Banati RB . Small animal SPECT and its place in the matrix of molecular imaging technolo gies. Phys Med Biol 2005;50:R45–R61. Kauczor H, Surkau R, Rober ts T. MRI using h yperpolarized noble gases. Eur Radiol 1998;8:820–7. Golman K, Petterson JS. Metabolic imaging and other applications of hyperpolarized 13C1. Acad Radiol 2006;13:932–42. Yu JX, K odibagkar VD, Cui W, Mason RP. 19F: a v ersatile reporter for non-in vasive ph ysiology and phar macology using magnetic resonance. Curr Med Chem 2005;12:819–48. Henkelman RM, Stanizs GJ , Graham SJ . Magnetization transfer in MRI: a review. NMR Biomed 2001;14:57–64. Ward KM, Aletras AH, Balaban RS. A new class of contrast agents for MRI based on proton chemical exchange dependent saturation transfer (CEST). J Magn Reson 2000;143:79–87. Aime S, Car rera C, Castelli DD , et al. Tunable imaging of cells labeled with MRI-P ARACEST agents. Angew Chem Int Ed 2005;44:1813–5. Woods M, Woessner DE, Sherry AD. Paramagnetic lanthanide complexes as PARACEST agents for medical imaging. Chem Soc Rev 2006;35:500–11. Aime S, Castelli DD , Terreno E. Highl y sensiti ve MRI chemical exchange saturation transfer agents using liposomes. Angew Chem Int Ed 2005;44:5513–5.

Chemistry of Molecular Imaging: An Overview

98. Rønn LCB, Olsen M, østergaard S, et al. Identif ication of a neuritogenic ligand of the neural cell adhesion molecule using a combinatorial librar y of synthetic peptides. Nat Biotechnol 1999;17:1000–5. 99. Cwirla SE, Balasubramanian P, Duffin DJ, et al. P eptide agonist of the thrombopoietin receptor as potent as the natural c ytokine. Science 1997;276:1696–9. 100. Wrighton NC, Farrell FX, Chang R, et al. Small peptides as potent mimetics of the protein hor mone er ythropoietin. Science 1996;273:458–64.

303

101. Tweedle MF. Adventures in multi valency – The Har ry S. F ischer memorial lecture CMR 2005, F rance, Evian, F rance. Contrast Media Mol Imaging 2006;1:2–9. 102. Crich SG, Bussolati B, Tei L, et al. Magnetic resonance visualization of tumor angio genesis b y tar geting neural cell adhesion molecules with the highl y sensitive gadolinium-loaded apoferritin probe. Cancer Res 2006;66:9196–201.

20 RADIOCHEMISTRY

OF

PET

HENRY F. VANBROCKLIN, PHD

Molecular imaging enables one to measure nor mal and aberrant molecular processes in vi vo in real time and longitudinally over time. Molecular imaging allows one to visualize, nonin vasively, the phar macokinetic and pharmacodynamic proper ties of ne w therapeutics including gene therap y and stem cell therap y.1–4 The ability to inter rogate these processes relies on discriminating molecular probes coupled with sensiti ve, highresolution scanners to provide quantitative measures of probe localization and concentration o ver time. Advances in instrumentation for the study of both small animals and humans, with improved sensitivity and resolution and the ability to acquire dual modality data sets (e g, positron emission tomo graphy [PET]/computed tomography and PET/magnetic resonance), ha ve enhanced the research en vironment for de velopment, evaluation, and application of new molecular probes.5–7 Advances in animal models of disease and genomic manipulation, including “knock-out, ” “knock-in,” and RNAi models, are providing fertile ground for evaluating and validating new probes.8–10 In addition, new regulatory guidelines for microdosing and f irst-in-man assessment of ne w tracers are dri ving a ne w w ave of chemistry and production of radiotracers for a v ariety of preclinical and clinical applications. 11–15 This g rowth in molecular imaging and the impor t of molecular biolo gy techniques o ver the last decade have fueled the de velopment of ne w probes and the concomitant chemistries that are needed to produce these probes. The a vailability of the shor t-lived positron-emitting isotopes carbon-11, nitro gen-13, oxygen-15, and fluorine-18 has increased and longer lived, “nonstandard” PET isotopes are being made available to more centers. 16 These ne w isotopes are gaining prominence and e xpanding the potential of

304

molecular imaging. In this chapter , PET isotope production and the subsequent application of these isotopes and labeled precursors to prepare radiotracers will be chronicled. The use of radioacti ve isotopes as a means to follow kinetic processes in li ving systems ma y be traced back to the Nobel Laureate Geor g de He vesy17 who studied the mo vement of lead isotopes, lead-210 and lead-212, in plants in 1923 and perfor med the f irst in vivo studies on the metabolism of phosphor us-32 in rats.18 The first human tracer studies were conducted by Blumgart and Yens19 in 1927 measuring b lood flo w with radium C, bismuth-214. The adv ent of the cyclotron in the earl y 1930s b y Er nest La wrence20 paved the w ay for the disco very of man y biolo gically relevant ar tificially produced isotopes including iron59,21 iodine-131,22,23 and technetium-99m 24,25 that have become in valuable nuclides for nuclear molecular imaging and therapy. No one could have predicted how valuable the cyclotron would become to moder n molecular imaging for the production of a v ariety of positron-emitting isotopes, especiall y the shor t-lived isotopes of carbon, nitro gen, o xygen, and fluorine. 26 The availability of small academic and hospital-based cyclotrons combined to help the f ield g row, and no w regional c yclotron f acilities ha ve increased the a vailability of PET tracers mostl y through the production and dispensing of 2-deo xy-2-[ 18F]fluoro-D-glucose ([18F]fluorodeoxyglucose, FDG, F igure 1), an indispensable molecular imaging agent. 27 The k ey to the continued future g rowth in nuclear molecular imaging is the ability to pro vide unique isotope-labeled probes that will fulf ill the promise of molecular imaging to personalize diagnosis and treatment of numerous diseases and disorders.

Radiochemistry of Positron Emission Tomography

305

Figure 1. Fluroine-18-labeled molecular imaging agents. ADAM = N,N-dimethyl-2-(2-amino-4-18F-fluorophenylthio)-benzylamine; FAZA = 1-(5-[18F]fluoro-5-deoxy-α-D-arabinofuranosyl)-2-nitroimidazole; FDOPA = 3,4-dihydroxy-6-fluoro-L-phenylalanine; FECNT = 2β-carbomethoxy-3β-(4-chlorophenyl)-8-(2-[18F]fluoroethyl)nortropane; FHBG = 9-(4-[18F]-fluoro-3-hydroxymethylbutyl)guanine; FIAU = 2ʹ′-[18F]fluoro-2ʹ′-deoxy-1-β-D-arabinofuranosyl-5-iodouracil.

DESIGNING POSITRON-LABELED PROBES There are a number of impor tant aspects related to the physical characteristics of the isotope and the biolo gic characteristics of the labeled compound that must be taken into consideration w hen selecting an appropriate isotope and designing PET radiotracers. Based on these characteristics, the radioisotope must be carefull y matched to the pharmacokinetics of the radiotracer in vi vo and the biologic target or process being inter rogated by the intended radiotracer application. The location of the isotope within the molecule is critical to maintain biolo gic acti vity and molecular inte grity, especiall y to metabolic de gradation. The specif ic activity of the radiotracer , a measure of the physical mass of the tracer associated with the quantity of radioisotope and a function of the isotope half-life, ma y define the utility of the tracer relati ve to toxicity, pharmacologic activity, or receptor b lockade as a result of e xcess nonradioactive mass. Ultimately, the ability to produce the

desired tracer is dictated b y the elemental composition of the tracer molecule and synthetic chemistr y accessibility. These aspects will be addressed in detail herein. Several positron-emitting isotopes that will be discussed throughout this chapter along with some of their physical proper ties are gi ven in Table 1. This is b y no means an e xhaustive list of positron-emitting isotopes that may be used for radiotracer production but represents a cross section of isotopes that are currently available and have been incor porated into molecular tracers. Fur thermore, many of these isotopes ma y be produced on small medical cyclotrons with only 10 to 20 MeV protons. 28–30 The isotopes in Table 1 are brok en into three major groups, lightweight elements, radiohalogens, and radiometals. These isotopes of fer a v ariety of radiolabeling options. Carbon, nitro gen, and o xygen are elements commonl y found in natural medicinal products and synthetic small organic molecules. Isotopic replacement of carbon-11 for carbon-12 in a biomolecule, for example, will produce a positron-emitting v ersion of

306

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 1. PHYSICAL PROPERTIES OF POSITRON-EMITTING RADIONUCLIDESa Nuclide

Decay Modeb

Carbon-11 Nitrogen-13 Oxygen-15

β+ (100%) β+ (100%) β+ (100%)

Max β+ Energy (MeV)

Half-Life

Theoretical Specific (Ci/µmol)c

0.96 1.20 1.7

20.4 min 9.97 min 122.2 s

9,220 18,800 92,600

109.7 96.7 16.2 3.6 4.2

1,710 1,920 195 52,200 31.1

Radiohalogens Fluorine-18 Bromine-75 Bromine-76 Iodine-122 Iodine-124

β+ (97%) β+ (76%) β+ (55%) β+ (78%) β+ (23%)

EC EC EC EC EC

(3%) (24%) (45%) (22%) (77%)

0.69 2.01 3.94 3.21 2.14

Radiometals Cobalt-55 Copper-60 Copper-61 Copper-62 Copper-64 Gallium-66 Gallium-68 Rubidium-82 Yttrium-86 Technetium-94m

β+ (76%) β+ (93%) β+ (61%) β+ (97%) β+ (18%) β+ (56%) β+ (89%) β+ (95%) β+ (32%) β+ (70%)

EC EC EC EC EC EC EC EC EC EC

(24%) (7%) (39%) (23%) (43%) β− (39%) (44%) (11%) (5%) (68%) (30%)

1.50 3.77 1.22 2.91 0.65 4.15 1.90 3.38 3.14 2.44

min min h min d

17.5 h 23.7 min 3.3 h 9.7 min 12.7 h 9.5 h 67.7 min 76.2 s 14.7 h 52 min

179 8,040 950 19,300 245 332 2,750 150,000 213 3,550

a

http://www.nndc.bnl.gov/nudat2/ accessed Mar 15, 2008.

b +

β , positron; β−, electron; EC: electron capture.

c

Calculated from the half-life.

that molecule with the same chemical, ph ysical, and biological characteristics. Halogens, especially fluorine, are increasingl y more common in synthetic biomolecules or in man y cases ma y be added to a molecular structure without signif icantly altering its biolo gic activity. Metals are infrequentl y used as elements in drug discovery as they generally require a chelate structure to complex the metal and attach it to the molecule. Radiometals, ho wever, possess a breadth of radionuclidic properties that are attractive for imaging and therapy. In man y cases, imaging characteristics ha ve been found for small molecule radiometal comple xes and then these imaging proper ties ha ve been honed b y development of analo gs (e g, [Cu]PTSM: Cu(II)-p yruvaldehyde-bis(N4-methylthiosemicarbazone), [Cu] ATSM: Cu(II)-diacetyl-bis(N4-methylthiosemicarbazone), and other bis(thiosemicarbazones)). 31–33 Radiometals are most useful for tagging large biomolecules such as peptides, proteins, antibodies, and aptamers w here the size of the chelated radioisotope does not interfere with the biologic characteristics.34,35 The radioisotopes in Table 1 sho w a range of halflives from 76 seconds ( 82Rb) to 4.2 da ys ( 124I). Although one may be limited in the choice of isotopes by the structure of the molecule being labeled, the range of half-lives provides the opportunity to match the physical lifetime of

the tracer , go verned b y the isotope half-life, to the biologic half-life of the tracer in the body and/or the biologic process or tar get being measured. F or example, one would not generally choose an isotope with a half-life of less than 12 hours for imaging w hole antibody (IgG) distribution as it tak es hours to da ys for the antibody to accumulate at the target and also clear from nontarget tissue. On the other hand, single-chain antibody fragments, about one sixth the mass of a full IgG, accumulate rapidly in tar geted tissues w hile being ef ficiently cleared from the body. Thus, single-chain fragments would be suitable for labeling with shor t-lived radioisotopes. A recent example of this is highlighted b y the de velopment of an anti-HER2 imaging agent from the commercial antibody Herceptin.36 One other aspect that is impor tant regarding the selection of the isotope relates to the potential exposure of the subject to e xcess radiation dose. It is important to balance the half-life so that it is long enough to measure the process of interest y et shor t enough to minimize radiation dose to the subject, as the isotope will remain in the body until it is deca yed or excreted. The isotope half-life determines the working preparation time for the radiotracer . Nito gen-13, o xygen-15, copper-62, rubidium-82, and iodine-122 possess relatively short half-li ves, less than 10 minutes, presenting limited options for their incor poration into biomolecules. These

Radiochemistry of Positron Emission Tomography

isotopes are either used directl y as radiotracers (e g, direct 82 Sr/82Rb generator), rubidium-82 infusion from the incorporated into radiotracers in situ in the accelerator target (eg, 13N-ammonia), or produced in-line as the isotope is eluted from the accelerator (eg, 15O-water) or generator (eg, 62 Cu-PTSM). Radioisotopes with half-li ves g reater than 20 minutes present more options for multistep radiotracer syntheses. Ho wever, there are yield loses associated with both time and the number of synthetic steps in volving the radioactive material. Therefore, the “r ule of thumb” maximum time for a radiosynthesis should be no more than 2 to 3 half-lives, and the isotope should be ideally introduced in the penultimate step of the reaction sequence. The theoretical specif ic acti vity or maximum specific activity, the ratio of the amount of mass associated with the radioisotope reported in Ci/mol or Bq/mol units, for an y isotope is in versely propor tional to the isotope half-life. This means that the shor ter the half-life, the greater the maximum specific activity for a given isotope. High specific activity is an important attribute of a tracer such that the radioprobe par ticipates in the biolo gic process but does not sho w pharmacologic or toxicologic effects, the tenet of De Hevesy’s tracer principle.17 All the positron emitters listed in Table 1 possess high theoretical specific activities > 30 Ci/ µmol; however, the measured specific activities of the labeled molecules are typically 2 to 4 orders of magnitude lo wer due to endo genous isotopes introduced during production and radiosynthesis. Specific activity of the tracer ma y be increased b y careful preparation of the isotope or tracer synthesis. Highl y purified starting materials and clean reaction v essels are essential to minimize introduction of nonradioacti ve isotopes into the process. This is especially true for handling carbon-11 carbon dioxide gas that is commonly produced in the c yclotron target. The air is full of nonradioacti ve carbon-12 carbon dio xide that must be eliminated from the reaction v essels to maximize the specif ic activity of the radiotracer. In spite of g reat effort to reduce isotopic contamination, it is v ery dif ficult to achie ve theoretical specific acti vity especiall y for carbon-11, nitro gen-13, oxygen-15, and fluorine-18 because of their ubiquitous nature. As the f inal radiotracers may contain nonradioactive mass, the specif ic acti vity decreases with deca y highlighting again the importance of time as a factor in tracer preparation and application. It is important to note that if the radiotracer is produced with theoretical specific activity, meaning all tracer molecules are labeled with a radioisotope, then the specif ic acti vity w ould not decrease with time as the deca y of the radiotracer w ould most lik ely destroy the labeled molecule and inherentl y

307

its ability to bind cellular proteins or be an enzyme substrate. One f inal unique ca veat relating to the high specific activity isotopes is that the y permit the preparation and injection of radiolabeled tracers that might otherwise be toxic at higher mass levels. Further examination of Table 1 reveals the decay properties of the isotopes including the positron ener gy and decay mode. The positron range, the maximum distance the positron tra vels before it combines with an electron and annihilates, is propor tional to the positron ener gy and an important factor for image spatial resolution, especiall y in small animal scanners. 37,38 Fluorine-18 and copper-64 possess the lowest positron energy with the least impact on resolution, whereas many of the “nonstandard” isotopes ha ve maximum positron ener gies g reater than 1.5 MeV . Maximum a posteriori (MAP) mathematical algorithms ha ve been applied to cor rect for the positron range and reco ver the lost image resolution 39,40 thereby improving small animal image quality. Given the size of human or gans and the current clinical scanner resolution of 5 to 7 mm, the effect of positron energy on image quality is less signif icant. The lightweight elements, carbon-11, nitrogen-13, and oxygen-15 are all pure positron emitters, and fluorine-18, copper-62, and r ubidium-82 have a > 95% positron deca y branch. A high branching fraction means that all or nearl y all the decays produce positrons that annihilate to gi ve two 511 KeV γ rays, most of which leave the body. The remaining isotopes in Table 1, with the e xception of copper -64, decay b y both electron capture and positron emission. When these isotopes decay, they may give off a spectrum of X-rays and γ rays, some with ener gies that may fall within the energy window of the PET scanner detectors leading to random coincidences or singles e vents that impact image wer quality.38 Additionally, an isotope possessing a lo positron branching fraction ma y impar t more dose to the body, through the alter nate decay pathways, meaning that less of the tracer may be injected into human subjects. Copper-64 has an additional deca y mode, electron ( β-) decay. Unlike gammas, electrons do not leave the body and deposit all their ener gy in the tissue, thereb y adding to the total body radiation dose. Electron emitters are often used as radiotherapeutics. Thus, in addition to being an imaging isotope, copper-64 possesses therapeutic qualities. 41–43 All the isotopes in Table 1 decay to stable daughter isotopes with the e xception of cobalt-55 and bromine-75. Cobalt-55 decays to the 2.7-year iron-55 that in turn decays to stable manganese-55. The majority of the radiation from the iron-55 deca ys are lo w ener gy auger electrons and X-rays that need to be considered for dosimetr y but do not interfere with positron detection. 44 Bromine-75 deca ys to the 120-da y selenium-75 that contributes signif icantly to

308

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

whole body dosimetr y with se veral medium ener gy γ rays in the 100 to 300 k eV range. 45–47 Selenium-75 decays to a stable daughter arsenic-75. The last impor tant aspects for radiophar maceutical design are chemical accessibility and the position of the label in the chemical str ucture. The remaining por tion of this chapter will be de voted to the discussion of chemical accessibility to labeled compounds. Briefl y, as mentioned above, the introduction of the radiolabel must be rapid and efficient otherwise the rate of nuclide deca y may compromise yield and specif ic activity. The position of the label is important, especiall y relati ve to compound metabolism. Untoward metabolism may compromise image quality and ultimately the infor mation retrie ved from the image b y reducing probe concentration, reducing specif ic acti vity, and complicating kinetic anal ysis with unw anted background counts. In some cases, the labeled metabolites themselves may be receptor ligands or enzyme substrates that will possess their own distribution that is indistinguishable from the parent radiophar maceutical. This case is exemplified by the metabolism of the high affinity serotonergic 5HT 1A ligand [ 11C]WAY100635 shown in F igure 2. WAY100635 w as f irst labeled in the metho xy position (a, Figure 2).48,49 However, when this molecule was injected into primates and humans, one of the major metabolites was [11C]WAY100634, a ligand with high af finity not onl y for the serotonergic 5HT 1A receptor but also for the α1-adrenergic receptors. 50 This untoward metabolism compromised kinetic analysis of the compound in the brain as the distribution of the tw o ligands w as impossib le to dif ferentiate. When label position w as mo ved to the carbon yl g roup 11 C]cyclohexyl(b, Figure 2), the labeled metabolite, [ methylketone, was eliminated from the body and did not cross the b lood brain bar rier. This example also illustrates the importance of specific activity. If the specific activity of the [ 11C]carbonyl-WAY100635 is lo w, then the unlabeled metabolite, WAY100634, could interfere with the binding of the labeled WAY100635 to the 5HT 1A receptors and negatively impact the anal ysis of receptor density and the volume of distribution.

Nature has pro vided the near -perfect set of positronemitting isotopes for molecular imaging applications. The elements that are commonl y found in biomolecules, 12C, 14 N, 16O, and 19F have corresponding nearly pure positronemitting isotopes with shor t half-lives, low positron energies, and high specif ic activity. Over the last tw o to three decades the availability of these PET isotopes from small medical cyclotrons and the methods to label probes with these isotopes ha ve increased dramaticall y. The need for new labeled tracers, the de velopment of ne w synthetic strategies, and the increased a vailability of nonstandard isotopes are adv ancing the f ield of radiophar maceutical chemistry and in-tur n e xpanding the boundaries and potential of molecular imaging.

CARBON-11 Production of Carbon-11 There are several nuclear reactions that have been used to produce carbon-11 including 14N(p,α)11C, 10B(d,n)11C, and 12C(γ,n)11C (nuclear reaction convention: stable stock isotope [accelerated par ticle, particle(s) ejected from the nucleus] new positron isotope; p = proton, d = deuteron, n = neutron, α = alpha, γ = gamma ray). These and other reaction pathw ays ha ve been e xtensively re viewed.51–53 The primar y reaction for routine preparation of carbon11 is the 14N(p,α)11C on natural nitrogen gas. The choice of this reaction pathw ay is e vident for a number of reasons when all the reaction pathw ays are compared. The 14 N(p,α) reaction occurs above 3 MeV with a high cross section thereby yielding curie quantities of high specif ic activity carbon-11 precursors. This reaction pathw ay avoids the use of carbon stock material, solid target material, and a deuteron beam that are required for the other nuclear reactions. Nitrogen-14 is 99.6% abundant in natural nitrogen gas so enrichment is not required. Addition of a trace amount of o xygen gas (< 2%) or h ydrogen gas (5–10%) with the nitro gen-14 af fords the in-tar get production of [ 11C]carbon dioxide ( 11CO2) or [ 11C]methane

Figure 2. The metabolism of the serotonergic 5HT1A radioligand WAY100635 into WAY100634 and cyclohexylmethylketone. WAY100635 has been labeled in two positions (a: methoxy carbon and b: carbonyl carbon) with carbon-11.

Radiochemistry of Positron Emission Tomography

(11CH4), respectively, during the irradiation. The purity of the tar get gases, ho wever, is critical as e ven the purest source gas has enough carbon to reduce the specif ic activity 100-fold.

Carbon-11-labeled Precursors In spite of the shor t 20.4-minute half-life, carbon-11 is the most v ersatile positron-emitting radioisotope with the g reatest number of labeled precursors that ma y be incorporated into the synthetic production of radiotracers. F igure 3 sho ws the v ariety of carbon-11 synthons that ha ve been for med for the synthesis of labeled probes. The tw o primar y precursors, 11CO2 and 11CH4 produced in situ in the cyclotron target, are shown in the highlighted rectangles. F rom these, a number of secondary precursors are possib le. The two key secondar y precursors, [ 11C]methyl iodide, 11CH3I, and [ 11C]carbon monoxide, 11CO, are highlighted in the circles. These secondary precursors may in tur n be used to for m additional labeled synthons or ma y be used in tracer synthesis. An in valuable compendium of carbon-11- and fluorine-18-labeled precursors and radiotracers with over 1,400 references has been compiled b y Dr . Ren Iwata at the Cyclotron and Radioisotope Center, Tohoku University.54 There are tw o methods routinel y used for the production of 11CH3I, a liquid phase reaction and a gas phase reaction. In the liquid phase, con version of 11CO2 to [11C]methanol ( 11CH3OH) b y LiAlH 4 reduction is followed by the addition of h ydriodic acid. 55,56 Upon heating, the gaseous 11CH3I is distilled into the subsequent reaction vessel. In the gas phase reaction, 11CH4 directly from the c yclotron or b y in-line catal ytic hydrogenation

309

of 11CO2 is reacted with iodine v apor in a heated quar tz tube to give the 11CH3I. Single pass 57 and recirculating 58 systems ha ve been de veloped to optimize the yield of 11 CH3I and provide a reliable reproducible source of this valuable intermediate. The specif ic activity of the liquid phase reaction is lo wer than the gas phase as a result of dissolved nonradioacti ve CO 2 carrier in the LiAlH 4, while the 11CH3I radiochemical yields tend to be higher for the liquid phase reaction. Automated synthesis units for both methods of 11CH3I production are commercially available. [11C]Methyl iodide is the synthetic stepping stone to many other v aluable labeling synthons sho wn in Figure 3 including [ 11C]methyl lithium ( 11CH3Li),59 [11C]methyl azide ( 11CH3N3), [ 11C]methyl cuprates ( 11CH3CuLi),60–63 [11C]nitromethane ( 11CH3NO2),64–66 [11C]methyl triflate (11CH3OTf),67,68 [11C]methyl magnesium bromide (11CH3MgBr) for Grignard reactions, 69 1-aza-5-stanna-5[11C]methyltricyclo [3.3.3.0]-undecane for Stille carboncarbon (C-C) bond couplings, 70 as w ell as [ 11C]methyl triphenylphosphoranes (Ph 3P11CH2),71 and arsenines (Ph3As11CH2) for Wittig alkene synthesis.72 [11C]Methyl triflate, a more reactive methylating agent than 11CH3I, offers expanded solvent choices and higher yields for alk ylation reactions at lo wer temperatures and shor ter reaction times.73–75 The gas phase preparation of 11CH3OTf is facilitated by passing 11CH3I or 11CH3Br, from 11CH4, over silver triflate heated at 200°C. Another series of synthons deri ved from 11CO2 are 76–80 the higher order aliphatic and benzyl halides. [11C]Carboxylation of alkyl and benzyl Grignard reagents gives the cor responding labeled acid magnesium bromides that are fur ther transfor med to the analo gous halide through the alcohol. Acid chlorides are also available through the acid magnesium bromide. 81–83

Figure 3. Carbon-11-labeled precursors. The primary precursors from the cyclotron target are in the rectangular boxes, and the major secondary precursors are in the circles.

310

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Hydrogen [ 11C]cyanide (H 11CN), [ 11C]cyanogen bromide ( 11CNBr), [ 11C]diazomethane 11CH2N2,84 and [11C]phosgene ( 11COCl2) are all deri ved from 11CH4. Catalytic amination of 11CH4 over platinum af fords the multipurpose precursor H 11CN.85–88 Further reaction of the H11CN with bromine 89,90 or perbromide 91 provides (11CNBr). The carbon yl g roup, present in man y biomolecules, may be introduced through [ 11C]COCl2 produced by exhaustive chlorination of 11CH4 to [ 11C]carbon tetrachloride ( 11CCl4) follo wed b y iron catal yzed o xidation92,93 or chlorination of [ 11C]CO b y platinum chloride or chlorine gas.94,95 The zinc, 96 charcoal,97 or mol ybdenum98 mediated reduction of 11CO2 is the primary source of 11CO for carbonylation reactions. [11C]Carbon monoxide, long known for its limited solubility and reacti vity,97–99 has, over the last decade, become an important secondary intermediate for the production of at least 10 classes of carbon yl containing compounds. 100 Improved manipulation using solid phase supports to trap the 11CO, and design of gasliquid phase microreactors has enab led the de velopment of key reaction pathw ays, using palladium and selenium catalysts, leading to several new compounds.100,101

Carbon-11-labeled Compounds The wide variety of synthetic precursors a vailable for carbon-11 labeling pro vides excellent synthetic accessibility to a breadth of molecular imaging radiotracers. The two

Figure 4.

major challenges to successful production of these probes are time and specif ic acti vity. To reduce synthesis time, many of the conversions shown in Figure 3 are perfor med as single pass in-line flo w transformations. These are performed in closed systems with high-quality reagents and gasses to eliminate ubiquitous CO 2 infiltration and endogenous carbon load. 102 The two main reaction types, N , O, or S alkylations or C-C bond for mation, that are achie vable with 11CH3I or11CH3OTf are sho wn in F igure 4. 75 A host of amines, alcohols, thiols, anilines, phenols, among others, ha ve provided a convenient route for the preparation of a v ariety of carbon-11-labeled probes. Examples of the nucleophilic reaction with all three heteroatoms types, sho wn in 11 Figure 4, include S-[ C-methyl]methionine,103,104 [11C]choline,105–107 and [ 11C]raclopride.73 These reactions are conducted in solution phase, in small bore reactor loops,73,108,109 on solid support cartridges104,107,110,111 or both loops and solid support.112 As seen in Figure 4, methylation reactions are selecti ve and ma y be car ried out in the presence of other heteroatoms in the precursor molecule. Although heteroatom meth ylation is f acile, it ma y not al ways pro vide a metabolicall y stab le probe. C-C bonds are less lik ely to be metabolicall y clea ved and therefore have been the focus of recent efforts to improve carbon-11 probe stability . Rapid palladium catal yzed cross-coupling reactions represent a signif icant advance in C-C bond for mation reactions with carbon-11. Three major cross-coupling reactions leading to carbon-11labeled tracers are sho wn in F igure 4. Stille coupling

Heteroatom alkylation and C-C bond forming reactions of [11C]methyl iodide.

Radiochemistry of Positron Emission Tomography

involves the reaction of 11CH3I with an ar yl, alkyl, vinyl, or alkynal tributylstannane,70,113 whereas Suzuki coupling entails similar chemistry with aryl, vinyl, or alkyl boronic acid substrates. 113,114 A recent re view 75 shows a number of probes labeled b y these cross-coupling reactions. The Sonogashira reaction couples 11CH3I to terminal alkynes. The preparation of an estradiol analo g is highlighted in Figure 4.115 Additional C-C bond for ming reactions are kno wn using precursors described in Figure 3. The Wittig reaction with carbon-11-labeled triphen ylphosphine and arsenine ylides gi ves alk ene compounds. 71,72 Metal-mediated [11C]cyanations with ar yl, vin yl, and aliphatic bromides with H11CN have also been reported.116–118 Carbon-11-labeled CO is catching up to meth yl iodide as a key secondary precursor for the preparation of several classes of compounds. As seen in F igure 5, the 11 CO synthon may be inser ted, using palladium and selenium catal ytic inter mediates, into a v ariety of small organic structures providing a palette of carbonyl containing probes. Although not all these reaction pathways have been fully explored, these new reactions have enabled the production of compound classes that w ere heretofore unobtainable with carbon-11. 100

311

There is no doubt that carbon-11 will continue to play a major role in the future of molecular imaging. Although carbon-11 is e xtremely v ersatile, its shor t half-life limits the availability of compounds outside of institutions with a c yclotron. On the other hand , carbon-11 is so attractive as a labeling isotope because most if not all biomolecules contain carbon, and isotopic labeling renders the compound indistinguishable from the nonlabeled molecule. Ne w labeled precursors will e xpand the carbon-11 chemical footprint and will mak e possible the fur ther preparation of labeled natural and man-made phar maceuticals for pharmacokinetic and phar macodynamic studies and tracers for monitoring physiologic processes as well as staging and monitoring of disease.

NITROGEN-13 AND OXYGEN-15 Production of Nitrogen-13 and Oxygen-15 Nitrogen was one of the f irst positron-emitting nuclides to be produced.119 There are a number of nitrogen-13 producing nuclear reaction pathw ays, but the most widel y used reactions are 12C(d,n)13N and 16O(p,α)13N.120–123 Depending on the star ting tar get material, the in-tar get products include nitro gen gas, [ 13N]N2, nitrates and nitrites, [ 13N]NOx (x = 2 or 3), and ammonia [13N]NH3.124–127 The proton irradiation of 16O is the most common method for making nitro gen-13 products with an o xygen gas or a w ater tar get. Addition of a radical scavenger, such as ethanol, to the w ater promotes the intarget production of [ 13N]NH3.128,129 The most common nuclear reactions for producing 14 N(d,n)15O, 15N(p,n)15O, and oxygen-15 include 16 O(p,pn)15O.130–132 Not all the small cyclotrons are capable of accelerating deuterons, so the 14N(d,n) on natural abundance nitro gen gas is limited to these sites. The 15 N(p,n) reaction on enriched nitro gen gas is more frequently used for the preparation of o xygen-15 species although nitrogen-15 gas must be enriched and is more costly. Addition of 0.1 to 4% O2 gas to the nitrogen target gives 15O-16O, addition of 0.1 to 2% CO 2 yields C16O15O,133,134 and addition of 5% H 2 produces H 215O in the target.134,135 Of these, 15O-16O is most often produced.

Nitrogen-13 and Oxygen-15 Precursors and Chemistry

Figure 5. Radioprobes from [11C]carbon monoxide. The R groups within a given molecule do not have to be the same.

Given the short 2 minute half-life of oxygen-15, the practical chemical transfor mation of labeled precursors into molecular imaging probes is limited. Most of the tracers

312

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

that are routinel y used for preclinical or clinical studies are either taken directly from the cyclotron or in-line conversions of the c yclotron products. The [ 15O]O2 or [15O]CO2 from the target is converted into [15O]H2O134,136 and [ 15O]butanol137 to measure blood flow and [ 15O]CO2 and [15O]CO133 to measure blood volume. Nitrogen-13 with a ~10-minute half-life af fords a little more time to work with the isotope, and it has been creatively incorporated into a limited number of labeled molecular radioprobes. The myocardial perfusion tracer [13N]NH3, produced in tar get as described abo ve, has also been used as a precursor for the preparation of the anticancer dr ugs [ 13N]cisplatin, ( 13NH3)2PtCl2,138 and the nitrosourea, 1-(2-chloroeth yl)-3-cyclohexyl-1[13N]nitrosourea, CCNU.139 An alternate path to BCNU, 1,3-bis-(2-chloroethyl)-3-cyclohexyl-1[13N]nitrosourea from [ 13N]nitrate w as de vised to improve the yields. 140 Antibiotics, [ 13N]streptozotocin, and [ 13N]nitrosocarbaryl have also been labeled b y the [13N]nitrate route. 141 Additionally, bioactive amines and amino acids, [ 13N]dopamine, [ 13N]putrescine, and [13N]γ-aminobutyric acid were produced by [13N]amination of organoborane polymers.142 Relative to the longer -lived positron-emitting isotopes, o xygen-15 and nitro gen-13 are more lik ely to have a suppor ting role in the molecular imaging applications. The challenges of time and the re gulatory requirements for radiotracers that will be administered to humans will limit the de velopment of probes labeled with these isotopes.

FLUORINE-18 Production of Fluorine-18 Similar to carbon, nitrogen, and oxygen, there are multiple (> 20) kno wn nuclear reaction pathw ays to produce fluorine-18.143 The two most common reactions are 18O(p,n)18F and 20Ne(d,α)18F. The target material for [ 18F]fluoride production b y the o xygen-18 pathw ay is generall y enriched oxygen-18 water although oxygen-18 and neon-20 gas may also be used.144–146 The natural abundance of oxygen-18 is 0.2% so enrichment is necessar y to use this isotope as a stock target material, whereas neon-20 with 90.5% natural abundance does not require enrichment. The irradiation of an oxygen-18 water target for 1 to 2 hours produces 5to 10 Ci (185–370 GBq) of [ 18F]fluoride ion with a specif ic activity of 10 3 to 10 4 Ci/mmol. The o xygen-18 gas and neon-20 gas tar get require a multiple step process to produce [18F]fluoride ion. The enriched [18O]O2 or [20Ne]neon

gas is loaded into the tar get and irradiated with protons or deuterons, respecti vely. At the end of bombardment, the gasses are recovered and the tar gets are rinsed with w ater to collect the [ 18F]fluoride ion. The tar get must be thoroughly dried before the ne xt ir radiation. The yield of [18F]fluoride ion is > 2 fold for oxygen-18 gas versus neon20 as e xpected based on the dif ferences in the reaction cross-sections, 700 mb for protons on o xygen v ersus 115 mb for deuterons on neon. 144,147,148 The [ 18F]fluoride ion in the w ater may be used directl y in the nucleophilic synthesis of labeled products b y evaporating the w ater in the presence of a base and phase transfer catal yst or concentrating using an anion exchange cartridge. Elemental [18F]fluorine gas ([ 18F]F2; 18F-19F) may be used for electrophilic fluorination reactions. It is produced from o xygen-18 or neon-20 gas. As with the gaseous tar gets described abo ve, the fluorine-18 will adhere to the tar get walls. Removal of the [ 18F]F2 from the target requires the introduction of elemental fluorine19 gas meaning that the specif ic acti vity of the [ 18F]F2 will be ~3 to 4 orders of magnitude less than [18F]fluoride ion, typically 1 to 10 Ci/mmol (37–370 MBq/mmol). 149 A single deuteron irradiation of 0.1 to 2% F2 gas in a high pressure neon tar get produces [ 18F]F2.147 The 18 O(p,n) reaction in volves a “doub le shoot” on the target.150 The tar get is f irst f illed with high pressure enriched 18O2 gas and ir radiated for an hour or more. After this f irst ir radiation, the 18O2 gas is cr yogenically recovered from the target leaving the fluorine-18 adhered to the target walls. Carrier fluorine gas in argon is loaded into the target. A second irradiation for 10 minutes f acilitates the scr ubbing of the fluorine-18 from the w alls by the F 2 gas and the for mation of 18F-19F gas b y isotopic exchange. As a result of the cross-section differences, the 18 19 F- F gas yield from the Ne(d ,α) is ~300mCi for an hour radiation versus 0.6 to 1 Ci for the 18O(p,n) reaction. The other attracti ve feature for the 18O reaction is the recovery of the 18O2 gas. Very little material is consumed during the target irradiation, so the 18O2 does not need to be replenished for several irradiations thereby conserving an enriched stock material that costs $200 to $400/L.

Fluorine-18-labeled Precursors Compared with carbon-11, the number of fluorine18-labeled precursors that are applied in radiotracer synthesis is less. The difference may be that fluorine-18 is incorporated directl y into the molecule in addition to being introduced through labeled prosthetics. The fluorine-18 half-life is nearl y 6 times longer than carbon-11,

Radiochemistry of Positron Emission Tomography

affording more time to perform complex or lengthy chemical syntheses and purif ication of the desired molecules. Multistep syntheses with the isotope ma y be under taken with the onl y concer n about the loss of acti vity due to chemical yield. Fluorine-18 precursors and labeled compounds may be transported away from the cyclotron facility thereb y e xpanding the a vailability of k ey molecular imaging agents. The electrophilic fluorination precursors are shown in Figure 6. The primar y precursor [ 18F]F2 is directly available from the cyclotron. Reaction of [18F]F2 through a column of acetic acid and potassium acetate gi ves acetyl hypofluorite (CH 3COO18F), a mild fluorinating agent.151,152 Perchlorofluoride ( 18FClO3) is produced b y reaction of KClO3 and [18F]F2.153 Only 50% of the [18F]fluorine is incor porated in the 18FClO3 as the [ 18F]F2 is split between the 18FClO3 and K18F. Heating [18F]F2 in the presence of Xe gas yields labeled x enon difluoride (Xe 18F19F, [18F]XeF2).154 The N-[ 18F]fluoropyridinium triflate is prepared by reacting [ 18F]F2 with N-trimethylsilylpyridinium triflate.155 These reactive precursors in turn may be used to incorporate fluorine-18 into a radiotracer or a synthetic intermediate.

Figure 6.

Fluorine-18-labeled precursors from [18F]F2 gas.

Figure 7.

Fluorine-18-labeled precursors from [18F]fluoride ion.

313

There are a fe w more reacti ve intermediates and precursors a vailable from [ 18F]fluoride ion (F igure 7). The transfer of the [ 18F]fluoride ion from the aqueous tar get water into an organic solvent where the nucleophilic reactions will tak e place is f acilitated by phase transfer catalysts, kryptofix 2.2.2, and tetrabutylammonium hydroxide. Evaporation of aqueous [18F]fluoride ion in the presence of either potassium carbonate/kr yptofix or tetrabutylammonium h ydroxide, among a fe w others, will gi ve K[18F]/kryptofix or tetrab utylammonium [ 18F]fluoride residue that is solub le in polar aprotic solv ents (typically DMSO, DMF, THF, dichloromethane, acetonitrile, or thodichlorobenzene). The reactive fluoride ion is a vailable to participate in nucleophilic aromatic and aliphatic substitution reactions with activated leaving groups to produce the labeled inter mediates/precursors shown in F igure 7. One interesting secondar y precursor that ma y be produced through [18F]fluoride ion is [ 18F]F2 gas. Bergman and Solin de veloped a method to electrolytically convert [18F]fluoride ion into [ 18F]F2 gas.156 The aqueous [ 18F]fluoride ion w as e vaporated in the presence of kryptofix and acetonitrile under a stream of nitrogen at 80°C. After dr ying, meth yliodode in acetonitrile was added to the heated v essel, and [ 18F]fluoromethane (CH318F) was produced. The CH318F gas was flushed from the v essel, purif ied through a gas chromatography column, and transfer red to a Teflon discharge chamber along with [ 19F]F2 gas in neon. A 20 to 30 kV po wer suppl y w as used to dis charge the chamber thereb y producing [ 18F]F2 gas. The specific activity ranged from 100 to 200 Ci/mmol, 2 to 3

314

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

orders of magnitude better specific activity than that of in-target produced [ 18F]F2 gas.

Fluorine-18 Molecular Imaging Agents There have been thousands of compounds labeled with fluorine-18 either b y nucleophilic substitution or b y electrophilic fluorination. There ha ve been mono graphs,157 reviews, and a compendium54 that chronicle the production of these compounds. There are far too many reactions and far too many classes of compounds that have been labeled to present a concise discussion of these tracers in a fe w paragraphs. A select, representati ve g roup of fluorine18-labeled compounds is presented in F igure 1. The most widely produced and distributed positron-emitting radiotracer is FDG. 27,158 FDG is principall y made through the nucleophilic [ 18F]fluoride route, ho wever, it has been produced from [ 18F]F2 gas and acetyl hypo[18F]fluorite.159 The nucleophilic route gi ves the one desired compound FDG, whereas the electrophilic route gives a mixture of FDG and [18F]fluorodeoxymanose that must be chromato graphically separated. Although specif ic activity is not a biochemical concern for this enzymatically accumulated tracer, the specific activity of FDG is considerab ly higher w hen labeled with [18F]fluoride ion. There are some tracers that are v ery difficult to produce from [18F]fluoride ion, often involving lengthy/multiple synthetic steps handling radioacti vity.160–162 Two of these tracers are sho wn in F igure 1, FDOP A (3,4-dih ydroxy-6-fluoro-L-phenylalanine) and FMT (6-fluorometa-tyrosine). The production of these tracers is a two-step reaction for the electrophilic route and se veral steps for the nucleophilic route. The site-specif ic electrophilic aromatic substitution reaction of [ 18F]F2 gas with a trialk ylstannane precursor gi ves a [ 18F]fluoroaryl intermediate that is deprotected with h ydrogen bromide to yield the desired FDOP A163–165 or FMT.166–168 These two tracers are, lik e FDG, enzyme substrates and converted into products that are retained in the cell or neuron. FDOPA and FMT are accumulated in the striatal regions of the brain w here amino acid decarbo xylase is present and active. The remaining tracers sho wn in F igure 1 are all produced by nucleophilic substitution reactions or a sequence of reactions that involves one of the labeled precursors that were shown in Figure 7. FES (16 α-[18F]fluoroestradiol) is a steroid with high af finity for estradiol receptors that are upregulated in hor mone-dependent breast cancer .169,170 [18F]fluorocholine is an analo g of choline that is tak en up

in cells and phosphorylated, the first step in cell membrane synthesis.171,172 FLT (3ʹ′-deoxy-3ʹ′-[18F]fluorothymidine), a thymidine analo g, is a mark er of cell proliferation. 173–175 FAZA (1-(5-[ 18F]fluoro-5-deoxy-α-D-arabinofuranosyl)2-nitroimidazole) and FMISO ([ 18F]fluoromisonidazole) are hypoxic tissue sensitive ligands.176–178 They accumulate in tissue based on the intracellular o xygen concentration. FIAU (2 ʹ′-[18F]fluoro-2ʹ′-deoxy-1-β-D-arabinofuranosyl-518 F]-fluoro-3-hydroxiodouracil) and FHBG (9-(4-[ ymethylbutyl)guanine) are th ymidine and ganc yclovir analogs, respecti vely, that are substrates for HSVtk, an enzyme that is used as a repor ter system for gene e xpression imaging. 179–184 FACBC (1-amino-3-[ 18F]fluorocyclobutane-1-carboxylic acid) is an unnatural amino acid 185,186 that is accumulated in malignant tumor tissue. Finally, FECNT (2 β-carbomethoxy-3β-(4-chlorophenyl)-8(2-[18F]fluoroethyl)nortropane) and 4-[ 18F]ADAM (N ,Ndimethyl-2-(2-amino-4-18F-fluorophenylthio)-benzylamine) are tw o representati ve ligands from a lar ge g roup of labeled probes that ha ve been produced to understand the action and interaction of neurotransmitter systems in the brain. FECNT is a dopaminergic transporter ligand,187 and 4-[18F]ADAM is a serotonergic transporter ligand.188 Throughout the development of the radiopharmaceutical chemistr y f ield, man y techniques and reactions ha ve been borrowed from the rich organic and medicinal chemistry literature to de velop ne w labeled probes. A recent example of one such set of reactions is sho wn in Figure 8. “Click chemistry” was developed in the Shar pless laboratory as means of bringing tw o halv es of a molecule together in close pro ximity, such as in a receptor binding pocket, where they would be “clicked” together to for m a new molecule.189,190 The resulting molecule would possess avidity for the receptor binding pock et that brought them together. The “click” reaction is based on the Huisgen 1,3 dipolar c ycloaddition of ter minal azides and ter minal alkynes.191,192 This reaction may be carried out in aqueous media. It is a mild reaction requiring Cu(I) to catal yze the cycloaddition reaction without the need to protect other functional groups in the molecule. The radiopharmaceutical community quickl y sa w the adv antage of this f acile reaction for fluorine-18 labeling of peptides and other water-soluble molecules. Several groups have already published initial results on the application of “click” chemistry to label folate or peptides. 193–197 As seen in F igure 8, the radiolabel ma y be incor porated into either the alk yne or the azide. The fluorine-18 is introduced onto the opposite end of the alk yl chain. The resulting [ 18F]fluoro-azide or-alkyne is subsequentl y reacted with the cor responding alkyne or azide on the biomolecule, respecti vely, to

Radiochemistry of Positron Emission Tomography

Figure 8.

315

Fluorine-18 “Click” chemistry.

produce the desired tracer . In one e xample, a series of [18F]fluoroalkynes w as produced b y heating the cor responding tosylate with K 18F in the presence of kr yptofix for 10 minutes. The resulting [ 18F]fluoroalkynes were distilled and added to the peptide with CuI (copper iodide), a base (diisoprop ylethylamine), and ascorbate to preser ve the o xidation state of the copper .195 The subsequent cycloaddition step takes place at room temperature for 10 minutes. Labeled peptide yields greater than 80 to 85% are achievable in 30 to 45 minutes. This reaction has provided a new pathway to label peptides and other molecules that may be susceptible to harsh reaction conditions.

RADIOMETALS Generator Produced Positron-emitting Isotopes Generator systems that produce v ery shor t-lived (< 30 min) radionuclides for diagnostic imaging studies provide a con venient source of these radioisotopes without the need for a dedicated , on-site c yclotron. Ho wever, the positron-emitting isotope generators (e g, 82Sr/82Rb, 68 Ge/68Ga, 62Zn/62Cu, 122Xe/122I) de veloped to date ha ve fallen shor t of the success of the 99Mo/99mTc generator system for single photon imaging. 31 The four well-known positron-emitting generator systems are sho wn in Table 2 along with the parent half-life, nuclear reaction(s) for the parent production, and the abundance of the tar get material that is used to produce the parent. Lik e all generator systems, the longer half-life parent isotope deca ys to the shorter-lived daughter isotope. In general, most of the positron systems suf fer from limited (underdeveloped) daughter chemistry, costly parent

production, high positron ener gy, and either too shor t a parent or daughter half-life. In contrast, some of the positive attributes of the 62Cu and 122I generators include ready availability without c yclotron and maintenance personnel costs, signif icantly lo wer radiation dose to patients, improved image quality with high photon flux, the capability of performing repeat studies or rapid sequential studies (3.6 min 122I has some adv antages over 10 min 13N or 8.7 min 62Cu), and the possibility of perfor ming multiple radiotracer procedures (ie, tw o or more dif ferent compounds in one study) within a shor t time frame. Rubidium-82, 76 seconds, is the decay-product of the 25-day strontium-82. It also possesses a high positron branching ratio. The v ery shor t half-life eliminates the possibility of using this isotope to tag biomolecules. The rubidium-82 is eluted from the commerciall y a vailable generator system with normal saline directly into the subject intravenous system. Rubidium behaves like an analog of potassium and is used for myocardial blood flow studies.198 It has also been used to evaluate the integrity of the blood brain barrier. Zinc-62 is the parent of the 9.7 minute pure positronemitter copper -62. The half-life of zinc-62 is 9.1 hours meaning that the useful life of the generator is onl y a few days. The copper -62 has been incor porated into the bis(thiosemicarbazone) ligands PTSM and ATSM shown in Figure 9. The 62Cu-PTSM has been applied to m yocardial b lood flo w studies. 33,199,200 62Cu-ATSM pro vides a measure of tissue hypoxia.201 Gallium-68 is one of the longest-lived daughter products from a PET generator system, 68 minutes. The parent isotope germanium-68 has a half-life of 275 days that means the generator ma y last a y ear or more before it needs to be replaced. Gallium-68 is eluted from the

316

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 2. GENERATOR-PRODUCED POSITRON-EMITTING ISOTOPES Daughter Isotope

Parent Isotope

Copper-62

62

Gallium-68

68

Rubidium-82

82

Iodine-122

122

Zn

Ge Sr Xe

Parent Half-life

Parent Production Reaction

Target Nuclide Natural Abundance (%)a

References

63

69.2

Fujibayashi and colleagues,249 Robinson and colleagues,250 Yagi and Kondo251

275 d

69

60.1

Loc’h and colleagues,252 Yano and Anger253

25 d

nat

25 d

127

9.1 h

Cu(p,2n)

Ga(p,2n) Mo(p,xn) I(p,xn)

100

Yano and colleagues254

100

Lagunas-Solar and colleagues209

a

http://www.nndc.bnl.gov/nudat2/ accessed Mar 15, 2008.

Accelerator-produced Radiometals

Figure 9. Structure of Cu-PTSM (Cu(II)-pyruvaldehyde-bis (N4-methylthiosemicarbazone) and Cu- ATSM (Cu(II)-diacetylbis (N4-methylthiosemicarbazone).

generator with a milliliter or more of concentrated hydrochloric acid that must be evaporated before reacting with a chelate. Gallium-68 has been used most often as a DO TA (1,4,7,10-tetraazac yclododecane-1,4,7,10tetraacetic acid) and DTP A (dieth ylenetriaminepentaacetic acid) chelate (see F igure 10) for labeling somatostatin, a tumor imaging agent or other antibodies or proteins.202–204 The 122Xe/122I system has some adv antages over the other generator systems in that iodine chemistr y is well developed; iodine can be directl y coupled to biolo gic substrates without the need for either bifunctional chelates or e xtensive synthesis; and iodine-labeled substrates can retain the biolo gic characteristics of the parent molecule. The 3.6-minute half-life does, ho wever, present some unique challenges for the labeling radiochemistry.205,206 The most practical reaction for the production of 122Xe is the bombardment of 127I with high energy protons. The f irst 122Xe/122I generator w as described b y the g roup at Brookha ven National Laboratory.207,208 In an e xtension of this preliminar y work, an operational 122Xe/122I generator system w as cooperatively developed at Lawrence Berkeley National Laboratory and Crock er Nuclear Laborator y.209,210 Rapid 122I-for-127I e xchange radioiodination w as demonstrated211,212 in the labeling of N ,N,Nʹ′-trimethylNʹ′-[2-hydroxy-3-methyl-5-iodobenzyl]-1,3-propanediamine (HIPDM), a cerebral perfusion tracer.213

The availability of positron-emitting radiometals expands the range of potential radiotracers be yond the shor t halflives of C, N, O, and F. As shown in Table 1, the radiometals have a breadth of half-li ves from 76 seconds to 17.5 hours. It is possib le to produce the subset of radiometals listed in Table 3 using a small medical c yclotron, meaning that there is potential for future a vailability.28,29 With the exception of cobalt-55, all the other repor ted nuclear reactions in Table 3 have target stock materials with a natural abundance of less than 50%. The natural abundance of nickel-64 and nickel-61 is less than 1.5%. This means that enriched targets will be needed to produce sufficient yields of these radioisotopes. F or e xample, the cost of nickel-64 for the production of copper -64 is ~$18/mg. A typical target is 40 mg or about $720 w orth of nickel-64. Following irradiation, the copper-64 is separated from the nickel-64 and ~85 to 95% of the nick el is reco vered for reuse.214,215 Recycling the expensive target material helps reduce the cost of isotope production. In order to har ness the imaging proper ties of these radiometals, one has to de velop a means of caging the radiometal ion in a chelate and tagging it to the desired biomolecule. This is accomplished using bifunctional chelate ligands such as those shown in Figure 10. DOTA and DTPA are v ersatile/promiscuous chelators as the y are kno wn to bind gallium, yttrium, and copper albeit with dif ferent stability constants. The remaining chelates in Figure 10, TETA (1,4,8,11-tetraazacyclotetradecane-1,4,8,11-tetraacetic acid) and B AT (bromoacetamidobenzyl-TET A), both chelate copper isotopes. The DO TA-NHS (DO TA-Nhydroxysuccinimide), TETA-NHS (TETA-N-hydroxysuccinimide), and B AT chelates all ha ve pendent g roups that will react with amines in peptides, proteins, and antibodies to give tagged macromolecules. There are three accelerator-produced isotopes of copper, copper-60, copper-61, and copper-64, which may be produced in high yield. All three isotopes are available by

Radiochemistry of Positron Emission Tomography

317

Figure 10. Radiometal Chelates. BAT = Bromoacetamidobenzyl-TETA; DOTA = 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid; DTPA = diethylenetriaminepentaacetic acid; TETA = 1,4,8,11-tetraazacyclotetradecane-1,4,8,11-tetraacetic acid. Table 3. ACCELERATOR-PRODUCED RADIOMETAL ISOTOPES Nuclide Cobalt-55

Copper-60 Copper-61 Copper-64

Gallium-66 Yttrium-86 Technetium-94m

Production Reaction

Target Nuclide Natural Abundance (%)a

58

68.1

54

Fe(d,n) 56 Fe(p,2n) 60 Ni(p,n) 61 Ni(p,n)

5.8 91.8 26.2 1.3

64

48.6

64

Ni(p,n) 66 Zn(p,n) 86 Sr(p,n)

1.1 27.8 9.9

94

9.3

Ni(p,α)

Zn(n,pn)

Mo(p,n)

References Spellerberg and colleagues,255 Sharma and colleagues,256 Lagunas-Solar and Jungerman257 — — McCarthy and colleagues214 McCarthy and colleagues,214 Szelecsenyi and colleagues258 McCarthy and colleagues,215 Szelecsenyi and colleagues,258 Zinn and colleagues259 — Szelecsenyi and colleagues232 Rosch and colleagues,227 Yoo and colleagues229 Bigott and colleagues,236 Qaim,237 Rosch and Qaim238

a

http://www.nndc.bnl.gov/nudat2/ accessed Mar 15, 2008.

the (p,n) reaction on stable enriched nickel-60, nickel-61, and nick el-64. Copper -60 has the highest positron branching ratio and the shortest half-life of 23.7 minutes. It also possesses the greatest positron energy of the three isotopes. Copper -61 has a 61% positron branch with a 1.22 MeV positron ener gy and a 3.3-hour half-life. The shorter half-life copper-60 and copper-61 are more attractive for labeling small molecule chelates or f ast clearing peptides rather than macromolecules that require time to

accumulate in the desired target tissue and clear from the blood pool and nontarget tissues. In spite of high positron energy and gammas from copper-60, small animal imaging revealed suitable image quality. Similarly, copper-61 also displa ys good image quality . Copper -60 has been used to label ATSM, F igure 9, for identifying h ypoxic tissues.214 Copper-61 w ould be w ell suited for labeling tracers including small peptides that w ould achie ve the maximum uptak e within 6 hours of injection, w hereas

318

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

copper-60 would work well with tracers with a maximum uptake within 2 hours. 214 Copper-64 has the longest half-life of the three positron-emitting copper isotopes with lo w positron energy comparable to fluorine-18. The 12.7-hour half-life provides the opportunity to label antibodies and large molecules with longer circulation and accumulation times. As noted earlier, copper -64 has onl y an 18 to 19% positron branch and is an electron emitter (40%). With these decay characteristics, copper -64 ma y be used for therap y and imaging. Betw een DO TA and TETA, copper -64 for ms more stab le comple xes with TETA. Thus, TETA has become the bifunctional chelate of choice for a v ariety of applications.216 Even though TETA comple xes ha ve favored stability, the metabolic loss of copper from these complexes in vi vo is still noted. 217,218 Several new classes of copper chelating ligands are being e valuated for improved in vi vo stability.219 Two of these are sho wn in Figure 11. The cross bridge (CB) ligands are DO TA and TETA molecules with an eth ylene CB between two opposite nitrogens. The new ligands CB-DO2A and CB-TE2A, developed by Anderson and colleagues,220,221 have stability constants that are similar to the non-cross bridged lig ands, yet the CB ligands are more iner t and are less susceptib le to in vivo transchelation than the non-CB ligands. The rank order of increasing transchelation is Cu-CB-TE2A > Cu-CB-DO2A > Cu-TETA > Cu-DOTA.218,222 The other ne w chelates in F igure 11 are the he xamine sarcophagine chelates (Sar ligands). 223–225 These ligands have been sho wn to chelate copper -64. Evaluation of their distribution in vivo demonstrated rapid (~30 min)

clearance from the b lood with low uptake in most major organs. In vitro stability measurements with copper -67labeled Sar sho wed 2% de gradation o ver 7 da ys in serum.226 These Sar ligands may be functionalized at the bridgehead carbon with a v ariety of g roups including NH3, OH, and ar yl amines as shown in Figure 11. These analogs will f acilitate the coupling of the Sar ligand to biomolecules. Yttrium-86 is a 14.7-hour isotope that is produced by the (p,n) reaction on enriched strontium tar gets.227–229 This 33% positron emitter has been comple xed with citrate and eth ylene-diamine tetrameth ylene phosphate (EDTMP) and evaluated in patients with cancer.230 From the yttrium-86 PET imaging, a quantitati ve measure of tracer distribution is determined and used to inform radiation treatment planning with the therapeutic isotope yttrium-90.230 Similar studies have been car ried out with [86Y]DOTA-D-Phe-Tyr-octreotide to deter mine the dosimetry of yttrium-90 therap y in tumors that o verexpress somatostatin receptors. 231 Gallium-66 is a 9.5-hour isotope produced b y the 66 Zn(p,n) reaction.232,233 Gallium has a 56% positron branch, but the positrons are v ery energetic at 4.15 MeV maximum energy. The gallium w as separated from the zinc b y diisopropyl ether extraction or cation exchange chromatography. The purif ied gallium-66 has been applied for studies with DOTA-octreotide and DOTA-biotin.234 Due to its ener getic positrons and gammas from the electron capture deca y, gallium-66 has therapeutic potential at high doses. 235 A positron labeled technetium w ould have signif icant appeal due to the well-known chemistry and the availability

Figure 11. New chelates for copper isotopes. CB-DO2A = 4,10-bis(carboxymethyl)-1,4,7,10-tetraazabicyclo[5.5.2]tetradecane; CB-TE2A = 4,11-bis(carboxymethyl)-1,4,8,11-tetraazabicyclo[6.6.2]hexadecane; diamSar = diamino-Sar; SarAr = benzylanilino-diamino-Sar; Sar = 3,6,10,13,16,19-hexaazabicyclo[6.6.6] eicosane.

Radiochemistry of Positron Emission Tomography

of kits that could be labeled with technetium. The (p,n) reaction on enriched mol ybdenum-94 produces the 52 minute technetium-94m. 236–238 Teboroxine and anti-CEA antibody fragment ha ve been successfull y labeled with technetium-94m.239,240 Quantitative imaging using technetium-94m and PET would provide better information relative to the single photon technetium-99m. The radiometals possess a range of deca y characteristics that match w ell with the tracer characteristics and the molecular processes that are being inter rogated. Some of the metals, such as yttrium-86, ma y be surrogates for therapeutic isotopes (e g, yttrium-90). Preimaging with the PET tracer will pro vide quantitative distribution infor mation that ma y be fed into treatment planning algorithms to ensure that adequate dose is being delivered w hen the therapeutic isotope is administered. The a vailability of some of these isotopes is cur rently limited, b ut as applications emer ge, these isotopes ma y become more prevalent.

MICRO-RADIOCHEMISTRY Automation of PET radiochemistr y processes be gan nearly 25 to 30 y ears ago shor tly after the f irst human doses of PET radiotracers w ere deli vered to the clinic. Chemists realized that the routine and reliab le production of these tracers could not be car ried out b y hand, especially from a radiation safety perspecti ve. Over the ensuing 25 y ears, re gulatory control and quality assurance has lar gely driven the f ield to produce full y automated synthesis devices that reproducibly and efficiently prepare curie quantities of radiolabeled tracers suitab le for clinical application. Cyclotrons toda y are capable of producing up to 15 Ci of 18F-fluoride ion obviating synthesis by hand due to the potential e xposure. The cur rent commercial synthetic de vices are lar ge and in many cases are dedicated to the production of a single product. They take up considerable hot cell space and are not modular or readil y adaptab le to preparation of multiple products. Also, once multiple curies of 18F-fluoride ion have been delivered to the device for a synthesis, it has not been possible to use that same device for another synthesis process for at least 24 hours. A historical vie w of automated synthesis de vices was published in 2003, 241 and more recently, a current perspective of radiochemistry automation has been written. 242 The move toward miniaturization and modular systems is being dri ven b y the need to produce multiple products in a single da y, the limited shielded space to house the device, and the device cost. Unlike FDG where

319

multiple doses are used dail y in the clinic, future tracers may onl y be used once or twice a da y or e ven once or twice a week. As such one does not need to mak e a large batch of the radiotracer for a small number of doses; it is not cost ef fective. The cur rent systems tak e up v aluable hot cell real estate, and the y cost $80,000 to $200,000 U.S. dollars. The space limitations, to adequatel y shield these systems, beg for miniaturization. Also, limited budgets to purchase ne w systems for each ne w tracer combined with limited prospect for cost reco very are necessitating alternatives. From the chemistry and regulatory perspective, miniaturization will reduce hazardous w aste b y reducing the amounts of solv ent and reagents used in the production process. The amount of reagent currently used in syntheses is 100 to 10000 times the amount of18F-fluoride ion or carbon-11 precursor. This means that reactions are not stoichiometric (reagents and reactants are unbalanced) and that the preparation will require separation of the small amount of radiolabeled product from the bulk reaction mixture. Reduction in reagents will mean reduced solv ent volumes that had been required to dissolve all the reaction components so that they are all in the same phase. Increasing reaction concentration, especiall y with respect to the radioisotope, may aid in increasing reaction potential and ultimately greater product yield through increased reaction surface area and enhancement of mass and heat transfer . Reduction of solvents and reactants also has benefits from a regulatory perspective. If one is able to reduce the quantity/concentration of solvents in the final dose to levels below re gulatory limits, then cer tain quality control tests, such as gas chromato graphy for residual solv ents, may be eliminated. This has the added benef it of getting the tracer to the clinic site f aster reducing deca y losses. The benef its from the cost, chemistr y, and re gulatory perspectives f ar outweigh the challenges of w orking on the microchemistr y scale. There are a fe w e xamples of the application of microscale PET radiochemistr y beginning to come into the literature. 243–245 There are also efforts in the pri vate sector to de velop commercial devices that operate in the microscale realm of fering the potential to produce smaller batches of PET radiotracers on-demand. Microscale chemistry has been applied to the production of FDG as a proof-of-principle e xperiment. Using microfluidics continuous flo w, microreactors ha ve been explored for the preparation of FDG. Lee and colleagues246 developed a microchip for the sequential preparation of FDG. The poly(dimethylsiloxane) (PDMS) chip, about the size of a penn y, is etched with microchannels and chambers much lik e a electrical circuit board.

320

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

The fluoride ion enters on one side of the chip and is concentrated in a miniature ion e xchange column. A solvent exchange from w ater to acetonitrile tak es place prior to the fluorination step in a reaction loop. The solv ent is exchanged back to w ater and then h ydrochloric acid is added to the reaction loop to deprotect the inter mediate, giving FDG. Using this chip, 190 µCi of FDG w as produced from 500 µCi of 18F-fluoride ion loaded, a yield of 38%. Only 1 µL of water from the cyclotron and ~0.32 mg of manose triflate star ting material w ere used for the reaction. Up to 1.74mCi of FDG w as produced with this synthesizer–on-a-chip device. The challenges of this system include the limitation on the amount of 18F-fluoride ion that may be introduced and the incompatibility of the PDMS matrix with organic solvents. The amount of product would be suf ficient for small animal studies but not human doses. Another microfluid reactor has been repor ted.247,248 This reactor was comprised of three pol ycarbonate layers that for med the reaction chamber and 100 µm channels for fluid flow. Two of these microreactors were combined for the preparation of FDG. The 18F-fluoride ion was resolubilized in DMF prior to loading into the reactant por t on the reactor . A manose triflate solution in DMF w as simultaneously pumped through the reactor with the 18Ffluoride ion solution at a rate of 250 µL/s. The outlet from the f irst reactor was connected to the inlet on the second reactor where sodium methanoate in methanol was mixed with the labeled intermediate to produce FDG. According to TLC results, 50% of the output solution from the second reactor was FDG with 10 to 20% unreacted 18F-fluoride ion and 20 to 30% labeled inter mediate. The limitations of this device are the need to prepare the fluoride solution prior to introduction in the chamber and the incomplete reaction that w ould require postreaction purification before use as a tracer. The promise of this microfluidic technolo gy is the ability to produce multiple products in a single day from a single batch of 18F-fluoride ion in a small footprint. The ability to produce GMP quality tracers using an automated system for efficiency and reliability will enable the delivery of unit doses of multiple tracers for clinical imaging studies. These devices would provide a cost ef fective approach to making small amounts of a tracer for a clinical study or small doses for animal studies. The microfluidic chemistr y af forded b y these de vices seems to w ork well. The major issues that need to be addressed are the concentration of the 18F-fluoride ion so that lar ger doses would be produced and at the end of the reaction the ability to purify the tracers.

CONCLUSION The def inition of molecular imaging espouses the application of molecular probes to inter rogate biolo gic processes and provide a readout of the state of the living system at the cellular and molecular le vel. Continued progress in molecular imaging is highly dependent on the development of sensitive and specific probes. PET radiotracers represent one series of probes that will enab le the advancement of molecular imaging. The sensiti vity of PET combined with the continued adv ances in high-resolution instrumentation will provide valuable information for understanding nor mal and disease-related processes as w ell as the de velopment of no vel therapeutics ultimately leading to individualized medicine.

REFERENCES 1. Aboagye EO, Price PM. Use of positron emission tomography in anticancer drug development. Invest New Drugs 2003;21:169–81. 2. Aboagye EO, Price PM, Jones, T. In vivo pharmacokinetics and pharmacodynamics in dr ug de velopment using positron-emission tomography. Drug Discov Today 2001;6:293–302. 3. Hutchinson OC, Collingridge DR, Barthel H, et al. Pharmacodynamics of radiolabelled anticancer drugs for positron emission tomography. Curr Pharm Des 2003;9:931–44. 4. Hutchinson OC, Collingridge DR, Barthel H, et al. Pharmacokinetics of radiolabelled anticancer dr ugs for positron emission tomo graphy. Curr Pharm Des 2003;9:917–29. 5. Beyer T, Townsend DW, Brun T, et al. A combined PET/CT scanner for clinical oncology. J Nucl Med 2000;41:1369–79. 6. Weber WA, Grosu AL, Czer nin J. Technology insight: adv ances in molecular imaging and an appraisal of PET/CT scanning. Nat Clin Pract Oncol 2008;5:160–70. 7. Tai YC, Laforest R. Instrumentation aspects of animal PET. Annu Rev Biomed Eng 2005;7:255–85. 8. Bogdanov AA Jr. Merging molecular imaging and RNA interference: early e xperience in li ve animals. J Cell Biochem 2008; 104:1113–1123. 9. Eckelman WC. The use of PET and knock out mice in the dr ug discovery process. Drug Discov Today 2003;404–10. 10. Eckelman WC. The use of gene-manipulated mice in the validation of receptor binding radiotracer. Nucl Med Biol 2003;30:851–60. 11. Bergstrom M, Grahnen A, Langstrom B. Positron emission tomography microdosing: a ne w concept with application in tracer and early clinical dr ug de velopment. Eur J Clin Phar macol 2003;59:357–66. 12. Garner RC, Lappin G. The phase 0 microdosing concept. Br J Clin Pharmacol 2006;61:367–70. 13. Frank R, Långström B , Antoni G, et al. The imaging continuum: bench to biomarkers to diagnostics. J Labelled Comp Radiopharm 2007;50:746–69. 14. Hoffman JM, Gambhir SS. Molecular imaging: the vision and opportunity for radiology in the future. Radiology 2007;244:39–47. 15. VanBrocklin HF. Radiopharmaceuticals for drug development: United States regulatory perspectiv`-e. Curr Radiopharm 2008;1:2–6. 16. Lewis JS, Welch MJ, Tang L. Workshop on the production, application and clinical translation of non-standard PET nuclides: a meeting report. Q J Nucl Med Mol Imaging 2008;52:101–6.

Radiochemistry of Positron Emission Tomography

17. Hevesy G. The absor ption and translocation of lead b y plants. Biochem J 1923;17:439–45. 18. Chievitz O, Hevesy G. Radioacti ve indicators in the study of phosphorus metabolism in rats. Nature 1935;136:754–5. 19. Blumgart HL, Yens OC. Studies on the v elocity of blood flow: I. The method utilized. J Clin Invest 1927;4:1–13. 20. Lawrence EO, Livington MS. The production of high speed protons without the use of high v oltages. Phys Rev 1931;38:834. 21. Livingood JJ, Seabor g GT. Radioacti ve isotopes of iron. Ph ys Re v 1938;54. 22. Livingood JJ , Seabor g GT . Radioacti ve isotopes of iodine. Ph ys Rev 1938;54:775–82. 23. Livingood JJ , Seabor g GT . Radioacti ve iodine isotopes. Ph ys Re v 1938;53:1015. 24. Perrier C, Se gre E. Radioacti ve isotopes of element 43. Nature 1937;140:193–4. 25. Segre E, Seabor g GT. Nuclear isomerism in element 43. Ph ys Re v 1938;54:772. 26. Wolf A, Jones W. Cycoltrons for biomedical radioisotope production. Radiochimic Acta 1983;34:1–7. 27. Ido T, Wan, C-N, Casella V, et al. Labeled 2-deoxy-2-fluoro-D-glucose analogs. I8F-Iabeled-2-deoxy-2-fluoro-D-glucose, 2-deo xy-2-fluoro-D-mannose and 14C-2-deoxy-2-fluoro-D-glucose. J Labelled Comp Radiopharm 1978;14:175–83. 28. Nickles RJ. A shotgun approach to the char t of the nuclides. Radiotracer production with an 11 MeV proton c yclotron. Acta Radiol Suppl 1991;376:69–71. 29. Nickles RJ. Production of a broad range of radionclides with an 11 MeV proton c yclotron. J Labelled Comp Radiophar m 1991; 30:120. 30. Nickles RJ . The production of a broader palette of PET tracers. J Labelled Comp Radiopharm 2003;46:1–27. 31. Green MA. A potential copper radiopharmaceutical for imaging the heart and brain: copper -labeled p yruvaldehyde bis(N4methylthiosemicarbazone). Int J Rad Appl Instr um B 1987; 14:59–61. 32. Green MA, Klippenstein DL, Tennison JR. Copper(II) bis(thiosemicarbazone) complexes as potential tracers for e valuation of cerebral and m yocardial b lood flo w with PET . J Nucl Med 1988; 29:1549–57. 33. John EK, Green MA. Str ucture-activity relationships for metallabeled b lood flo w tracers: comparison of k eto aldeh yde bis(thiosemicarbazonato)copper(II) deri vatives. J Med Chem 1990;33:1764–70. 34. Banerjee S, Pillai MR, Ramamoor thy N . Ev olution of Tc-99m in diagnostic radiophar maceuticals. Semin Nucl Med 2001; 31:260–77. 35. Kaden TA. Labelling monoclonal antibodies with macroc yclic radiometal comple xes. A challenge for coordination chemists. Dalton Trans 2006;30:3617–23. 36. Smith-Jones PM, Solit DB , Akhurst T, et al. Imaging the phar macodynamics of HER2 de gradation in response to Hsp90 inhibitors. Nat Biotechnol 2004;22:701–6. 37. Laforest R, Ro wland DJ, Welch MJ. MicroPET imaging with nonconventional isotopes. IEEE Trans Nucl Sci 2002;49:2119–26. 38. Welch MJ, Laforest R, Le wis JS. Production of non-standard PET radionuclides and the application of radiophar maceutical labeled with these nuclides. In: Schubiger PA, Lehmann L, Friebe M, editors. PET Chemistr y. Berlin Heidelber g: Springer -Verlag; 2007. p. 159–181. 39. Chatziioannou A, Qi J, Moore A, et al. Comparison of 3-D maximum a posteriori and f iltered backprojection algorithms for highresolution animal imaging with microPET . IEEE Trans Med Imaging 2000;19:507–12. 40. Ruangma A, Bai B , Lewis JS, et al. Three-dimensional maximum a posteriori (MAP) imaging with radiopharmaceuticals labeled with three Cu radionuclides. Nucl Med Biol 2006;33:217–26.

321

41. Philpott GW, Schw arz SW, Anderson CJ , et al. RadioimmunoPET : detection of colorectal carcinoma with positron-emitting copper 64-labeled monoclonal antibody. J Nucl Med 1995;36:1818–24. 42. Lewis J , Laforest R, Buettner T, et al. Copper -64-diacetyl-bis(N4methylthiosemicarbazone): an agent for radiotherap y. Proc Natl Acad Sci USA 2001;98:1206–11. 43. Lewis JS, Connett JM, Garbo w JR, et al. Copper -64-pyruvaldehydebis(N(4)-methylthiosemicarbazone) for the pre vention of tumor growth at wound sites follo wing laparoscopic sur gery: monitoring therapy response with microPET and magnetic resonance imaging. Cancer Res 2002;62:445–9. 44. Jansen HM, Knollema S, v an der Duin L V, et al. Phar macokinetics and dosimetr y of cobalt-55 and cobalt-57. J Nucl Med 1996; 37:2082–6. 45. Qaim SM. Recent de velopments in the production of F-18, Br -75, Br-76, Br-77 and I-123. Appl Radiat Isot 1986;37:803–10. 46. Qaim SM, Stocklin G. Production of some medically important shortlived neutron-def icient radioisotopes of halo gens. Radiochim Acta 1983;34:25–40. 47. Welch MJ, McElvany KD. Radionuclides of bromine for use in biomedical studies. Radiochim Acta 1983;34:41–6. 48. Hume SP , Ashworth S, Opacka-Juf fry J , et al. Ev aluation of [O-methyl-3H]WAY-100635 as an in vivo radioligand for 5-HT1A receptors in rat brain. Eur J Phar macol 1994;271:515–23. 49. Pike VW, McCarron JA, Lammerstma AA, et al. F irst delineation of 11 5-HT1A receptors in human brain with PET and [ C]WAY100635. Eur J Pharmacol 1995;283:R1–3. 50. Osman S, Lundkvist C, Pik e VW, et al. Characterization of the radioactive metabolites of the 5-HT1A receptor radioligand , [O-methyl-11C]WAY-100635, in monk ey and human plasma b y HPLC: comparison of the beha viour of an identif ied radioactive metabolite with parent radioligand in monk ey using PET . Nucl Med Biol 1996;23:627–34. 51. Ferrieri RA. Production and application of synthetic precursors labeled with carbon-11 and fluorine-18. In: Welch MJ, Redvanly CS, editors. Handbook of radiophar maceuticals: radiochemistr y and applications. West Susse x, England: John Wiley and Sons, Ltd.; 2003. p. 229–82. 52. Schlyer DJ. Production of radionuclides in acceleators. In: Welch MJ, Redvanly CS, editors. Handbook of radiophar maceuticals: radiochemistry and applications. West Sussex England: John Wiley and Sons, Ltd.; 2003. p. 1–70. 53. Wolf AP, Redv anly CS. Carbon-11 and Radiophar maceuticals. Int J Appl Radiat Isot 1977;28:29–48. 54. Iwata R. Reference Book for PET Radiopharmaceuticals. 2004. Available at: http://kakuyaku.c yric.tohoku.ac.jp/public/preface2004 .html (accessed March 20, 2008). 55. Crouzel C, Langstrom B , Pike VW, Coenen HH. Recommendations for a practical production of [C-11] meth yl-iodide. Appl Radiat Isot 1987;38:601–3. 56. Langstrom B, Lundqvist H. Preparation of C-11-methyl iodide and its use in synthesis of C-11-meth yl-L-methionine. Int J Appl Radiat Isot 1976;27:357–63. 57. Larsen P , Ulin J , Dahlstrom K, Jensen M. Synthesis of [C-11]iodomethane by iodination of [C-11]methane. Appl Radiat Isot 1997;48:153–7. 58. Link JM, Krohn KA, Clark JC. Production of [C-11]CH 3I by single pass reaction of [C-11]CH4 with I2. Nucl Med Biol 1997;24:93–7. 59. Reiffers S, Vaalburg W, Wiegman T, et al. C-11 labeled methyllithium as methyl donating agent—the addition to 17-k eto steroids. Int J Appl Radiat Isot 1980;31:535–9. 60. Kihlberg T, Neu H, Langstrom B. [C-11]methyl(2-thienyl)cuprates, new C-11-precursors used in the syntheses of C-11-labelled meth yl ketones and octane. Acta Chem Scand 1997;51:791–6. 61. Lidstrom P , Neu H, Langstrom B . Syntheses of [21- 11C] and (21-13C)Progesterone. J Labelled Comp Radiophar m 1997;39: 695–704.

322

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

62. Neu H, Bonasera TA, Langström B . Lithium[ 11C] meth yl (2-thienyl) cuprate-LiCN in 1, 4-additions to α, β-unsaturated 11 13 ketones. C and C labeling of the andro gen mesterolone. J Labelled Comp Radiopharm 1998;41:227–34. 63. Neu H, Kihlber g T, Langstrom B . [ 11C]Methyl(2-thienyl)cuprates, new 11C-precursorsin the synthesis of meth yl 11C-labeled f atty acids and k etones. J Labelled Comp Radiophar m 1995; 39:695–701. 64. Schoeps KO, Langstrom B, Stone-Elander S, Halldin C. Synthesis of (1-C-11)D-glucose and (1-C-11)D-mannose from online produced (C-11)nitromethane. Appl Radiat Isot 1991;42:877–83. 65. Schoeps KO, Halldin C. Synthesis of racemic [alpha-C-11]amphetamine and [alpha-C-11]phenethylamine from [C-11] nitroalkanes. J Labelled Comp Radiopharm 1992;31:891–901. 66. Schoeps KO, Stone-Elander S, Halldin C. Online synthesis of [C-11] nitroalkanes. Appl Radiat Isot 1989;40:261–2. 67. Jewett DM. A simple synthesis of [C-11] methyl triflate. Appl Radiat Isot 1992;43:1383–5. 68. Mock BH, Mulholland GK, Vavrek MT. Convenient gas phase bromination of [C-11]methane and production of [C-11]meth yl triflate. Nucl Med Biol 1999;26:467–71. 69. Elsinga PH, K eller E, DeGroot TJ, et al. Synthesis of [C-11] meth yl magnesium iodide and its application to the introdcition of [C-11] N-tert-butyl g roups and [C-11]-sec-alcohols. Appl Radiat Isot 1995;46:227–31. 70. Forngren T, Samuelsson L, Langstrom B. A C-11-Methyl stannane (5-[C11]methyl-1-aza-5-stanna-bicyclo[3.3.3]undecane) for use in palladium-mediated [C-11]C-C bond for ming reactions with organohalides. J Labelled Comp Radiopharm 2004;47:71–8. 71. Kihlberg T, Gullberg P, Langstrom B . [ 11C]Methylenetriphenylphosphorane, a new 11C-precursor, used in a one-pot Wittig synthesis of [beta- 11C]styrene. J Labelled Comp Radiophar m 1990; 28:1115–20. 72. Zessin J, Steinbach J , Johannsen B . Synthesis of triphen ylarsonium [C-11]methylide, a new C-11-precursor. Application in the preparation of [2-C-11]indole. J Labelled Comp Radiophar m 1999; 42:725–36. 73. Iwata R, Pascali C, Bogni A, et al. A simple loop method for the automated preparation of [C-11]raclopride from [C-11]methyl triflate. Appl Radiat Isot 2001;55:17–22. 74. Nagren K, Muller L, Halldin C, et al. Impro ves synthesis of come commonly used PET radioligands b y the use of [C-11]meth yl triflate. Nucl Med Biol 1995;22:235–9. 75. Wuest F, Ber ndt M, Kniess T. Carbon-11 labeling chemistr y based upon [11C]Methyliodide. In: Schubiger PA, Lehmann L, Friebe M, editors. PET Chemistr y: the dri ving force in molecular imaging. Heidelberg: Springer-Verlag; 2007. p. 183–213. 76. Halldin C, Bjurling P , Stalnack e CG, et al. C-11-labelling of dimethylphenethylamine in 2 dif ferent positions and biodistribution studies. Appl Radiat Isot 1989;40:557–60. 77. Antoni G, Langstrom B. Synthesis of DL-[3-C-11]valine using [2-C11]isopropyl iodide, and preparation of L-[3-C-11]valine by treatment with D-amino-acid o xidase. Appl Radiat Isot 1987; 38:655–9. 78. Langstrom B, Antoni G, Gullber g P, et al. The synthesis of 1-C-11labeled ethyl, propyl, butyl and isobutyl iodides and e xamples of alkylation reactions. Journal 1986;37:1141–5. 79. Ishiwata K, Ishii SI, Shinoda M, et al. Automated synthesis of radiochemically pure C-11-labeled eth yl, prop yl and butyl iodides. Appl Radiat Isot 1999;50:693–7. 80. Fasth KJ , Antoni G, Langstrom B . Synthesis of some alpha-C-11labeled substituted benzyl iodides. Appl Radiat Isot 1990; 41:611–3. 81. Ehrin E, Luthra SK, Crouzel C, Pike VW. Preparation of C-11 labeled prazosin, a potent and selecti ve alpha-1-adrenoreceptor antagonist. J Labelled Comp Radiophar m 1988;25:177–83.

82. Luthra SK, Pik e VW, Brady F . Preparation of some nca [1-C11]acid chlorides as labeling agents. Appl Radiat Isot 1990;41:471–6. 83. Ravert HT, Matthe ws WB, Musachio JL, Dannals RF . Synthesis of C-11 labeled meth ylcarbamates from [C-11] meth ylchloroformate. J Labelled Comp Radiopharm 1995;36:365–71. 84. Crouzel C, Amano R, F ournier D . Synthesis of C-11 labeled diazomethane. Appl Radiat Isot 1987;38:669–70. 85. Christman DR, Wolf A, Finn RD, Karlstrom KI. Production of carrierfree H11CN for medical use and radiopharmaceutical syntheses. 9. J Nucl Med 1973;14:864–6. 86. Iwata R, Ido T, Takahashi T, et al. Optimization of [C-11]HCN production and no-carrier-added [1-C-11]amino acid synthesis. Appl Radiat Isot 1987;38:97–102. 87. Lamb JF, James RW, Winchell HS. Recoil synthesis of high specif ic activity C-11-cyanide. Int J Appl Radiat Isot 1971;22:475–9. 88. Finn RD, Christman DR, Ache HJ, Wolf A. Preparation of c yanide C-11 for use in synthesis of or ganic radiopharmaceuticals. II. Int J Appl Radiat Isot 1971;22:735. 89. Westerberg G, Langstrom B. Synthesis of meta-iodobenzyl [C-11]guanidine. J Labelled Comp Radiopharm 1997;39:525–9. 90. Westerberg G, Langstrom B . Synthesis of [C-11] c yanogen-bromide and (C-13)-cyanogen bromide, useful electrophilic labeling precursors. Acta Chem Scand 1993;47:974–8. 91. Westerberg G, Langstrom B . On-line production of [C-11]c yanogen bromide. Appl Radiat Isot 1997;48:459–61. 92. Landais P, Crouzel C. A ne w synthesis of C-11 labeled phosgene. Appl Radiat Isot 1987;38:297–300. 93. Nishijima K, K uge Y, Seki K, et al. A simplif ied and improved synthesis of [ 11C]phosgene with iron and iron (III) o xide. Nucl Med Biol 2002;29:345–50. 94. Roeda D, Westera G. A UV-induced online synthesis of C-11-phosgene and the preparation of some of its deri vatives. Int J Appl Radiat Isot 1981;32:931–2. 95. Crouzel C, Roeda D , Berridge K, et al. C-11-labelled phosgene—an improved procedure and synthesis de vice. Int J Appl Radiat Isot 1983;34:1558–9. 96. Christman DR, Finn RD, Karlstrom KI, Wolf A. Production of ultra high activity C-11-labeled h ydrogen-cyanide, carbon-dio xide, carbonmonoxide and methane via N-14(p,alpha)C-11 reaction. 15. Int J Appl Radiat Isot 1975;26:435–42. 97. Al-Qahtani MH, Pik e VW. P alladium(II)-mediated C-11-carbon ylative coupling of diar yliodonium salts with or ganostannanes—a new, mild and rapid synthesis of ar yl [C-11]ketones. J Chem Soc Perkins Trans 2000;1:1033–6. 98. Zeisler SK, Nader M, Theobald A, Oberdorfer F. Conversion of no-carrier-added [C-11]carbon dio xide to [C-11]carbon mono xide on molybdenum for the synthesis of C-11-labelled aromatic k etones. Appl Radiat Isot 1997;48:1091–5. 99. Al-Qahtani MH, Pik e VW. Rapid mild syntheses of [C-11]benzophenones by Pd(0)-catalysed C-11-carbonylative coupling of iodoarenes with phen yltributylstannane in DME-w ater. J Labelled Comp Radiopharm 2000;43:825–35. 100. Langstom B , Itsenk o O , Rahman O . [C-11]Carbon mono xide, a versatile and useful precursor in labelling chemistr y for PET ligand de velopment. J Labelled Comp Radiophar m 2007; 50:794–810. 101. Miller PW, Long NJ, de Mello AJ, et al. Rapid multiphase carbonylation reactions b y using a microtube reactor: applications in positron emission tomo graphy 11C-radiolabeling. Ange w Chem Int Ed 2007;46:2875–8. 102. Iwata R, Ido T, Ujiie A, et al. Comparati ve-study of specif ic activity of [C-11]methyl iodide—a search for the source of carrier carbon. Appl Radiat Isot 1988;39:1–7. 103. Långström B , Antoni G, Gullber g P , et al. Synthesis of L- and D-[methyl-11C]methionine. J Nucl Med 1987;28:1037–40.

Radiochemistry of Positron Emission Tomography

104. Pascali C, Bo gni A, Iwata R, et al. High ef ficiency preparation of L-[S-methyl-C-11]methionine b y on-column [C-11]meth ylation on C18 Sep-Pak. J Labelled Comp Radiophar m 1999;42:715–24. 105. Friedland RP, Mathis CA, Budinger TF, et al. Labeled choline and phosphorylcholine: body distribution and brain autoradio graphy: concise communication. J Nucl Med 1983;24:812–5. 106. Rosen MA, Jones RM, Yano Y, Budinger TF. Carbon-11 choline: synthesis, purif ication, and brain uptak e inhibition b y 2-dimeth ylaminoethanol. J Nucl Med 1985;26:1424–8. 107. Pascali C, Bo gni A, Iwata R, et al. [C-11]Meth ylation on a C-18 Sep-Pak car tridge: a con venient w ay to produce [N-meth yl-C11]choline. J Labelled Comp Radiophar m 2000;43:195–203. 108. Watkins GL, Je wett DM, Mulholland GK, et al. A captive solv ent method for rapid N-[C-11]meth ylation of secondar y amides— application to the benzodiazepine, 4 ʹ′-chlorodiazepam (R O54864). Appl Radiat Isot 1988;39:441–4. 109. Wilson AA, Garcia A, Jin L, Houle S. Radiotracer synthesis from [11C]-iodomethane: a remarkab ly simple capti ve solvent method. Nucl Med Biol 2000;27:529–32. 110. Wilson AA, DaSilva JN, Houle S. Solid-phase radiosynthesis of [C11]WAY 100635. J Labelled Comp Radiopharm 1996;38:149–54. 111. Iwata R, Pascali C, Yuasa M, et al. Online [C-11] meth ylation using [C-11] meth yl-iodide for the automated preparation of C-11 radiopharmaceuticals. Appl Radiat Isot 1992;43:1083–8. 112. Iwata R, Pascali C, Bogni A, et al. A combined loop-SPE method for the automated preparation of [C-11]do xepin. J Labelled Comp Radiopharm 2002;45:271–80. 113. Andersson Y, Cheng AP, Langstrom B. Palladium-promoted coupling reactions of [C-11] meth yl-iodide with or ganotin and organoboron compounds. Acta Chem Scand 1995;49:683–8. 114. Hostetler ED, Terry GE, Bur ns HD. An improved synthesis of substituted [C-11]toluenes via Suzuki coupling with [C-11] meth yl iodide. J Labelled Comp Radiophar m 2005;48:629–34. 115. Wust F, Zessin J, Johannsen B. A new approach for C-11-C bond formation: synthesis of 17 alpha-(3 ʹ′-[C-11]prop-1-yn-1-yl)-3methoxy-3,17 beta-estradiol. J Labelled Comp Radiophar m 2003;46:333–42. 116. Antoni G, Langstrom B. Synthesis of C-11 labeled alpha, beta-unsaturated nitriles. Appl Radiat Isot 1992;43:903–5. 117. Andersson Y, Langstrom B . Transition metal-mediated reactions using [C-11] cyanide in synthesis of C-11 labeled aromatic-compounds. J Chem Soc Perkin Transact 1994;1:1395–400. 118. Andersson Y, Tyrefors N, Sihver S, et al. Synthesis of a C-11-labelled derivative of the N-meth yl-D-aspartate receptor antagonist MK801. J Labelled Comp Radiophar m 1998;41:567–76. 119. Joliot F , Curie I. Artificial production of a ne w kind of radioelement. Nature 1934;133:201–2. 120. Ferrieri RA, Schlyer DJ, Wieland BW, Wolf A. Online production of nitrogen-N-13 gas from a solid enriched target-C-13 and its application to ammonia-N-13 synthesis using microwave-radiation. Int J Appl Radiat Isot 1983;34:897–900. 121. Straatman MG. A look at 13N and 15O in radiophar maceutcals. Int J Appl Radiat Isot 1977;28:13–20. 122. Welch MJ , Lifton JF . The f ate of nitro gen-13 for med b y the 12 C(d,n)13N reaction in inorganic carbides. J Am Chem Soc 1971; 93:3385–8. 123. Whitehead AB, F oster JS. Nuclear cross section for the 16 O(p,alpha)13N reaction. Can J Phys 1958;36:1276. 124. Nickles RJ, Gatley SJ, Hichwa RD, et al. Synthesis of n-13-labeled nitrous-oxide. Int J Appl Radiat Isot 1978;29:225–7. 125. Wieneke J , Nebeling B . Impro ved method for N-13 application in short-term studies on NO 3-fluxes in barle y and squash plants. Z Pfanzenernahr Bodenk 1990;153:117–23. 126. Welch MJ, Straatman MG. The reactions of recoil 13N atoms with some or ganic compounds in the solid and liquid phases. Radiochim Acta 1973;20:124–9.

323

127. Parks NJ , Krohn KA. The synthesis of 13N ammonia, dinitro gen, nitrite, and nitrate using a single c yclotron tar get system. Int J Appl Radiat Isot 1978;29:754–6. 128. Berridge M, Landmeier BJ . In-tar get production of [N-13] ammonia—target design, products, and operating parameters. Appl Radiat Isot 1993;44:1433–41. 129. Wieland B, Bida G, P adgett H, et al. In-tar get production of [N-13] ammonia via proton ir radiation of dilute aqueous ethanol and acetic-acid mixtures. Appl Radiat Isot 1991;42:1095–8. 130. Beaver JE, Finn RD, Hupf HB. New method for production of highconcentration oxygen-15 labeled carbon-dioxide with protons. Int J Appl Radiat Isot 1976;27:195–7. 131. Sajjad M, Lambrecht RM, Wolf AP. Cyclotron isotopes and radiopharmaceuticals XXXIV. Excitation function for the 15N(p,n)15O reaction. Radiochim Acta 1984;36:159–62. 132. Vera Ruiz H, Wolf A. Excitiation function for 15O production via the 14 N(d,n)15O reaction. Radiochim Acta 1977;24:65–7. 133. Welch MJ, Ter Pergossian M. Preparation of short half-lived radioactive gases for medical studies. Radiat Res 1968;36:580–7. 134. Clark JC, Buckingham PD. Short lived radioactive gasses for clinical use. London: Butterworths; 1975. 135. Vera Ruiz H, Wolf A. Direct synthesis of O-15 labeled water at high specific activities. J Labelled Comp Radiophar m 1978;15:185–9. 136. Welch MJ , Lifton JF , Ter P ergossian M. Preparation of millicurie quantities of o xygen-15 labeled w ater. J Labelled Comp Radiopharm 1969;5:168–72. 137. Kabalka GW , Lambrect RM, Sajjad M, et al. Synthesis of o-15-labeled butanol via or ganoborane chemistr y. Int J Appl Radiat Isot 1985;36:853–5. 138. Holschbach M, Hamk ens W, Steinbach A, et al. [N-13]Cisplatin: a fast and ef ficient on-line synthesis using a solid state suppor t. Appl Radiat Isot 1997;48:739–44. 139. Pettit WA, Mor tara RH, Digenis GA, Reed MF . Preparation of nitroso-N-13-labeled nitrosoureas. J Med Chem 1975; 18:1029–31. 140. Pettit WA, Tilbury RS, Digenis GA, Mor tara RH. Con venient synthesis of N-13-BCNU . J Labelled Comp Radiophar m 1977;13:119–22. 141. Digenis GA, McQuinn RL, F reed B , et al. Preparation of N-13labeled streptozotocin and nitrosocarbar yl. J Labelled Comp Radiopharm 1979;16:95–6. 142. Kabalka GW, Goodman MM, Green JF, et al. Synthesis of nitro gen13 labelled amines using or ganoborane pol ymers. J Labelled Comp Radiopharm 1993;32:165. 143. Qaim SM, Clark JC, Crouzel C, et al. PET radionuclide production. In: Stocklin G, Pik e VW, editors. Radiophar maceuticals for positron emission tomography—methodological aspects, Dordrecht: Kluwer; 1993. p. 1–43. 144. Blessing G, Coenen H, F ranken K, Qaim S. Production of ( 18F)F2, H18F and 18Faq using the 20Ne(d,alpha)18F process. Appl Radiat Isot 1986;37:1135. 145. Ruth T, Wolf A. Absolute cross sections for the production of 18F via the 18O(p,n)18F reaction. Radiochim Acta 1979;26:21–4. 146. Ruth TJ, Buckley KR, Chun KS, et al. A proof of principle for targetry to produce ultra high quantities of F-18-fluoride. Appl Radiat Isot 2001;55:457–61. 147. Casella V, Ido T, Wolf A, et al. Anhydrous F-18 labeled elemental fluorine for radiophar maceutical preparation. J Nucl Med 1980;21:750. 148. Helus F, Maierborst W, Sahm U. F-18 cyclotron production methods. Radiochem Radioa Let 1979;38:395–410. 149. Hess E, Blessing G, Coenen HH, Qaim SM. Improved target system for production of high purity [F-18]fluorine via the O-18(p,n)F-18 reaction. Appl Radiat Isot 2000;52:1431–40. 150. Nickles RJ, Daube ME, Ruth TJ. An 18O2 target for the prodcution of (18F)F2. Int J Appl Radiat Isot 1984;35:117–22.

324

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

151. Chirakal R, F irnau G, Gar nett ES. Sequential production of electrophilic and nucleophilic fluorinating agents from [ 18F]fluorine gas. Appl Radiat Isot 1988;39:1099–101. 152. Jewett DM, P otoki JF, Ehrenkaufer RE. A gas-solid phase microchemical method for the synthesis of acetyl h ypofluorite. J Fluor Chem 1984;24:477–81. 153. Ehrenkaufer RE, MacGre gor RR. Synthesis of [ 18F]perchloro fluoride and its reaction with functionalized ar yl lithiums. Int J Appl Radiat Isot 1983;34:613–5. 154. Chirakal R, F irnau G, Schrobilgen GJ , et al. The synthesis of [18F]xenon difluoride from [ 18F]fluorine gas. Appl Radiat Isot 1984;35:401–4. 155. Oberdorfer F , Hofmann E, Maier -Borst W. Preparation of 18 F-labeled N-fluorop yridinium triflate. J Labelled Comp Radiopharm 1987;25:999–1005. 156. Bergman J, Solin O. Fluorine-18-labeled fluorine gas for synthesis of tracer molecules. Nucl Med Biol 1997;24:677–83. 157. Kilbourn MR. Fluorine-18 labeling of radiophar maceuticals. Washington, DC: National Academy Press; 1990. 158. Reivich M, K uhl D , Wolf A, et al. The ( 18F)-fluorodeoxyglucose method for the measurement of local cerebral glucose utilization in man. Circ Res 1979;44:127–37. 159. Fowler JS, Ido T. Design and synthesis of 2-deozy-2-[ 18F]fluoro-Dglucose (18FDG). In: Welch MJ, Redvanly CS, editors. Handbook of Radiopharmaceuticals. Radiochemistry and Applications, West Sussex, England: John Wiley and Sons, Ltd; 2003. p. 307–21. 160. Lemaire C. Production of L-[18F]fluoro amino acids for protein synthesis: overview and recent developments in nucleophilic syntheses. In: Mazo yer, BEA, editor . PET Studies on Amino Acid Metabolism and Protein Synthesis, Netherlands: Kluw er Academic Publishers; 1993. p. 89–108. 161. Lemaire C, Damhaut P, Plenevaux A, et al. Asymmetric synthesis of 4-[18F]fluoro-L-m-tyrosine via aromatic fluorination. Sci P ap 1991;32:935. 162. Lemaire C, Damhaut P , Plene vaux A, Comar D . Enantioselecti ve synthesis of 6-[fluorine-18]-fluoro-L-dopa from no-car rier-added fluorine-18-fluoride. J Nucl Med 1994;35:1996–2002. 163. Adam MJ , Ji van S. Synthesis and purif ication of L-6-[ 18F]fluorodopa. Appl Radiat Isot 1988;39:1203–10. 164. Luxen A , Guillaume M, Mele ga W, et al. Production of 6-[18F]fluoro-L-DOPA and its metabolism in vi vo—a critical review. Nucl Med Biol 1992;19. 165. Namavari M, Bishop A, Satyamurthy N, et al. Regioselective radiofluorodestannylation with [18F]F2 and [18F]CH 3COOF: a high yield synthesis of 6-[18F]fluoro-L-dopa. Appl Radiat Isot 1992;43:989–96. 166. DeJesus O, Sunderland J , Nickles J , et al. Synthesis of radiofluorinated analogs of m-tyrosine as potential L-dopa tracers via direct reaction with acetylhypofluorite. Appl Radiat Isot 1990;41:433–7. 167. Nahmias C, Wahl L, Chirakal R, et al. A probe for intracerebral aromatic amino-acid decarboxylase activity: distribution and kinetics of [18F]6-fluoro-L-m-tyrosine in the human brain. Mo v Disord 1995;10:298–304. 168. Namavari M, Satyamur thy N , Phelps M, Bar rio J . Synthesis of 6-[18F] and 4-[ 18F]fluoro-L-m-tyrosines via re gioselective radiofluorodestannylation. Appl Radiat Isot 1993;44:527–36. 169. Kiesewetter D, Katzenellenbogen J, Kilbourn M, Welch M. Synthesis of 16-fluoroestro gens b y unusuall y f acile fluoride ion displacement reactions: prospects for the preparation of fluorine-18 labeled estrogens. J Org Chem 1984;49:4900–5. 170. Kiesewetter D, Kilbourn M, Landvatter S, et al. Preparation of four fluorine-18 labeled estrogens and their selective uptakes in target tissues of immature rats. J Nucl Med 1984;25:1212–21. 171. DeGrado TR, Baldwin SW, Wang S, et al. Synthesis and e valuation of 18F-labeled choline analo gs as oncolo gic PET tracers. J Nucl Med 2001;42:1805–14.

172. DeGrado TR, Coleman RE, Wang S, et al. Synthesis and e valuation of 18F-labeled choline as an oncologic tracer for positron emission tomography: initial findings in prostate cancer. Cancer Res 2001; 61:110–7. 173. Bading JR, Shields AF. Imaging of cell proliferation: status and prospects. J Nucl Med 2008;49(Suppl 2):64S–80. 174. Grierson JR, Shields AF. Radiosynthesis of 3ʹ′-deoxy-3ʹ′-[(18)F]fluorothymidine: [(18)F]FLT for imaging of cellular proliferation in vivo. Nucl Med Biol 2000;27:143–56. 175. Shields AF, Grierson JR, Dohmen BM, et al. Imaging proliferation in vivo with [F-18]FLT and positron emission tomography. Nat Med 1998;4:1334–6. 176. Mathias CJ, Welch MJ, Kilbour n MR, et al. Radiolabeled h ypoxic cell sensitizers: tracers for assessment of ischemia. Life Sci 1987;41:199–206. 177. Reischl G, Dorow DS, Cullinane C, et al. Imaging of tumor h ypoxia 18 with [ 124I]IAZA in comparison with [ F]FMISO and [18F]FAZA—first small animal PET results. J Phar m Phar m Sci 2007;10:203–11. 178. Reischl G, Ehrlichmann W, Bie g C, et al. Preparation of the hypoxia imaging PET tracer [ 18F]FAZA: reaction parameters and automation. Appl Radiat Isot 2005;62:897–901. 179. Alauddin MM, Conti PS. Synthesis and preliminar y e valuation of 9-(4-[F-18]-fluoro-3-hydroxymethylbutyl)guanine ([F-18]FHBG): a new potential imaging agent for viral infection and gene therap y using PET. Nucl Med Biol 1998;25:175–80. 180. Li HF, Stockle M, Wagner R, et al. Synthesis of [F-18]FIA U, as a potential PET tracer for gene expresssion of herpes simplex virus type 1 thymidine kinase (HSV-1-tk). Eur J Nucl Med Mol Imaging 2004;31:S385–385. 181. Shiue GG, Lee RL, Shiue C, et al. 9-[(3-[F-18]fluoro-1-h ydroxy-2propoxy)methyl]guanine ([F-18]FHPG) and 9-(4-[F-18]fluoro-3 hydroxymethylbutyl)guanine ([F-18]FHBG) as gene therap y imaging agents: a comparison. J Nucl Med 2000;41:233P–233. 182. Mangner TJ, Kleck er R W, Anderson L, Shields AF. Synthesis of 2ʹ′-deoxy-2ʹ′-[18F]fluoro-beta-D-arabinofuranosyl nucleosides, [18F]FAU, [18F]FMAU, [18F]FB AU and [18F]FIA U, as potential PET agents for imaging cellular proliferation. Synthesis of [18F]labelled F AU, FMA U, FB AU, FIA U. Nucl Med Biol 2003;30:215–24. 18 183. Min JJ , Iy er M, Gambhir SS. Comparison of [ F]FHBG and [14C]FIAU for imaging of HSV1-tk reporter gene expression: adenoviral infection vs stab le transfection. Eur J Nucl Med Mol Imaging 2003;30:1547–60. 184. Tjuvajev JG, Doubrovin M, Akhurst T, et al. Comparison of radiolabeled nucleoside probes (FIA U, FHBG, and FHPG) for PET imaging of HSV1-tk gene e xpression. J Nucl Med 2002; 43:1072–83. 185. Shoup TM, Olson J, Hoffman JM, et al. Synthesis and e valuation of [18F]1-amino-3-fluorocyclobutane-1-carboxylic acid to image brain tumors. J Nucl Med 1999;40:331–8. 186. McConathy J , Voll RJ , Yu W, et al. Impro ved synthesis of anti[18F]FACBC: impro ved preparation of labeling precursor and automated radiosynthesis. Appl Radiat Isot 2003;58:657–66. 187. Goodman MM, Kilts CD, Keil R, et al. 18F-labeled FECNT: a selective radioligand for PET imaging of brain dopamine transpor ters. Nucl Med Biol 2000;27:1–12. 188. Shiue GG, Choi SR, F ang P, et al. N,N-dimethyl-2-(2-amino-4-(18Ffluorophenylthio)-benzylamine (4-18F-ADAM): an improved PET radioligand for serotonin transpor ters. J Nucl Med 2003; 44:1890–7. 189. Kolb HC, Finn MG, Shar pless KB. Click chemistry: diverse chemical function from a fe w good reactions. Angew Chem Int Ed 2001;40:2004–21. 190. Kolb HC, Sharpless KB. The growing impact of click chemistry on drug discovery. Drug Discov Today 2003;8:1128–37.

Radiochemistry of Positron Emission Tomography

191. Rostovtsev VV, Green LG, Fokin VV, Sharpless KB. A stepwise huisgen c ycloaddition process: copper(I)-catal yzed re gioselective ldquoLigationrdquo of azides and terminal alkynes. Angew Chem Int Ed 2002;41:2596–9. 192. Tornoe CW, Christensen C, Meldal M. P eptidotriazoles on solid phase: [1,2,3]-triazoles b y regiospecific copper(i)-catalyzed 1,3dipolar cycloadditions of terminal alkynes to azides. J Or g Chem 2002;67:3057–64. 193. Glaser M, Arstad E. Click labeling with 2-[ 18F]fluoroethylazide for PET. J Labelled Comp Radiophar m 2007;50 (Suppl 1):S36. 194. Li Z, Wu Z, Chen K, et al. Click chemistry for F-18-labeling of RGD peptides and microPET imaging of tumor inte grin alpha vbeta3 expression. Bioconjug Chem 2007;18:1987–94. 195. Marik J , Sutclif fe JL. Click for PET : rapid preparation of [F-18]fluoropeptides using Cu-I catal yzed 1,3-dipolar c ycloaddition. Tetrahedron Lett 2006;47:6681–4. 196. Mindt TL, Ross T, Schibli R. Efficient strategy for the parallel development of multiple modality imaging probes using “click chemistry.” J Labelled Comp Radiophar m 2007;50(Suppl 1):S34. 18 197. Wust F , Ramienda T, Ber gmann R. Synthesis of F -lab led neurotensin (8–13) via copper -mediated 1,3-dipolar (3+2) cycloaddition reaction. J Labelled Comp Radiophar m 2007; 50(Suppl 1):S36. 198. Yano Y, Cahoon JL, Budinger TF. A precision flow-controlled Rb-82 generator for bolus or constant-infusion studies of the hear t and brain. J Nucl Med 1981;22:1006–10. 199. Mathias CJ, Margenau WH, Brodack JW, et al. A remote system for the synthesis of copper -62 labeled Cu(PTSM). Int J Rad Appl Instrum [A] 1991;42:317–20. 200. Mathias CJ, Welch MJ, Raichle ME, et al. Ev aluation of a potential generator-produced PET tracer for cerebral perfusion imaging: single-pass cerebral e xtraction measurements and imaging with radiolabeled Cu-PTSM. J Nucl Med 1990;31:351–9. 201. Fujibayashi Y, Taniuchi H, Yonekura Y, et al. Copper -62-ATSM: a new hypoxia imaging agent with high membrane permeability and low redox potential. J Nucl Med 1997;38:1155–60. 202. Green MA. Metal radionuclides in diagnostic imaging b y positron emission tomography (PET). In: Abrams MJ, Murrer BA, editors. Advances in Metals in Medicine, Greenwich, Connecticut: J AI Press, Inc; 1993. p. 75–114. 203. Green MA, Welch MJ. Gallium radiopharmaceutical chemistry. Int J Rad Appl Instrum B 1989;16:435–48. 204. Maecke HR, Andre JP. 68Ga-PET radiophar macy: a generator -based alternative to 18F-radiopharmacy. In: Schubiger P A, Lehmann L, Friebe M, editors. PET Chemistr y: The Driving Force in Molecular Imaging, Heidelberg: Springer-Verlag; 2007. p. 215–242. 205. Moerlein S, Mathis C, Yano Y. Evaluation of aromatic iododemetallation reactions as methods for labeling radiophar maceuticals with iodine-122. J Labelled Comp Radiophar m 1986;23:1237–8. 206. Moerlein S, Mathis C, Yano Y. Comparati ve e valuation of electrophilic aromatic iododemetallation techniques for labeling radiopharmaceuticals with iodine-122. Appl Radiat Isot 1987;38:85–90. 207. Mausner L, Prach T, Richards P. Production of radionuclides for generator systems. In: Knapp FF , Butler TA, editors. Radionuclide Generators, Washington, DC: American Chemical Society; 1984. p. 77–95. 208. Richards P, Ku T. The 122Xe-122I system: a generator for the 3.62-min positron emitter, 122I. Int J Appl Radiat Isot 1979;30:250–4. 209. Lagunas-Solar M, Car vacho O , Har ris LJ , Mathis CA. Cyclotron production of 122Xe(20. 1 h)→122I (β+ 77%; EC 23%; 3.6 min) for positron emission tomo graphy. Cur rent methods and potential developments. Appl Radiat Isot 1986;37:835–42. 210. Mathis CA, Lagunas-Solar MC, Sargent T, et al. A 122Xe-122 I generator for remote radio-iodinations. Appl Radiat Isot 1986; 37:258–60.

325

211. Jagust WJ, Mathis CA, Seab JP, et al. PET studies of CO 2 reactivity using I-122-HIPDM as a quantitati ve b lood flo w tracer. J Nucl Med 1988;29:773. 212. Mathis C, Moerlein S, Yano Y, et al. Rapid labeling of radiopharmaceuticals with Iodine-122. J Labelled Comp Radiophar m 1986;23:1259–61. 213. Kung H, Tramposch K, Blau M. A new brain perfusion imaging agent: [I-123]HIPDM:N,N,Nʹ′-Trimethyl-Nʹ′-[2-Hydroxy-3-Methyl-5Iodobenzyl]-1,3-propanediamine. J Nucl Med 1983;24:66–72. 214. McCarthy DW, Bass LA, Cutler PD , et al. High purity production and potential applications of copper-60 and copper-61. Nucl Med Biol 1999;26:351–8. 215. McCarthy DW, Shefer RE, Klinkowstein RE, et al. Efficient production of high specif ic activity 64Cu using a biomedical c yclotron. Nucl Med Biol 1997;24:35–43. 216. Rogers BE, Anderson CJ, Connett JM, et al. Comparison of four bifunctional chelates for radiolabeling monoclonal antibodies with copper radioisotopes: biodistribution and metabolism. Bioconjug Chem 1996;7:511–22. 217. Bass LA, Wang M, Welch MJ, Anderson CJ. In vivo transchelation of copper-64 from TETA-octreotide to supero xide dismutase in rat liver. Bioconjug Chem 2000;11:527–32. 218. Boswell CA, Sun X, Niu W, et al. Comparati ve in vi vo stability of copper-64-labeled cross-bridged and conventional tetraazamacrocyclic complexes. J Med Chem 2004;47:1465–74. 219. Wadas TJ, Wong EH, Weisman GR, Anderson CJ. Copper chelation chemistry and its role in copper radiophar maceuticals. Cur r Pharm Des 2007;13:3–16. 220. Anderson CJ, Wadas TJ, Wong EH, Weisman GR. Cross-bridged macrocyclic chelators for stable complexation of copper radionuclides for PET imaging. Q J Nucl Med Mol Imaging 2008; 52:185–92. 221. Sprague JE, Peng Y, Fiamengo AL, et al. Synthesis, characterization and in vi vo studies of Cu(II)-64-labeled cross-bridged tetraazamacrocycle-amide complexes as models of peptide conjugate imaging agents. J Med Chem 2007;50:2527–35. 222. Boswell CA, McQuade P , Weisman GR, et al. Optimization of labeling and metabolite anal ysis of copper -64-labeled azamacrocyclic chelators b y radio-LC-MS. Nucl Med Biol 2005;32:29–38. 223. Geue RJ, Hambley TW, Harrowsfield JM, et al. Metal ion encapsulation: cobalt cages deri ved from polyamines, formaldehyde, and nitromethane. J Am Chem Soc 1984;106:5478–88. 224. Sargeson AM. The potential for the cage comple xes in biolo gy. Coord Chem Rev 1996;151:89–114. 225. Smith S. Molecular imaging with copper -64. J Inor g Biochem 2004;98:1874–901. 226. Di Bar tolo N , Sar geson AM, Smith SV . Ne w 64Cu PET imaging agents for personalised medicine and drug development using the hexa-aza cage, SarAr. Org Biomol Chem 2006;4:3350–7. 227. Rosch F, Qaim SM, Stocklin G. Production of the positron emitting radioisotope Y-86 for nuclear medical application. Appl Radiat Isot 1993;44:677–81. 228. Rosch F, Qaim SM, Stocklin G. Nuclear-data relevant to the production of the positron emitting radioisotope Y-86 via the Sr-86(p,n)processes and (nat)Rb(He-3,xn)-processes. Radiochim Acta 1993;61:1–8. 229. Yoo J, Tang L, Perkins TA, et al. Preparation of high specific activity 86 Y using a small biomedical c yclotron. Nucl Med Biol 2005;32:891–7. 230. Rosch F, Herzog H, Plag C, et al. Radiation doses of yttrium-90 citrate and yttrium-90 EDTMP as deter mined via analo gous yttrium-86 complexes and positron emission tomography. Eur J Nucl Med Mol Imaging 1996;23:958–66. 231. Herzog H, Rosch F , Stocklin G, et al. Measurement of phar macokinetics of yttrium-86 radiophar maceuticals with PET and radiation

326

232.

233.

234.

235.

236.

237. 238.

239.

240.

241.

242.

243.

244.

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

dose calculation of analogous yttrium-90 radiotherapeutics. J Nucl Med 1993;34:2222–6. Szelecsenyi F, Boothe TE, Tavano E, et al. Compilation of crosssections thick target yields for Ga-66, Ga-67 and Ga-68 production using Zn tar gets up to 30-MeV proton ener gy. Appl Radiat Isot 1994;45:473–500. Zweit J, Shar ma H, Do wney S. Production of gallium-66, a shor tlived, positron emitting radionuclide. Int J Rad Appl Inst [A] 1987;38:499–501. Ugur O , K othari PJ, F inn RD, et al. Ga-66 labeled somatostatin analogue DOTA-DPhe1-Tyr3-octreotide as a potential agent for positron emission tomo graphy imaging and receptor mediated internal radiotherapy of somatostatin receptor positi ve tumors. Nucl Med Biol 2002;29:147–57. Lewis MR, Reichert DE, Laforest R, et al. Production and purif ication of gallium-66 for preparation of tumor -targeting radiopharmaceuticals. Nucl Med Biol 2002;29:701–6. Bigott HM, Laforest R, Liu X, et al. Advances in the production, processing and microPET image quality of technetium-94m. Nucl Med Biol 2006;33:923–33. Qaim SM. Production of high purity ( 94m)Tc for positron emission tomography studies. Nucl Med Biol 2000;27:323–8. Rosch F, Qaim SM. Nuclear -data relevant to the production of the positron emitting technetium isotope 94mTc via the mo-94(p,n)reaction. Radiochim Acta 1993;62:115–21. Griffiths GL, Goldenberg DM, Roesch F, Hansen HJ. Radiolabeling of an anti-carcinoembr yonic antigen antibody F ab’ fragment (CEA-Scan) with the positron-emitting radionuclideTc-94m. Clin Cancer Res 1999;5:3001s–3. Nickles RJ, Nunn AD, Stone CK, Christian BT . Technetium-94mteboroxime: synthesis, dosimetry and initial PET imaging studies. J Nucl Med 1993;34:1058–66. Alexoff DL. Automation for the synthesis and application of PET radiopharmaceuticals. In: Welch MJ, Redvanly CS, editors. Handbook of Radiophar maceuticals: Radiochemistr y and Applications, West Sussex, England: John Wiley and Sons, Ltd; 2003. p. 283–306. Krasikova R. Synthesis modules and automation in F-18 labeling. In: Schubiger PA, Lehmann L, F riebe M, editors. PET Chemistr y: The Driving Force in Molecular Imaging, Heidelber g: SpringerVerlag; 2007. p. 289–316. Audrain H. P ositron emission tomo graphy (PET) and microfluidic devices: a breakthrough on the microscale? Angew Chem Int Ed 2007;46:1772–5. Lu SY, Pike VW. Micro-reactors for PET tracer labeling. In: Schubiger PA, Lehmann L, F riebe M, editors. PET Chemistr y: The

245.

246.

247. 248.

249.

250.

251. 252. 253. 254. 255.

256.

257.

258.

259.

Driving Force in Molecular Imaging, Heidelberg: Springer-Verlag; 2007. p. 271–87. Steel CJ, O’Brien AT, Luthra SK, Brady F. Automated PET radiosyntheses using microfluidic de vices. J Labelled Comp Radiophar m 2007;50:308–11. Lee CC, Sui G, Elizaro v A, et al. Multistep synthesis of a radiolabeled imaging probe using inte grated microfluidics. Science 2005;310:1793–6. Gillies JM, Prenant C, Chimon GN , et al. Microfluidic technolo gy for PET radiochemistry. Appl Radiat Isot 2006;64:333–6. Gillies JM, Prenant C, Chimon GN, et al. Microfluidic reactor for the radiosynthesis of PET radiotracers. Appl Radiat Isot 2006; 64:325–32. Fujibayashi Y, Matsumoto K, Yonkekura Y, et al. A new zinc-62 copper62 generator as a copper -62 source for pet radiophar maceuticals. J Nucl Med 1989;30:1838–42. Robinson GD, Zielinski FW, Lee AW. The zinc-62/copper-62 generator: a convenient source of copper-62 for radiopharmaceuticals. Int J Appl Radiat Isot 1980;31:111–6. Yagi M, K ondo K. Cu-62 generator . Int J Appl Radiat Isot 1979;30:569–70. Loc’h C, Maziere B , Comar D. A new generator for ionic gallium68. J Nucl Med 1980;21:171–3. Yano Y, Anger HO . A gallium-68 positron co w for medical use. J Nucl Med 1964;5:484–7. Yano Y, Chu P, Budinger TF, et al. Rubidium-82 generators for imaging studies. J Nucl Med 1977;18:46–50. Spellerberg S, Reimer P, Blessing G, et al. Production of Co-55 and Co-57 via proton induced reactions on highl y enriched Ni-58. Appl Radiat Isot 1998;49:1519–22. Sharma H, Zweit J, Smith AM, Downey S. Production of Cobalt-55, a shor t-lived, positron emitting radiolabel for b leomycin. Appl Radiat Isot 1986;37:105–9. Lagunas-Solar MC, Jungerman JA. Cyclotron production of car rierfree cobalt-55, a new positron-emitting label for b leomycin. Int J Appl Radiat Isot 1979;30:25–32. Szelecsenyi F, Blessing G, Qaim SM. Excitation-functions of proton-induced nuclear -reactions on enriched Ni-61 and Ni-64— possibility of production of no-carrier-added Cu-61 and Cu-64 at a small cyclotron. Appl Radiat Isot 1993;44:575–80. Zinn K R, Chaudhuri TR, Cheng TP, et al. Production of no-carrier-added 64Cu from zinc metal ir radiated under boron shielding. Cancer 1994;73:774–8.

21 RADIOCHEMISTRY OF SPECT: 99M 111 EXAMPLES OF TC AND IN COMPLEXES HANK F. KUNG, PHD

Radionuclides for single-photon emission computer tomography (SPECT) ha ve a relati vely longer half-life compared with the ones for positron emission tomography (PET). It is preferred to have a medium gamma ray (γ-ray) energy (100 to 300 KeV) for SPECT imaging. Most of the commercial gamma cameras for SPECT imaging are tuned for optimal counting of 140 K eV, the γ-ray emitted b y 99m Technetium (99mTc). Generally, it is well recognized that PET has a higher resolution, higher sensitivity, and a better quantitation capability than SPECT. But there are more hospitals equi pped with SPECT sc anners; therefore, SPECT is more practical as a routine procedure. Currently, other imaging modalities, such as magnetic resonance imaging (MRI) and computed tomo graphy (CT), could potentially supplant or substitute for SPECT studies. Commonly used radionuclides for SPECT imaging are listed in Table 1. The radiochemistry of SPECT imaging agents is largely based on 99mTc. In the past 30 y ears, 99mTc (T1/2 = 6 h) has been the most widely used radionuclide in nuclear medicine for SPECT imaging. 99mTc is produced from a generator after the decay of Molybdenum-99 (99Mo) (T1/2 = 67h). The generator is based on a simple column chromatography system, which separates the daughter , 99mTc, from the parent radionuclide, 99Mo (Figure 1). The generator was developed

in the 1960s and became a mature commercialized product in the 1970s. The transition of meta-stab le 99mTc to stab le 99 Tc is accompanied b y the emission of a medium ener gy γ-ray (140 KeV) with essentially no other radiation (leading to a low radiation dose to patients). Because the 99mTc is a man-made radionuclide (no naturall y occur ring 99Tc), the specific activity of 99mTc is generally very high. In the past several decades, the 99mTc/99Mo generator has pro vided a steady supply of imaging agents for routine clinical procedures.1,2 In consequence, the majority of cur rent imaging devices for single photon emission tomo graphy have been calibrated and optimized specif ically to the 140 K eV γ-ray emitted by this radionuclide.

TECHNETIUM CHEMISTRY Reduction and Chelation Reactions of 99mTc The 99mTc/99Mo generator is de veloped using an alumina column chromato graphy system. When the generator is eluted with saline, the desired 99mTc is separated from the column as sodium per technetate (NaTcO 4), w hich is the starting material for all 99mTc radiopharmaceuticals. The oxidation state of the technetium (Tc) in the per technetate

Table 1. RADIONUCLIDES FOR SPECT IMAGING AGENTS Radionuclides 99m

Tc 123 I 111 In 201 Tl

T1/2

γ Ray

99Mo

AI2O3 Column

6h 13 h 67 h 72.5 h

140 KeV 159 KeV 171 and 245 KeV 70–90 keV

T1/2 67 h Parent

eluted with saline

99mTc

T1/2 6 h Daughter

Figure 1. 99mTc/99Mo generator is based on an alumina column, which is eluted with saline separating the parent, 99Mo, from the daughter, 99mTc.

327

328

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

is +7, the highest oxidation state possible. Therefore, chemical preparations of Tc usually begin with a reduction reaction. Technetium is a g roup VII B element of the periodic tab le, and it has an atomic number (Z) of 43. The electronic configuration of Tc leads to higher empty orbitals, 5s and 4d , which are receptive for pi (π) bonding. Most Tc chemistry is related to the for mation of coordinate-co valent bonding through electronic-rich atoms (soft donor ligands), such as O, S, N , P, and C (the process is generall y called chelation).3–8 Depending on reaction conditions, reducing agents and chelating agents, the o xidation state of Tc in the f inal complexes can range from 0 to +6 (F igure 2). The formation of multiple coordinate co valent bonds (chelation) through the use of soft donors stabilizes the reduced Tc core. The most commonly used reducing agent in the preparation of radiophar maceuticals is stannous (+2) chloride. After reducing pertechnetate to a lower oxidation state, the stannous (+2) ion is o xidized to stannic (+4) ion (see Figure 2). Without the presence of a chelating g roup to form the coordinate covalent bonds, pertechnetate is usually reduced to the +4 oxidation state as TcO2 (an insoluble colloidal material). In preparation of Tc radiopharmaceuticals for routine clinical use, the problem most often encountered is incomplete reduction due to the presence of air (oxygen), an oxidant. This can lead to the for mation of technetium dioxide or the oxygen can re-oxidize the “reduced Tc” back to pertechnetate. The re-oxidized 99mTc pertechnetate often shows up in the th yroid and stomach making the scan unacceptable.

CHELATION OF RADIOMETALS: 99MTC Myocardial Perfusion Imaging Agents For the past 20 years, SPECT imaging of myocardial perfusion with 99mTc has been the most commonl y used procedure in nuclear medicine. In 2006, approximately 9 million such studies w ere perfor med in the United States. These studies account for appro ximately 50% of all nuclear medicine procedures. 9,10 SPECT myocardial imaging studies play a critical role in the risk assessment of both asymptomatic and symptomatic individuals. It is Sn2

O Na

O

Tc O

7

O

Sn4 Tc at a lower oxidation state at 6, 5, 4, 3, 2, 1, or 0

Oxidation - Reduction Reaction

Figure 2. 99mTc pertechnetate is reduced by stannous (+2) as the reducing agent. In the process of 99mTc labeling, stannous (+2) ion is oxidized to stannic (+4) ion, whereas the pertechnetate at the +7 oxidation state is reduced to a lower oxidation state.

now a w ell-established nonin vasive imaging modality that is an impor tant par t of the e valuation of patients with chest pain w ho are suspected of ha ving coronar y artery disease. SPECT instr umentation and software are continually upgraded to provide better resolution, higher counting ef ficiency, and robust attenuation cor rection for m yocardial p erfusion s tudies.11 Development o f lipophilic plus one char ged 99mTc comple xes (used for myocardial perfusion imaging) is considered one of the major contributions of radiopharmaceutical chemistry to nuclear medicine. It is generall y accepted that plus one char ged 201-Thallium (201Tl) (T1/2 = 73 h, 70 to 90 keV) is also an excellent tracer for measuring re gional myocardial perfusion in conjunction with rest and stress tests. 12,13 On the basis of size/charge ratio, it is believed that Tl+1 is a close analog of K + and the ions are transpor ted across the cell membrane b y the Na +/K+-ATP pump. After intra venous (IV) injection, it displays a very high f irst-pass extraction (> 80%) into m yocardium, and its distribution reflects regional myocardial blood flow. The first-pass extraction is a measure of ef ficiency of an y agent that can pass through the m yocardium and enter the m yocardial cells following the b lood flow. However, the γ-ray energies of 201 Tl are too lo w (70 to 90 k eV, Hg X-ra ys) for optimal γ detection (all of SPECT cameras are optimized for 140 KeV, the γ-ray emitted by 99mTc), and the attenuation is relatively high for optimal SPECT imaging. In addition, due to its long ph ysical half-life, patient recei ves a relatively high radiation dose. Scattering also increases the noise level of the final images, which makes it more difficult to quantify the myocardial perfusion images. Despite these shortcomings, Tl1+ is still commonl y exploited for measuring myocardial perfusion in patients because it has excellent initial uptake and redistribution for estimation of viable m yocardium. Ho wever, because of the ph ysical properties of 99mTc (T 1/2 = 6 hours, 140 keV), 99mTc methoxyisobutylisonitrile [(MIBI); Sestamibi] and 99mTc tetrafosmin, w hich are F ood and Dr ug Administration (FDA)-approved, they are the agents of choice for routine myocardial perfusion studies (F igure 3). Both agents are lipophilic and contain a plus one charged Tc core. The kit formulations for these Tc radiopharmaceuticals are listed in F igure 3. It is belie ved that these plus one char ged 99m Tc agents enter normal myocardial cells by a simple diffusion mechanism (their first-pass extraction rate is approximately 50%, w hich is lo wer than that of Tl1+). Once inside the cell, they are trapped in the mitochondria (Figure 4).14–16 It is important to note that both MIBI and tetrafosmin display relatively high liver uptake and reten99m tion. Li ver uptak e and retention for Tc-MIBI and

Radiochemistry of Single-Photon Emission Computer Tomography

tetrafosmin can be reduced by the addition of a h ydrolyzable ester group on the side chain (CPI). The CPI with this ester linkage showed moderate washout from the heart and rapid hepatobiliar y clearance in animal models and human volunteers.17–19 Additional no vel m yocardial perfusion imaging agents developed from 99mTc chelates have been reported to circumvent the undesired li ver uptake and improve in vivo kinetics. 20–22 In the past 15 y ears, a ne w g roup of 99m Tc chelates based on Tc tricarbonyl (Tc(CO)3) derivatives has been reported. It provides ample venues for new chelation chemistr y.23–25 However, the y ha ve not been commonly adopted for human studies, because MIBI and tetrafosmin have already been widely accepted by nuclear cardiologists. In vi vo biodistribution studies in suitab le animal models are typically the ultimate test for the effectiveness of ne w agents. In the late 1980s, Deutsch’ s g roup performed a “Noah’s Ark experiment” in which plus one charged 99mTc comple xes w ere tested as m yocardial imaging agents in multiple animal species. 26 The experiment showed the risk of using any animal models to evaluate the kinetics of m yocardial uptake of v arious 99mTc plus one comple xes. The biodistribution patter ns w ere R

R N

N

R N

C C C Tc 1

OCH3 R: CH2 C CH3

C C C N R

N

EtO EtO

N

CH3

EtO EtO

R R MIBI (Sestamibi, Cardiolite ® )

OEt OEt

P CI P Tc 1 P CI P

OEt OEt

Tetrafosmin (Myoview ® )

Figure 3. Chemical structures and formulations of 99mTc MIBI (Sestamibi, Cardiolite®) and Tetrafosmin (Myoview®). They are two of the most commonly used 99mTc radiopharmaceuticals for nuclear medicine procedures.

Plus one charged ions Myocyte Trapping K11 T111

1

1

Na /K ATPase

Rb11

K11 T111

329

species dependent, and no one model could adequatel y predict human m yocardial uptake and retention. 27 These findings may also be e xtended to de velopment of other radiopharmaceuticals.28–30

Brain Imaging Agents for Regional Perfusion A number of reported complexes show substantial brain uptake and retention, including [99mTc] hexamethyl propyleneamine o xime (HMP AO),31,32 [99mTc]L,L-ethylene cysteine dimer (ECD),33–37 and [99mTc] boric acid adducts of technetium o ximes (B ATO) comple xes.38,39 Among these, [99mTc]HMPAO and [99mTc]ECD are the most wellstudied agents. Both ha ve recei ved FD A appro val for human use as regional brain perfusion imaging agents. The chelation chemistry of ECD and HMP AO is based on the formation of a TcO3+ center core (Figure 5). Each has a set of electron-donating sof t ligands, N4 and N2S2 atoms, respectively, and each forms a neutral 99mTc complex using a very unique combination of chelating g roups. A wealth of information has been reported for SPECT imaging of regional brain perfusion in normal and disease states.40–43 The 99mTc brain imaging agents de veloped so far focus on measuring re gional perfusion or its changes due to a par ticular disease state. The brain uptak e of [99mTc]HMPAO and ECD is based on simple dif fusion due to high lipophilicity . Ho wever, the trapping mechanisms for brain retention of [ 99mTc]HMPAO and ECD are quite different. Although ECD is trapped by enzymatic degradation of ester g roups in vivo, there are esterases in the brain cells, which can hydrolyze the ethyl ester groups of ECD leading to the for mation of acid. The acid derivative(s) will be more hydrophilic and will be trapped inside the cells because it cannot readil y diffuse out of the cell membrane. HMPAO relies on in vi vo instability (in vivo hydrolysis of relati vely unstab le [ 99mTc]HMPAO

Lipid-solubleTc-99m MIBI or Tc-99m Tetrafosmin Myocyte Tc

11

Rb11

11

Tc

Binding to Cytosol Fractions TRAPPING

ATPase Inhibitors MIBI mechanism

Figure 4. Mechanisms of uptake and retention of 201Tl, 99mTc MIBI, and Tetrafosmin in myocardial cells. Plus one charged metal ions such as K+, Tl+ or Rb+ penetrate cell membranes by using an active transport system tightly coupled to Na+/K+ ATPase; but the lipophilic plus one charged MIBI and Tetrafosmin.

330

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

NH HN N HO

N

H

NO N Tc N N

Sn(II) TcO4

O

OH

N

H 99mTc]HMPAO (Ceratec ® )

[ H C2H5OOC N

H N

COOC2H5

Sn(II) TcO4

SH HS

H C2H5OOC N O N Tc S S

N

SH HS

O

COOC2H5

[99mTc]ECD (Neurolite® )

Figure 5. Radiolabeling and formation of commercially available [99mTc]HMPAO (Ceratec®) and [99mTc]ECD (Neurolite®) for brain perfusion imaging.

complex) for retention.44 Rapid advances in MRI and fast CT ma y pro vide f aster and more ef ficient procedures, with superior anatomical resolution, for the diagnosis of changes in regional brain perfusion.

Brain Imaging Agents—Receptor or Binding Site-Specific Imaging Agents Many researchers have noted that 99mTc could be useful in labeling agents for receptor or binding site-specif ic imaging.45–49 Several reports have shown that it is possible to incor porate [Tc vO]+3N2S2 into potential binding site–selective imaging agents. 50,51 Generally, these agents are classif ied into tw o cate gories: pendent approach, in which the 99mTc complexing moiety hangs from the main body of the molecule responsib le for binding to the pocket of the receptor-ligand binding site; or inte grated approach, in w hich the 99mTc complex is inte grated into the receptor specif ic ligand. 47 99m Tc central ner vous system (CNS) receptor imaging agents use a small neutral Tc chelating core to maximize the initial deli very of radioacti ve tracers into the brain following an IV injection while the resulting complex maintains a high binding affinity and selectivity toward the targeted binding sites or receptors. Without suitab le proper ties (small size and optimal lipophilicity) to penetrate the intact blood-brain barrier (BBB), no 99mTc labeled agents can be used as imaging agents for specific binding sites in the brain. The development of a useful 99mTc imaging agent with the above biological proper ties is a seemingl y o verwhelming challenge.28,45 N2S2 ligands have been selected as the basic chelating systems for complexing reduced 99mTc because ligand systems consisting of soft donor atoms, N and S, for m highly stable [TcvO]3+ complexes with pyramidal structures. During the comple x for mation, tw o or three protons will be

H

H [TcvO] 3H

3

N O N Tc S S

0 N O N Tc S S

R

Mol Wt. 282

Figure 6. Formation of [TcvO]+3 complexes based on the N2S2 (BAT) ligands. Mono N-substitution (indicated by an arrow) is the connecting site for the receptor binding moiety based on the pendent approach.

removed countering the +3 or +2 char ges at the center Tc core and creating a comple x with an o verall net char ge of zero. Therefore, the ionizability of the S-H and the amine or amide N-H g roups can deter mine the f inal net charge giving them a strong influence on the ultimate biodistribution of these imaging agents (Figure 6). To prepare neutral and lipophilic [Tc vO]3+ complexes based on the N2S2 (BAT) ligands, three protons (H) must be ionized. It has been sho wn that neutral TcvON2S2 comple xes can be prepared with a predictable structure: a [Tc vO]3+ center core and a square pyramidal str ucture.52 This ligand system can be modified in various ways. For example, changing an amine (ie, NR 3) to an amide g roup (ie, R-C(=O)NR 2) will decrease the lipophilicity b y a f actor of 10 to 100. A gem dimethyl substitution (ie, at the carbon ne xt to the sulfur atom) will increase the lipophilicity b y a f actor of 10 to 100 and will significantly increase the stability of the Tc vON2S2 complexes by preventing the nucleophilic attack from the sixth coordinating site. 53 It is important to balance all of these f actors in light of the stringent requirements for 99mTc CNS receptor imaging (ie, the ligand needs to be neutral, lipophilic, and small enough to be dif fusible through the intact BBB and localized in receptor sites with high af finity and selectivity). The basic TcvON2S2 complex, with or without amide and gem-dimeth yl g roups, pro vides fle xibility for designing comple xes with additional str uctural modifications for receptor binding. Several researchers have sought to create dopamine transporter (DAT) ligands to e valuate changes in presynaptic D AT sites in vi vo and in vitro, especiall y for patients with Parkinson’s disease (PD), which is characterized by a selecti ve loss of dopamine neurons in the basal ganglia and substantia nig ra. Recent pub lications using [11C]-CFT and [ 123I]-β-CIT suggest a strong correlation between a decrease in localization in the putamen a rea an d P D s ymptoms.54–56 The results ha ve encouraged the fur ther development of these agents for diagnosing and monitoring the treatment of patients with PD. Cur rently, [ 123I]-β-CIT,57,58 [123I]-FP-β-CIT,59,60

Radiochemistry of Single-Photon Emission Computer Tomography

H3C*

H3C

N

COOCH3

N

N

COOCH3

*I FP

CIT

CFT N

F

*I

F

N

COOCH3

*I

COOCH3

CIT

H3C

331

S O S N

Tc N

N CI

COOCH3

[99mTc]-TRODAT-1

*I CI IPT

F Altropane

Figure 7. Chemical structures of labeled tropanes, [11C]CFT and [123I]β-CIT, FP-β-CIT, IPT and altropane and [99mTc]TRODAT-1, as dopamine transporter (DAT) imaging agents.

[123I]-IPT,61–63 and [ 123I]-altropane64,65 are being de veloped to meet this pur pose (Figure 7). Because 123I is a cyclotron-produced isotope, w hich is more e xpensive than 99mTc, 99mTc labeled agents are highly desirable for routine clinical study . In addition, 99mTc agents can be made more readily available and can be more easily used within the cur rent infrastr ucture of nuclear medicine practice. All of these f actors pro vide strong incenti ves for their development. In addition, it is often observed that the patients with PD show asymmetric reduction of uptake between the left and right striatum. The f irst CNS D AT binding site-specif ic imaging agent, [99mTc]TRODAT-1, was developed based on [TcvO]3+N2S2 and [TcvO]3+NS3 cores. Initial clinical studies clearl y suggested that this agent sho ws promise as a diagnostic tool for PD and other CNS neurode generative diseases, w hich displa y a reduction of dopamine neurons.66–68 Several other research g roups ha ve repor ted 99m Tc labeled tropanes as potential D AT imaging agents.49,69 None of these has been successfull y tested in humans. Recent reports on several new tropane derivatives in humans also sho wed ne gative results. 70,71 Neutral and lipophilic c yclopentadienyltricarbonyl Tc/Rhenium (Re) complexes (the y are often called piano stool comple xes) have been reported as alternative chelating groups.45,49,72–74 Up to no w, only [ 99mTc]TRODAT-1 has been successfull y used for human brain imaging (Figure 8).66 Due to the presence of two optical centers in the molecule, one of the major stereochemistr y issues related to [99mTc]TRODAT-1 is the presence of tw o distincti ve diastereomers (F igure 9 A, B). Both diastereomers display selective uptake in the striatal area, suggesting that the localization is consistent with the D AT distribution. Racemic mixtures of [ 99mTc]TRODAT-1 have been used for all of the studies on humans. However, the presence of diastereomers in [ 99mTc]TRODAT-1 and their biolo gical properties in human brain require fur ther consideration.

Figure 8. Transaxial, SPECT images of human brain at 3 hours post IV injection of 20 mCi of [99mTc]TRODAT-1 for normal and Parkinsonian subject, respectively. In the normal subject, a high accumulation of [99mTc]TRODAT-1 was observed in caudate and putamen, where dopamine transporters are concentrated, whereas the Parkinsonian patient displayed a dramatically decreased uptake in this region of the brain. Courtesy of Mozley PD et al.67

To identify and characterize the diastereomers, the corresponding Re complexes were prepared and the structures w ere deter mined. Diastereomers of [ 99mTc]TRODAT-1 were separated on HPLC using an optical column (Chiracel-AD), and the biodistribution of each indi vidual isomer w as e valuated in rats. The designation of peak A and peak B is cor related with the Re comple xes, with which the X-ra y cr ystallography str ucture deter mination was achieved (see Figure 9). The HPLC profiles of Re derivatives A and B cor respond to those of peak A and peak B, respectively (Figure 10). Biodistribution studies performed on the separated isomers sho w that the y ha ve strikingl y dif ferent properties. Separation and biodistribution of the 99mTc isomers sho wed that [ 99mTc]TRODAT-1A has higher initial brain uptak e in rats than [ 99mTc]TRODAT-1B (0.5% dose/or gan vs 0.28% dose/or gan, respecti vely). After 60 minutes, the ratios of ST/CB (%dose/g ratio between striatum and cerebellum = tar get/nontarget ratio) were 2.72 and 3.79 for [ 99mTc]TRODAT-1A and B, respecti vely. This higher ratio of isomer B is also reflected in a higher binding af finity seen in in vitro

332

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

C(13)

CI(1) S

CH3

O

C(19) S

C(18)

Tc

N 2b

C(17) N

N

C(11) C(20)

N(3)

C(21) C(14) C(15) C(16)

C(14) C(10) C(11)

3b

C(8) C(7) O(1) C(4) C(3) N(2)

CI

Tc-99m TRODAT-1 N(1) C(2) C(1)

Re(1)

C(12) C(9)N(3)

C(5)

A

C(13)

C(21) C(20)

S(2)

C(9) O(1) C(15) C(8) C(7) C(4) N(2)

C(16)

C(19)

C(5)

C(18)

C(17)

B

C(6) S(1)

C(10)

C(6)

C(3) N(1) C(2) Re(1) C(1)

S(2)

S(1)

CI(1)

Figure 9. The N-alkyl substituted [TcvO]3+N2S2 complexes formed diastereomers. X-ray structure analyses on the two Re-TRODAT-1 isomers showed that both complexes have a syn configuration (ie, the TcO bond is on the same side as that of the N-substituted functional group of the [TcvO]3+N2S2 center core). Isomers A and B displayed corresponding HPLC profiles to those of peak A and peak B of [99mTc]TRODAT-1 confirming that it is likely the 99mTc isomers may have similar structures.

A A B HPLC Separation

0 2 4 6 8 10 12 14 16 18 20 ret.time (min) HPLC profile of racemic mixture by PRP, 1 column, CH#CN/buffer (pH 7) 80:20, 1 mL/min, ret.time 12 min. Radiochemical Purity > 97%

0

10

20

30

40 50 60 ret.time (min)

HPLC profiles of Isomers separated by ADcolumn, hexane/ EIOH 3:1, 1 mL/min collected in a ratio of A : B = 3 : 4 RCP of each compound > 98%

Figure 10. HPLC profiles of diastereomers of [ 99mTc]TRODAT-1: purified peak A and peak B. The positions of peak A and peak B on HPLC profiles were confirmed by the corresponding Re complexes.52

binding studies (Ki = 8.42 and 13.87 nM for Re TRODAT-1A and B , respecti vely, see F igure 9). The lipophilicity is higher for isomer A than for B [partition coefficient (PC) (a measure of lipophilicity) is 305 and 229, respectively]. X-ray structure analyses of the ReO complexes sho wed that the y are both syn-comple xes, yet due to the chirality of the tropan yl moiety, they are diastereomers. SPECT imaging studies in a female baboon using [99mTc]TRODAT-1 diastereomers showed that peak A displayed the highest uptake and contrast in caudate and putamen areas, w here DATs are known to be located. The racemic mixture pro vided high-quality images comparable to that obtained with the pure peak A. Ev en though peak B ga ve higher striatum/cerebellum ratios in rats, SPECT images in baboon did not exhibit a better image. This is most lik ely due to the lower brain uptak e of peak B , w hich led to a lo wer count rate in the brain. The results suggest that higher brain uptake may be more impor tant than higher target

to nontarget ratio. 52 The separation of diastereomers of [99mTc]TRODAT-1 and their Re complexes clearly indicated the impor tant roles the stereochemistr y of these diastereomers pla yed in af fecting in vi vo biodistribution. For de veloping receptor or site-specif ic imaging agents, it is important to consider the potential mixture of diastereomers and the confounding f actors that may influence image inter pretation. Clinical studies of [99mTc]TRODAT-1 clearly suggest that this agent shows promise as a diagnostic tool for PD and other CNS neurodegenerative diseases, for w hich a reduction of dopamine neurons is indicated. 67 Additional 99mTc tropane deri vatives ha ve been reported, but none has been sho wn to ha ve better in 99m vivo imaging proper ties in humans than [ Tc] TRODAT-1.70–72,75–77 In the past few years, despite a large number of brain receptor specif ic compounds ha ving been prepared and tested, none showed any success in imaging brain receptor or binding sites. 45,49 The reasons for this f ailure are not known at this time.

RENAL IMAGING AGENTS The utility of TcvO center core is not limited to the preparation of 99mTc brain imaging agents. The most commonly used renal imaging agent is the 99mTc-MAG3 (Bertiatide®) (Figure 11). 78 The chelating g roup consists of 1 sulfur and 3 nitro gen atoms (SN3) bound to a TcvO center core. The core is relatively stable, and the addition of the e xtra acid functional g roup mak es the comple x hydrophilic and readily secreted through kidney. In addition to 99mTc-MAG3, 99mTc-EC (see Figure 11), on which both of the eth yl ester g roups of ECD (see F igure 5) are hydrolyzed to two acid g roups, is also successfull y used as a kidney imaging agent. 79,80

Radiochemistry of Single-Photon Emission Computer Tomography

pentaacetic acid (DTPA), and DOTA. All of these form multiple bonds between indium and oxygen or nitrogen (see Chapter 23, “Ne wer Bioconjugation Methods” b y Meares.) Currently, 111In indium-o xine is a commerciall y available star ting material for labeling peptides or small molecules. In the nuclear medicine laborator y, it is often used for labeling white blood cells. After labeling, the 111In white blood cells are re-injected into the same patients for detecting infection sites. Ho wever, 111 In indium-oxine can also serve as an intermediate for labeling peptides. When peptides are conjugated with a strong metal chelating g roup such as DTP A or DO TA moiety, 111In is transferred from the oxine to other chelating g roups on the peptides b y for ming stronger and more stab le metal comple xes with the peptides, such as Octreotide (see Figures 12 and 13). Somatostatins are a group of neuropeptides derived from their precursor, pro-somatostatin, which is mainly expressed in nor mal endocrine, gastro-intestinal, immune and neuronal cells in the brain, and peripheral organs. Somatostatin e xerts its biolo gical actions b y interacting w ith cell membrane-bound

CHELATION OF RADIOMETALS—111IN Indium-111 ( 111In) is a single photon-emitting isotope (T1/2 = 2.83 d , γ-rays 171 and 245 K eV, Group III atoms). It is also a metal ion, w hich requires metal chelation for changing ph ysical and chemical properties. It contains a +3 charge and requires the formation of comple xes through chelation reactions. It prefers to for m coordinate covalent bonds with “hard” donor atoms, such as o xygen and nitro gen. The most commonly used chelating agents for preparation of In3+ radiopharmaceuticals are o xine, dieth ylenetriamine

S

O

CH2- COOH

H

NH O

N O N Tc S S 99mTc-EC(LL)

HOOC

Tc O

N

99mTc-MAG3

N O

COOH

(Bertiatide®) Figure 11. Chemical structures of 99mTc-MAG3 and 99mTc-EC. Both were successfully tested, and they are currently being used as kidney imaging agents.

111InCI

3

solution N

333

DTPAconjugated Peptides

N

111In-DTPA-Peptides

In

OH

N

8-Hydroxyquinoline (Oxine) N

Trans-chelation Reaction

Intermediate In complex

D-Phe-c[Cys-Phe-D-Trp-Lys-Thr-Cys]-Thr-OL (BOC)2O D-Phe-c[Cys-Phe-D-Trp-Lys(BOC)-Thr-Cys]-Thr-OL DTPA-dianhydride

DTPA-D-Phe-c[Cys-Phe-D-Trp-Lys(BOC)-Thr-Cys]-Thr-OL TFA DTPA-D-Phe-c[Cys-Phe-D-Trp-Lys-Thr-Cys]-Thr-OL In(Oxine)3 DTPA-D-Phe-c[Cys-Phe-D-Trp-Lys-Thr-Cys]-Thr-OL

Figure 12. Labeling of 111In-Octreotide, a somatostatin receptor imaging agent (111In-Octreoscan®), by using metal exchange reaction between 111In-Oxine and DTPA-Octreotide.

334

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

O O

N N O In O N O O O

111In-DTPA

D-Phe

Phe

DPhe

D-Trp O

Cys H H O NH N N N O H O O S S O HHN H N Lys N N HO O H O HO Thr(ol) Cys HO Thr O O N H

O O

N N

In

N N

111In-DTPA-D-Phe-c[Cys-Phe-D-Trp-Lys-Thr-Cys]-Thr-OL

Cys H O N

O N H

O NH2

OH

O O

111In-DOTA

HO HO Thr(ol)

OS H N

DTrp

Tyr

S

O Cys

H N N H O O

N H HO

NH O

HN H N Lys O

NH2

Thr

111In-DOTA-D-Phe-c[Cys-Phe-D-Trp-Lys-Thr-Cys]-Thr-OL

Figure 13. Labeling of Octreotide (111In-Octreoscan®) can be achieved by using a modified peptide containing either DTPA or DOTA as the chelating group for complexing with 111In.

G-protein-coupled receptors. These somatostatin receptors (SSTRs) are functionally coupled to multiple cellular ef fector systems (there are f ive different subtypes of SSTRs). Ov erexpression of SSTRs is strongl y associated with v arious endocrine and nonendocrine tumors. Cur rently, FD A-approved SSTR-tar geting imaging agents, 111In-Octreoscan® (see Figures 12 and 13) and 99mTc-NeoTect ®, are peptide based and designed for SPECT imaging of tumors with a high expression of SSTRs. In the process of membrane signal transduction involving G-protein-coupled receptors, there are tw o major types of ligand-receptor interactions. When an agonist ligand binds to G-protein-coupled receptors, in the case of SSTR imaging agents, such as Octreotide (111In-Octreoscan®), the binding triggers a chain of events of second-messenger, leading to other cascades of biochemical changes. When an antagonist ligand binds to G-protein-coupled receptors, it occupies the binding site, prevents further binding, but activates no other biochemical events. The binding of antagonists to the receptor’s binding pock et could be more fle xible and with higher af finity, pro viding an enhanced signal-to-noise ratio at the tar get site. Both agonist- and antagonistbased imaging agents could form a receptor-ligand complex. This event could subsequently lead to internalization, trapping inside the cells. As such the labeled imaging agent is sequestered at the receptor binding sites pro viding a strong signal for imaging. Recently, it has been suggested that instead of agonists of SSTR, the antagonists might ser ve as a better tar geting ligand for imaging SSTR receptors in tumors. 81 The 111 In-DTPA labeled [T yr(3),Thr(8)]-octreotide, an antagonist ligand, appeared to label considerably more binding sites and displa yed more prominent signalto-noise ratio. The clinical application of such antagonists of endocrine tumors remains to be e valuated.

Many other types of endocrine peptides ha ve been tested as potential tumor diagnostic imaging and therapeutic agents. 82–84 It is an emer ging f ield in which the chelating chemistry of both Tc and In will play important roles in de veloping new probes tar geting tumor surf ace binding sites.

ACKNOWLEDGMENT The author thanks Dr . Karl Ploessl for his technical assistance in assembling the figures. He also thanks Dr. Carita Huang and Ms. Janet Bar row for their editorial assistance.

REFERENCES 1. Steigman J, Eckelman WC. The chemistry of technetium in medicine. Washington, DC: National Academy; 1992. 2. Eckelman WC, Volkert WA. In vi vo chemistr y of technetium-99mlabeled chelates. Int J Rad Appl Instr um [A] Appl Radiat Isot 1982;33:945–51. 3. Davison A, Jones AG. The chemistr y of technetium(V). Int J Appl Rad Isot 1982;33:875–81. 4. Jones AG, Davison A. The chemistry of technetium I, II, III and IV . Int J Appl Radiat Isot 1982;33:867–74. 5. Liu S. The role of coordination chemistry in the development of targetspecific radiopharmaceuticals. Chem Soc Rev 2004;33:445–61. 6. Mease RC, Lambert C. Newer methods of labeling diagnostic agents with Tc-99m. Semin Nucl Med 2001;31:278–85. 7. Alberto R. [Tc(CO)(3)](+) chemistr y: a promising ne w concept for SPET? Eur J Nucl Med Mol Imaging 2003;30:1299–302. 8. Alberto R, Schib li R, Waibel R, et al. Basic aqueous chemistr y of [M(OH2)3(CO)3]+ (M=Re, Tc) directed to wards radiophar maceutical application. Coord Chem Rev 1999;190–192:901–19. 9. Berman DS, Sha w LJ, Hachamo vitch R, et al. Comparati ve use of radionuclide stress testing, coronary artery calcium scanning, and noninvasive coronar y angiography for diagnostic and pro gnostic cardiac assessment. Semin Nucl Med 2007;37:2–16. 10. Makaryus AN, Diamond JA. Nuclear stress testing in elderly patients: a review of its use in the assessment of cardiac risk, par ticularly in patients undergoing preoperative risk assessment. Drugs Aging 2007;24:467–79.

Radiochemistry of Single-Photon Emission Computer Tomography

11. Madsen MT . Recent adv ances in SPECT imaging. J Nucl Med 2007;48:661–73. 12. Lodge MA, Bengel FM. Methodolo gy for quantifying absolute myocardial perfusion with PET and SPECT . Cur r Cardiol Rep 2007;9:121–8. 13. Beller GA, Ber gmann SR. My ocardial perfusion imaging agents: SPECT and PET. J Nucl Cardiol 2004;11:71–86. 14. Berman DS, Kiat H, Maddahi J . The new 99mTc m yocardial perfusion imaging agents: 99mTc-sestamibi and 99mTc-teboroxime. Circulation 1991;84:I7–21. 15. Ayalew A, Marie PY , Menu P , et al. A comparison of the o verall first-pass kinetics of thallium-201 and technetium-99m MIBI in normoxic and low-flow ischaemic myocardium. Eur J Nucl Med 2000;27:1632–40. 16. Lodge MA, Braess H, Mahmoud F , et al. De velopments in nuclear cardiology: transition from single photon emission computed tomography to positron emission tomography-computed tomography. J Invasive Cardiol 2005;17:491–6. 17. Kronauge JF, Leon AS, Verdera ES, et al. Human pharmacokinetics and radiopharmacological studies of the m yocardial perfusion agent: technetium (2-carbomethoxy- 2-isocyano-propane)6+. J Nucl Med 1992;33:1949–57. 18. Kronauge JF, Noska MA, Da vison A, et al. Interspecies v ariation in biodistribution of technetium (2-carbometho xy-2-isocyanopropane)6+. J Nucl Med 1992;33:1357–65. 19. Verdera S, Leon A, Kronauge JF, et al. Radiopharmacological studies of 99mTc-CPI: experience with isolated rat atrial tissue. Nucl Med Biol 1993;20:127–32. 20. Liu S. Ether and cro wn ether-containing cationic 99mTc comple xes useful as radiophar maceuticals for hear t imaging. Dalton Trans 2007;1183–93. 21. Pasqualini R, Duatti A, Bellande E, et al. Bis(dithiocarbamato) nitrido technetium-99m radiopharmaceuticals: a class of neutral m yocardial imaging agents. J Nucl Med 1994;35:334–41. 22. Dilworth JR, Archer CM, Latham IA, et al. Lipophilic technetium99m-nitride radiophar maceuticals as potential m yocardial imaging agents. Nucl Med Biol 1991;18:547–50. 23. Mindt TL, Struthers H, Brans L, et al. \' 'Click to Chelate\'': synthesis and installation of metal chelates into biomolecules in a single step. J Am Chem Soc 2006;128:15096–7. 24. Banerjee SR, Maresca KP, Francesconi L, et al. New directions in the coordination chemistr y of (99m)Tc: a reflection on technetium core structures and a strate gy for ne w chelate design. Nucl Med Biol 2005;32:1–20. 25. Alberto R, P ak JK, v an Staveren D, et al. Mono-, bi-, or tridentate ligands? The labeling of peptides with 99mTc-carbonyls versatile synthetic approach to ne w bifunctional chelating agents tailor made for labeling with the f ac-[M(CO)(3)](+) core (M = Tc, (99m)Tc, Re): synthesis, in vitro, and in vi vo beha vior of the model complex [M(APPA)(CO)(3)] (APPA = [(5-amino-pentyl)pyridin-2-yl-methyl-amino]-acetic acid). Biopol ymers 2004;76: 324–33. 26. Deutsch E, Ketring AR, Libson K, et al. The Noah’s Ark experiment: species dependent biodistributions of cationic 99mTc comple xes. Int J Rad Appl Instrum B 1989;16:191–232. 27. Cutler CS, Giron MC, Reicher t DE, et al. Ev aluation of gallium-68 tris(2-mercaptobenzyl)amine: a complex with brain and myocardial uptake. Nucl Med Biol 1999;26:305–16. 28. Wong DF, Pomper MG. Predicting the success of a radiophar maceutical for in vivo imaging of central ner vous system neuroreceptor systems. Mol Imaging Biol 2003;5:350–62. 29. Mathis CA, Lopresti BJ , Klunk WE. Impact of am yloid imaging on drug de velopment in Alzheimer’s disease. Nucl Med Biol 2007;34:809–22. 30. Eckelman WC, Kilbour n MR, Mathis CA. Radiophar maceutical space. Nucl Med Biol 2006;33:829.

335

31. Neirinckx RD , Canning LR, Piper IM, et al. Technetium-99m-d, l-HM-PAO: a ne w radiopharmaceutical for SPECT imaging of regional cerebral b lood perfusion. J Nucl Med 1987; 28:191–202. 32. Neirinckx RD. Evaluation of regional cerebral blood flow with 99mTcd, 1 HM-PAO and SPECT. Neurosurg Rev 1987;10:181–4. 33. Waldemar G, Walovitch RC, Andersen AR, et al. [ 99mTc]-bicisate (neurolite) SPECT brain imaging and co gnitive impair ment in dementia of the Alzheimer type: a blinded read of image sets from a multicenter SPECT trial. J Cereb Blood Flo w Metab 1994;1:S99–105. 34. Walovitch RC, Hill TC, Gar rity ST , et al. Characterization of technetium-99m-L,L-ECD for brain perfusion imaging, P art 1: pharmacology of technetium-99m ECD in non-human primates. J Nucl Med 1989;30:1892–901. 35. Walovitch RC, Williams SJ, Lafrance ND . Radiolabeled agents for SPECT imaging of brain perfusion. Nucl Med Biol Int J RadAppl Instrum B 1990;17:77–83. 36. Walovitch RC, Franceschi M, Picard M, et al. Metabolism of [99mTc]L,L-ethyl c ysteinate dimer in health y v olunteers. Neurophar macology 1991;30:283–92. 37. Walovitch RC, Cheesman EH, Maheu LJ , Hall KM. Studies of the retention mechanism of the brain perfusion imaging agent [99mTc]bicisate ([99mTc]-ECD). J Cereb Blood Flow Metab 1994;1:S4–11. 38. Treher EN , F rancesconi LC, Gougoutas JZ, et al. Monocapped tris(dioxime) complexes of technetium (III): synthesis and str uctural characterization of TcX(dioxime)3B-R (X = Cl, Br; dioxime = dimeth ylglyoxime, c yclohexanedionedioxime; R = CH 3, C4H9). Inorg Chem 1989;28:3411. 39. Ramalingam K, Jurisson SS, K uczynski BL, et al. B ATO complexes derived from dimethoxy dioximes: synthesis, characterization and biodistribution. Nucl Med Biol 1995;22:625–34. 40. Mountz JM, Liu HG, Deutsch G. Neuroimaging in cerebrovascular disorders: measurement of cerebral physiology after stroke and assessment of strok e reco very. Semin Nucl Med 2003; 33:56–76. 41. Van Heer tum RL, Tikofsky RS. P ositron emission tomo graphy and single-photon emission computed tomo graphy brain imaging in the evaluation of dementia. Semin Nucl Med 2003;33:77–85. 42. Lee HY, Paeng JC, Lee DS, et al. Efficacy assessment of cerebral arterial bypass surgery using statistical parametric mapping and probabilistic brain atlas on basal/acetazolamide brain perfusion SPECT. J Nucl Med 2004;45:202–6. 43. Ito H, Inoue K, Goto R, et al. Database of nor mal human cerebral blood flow measured by SPECT: I. Comparison between I-123IMP, Tc-99m-HMPAO, and Tc-99m-ECD as refer red with O-15 labeled water PET and voxel-based morphometry. Ann Nucl Med 2006;20:131–8. 44. Zerarka S, Pellerin L, Slosman D, Magistretti PJ. Astrocytes as a predominant cellular site of (99m)Tc-HMP AO retention. J Cereb Blood Flow Metab 2001;21:456–68. 45. Johannsen B, Pietzsch HJ . Development of technetium-99m-based CNS receptor ligands: have there been any advances? Eur J Nucl Med Mol Imaging 2002;29:263–75. 46. Jurisson SS, L ydon JD. Potential technetium small molecule radiopharmaceuticals. Chem Re v (W ashington, DC) 1999; 99:2205–18. 47. Hom RK, Katzenellenbo gen J A. Technetium-99m-labeled receptor specific small-molecule radiophar maceuticals: recent de velopments and encouraging results. Nucl Med Biol 1997;24:485–98. 48. Kung HF, Kung MP, Wey SP, et al. Clinical acceptance of a molecular imaging agent: a long march with [(99m)Tc] TRODAT. Nucl Med Biol 2007;34:787–9. 49. Schibli R, Schubiger P A. Cur rent use and future potential of organometallic radiopharmaceuticals. Eur J Nucl Med Mol Imaging 2002;29:1529–42.

336

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

50. Lever SZ, Baidoo KE, Mahmood A, et al. No vel technetium ligands with affinity for the muscarinic cholinergic receptor. Nucl Med Biol 1994;21:157–64. 51. Minutolo F , Katzenellenbo gen J A. A con venient three-component synthesis of substituted cyclopentadienyl tricarbonyl rhenium complexes. J Am Chem Soc 1998;120:4514–5. 52. Meegalla SK, Plössl K, Kung M-P, et al. Specificity of diastereomers of [ 99mTc]TRODAT-1 as dopamine transpor ter imaging agents. J Med Chem 1998;41:428–36. 53. Oya S, Plössl K, K ung M-P, et al. Small and neutral TcO(V) B AT, bisaminoethanethiol (N 2S2) comple xes for de veloping ne w brain imaging agents. Nucl Med Biol 1998;25:135–40. 54. Frost JJ, Rosier AJ, Reich SG, et al. P ositron emission tomo graphic imaging of the dopamine transpor ter with [ 11C]-WIN 35,428 reveals marked declines in mild P arkinson’s disease. Ann Neurol 1993;34:423–31. 55. Innis RB, Seibyl JP, Scanley BE, et al. Single photon emission computed tomo graphic imaging demonstrates loss of striatal dopamine transporters in Parkinson’s disease. Proc Natl Acad Sci U S A 1993;90:11965–9. 56. Innis RB. Single photon emission tomo graphy imaging of dopamine terminal inner vation: a potential clinical tool in P arkinson’s disease. Eur J Nucl Med 1994;21:1–5. 57. Seibyl JP , Marek KL, Quinlan D , et al. Decreased singlephoton emission computed tomo graphic [ 123I]beta-CIT striatal uptake correlates with symptom severity in Parkinson’s disease. Ann Neurol 1995;38:589–98. 58. Seibyl JP, Marek K, Shef f K, et al. Test/retest reproducibility of iodine-123-betaCIT SPECT brain measurement of dopamine transporters in Parkinson’s patients. J Nucl Med 1997;38:1453–9. 59. Abi-Dargham A, Gandelman MS, DeErausquin GA, et al. SPECT imaging of dopamine transporters in human brain with iodine-123fluoroalkyl analogs of beta-CIT. J Nucl Med 1996;37:1129–33. 60. Booij J , Tissingh G, Winogrodzka A, et al. Practical benef it of [123I]FP-CIT SPET in the demonstration of the dopaminer gic deficit in Parkinson’s disease. Eur J Nucl Med 1997;24:68–71. 61. Goodman MM, Kung M-P, Kabalka GW, et al. Synthesis and characterization of radioiodinated N-(3-iodopropen-1-yl)-2b-carbomethoxy-3b-(4-chlorophenyl)tropanes: potential dopamine reuptake site imaging agents. J Med Chem 1994;37:1535–42. 62. Mozley PD, Stubbs JB, Kim H-J, et al. Dosimetr y of an iodine-123labeled tropane to image dopamine transpor ters. J Nucl Med 1996;37:151–9. 63. Tatsch K, Schwarz J, Mozley PD, et al. Relationship between clinical features of P arkinson’s disease and presynaptic dopamine transporter binding assessed with [123I]IPT and single-photon emission tomography. Eur J Nucl Med 1997;24:415–21. 64. Fischman AJ, Bonab AA, Babich JW, et al. Rapid detection of P arkinson’s disease b y SPECT with altropane: a selecti ve ligand for dopamine transporters. Synapse 1998;29:128–41. 65. Madras BK, Gracz LM, F ahey MA, et al. Altropane, a SPECT or PET imaging probe for dopamine neurons: III. human dopamine transporter in postmor tem nor mal and P arkinson’s diseased brain. Synapse 1998;29:116–27. 66. Kung HF, Kim H-J , K ung M-P, et al. Imaging of dopamine transporters in humans with technetium-99m TRODAT-1. Eur J Nucl Med 1996;23:1527–30.

67. Mozley PD, Schneider JS, Acton PD, et al. Binding of [99mTc]TRODAT-1 to dopamine transpor ters in patients with parkinson’ s disease and in healthy volunteers. J Nucl Med 2000;41:584–9. 68. Acton PD, Meyer PT, Mozley PD, et al. Simplif ied quantif ications of dopamine transporters in humans using [99mTc]TRODAT-1 and single photon emission tomography. Eur J Nucl Med 2000;27:1714–8. 69. Meltzer PC, Blundell P, Jones AG, et al. A technetium-99m SPECT imaging agent which targets the dopamine transporter in primate brain. J Med Chem 1997;40:1835–44. 70. Vanbilloen HP , Kief fer DM, Cle ynhens BJ , et al. Ev aluation of (99m)Tc-labeled tropanes with alkyl substituents on the 3betaphenyl ring as potential dopamine transpor ter tracers. Nucl Med Biol 2006;33:413–8. 71. Vanbilloen HP, Kieffer D, Cleynhens BJ, et al. Development and biological evaluation of 99mTc-BAT-tropane esters. Nucl Med Biol 2005;32:607–12. 72. Kieffer DM, Cleynhens BJ, Vanbilloen HP, et al. Synthesis and biological e valuation of a technetium-99m(I)-tricarbon yl-labelled phenyltropane derivative. Bioorg Med Chem Lett 2006;16:382–6. 73. Katzenellenbogen J A, Minutolo F , Spradau TW, Skaddan MB . Preserving bioactivity of small molecules labeled with technetium and rhenium: an organometallic approach. Q J Nucl Med 1998;42:3. 74. Alberto R, Egli A, Schibli R, et al. F rom [TcO4] to an or ganometallic aquo-ion: synthesis and chemistr y of [Tc(OH2)3(CO)3]. Q J Nucl Med 1998;42:9. 75. Meltzer PC, Blundell P , Zona T, et al. A second-generation 99m technetium single photon emission computed tomo graphy agent that provides in vi vo images of the dopamine transpor ter in primate brain. J Med Chem 2003;46:3483–96. 76. Kieffer DM, Vanbilloen HP, Cleynhens BJ, et al. Biological evaluation of a technetium-99m-labeled integrated tropane-BAT and its piperidine congener as potential dopamine transpor ter imaging agents. Nucl Med Biol 2006;33:125–33. 77. Cleynhens BJ , de Groot TJ, Vanbilloen HP, et al. Technetium-99m labelled integrated tropane-BAT as a potential dopamine transporter tracer. Bioorg Med Chem 2005;13:1053–8. 78. Fritzberg AR, Kasina S, Eshima D , Johnson DL. Synthesis and biological e valuation of technetium-99m MA G3 as a hippuran replacement. J Nucl Med 1986;27:111–6. 79. Verbruggen AM, Nosco DL, Van Nerom CG, et al. Technetium-99mL,L-ethylenedicysteine: a renal imaging agent. I. Labeling and evaluation in animals. J Nucl Med 1992;33:551–7. 80. Verbruggen AM. Radiophar maceuticals: state of the ar t. Eur J Nucl Med 1990;17:346–64. 81. Ginj M, Zhang H, Waser B, et al. Radiolabeled somatostatin receptor antagonists are preferable to agonists for in vi vo peptide receptor targeting of tumors. Proc Natl Acad Sci U S A 2006;103: 16436–41. 82. Mariani G, Erba P A, Signore A. Receptor-mediated tumor tar geting with radiolabeled peptides: there is more to it than somatostatin analogs. J Nucl Med 2006;47:1904–7. 83. Rufini V, Calcagni ML, Baum RP. Imaging of neuroendocrine tumors. Semin Nucl Med 2006;36:228–47. 84. Wild D , Behe M, Wicki A, et al. [L ys40(Ahx-DTPA111In)NH2]exendin-4, a very promising ligand for glucagon-lik e peptide-1 (GLP-1) receptor tar geting. J Nucl Med 2006;47:2025–33.

22 NANOCHEMISTRY

FOR

MOLECULAR IMAGING

YUN XING, PHD AND JIANGHONG RAO, PHD

Nanotechnology is an area of science de voted to the manipulation of atoms and molecules leading to the construction of structures in the size range of nanometer scale (usuall y 100 nm or smaller), w hich retain unique proper ties not seen in bulk materials or single molecules. Recent adv ances in nanotechnolo gy ha ve led to successful applications in medicine and hold the promise of re volutionary changes in disease detection, monitoring, and treatment. Successful examples include quantum dots (fluorescence imaging), iron oxide nanopar ticles (MRI), gold nanopar ticles (Raman scattering), carbon nanotubes, dendrimers, and others. In this chapter, we will use semiconductor quantum dots as an e xample to show the chemistry of making nanostructures for molecular imaging purposes. We will cover the synthesis, w ater solubilization, biofunctionalization, and animal imaging applications. In the end, w e will also discuss the limitations, issues, and perspectives for nanostr ucture-based molecular imaging. The development of a wide range of nanoscale technologies is be ginning to change the foundations of disease diagnosis, treatment, and pre vention. These technological innovations, refer red to as nanomedicine by the National Institutes of Health (Bethesda, MD , USA), have great potential to lead to major advances in disease detection, diagnosis, and treatment, and eventually personalized therapy and disease management. The basic rationale is that metal, semiconductor , and pol ymeric particles have novel optical, electronic, magnetic, and/or str uctural proper ties that are often not a vailable from indi vidual molecules or bulk solids. 1 Recent research has developed functional nanoparticles that are covalently link ed to biolo gical molecules such as peptides, proteins, nucleic acids, or small-molecule ligands.2,3 Medical applications ha ve also appeared , such as the use of super paramagnetic iron o xide

nanoparticles as a contrast agent for l ymph node prostate cancer detection 4 and the use of pol ymeric nanoparticles for targeted-gene delivery to tumor vasculatures.5 Depending on the pur pose, nanostr uctures for medicine come in v arious materials and shapes. Ho wever, the general route of chemistr y is similar, from the synthesis of the core material, w ater solubilization (for structures made of h ydrophobic materials), to biofunctionalization with tar geting moieties or therapeutic agents. In this chapter, we have chosen to use semiconductor quantum dots (QDs) as an e xample to illustrate nanochemistry for molecular imaging. We will co ver QD core synthesis, w ater solubilization, biolo gical functionalization, and animal imaging applications and finish with a discussion on the issues/limitations and future perspecti ves. We hope this will pro vide generic understanding of nanochemistr y for molecular imaging and ser ve as a platfor m for the de velopment of other types of nanostructures for imaging pur poses. Semiconductor QDs ha ve capti vated scientists and engineers over the past two decades owing to their fascinating optical and electronic proper ties, which are not a vailable from either isolated molecules or bulk solids. Recent research has stimulated considerable interest in developing these quantum-conf ined nanocrystals as optical probes for biomedical applications. Compared with small or ganic dyes and fluorescent proteins, semiconductor QDs of fer several unique advantages, such as size- and compositiontunable emission from visible to infrared wavelengths, large absorption coef ficients across a wide spectral range, and very high levels of brightness and photostability.6 The longterm photostability and superior brightness of QDs mak e them ideal candidates for live animal targeting and imaging. In addition, recent studies ha ve sho wn that QDs ha ve two-photon e xcitation cross sections of magnitude lar ger than those of conventional fluorescent probes now in use, a further plus for deep tissue imaging. 7 337

338

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

QD STRUCTURE Quantum dots are some what spherical nanocr ystals in the size range of 1 to 10 nm in diameter (F igure 1A). 8,9 Semiconductor nanocr ystals can also be produced with other shapes such as rods and tetrapods, 10 but spherical QDs are often widel y used for biolo gical applications. These par ticles are generall y made from hundreds to thousands of atoms of group II and VI elements [eg, cadmium selenide (CdSe) and cadmium telluride (CdTe)] or group III and V elements [e g, indium phosphide (InP) and indium arsenide (InAs)]. Recent adv ances ha ve allowed the precise control of par ticle size, shape (dots, rods, or tetrapods),11–14 and internal structure (core shell, gradient alloy or homogeneous alloy).15,16

The no vel optical proper ty of QDs arises from the so-called “quantum conf inement” ef fect of the semiconductor materials concer ning the size- and compositiondependence of the semiconductor bandgap ener gy. This effect is readily observed when one or more dimensions of a semiconductor are reduced to the nanometer re gime. Absorption of a photon with ener gy above the semiconductor bandgap ener gy results in the creation of an electron-hole pair (or e xcitation). The absor ption has an increased probability at higher ener gies (ie, shorter wavelengths) and results in a broadband absor ption spectrum, in mark ed contrast to standard fluorophores (Figure 1C). F or nanocr ystals smaller than the so-called Bohr excitation radius (a fe w nanometers), ener gy levels are quantized, with v alues directl y related to the size of C

O

O

O

P O

OH O

P

C

C O H

P O O

P

O

O

O P

O

C NH

O

Quantum dot

P

O P

C NH

O P

O

P

H C O C

O C N

OH

C O

O

0.6 0.4 0.2

500

600

700

1.0

O

P

O

P

O P

O

P

O

0.8

400

PEG (– CH2 –CH2 – O –)n

C O H

C

A

0.0

H C O

P

O

O

P O

O

O

O

O P

O

OH

C

H

C

OH

Amphiphilic polymer coating

B

0.8 0.6 0.4 0.2

Affinity ligands: antibody, peptide, small molecule drug, inhibitor, etc

400

D

O

O

O

O

H N

Normalized intensity

OH

H C O

C NH

C O H

OH

Normalized intensity

N H C

C

C

B

QD capping ligand TOPO

NH

O

O

1.0 Normalized intensity

O

OH

C NH

C O H

O

O

C

C

OH

H C O

A

1.2

500 600 Wavelength (nm)

700

QD 608-streptavidin

1 QD 608-streptavidin with antifade medium

0.8 0.6

Alexa 488-streptavidin with antifade medium

0.4 0.2

Alexa 488-streptavidin

0 0

20

40

60

80 100 120 140 160 180 Time (s)

Figure 1. Quantum dot structure and novel optical properties. A, Schematic of a multifunctional quantum dot (QD) probe containing the capping ligand trioctylphosphine oxide (TOPO), an encapsulating copolymer layer, tumor-targeting ligands (such as peptides, antibodies or smallmolecule inhibitors) and polyethylene glycol (PEG). Reproduced with permission from Gao X et al.20 B, Size-tunable emission spectra of QDs. This image shows 10 distinguishable emission colors of ZnS-capped CdSe QDs excited with a near-UV lamp. From left to right (blue to red), the emission maxima are located at 443, 473, 481, 500, 518, 543, 565, 587, 610, and 655 nm. Reproduced with permission from Nie S et al.1 C, Excitation (dotted line) and fluorescence (solid line) spectra of fluorescein (top) and a typical water-soluble QD (bottom). Reproduced with permission from Bruchez MJ et al.38 D, Superior photostability of QDs as compared to organic dyes. Reproduced with permission from Wu X et al.37

Nanochemistry for Molecular Imaging

the nanoparticle. The radiati ve recombination of an excitation (characterized b y a long lifetime, > 10 ns) 17 leads to the emission of a photon in a nar row, symmetric energy band. This dependence of light emission on particle size allows the development of new fluorescence emitters with precisel y tuned emission w avelengths (F igure 1B). For example, the semiconductor CdSe has a bulk bandgap of 1.7 eV (corresponding to 730 nm light emission). Quantum dots of this material can be tuned to emit between 450 and 650 nm b y changing the nanocr ystal diameter from 2 to 7 nm. The composition of the material ma y also be used as a parameter to alter the bandgap of a semiconductor. Quantum dots with a diameter of 5 nm can be tuned to emit between 610 and 800 nm b y varying the molar ratio of Se and Te in the alloy CdSexTe1−x.18 The classic and most commonly used QDs consist of a cadmium selenide (CdSe) core and a shell la yer made of zinc sulfide (ZnS) or cadmium sulfide (CdS). Fluorescent properties are determined by the core materials, and the shell layer removes surface defects and prevents nonradiative deca y (radiationless dissipation of the e xcited energy), leading to a signif icant improvement in the particle stability and fluorescence quantum yields. F or biological imaging applications, these hydrophobic dots can be made w ater-soluble by e xchanging with bifunctional ligands (mostly thiol and phosphine mono and multidentate ligands) or b y using amphiphilic pol ymers that contain both a h ydrophobic se gment or side chain (mostl y hydrocarbons) and a hydrophilic segment or group (such as pol yethylene gl ycol [PEG] or multiple carbo xylate groups). In addition, biomolecules such as antibodies and peptides can be attached to the QDs to achie ve specif ic labeling and targeting (Figure 1A).

NOVEL OPTICAL PROPERTIES As briefl y noted abo ve, QDs are made from inor ganic semiconductors and ha ve no vel optical proper ties that can be used to optimize thesignal-to-background ratio in fluorescence imaging. In comparison with or ganic dyes and fluorescent proteins, QDs ha ve se veral adv antages and unique applications. F irst, QDs ha ve v ery lar ge molar e xtinction coef ficients in the order of 0.5–5 × 106 M−1cm−1,19 about 10–50 times lar ger than that of organic dyes (5–10 × 104 M−1cm−1). Therefore, QDs are able to absorb 10–50 times more photons than or ganic dyes at the same e xcitation photon flux (ie, the number of incident photons per unit area), leading to a signif icant improvement in the probe brightness; this allows for brighter emission under photon-limited in vi vo conditions (where light intensities are se verely attenuated b y

339

scattering and absorption). In theory, the lifetime-limited emission rates for single QDs are 5 to 10 times slower than those of small or ganic dyes because of their longer excited state lifetimes (20 to 50 ns vs less than 10 ns). In practice, however, fluorescence imaging usually operates under absorption-limited conditions in which the rate of absorption is the main limiting f actor of fluorescence emission (vs. the emission rate of the fluorophore). As a result, indi vidual QDs ha ve been found to be 10 to 20 times brighter than or ganic dyes.20 Second, QDs are several thousand times more stable against photobleaching (the loss of fluorescence due to photo-induced chemical damages) than organic dyes (Figure 1D) and are thus well suited for continuous-tracking studies o ver a long period of time. In addition, the relati vely longer e xcited state lifetimes of QDs can be used to separate the QD fluorescence from backg round fluorescence, in a technique kno wn as time-domain imaging 21; because QDs emit light slowly enough, most of the backg round autofluorescence emission is o ver by the time QD emission occurs. Third, the lar ge Stokes shifts of QDs (measured by the distance betw een the e xcitation and emission peaks) can be used to further improve detection sensitivity. This factor becomes especially important for in vivo molecular imaging due to the high autofluorescence background often seen in comple x biomedical specimens. The Stokes shifts of semiconductor QDs can be as large as 300 to 400 nm, depending on the w avelength of the e xcitation light. Or ganic dy e signals with a small Stokes shift are often buried b y strong tissue autofluorescence, w hereas QD signals with a lar ge Stokes shift are clearl y reco gnizable abo ve the background. This de gree of “color contrast” is onl y available to QD probes because the signals and background can be easil y separated b y wavelength-resolved or spectral imaging.20 A further advantage of QDs is that multicolor QD probes can be used to image and track multiple molecular targets simultaneously. This is a very desirable feature because most comple x human diseases such as cancer and atherosclerosis often in volve a number of genes and proteins rather than a single one. The ability to track a panel of molecular markers at the same time will allo w scientists to better understand , classify, and differentiate comple x human diseases than using a single biomark er each time. Multiple parameter imaging, ho wever, represents a signif icant challenge for magnetic resonance imaging (MRI), positron emission tomography (PET), computed X-ra y tomo graphy (CT), and related imaging modalities. By contrast, fluorescence optical imaging provides both signal intensity and wavelength infor mation, and multiple w avelengths or

340

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

colors can be resolved and imaged simultaneously (color imaging). Therefore, different molecular or cellular targets can be tagged with dif ferent colors. In this regard, QD probes are particularly attractive because their broad absorption prof iles allo w simultaneous e xcitation of multiple colors and their emission w avelengths can be continuously tuned by varying particle size and chemical composition. For or gan and v ascular imaging in w hich micrometer-sized par ticles could be used , opticall y encoded beads (pol ymer beads embedded with multicolor QDs at controlled ratios) could allo w multiplexed molecular profiling in vivo at high sensitivities.22,23 The most common scheme of QD production consists of four major steps: (1) synthesis of the QD core, most often CdSe, in a high-temperature organic solvent; (2) g rowth of an inor ganic shell (usuall y ZnS) on the core to protect the optical proper ties of the core; (3) water solubilization of QDs b y ligand e xchange or polymeric encapsulation; (4) functionalization of the QD surface with biologically active molecules. Each of the four steps is discussed in the follo wing section.

SYNTHESIS Core Synthesis The most effective and reproducible procedure for synthesis of monodisperse QD in volves the addition of semiconductor precursors to a liquid coordinating solvent at high temperature, f irst described by Murray and colleagues.24 In a typical synthesis of CdSe QDs, roomtemperature precursors, dimeth ylcadmium and elemental selenium dissolv ed in liquid trioctylphosphine (TOP), are s wiftly injected into hot (290 to 350°C) trioctylphosphine o xide (T OPO), immediatel y initiating nucleation of QD crystals. The presence of coordinating solvents, TOPO or TOP, prevents the for mation of bulk semiconductors by binding to the QD surf aces through their basic functional g roups (phosphines, phosphine oxides). The alkyl chains from the coordinating ligands extend away from the QD surf ace, rendering them sterically stable as colloids and dispersib le in many nonpolar solvents such as chlorofor m and he xane. By tuning the reaction parameters (precursors, coordinating ligands, etc), QDs with diameters between 2 and 8 nm and emission spanning the entire visib le spectr um can be easily synthesized with just one composition (CdSe). Quantum dots emitting all the way from 400 to 4000 nm can be made b y v arying the composition (ZnS, CdS, CdSe, CdTe, lead sulf ide (PbS), lead selenide (PbSe)) and their ratios in the allo ys.15,25–27 The quantum yield

(ratio of the number of photons emitted to the number of photons absorbed, essentially the emission efficiency of a fluorophore) of the nanocrystal core synthesized as above is relatively low (less than 10%) due to the presence of gap surf ace states arising from surf ace nonstoichiometry and unsaturated bonds. 24,28

Shell Synthesis Usually, a thin (a fe w atoms thick) shell of higher bandgap semiconductor material, such as ZnS or CdS, is epitaxially grown around the core to achieve better photo stability and higher quantum yield. A higher bandgap material eliminates the surf ace defects. This la yer improves quantum yield and protects the CdSe core from photooxidation, which is e xtremely impor tant for minimizing cytotoxicity as well as for enhancing photostability. To produce a ZnS shell, a solution of dimeth ylzinc and he xamethyldisilathiane (in tri-n-octylphosphine) is slowly added into the reaction solution of CdSe. Thickness of the shell is mediated by the amount of dimethylzinc and he xamethyldisilathiane injected into the solution. By carefull y designing the shell, quantum yields of QDs can be signif icantly improved (from 5 up to 90%). 28,29 The presence of a shell also improves photostability of QDs b y several orders of magnitude relative to conventional fluorophores.

WATER SOLUBILIZATION Quantum dots made in or ganic solvents are coated with alkyl chains that render solubility only in nonpolar organic solvents. To make QDs useful for biological purposes, it is essential to modify the surf ace of QDs to mak e them water-soluble. An ideal modification will meet the following criteria: (1) has good shelf stability , no flocculating over long ter m of storage; (2) ef ficiently con verts the insoluble QDs to w ater-soluble; (3) maintains the brightness, that is, high quantum yield of the QDs; and (4) keeps the overall size of the QDs small, best if belo w 10 nm. Unfor tunately, none of the cur rent coating strategies meet all four criteria, each with their adv antages and shortcomings. Different QD solubilization strate gies ha ve been devised over the past decade, which can be grouped into two major categories (Figure 2). The f irst one uses “ligand exchange” and involves the substitution of the native TOP/TOPO with bifunctional ligands, each presenting a surface-anchoring moiety to bind to the inor ganic QD surface and an opposing h ydrophilic end g roup that

NH -R

NH-R

OH

C C

P PP PPP

P

-R NH

O C

OH

O C

O C O C O C

OH

C O

O

C

H

peptides

dendrons

C

NH-R

Silica shell

O

OH

P

C O C CO

CO C

O C

OH

P

P P P P

HS

OH

C

Oligomeric phosphine

C O

HS

NH-R

NH

-R

hydrophilic domain Adhesive domain

N

N

O

N

N O

Si

Si

O

O

O N

O Si

Si

Si

COOH

COOH

COOH

O

O

N

Si

O NH O O NH O O NH

O

OH CO

COOH

CO OH

O

OH

OH

DHLA

OH

-R NH

OH

OH

HS

HS

P

P

P

SH

HS

OH

OH

C

P PP P

OH

P P P PP P

P P P P

NH-R

OH

R

O N

NH-R

C O C O C

CdSe

OH

C

P

NH -

SH

HS

OH

CO

ZnS

Ligand exchange

HS

-R NH

O

Phospholipids, polysaccharides, polyanhydrides)

TOPO coated QD, insoluble in water

Mono thiol ligands

C

341

C

(Amphiphilic block polymers

P P P

O

CO

C

O

P P P P

P

P

C O CO C

C

P P P

CdSe

Encapsulation

C

O

O

P P

ZnS

-R NH

OH

OH

Nanochemistry for Molecular Imaging

NH2

Figure 2. Quantum dot (QD) water solubilization strategies. Both encapsulation and ligand exchange schemes are shown. Encapsulation preserves the TOPO coating on QD, the encapsulating agent contains a hydrophobic domain that intercalates with trioctylphosphine oxide (TOPO) and a hydrophilic domain that renders water solubility. Many types of encapsulating molecules have been used including amphiphilic block polymers (shown in the figure), phospholipids, polysaccharides, and polyanhydrides (see text for references). Ligand exchange replaces the TOPO layer with a bifunctional ligand, which contains functional groups that interact with the QD surface on one end (eg, –SH) and hydrophilic functional groups on the other end; depicted in the figure are some commonly used ligands (see text for references).

renders w ater compatibility . An ar ray of mono and multidentate thiol and phosphine ligands has been tested and will be discussed in detail in the ne xt section. The second method preser ves the nati ve TOP/TOPO on the QDs and uses v ariants of amphiphilic dib lock and triblock copol ymers and phospholipids to tightl y interleave/interdigitate the alk ylphosphine ligands through hydrophobic interaction, w hereas the h ydrophilic outer block permits aqueous dispersion and further derivatization. In addition, combinations of layers of different molecules conferring the required colloidal stability to QDs have also been exploited.30,31 In the ligand e xchange scheme, TOPO-coated QDs are mix ed with a solution containing a bifunctional ligand, w hich competes with TOPO for binding to a

metal atom on the QD surf ace. With excess bifunctional ligands (e g, mercaptoacetic acid) in the solution, the thiol functional g roups outcompete the phosphonic oxides (from the TOPO) for binding to the QDs, and the QDs become hydrophilic. Various types of ligands ha ve been used, including simple thiol-containing molecules such as mercaptoacetic acid,32 cysteine,33 dithiothreitol,34 dihydrolipoic acid (DHLA), 35 oligomeric phosphine, 25 dendrons,30 and peptides. 36 This method results in smaller QDs, generall y not too much bigger than the organic QDs. Drawbacks of the ligand exchange method are relati vely lo w quantum yields, pH sensiti vity and poor stability in biolo gical buf fers, or insuf ficient robustness for fur ther chemical modif ications to introduce biorecognition molecules.

342

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

In the second method , coordinating ligands on the QD surf ace are retained and used to interact with amphiphilic b lock copol ymers20,37 in silica shells, 38 phospholipids micelles,39 amphiphilic polysaccharides,40 or polyanhydrides.41 The hydrophobic domains of these chemicals strongl y interact with TOP or TOPO on the QD surf ace, w hereas the h ydrophilic g roups f ace outwards and render the w ater-soluble QDs. Note that the coordinating or ganic ligands (T OP or TOPO) are retained on the inner surf ace of QDs, a feature that is important for maintaining the optical proper ties of QDs and for shielding the core from the outside environment. This approach is more ef fective than ligand exchange in maintaining the QD optical proper ties (higher quantum yield) and storage stability in aqueous buf fer, and the drawback is the increase in the o verall size of the QDs. For example, phospholipids and b lock copolymer coatings tend to increase the diameter of CdSe/ZnS QDs from ~4 to 8 nm before encapsulation to ~20 to 30 nm, a size that although smaller than most mammalian cells can still limit intracellular mobility and ma y preclude distance-sensitive applications such as fluorescence resonance ener gy transfer .42 In animal imaging applications, an increase in size ma y present a hurdle b y impeding the extravasations of QDs to the tar geted sites and b y causing rapid uptak e b y the reticuloendothelial system (RES) and thus decreasing the bioa vailability of the nanoparticles.

BIOFUNCTIONALIZATION To mak e QDs more useful for molecular imaging and other biological applications, QDs need to be conjugated to biological molecules without perturbing the biological function of these molecules. Se veral successful approaches have been used to link biolo gical molecules to QDs, including nonspecif ic adsor ption, electrostatic interactions, mercapto (–SH) e xchange, and co valent linkage (F igure 3). 43 It has been repor ted that simple small molecules, such as oligonucleotides44,45 and various serum albumins, 46 are readily adsorbed to the surf ace of water-soluble QDs. This adsor ption is nonspecif ic and depends on ionic strength, pH, temperature, and the surface char ge of the molecule. Mattoussi and colleagues presented a method of conjugating proteins to QD surfaces using electrostatic interactions. The protein of interest w as engineered to contain a positi vely char ged domain (pol yhistidine), w hich in tur n electrostaticall y interacted with the negatively charged surface of DHLAcapped QD. The protein-QD conjugates prepared in this way were stable, and the fluorescence quantum yield was

35 even higher than that from the nonconjugated QDs. Biological molecules containing thiol g roups can be conjugated to the QD surf ace through a mercapto exchange process. 47–51 Unfortunately, the same prob lem of using thiol as an anchoring g roup on a ZnS surf ace occurs because the bond between Zn and thiol is not very strong and is dynamic. As a result, biomolecules can readily dissociate from the nanopar ticle surface, causing QDs to precipitate from the solution. A more stable linkage is obtained by covalently linking biomolecules to the functional groups on the QD surf ace using cross-linking molecules.6,32,34,38,52 Most w ater solubilization methods result in QDs co vered with carbo xylic acid , amino, or thiol g roups. Under these situations, it is easy to link QDs to biolo gical molecules that also ha ve these reaction groups. F or e xample, the cross-link er 1 -ethyl3-(3-dimethylaminopropyl) carbodiimide (EDC) is commonly used to link –NH 2 and –COOH g roups, whereas 4-(N-maleimidometh yl)-cyclohexane carbo xylic acid N-hydroxysuccinimide ester (SMCC) can be used to cross-link –SH and –NH 2 groups. Using the abo ve methods, there ha ve been numerous repor ts of conjugating QDs with various biological molecules, including biotin,38 oligonucleotides,53 peptides36 and proteins, including avidin/streptavidin,37 albumin,54 adapter proteins (eg, protein A, protein G), and antibodies. 37,52 In addition, the native functional g roups (–COOH, –NH 2, or –SH) on a water-soluble QD surf ace can be fur ther con verted to other functional g roups to allo w more v ersatile conjugation of QDs to biomolecules (site-specif ic conjugation and molecules that are sensiti ve to EDC or SMCC modification). For instance, carboxylic acids on QDs have been converted to hydrazides, allowing attachments of biomolecules containing sugar g roups.52 Recently, Zhang and colleagues55 presented another strate gy for site-specif ic conjugation of biomolecules to QD surf ace. This study used the specific and stable binding between HaloTag protein (HTP) and its ligand. Quantum dots w ere f irst functionalized with HaloT ag ligands, and the protein of interest (eg, Renilla luciferase) was genetically fused to a HTP. When mixed together, QDs and Rluc8 can be immobilized on QDs through the HTP-HaloTag ligand linkage. In most cases, the biological functions of these molecules have been preserved during the conjugation process.

ANIMAL IMAGING APPLICATIONS OF QDS Fluorescent proteins and small or ganic dyes have been used as a fluorescent contrast agent for li ving animal imaging. Ho wever, compared with other imaging modalities, such as PET and MRI, fluorescence imaging

Nanochemistry for Molecular Imaging

343

biotin Peptide recognition sequence

Direct linkage

TOPO QD (insoluble)

Peptide (with adhesive domain), thiolated-DNA

DOTA quencher

PEG

Electrostatic interaction Positively charged proteins

2

C O

C

OH

2

O

Biomolecule with free NH2, such as proteins

C OH O

O C O

NH

NH 2

streptavidin

C

C

O

O

O C NH 2

2

C NH

O

C

O

H O

O

O

ec Biotin ul es

2

m ol

C NH

er

H

EDC

Li nk

2

NH C

H O

C

C

O

OH

Polymer coated QD with – COOH or – NH2 groups (showing both)

O

C

C NH2

O

O

O

C NH

Covalent binding

OH

C NH 2

C

S

O

O

O

O

C NH 2

C NH

NH 2

C

O

O

l, hio t e fre ibody h wit ant e l d cu uce e l d mo s re o i B ha C c C su M

NH

NH 2

Negatively charged QD (water-soluble)

C

C

H

O

O

O

C NH 2

H O

Halo ligand

Adapter protein, protein A/G

Followed by specific binding between linker and desired biomolecules (eg, streptavidin & biotinylated oligonucleotides; halo ligand & Halo Tag protein; adaptor protein & antibodies)

Figure 3. Common methods used for quantum dot (QD) biofunctionalization including direct linkage to the TOPO-coated QDs (eg, thiolated DNA via ligand exchange, and peptides with adhesive domains), electrostatic interactions, and covalent linking (see text for references).

is still limited b y the poor transmission of visib le light through biolo gical tissue. One w ay to get around this penetration prob lem is to use near -infrared (NIR, 650 to 900 nm) light because hemo globin and w ater, the major absorbers of visib le and infrared light, respectively, ha ve their lo west absor ption coef ficient in the NIR region.56 Few organic dyes, however, emit brightly in this spectral region, and they further suffer from the photobleaching problem. On the contrar y, QDs can be synthesized as bright and stab le fluorescent labels that

emit in the NIR spectr um b y v arying their size and composition. Because visib le QDs are more synthetically advanced, they are used in most animal imaging studies, but a few recent studies have started using NIR QDs.57,58 Although still far from its mature stage, these studies have shown the great performance and promise of NIR QDs as fluorescent imaging agents in li ving animals. In the follo wing section, w e will summarize the major successful applications of QDs in nontargeted and targeted animal imaging.

344

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Non-Targeted Animal Imaging Quantum dots have been used for nontargeted imaging in various animal models. Ballou and colleagues 59 injected PEG-coated QDs into the mouse blood stream and investigated how the surface coating would affect their circulation lifetime. In contrast to small or ganic dyes, which are eliminated from the circulation within minutes after injection, PEG-coated QDs w ere found to sta y in the blood circulation for an e xtended period of time (halflife more than 3 hours). This long-circulating feature is due to the relati vely lar ge size of PEG-coated QDs, which f alls within an inter mediate size range; the y are small enough and suf ficiently hydrophilic to slow down opsonization and reticuloendothelial uptak e, but lar ge enough to a void renal clearance. Amazingly, these QDs maintained their fluorescence e ven after 4 months in vivo. In 2003, Larson and colleagues 7 intravenously injected green QDs (550 nm) in a living mouse and visualized them dynamically through the skin (in capillaries hundreds of micrometers deep) b y using tw o-photon microscopy (F igure 4A). In addition to the superior brightness and photostability, this study also found that QDs have two-photon e xcitation cross sections as high as 47,000 Goeppert–Mayer units, by far the largest of any label used in multiphoton microscop y. Two-photon excitation allows g reater tissue penetration due to e xcitation in the NIR spectral range (eg, 900 nm), but few fluorophores are bright enough for these pur poses; QDs, with their large multiphoton e xcitation cross section, appear to be ideal probes for multiphoton microscopy imaging. Improved tissue penetration can also be achieved by tuning the QD emission to the NIR windo w. Kim and colleagues58 prepared a no vel core-shell nanostr ucture called type II QDs with f airly broad emission at 850 nm and a moderate quantum yield of ~13%. In contrast to type I QDs, the shell materials in type II QDs ha ve valence and conduction band ener gies both lo wer than those of the core materials. As a result, the electrons and holes are ph ysically separated , and the nanopar ticles emit light at reduced ener gies (longer w avelengths). Near-infrared QDs were injected intradermally in the left paw of a li ving mouse. The results showed rapid uptake (5 min) of QDs into nearb y lymph nodes and that the y could be imaged vir tually background-free (Figure 4B). The same study also sho wed the feasibility of imageguided surgery in a big animal model. 400 pmole of NIR QDs injected intrader mally per mits sentinel l ymph nodes 1 cm deep to be imaged easil y in real time (and removed sur gically) using e xcitation fluorescence rates of only 5 mW/cm 2. Quantum dots ha ve also been used

for cell-tracking studies. 20,39,60,61 Quantum dots w ere delivered into li ve mammalian cells via three dif ferent mechanisms: non-specif ic pinoc ytosis, microinjection, and peptide-induced transpor t (e g, using the protein transduction domain of HIV -1 Tat peptide, Tat-PTD).62 A surprising f inding was that two billion QDs could be delivered into the nucleus of a single cell, without compromising its viability, proliferation, or mig ration.39,61,62 The ability to image single-cell migration and differentiation in real time is belie ved to be impor tant to se veral research areas such as embryogenesis, cancer metastasis, stem-cell therapeutics, and l ymphocyte immunolo gy. These studies showed the potential of using QDs to track cell (one pol ymer coated QD is about the size of one thousandth of a mammalian cell), tissue, and or gan development over e xtended periods of time, a task not possible with small molecule or ganic dy es that are quickly photobleached.

Targeted Animal Imaging The above studies showed the capability of QDs for living animal imaging but had only shown image contrast at the tissue/organ level. The goal of molecular imaging is to generate an image contrast due to the molecular difference in dif ferent tissue and or gans. This requires a probe that has a targeting moiety to generate contrast only in locations specif ied b y the tar geting probe. Akerman and colleagues 51 were among the f irst to explore the possibility of using QD-peptide conjugates to tar get tumor v asculatures in vi vo. Quantum dots coated with peptides tar geting the lung v asculature, blood v essel, tumor cell, or l ymphatic v essels w ere injected systemically into mice. Although they were not able to image the QDs in a li ving animal, histolo gic sections of different organs after 5 or 20 min of circulation showed that QDs homed to tumor vessels guided by the peptides, but not to sur rounding tissues, probably due to their lar ger size relati ve to or ganic dy es, which would stain sur rounding tissues. Whole animal imaging with molecular-level detection was realized by Gao and colleagues20 in 2004 using red fluorescent QDs conjugated to antibodies specif ic to prostate-specif ic membrane antigen (PSMA) on a human prostate cancer induced in a mouse. Quantum dot conjugates used in this study contained an amphiphilic triblock copolymer for in vi vo protection, tar geting ligands (anti-PSMA) for tumor antigen recognition, and multiple PEG molecules for impro ved biocompatibility and circulation. The QD-tagged PSMA antibodies reco gnized and

Nanochemistry for Molecular Imaging

A

345

B

C D

Figure 4. Animal imaging applications using quantum dots (QDs). A, Two-photon fluorescence imaging of capillaries at the base of the dermis with QD (1 µM) or fluorescein (40 µM) intravenously delivered to the animal. Adapted from Larson et al.7 B, NIR QD fluorescence guided dissection of a lymph node in a pig with 400 pmol of NIR QDs injected intradermally in the right groin. Reproduced with permission from Kim S et al.58 C, Targeted tumor imaging using QD antibody conjugates. Reproduced with permission from Gao et al.20 D, Tumor imaging using QD705 RGD bioconjugates; arrows indicate tumor sites. Reproduced with permission from Cai W et al.57

bound at the tumor site and were clearly imaged in vivo (Figure 4C). There are two possible mechanisms for the preferential accumulation of QDs at tumor sites: passive targeting due to the enhanced per meability and retention (EPR) ef fect, and acti ve targeting because of the antibody against PSMA, w hich is a cell surf ace marker for both prostate epithelial cells and neo vascular endothelial cells. Because the QDs emit in the visible range, a spectral unmixing algorithm w as used to separate QD signal from background autofluorescence. More recentl y, Cai and colleagues 57 used NIR QDs for tumor imaging b y tar geting angio genesis, the formation of ne w blood vessels from pre-e xisting vasculature. Amine-modified QD705 (emission maximum at 705 nm) w as conjugated to αvβ3 integrin-targeting cyclic ar ginine–glycine–aspartic acid (RGD) peptide and injected intravenously into living mice. Tumor fluorescence reached maximum at 6 h postinjection with good contrast (F igure 4D). Because angio genesis is common to all tumors, this technique ma y aid cancer detection and management in general. It is w orth noting that in both studies, a signif icant portion of the injected QDs mo ved to the RES, including the li ver, spleen, and lung.

RESONANCE ENERGY TRANSFER-BASED DETECTION AND IMAGING USING QDS QD Fluorescence Resonance Energy Transfer Fluorescence resonance ener gy transfer (FRET) is a process in w hich ener gy is transfer red from an e xcited donor to an acceptor via a resonant, near -field dipole–dipole interaction. 63 FRET is v ery sensitive to the distance between donor and acceptor and has been used to study biomolecule confor mation, dynamics, and interactions. Prob lems associated with FRET betw een or ganic fluorophores include f ast photob leaching and spectral overlap between the acceptor and donor , but the y can be solved with QDs as the FRET donor .42,47,64,65 Willard and colleagues65 reported using QDs as a FRET donor in a protein–protein binding assa y. Quantum dots w ere conjugated to bo vine ser um albumin (BSA) as the FRET donors, and tetrameth ylrhodamine (TMR) w as linked to the protein as the acceptor . Enhanced TMR fluorescence was observed as energy transferred from the QDs to TMR. Medintz and colleagues 66 engineered a histidine tag on maltose-binding protein (MBP) that bound

346

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

electrostatically to the QDs, w hich ser ved as the FRET donor (Figure 5A). With a quencher bound to the maltosebinding site, QD fluorescence w as inhibited but could be recovered by adding maltose to displace the quencher . In 2004, the same group did an in-depth study using the MBP system by varying the QD size and the quantity of acceptor dye.64 Increasing the number of the acceptor dye molecules or the de gree of spectral o verlap (by changing QD size) led to a substantial enhancement in the ener gy transfer efficiency. This study showed that QDs can be used as an efficient energy donor in the FRET system and that b y tuning their size, QDs can transfer ener gy to a number of different or ganic dy e molecules. F rom these studies, the unique advantages of QDs as FRET donors are apparent: size-tunable emission spectra can be used to improve spectral overlap with a par ticular acceptor dy e, and the multivalency ability allows multiple acceptor dye molecules on a single QD donor , w hich will substantiall y improve the FRET efficiency. Quantum dots could e ven drive biosensors through a tw o-step FRET mechanism o vercoming inherent donor–acceptor distance limitations, as sho wn in another work of the same researchers. 66 In this assemb ly, each 530-nm QD is surrounded by ~10 MBPs (each monolabeled with Cy3). A second dy e, β-cyclodextrin-Cy3.5 (β-CD-Cy3.5), specif ically binds in the MBP central binding pocket. Excitation of the QD leads to FRET excitation of the MBP-Cy3, which in turn FRET-excites the β-CD-Cy3.5.

Recently, QD–FRET has been successfull y applied on the proteolytic activity detection b y Medintz and colleagues. 42 In this scenario, a rationall y designed multifunctional peptide sequence was engineered between the donor (QD) and acceptor (or ganic dy e). The peptide sequence consists of four functional domains including (1) pol yhistidine sequence for self-assemb ly onto the DHLA capped QDs; (2) a helix-link er spacer to pro vide rigidity; (3) an exposed protease recognition/cleavage sequence; and (4) a C-terminal site-specif ic location (c ysteine thiol) for dy e attachment. When the donor–acceptor h ybrid is inte grated, FRET takes place, and the emission spectr um is a composite of the QD and acceptor emissions. Upon the addition of the acti ve protease (caspase-1, collagenase, ch ymotrypsin, or thrombin), the acceptor dye is released from the donor (QD), and the observed fluorescent emission is purely QD emission. Because all the results so f ar are from e xperiments performed in buffers, and not in biological samples such as b lood or ser um, it remains to be deter mined whether these QD-FRET nanoassemb lies can be applied for in vivo imaging of proteolytic activity. Paradoxically, the excellent QD donor properties (long fluorescence lifetime, broad absor ption, and high e xtinction coefficient) may almost preclude their role as a FRET acceptor for or ganic dyes.64 However, this does not limit their use as acceptors in other applications such as bioluminescence resonance energy transfer as discussed below.

A

B

Figure 5. Resonance energy transfer based detection and imaging using QDs. A, QD FRET. Formation of QD560-MBP-β-CD-QSY9 nanoassembly results in quenching of the QD emission. Added maltose displaces β-CD-QSY9 from the nanoassembly, resulting in an increase in direct QD emission. Reproduced with permission from Medintz IL et al.66 B, QD BRET. QD655 is covalently linked to a BRET donor, Luc8. The bioluminescence energy of the Luc8-catalyzed oxidation of coelenterazine is transferred to QDs in close proximity, resulting in the QD emission (655 nm). Reproduced with permission from So MK et al.70

Nanochemistry for Molecular Imaging

QD Bioluminescence Resonance Energy Transfer-Based Animal Imaging Bioluminescence resonance ener gy transfer (BRET) is a naturally occurring phenomenon whereby a light-emitting protein (the donor, eg, R. reniformis luciferase) nonradiatively transfers energy to a fluorescent protein (the acceptor, e g g reen fluorescent protein [GFP]) in a close proximity. BRET is analo gous to FRET e xcept that the energy comes from a biochemical reaction catal yzed b y the donor enzyme (e g, R. reniformis luciferase-mediated oxidation of its substrate coelenterazine) rather than the absorption of e xcitation photons. Compared to fluorescence, bioluminescence has extremely high sensitivity for in vivo imaging pur poses.67 BRET has been successfull y used to study protein–protein interaction in living cells and animals.68,69 In a study b y De and Gambhir , a hRluc8 (donor)–GFP2 (acceptor) BRET pair w as used to study FKBP12 (fused with hRluc8) and FRB (fused with GFP 2) interaction inside li ving mice. 68 BRET signal could be detected by using a cooled charged-coupled device (CCD) camera. Recent work in the Rao laborator y has shown the feasibility of using QDs as the acceptor in a BRET system.70 In this study, QDs were covalently conjugated by EDC coupling to the donor , Luc8 protein (sometimes called Rluc8 in literature), an eight-mutation variant of the bioluminescent R. reniformis luciferase developed in the Gambhir lab.71 The protein emits blue light with a peak at 480 nm upon the addition of the substrate, coelenterazine. If the QDs are in a close pro ximity of the protein, they can be e xcited and emit at its emission maximum (Figure 5B). The advantage of using bioluminescence versus fluorescence lies in the f act that no external excitation is needed. This self-illuminating feature allo ws cancer imaging in deeper tissue where light resources are limited. Because no excitation light is needed, the autofluorescence problem is automaticall y solv ed. In addition, other QDs (including those emitting in the NIR re gion, eg, 705 and 800 nm QDs ha ve been tested) can be used as a BRET acceptor, w hich allo ws for multiple xed imaging. Compared with existing QDs, self-illuminating QD conjugates have greatly enhanced sensitivity in small animal imaging, with an in vi vo signal-to-background ratio of > 1000 for 5 pmole of conjugates subcutaneously injected. One critical issue in making these self-illuminating QDs is the size of the nanopar ticle because lik e FRET, BRET is also a distance-dependent process. An increase in the size of QD conjugate results in a greater distance between the protein and the fluorescent semiconductor core and thereb y decreases the ener gy transfer ef ficiency signif icantly. For instance, the authors ha ve obser ved that increasing the

347

protein-nanoparticle distance by only 2 to 3 nm causes the 70 By attaching BRET ratio to drop from 1.29 to 0.37. targeting moieties such as tumor homing antibodies or peptides to the BRET assembly, it is possible to use BRET for targeted-tumor imaging in living animals. More recently, QD-BRET was successfully applied to proteolytic activity detection in buffer with a slightly different coupling scheme. 72 In this study , a 15 amino-acid peptide (GGPLGVRGGHHHHHH), containing the matrix metalloproteinase 2 (MMP-2) substrate and a six-histidine tag, was genetically fused to the C ter minus of the BRET donor, Luc8. In the presence of Nick el2+ cation, the carboxylic acids on the QDs will bind the metal ions and form complexes with the 6x His tag on the Luc8 fusion protein. BRET will take place and produce light emission from the QDs. The cleavage of the amide bond between Gly and Val by MMP-2 will release the 6x His tag from the fusion Luc8, and thus no BRET will occur . In the presence of active MMP-2, the protein w as released from the conjugates and BRET did not happen. A similar concept can be applied for in vi vo protease detection and imaging with a different coupling strategy, which ensures site-specific conjugation (ie, protease substrate positioned between QD and Luc8) and meanw hile mak es the conjugates stab le and resistant to interference from other proteins present inside the living body.

ISSUES AND PERSPECTIVES Although QDs provide a class of exciting new fluorescent probes that open up man y opportunities for fluorescence imaging, several issues remain and need to be resolv ed before QDs can be widel y used for tar geted imaging of tumor or other diseases in human. Some of these issues are applicab le to all nanostr uctures intended for molecular imaging purposes.

Less RES Uptake and Longer Circulation Time The f irst issue is their relati vely shor t circulation halflife, preventing long-term imaging or cell-tracking studies, and this applies to all nanostr uctures injected systemically into a li ving body. Literature on the in vi vo studies using QDs for imaging ha ve shown that their circulation half-time is influenced strongl y b y the surf ace chemistry and size and that the y are cleared from the circulation primarily by phagocytosis of the nanopar ticle by the RES in the li ver, spleen, and l ymph nodes. 20,59 Coating with PEG increases the circulation half-life, and

348

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

attachment to tar geting moieties (such as antibodies) reduces the dose needed to generate contrast betw een normal and tumor tissue. 20 Even with these strate gies incorporated, the majority of nanoparticles still end up in the RES. A recent study using commercial nontar geted amino (PEG) QD705 for imaging glioma in rat suggested that the maximal RES phagoc ytosis of QDs was reached between 3.4 and 8.5 nmol doses. 73 Significant tumor uptake of QDs w as obser ved for doses higher than 8.5 nmol. Such a high dose might pose a health hazard to living subjects. Therefore, there is an ur gent need for engineering strate gies to impro ve the circulation halftime of QDs because the longer the y can stay in the circulation system the better are the chances that the y can get to the tumor site. This might be achieved by minimizing the opsonization or other components of the RES. Intending to solve this problem, a few groups have started studying the interaction betw een blood components and nanoparticles.74,75 The rationale is that once entered into the blood stream, nanopar ticles are immediately covered by plasma proteins; w hat the RES system “sees” and what defines the identity of the nanoparticle is largely the protein corona around the par ticle, not the core material. Elucidation of the prof ile of proteins adsorbed on nanoparticles has the potential to f acilitate engineering a surface chemistr y that is less prone to opsonization and the RES uptak e. The alter native approach is to mak e smaller QDs that completel y e vade the RES. A recent study by the F rangioni laboratory33 has shown that QDs with a hydrodynamic size of 5.5 nm or smaller can evade the RES or gans (no accumulation in the li ver, spleen, or lung) and be cleared b y the renal system. Quantum dots used in this study w ere cysteine coated 76 with no tar geting or other biofunctional g roups, therefore it remains a challenge ho w to mak e biofunctionalized QDs w hile keeping the size belo w 5.5 nm. Another issue with nanoparticle-based molecular imaging is in vivo delivery and efficient extravasation to tumor sites. Using intravital microscopy, Smith and colleagues 77 discovered that QDs with tumor tar geting RGD peptide w ere not ab le to extravasate in an SKOV-3 mouse ear tumor model, as specific binding onl y occur red in the tumor neo vasculature with aggregated QD conjugates.

Deeper Tissue Penetration Although the superior brightness and photostability of QDs made them attracti ve candidates for in vi vo animal imaging, most of the cur rent QDs still emit within the visible range. The ideal QDs for deep tissue imaging, that is, high-quality QDs with NIR-emitting proper ties, are

not yet available. Recent developments include a promising water-based synthesis method that yields intrinsically water-soluble par ticles that emit from the visib le to the NIR spectr um, but the par ticles have yet to be tested in biological environments. Most materials (e g, PdS, PdSe, cadmium mercur y telluride (CdHgT e), and cadmium selenium telluride (CdSeTe)) are either not bright enough or stable enough for biomedical imaging applications. As such, there is an urgent need to develop bright and stable NIR-emitting QDs that are broadl y tunable in the f ar-red and infrared spectral re gions.20 Theoretical modeling studies by Lim and colleagues78 indicate that two spectral windows are e xcellent for in vi vo QD imaging, one at 700 to 900 nm and another at 1200 to 1600 nm. In addition to high-quality NIR QDs, multiphoton fluorescence microscopy and novel illuminating mechanisms, such as bioluminescence resonance ener gy transfer , can all be used to achieve deeper tissue penetration.

Toxicity One of the major issues that hinder the application of QDs to human subjects is the concer n about their safety: cadmium and selenium are potential hazards causing neurological and genitourinar y to xicity. Indeed, in vi vo toxicity is likely to be a key factor in determining whether QD imaging probes w ould be appro ved b y re gulatory agencies for human clinical use. Cell culture studies 79 indicate that CdSe QDs are highl y toxic to cultured cells under UV illumination for extended periods of time. This result is not surprising because the energy of UV irradiation is close to that of a co valent chemical bond and dissolves the semiconductor particles in a process known as photolysis, releasing toxic cadmium ions into the culture medium. In the absence of UV ir radiation, QDs with a stable polymer coating are not lik ely to be to xic to cells and animals. In vi vo studies b y Ballou and colleagues 59 also conf irmed the nonto xic nature of stab ly protected QDs. Still, there is an urgent need to systematically study the long-ter m to xicity and in vi vo de gradation mechanisms of QD probes. The polymer-protected QDs might be cleared from the body by slow filtration and excretion out of the body. This and other possible mechanisms must be carefully examined before an y human applications in tumor or v ascular molecular imaging tak e place. While modification of QD surfaces can help QDs clear from the body within a reasonable time, an alternative is to replace the toxic elements of QDs but retain similar optical properties (eg, high quantum yield , stability). 43 Indeed, there are some recent de velopments on noncadmium based QDs80 and doped dots. 81 These ne w types of QDs ma y

Nanochemistry for Molecular Imaging

hold promises for overcoming the toxicity issue and lead to the e ventual use for in vi vo molecular imaging in clinics.

Perspectives Quantum dots as a novel fluorescent probe have proved to be tremendousl y useful in man y areas of biolo gical and medical research, especially multiplexed tissue/cell labeling, live-cell imaging as well as in vivo imaging. However, as an in vi vo imaging agent, QDs are still at a v ery premature stage. Several issues (including toxicity, size issue, and clearance by RES) remain before its potential can be fully e xploited in this arena and applied to human subjects. Although the perfor mances of visib le QDs are greatly improved over conventional fluorophores, an ideal QD fluorophore should emit in the NIR/f ar-red re gion with high quantum yield and e xcellent stability. Meanwhile, techniques such as tw o-photon excitation7 can be used to f acilitate the use of visib le dots for better tissue penetration. In summar y, QDs technolo gy for in vi vo cancer imaging is still an area of acti ve research and will require the ongoing collaboration of chemists, biolo gists, and material scientists.

OTHER TYPES OF NANOSTRUCTURES FOR IN VIVO MOLECULAR IMAGING In addition to QDs, man y other types of nanopar ticles have also been de veloped for molecular imaging purposes, including magnetic iron o xide nanopar ticles4,82 (for MRI), carbon nanotubes, 83–85 and surf ace-enhanced Raman probes for tumor tar geting and imaging. 86,87 In general, the same principles in volved with QDs and described in this chapter ma y be similarl y applied to those nanoparticles. We choose iron o xide nanoparticles and carbon nanotubes as tw o additional e xamples and briefly discuss their synthesis and molecular imaging applications.

Magnetic Iron Oxide Nanoparticles (MNP) Magnetic iron o xide nanopar ticles constitute another large class of molecular MRI agents due to the ability of superparamagnetic particles in modulating the uniformity of a magnetic f ield.82 Typically, magnetic nanopar ticle probes for biomedical applications are comprised of nanoscale super paramagnetic iron o xide super paramagnetic iron o xide (SPIO) cores (3 to 5 nm) of magnetite

349

and/or maghemite encased in de xtran, a synthetic pol ymer, or monomer coatings.88 The first generation of MNP consists of a thin layer of dextran (eg, Feridex) and has the propensity to agg regate. Subsequent de velopment in surface coating with e xtensive pol ymers results in monocrystalline iron oxide nanoparticles (MION). MION used in in vi vo molecular imaging are generall y smaller than 50 nm in diameter . A more recent de velopment for targeted molecular imaging uses a highly stabilized, crosslinked deri vative of MION [cross-link ed iron o xide (CLIO)].89 CLIO has amino functional g roups, allowing functionalization b y attaching fluorophores for dualmodality imaging (MRI fluorescence) or targeting ligands such as annexinV for apoptosis imaging. The nontargeting MNP have been used e xtensively in the clinical arena for imaging the li ver and l ymphatic systems due to the l ymphotropic feature of MNP .4 Targeted MNP such as the CLIOs have been used for cardiovascular molecular imaging as well as tumor imaging. 88,89

Carbon Nanotube (CNT) CNTs are carbon-based tubular structures, which are only a few nanometers in diameter and microns or e ven up to millimeters in length. Carbon nanotubes possess e xtraordinary proper ties, including high electrical and ther mal conductivity, g reat strength, and rigidity , and are being developed for a w ealth of applications, including f ield emission, ener gy storage, molecular electronics, and atomic force microscop y.85 Bare CNTs are insolub le in water and can become water-soluble after surface coating with phospholipids 83 or peptides. 85 Although still at the exploratory stage, CNTs have shown promising results in molecular imaging. Singh and colleagues 85 functionalized single w all CNT (SWNT) with the chelating molecule dieth ylenetriaminepentaacetic (DTPA) and labeled them with indium ( 111In) for imaging pur poses. Tissue biodistribution results sho wed that SWNTs w ere not retained in the li ver or spleen but w ere rapidl y cleared from systemic blood circulation through the renal e xcretion route. Targeted tumor imaging with SWNT w as achieved by Liu and colleagues 83 in 2007. In this study , SWNTs were functionalized with phospholipids bearing PEG chains. The ter minus of the PEG chain w as then linked to a radioactive tracer (64Cu) or targeting molecule (RGD peptide). Ef ficient tar geting of inte grin positi ve tumor was confirmed by in vivo PET, ex vivo biodistribution, and Raman spectroscop y, and w as attributed to the multivalency ef fect of SWNT . Another e xample of SWNT for in vivo molecular imaging comes from a more recent study b y Liu and colleagues. 84 Results from this

350

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

study showed that SWNT with branched PEG chains had no toxic side ef fects on the animal and can be e xcreted and cleared via the biliar y and renal pathways.

SUMMARY The rapid progress of the research in nanomedicine is critically dependent on the successful preparation of nanomaterials with the desired properties for intended biomedical and biological applications and requires the seamless collaboration of chemists, biologists, and material scientists. In this chapter , using semiconductor nanocr ystals, QDs, as an example, we have reviewed the chemistry of the QD synthesis, water solubilization, biofunctionalization, their applications in molecular imaging and also discussed the limitations, issues, and perspecti ves for nanostr ucturebased molecular imaging. With the limited space of the chapter and the enor mously large body of e xisting literature on the subject, this coverage is not meant to and cannot be e xtensive; it just ser ves as a general guide to the basic aspects of chemistry involved with the synthesis and functionalization of nanopar ticles for molecular imaging applications. Although much of our discussion focuses on the QDs, the general principles may be applicable to other nanoparticles. Some of the other types of nanopar ticles and their applications for molecular imaging ha ve been separately discussed in other chapters, e g Chapter 35, “Fluorocarbon Agents for Quantitative Multimodal Molecular Imaging and Targeted Therapeutics” on nanopar ticles and Chapter 36, “ Aptamers for Molecular Imaging” on perfluorocarbon nanopar ticles. Interested readers are advised to consult with relevant chapters and literature for more details.

REFERENCES 1. Nie S, Xing Y, Kim GJ, Simons JW. Nanotechnology applications in cancer. Annu Rev Biomed Eng 2007;9:257–88. 2. Alivisatos P. The use of nanocr ystals in biolo gical detection. Nat Biotechnol 2004;22:47–52. 3. Michalet X, Pinaud FF , Bentolila LA, et al. Quantum dots for li ve cells, in vi vo imaging, and diagnostics. Science 2005;307: 538–44. 4. Harisinghani MG, Barentsz J , Hahn PF, et al. Nonin vasive detection of clinicall y occult l ymph-node metastases in prostate cancer . N Engl J Med 2003;348:2491–9. 5. Hood JD, Bednarski M, Frausto R, et al. Tumor regression by targeted gene delivery to the neovasculature. Science 2002;296:2404–7. 6. Chan WCW, Maxwell DJ, Gao XH, et al. Luminescent quantum dots for multiple xed biolo gical detection and imaging. Cur r Opin Biotechnol 2002;13:40–6. 7. Larson DR, Zipfel WR, Williams RM, et al. Water-soluble quantum dots for multiphoton fluorescence imaging in vi vo. Science 2003;300:1434–6.

8. Alivisatos AP. Perspectives on the physical chemistry of semiconductor nanocrystals. J Phys Chem 1996;100:13226–39. 9. Sutherland AJ. Quantum dots as luminescent probes in biological systems. Curr Opin Solid State Mater Sci 2002;6:365–70. 10. Peng XG. Mechanisms for the shape-control and shape e volution of colloidal semiconductor nanocr ystals. Adv Mater 2003; 15:459–63. 11. Manna L, Milliron DJ, Meisel A, et al. Controlled growth of tetrapodbranched inorganic nanocrystals. Nat Mater 2003;2:382–5. 12. Milliron DJ , Hughes SM, Cui Y, et al. Colloidal nanocr ystal heterostructures with linear and branched topolo gy. Nature 2004;430:190–5. 13. Dick KA, Deppert K, Larsson MW, et al. Synthesis of branched ‘nanotrees’ by controlled seeding of multiple branching e vents. Nat Mater 2004;3:380–4. 14. Yu WW, Wang YA, Peng XG. Formation and stability of size-, shape-, and str ucture-controlled CdT e nanocr ystals: ligand ef fects on monomers and nanocrystals. Chem Mater 2003;15:4300–8. 15. Hines MA, Scholes GD . Colloidal PbS nanocr ystals with sizetunable near -infrared emission: obser vation of post-synthesis self-narrowing of the par ticle size distrib ution. Adv Mater 2003;15:1844–9. 16. Bailey RE, Nie S. Alloyed semiconductor quantum dots: tuning the optical properties without changing the particles. J Am Chem Soc 2003;125:7100–6. 17. Efros AL, Rosen M. The electronics str ucture of semiconductor nanocrystals. Annu Rev Mater Sci 2000;30:475–521. 18. Smith AM, Gao X, Nie S. Quantum dot nanocrystals for in vivo molecular and cellular imaging. Photochem Photobiol 2004;80: 377–85. 19. Leatherdale CA, Woo WK, Mikulec FV, Bawendi MG. On the absorption cross section of CdSe nanocr ystal quantum dots. J Ph ys Chem B 2002;106:7619–22. 20. Gao X, Cui Y, Levenson RM, et al. In vivo cancer targeting and imaging with semiconductor quantum dots. Nat Biotechnol 2004;22: 969–76. 21. Dahan M, Laurence T, Pinaud F, et al. Time-gated biological imaging by use of colloidal quantum dots. Opt Lett 2001;26:825–7. 22. Chien GL, Anselone CG, Davis RF, Van Winkle DM. Fluorescent vs radioactive microsphere measurement of re gional m yocardial blood flow. Cardiovasc Res 1995;30:405–12. 23. Han M, Gao X, Su JZ, Nie S. Quantum-dot-tagged microbeads for multiplexed optical coding of biomolecules. Nat Biotechnol 2001;19:631–5. 24. Murray CB, Norris DJ, Bawendi MG. Synthesis and characterization of nearl y monodisperse CdE (E = S, Se, Te) semiconductor nanocrystals. J Am Chem Soc 1993;115:8706–15. 25. Kim S, F isher B , Eisler HJ , Ba wendi MG. Type-II quantum dots: CdTe/CdSe(core/shell) and CdSe/ZinT e(core/shell) heterostr uctures. J Am Chem Soc 2003;125:11466–7. 26. Pietryga JM, Schaller RD, Werder D, et al. Pushing the band gap envelope: mid-infrared emitting colloidal PbSe quantum dots. J Am Chem Soc 2004;126:11752–3. 27. Zhong XH, F eng YY, Knoll W, Han MY . Alloyed ZnxCd(1-x)S nanocrystals with highl y nar row luminescence spectral width. J Am Chem Soc 2003;125:13559–63. 28. Hines MA, Guy ot-Sionnest P . Synthesis and characterization of strongly luminescing ZnS-Capped CdSe nanocr ystals. J Phys Chem 1996;100:468–71. 29. Reiss P , Bleuse J , Pron A. Highl y luminescent CdSe/ZnSe core/shell nanocrystals of low size dispersion. Nano Lett 2002; 2:781–4. 30. Guo WH, Li JJ , Wang YA, P eng XG. Luminescent CdSe/CdS core/shell nanocrystals in dendron boxes: superior chemical, photochemical and ther mal stability . J Am Chem Soc 2003;125: 3901–9.

Nanochemistry for Molecular Imaging

31. Sukhanova A, De vy J , Venteo L, et al. Biocompatib le fluorescent nanocrystals for immunolabeling of membrane proteins and cells. Anal Biochem 2004;324:60–7. 32. Chan WC, Nie S. Quantum dot bioconjugates for ultrasensiti ve nonisotopic detection. Science 1998;281:2016–8. 33. Choi S, Liu W, Misra P, et al. Renal clearance of quantum dots. Nat Biotechnol 2007;25:1165–70. 34. Pathak S, Choi SK, Arnheim N , Thompson ME. Hydro xylated quantum dots as luminescent probes for in situ h ybridization. J Am Chem Soc 2001;123:4103–4. 35. Mattousi H, Mauro JM, Goldman ER, et al. Self-assemb ly of CdSeZnS quantum dots bioconjugates using an engineered recombinant protein. J Am Chem Soc 2000;122:12142–50. 36. Pinaud F, King D, Moore HP, Weiss S. Bioactivation and cell tar geting of semiconductor CdSe/ZnS nanocrystals with phytochelatinrelated peptides. J Am Chem Soc 2004;126:6115–23. 37. Wu XY, Liu HJ, Liu JQ, et al. Immunofluorescent labeling of cancer marker Her2 and other cellular tar gets with semiconductor quantum dots. Nat Biotechnol 2003;21:41–6. 38. Bruchez MJ, Moronne M, Gin P, et al. Semiconductor nanocrystals as fluorescent biological labels. Science 1998;281:2013–6. 39. Dubertret B, Skourides P, Norris DJ, et al. In vi vo imaging of quantum dots encapsulated in phospholipid micelles. Science 2002; 298:1759–62. 40. Osaki F, Kanamori T, Sando S, et al. A quantum dot conjugated sugar ball and its cellular uptak e: on the size ef fects of endoc ytosis in the subviral region. J Am Chem Soc 2004;126:6520–1. 41. Pellegrino T, Manna L, K udera S, et al. Hydrophobic nanocr ystals coated with an amphiphilic pol ymer shell: a general route to water-soluble nanocrystals. Nano Lett 2004;4:703–7. 42. Medintz IL, Clapp AR, Br unel FM, et al. Proteol ytic activity monitored b y fluorescence resonance ener gy transfer through quantum-dot-peptide conjugates. Nat Mater 2006;5:581–9. 43. Alivisatos P, Gu W, Larabell C. Quantum dots as fluorescent probes. Annu Rev Biomed Eng 2005;7:55–76. 44. Lakowicz JR, Gryczynski I, Gryczynski Z, et al. Time-resolved spectral observations of cadmium-enriched cadmiumsulf ide nanoparticles and the ef fects of DN A oligomer binding. Anal Biochem 2000;280:128–36. 45. Mahtab R, Harden HH, Murphy CJ. Temperature- and salt-dependent binding of long DN A to protein-sized quantum dots: ther modynamics of “inorganic protein”-DNA interactions. J Am Chem Soc 2000;122:14–7. 46. Hanaki K, Momo A, Oku T, et al. Semiconductor quantum dot/albumin comple x is a long-life and highl y photostab le endosome mark er. Biochem Bioph ys Res Commun 2003;302: 496–501. 47. Patolsky F, Gill R, Weizmann Y, et al. Lighting-up the dynamics of telomerization and DNA replication by CdSe-ZnS quantum dots. J Am Chem Soc 2003;125:13918–9. 48. Mitchell GP, Mirkin CA, Letsinger RL. Pro grammed assemb ly of DNA functionalized quantum dots. J Am Chem Soc 1999; 121:8122–3. 49. Rosenthal SJ, Tomlinson I, Adkins EM, et al. Targeting cell surf ace receptors with ligand conjugated nanocr ystals. J Am Chem Soc 2002;124:4586–94. 50. Winter JO, Liu TY, Korgel BA, Schmidt CE. Reco gnition molecule directed interf acing betw een semiconductor quantum dots and nerve cells. Adv Mater 2001;13:1673–7. 51. Akerman ME, Chan WC, Laakkonen P, et al. Nanocr ystal targeting in vivo. Proc Natl Acad Sci U S A 2002;99:12617–21. 52. Xing Y, Chaudr y Q, Shen C, et al. Bioconjugated quantum dots for multiplexed and quantitati ve immunohistochemistr y. Nat Protoc 2007;2:1152–65. 53. Xiao Y, Bark er PE. Semiconductor nanocr ystal probes for human metaphase chromosomes. Nucleic Acids Res 2004;32:e28.

351

54. Gao X, Chan WC, Nie S. Quantumdot nanocr ystals for ultrasensitive biological labeling and multicolor optical encoding. J Biomed Opt 2002;7:532–7. 55. Zhang Y, So MK, Loening AM, et al. HaloT ag protein-mediated sitespecific conjugation of bioluminescent proteins to quantum dots. Angew Chem Int Ed Engl 2006;45:4936–40. 56. Weissleder R. A clearer vision for in vi vo imaging. Nat Biotechnol 2001;19:316–7. 57. Cai W, Shin DW, Chen K, et al. Peptide-labeled near-infrared quantum dots for imaging tumor v asculature in li ving subjects. Nano Lett 2006;6:669–76. 58. Kim S, Lim YT, Soltesz EG, et al. Near -infrared fluorescent type II quantum dots for sentinel l ymph node mapping. Nat Biotechnol 2004;22:93–7. 59. Ballou B , Lagerholm BC, Er nst LA, et al. Nonin vasive imaging of quantumdots inmice. Bioconjug Chem 2004;15:79–86. 60. Hoshino A, Hanaki K, Suzuki K,Yamamoto K. Applications of T-lymphoma labeled with fluorescent quantum dots to cell tracing markers in mouse body . Biochem Bioph ys Res Commun 2004;314:46–53. 61. Voura EB, Jaiswal JK, Mattoussi H, Simon SM. Tracking metastatic tumor cell extravasation with quantum dot nanocr ystals and fluorescence emission-scanning microscop y. Nat Med 2004;10: 993–8. 62. Lewin M, Carlesso N , Tung CH, et al. Tat peptide-derivatized magnetic nanoparticles allow in vivo tracking and recovery of progenitor cells. Nat Biotechnol 2000;18:410–4. 63. Van Der Meer BW, Coker G III, Chen S-Y. Resonance energy transfer: theory and data. New York: John Wiley & Sons; 1994. 64. Clapp AR, Medintz IL, Uy eda HT, et al. Quantum dot-based multiplexed fluorescence resonance ener gy transfer. J Am Chem Soc 2005;127:18212–21. 65. Willard DM, Carillo LL, Jung J , Van Orden A. CdSe-ZnS quantum dots as resonance ener gy transfer donors in a model proteinprotein binding assay. Nano Lett 2001;1:469–74. 66. Medintz IL, Clapp AR, Mattoussi H, et al. Self-assemb led nanoscale biosensors based on quantum dot FRET donors. Nat Mater 2003;2:630–8. 67. Contag CH, Bachmann MH. Advances in in vi vo bioluminescence imaging of gene expression. Annu Rev Biomed Eng 2002;4:235–60. 68. De A, Gambhir SS. Nonin vasive imaging of protein-protein interactions from li ve cells and li ving subjects using bioluminescence resonance energy transfer. FASEB J 2005;19:2017–9. 69. De A, Loening AM, Gambhir SS. An improved bioluminescence resonance energy transfer strategy for imaging intracellular events in single cells and living subjects. Cancer Res 2007;67:7175–83. 70. So MK, Xu C, Loening AM, et al. Self-illuminating quantum dot conjugates for in vi vo imaging. Nat Biotechnol 2006; 24:339–43. 71. Loening AM, F enn TD, Wu AM, Gambhir SS. Consensus guided mutagenesis of Renilla luciferase yields enhanced stability and light output. Protein Eng Des Sel 2006;19:391–400. 72. Yao H, Zhang Y, Xiao F, et al. Quantum dot/bioluminescence resonance ener gy transfer based highl y sensiti ve detection of proteases. Angew Chem Int Ed Engl 2007;46:4346–9. 73. Jackson H, Muhammad O , Daneshv ar H, et al. Quantum dots are phagocytized by macrophages and colocalize with e xperimental gliomas. Neurosurgery 2007;50:524–9. 74. Cedervall T, Lynch I, Lindman S, et al. Understanding the nanopar ticle-protein corona using methods to quantify e xchange rates and affinities of proteins for nanopar ticles. Proc Natl Acad Sci U S A 2007;104:2050–5. 75. Kim HR, Andrieux K, Delomenie C, et al. Analysis of plasma protein adsorption onto PEGylated nanopar ticles b y complementar y methods: 2-DE, CE and Protein Lab-on-chip system. Electrophoresis 2007;28:2252–61.

352

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

76. Liu W, Choi HS, Zimmer JP , et al. Compact c ysteine-coated CdSe (ZnCdS) quantum dots for in vi vo applications. J Am Chem Soc 2007;129:14530–1. 77. Smith BR, Cheng Z, De A, et al. Real-time intra vital imaging of RGD-quantum dot binding to luminal endothelium in mouse tumor neovasculature. Nano Lett 2008;8:2599–606. 78. Lim YT, Kim S, Naka yama A, et al. Selection of quantum dot w avelengths for biomedical assa ys and imaging. Mol Imaging 2003; 2:50–64. 79. Derfus AM, Chan WCW, Bhatia SN. Probing the cytotoxicity of semiconductor quantum dots. Nano Lett 2004;4:11–18. 80. Zimmer JP, Kim SW , Ohnishi S, et al. Size series of small indium arsenide-zinc selenide core-shell nanocr ystals and their application to in vivo imaging. J Am Chem Soc 2006;128:2526–7. 81. Pradhan N, Battaglia DM, Liu Y, Peng X. Efficient, stable, small, and water-soluble doped ZnSe nanocr ystal emitters as non-cadmium biomedical labels. Nano Lett 2007;7:312–7. 82. Sosnovik DE, Weissleder R. Emer ging concepts in molecular MRI. Curr Opin Biotechnol 2007;18:4–10. 83. Liu Z, Cai W, He L, et al. In vi vo biodistribution and highly efficient tumour targeting of carbon nanotubes in mice. Nat Nanotechnol 2007;2:47–52.

84. Liu Z, Da vis C, Cai W, et al. Circulation and long-ter m f ate of functionalized, biocompatible single-walled carbon nanotubes in mice probed b y Raman spectroscop y. Proc Natl Acad Sci U S A 2008;105:1410–5. 85. Singh R, P antarotto D, Lacerda L, et al. Tissue biodistribution and blood clearance rates of intra venously administered carbon nanotube radiotracers. Proc Natl Acad Sci U S A 2006; 103:3357–62. 86. Qian X, Peng XH, Ansari DO, et al. In vivo tumor targeting and spectroscopic detection with surf ace-enhanced Raman nanopar ticle tags. Nat Biotechnol 2008;26:83–90. 87. Keren S, Zavaleta C, Cheng Z, et al. Nonin vasive molecular imaging of small li ving subjects using Raman spectroscop y. Proc Natl Acad Sci U S A 2008;105:5844–9. 88. Thorek DL, Chen AK, Czupr yna J, Tsourkas A. Super paramagnetic iron oxide nanopar ticle probes for molecular imaging. Ann Biomed Eng 2006;34:23–38. 89. Wunderbaldinger P, Josephson L, Weissleder R. Crosslink ed iron oxides (CLIO): a ne w platform for the de velopment of targeted MR contrast agents. Acad Radiol 2002;9:S304–6.

23 NEWER BIOCONJUGATION METHODS CLAUDE F. MEARES, PHD

For molecular imaging, the core of bioconjugate chemistry is preparing ne w molecules or par ticles that illuminate biological tar gets in vi vo. This chapter focuses on ne w approaches to attach probes to molecules that accumulate in target organs or disease sites. Imaging probes come in a variety of for ms, from small molecules that accumulate in interesting sites (e g, 99mTc-methylenediphosphonate (MDP) in bone) to genes that code for the e xpression of signal-generating proteins (e g, luciferase, g reen fluorescent protein). In between are bioconjugates, generally produced by a combination of chemistr y and biolo gy, which can cause chosen tar gets to light up against a dark background. The most f amiliar examples include peptides and antibody fragments, sometimes attached to nanopar ticle frameworks to enhance performance. Another molecular imaging challenge that can be met b y bioconjugate chemistr y is the impro vement of pharmacokinetics so that e ver-smaller tar gets can be observed. F or e xample, the need for clearance from nontarget organs can be met by conjugating polar groups such as carbohydrates31 or the circulating lifetime can be extended by conjugating pol ymers such as pol yethylene glycol (PEG). 3 The cur rent state of the ar t for preparing molecular imaging probes is highl y advanced, for example, gi ven a selecti ve ligand for a receptor of interest, there are rarel y an y insuperab le bar riers to de velop a successful probe. Like synthetic chemistr y, the subject ma y be brok en down into a collection of reactions (more broadly defined to include the use of molecular genetics to produce reagents), and each single reaction step is relati vely straightforward, but putting these reactions together in the best order (or designing the gene and car rying out its expression) requires both skill and e xperience. Descriptions of standard bioconjugation techniques are a vailable in familiar references such as the book b y Her manson21 or the review by Means and Feeney.24 They have evolved

over many years and generall y focus on the synthesis of electrophilic reagents to react with electron-rich protein nucleophiles, such as c ysteine or l ysine side chains to form stab le co valent bonds. Such reagents, along with protocols for their use, can be purchased from commercial suppliers and are onl y briefly covered here. Table 1 shows some e xamples of w hat might be considered classical approaches to bioconjugation applicab le to molecular imaging. Let us consider some concrete examples of bioconjugation that illustrate the basic concepts. We star t with a small peptide, w here purel y chemical techniques are applied, mo ve to a receptor -binding protein w here the bioconjugate chemistr y occurs in aqueous solution and then describe a nanoparticle bioconjugate. The synthetic peptide octreotide and its analogues have been developed and highly ref ined to be tak en up by cells that overexpress the somatostatin receptor type-2, w hich is a cancer mark er. Octreotide (F igure 1) is a small c yclic octapeptide that ma y be thought of as a receptor -binding core linked to an alpha-amino g roup that is a per missible site for chemical modif ication. Chemical planning is complicated by the f act that the receptor -binding core contains another amino g roup on a l ysine side chain; ho wever, octreotide and its bioconjugates can be produced b y solid phase peptide synthesis, w hich allo ws us to protect the lysine amino g roup and operate onl y on the alpha-amino group. Many different modifications have been tried at the alpha-amino g roup with the aim of producing a receptor imaging bioconjugate. These include, among man y others, a dieth ylenetriaminepentaacetic acid (DTP A) residue for imaging with indium-111 (Octreoscan ®), a 1, 4, 7, 10tetraazacyclododecane-N, N', N'', N''' -tetraacetic acid (DOTA) residue for imaging with an assor tment of metal radionuclides (DOTATOC®), and a v ariety of substituents for other tags (F igure 1). 29 In each case, the chemistr y is similar, for example, the amino g roup of octreotide can be 353

354

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 1. SOME CLASSICAL EXAMPLES OF BIOCONJUGATION How to do it21

What to do Add a strong metal-binding group (for yttrium, lanthanides, etc) to a free amino group.11

Add a fluorescent dye to a free amino group.21

Using the Web, select dye from many commercially available, and follow manufacturer’s protocol.

Add a PEG polymer to a free amino group.3

See Figure 2.

Add F-18 to a free amino group.20

modified by an acti vated carboxyl g roup on DO TA using the methods of solid phase peptide synthesis.11 The DOTAoctreotide product is then removed from the beads, purified by chromatography, and radiolabeled by adding appropriate metal ions in aqueous solution. It is possib le to have all the chemical steps before radiolabeling performed by commercial laboratories that specialize in peptide synthesis. Small molecules such as octreotide bioconjugates are synthesized and purified by the methods of synthetic organic chemistry and do not come into contact with biological cells until they are used for imaging. In contrast, biological macromolecules such as antibodies are expressed in cells before purif ication and conjugation. The principal dif ference between the chemistr y of lar ge molecules and small molecules tends to be a matter of the statistics of chemical modification. Rather than a single amino g roup, a protein might ha ve 10 to 100 amino groups with similar chemical reacti vity, and a vir us particle might have thousands. The protecting-group strategies available in peptide synthesis are not applicab le to proteins under normal conditions. Engineering strategies have been de veloped for cloning unique reacti ve sites into biological macromolecules (examples are given later in the chapter), but in everyday practice, it can be necessary to prepare protein conjugates that are a comple x mixture of molecules conjugated at different sites and to different de grees. Similar considerations appl y to most synthetic nanopar ticles and also to dendrimers in some cases. For introductory purposes, we compare conjugating a protein in aqueous solution to synthesizing a conjugated peptide on a resin.

Bioconjugate chemistr y in aqueous solution is frequently dominated b y the need to a void modifying a large and variable number of amino g roups or other common chemicall y reacti ve sites. A fe w such reacti ve sites may occur in or near the binding sites of antibodies, whereas larger numbers are found in less important regions that can be modified with little effect on biological properties. The challenge of labeling is usuall y addressed empirically by exploring a range of reaction conditions and b y analyzing the proper ties of the resulting conjugates. In most cases, an excess of chemical reagent over macromolecule is required because such reagents hydrolyze in aqueous solution, but the aim is to modify onl y a few sites on the macromolecule. A useful e xample is conjugation of PEG to a particle or a macromolecule to increase its circulating lifetime. The preparation and characterization of PEG-interferon for the treatment of hepatitis pro vides an example of the chemistr y and the challenges. 3,13,14 The recombinantly produced human interferon- α2a has an inconveniently brief lifetime in circulation. Conjugation of a single lar ge 40KD PEG-NHS ester in aqueous solution with human interferon- α2a (see F igure 2) can lead to 12 positional isomers, cor responding to PEG conjugation at the N ter minal alpha-amino or at one of the 11 l ysine epsilon-amino groups; practical quantities of six isomers, each conjugated at a distinct l ysine, w ere obtained and studied by Dhalluin and colleagues. 14 Products containing multiple PEG groups are also possible, but in this example were not isolated. Dhalluin and colleagues compared the rates of receptor binding b y these “pe gylated” interferon isomers, f inding lar ge dif ferences depending on the

Newer Bioconjugation Methods

Figure 1.

355

Structure of the synthetic peptide octreotide (top) and three example bioconjugates.16,38

location of the PEG. 13 The pe gylated interferons tend to have significantly lower affinities for their target, but they also circulate for much longer periods; the net result is that an unfractionated mixture of pe gylated interferon isomers shows much impro ved therapeutic ef fects relati ve to the unmodified protein.3 An extension of this chemistr y is the preparation of PEG reagents car rying receptor ligands at their unreactive ends. F or example, Xiong and colleagues 39 labeled adenovirus par ticles with an ar ginine–glycine–aspartic acid inte grin ligand link ed to PEG-NHS, using the chemistry in F igure 2. Proteins tend to be in the same size range as nanoparticles, and nanoparticles often need to be modified by PEG to improve their biological properties. The multiplicity of potential reacti ve g roups is often more pronounced with synthetic nanopar ticles,

which ha ve more unifor m str uctures than proteins. An interesting example of using this multiplicity to adv antage is the coupling of copies of a small molecule from a library to dozens of sites on a nanoparticle and discovering unsuspected biolo gical proper ties.32,37 Again, the chemistry used is acylation of abundant amino groups on the particle by an active ester as in Figure 2 or by related chemistry, but here the aim is to conjugate as man y amino g roups as possib le. The a vidity due to multiple weak binding interactions pro vides an oppor tunity to develop very selective probes. Now let us proceed. In k eeping with the aims of this book, imaging chemistr y is emphasized rather than biomaterials or other bioconjugate applications. The flow of the subject matter is from the more biological to the more chemical methodologies.

356

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 2. A common bioconjugation reaction, between PEG–NHS and an amino group on a protein. N-hydroxy-succinimide (NHS) is formed as a product, along with the bioconjugate.

FUSION PROTEINS Modern techniques often include manipulation of the gene that codes for a tar geting protein to prepare mutants or fusion proteins that are suitab le for further bioconjugation. There are man y peptide “tags” a vailable, par ticularly for affinity capture * in protein purif ication or in w estern blotting, and there are some tags that appear w ell suited to imaging applications. F or e xample, genetic fusion of the C-terminus of a tar geting protein with an engineered peptide sequence that codes for enzymatic gl ycosylation is a promising approach to using carboh ydrate chemistr y for probe attachment or perhaps for modifying pharmacokinetics. Using pol ypeptide-R-N-acetylgalactosaminyltransferase II (ppGalN Ac-T2), Ramakrishnan and colleagues 26 transfer a C2-modif ied galactose that has a chemical handle, such as k etone or azide, from its respecti ve uridine diphosphate (UDP) sugar to the Ser/Thr residue(s) of an acceptor pol ypeptide fused to a nongl ycoprotein (Figure 3). A v ariety of sequences ma y be used for the acceptor pol ypeptide, star ting with the threonine-rich peptide PTTDSTTPAPTTK† as shown in Figure 3.

DOCK AND LOCK Another interesting genetic tagging application is “dock and lock” technolo gy, w here complementar y peptide tags on tw o dif ferent proteins can spontaneously combine and for m a stab le cross-link. Rossi and colleagues 28 use the natural binding betw een the regulatory subunits of cAMP-dependent protein kinase and the anchoring domains of A kinase anchor proteins, followed by spontaneous disulf ide bond for mation for constructing co valent conjugates from modular

*

subunits (F igure 4). Rossi and colleagues 28 validated the method b y producing bispecif ic, tri valent-binding complexes composed of three stab ly link ed Fab fragments for selecti ve deli very of radiotracers to human cancer x enografts, resulting in rapid cancer tar geting and imaging and pro viding e xceptionally high tumor/blood ratios. 2 Backer and colleagues describe another adapter/docking tag system, based on mutated fragments of human RNase I that spontaneousl y bind to each other and for m a conjugate with a disulf ide bond between complementar y c ysteine residues. This selfassembled dock and lock system uses the fusion C-tag, an amino-acid fragment of human ribonuclease I containing residues 1 to 15 with an R4C amino acid substitution, to bind a 107-amino-acid adapter protein (Ad-C), the fragment of human RNase I containing residues 21 to 127 with a V118C substitution. Two different C-tagged recombinant proteins, human v ascular endothelial g rowth f actor and an N-ter minal fragment of anthrax lethal factor, retain functional activities after spontaneous conjugation of Ad-C to N-ter minal or C-terminal C-tag. Ad-C modif ied with pegylated phospholipid and inser ted into the lipid membrane of drug-loaded liposomes retains the ability to conjugate C-tagged proteins, yielding tar geted liposomes decorated with functionally active proteins. There is also an engineered adapter with an additional c ysteine residue for site-specif ic modif ication, which can be dock ed to another protein and then labeled with an imaging probe. In pre vious work, De yev and colleagues 12 used the ribonuclease bar nase (12 kDa) and its inhibitor barstar (10 kDa), which form a very tight noncovalent complex, as tags to link antibody fragments to gether.

Affinity capture is the capture of a ligand b y its receptor, such as capture of biotin b y avidin, or histidine oligomers by nickel-NTA, or the V5 epitope by an anti-V5 antibody. †Amino acid residues are represented here b y the one-letter code. A good Web site for this is http://en.wikipedia.or g/wiki/List_of_standard_amino_acids.

Newer Bioconjugation Methods

357

Figure 3. Enzymatic glycosylation of a nonglycoprotein that has been genetically tagged with a peptide substrate for the cloned enzyme polypeptide-R-N-acetylgalactosaminyltransferase II. Modified galactose residues such as the illustrated azido compound can be transferred from the appropriate UDP-sugar to multiple threonine residues on the peptide tag.26

Figure 4. Using peptides with mutual affinity, and disulfide linkages, to form a stable trivalent protein.28 A dimerizing peptide tag containing the amino-acid sequenceii CGHIQIPPGLTELLQGYTVEVLRQQPPDLVEFAVEYFTRLREARA is genetically added to the C-terminus of the heavy chain of a targeting Fab fragment B, which is consequently expressed as a B-B dimer with two free cysteine sulfhydryl groups in position for the next step. Further, a complementary peptide containing the sequence CGQIEYLAKQIVDNAIQQAGC, with two cysteine residues that will bind close to those in the B-B dimer, is genetically added to the C-terminus of the heavy chain of a different Fab fragment A. The latter is expressed and added to the B-B dimer in vitro. Spontaneously, a stable disulfide-linked trivalent complex forms, cross-linking all three Fab fragments.

INFINITE AFFINITY More of a probe-capture strate gy than bioconjugation, antibodies with inf inite af finity w ere de veloped b y the Meares group to make specif ic binding pairs that do not dissociate.6–8 This concept is reduced to practice b y genetically engineering the binding site of an antibody or other binding protein so that upon binding a specif ically modified synthetic ligand , a stab le co valent bond is formed, per manently attaching the tw o (F igure 5). We have de veloped this technolo gy for the capture of 5 imaging/therapy probes based on metal chelates, whereas others have explored diverse applications.23

NATIVE LIGATION The technology of nati ve chemical ligation ‡ of peptide or protein fragments to form polypeptide backbones has been

highly de veloped since the original w ork of Da wson and colleagues.9 Here, a molecule with a C-ter minal thioester reacts with an N-terminal cysteine residue on another molecule, f irst transfer ring its ac yl g roup to mak e a thioester linkage to the c ysteine side chain and then transfer ring the acyl g roup to the nearb y ter minal amino g roup to for m a peptide bond (F igure 6). F or e xample, Gro gan and colleagues17 and Reulen and colleagues27 use this chemistry to prepare protein-liposome conjugates using c ysteine lipids. Intein fusion proteins can be captured on an af finity column and treated with mercaptoethanesulfonic acid to release the desired protein, such as g reen fluorescent protein, in thioester linkage to mercaptoethanesulfonate. The protein thioester can ligate to a c ysteinyl lipid to for m a native peptide bond. Alternatively, K ushnir and colleagues 22 produce a protein with a c ysteine N-terminus and conjugate it to a probe-thioester by analogous chemistry.

‡Native ligation refers to the joining of tw o polypeptides to make a larger structure by forming a native peptide bond, as distinct from a synthetic cross-link.

358

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

SH

S

SH +

antibody ligand

reversible complex

irreversible complex

Figure 5. Engineering an antibody–ligand binding pair by genetically placing a reactive amino acid side chain adjacent to the antibody binding site and synthesizing a complementary reactive group on the ligand.6 The complex forms due to the affinity of the antibody for the ligand; a cross-linking reaction, accelerated by the local concentration of the two reactive groups in the complex, prevents dissociation of the ligand from the antibody.

Figure 6. Native chemical ligation of thioesters to N-terminal cysteine residues to form a stable amide or peptide bond. Groups A and B can be protein or peptide, but also can be various other molecules, from probes to lipids. The only common feature is a thioester and an N-terminal cysteine.

ENZYMATIC TRANSPEPTIDATION The enzyme transglutaminase can be used to replace the side-chain amide of glutamine with synthetic amines. 15 Other enzymes such as the transpeptidases act on the polypeptide backbone to ligate tw o molecules to gether. Sortase is an interesting e xample, which Parthasarathy and colleagues 25 have recently shown to attach proteins carrying the appropriate tag sequence to a v ariety of amine-bearing compounds, ranging from other proteins to polystyrene beads (Figure 7). The sortase recognition sequence is shor t and simple, LPETG; it can be conjugated readil y to an N-ter minal GGG sequence on another protein or , with less ef ficiency, to primar y amino groups in general.

ORGANIC CHEMISTRY Click Chemistry Since the pub lication of the w ork b y Wang and colleagues,36 tagging probes and biomolecules with terminal azides and alkynes has become a very popular route to bioconjugation reactions that join a wide v ariety of targeting and probe functions involving either modified natural or totall y synthetic molecules via copper catalyzed for mation of c yclic triazoles. This is the

focus of a separate chapter in this book, so the description here is brief. We note that recent de velopments by van Berk el and colleagues 35 and b y Baskin and colleagues4 in ef ficiently for ming the same or similar products without added copper promise to enhance the value of this reaction (Figure 8). The scheme described in a study b y van Berkel and colleagues 35 appears particularly simple, in volving the addition of furan to an electron-deficient alkyne, followed by cycloaddition of azide and loss of furan, to gi ve the e xpected triazole product.

Staudinger Ligation This is an alter native method leading to an amide or peptide bond under mild conditions in w ater, applied to attach probes to cells by Saxon and Bertozzi.30 Like click chemistry, one reactant is an azide. Lik e native ligation, the other reactant is a thioester. Unlike either, a phosphine group, w hich reduces the azide, must be intimatel y involved. The strategy for including the phosphine is k ey to useful bioconjugation, as illustrated b y Tam and colleagues.33 The e xample in F igure 9 car ries out a “traceless” ligation in the course of w hich nonbiological groups are eliminated , producing a nati ve protein from two peptides.

Newer Bioconjugation Methods

359

Figure 7. Protein ligation (right arrow) mediated by the transpeptidase sortase A. The C-terminal LPETG sequence is cleaved between T and G, forming an acyl-enzyme intermediate that can be attacked by nucleophiles. Even beads carrying amino groups can serve as nucleophiles in this reaction (left arrow).

Figure 8.

Metal-free analog of click chemistry, using a furan adduct of an alkyne.35

Figure 9. Traceless Staudinger ligation to join two peptides, showing the thioester favored by Tam et al.33 This procedure is expected to be applicable to nonpeptides as well.

Diels-Alder Conjugation

ORGANOMETALLIC COUPLING

An unusual reaction that tak es adv antage of the lar ge number of maleimide-tagged probes available commercially, Diels-Alder bioconjugation requires the synthesis of a diene-tagged biomolecule, as illustrated by the work of Häner and Tona18,19 with nucleic acids (Figure 10) or Waldmann and colleagues with proteins and peptides. 10

While the y ha ve not y et been applied to imaging, organometallic reagents are beginning to be used to modify the side chains of aromatic residues in proteins in aqueous solution. Recent examples include the modification of tryptophan by Antos and F rancis1 using rhodium reagents and the modif ication of tyrosine b y Tilley and F rancis34 using palladium reagents. These explorations represent interesting

360

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 10. The classic Diels–Alder reaction links a conjugated diene to a dienophile, an electron-poor alkene. Here is an example of Diels–Alder bioconjugation of a maleimide-tagged probe molecule to a synthetic DNA hairpin containing a diene in the loop region.19 The reaction proceeds under very mild aqueous conditions, but requires several days.

Figure 11. Use of an electrophilic p-allyl palladium complex to selectively alkylate tyrosine side chains in a protein in aqueous solution.34 Reasonable yields are obtained at pH=9, room temperature, 45 min.

new chemistr y that ma y f ind a place among the useful bioconjugation reactions in the future (Figure 11).

SUMMARY Today, we have access to a set of molecular tools and techniques that per mit a v ariety of approaches to molecular imaging, a variety that may appear bewildering to the nonspecialist. In de veloping a molecular imaging procedure, one way to na vigate through the details is to assume the point of view most familiar to you. If your background is chemistry, then synthesizing small-molecule probes has a long and successful histor y; the major challenges to this approach lie in the numerous possibilities for poor pharmacokinetics of the products and the potential for long development times. If your background is molecular biology, then engineered fusion proteins ha ve many attractive features including binding specif icity and relati vely predictable phar macokinetics; the major challenges lie in expressing useful quantities of biologically active proteins and avoiding immunogenicity. If your background is engineering, then nanopar ticles ha ve interesting proper ties, particularly for MRI contrast agents; the challenges here include control of pharmacokinetics and metabolism. The most promising oppor tunities for the future appear to lie in combinations of chemistr y, molecular biolo gy and engineering that will allo w multiplexed imaging of diagnostic sets of molecular tar gets, such as gene products under control of different promoters, as well as therapy of the identif ied cells. Some precursors to that technolo gy may be found in this chapter.

REFERENCES 1. Antos JM, F rancis MB . Selecti ve tr yptophan modif ication with rhodium carbenoids in aqueous solution. J Am Chem Soc 2004;126:10256–7. 2. Backer MB, Patel V, Jehning BT, Backer JM. Self-assemb led “dock and lock” system for linking pa yloads to tar geting proteins. Bioconjug Chem 2006;17:912–9. 3. Bailon P, Palleroni A, Schaffer CA, et al. Rational design of a potent, long-lasting for m of interferon: a 40 kDa branched pol yethylene glycol-conjugated interferon α-2a for the treatment of hepatitis C. Bioconjug Chem 2001;12:195–202. 4. Baskin JM, Prescher JA, Laughlin ST, et al. Copper-free click chemistry for dynamic in vi vo imaging. Proc Natl Acad Sci USA 2007;104:16793–7. 5. Butlin NG, Meares CF. Antibodies with inf inite affinity: origins and applications. Acc Chem Res 2006;39:780–7. 6. Chmura AJ, Orton MS, Meares CF. Antibodies with inf inite affinity. Proc Natl Acad Sci USA 2001;98:8480–4. 7. Corneillie TM, Lee KC, Whetstone PA, et al. Irreversible engineering of the multielement-binding antibody 2D12.5 and its complementary ligands. Bioconjug Chem 2004;15:1392–1402. 8. Corneillie TM, Whetstone P A, Lee KC, et al. Con verting w eak binders into inf inite binders. Bioconjug Chem 2004;15:1389–91. 9. Dawson PE, Muir TW, Clark-Lewis I, Kent SB. Synthesis of proteins by native chemical ligation. Science 1994;266:776–9. 10. de Araújo AD, Palomo JM, Cramer J, et al. Diels-alder ligation of peptides and proteins. Chemistry 2006;12:6095–6109. 11. De Leon-Rodriguez LM, K ovacs Z. The synthesis and chelation chemistry of DO TA-peptide conjugates. Bioconjug Chem 2008;19:391–402. 12. Deyev SM, Waibel R, Lebedenk o EN , et al. Design of multi valent complexes using the bar nase*barstar module. Nat Biotechnol 2003;21:1486–92. 13. Dhalluin C, Ross A, Huber W, et al. Structural, kinetic, and thermodynamic analysis of the binding of the 40 kDa PEG-interferon- α2a and its individual positional isomers to the extracellular domain of the receptor IFNAR2. Bioconjug Chem 2005;16:518–27. 14. Dhalluin C, Ross A, Leuthold LA, et al. Str uctural and bioph ysical characterization of the 40 kDa PEG-interferon- α2a and its individual positional isomers. Bioconjug Chem 2005;16:504–17.

Newer Bioconjugation Methods

15. Folk JE. Mechanism and basis for specif icity of transglutaminasecatalyzed epsilon-gamma-glutam yl l ysine bond for mation. Adv Enzymol Relat Areas Mol Biol 1983; 54:1–56. 16. Forster GJ, Engelbach MJ, Brockmann JJ, et al. Preliminar y data on biodistribution and dosimetr y for therap y planning of somatostatin receptor positive tumours: comparison of (86)Y-DOTATOC and (111)In-DTPA-octreotide. Eur J Nucl Med 2001;28:1743–50. 17. Grogan MJ, Kaizuka Y, Conrad RM, et al. Synthesis of lipidated g reen fluorescent protein and its incorporation in supported lipid bilayers. J Am Chem Soc 2005;127:14383–7. 18. Häner R, Tona R. Functionalisation of a diene-modified hairpin mimic via the Diels-Alder reaction. Chem Commun 2004;1908–9. 19. Häner R, Tona R. Synthesis and bioconjugation of diene-modif ied oligonucleotides. Bioconjug Chem 2005;16:837–42. 20. Haubner R, Kuhnast B, Mang C, et al. [18F]Galacto-RGD: synthesis, radiolabeling, metabolic stability , and radiation dose estimates. Bioconjug Chem 2004;15:61–9. 21. Hermanson GT. Bioconjugate techniques. 2nd ed. San Die go (CA): Academic Press; 2008. 22. Kushnir S, Marsac Y, Breitling R, et al. Rapid production of functionalized recombinant proteins: mar rying ligation independent cloning and in vitro protein ligation. Bioconjug Chem 2006;17:610–7. 23. Levitsky K, Boersma MD , Ciolli CJ , Belshaw PJ. Exo-mechanism proximity-accelerated alk ylations: in vestigations of link ers, electrophiles and surf ace mutations in engineered c yclophilincyclosporin systems. Chembiochem 2005;6:890–9. 24. Means GE, F eeney RE. Chemical modif ications of proteins: histor y and applications. Bioconjug Chem 1990;1:2–12. 25. Parthasarathy R, Subramanian S, Boder ET. Sortase a as a novel molecular “stapler” for sequence-specif ic protein conjugation. Bioconjug Chem 2007;18:469–76. 26. Ramakrishnan B, Boeggeman E, Qasba PK. Novel method for in vitro O-glycosylation of proteins: application for bioconjugation. Bioconjug Chem 2007;18:1912–8. 27. Reulen SWA, Brusselaars WWT, Langereis S, et al. Protein-liposome conjugates using c ysteine-lipids and nati ve chemical ligation. Bioconjug Chem 2007;18:590–6.

361

28. Rossi EA, Goldenber g DM, Cardillo TM, et al. Stab ly tethered multifunctional str uctures of def ined composition made b y the dock and lock method for use in cancer targeting. Proc Natl Acad Sci USA 2006;103:6841–6. 29. Rufini V, Calcagni ML, Baum RP. Imaging of neuroendocrine tumors. Semin Nucl Med 2006;36:228–47. 30. Saxon E, Ber tozzi CR. Cell surf ace engineering b y a modif ied Staudinger reaction. Science 2000;287:2007–10. 31. Schottelius M, Wester HJ, Reubi JC, et al. Impro vement of phar macokinetics of radioiodinated Tyr3-octreotide by conjugation with carbohydrates. Bioconjug Chem 2002; 13:1021–30. 32. Sun EY , Josephson L, K elly KA, Weissleder R. De velopment of nanoparticle libraries for biosensing. Bioconjug Chem 2006;17:109–113. 33. Tam A, Soellner MB , Raines R T. Water-soluble phosphinothiols for traceless staudinger ligation and inte gration with e xpressed protein ligation. J Am Chem Soc 2007; 129:11421–30. 34. Tilley SD , F rancis MB . Tyrosine-selective protein alk ylation using pi-allylpalladium complexes. J Am Chem Soc 2006;128:1080–81. 35. van Berkel SS, Dirks ATJ, Debets MF, et al. Metal-free triazole formation as a tool for bioconjugation. Chembiochem 2007;8:1504–08. 36. Wang Q, Chan TR, Hilg raf R, et al. Bioconjugation b y copper(I)catalyzed azide-alk yne [3 + 2] c ycloaddition. J Am Chem Soc 2003;125:3192–3. 37. Weissleder R, K elly K, Sun EY , et al. Cell-specif ic tar geting of nanoparticles by multivalent attachment of small molecules. Nat Biotechnol 2005;23:1418–23. 38. Wester HJ , Schottelius M, Scheidhauer K, et al. PET imaging of somatostatin receptors: design, synthesis and pre-clinical e valuation of a no vel 18F-labelled , carboh ydrated analo gue of octreotide. Eur J Nucl Med Mol Imaging 2003;30:117–22. 39. Xiong Z, Cheng Z, Zhang X, et al. Imaging chemically modified adenovirus for tar geting tumors e xpressing inte grin alpha vbeta3 in living mice with mutant her pes simple x vir us type 1 th ymidine kinase PET reporter gene. J Nucl Med 2006;47:130–9.

24 TARGETED ANTIBODIES

AND

PEPTIDES

MICHAEL R. LEWIS, MS, PHD, CATHY S. CUTLER, PHD, AND SILVIA S. JURISSON, PHD

Since Paul Ehrlich postulated the “magic bullet” concept in 1906, a compound targeting a disease delivered with an agent of selecti vity, numerous tar geted dr ugs and agents have been de veloped for molecular imaging and therap y of catastrophic diseases. In 1908, Ehrlich and Il ya Ilyich Mechnikov were awarded the Nobel Prize in Ph ysiology or Medicine “in reco gnition of their w ork on immunity.” Ehrlich believed that his prize-winning antisera contained distinct, acti ve chemical substances, w hich he called “antibodies.” Fur thermore, he de veloped the “side-chain theory” of molecular tar geting, stating that his so-called “magic bullets” bind to these side-chains “lik e a key in a lock.” Thus, Ehrlich’s w ork of the late 1800s and earl y 1900s foretold and under pinned the moder n principles by which we understand how targeted antibodies and peptides, respectively, bind their antigens and receptors with high af finity, specif icity, and selecti vity. In 1984, Niels Jerne, Georges Köhler, and César Milstein w ere awarded the Nobel Prize in Ph ysiology or Medicine, in par t for their “disco very of the principle for the production of monoclonal antibodies” (mAbs), that is, antibodies produced by cells derived from a single pro genitor cell. This discovery ushered in a new era of molecular medicine, as mAbs were widely believed to be the ultimate “magic bullets” for nonin vasive diagnosis and treatment of disease, mainly cancer. In the area of disease targeting, there exists a close and inseparab le relationship betw een molecular imaging and therapy. Despite the optimism and enthusiasm with w hich radioactively labeled mAbs were heralded as new molecular imaging agents, the y have suffered from a number of limitations. The major limitation of intact antibodies as planar gamma scintig raphy agents is their long residence time in the b lood, obscuring high imaging contrast in patients for periods as long as da ys to weeks. When translated to radioimmunotherapy of epithelial cancers, the long circulating times of mAbs often result in dose-limiting 362

toxicity, usuall y bone mar row suppression, before therapeutic efficacy can be achieved. In contrast, radioactive peptides typically exhibit rapid targeting of diseased tissue and rapid w hole body clearance, mark edly impro ving imaging contrast. Ho wever, radiolabeled peptides can suffer from differential uptake in various sites of disease, such as primar y tumors v ersus distant metastases, and rapid w ashout from tar get tissue, sometimes creating dif ficulties in optimizing imaging times for all patients. Moreo ver, radiolabeled peptides often do not ha ve absolute tar get tissue uptak e as high as antibodies, and translation to therapy usually requires large injected doses and progressive systemic toxicity. Thus, there ha ve traditionall y been tw o “camps” of molecular imaging and therap y in this area: members of the “antibody camp” and members of the “peptide camp, ” each of w hom ha ve tak en a v ariety of chemical and biochemical approaches to circumvent the shortcomings of their respecti ve imaging or therapeutic modalities. Research in antibody imaging has focused primaril y on bioengineering of smaller fragments, such as single-chain constructs, which have more rapid targeting and clearance properties. In peptide imaging, researchers ha ve focused on the dif ferences in the tar geting proper ties of agonist versus antagonist agents and peptide-radiolabel link ers that function as pharmacokinetic modifiers, even cleavable linkers designed for intracellular residualization. Thirdly, predominantly in e xperimental v enues, there is an “antibody pre-targeting camp” that seeks to combine the high target tissue uptake of antibodies with the rapid clearance of peptides. Re gardless of this seemingl y ideal combination, antibody pre-targeting has been beset with limitations of comple x administration protocols, diminished patient compliance, immuno genic tar geting agents, lo w af finity, and dif ficult in vi vo chemistr y. The adv antages and disadvantages of each of these deli very platforms for molecular imaging are discussed next in this chapter.

Targeted Antibodies and Peptides

TARGETED ANTIBODIES The introduction of monoclonal antibody technology was initially very attractive because the idea, re garded as the “full realization of the magic bullet concept, ” seemed so simple. However, a number of challenges and limitations belie the conceptual simplicity of these so-called “magic bullets.” F irst, the emission characteristics of the most commonly used radionuclide, 131I, result in unw anted radiation burdens to patients and the distinct possibility of therapeutic ef fects during the course of molecular imaging, that is, if the agent has potential therapeutic properties as well, treatment planning could be biased by the imaging session. Second , intact mAbs suf fer from a number of major prob lems in tar geting solid tumors, including slo w tumor uptak e and b lood clearance, lo w absolute tumor accumulation in humans, une ven tumor penetration resulting from slow diffusion constants, high interstitial pressure of tumors, a “binding site bar rier” in which mAbs tend to bind to the f irst antigen molecules encountered, and antigenic heterogeneity of tumors. One or a combination of these f actors usually results in poor imaging contrast w hen using intact antibodies. An example of clinical cancer imaging with an 111In-labeled antibody is shown in Figure 1 (left side). However, hematologic malignancies are good candidates for antibody therapy because the y are b lood-borne diseases and the b lood is the major residence compar tment of intact antibodies.

Figure 1. Planar gamma scintigraphy of two cancer patients. The patient on the left was imaged 48 hours post-injection of the radiolabeled antibody 111In-ibritumomab tiuxetan. Normal uptake in the liver (L) and blood pool is evident, as well as uptake in a malignant abdominal mass (T). The patient on the right was imaged 4 hours postinjection of the radiolabeled peptide 111In-DTPA-octreotide. Normal uptake in the spleen is evident on the right, as well as urinary excretion. Also, evident is a malignant abdominal mass (T) and extensive liver metastases (L). Note how differences in imaging contrast were obtained with each drug.

363

The “building b locks” of an ef fective therapeutic radiolabeled antibody , or radioimmunoconjugate, ha ve been re viewed b y Schubiger and colleagues, 1 but it is important to k eep in mind that the same principles that apply to ef fective radioimmunotherap y also appl y to radioimmunoscintigraphy. The v arious components of a radioimmunoconjugate that must be optimized for successful imaging and therapy are the radionuclide (Table 1), the chelating agent (F igure 2) in the case of metallic radionuclides, and the v ehicle, or antibody constr uct. Because of the slow tumor tar geting of intact antibodies, the ph ysical half-life of the radionuclide should match the biolo gical half-life of the mAb constr uct, the chelator should bind radiometals with high in vi vo stability , and the v ehicle should exhibit optimal tumor uptak e and non-tar get tissue clearance over a relatively short time interval. When selecting a radionuclide for antibody-based imaging, several factors are of primar y importance. These factors include the nature of radioacti ve emissions, the ratio of penetrating to non-penetrating radiation, the ph ysical half-life of the radionuclide, the biolo gical stability of the radionuclide, the ph ysical stability of the daughter nuclide(s), and the a vailability of the radionuclide. For example, one of the most readil y a vailable radionuclides, 131I, emits both penetrating gamma ra ys and nonpenetrating, cytotoxic beta minus (β−) particles. Dosimetry calculations in humans 2 show that the ratio of absorbed doses from 131I penetrating and non-penetrating radiation is appro ximately 50% each. Deca y of 131I results in an unwanted burden from non-penetrating radiation during strictly diagnostic procedures. Fur thermore, the c ytotoxic effects of 131I β− radiation invoke the possibility that a therapeutic ef fect ma y occur that under mines the diagnostic information obtained unless v ery lo w injected doses are administered. Therefore, most antibodies ha ve been labeled with 111In or 99mTc for radioimmunoscintig raphy. These radionuclides ha ve much higher ratios of penetrating radiation than 131I, and their lower energy gamma emissions result in higher resolution and image quality. In the case of 99mTc, a direct labeling approach b y reducing interchain disulf ide bonds and allo wing the radiometal to bind directl y to the resulting sulfh ydryl groups has been used, analogous to that used with rhenium radioisotopes.3 This labeling method is achie vable because metals such as technetium and rhenium have several possible o xidation states, as w ell as a high af finity for sulfur . However, if an antibody fragment, such as F(ab’) 2 generated by pepsin cleavage of intact antibodies, is labeled with 99m Tc in this manner , disulf ide reduction produces tw o F(ab)’ half molecules, which bind the radiometal to form an undesirable mixture of products, resulting in loss of

364

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 1. PROPERTIES OF DIAGNOSTIC RADIONUCLIDES108–112 Radionuclide

Maximum Particle Emission (%)

Gamma Emission (%)

Half-life

18

0.635 MeV β+

511 keV (200%)

109.8 min

64

0.653 MeV β+ (19%)

511 keV (38.6%)

12.7 h

F Cu

EC decay (41%) 0.579 MeV β− (40%) 68

Ga

1.899 MeV β+ (90%) EC decay (10%)

86

Y

1077 keV (3.3%)

67.63 min

511 keV (176%)

3.15 MeV β+ (34%)

1921 keV (20.8%)

EC decay (66%)

1854 keV (17.2%)

14.74 h

1153 keV (30.54%) 1077 keV (82.5%) 777 keV (22.45%) 703 keV (15.4%) 511 keV (67.2%) 99m

Tc

IT

140 keV (89.3%)

6.01 h

111

In

EC decay

245 keV (94%)

2.83 d

171 keV (91%) 123

EC decay

159 keV (100%)

131

0.971 MeV β−

284.3 keV (7.5%)

I I

13.2 h 8.02 d

364.5 keV (100%) 637 keV (8.8%) EC = electron capture; IT = isometric transition.

immunoreactivity. In contrast, more promising results have been obtained using the bifunctional chelate approach, in which a 99mTc-binding ligand, based on a N ,S multidentate system such as MA G3 (Figure 2), is appended to the antibody. The use of MA G3 for 99mTc labeling of antibodies and fragments requires a pre-labeling reaction sequence, in which the radiometal is f irst complexed with the bifunctional chelating agent, which is then activated in the form of a reactive ester and conjugated to the mAb . This reaction preserves high immunoreacti vity, but it is a multistep method that can result in decreased radiolabeling yield in some cases. 111 In(III), For simple tri valent radiometals, such as polyamino pol ycarboxylate ligands, such as EDT A, DTPA, and DOTA (Figure 2) can be used for antibody labeling.These bifunctional chelating agents can be used in a post-labeling reaction sequence, in w hich the chelator is f irst attached to the antibody, and then the radiometal is complexed to the conjugate in a “kit” fashion in which the two are mixed and incubated. Conjugation of DTPA is often performed by reaction of the antibody with the bic yclic anh ydride of the ligand. 4

However, this method can result in cross-linking of antibodies and loss of immunoreacti vity. Therefore, acti ve ester 5 and isothiocyanate6 derivatives of DTP A have been used to circumvent cross-linking. DOTA conjugation methods have also used these activating groups, including isothiocyanates7,8 and active esters.9 Another impor tant consideration in the selection of bifunctional chelating agents for radiometal labeling of antibodies is in vi vo stability. For example, DTPA has high ther modynamic stability for 111In, but the most important consideration in vi vo is kinetic stability or the off rate of the metal-chelator dissociation in biolo gical systems. The high ther modynamic stability of DTP A in binding 111In results from a combination of a fast on rate and a fast off rate. This property can lead to in vi vo transchelation to transfer rin in the b loodstream, resulting in high liver uptake mediated by transferrin receptors. On the other hand, the high thermodynamic stability of DOTA for 111 In and other radiometals results from a slow on rate and a very slo w off rate, making DO TA considerab ly more kinetically stab le in vi vo for a v ariety of radiometals.

Targeted Antibodies and Peptides

Various Chelates Used for Radiometals:

Radioiodine Chemistry

DOTA

DTPA

HOOC

COOH N

N

N

N COOH

O COOH

HOOC N

N

N

N

N

N

O

HOOC

O

+

O

O

Na∗I/Iodogen

N O

Radioiodination generally involves using the tributyltin as the leaving group and oxidizing radioiodide (I-131, I-123, I-125, etc.) with lodogen, for example, for the eletrophilic substitution reaction on N-succinimidyl 3-iodo benzoate (SIB). The substitution is generally on the meta (as shown) position relative to the benzoic acid functionality, although the para position is also possible.

COOH N

∗I

N

COOH

NOTA COOH

Bu3Sn

COOH

HOOC

TETA HOOC

O

O N

HOOC

365

N N

COOH

HOOC

Cross-bridged TE2A

COOH

Sargeson Cage: Sar(Ar) Radiofluorine Chemistry NH2

HN

O Me3N+

18F

Ot−Bu

TfO−

N N N

O

K[222]+18F −

Ot-Bu

COOH N

N

N N N N

N

HOOC

NH2

O

Tc-HYNIC core

NH

Br

N

HN

L

O COOH

L

Tc

OC

L

L L Note: Tc=N=N bonds should be conjugated not double bonds. L = ancillary ligands such as tricine.

Tc-tricarbonyl core

H2O

2

O 18F

Me4NOH

OH

O K[222]/[18F]KF OH TEA

18F

N

HN

N SH

+NMe

O O

O

N

O

HCI

BF4−

O

OH

O

O

18F

O MAG3 (N3S core)

N

Me2N

Note: Hydrogen atoms on each cage N are omitted for clarity.

O

O

O

The above figure shows the synthesis of [18F]FPA, F-18 fluoropropionic acid, an alternative to fluorobenzoic acid (above). This method requires subsequent acivation for conjugation to peptides.

EDTA

OH2 OH2 Tc CO

HOOC

CO

HOOC

COOH N

N COOH

Note: The three coordinated water molecules are readily replaced by a tridentate chelate, which would serve to link to the peptide or antibody.

Figure 2.

Various chelates used for radiometals, as well as labeling reactions for radioiodine and radiofluorine. (Taken from Refs. 113–117.)

However, the slo w on rate of DO TA often requires immunoconjugates to be heated for stab le comple xation of radiometals, w hich can cause antibody denaturation and loss of biolo gical activity. DOTA has also been used for labeling antibodies with 64Cu for positron emission tomography (PET). 10 Copper is a redo x and biolo gically active metal, and w hile 64Cu-DOTA antibody conjugates may be kinetically stable ex vivo, in vivo reduction of 64Cu can result in unwanted accumulation in liver and kidneys. In addition, DOTA is not selective for 64Cu, and the presence of stab le contaminants in preparations of the radiometal can inhibit ef ficient labeling. Deri vatives of the 14-membered macrocycle TETA (Figure 2), which can

be labeled at room temperature, 11,12 are therefore often used to bind Cu(II) in a stab le octahedral conf iguration. Moreover, TETA is relatively selective for binding 64Cu in the presence of metal contaminants. In addition to the radionuclide and bifunctional chelating agent, the other major consideration in mAbbased molecular imaging is the nature of the tar geted antibody construct or fragment. Antibodies can be modified in size or composition b y genetic engineering or enzymatic clea vage to alter their biolo gical proper ties. Cleavage of 160-kDa intact IgG mAbs with pepsin or papain generates 110-kDa F(ab’) 2 or 55-kDa F ab fragments, respectively. The lower molecular weight of these

366

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

fragments causes their targeting and clearance properties to be more rapid than IgGs, and liver uptake of these fragments is lower than that of intact mAbs. Ho wever, when labeled with radiometals, kidney uptake of enzymatically generated fragments can be quite high. In addition to high li ver and kidne y uptake of intact IgG antibodies and their F(abʹ′)2 or Fab fragments, another disadvantage of murine mAbs is their immuno genicity, resulting in human anti-mouse antibody (HAMA) reactions occurring at an incidence of 50% per dose. The first step tak en to circumv ent this neutralizing ef fect w as to construct human-murine chimeric mAbs, in w hich the mouse constant domains are replaced with human sequences and onl y the antigen-binding v ariable re gions are derived from mouse sequences. 13 Unfortunately, this strategy did not completel y resolv e the immuno genicity problem, as human anti-chimeric antibody (HA CA) responses are also frequentl y obser ved. The next step in engineering of mAbs w as based on the confor mational similarity of the murine and human frame work re gions surrounding the complementarity-deter mining regions in the antigen-combining sites. However, the resulting threedimensional conformation of these humanized antibodies are sometimes suf ficiently dif ferent from those of the murine or chimeric constr uct such that the af finity is significantly reduced. Following the same logic that produced enzymatically generated fragments, namel y that smaller molecules will partition more rapidly in vivo, genetically engineered antibody fragments ha ve also been made and e valuated. Colcher and colleagues14 performed the first in vivo evaluation of a bioengineered , 27-kDa single-chain antibody (scFv), consisting of the variable regions of the IgG heavy and light chains connected b y an oligopeptide link er, in tumor-bearing mice. Although these authors obser ved more rapid w hole body clearance of the scFv constr uct compared to the cor responding F(ab’) 2 fragment, no differences in affinity or tumor uptake were found. However, most subsequent studies ha ve sho wn v ery lo w absolute tumor uptake of scFvs in tumor xenograft models. Further engineering efforts have sought to increase tumor uptak e by constructing di-, tri-, and tetrameric scFvs, with higher molecular weight, valence, and avidity.15,16 A particularly interesting bioengineered fragment, the homodimeric 80kDa “minibody” de veloped by Wu and colleagues, consists of tw o murine scFvs attached to the human disulfide-linked hinge re gion and tw o human CH 3 domains, which serve as h ydrophobic regions that maintain structural stability. This construct has been e valuated for 64Cu high resolution rodent PET 17 and 123I single photon emission computed tomo graphy (SPECT) in

patients.18 Molecular imaging of the minibody in mice and patients sho wed tumor uptak e equi valent to intact IgG, with f aster w hole body clearance and mark edly improved imaging contrast. Ho wever, when labeled with 64 Cu, liver uptake in mice was quite high. Further impro vements in molecular imaging using targeted antibodies might be achie ved using principles described by the “imaging f igure of merit” proposed b y Williams and colleagues. 19 These authors compared numerically five members of the mAb family, IgG1, scFv, diabody (divalent scFv), minibody , and F(ab ʹ′)2, reactive against the same epitope of the same antigen, labeled with dif ferent imaging radionuclides. It w as found that the optimal imaging f igures of merit, as a function of time post-administration, w ere as follo ws: 131I for IgG 1, 123 I for F(abʹ′)2 and minibody, and 18F for the diabody. The scFv had lo w imaging f igures of merit for all radionuclides, likely the result of v ery low tumor uptak e. These results suggest that if tar geted antibody-based molecular imaging is to be improved in the future, the physical halflife of the radionuclide should be matched with a fragment geneticall y engineered for a matched biolo gical half-life.

TARGETED PEPTIDES Peptides are increasingl y being used as v ectors for radiodiagnostics due to the high number of receptors found to be overexpressed in human cancers in vivo and their f aster clearance rates and rapid tissue and tumor penetration. In addition, peptides rarel y induce an immune response and can be produced more easil y and less expensively than antibodies. Peptides do have some drawbacks, for example, endogenous peptides can cause unwanted physiological side effects even at low concentrations; high kidne y uptak e is often obser ved, lo wer tumor uptak e is obtained , and e ven slight changes in conformation due to amino acid substitution or modification can result in huge changes in their af finity. In the past, agonists, or peptides that acti vate receptors, were predominantl y used as ligands, as the y under go receptor-mediated internalization upon binding, thereby optimizing residualization of radioacti vity in tar get tissues. Ph ysiological side ef fects can be minimal if the peptide is administered at a so-called “tracer le vel,” where the mass dose is so small as not to cause ph ysiological ef fects. More recentl y, antagonists, or peptides that bind receptors without acti vation, have been investigated not onl y for minimizing phar macological side effects, but also because a much larger number of binding sites per receptor molecule are generall y present.

Targeted Antibodies and Peptides

A number of factors must be considered when designing radiolabeled peptides for molecular imaging, man y of which are similar to those previously discussed for antibody v ectors: w hich radionuclide; ho w to incor porate the radionuclide; is the uptak e specif ic and selecti ve; how stable is the peptide in vi vo; and are modif ications to the peptide required to impro ve stability and biodistribution properties. A major hurdle encountered in using native peptides is their relatively short biological half-lives, a few minutes, for radioimaging. This is nor mally due to rapid proteol ysis by enzymes nor mally present in the b lood stream. Modif ications can be made to the peptide to lengthen its biolo gical half-life by determining where enzymatic clea vage occurs and switching the L-amino acids for D-amino acids, unnatural amino acids, or peptide-like structures called mimetics. Capping or modifying the C- or N-ter minus has also been shown to minimize enzymatic clea vage. In cer tain cases, the peptide can be fragmented to remo ve regions that are less stable and only maintain the regions that are important for binding to the receptor . Although this technique has been shown to increase biolo gical half-life, as in the case with antibody fragments, removing portions of the peptide can reduce af finity or lead to binding to fe wer receptor subtypes. Cyclization of the peptide can often be performed in a reasonably straightforward manner by judicious placement of disulfide-forming c ysteine residues. Cyclization may also reduce proteolysis and can enhance receptor binding and cellular uptake. It is often necessar y to modify the nati ve peptide to change its biodistribution. F or instance, most peptides clear via the kidneys, and if the presence of tumors in the kidneys is to be detected , it might be necessar y to mak e the native peptide more hydrophobic to clear via the liver and gastrointestinal tract. In most cases, the re verse is noted; due to the need to detect tumors or metastases in the li ver, it is necessar y to mak e the peptide more hydrophilic to clear via the urinar y tract. Man y of these modifications can be made through a linker that is used to attach the chelator to the peptide. The physical and chemical proper ties of the radionuclide go vern the radiolabeling and purif ication methods for the par ticular tar geted peptide. Many human tumors o ver-express somatostatin receptors. Somatostatin-14 and -28 are the tw o most prominent native for ms of somatostatin, but their v ery short biological half-life (< 4 min) precludes their use as targeting v ectors. A number of analo gues ha ve been designed that e xhibit high af finity for the somatostatin receptor and ha ve much longer in vi vo residence times.20 The majority are octapeptides that contain the

367

fundamental amino acid binding sequence Phe 7, Trp8, Lys9, and Thr10, which express high af finity for tumor associated subtype 2 (sst2) and subtype 5 (sst5) receptors and low affinity for more endo genous subtype 3 (sst 3) receptors. P eptides with af finity for multiple subtypes are under development. Sandoz de veloped the eight amino acid analo gue, octreotide, a modif ied analo gue of somotastatin-14, b y replacing the L-amino acids Phe and Trp with D-amino acids and incor porating an amino alcohol at the C ter minus to decrease enzymatic clea vage. Octreotide with its enhanced biological half-life (1.5-2 h) exhibited increased tumor uptake. The first imaging analogue of octreotide was developed b y iodination of the modif ied peptide sequence. An analo gue of octreotide w as synthesized that replaced the third amino acid Phe with a Tyr that w as then iodinated with 123I by electrophilic substitution of the iodine onto the aromatic ring of Tyr.21 The analogue 123 I-Tyr3-octreotide exhibited very high affinity for the sst2 receptor, however, it was too hydrophobic and suffered from dehalo genation, that is, loss of 123I, w hich led to signif icant uptake in the li ver and gastrointestinal tract that hampered lesion detection in these organs. The dehalo genation w as due to the sst receptor being internalized into the cell, with subsequent de gradation of the peptide in the lysosome. This feature is common for iodinated peptides and antibodies that are inter nalized, resulting in cellular efflux of radioactive iodine or iodotyrosine. An additional disadv antage w as the expense and limited availability of 123I. To introduce a longer li ved, residualizing label, the octreotide analogue was modif ied by adding the chelator DTPA to the amine ter minus for the stab le attachment of the radionuclide 111In. Indium-111-DTP A-octreotide is excreted mainly through the kidneys, albeit it more slowly than 123I-Tyr3-octreotide, w hich clears via the hepatobiliary system. 21 Indium-111-DTPA-octreotide is mark eted as OctreoScan ® for imaging and diagnosing somatostatin receptor-positive tumors. The impro vement in imaging contrast using 111In-DTPA-octreotide, as opposed to an 111 In-labeled antibody, is sho wn in F igure 1 (right side). Although the modif ications to somatostatin led to an increased biological half-life and increased tumor uptak e, it also resulted in a compromised tumor -targeting effect. Somatostatin has nanomolar af finity for all f ive receptor subtypes; the modifications lead to a decrease in the affinity for sst 3 and sst 5 and also a loss of affinity for sst 1 and sst4. Nanomolar af finity w as onl y retained for sst 2. The uptake of 111In-DTPA-octreotide has been shown to correlate directly with sst 2 expression. Low uptake is observed

368

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

in tumors e xpressing a high density of the other sst subtypes and only moderate expression of sst 2. 111 In-DTPA-octreotide, Based on the success of investigators began to identify methods to stab ly incorporate 99mTc to octreotide. NeoTect, the Food and Dr ug Administration (FD A)-approved [ 99mTcO]depreotide agent, was synthesized b y removing the disulf ide bond between the tw o Cys residues and for ming a bond between the Phe and Cys on the c yclic hexapeptide targeting vector. The loop was made more h ydrophobic by substituting a Tyr for a Val. This was linked to a linear tetrapeptide chelator motif, w hich for ms a coordination chelation complex with technetium at the Dap-L ys-Cys sequence (Dap = β-diaminopropionic acid). 22,23 The chelation around Tc is denoted as N 3S, as the coordinating g roups in the peptide include three nitro gens (tw o amide nitro gen atoms of L ys and Cys and one amine nitrogen of Dap) and one sulfur (the thiol sulfur of Cys22). The 99mTc-labeled agent e xhibited nanomolar binding affinity to sst and had higher tumor uptak e than 111 In-DTPA-octreotide and slo wer clearance from normal tissues via the kidne ys.24,25 [99mTcO]depreotide binds to a wider range of sst subtypes (2,3, and 5 26) and therefore may be useful in diagnosing a broader range of tumors. Ligands containing chiral Tc centers tend to form diastereomers, either syn or anti to the Tc = O

Figure 3.

group. Recent w ork has resulted in the separation and identification of depreotide diastereomers, sho wing the syn isomer to be predominant ( anti:syn 10:90), with the syn isomer having a slightly higher affinity (IC 50 = 0.15 nM) and higher tumor uptak e (6.58% ID/g) for the somatostatin receptor , compared to the anti isomer (IC50 = 0.89 nM) and (3.38% ID/g). The binding of both isomers is higher than the free peptide (IC 50 = 7.4 nM) and may be due to conformation changes that are a result of metal chelation or b y the selecti ve folding obser ved for the syn diastereomer molecule. 23 The DTPA chelate of octreotide w as replaced with DO TA to generate the peptide conjugate DO TATOC. The str ucture of DO TATOC is sho wn in F igure 3 (top str ucture). Labeling DOTATOC with dif ferent metals can ha ve a profound influence on tar get tissue uptak e and renal clearance. DOTATOC labeled with 67Ga sho wed higher tumor uptake and f aster kidne y clearance than 90Y-DOTATOC.27 This result was surprising in view of the similar chemistry of 67Ga and 90Y (both are +3 cations) but e vidently is due to differences in the coordination chemistry of the metals. 27 Owing to the higher sensitivity and spatial resolution of PET, compared to SPECT , positron emitters such as 68 Ga, 64Cu, and 18F have been attached to peptides and evaluated in humans. Gallium-68 can be made routinel y

Structures of DOTATOC (top) and DOTA-8Aoc-BBN(7–14)NH2 with an 8 carbon spacer (bottom).

Targeted Antibodies and Peptides

available from a 68Ge/68Ga generator. The chemistr y of gallium, a +3 cation, is v ery similar to that of In and many of the same chelates that stably bind In can be used +3 with Ga. A major dif ference is that the size of Ga (0.47 Å) ion is very close to that of Fe+3 (0.49 Å) and they share similar ionization potential, coordination number , and electronic conf iguration; therefore, transchelation is more of a concern with Ga. Gallium can be stabilized b y either four, f ive, or six coordinate ligands with six being the most stab le. The availability of 68Ga via a generator allows for PET imaging without the need of an onsite cyclotron. Gallium can be attached to peptides and other biomolecules using the DO TA chelator . Gallium-68 radiopharmaceuticals would be of particular value in producing imaging sur rogates of radiometallated (including 177 Lu, 90Y, etc) therapeutic agents for treatment of cancers. With the rapid spread of PET capabilities in nuclear medicine clinics, there will be an enor mous acceleration of demand for FDA-approved 68Ga labeled bioconjugates over the next decade. 111 A study comparing images obtained with InDTPA-octreotide to those obtained with 68Ga-DOTATOC, a DOTA derivatized somatostatin analo gue, showed that 68 Ga-DOTATOC resulted in higher tumor to non-tumor contrast, resulting in the detection of more tumor sites 111 (100%), especiall y small tumors, than with InDTPA-octreotide, in w hich onl y 85% w ere obser ved.28 Gallium-68-DOTATOC has sho wn g reat promise for imaging neuroendocrine tumors, par ticularly in staging and predicting patient dosimetr y and outcomes to therapeutic analo gues.28–30 An octreotide deri vative designed specifically for binding to Ga, NODAGATOC, was developed that contains the chelator NO TA.31 NODAGATOC was labeled in high yields and high specif ic activity with both In and Ga and exhibited high internalization and high affinity to both sst 2 and sst 5. The complex also e xhibited good biodistribution characteristics such as f ast clearance from the kidne ys and nor mal tissues. A concer n is that 111 In- or 68Ga-NODAGATOC may not accuratel y reflect the distribution of the DOTA derivatives that are the most commonly used in therapeutic applications. Ne w derivatives of somatostatin peptides are being de veloped with a broader affinity profile to somatostatin receptors to widen their ability to diagnose tumors that o ver-express more than just the sst 2 subtype. One successful deri vative DOTANOC (NOC = Nal 3-octreotide, a sst 2, sst 3, and sst 5 binding peptide) has sho wn potential in diagnosing a broader range of tumors when labeled with 68Ga.32 A v ariety of macroc yclic chelators ha ve been used to bind copper to somatostatin peptides, most notably DOTA, TETA, and CPT A. Imaging studies w ere

369

performed using 64Cu-TETA-D-Phe1-octreotide and 64 Cu-CPTA-D-Phe1-octreotide in patients with neuroendocrine tumors 33 and, w hen compared to 111In-DTPAD-Phe1- octreotide, they were shown to be more sensitive, especially for detection of small metastases. Ho wever, further studies in animals showed that the copper dissociated both from the TETA, and more so from the CPT A, resulting in higher tumor to non-tumor ratios than 111InDTPA-D-Phe1-octreotide. The cross-bridged macrocyclic ligand CB-TE2A w as conjugated to Tyr3-octreotate and shown to bind copper with high stability . The 64Cu-CBTE2A-D-Phe1-Tyr3-octreotate e xhibits nanomolar af finity for somatostatin receptors, as w ell as tumor uptak e four times g reater than that obser ved for 64Cu-TETATyr3-octreotate, but with slo wer kidne y clearance. 34 Further modifications and studies are underway to determine if these comple xes will prove useful in the clinical setting. Indium-111 is often used as an imaging sur rogate for 90 Y, which is a pure beta minus emitter . Studies indicate that 111In is handled differently by the body and can lead to an underestimation of the dose to critical organs.35 This has led to the development and evaluation of the positron emitter 86Y. The use of this radionuclide is limited b y its availability and difficulty in obtaining it in a highl y pure form. It has been evaluated as a surrogate for 90Y-DOTATOC and, when bound to DO TATOC, 86Y was shown to ha ve high stability in vi vo and to clear rapidl y from nor mal tissues via the kidne ys into the urine. A comparati ve study of 86 Y-DOTATOC to 111In-DTPA-octreotide showed the doses to critical or gans w ere similar for both agents; ho wever, dosimetry to the tumor w as underestimated b y 111InDTPA-octreotide.36 Yttrium-86-DOTATOC appears to give better dosimetry due to the identical biodistribution of the imaging radionuclide and chelator molecule used for the therapeutic radionuclide 90Y.36 Fluorine-18 can often be substituted for a h ydrogen or hydroxyl group, forming a C-F bond resistant to defluorination and resulting in little to no change in the peptide’ s affinity or in vivo behavior. Initial attempts to label D-Phe1octreotide with 18F were made by using N-succinimidyl-fluorobenzoate (SFB) and 18F-propionic acid. 37,38 Evaluation of these comple xes in rodent models sho wed poor tumor uptake and retention, as w ell as unf avorable biodistributions. The 18F-fluoropropionyl-octreotide complex was developed, and its beha vior was compared with that of 86Y-DTPA-octreotide and 67Ga-DFO-octreotide. The 18 F-fluoropropionyl deri vative sho wed impro ved tumor uptake; however, its tumor retention was too low, and it was hampered b y its e xcretion through the hepatobiliar y system.39 Previous w ork indicated the iodinated deri vative

370

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

could be made less h ydrophobic b y the conjugation of carbohydrates reducing the hepatobiliar y e xcretion.40 A similar analo gue w as prepared using 18F-FP-GlucTOCA, w hich e xhibited nanomolar af finity for the sst 2 receptor and lower hydrophobicity (log P = −1.7).41 Glycosylated 18F-Tyr3-octreotide sho wed impro ved biodistribution in animals; however, in a pilot patient study, it suffered from defluorination. In an attempt to pre vent defluorination, a fluorobenzilidene-o xime carboh ydrate analo gue, TOCA, was prepared, which was stable against defluorination, exhibited high tumor uptake, rapid internalization, and higher tumor to nor mal tissue ratios compared to the fluoropropionylated analogue.42 There is a signif icant interest in de veloping “matched pairs” for diagnostic and therapeutic applications, based on radiometals that ha ve similar radiolabeling chemistr y, such as 99mTc and 186/188Re. Although progress has been made, signif icant challenges remain. The majority of 99mTc/186/188Re pairs tend to be highl y lipophilic, resulting in predominant clearance through the hepatobiliary system and thus inhibiting the ability to image tumors in the abdominal re gion.43 A number of methods have been evaluated to increase the hydrophilicity of these conjugates to impro ve their biodistribution and increase their excretion through the urinary tract. An N3S chelator was attached to a bombesin derivative, BBN(7–14)NH2, via a series of linkers that varied from 0 to 11 carbon spacers. Conjugates containing the 3, 5, and 8 carbon spacers retained high af finity and specif icity for gastrin-releasing peptide (GRP) receptors, and with the 0 and 11 carbon spacers, the affinity was reduced.44 Recently tricarbonyl Tc(I) complexes with a variety of donor groups have been reported.45,46 The N-terminus of BBN(7–14)NH2 was functionalized with (Nα-histidinyl)acetic acid and radiolabeled with the [ 99mTc(H2O)3(CO)3]+ synthon, w hich is commercially available as a kit. This allows for stable radiolabeling of a v ariety of biomolecules with 99mTc. High affinity and specif icity were obtained for this conjugate in PC-3 tumors in a nude mouse model, ho wever low tumor localization was observed. Agonists have typically been used in the past as they undergo inter nalization and residualization of the attached radiometal within the tumor cell. A potent antagonist, [99mTc]demobesin-1, has shown high binding and af finity to the GRP receptor .47 High tumor uptak e was noted in a nude mouse model, prompting the de velopment of optimized deri vatives. These new derivatives showed nanomolar af finity, with high tumor uptak e and fast washout from normal tissues and excretion predominantly through the urinar y system. Images acquired with 99mTc-Demobesin-3 showed high tumor to

background ratios with only slight uptake in the kidneys. These studies indicate that inter nalization ma y not be necessary. Peptides containing 0 to 11 carbon spacers placed between DO TA and BBN(7–14)NH 2 were radiolabeled with 111In, and comparati ve studies w ere perfor med in nude mice bearing PC-3 tumors. 48 The results for GRP receptor af finity w ere similar to those obser ved for the 99m Tc analogues, in that no difference was noted for the 3, 5, and 8 spacers; however, affinity was greatly affected by the 0 and 11 carbon spacers. Comparing the results to those obtained for the 99mTc analogues, it w as seen that 99m Tc forms a more labile comple x than 111In. The structure of BBN(7–14)NH 2 with an 8 carbon spacer is sho wn in Figure 3 (bottom structure). 68 BBN(7–14)NH2 has been labeled with Ga b y attachment of the chelator DO TA via a PEGylated spacer, resulting in the PEGylated deri vative 68 Ga-DOTA-PESIN.49 PEG4 was chosen, as pre vious studies had indicated its length was optimal. PEGylated peptides have shown higher tumor uptake and improved biodistribution proper ties. The radiolabeled peptide exhibited high to moderate af finity for the GRP receptor and neuromedin B receptors, b ut no affinity for the bombesin-3 receptor (BB3). F ast clearance from b lood and receptor ne gative tissues w as observed, with clearance through the urinar y system. High tumor uptak e, similar to that obtained b y the best BBN peptides, w as observed at 1 hour and combined with the f ast washout resulted in excellent PET images. PET images of a nude mouse PC-3 tumor model sho wed uptak e onl y in the receptor-rich pancreas, tumor , and kidne ys. F ast washout w as initiall y obser ved for the tumor and w as believed to arise from metabolism of the radioligand, as it is not inter nalized.50 The high uptak e in tumor and superior tumor to nor mal tissue ratios indicate that this peptide derivative should be evaluated further. The initial analo gues of ar ginine-glycine-aspartic acid (RGD) peptides used for imaging αvβ3 were of iodinated derivatives. A sugar amino acid, 3-acetamido2,6-anhydro-4,5,7-tri-O-benzyl-3-deoxy-beta-D-glyceroD-gulo-heptonic acid , w as added to the c yclized peptides c[RDGDyK] and c[RDGDyV], to w hich a D-Tyr was added for iodination with 125I.51 The leucine was substituted with a sugar functionalized l ysine. The incorporation of the galacto g roup resulted in improved selectivity; ho wever, tumor uptak e in animal models was lo w. In addition, uptak e in the th yroid w as high, indicating deiodination of the peptide had occur red, leading to the incor poration of [ 125I]iodide into iodinerich thyroid hormones.

Targeted Antibodies and Peptides

A linear peptide containing the RGD sequence w as recently prepared by a novel rapid solid phase synthesis.52 Fluorine-18 w as attached using [ 18F]fluorobenzoic acid and was shown to have no effect on binding af finity. The complex w as unstab le and the tumor uptak e w as nonspecific. Linear RGD peptides tend to e xhibit lower than nanomolar affinity, low specificity and rapid degradation, which mak e them undesirab le as tar geting v ectors.53 Cyclization of RGD peptides leads to increased receptor affinity and selectivity and, in certain cases, higher receptor binding can also be achie ved. In 2003, a c yclic RGD peptide was labeled with 18F by direct electrophilic fluorination with [ 18F]acetyl h ypofluorite.54 It e xhibited high affinity similar to the nati ve peptide and specif ic tumor uptake. However, this analo gue had lo w specif ic activity due to the need of adding car rier during the preparation. Furthermore, the peptide w as de graded and e xhibited biliary excretion. As previously demonstrated for somatostatin receptor targeting, the introduction of glycosylation to the RGD peptide w as used to impro ve the o verall in vivo distribution. [ 18F]fluoropropionate was used to label a galacto-RGD conjugate. The gl ycosylation introduced more h ydrophilicity to the comple x, resulting in better tumor to nor mal tissue ratios and impro ved biodistribution. Replacement of the Tyr with Phe resulted in f ar less degradation, as demonstrated by catabolism studies showing 75 to 87% of the e xtracted material to be the intact peptide. The [18F]galacto-RGD w as used for imaging melanoma and osteosarcoma tumors in mouse models and is currently in clinical trials for tumor imaging. 55,56 99m The HYNIC chelator has been used to bind Tc rapidly in high yields and lo w concentrations and it allows for easy modification of biodistribution properties by the use of dif ferent co-ligands. 57 The co-ligand has been shown to have a signif icant impact on the stability , tumor uptake, and excretion kinetics of the RGD peptide. Both HYNIC and DO TA ha ve been used to attach radiometals to RGD peptides. Comparisons of HYNIC and DOTA labeled with 99mTc and other radiometals have shown no significant differences in binding affinity to the vascular αvβ3 integrin targeted by RGD peptides. Ho wever, the radiochemical composition of the conjugates did impact the uptake in normal organs, particularly the liver and kidneys.58 Multivalent c yclic RGD peptides ha ve been prepared and result in improved affinity to the αvβ3 integrin. The multivalent concept has been used pre viously with recombinant antibody fragments, w hich resulted in affinities 3 to 5 times that of the monomers.59,60 Comparative studies of radiolabeled RGD dimers resulted in higher tumor uptak e and longer residence times than

371

their corresponding monomeric forms.53 RGD tetramers have also been prepared and demonstrated that the increase in peptide multiplicity results in a signif icant enhancement in αvβ3 integrin binding af finity.61 The length of the multimer linker should be maintained from 25 to 30 bond PEG lengths to allo w for simultaneous binding.62 The cyclic RGD tetramers ha ve exhibited the highest tumor uptake and retention, as well as the fastest blood clearance.

ANTIBODY PRE-TARGETING Radiolabeled mAbs ha ve sho wn considerab le promise for radioimmunotherapy of blood-borne cancers, particularly non-Hodgkin’s lymphoma.63 However, because of their slow blood clearance, radiolabeled mAbs generally do not localize to solid tumors in sufficient quantities to provide consistent therapeutic ef ficacy without signif icant bone mar row to xicity. Fur thermore, these same properties generally cause radiolabeled mAbs to exhibit relatively poor imaging contrast. Future impro vements in diagnostic imaging and targeted radiotherapy of cancer will likely require implementation of se veral innovative approaches, including the use of dif ferent radionuclides, new targeting molecules, and novel delivery systems. F or e xample, radiolabeled peptides and small molecules offer the prospect of rapid and specif ic tumor targeting, with extremely rapid whole-body elimination of radioacti vity, of fering the potential for considerable impro vements in imaging contrast o ver antibodies. On the other hand, rapid blood clearance and excretion of small radiolabeled compounds ma y preclude high tumor uptak e and compromise imaging and therapeutic efficacy. Antibody pre-tar geting is an approach in w hich an unlabeled antibody conjugate or fusion protein, containing a non-natural receptor , is f irst administered and allowed to accumulate in tumors and then radionuclide imaging or therapy is achieved, given via a small effector molecule that binds rapidl y with high af finity to the mAb-receptor constr uct at the tumor site, “decorating” the non-natural receptors on the tumor cells with the molecular imaging or therap y agent. When successful, this process results in immediate accumulation of radioacti vity in the tumor , causing substantial impro vements in tumor-to-blood ratio, imaging contrast, normal organ toxicity, tumor absorbed dose, and therapeutic ef ficacy. Thus, pre-targeting combines the desirab le properties of high tumor uptake of antibodies with rapid in vivo distribution of radiolabeled small molecules, with e xtremely fast w hole-body clearance of radioacti vity. Antibody

372

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

pretargeting has recentl y been re viewed b y Goldenber g and colleagues.64 Several types of “receptor/ef fector” approaches ha ve been de veloped for antibody pre-tar geting, including mAb/hapten,65–71 biotin/avidin or strepta vidin,72–85 and oligonucleotide/antisense oligonucleotide analo gue86–89 systems. The high affinity noncovalent binding of biotin to avidin (~10 15 M−1) or streptavidin (~10 13 M−1) makes this system attracti ve for mAb pre-tar geting methods. Both mAb-biotin and mAb-avidin conjugates have been investigated for pre-tar geting of radiolabeled a vidin and biotin, respectively. However, radiolabeled avidin and streptavidin have shown considerable kidney uptake, as w ell as crosslinking of circulating biotin ylated antibody, which results in high li ver uptake and interferes with tumor tar geting.73 Most attempts to clear biotin ylated mAb with cold a vidin have produced onl y modest impro vements in tumor -toblood ratios. 90,91 In addition, dif fusion of radiolabeled streptavidin into tumors pretreated with mAb-biotin conjugates occurs slo wly over a period of 24 hours, 75 demonstrating that strepta vidin can encounter signif icant retardation in its tumor penetration. 92 Pharmacokinetic modeling studies of directl y labeled mAb, labeled biotin after pre-tar geting, and labeled strepta vidin after pre-tar geting ha ve been performed.93 These studies indicated that, in the absence of antigen inter nalization, the protocol involving radiolabeled biotin and mAb-strepta vidin conjugate yields the highest tumor -to-blood ratio and mean residence time of radioacti vity in the tumor . Thus, three k ey features of a desired radiolabeled ef fector molecule emerged from these f indings: (1) it must be small, hydrophilic, and rapidl y diffusible, (2) it must under go rapid renal elimination, and (3) it must ha ve minimal uptake in normal tissues. 94 The first mAb used for clinical biotin/streptavidin pretargeting, NR-LU-10, reacts with a 40-kDa gl ycoprotein antigen, Ep-CAM, found on human adenocarcinomas such as ovarian, breast, prostate, colon, and small cell lung cancer. Tumor targeting of 99mTc- and 186Re-labeled NR-LU-10 has been demonstrated in tumor-bearing animal models 95 and in patients. 96–98 A strepta vidin chemical conjugate of NR-LU-10 (SA-NR-LU-10) w as prepared and e valuated 81 for pre-tar geted RIT in mouse tumor models and patients.82 SA-NR-LU-10 e xhibited tumor uptak e and blood clearance equi valent to unmodif ied intact mAb . After 20 to 24 hours, treatment with a clearing agent removed 90 to 95% of circulating SA-NR-LU-10. The effector molecule, 111In-DOTA-biotin, gi ven 6 to 8 hours after the clearing agent, sho wed rapid b lood clearance and low normal organ uptake, with urinary excretion of > 80%

of the injected dose in 2 hours. Injected doses and dose schedules of these agents w ere optimized for mice 81 and humans82 to provide maximum tumor uptake and clearance of SA-NR-LU-10 without compromising tumor targeting of radiolabeled DOTA-biotin. The tumor tar geting and therapeutic proper ties of SA-NR-LU-10 and radiometal-labeled DO TA-biotin show promise in animal models, but potential prob lems exist that may ultimately limit the utility of this mAb for pre-targeted radioimmunotherap y. During the course of these studies, it w as found that NR-LU-10 cross-reacts with normal intestinal tissue in humans, and dose escalation of SA-NR-LU-10-pre-tar geted 90Y-DOTA-biotin in Phase I clinical trials was limited by gastrointestinal toxicity.82 Furthermore, streptavidin is a highly immunogenic bacterial protein, and neutralizing anti-strepta vidin responses can be elicited in nearly 100% of immunocompetent patients with non-hematologic malignancies. Pre-targeting strategies other than biotin/streptavidin suffer from dra wbacks as w ell. In bispecif ic antibodyhapten systems, one antigen-combining site binds a metal chelate and the other binds a tumor -associated antigen. Such systems can sho w lo w-affinity binding of the chelate, usually necessitating the use of bivalent haptens. However, e ven in the case of bi valent haptens, it is difficult to engineer bispecif ic antibodies to ha ve high affinity for both the tumor and the chelate, as neither has the advantage of binding avidity to enhance affinity. Perhaps the use of tetra valent single-chain antibodies, with two tumor and two chelate binding sites, could overcome this difficulty, but to the authors’knowledge, no construct of this type has been e valuated in vivo. Oligonucleotide analogue pre-targeting systems use a “sense” oligonucleotide co valently attached to the antibody and a complementary “antisense” sequence car rying the imaging label. These systems typicall y displa y high binding af finity of complementar y sequences. Ho wever, unlike biotin/strepta vidin binding, oligonucleotide hybridization is a relati vely slow process, it has required extensive optimization of such systems to yield even moderate tumor uptake in mouse xenograft models, and patient studies still remain a future goal. Perhaps the most e xciting and promising recent advance in antibody pre-tar geting is the de velopment of “antibodies with inf inite af finity” b y Meares and colleagues.99,100 In this inno vative variation on mAb/hapten systems, these authors evaluated a series of 111In chelates modified with alkylating moieties for in vivo stability, as indicated by slow reaction with serum albumin and rapid whole-body clearance. It w as found that an all ylamido derivative of benzyl EDTA had optimal in vi vo stability.

Targeted Antibodies and Peptides

Meares and colleagues also engineered IgG and F ab mAbs against these chelates, such that a c ysteine residue was substituted for serine in the light chain v ariable region of the chelate-binding site. Following noncovalent binding of the 111In chelate hapten, alkylation of the substituted c ysteine b y the all ylamido benzyl EDT A produced a covalent linkage between hapten and mAb, thus affording “infinite” affinity. Extension of this concept to bispecific antibodies may overcome many problems associated with con ventional mAb/hapten, biotin/a vidin or streptavidin, and oligonucleotide analogue strategies, creating a w ay for pre-tar geting to mo ve forw ard successfully into the future.

FUTURE DIRECTIONS One might get the impression that nuclear imaging (gamma scintig raphy, SPECT, PET) using tar geted antibodies and peptides is a relatively mature field at present. However, breakthroughs in antibody engineering, phage display technology, and ne w pre-targeting strategies will likely ensure that inno vations in gamma scintig raphy, SPECT, and PET with mAbs and peptides will continue to be made w ell into the future. In addi tion, ne w nonnuclear imaging modalities are currently being developed that rely on mAbs and peptides for tar geted delivery. A nascent, but g rowing, area of research is tar geted magnetic resonance imaging (MRI), where cell-penetrating peptides101 and internalizing antibodies 102 are being conjugated to MRI contrast agents, designed for intracellular delivery to diseased tissue.Also, a great deal of recent effort has been aimed at the use of nanopar ticles in molecular imaging and therap y; again antibody- 103 and peptideconjugated104 nanoparticles are being developed for targeting these agents to catastrophic disease states. Furthermore, optical technologies are rapidly emerging not only as viable techniques for molecular imaging, b ut also for highthroughput screening of lead compounds, tar geted mAbs, and peptides. 101,105 Conjugation of these molecules with near-infrared fluorescent dyes are playing a leading role in this area of research. Finally, one of the newest areas of antibody targeting applied to molecular imaging is the use of quantum dots,106,107 which have the advantage of extremely high fluorescence quantum yields for optical imaging and hold great promise for such applications. Not only are these new molecular imaging modalities e xciting in their potential to bring about another “renaissance” in the use of targeted antibodies and peptides, but the y also hold the promise that Ehrlich’s “magic bullet” concept ma y remain important in the use of such delivery platforms for molecular imaging in the future.

373

ACKNOWLEDGMENT The authors would like to thank the Depar tment of Veterans Affairs, for the use of f acilities and resources at the Harry S Truman Memorial Veterans’ Hospital in Columbia, MO.

REFERENCES 1. Schubiger PA, Alberto R, Smith A. Vehicles, chelators, and radionuclides: choosing the “b uilding blocks” of an ef fective therapeutic radioimmunoconjugate. Bioconjug Chem 1996;7:165–79. 2. Langmuir VK. Radioimmunotherapy: clinical results and dosimetric calculations. Nucl Med Biol 1992;19:213–25. 3. Griffiths GL, Goldenberg DM, Knapp FF, et al. Direct radiolabeling of monoclonal antibodies with generator -produced rhenium-188 for radioimmunotherap y: labeling and animal biodistribution studies. Cancer Res 1991;51:4594–602. 4. Hnatowich DJ, Layne WW, Childs RL, et al. Radioacti ve labeling of antibody: a simple and efficient method. Science 1983;220:613–5. 5. Paxton RJ , Jak owatz JG, Beatty JD , et al. High-specif ic-activity 111 In-labeled anticarcinoembr yonic antigen monoclonal antibody: improved method for the synthesis of diethylenetriaminepentaacetic acid conjugates. Cancer Res 1985;45:5694–9. 6. Brechbiel MW , Ganso w O A, Atcher R W, et al. Synthesis of 1-(p-isothiocyanatobenzyl) derivatives of DTPA and EDTA. Antibody labeling and tumor -imaging studies. Inor g Chem 1986;25:2772–81. 7. Moi MK, Meares CF , DeNardo SJ . The peptide w ay to macroc yclic bifunctional chelating agents: synthesis of 2-( p-nitrobenzyl)1,4,7,10-tetraazacyclododecane-N,Nʹ′,Nʹ′ʹ′,Nʹ′ʹ′ʹ′-tetraacetic acid and study of its yttrium(III) comple x. J Am Chem Soc 1988;110:6266–7. 8. Sumerdon GA, Rogers PE, Lombardo CM, et al. An optimized antibody-chelator conjugate for imaging of carcinoembryonic antigen with indium-111. Int J Radiat Appl Instrum Part B Nucl Med Biol 1990;17:247–54. 9. Lewis MR, Raubitschek A, Shi vely JE. A f acile, w ater-soluble method for modif ication of proteins with DO TA. Use of elevated temperature and optimized pH to achie ve high specif ic activity and high chelate stability in radiolabeled immunoconjugates. Bioconjug Chem 1994;5:565–76. 10. Lewis MR, Kao JY , Anderson A-LJ, et al. An improved method for conjugating monoclonal antibodies with N-hydroxysulfosuccinimidyl DOTA. Bioconjug Chem 2001;12:320–4. 11. Anderson CJ , Connett JM, Schw arz SW, et al. Copper -64-labeled antibodies for PET imaging. J Nucl Med 1992;33:1685–91. 12. Moi MK, Meares CF, McCall MJ, et al. Copper chelates as probes of biological systems: stab le copper comple xes with a macroc yclic bifunctional chelating agent. Anal Biochem 1985;148:249–53. 13. Jones PT, Dear PH, F oote J , et al. Replacing the complementarity determining regions in a human antibody with those of a mouse. Nature 1986;321:522–5. 14. Colcher D , Bird R, Roselli M, et al. In vi vo tumor tar geting of a recombinant single-chain antigen-binding protein. J Natl Cancer Inst 1990;28:1191–7. 15. Lawrence LJ, Kortt AA, Iliades P, et al. Orientation of antigen binding sites in dimeric and trimeric single chain Fv antibody fragments. FEBS Lett 1998;425:479–84. 16. Le Gall F , Kipriyanov SM, Moldenhauer G, Little M. Di-, tri- and tetrameric single chain Fv antibody fragments against human CD19: ef fect of v alency on cell binding. FEBS Lett 1999;453:164–8.

374

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

17. Wu AM, Yazaki PJ, Tsai S-w, et al. High-resolution microPET imaging of carcinoembr yonic antigen-positi ve x enografts b y using a copper-64-labeled engineered antibody fragment. Proc Natl Acad Sci U S A 2000;97:8495–500. 18. Wong JYC, Chu DZ, Williams LE, et al. Pilot trial e valuating an 123 I-labeled 80-kilodalton engineered anticarcinoembr yonic antigen antibody fragment (cT84.66 minibody) in patients with colorectal cancer. Clin Cancer Res 2004;10:5014–21. 19. Williams LE, Wu AM, Yazaki PJ, et al. Numerical selection of optimal tumor imaging agents with application to engineered antibodies. Cancer Biother Radiopharm 2001;16:25–35. 20. Bauer W, Briner U, Doepfner W, et al. SMS 201–995: a v ery potent and selecti ve octapeptide analo gue of somatostatin with prolonged action. Life Sci 1982;31:1133–40. 21. Krenning E, Kwekkeboom DJ, Bakker WH, et al. Somatostatin recep111 tor scintig raphy with [ In-DTPA-D-Phe1] and [123I-Tyr3]-octreotide: the Rotterdam e xperience with more than 1000 patients. Eur J Nucl Med 1993;20:716–31. 2 2. Cantorias M, Howell RW, Todaro L, et al. MO tripeptide diastereomers (M = 99/99mTc, Re): models to identify the str ucture of 99mTc peptide targeted radiopharmaceuticals. Inorg Chem 2007;46:7326–40. 23. Cyr J, Pearson D, Nelson C, et al. Isolation, characterization, and biological evaluation of syn and anti diasteromers of [ 99mTc]Technetium depreotide: a somatostatin receptor binding tumor imaging agent. J Med Chem 2007;50:4295–303. 24. Pearson D, Lister-James J, Mcbride W, et al. Somatostatin receptor binding peptides labeled with technetium-99m: chemistr y and initial biological studies. J Med Chem 1996;39:1361–71. 25. Vallabhahajosula S, Moyer B, Lister-James J, et al. Preclinical e valuation of technetium-99m-labeled somatostatin receptor -binding peptides. J Nucl Med 1996;37:1016–22. 26. Virgolini I, Leimer M, Handmak er H, et al. Somatostatin receptor subtype specif icity and in vi vo binding of a no vel tumor tracer , 99m Tc-P829. Cancer Res 1998;58:1850–9. 27. Heppeler A, F roidevaux S, Maeck e HR, et al. Radiometal-labelled macrocyclic chelator -derivatised somatostatin analo gue with superb tumour -targeting proper ties and potential for receptor mediated internal radiotherapy. Chem Eur J 1999;5:1974–81. 28. Hofmann M, Maecke H, Borner AR, et al. Biokinetics and imaging with the somatostatin receptor PET radioligand 68Ga-DOTATOC: preliminary data. Eur J Nucl Med 2001;28:1751–7. 29. Kowalski J, Henze M, Schuhmacher J, et al. Evaluation of positron emis68 sion tomo graphy imaging using [ Ga]-DOTA-D-Phe1-Tyr3octreotide in comparison to [111In]-DTPAOC SPECT. First results in patients with neuroendocrin tumors. Mol Imaging Biol 2003;5:42–8. 30. Henze M, Dimitrakopou-Strauss A, Milker-Zabul S, et al. Characterization of [ 68Ga]DOTA-D-Phe1Tyr3-octreotide kinetics in patients with meningiomas. J Nucl Med 2005;46:763–9. 31. Eisenwiener KP, Prata MIM, Buschmann I, et al. NODAGATOC, a new chelator-coupled somatostatin analo gue labeled with [ 67/68Ga] and [111In] for SPECT , PET, and tar geted therapeutic applications of somatostatin receptor (hsst2) e xpressing tumors. Bioconjug Chem 2002;13:530–41. 32. Antunes P, Ginj M, Walter M, et al. Influence of dif ferent spacers on the biological profile of a DOTA-somatostatin analogue. Bioconjug Chem 2007;18:84–92. 33. Anderson CJ, Pajeau TS, Edwards WB, et al. In vitro and in vivo evaluation of copper -64-octreotide conjugates. J Nucl Med 1995;36:2315–25. 34. Anderson CJ, Dehdashti F, Cutler PD, et al. 64Cu-TETA-octreotide as a PET imaging agent for patients with neuroendocrine tumors. J Nucl Med 2001;42:213–21. 35. Smith-Jones PM, Omer SS, Navarro V, et al. Comparative biodistributions of 111In-DOTA-J591 and 177Lu-DOTA-J591: ramif ications for 90Y dosimetry. J Nucl Med 2001;42:241. 36. Forster G, Engelbach M, Brockmann J , et al. Preliminar y data on biodistribution and dosimetry for therapy planning of somatostatin

receptor positi ve tumors: comparison of 86Y-DOTATOC and 111 In-DTPA-octreotide. Eur J Nucl Med 2001;28:1743–50. 37. Vaidyanathan G, Zalutsk y M. Labeling proteins with fluorine-18 using N-succinimidyl 4-[ 18F]fluorobenzoate. Nucl Med Biol 1994;19:275–81. 38. Guhlke S, Wester H, Bruns C, Stocklin G. (2-[ 18F]fluoropropionyl(D)Phe1)-octreotide, a potential radiophar maceutical for quantitative somatostatin receptor imaging with PET : synthesis, radiolabeling, in vitro v alidation and biodistribution in mice. Nucl Med Biol 1994;21:818–25. 39. Wester H, Brockman J, Rosch F, et al. PET-pharmacokinetics of 18Foctreotide: a comparison with 67Ga-DFO-octreotide and 86YDTPA-octreotide. Nucl Med Biol 1997;24:275–86. 40. Schottelius M, Wester H, Reubi JC, et al. Improvement of pharmacokinetics of radioiodinated Tyr3-octreotide by conjugation with carbohydrates. Bioconjug Chem 2002;13:1021–30. 41. Wester H, Schottelius M, Scheidhauer K, et al. P et imaging of somatostatin receptors: design, synthesis, and preclinical e valuation of a novel 18F-labelled, carbohydrated analogue of octreotide. Eur J Nucl Med 2003;30:117–22. 42. Schottelius M, Poethko T, Herz M, et al. F irst 18F-labeled tracer suitable for routine clinical imaging of sst receptor-expressing tumors using positron emission tomo graphy. Clin Cancer Res 2004;10:3593–606. 43. Decristoforo C, Mather S. The influence of chelator on the pharmaco99m kinetics of Tc-labeled peptides. Q J Nucl Med 2002;46:195–205. 44. Smith C, Gali H, Sieckman GL, et al. Radiochemical in vestigations of 99m Tc-N3S-X-BBN(7-14)NH2: an in vitro/in vi vo str ucture-activity relationship study w here x = 0,3,5,8 and 11-carbon tethering moieties. Bioconjug Chem 2003;14:93–102. 45. Schibli R, Katti K, Volkert WA, Barnes C. Novel coordiantion behavior of f ac-[ReBr3(CO)3]2− with 1,3,5-triaza-7-phosphaada-mantane (PTA). Systematic investigation on stepwise replacement of the halides by PTA ligand. Phase transfer studies and X-ray crystal str ucture of [Net 4][ReBr2(PTA)(CO)3], [ReBr(PT A)2(CO)3], and [Re(PTA)3(CO)3]PF6. Inorg Chem 1998;37:5306–12. 46. Alberto R, Egli A, Schibli R, et al. F rom [TcO 4−] to an organometallic aqua-ion: synthesis and chemistr y of [ 99mTc(OH2)3(CO)3]+. Padova, Italy: SGE; 1999. 47. Nock B, Nikolopou A, Chiotellis E, et al. [ 99mTc] Demobesin 1, a novel potent bombesin analo gue for GRP receptor -targeted tumour imaging. Eur J Nucl Med 2003;30:247–58. 48. Hoffman TJ, Gali H, Smith CJ, et al. Novel series of Indium-111 labeled bombesin analo gs as potential radiophar maceuticals for specif ictargeting of gastrin releasing peptide receptors expressed on human prostate cancer cells. J Nucl Med 2003;44:823–31. 49. Schumacher J, Zhang H, Doll J, et al. GRP-receptor targeted PET of a 68 rat pancreas carcinoma x enograft in nude mice with a Galabeled bombesin(6–14) analog. J Nucl Med 2005;46:691–9. 50. Zhang H, Schumacher J, Waser B, et al. DOTA-PESIN, a DOTA-conjugated bombesin deri vative designed for the imaging and tar geted radionuclide treatment of bombesin receptor-positive tumours. Eur J Nucl Med 2007;34:1198–208. 51. Haubner R, Wester H, Burkhar t F , et al. Gl ycosylated RGD-containing peptides: tracer for tumor tar geting and angiogenesis imaging with impro ved biokinetics. J Nucl Med 2001;42:326–36. 52. Sutcliffe-Goulden J, O’Dougher ty M, Marsden P, et al. Rapid solidphase synthesis and biodistribution of 18F-labeled linear peptides. Eur J Nucl Med 2002;29:754–9. 53. Liu S, Robinson S, Edw ards D. Radiolabeled αvβ3 integrins antagonists as radiophr maceuticals for tumor radiotherap y. Top Cur r Chem 2005;252:117–53. 54. Ogawa M, Hatano K, Oishi S, et al. Direct electrophilic fluorination of a c yclic RGD peptide for in vi vo αvβ3 integrin related tumor imaging. Nucl Med Biol 2003;30:1–9.

Targeted Antibodies and Peptides

55. Zijlstra A, Guna wan J, Burcher t W. Synthesis and e valuation of a 18 F-labelled recombinant anne xin-V deri vative, for identif ication and quantification of apoptotic cells with PET. Appl Radiat Isot 2003;58:201–7. 56. Yagle K, Ear y JF, Tait J, et al. Ev aluation of 18F-annexin V as a PET imaging agent in an animal model of apoptosis. J Nucl Med 2003;46:658–66. 57. Liu S. Hynic deri vatives as bifunctional coupling agents for 99mTclabeling of small biomolecules. Top Curr Chem 2005;252:193–216. 58. Liu S, Robinson S, Edw ards DS. Integrin αvβ3 directed radiopharmaceuticals for tumor imaging. Dr ugs Future 2003;28:551–64. 59. Goel A, Barano wska-Kortylewicz J , Hinrichs S, et al. 99m Tc-labeled di valent and tetra valent CC49 single-chain Fv’ s: novel imaging agents for rapid in vi vo localization of human colon carcinoma. J Nucl Med 2001;42:1519–27. 60. Viti F, Tarli L, Gio vannoni L, et al. Increased binding af finity and valence of recombinant antibody fragments lead to impro ved targeting of tumoral angiogenesis. Cancer Res 1999;59:347–52. 61. Poethko T, Thumshirn G, Hersel U , et al. Impro ved tumor uptak e, tumor retention and tumor/backg round ratios of pe gylated RGD multimers. J Nucl Med 2003;44:46P. 62. Liu S. Radiolabeled multimeric c yclic RGD peptides ad inte grin αvβ3 targeted radiotracers for tumor tar geting. Mol Phar m 2006;3:472–87. 63. Eary JF . Radioimmunotherap y of B-cell l ymphoma. Ann Oncol 1991;2 Suppl 2:187–90. 64. Goldenberg DM, Shark ey RM, P aganelli G, et al. Antibody pre-targeting advances radioimmunodetection and radioimmunotherapy. J Clin Oncol 2006;24:823–34. 65. Goodwin D A, Meares CF , McCall MJ , et al. Pre-tar geted immunoscintigraphy of murine tumors with Indium-111-labeled bifunctional haptens. J Nucl Med 1988;29:226–34. 66. Le Doussal J-M, Chetanneau A, Gr uaz-Guyon A, et al. Bispecif ic monoclonal antibody-mediated tar geting of an Indium-111labeled DTPA dimer to primar y colorectal tumors: phar macokinetics, biodistribution, scintigraphy and immune response. J Nucl Med 1993;34:1662–71. 67. Goodwin DA, Meares CF, Watanabe N, et al. Phar macokinetics of pretargeted monoclonal antibody 2D12.5 and 88Y-Janus-2-(p-nitrobenzyl)-1,4,7,10-tetraazacyclododecanetetraacetic acid (DO TA) in BALB/c mice with KHJJ mouse adenocarcinoma: a model for 90Y radioimmunotherapy. Cancer Res 1994;54:5937–46. 68. Kranenborg MHGC, Boer man OC, Grif fiths GL, et al. Two-step tar geting of RCC tumors in mice: impro ved tumor uptake and retention with a bi valent chelate. J Nucl Med 1998;39:78P. 69. Barbet J , P eltier P , Bardet S, et al. Radioimmunodetection of medullary th yroid carcinoma using indium-111 bi valent hapten and anti-CEA × anti-DTPA-indium bispecif ic antibody. J Nucl Med 1998;39:1172–8. 70. Klivényi G, Schuhmacher J, Patzelt E, et al. Gallium-68 chelate imaging of human colon carcinoma xenografts pre-targeted with bispecific anti-CD44 V6/anti-gallium chelate antibodies. J Nucl Med 1998;39:1769–76. 71. Gautherot E, Le Doussal J-M, Bouhou J , et al. Deli very of therapeutic doses of radioiodine using bispecif ic antibody-targeted bivalent haptens. J Nucl Med 1998;39:1937–43. 72. Kalofonos HP, Rusckowski M, Siebecker DA, et al. Imaging of tumor in patients with indium-111-labeled biotin and streptavidin-conjugated antibodies: preliminar y communication. J Nucl Med 1990;31:1791–6. 73. Paganelli G, Magnani P, Zito F, et al. Three-step monoclonal antibody tumor tar geting in carcinoembr yonic antigen-positi ve patients. Cancer Res 1991;51:5960–6. 74. Dosio F, Magnani P, Paganelli G, et al. Three-step tumor pre-targeting in lung cancer immunoscintig raphy. J Nucl Biol Med 1993;37:228–32.

375

75. Saga T, Weinstein JN , Jeong JM, et al. Two-step tar geting of experimental lung metastases with biotin ylated antibody and radiolabeled streptavidin. Cancer Res 1994;54:2160–5. 76. Axworthy DB, Beaumier PL, Bottino BJ , et al. Preclinical optimization of pre-tar geted radioimmunotherap y components: high ef ficiency, curati ve 90Y deli very to mouse tumor x enografts. Tumor Target 1996;2:156–7. 77. Paganelli G, Orecchia R, Jereczek-Fossa B, et al. Combined treatment of advanced oropharyngeal cancer with external radiotherapy and three-step radioimmunotherapy. Eur J Nucl Med 1998;25:1336–9. 78. Foulon CF, Alston KL, Zalutsk y MR. Astatine-211-labeled biotin conjugates resistant to biotinidase for use in pre-tar geted radioimmunotherapy. Nucl Med Biol 1998;25:81–8. 79. Cremonesi M, F errari M, Chinol M, et al. Three-step radioimmunotherapy with yttrium-90 biotin: dosimetr y and phar macokinetics in cancer patients. Eur J Nucl Med 1999;26:110–20. 80. Foulon CF, Bigner DD, Zalutsky MR. Preparation and characterization of anti-tenascin monoclonal antibody-strepta vidin conjugates for pre-targeting applications. Bioconjug Chem 1999;10:867–76. 81. Axworthy DB, Reno JM, Hylarides MD , et al. Cure of human carcinoma xenografts by a single dose of pre-tar geted yttrium-90 with negligible toxicity. Proc Natl Acad Sci U S A 2000;97:1802–7. 82. Knox SJ, Goris ML, Tempero M, et al. Phase II trial of yttrium-90DOTA-biotin pre-targeted by NR-LU-10 antibody/streptavidin in patients with metastatic colon cancer . Clin Cancer Res 2000;6:406–14. 83. Weiden PL, Breitz HB, Press O, et al. Pre-targeted radioimmunotherapy (PRIT ™) for treatment of non-hodgkin’ s l ymphoma (NHL): initial phase I/II study results. Cancer Biother Radiophar m 2000;15:15–29. 84. Lewis MR, Wang M, Axworthy DB, et al. In vi vo evaluation of pretargeted 64Cu for tumor imaging and therap y. J Nucl Med 2003;44:1284–92. 85. Lewis MR, Zhang J , Jia F , et al. Biolo gical comparison of 149Pm-, 166 Ho-, and 177Lu-DOTA-biotin pre-targeted by CC49 scFv-streptavidin fusion protein in x enograft-bearing nude mice. Nucl Med Biol 2004;31:213–23. 86. Kuijpers WHA, Bos ES, Kaspersen FM, et al. Specif ic recognition of antibody-oligonucleotide conjugates b y radiolabeled antisense nucleotides: a no vel approach for tw o-step radioimmunotherapy of cancer. Bioconjug Chem 1993;4:94–102. 87. Bos ES, K uijpers WHA, Meesters-Winters M, et al. In vitro e valuation of DN A-DNA h ybridization as a tw o-step approach in radioimmunotherapy of cancer. Cancer Res 1994;54:3479–86. 88. Wang Y, Chang F , Zhang Y, et al. Pre-tar geting with amplif ication using pol ymeric peptide nucleic acid. Bioconjug Chem 2001;12:807–16. 89. Liu G, Mang’era K, Liu N , et al. Tumor pre-targeting in mice using 99m Tc-labeled mor pholino, a DN A analo g. J Nucl Med 2002;43:384–91. 90. Yao Z, Zhang M, K obayashi H, et al. Impro ved targeting of radiolabeled streptavidin in tumors pre-targeted with biotinylated monoclonal antibodies through an a vidin chase. J Nucl Med 1995;36:837–41. 91. Zhang M, Sakahara H, Yao Z, et al. Intra venous a vidin chase improved localization of radiolabeled strepta vidin in intraperitoneal x enograft pre-tar geted with biotin ylated antibody . Nucl Med Biol 1997;24:61–4. 92. Sung C, v an Osdol WW, Saga T, et al. Strepta vidin distribution in metastatic tumors pre-tar geted with a biotin ylated monoclonal antibody: theoretical and e xperimental pharmacokinetics. Cancer Res 1994;54:2166–75. 93. Sung C, van Osdol WW. Pharmacokinetic comparison of direct antibody targeting with pre-targeting protocols based on streptavidinbiotin binding. J Nucl Med 1995;36:867–76. 94. Goodwin DA. Tumor pre-targeting: almost the bottom line. J Nucl Med 1995;36:876–9.

376

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

95. Goldrosen MH, Biddle WC, Pancook J, et al. Biodistribution, pharmacokinetic, and imaging studies with 186Re-labeled NR-LU-10 whole antibody in LS174T colonic tumor -bearing mice. Cancer Res 1990;50:7973–8. 96. Breitz HB, Weiden PL, Vanderheyden J-L, et al. Clinical e xperience with rhenium-186-labeled monoclonal antibodies for radioimmunotherapy: results of phase I trials. J Nucl Med 1992;33:1099–109. 97. Breitz HB , Durham JS, F isher DR, Weiden PL. Radiation-absorbed dose estimates to nor mal organs following intraperitoneal Re-186labeled monoclonal antibody:meth-ods and results. Cancer Res (Suppl) 1995;55:5817s–22s. 98. Breitz HB, Tyler A, Bjorn MJ, et al. Clinical experience with Tc-99m nofetumomab mer pentan (v erluma) radioimmunoscintig raphy. Clin Nucl Med 1997;22:615–20. 99. Chmura AJ, Orton MS, Meares CF. Antibodies with infinite affinity. Proc Natl Acad Sci U S A 2001;98:8480–4. 100. Chmura AJ, Schmidt BD , Corson DT , et al. Electrophilic chelating agents for irreversible binding of metal chelates to engineered antibodies. J Control Release 2002;78:249–58. 101. Bullok KE, Gammon ST, Violini S, et al. Permeation peptide conjugates for in vi vo molecular imaging applications. Mol Imaging 2006;5:1–15. 102. Guccione S, Li KC, Bednarski MD . Molecular imaging and therap y directed at the neovasculature in pathologies. How imaging can be incorporated into v ascular-targeted deli very systems to generate active therapeutic agents. IEEE Eng Med Biol Mag 2004;23:50–6. 103. Huang X, El-Sayed IH, Qian W, El-Sayed MA. Cancer cell imaging and photother mal therap y in the near -infrared re gion b y using gold nanorods. J Am Chem Soc 2006;128:2115–20. 104. Nitin N, LaConte LE, Zurkiya O , et al. Functionalization and peptide-based delivery of magnetic nanopar ticles as an intracellular MRI contrast agent. J Biol Inor g Chem 2004;9:706–12. 105. Persigehl T, Heindel W, Bremer C. MR and optical approaches to molecular imaging. Abdom Imaging 2005;30:342–54. 106. Kaul Z, Yaguchi T, Harada JI, et al. An antibody-conjugated internalizing quantum dot suitab le for long-ter m li ve imaging of cells. Biochem Cell Biol 2007;85:133–40.

107. Sun J, Zhu MQ, Fu K, et al. Lead sulf ide near-infrared quantum dot bioconjugates for tar geted molecular imaging. Int J Nanomedicine 2007;2:235–40. 108. Browne E, Dairiki JM, Doeb ler RE, et al. In: Lederer CM, Shirle y VS, editors. Table of isotopes. 7th ed. New York: Wiley; 1978. 109. Erdtmann G, Soyka W. The gamma rays of the radionuclides. Weinheim, Germany: Verlag Chemie; 1978. 110. Walker FW, Parrington JR, Feiner F. Nuclides and isotopes. In: Char t of the nuclides. 14th ed. GE nuclear ener gy, general electric compan y, nuclear energy operations, 175 cur tner avenue, M/C 397, San Jose, CA 95125 (U S A), Revised 1989. 111. Weber DA, Eckerman L, Dillman L T, Ryman JC. MIRD: radionuclide data and decay schemes. New York: Society of Nuclear Medicine; 1989. 112. Firestone RB. In: Shirle y VS, Baglin CM, Chu SYF , Zipin J, editors. Table of the isotopes. New York: Wiley; 1996. 113. Garg PK, Alston KL, Welsh PC, Zalutsky MR. Enhanced binding and inertness to dehalo genation of α-melanotropic peptides labeled using N-succinimidyl 3-iodobenzoate. Bioconjug Chem 1996;7:233–9. 114. Qu W, Kung M-P, Hou C, Oya S, K ung HF. Quick assemb ly of 1,4diphenyltriazoles as probes tar geting β-amyloid agg regates in Alzheimer’s disease. J Med Chem 2007;50:3380–7. 115. Pozzi OR, Sajaroff EO, Edreira MM. Influence of prosthetic radioiodination on the chemical and biological behavior of chemotactic peptides labeled at high specif ic acti vity. Appl Radiat Isot 2006;64:668–76. 116. Maeding P, Fuechtner F, Wuest F. Module-assisted synthesis of of the bifunctional labeling agent N-succinimidyl 4-[ 18F]fluorobenzoate ([18F]SFB). Appl Radiat Isot 2005;63:329–32. 117. Marik J, Hausner SH, F ix LA, Gagnon KJ , Sutcliffe JL. Solid-phase synthesis of 2-[ 18F]fluoropropionyl peptides. Bioconjug Chem 2006;17:1017–21.

25 13

HYPERPOLARIZED C MAGNETIC RESONANCE IMAGING—PRINCIPLES AND APPLICATIONS JAN HENRIK ARDENKJÆR-LARSEN, PHD, KLAES GOLMAN, PHD, AND KEVIN M. BRINDLE, DPHIL

Among the techniques that can be used to image biologic tissues noninvasively in vi vo, magnetic resonance (MR) is unique in the w ealth of chemical infor mation it provides. Since the demonstrations in the earl y 1970s that magnetic resonance spectra could be obtained from intact biologic systems1,2 there has been an expectation, among many practitioners, that magnetic resonance spectroscopy (MRS) could ha ve a signif icant impact in the clinic. Although there have been some notable successes,3 the technique has not entered into routine clinical use. This is due primarily to a lack of sensitivity, which means that the spectroscopic image v oxels are in the milliliter range with data acquisition times of tens of minutes. The problem is particularly severe for 13C MRS because of its low gyromagnetic ratio, w hich w as disappointing since 13 C MRS is potentiall y the most useful MRS technique for monitoring tissue metabolism and its per turbation in disease. F ollowing the administration of a 13C-labeled substrate (13C is only 1.1% naturally abundant), 13C MRS can be used to follow the distribution of the label among cellular metabolites, the 13C spectra sho wing not onl y which molecule is labeled but also w hich position in the molecule is labeled. 4 The recent introduction of dynamic nuclear polarization (DNP), 5 which can increase the sensitivity of detection of 13C and other nuclei by more than 10,000 ×, could make reality the dream of MRS as a routine clinical tool for diagnosing disease and monitoring its response to treatment. This chapter describes the physics behind the process and the applications to date in imaging of cardiac and tumor tissues, concluding with a section discussing the prospects for future success in the clinic.

PHYSICS AND INSTRUMENTATION Two methods of h yperpolarizing nuclear spins in solution have been developed over the last decade. The two methods are quite dif ferent in their ph ysics, chemistry, and the instr umentation required. The methods are: (1) parah ydrogen-induced polarization (PHIP) and (2) DNP in the solid state followed by dissolution. Both methods are described in detail. Other methods ha ve been proposed,6 but none have been developed for practical application. Nob le gas polarization b y optical pumping7 has been applied during the last decade, mainly for lung imaging, but it is not co vered in this chapter. There are some common elements to an y hyperpolarization method that need to be considered and w hich will ultimately limit the applicability of the technique.

Target Nucleus Many nuclei can be h yperpolarized by the tw o methods mentioned above: 6Li,8 19F,9 13C, 15N,10 and 89Y.11 Other nuclei, such as 1H, 31P, and 129Xe, are potential targets, but nuclei with spin g reater than 1/2, w hich will ha ve shor t relaxation times, are unattracti ve for h yperpolarization studies in vivo. The high gyromagnetic ratio nuclei lik e 1 H, 19F, and 31P will also typically have relatively short T1. The obvious nucleus of choice is 13C since many cellular metabolites can be labeled at positions in w hich the nucleus has a long relaxation time (see belo w). Another advantage of 13C is the lar ge chemical shift range and lack of background signals.

377

378

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Time Window The decay of the hyperpolarized spins to thermal equilibrium limits the time o ver which they can be studied. The signal (polarization) that is created in the polarization process decays to ther mal equilibrium with a rate (relaxation time) that depends on a number of f actors. Some factors can be controlled in the dissolution process, for example, temperature, magnetic f ield, pH, and paramagnetic impurities. Others are determined by the structure of the labeled molecule and the location of the label. In order to achie ve an acceptab le relaxation time, the prefer red labeling position is in a quater nary carbon, carbon yl, or carboxylic acid, in molecules with lo w molecular w eight (less than 200 to 300 Da). The reason is that dipolar relaxation from intramolecular and inter molecular spins ( 1H predominantly) is minimized b y being at least one bond length a way, and the small size ensures rapid ther mal motion, which averages the interaction. For carbonyls and carboxylic acids, chemical shift anisotropy also contribute to the relaxation rate, and at high f ields (> ~7 T), it will often reduce the T1. Other factors, such as protein binding, can reduce the T1 in vi vo and cause rapid loss of the hyperpolarized signal. This can be v ery pronounced for some molecules due to specif ic or nonspecif ic binding. For small molecules, the nuclear longitudinal relaxation time for carbonyls or carboxylic acids is typically 30 to 60 seconds in vivo. The T1 is then limited mainl y by dipolar relaxation with water protons. Partial deuteration can be a means to prolong T1 when intramolecular dipolar relaxation is signif icant. An understanding of the relaxation processes for nuclear spins is an impor tant aspect of hyperpolarization at all stages of the process. Methods have been proposed that, in principle, could preser ve the signal for periods longer than the T1 of the nuclear spin of 12,13 interest b y storing the spin order in singlet states. However, these ideas are dif ficult to implement in practice, especially for in vi vo imaging, due to the high radio frequency (RF) power required.

Imaging The fundamental adv antage of MR to pro vide spectral, spatial, and temporal infor mation for molecular assignment, quantif ication, localization, and kinetics means that important decisions have to be made in terms of the imaging strate gy. In some cases, spatial localization is not required beyond the sensiti ve region of the recei ver coil. The recei ver coil localizes the area of interest and the response can be considered homo genous within this v olume. Reduced spatial encoding enab les f aster temporal

acquisition and a wider spectral bandwidth. Co verage of the full chemical shift range of the nucleus is impor tant during initial studies with a ne w probe when little knowledge of the biochemistr y under the applied conditions is available; or when metabolism is suspected to be perturbed by disease or other stimuli. A typical strate gy w ould involve studying the molecule of interest using dynamic low f lip angle spectroscop y to identify an y labeled metabolites that might be for med. All spectra in a time series can be added together for maximum signal-to-noise, with little penalty compared with a single acquisition using a 90° pulse. Of course, each e xcitation pulse will destro y some of the polarization, which will in addition be decaying with a time constant gi ven by the T1. Therefore, a v ariable flip angle scheme can be used to equalize the signal-to-noise ratio over the time course and using the entire a vailable signal (ie, ending with a 90° flip angle). The time series can give further information about the kinetics of the metabolic reactions by use of appropriate kinetic modeling. Chemical shift imaging (CSI) has been the w orkhorse for spectral-spatial imaging and can be used with low flip angle or v ariable flip angle schemes. A range of two-dimensional and three-dimensional (3-D) implementations can be found in the literature. 14,15 If the metabolism of the labeled compound is w ell def ined, prior knowledge of the chemical shifts ma y be used to reduce the sampling of the spectral dimension. This will fur ther accelerate data acquisition and also allow sequences with 16,17 inherently higher signal-to-noise ratios. These sequences are based on spin echoes (RARE or steadystate free-precession ), w hich take advantage of the long T2 that is usuall y a characteristic of lo w-gamma nuclei like 13C and 15N. The f act that the h yperpolarization is nonrecoverable means that the imaging method should use ef ficiently the a vailable signal, and pre vents signal acquisition beyond the limits of T1 and T2. For an injected substance, imaging must take place within a few minutes after injection. In the human body , a substance that is injected intra venously reaches the right v entricle of the heart and the lungs in ~4 seconds, the left ventricle of the heart in ~10 seconds, and the other major organs in 10 to 40 seconds. Most signal will, therefore, be found in heart, lungs, brain, li ver, and kidne ys at the time of image acquisition. Most animal scanners ha ve multinuclear imaging capability. Apart from the challenges with the imaging strategy and pulse sequences, there is little dif ference from con ventional multinuclear imaging. Most clinical MRI scanners can be upgraded for multinuclear imaging, but the y ha ve typicall y a more limited librar y of pulse sequences. However, for large animal or clinical imaging,

Hyperpolarized

13

C Magnetic Resonance Imaging

379

Figure 1. Dynamic nuclear polarization prototype polarizers, design and principles, as used in most hyperpolarization experiments to date. A superconducting magnet charged to 3.35 T, mechanical pumps (left-hand side) to reduce the temperature of liquid helium to ~1.2 K, a microwave source to irradiate the sample at ~94 GHz, and to the right the dissolution stick, which is introduced at the end of the experiment to dissolve the frozen material in the sample container in lower right corner. Reproduced with permission from Ardenkjaer-Larsen JH et al.5

some technical limitations ma y need to be resolv ed in order to full y exploit the available signal. The low gyromagnetic ratios of 13C means that a four times higher gradient strength is required relati ve to 1H for the same spatial resolution and bandwidth. The longer T1 and T 2 may relax this requirement, but high-resolution angiographic applications will still be restricted b y cur rent clinical g radient perfor mance. Another limitation is the pulse length obtainable with the available RF power. The lower gyromagnetic ratio requires a proportionally higher B1 field for the same pulse length.

DNP In recent years, a novel method for polarizing nuclear spins in molecules or ions has been de veloped.18 The method takes advantage of DNP in the solid state followed by rapid dissolution in a suitab le solv ent.5,19 The polarization is

retained in the dissolution step, creating a solution with a nonthermal nuclear polarization approaching unity. DNP is the name of a class of spin ph ysics e xperiments f irst described theoreticall y b y Ov erhauser in 195320 and a fe w months later demonstrated b y Car ver and Slichter 21 in metallic lithium. Ov erhauser predicted that saturating the conduction electrons of a metal w ould cool the other spin system or in other w ords, dynamically polarize the nuclear spins. The predictions by Overhauser for metals were soon extended to electron spins in solution by Abragam, 22 and most nuclear magnetic resonance (NMR) spectroscopists are today familiar with the nuclear and the electronic Overhauser effect. However, this effect is limited to solutions, w here relaxation processes couple the spin systems via the lattice (ther mal motions). Soon after, the solid effect was described for spin systems in the solid state coupled by dipolar interactions.23 Later, DNP in the solid state w as extended mechanistically to processes

380

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

involving several electron spins and thermal mixing.24 The theory of DNP in the solid state, ho wever, has f ailed to provide a quantitati ve description generall y and has not made much progress since the work described in Abragam and Goldman. 24 In the solid state, the high electron-spin polarization is in par t transfer red to the nuclear spins b y microwave irradiation close to the resonance frequency of the electron spin. The efficiency of this process depends on several parameters characterizing the various spins but also on technical factors such as microwave frequency and power. DNP has mainl y been applied to making polarized targets for neutron scattering experiments, and it has been demonstrated that the nuclear polarizations of 1H and 13C could be increased to almost 100% and to ~50%, respectively, in the solid state b y means of DNP at low temperature.25,26 The mechanism requires the presence of unpaired electrons, which are added to the sample as, for example, an or ganic free radical. In order for the DNP process to be ef fective, the radical must be homo geneously distributed within the sample. To achie ve this glass for mers, for e xample, gl ycerol or DMSO , can be added to pre vent cr ystallization and produce an amorphous solid upon cooling of the sample. This step is a major challenge for h yperpolarization b y DNP . Man y compounds will ha ve a tendenc y to cr ystallize at high concentrations for ming hetero geneous domains, w hich polarize poorly. Since it is often a requirement for in vivo studies to achieve a high concentration of the compound after dissolution, it is necessar y to be able to prepare the compound in a concentrated for m. If an additi ve (glass former) is used, this has to be biolo gically compatible or removed from the solution after polarization. The source of the unpaired electron is typicall y an organic free radical, but a few metal ions have been used successfully for DNP, Cr(V) in particular.21,22,27 The choice of radical will depend on a number of f actors. Firstly, the radical needs to be chemically stable and dissolve readily in the matrix of interest. Secondl y, the electron paramagnetic resonance (EPR) spectrum of the radical should have characteristics that allow DNP to be effective for the nucleus of interest. In the f irst instance, this means that the EPR spectrum should ha ve a line width that e xceeds the Larmor frequency of the nuclear spin. In practice, the abo ve criteria means that tw o classes of radicals are a vailable, the nitroxides and the trityls. 28 The nitroxides are a class of radicals that have been studied extensively by EPR and which have been used for DNP in many systems. Nitroxides are characterized b y having a broad EPR spectr um (Δg/g ≈ 4.0 × 10–3),23,29 covering the Larmor frequency of all nuclear spins. Some of them have reasonable chemical

stability and come with dif ferent degrees of hydrophilicity. Another class of radicals with superior proper ties for direct polarization of low-gamma nuclei like 13C, 15N, and 2 H are the trityls ( Δg/g ≈ 0.25 × 10–3).23 The trityls also have a range of h ydrophilicities, and some of them are chemically very stable. The instrumentation for hyperpolarization by DNP was first described b y Ardenkjaer-Larsen and colleagues. 5 The polarizer builds to a lar ge extent on principles estab lished by the polarized tar get community. Most solid-state DNP has been performed at magnetic fields between 0.35 T30 and 16.5 T 31,32 and at temperatures from belo w 1 K to room temperature. At temperatures belo w a fe w degrees Kelvin and magnetic f ield strengths abo ve a fe w Tesla, electron spins are fully polarized and large nuclear polarizations can be obtained. Since polarizations close to unity are the ultimate aim for in vi vo applications, it is impor tant to choose initial operating conditions that have been proven to provide high nuclear polarization, but at the same time, are easil y obtainable using standard instrumentation. Temperatures of ~1 K can be achieved by pumping on liquid helium. In the original work, the liquid helium was supplied to the sample space through a needle valve from the magnet cryostat, but in a recent publication, an alternative arrangement that used a separate helium dewar was described.29 A magnetic f ield strength of 3.35 T w as chosen to put the micro wave frequency for DNP , with a g = 2 paramagnetic agent, at 94 GHz (W-band), where microwave sources w ere readily available. Once adequate solid-state polarization has been obtained, the sample is dissolved in a suitable buffer. The dissolution may involve neutralization of the agent with acid or base, depending on the solid sample preparation. The solution is buf fered to maintain control of pH and achie ve a ph ysiologically acceptab le for mulation. The dissolution has to be v ery efficient to preser ve the nuclear polarization in this process. F ormulating the solid sample as beads or powder may improve the dissolution (polarization and concentration), but understanding and optimizing the fluid dynamics, as w ell as providing the necessary heat, is essential for best performance of more dif ficult agents. Relaxation during the dissolution process can depend on many factors. To minimize relaxation, dissolution is perfor med in the high magnetic f ield of the polarizer , but abo ve the liquid helium. Any paramagnetic ions that could cause relaxation are chelated b y adding, for e xample, eth ylenediaminetetraacetic acid to the dissolution solv ent. Finally, the solution ma y under go a chromato graphy step to remove the paramagnetic agent (radical) in volved in the DNP process. In most cases, the radical does not cause

Hyperpolarized

significant relaxation after dissolution, and ma y also be safe to inject into animals, but can simply be removed by ion-exchange or re verse-phase chromatography in most cases.

PHIP In the late 1980s, Bo wers and Weitekamp33,34 discovered, both theoretically and experimentally, that the hydrogenation of small or ganic molecules with parah ydrogen led to a highly ordered spin state, w hich manifested as v ery lar ge NMR signals for the cor responding protons. The names of PASADENA34 and ALTADENA,35 originally introduced by Weitekamp, have now been superseded b y the name PHIP, for parah ydrogen-induced polarization. This phenomenon arises from the quantum statistical mechanical proper ties of dihydrogen. Of the four possible spin isomers of dihydrogen, the singlet state of the nuclear spins, called parah ydrogen, has the lowest energy. At room temperature, there is a random mixture of the four spin isomers, gi ving 25% of the para form. When the temperature is lowered, the fraction of parahydrogen increases and approaches unity at ~20 K. If a molecule of parah ydrogen is allo wed to attach itself to another molecule b y a chemical reaction, it will initiall y retain the spin cor relation between the tw o protons but, in most cases, the symmetry of the hydrogen molecule will be broken due to couplings with other spins or dif ferences in chemical shift. The 1H spectrum will be strongl y enhanced compared with a ther mal equilibrium spectr um and will show a characteristic antiphase patter n. The nuclei are not polarized, rather the y are in a state of increased spin order . This spin order can be converted into net polarization in various w ays.36 The PHIP phenomenon has been re viewed recently b y se veral authors. 37–39 The chemistr y of h ydrogenation and catal yst de velopment is a complicated and limiting aspect of the technique. F or imaging applications, complete hydrogenation at high concentrations in aqueous systems has to be achie ved. A number of substrates ha ve been demonstrated for angio graphic and perfusion studies, but there has been limited success with endogenous metabolites thus f ar. The instrumentation for PHIP is, on the other hand, quite simple.36

CARDIOVASCULAR APPLICATIONS OF DNP

13

C Magnetic Resonance Imaging

381

is obtainable. The solution of the hyperpolarized molecule will typicall y be highl y concentrated (0.3 to 1.0 M) and currently with a polarization level of ~20%. It is, therefore, possible to track the position of a catheter containing this solution in “real time” during an interventional MRI examination.40 The lack of backg round signal with h yperpolarized 13C imaging and consequentl y high CNR mak es 13 C-labeled substances potentiall y attractive for inter ventional endovascular MRI procedures. 41 By using an MRI system capable of multinuclear transmit and recei ve, 13C catheter images ma y be acquired simultaneousl y with a proton “road map” image by online image fusion. Such 13C projection images overlaid on a 1H road map may be used directly for interventional guidance.42 An example of such a 3-D catheter reconstruction and image fusion is shown in Figure 2, w here the catheter f illed with a h yperpolarized 13 C-labeled substance w as track ed when moving through the aor ta and into the renal ar tery of a pig. When the catheter tip had reached a position pro ximal to the right kidney, a few milliliters of the contrast agent were injected into the kidne y. This indicates a potentiall y interesting feature of the h yperpolarization technique because it is now possib le not onl y to visualize the catheter used for interventional MRI procedures with high spatial resolution but also to localize the position and distribution of the injected substance. During chemical ab lation procedures and therapy, this is a critical factor, and it is not possible to do this using any other imaging technique. If one chooses a 13C-labeled substance that is metabolized by cells, then the method opens another interesting possibility for the inter ventional procedure since the metabolism of the chosen substance ma y infor m on the viability and the metabolic status of the cells in the area where the substance is located. A map of the metabolic status will be an impor tant guide to w here a possib le biopsy should be taken and where a metabolically abnormal tissue, for example, a tumor, is located. Viability and metabolic activity are tw o parameters of major medical importance as the y may infor m about the general function of the tissue of interest. The use of h yperpolarized substances under some conditions allows localization and quantification of these parameters, something w hich has not until now been possib le using other clinical imaging techniques.

Cardiac Metabolism Interventional Imaging with Hyperpolarized 13C Immediately after preparation of the solution containing the hyperpolarized molecule, an extremely high MR signal

Many imaging modalities can be used to e valuate diseases of the heart; however, it has been suggested that MRI could be used as a “one-stop-shop” for disease diagnosis. State-of-the-art MRI is ab le to depict cardiac

382

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 2. Catheter tracking of the renal arteries in a pig. A bolus of 2-hydroxyethylacrylate was injected into the kidney via a separate channel of the catheter. The 13C images series (color) were acquired using a true fast imaging steady-state free-precession sequence with in-plane resolution 2 × 2 mm2 and acquisition time 329 ms/image.

anatomy, wall thickness, motion, and function. By injecting a Gd 3+-contrast medium, it is possib le to assess myocardial perfusion 43 and through late enhancement methods, to demonstrate the e xtent of inf arcted myocardium.44 However, for a method to be clinicall y acceptable, it needs to be capable of delineating the morphologic re gion of interest and , w hen perfor med routinely, it should allo w ready dif ferentiation of diseased tissue from normal tissue. Selection of an appropriate 13 C-labeled molecule and the use of hyperpolarized MRI should allo w this. Experimental MR angio graphy with hyperpolarized 13C-labeled substances has sho wn high spatial and temporal resolution 45,46 and the f acility to perform quantitati ve perfusion measurements. 47,48 By using p yruvate as the 13C-labeled substance and performing cardiac CSI within 1 minute after injection, it was possible to evaluate the metabolic status and the viability of myocardial cells. 49 Pyruvate is a substance that is at a central crossroads of ener gy metabolism in hear t muscle,50 and indeed, in nearly all tissues of the body. Its metabolic fate in heart muscle has been described extensively, both under nor mal51 and in disease conditions. 52 Pyruvate is reduced to lactate b y the reduced coenzyme nicotinamide adenine dinucleotide (NADH), in the reaction catal yzed b y the enzyme lactate deh ydrogenase (LDH). The resulting o xidized coenzyme (N AD)+ is reused in the gl ycolytic pathway, which is an impor tant source of adenosine tri-phosphate (A TP) and also of biosynthetic inter mediates. Reo xidation of N ADH b y pyruvate allo ws the gl ycolytic pathw ay to produce ATP under anaerobic conditions, w hich is par ticularly

important in tumors, which often have a poor and rapidly fluctuating blood supply and as a consequence are often hypoxic.53 Pyruvate can also be transaminated to for m alanine in the reaction catal yzed b y alanine aminotransferase. The source of the amino group for this reaction is glutamate, and the reaction is par ticularly important in muscle, where during fasting, amino groups resulting from protein de gradation in the muscle are incorporated into alanine, w hich then passes via the circulation to the liver, where it is used to mak e glucose in the pathw ay of gluconeo genesis. The reactions catalyzed by LDH and alanine aminotransferase are thought to be near to equilibrium in most tissues, and therefore, these enzymes can catal yze rapid e xchange of the 13C label in [1-13C]pyruvate with lactate and alanine, respectively. Pyr uvate can also be o xidatively decarbo xylated to produce acetyl CoA in the reaction catalyzed by pyruvate dehydrogenase. This ir reversible reaction commits the carbon skeleton of pyruvate to oxidation in the Krebs cycle to for m CO 2. The resulting reduced coenzymes, NADH and flavin adenine dinucleotide (FADH2) donate their electrons to the electron transpor t chain on the inner mitochondrial membrane, w here the ener gy released from their subsequent transfer to molecular oxygen, to for m w ater, is used to dri ve ATP synthesis. This is a fundamental ener gy process in most tissues, particularly those with a high energy demand, such as the heart. Because of the high SNR generated b y hyperpolarization, it is possib le to use MRI to visualize and map the labeled metabolites for med during the f irst minute after intra venous injection of h yperpolarized

Hyperpolarized

13

C Magnetic Resonance Imaging

383

Figure 3. The proton image, the semiquantitative Gd3+-perfusion map, the Gd3+-delayed enhancement image, together with the alanine and bicarbonate maps obtained in a pig 2 hours after 15 min of coronary artery occlusion. The Gd3+-perfusion map, the alanine map and the bicarbonate map have been superimposed on the proton image. Data reproduced with permission from Golman K et al.49

[1-13C]pyruvate. If the p yruvate is injected 2 hours after a local 15 minutes ischemia in the m yocardium (Figure 3), CO 2 production is reduced in the re gion of ischemia and this seems to be more sensitive to ischemia than an y of the other w ell-established parameters that characterize cardiac function (m yocardial mor phology, cardiac motion, perfusion, and delayed enhan cement). 49 Radioactive methods, such as single-photon emission computed tomo graphy (SPECT) and positron emission tomography (PET), can to some e xtent provide the same information. Both SPECT and PET ma y be used to e valuate hear t perfusion and to some de gree the amount of cell death. 54 PET imaging can depict uptak e of the glucose analo g, 18F-fluorodeoxyglucose (FDG), across the cell membrane but cannot dif ferentiate betw een loss of cell membrane transpor t activity and changes in the later stages of the energy-generating metabolic process.54 Only MRI gi ves direct infor mation on the chemical transformation of the injected molecule.

APPLICATIONS OF DNP IN ONCOLOGY Cancer could be considered a metabolic disease as changes in metabolism clearl y contribute to malignant pro gression.55 Tumor cells ha ve long been kno wn to ha ve high rates of aerobic gl ycolysis, the so-called “W arburg effect,”53 and our understanding of the biochemical basis for this increased flux has g rown rapidl y in the last fe w years. Se veral onco genes ha ve been sho wn to increase directly the e xpression of gl ycolytic enzymes and to increase them indirectly by up-regulating expression of the transcription f actor, h ypoxia-inducible f actor (HIF-1 α), which causes coordinate increases in the e xpression of most of the glycolytic enzymes.56,57 HIF expression is normally increased by tissue hypoxia, but in tumors, oncogene activation can increase its e xpression, even under aerobic conditions, leading to increased aerobic gl ycolysis. This is thought to provide a sur vival advantage to the tumor cells

in the frequentl y anaerobic microen vironment of a solid tumor. In addition, since the intermediates in the glycolytic pathway are in volved in the synthesis of man y cell constituents, such as lipids and nucleotides, the increased glycolytic flux will support the higher rates of biosynthesis required by rapidly growing tumor cells. The tricarboxylic acid (TCA) cycle is also involved in tumorigenesis since it has been sho wn that the TCA c ycle enzymes, succinate dehydrogenase (SDH) and fumarate hydratase (FH), act as tumor suppressors b y promoting HIF de gradation.58 Hydroxylation of HIF-1 α at two specif ic prolyl residues promotes association with v on Hippel-Lindau f actor and subsequent proteol ytic de gradation b y the proteosome. 59 Loss of SDH or FH results in accumulation of succinate or fumarate, respecti vely, either of w hich inhibits HIF-1 α prolyl h ydroxylase and thus stabilizes HIF . Phospholipid and amino acid metabolism are also altered in tumors. Ras oncogene transfor mation can stimulate choline kinase activity, leading to increased levels of phosphocholine, and numerous 1H and 31P MRS studies ha ve demonstrated an increase in the levels of phosphocholine- and choline-containing compounds when compared with nor mal tissues.60 Tumors frequentl y sho w increased amino acid uptak e, which seems to be predominantly due to increased expression of the L-type amino acid transpor ter.61 This altered metabolism of tumor cells has been exploited to image tumors and to detect their earl y responses to treatment. In an era w here the introduction of new drugs is star ting to increase the number of treatment options for patients with cancer , techniques for detecting earl y responses will become increasingl y important, as the y should allo w the oncolo gist to select the most appropriate treatment at an early stage.62,63 Currently, treatment response is often e valuated by looking for evidence of tumor shrinkage, ho wever, this may take weeks or e ven months to become apparent and , in some cases, for e xample, with antiangio genic dr ugs, may not occur at all despite a positi ve response to therapy.

384

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

The most widely used technique clinically for imaging tumor metabolism and the response of tumors to treatment has been PET. The elevated aerobic gl ycolysis in tumors has been imaged indirectly using FDG. FDG is a glucose analog that is tak en up by cells, on the plasma membrane glucose transpor ter, and then phosphor ylated to fluorodeo xyglucose 6-phosphate b y he xokinase, the first enzyme of the gl ycolytic pathway. Because FDG is not signif icantly fur ther metabolized , the positronemitting isotope is trapped in the cell, w here it can be imaged, at relati vely low resolution (2 to 3 mm), using PET. FDG-PET can be used to locate tumors and has been widely used to detect response to treatment in lung, esophageal, lymphoma, breast, and o varian cancers. 63 In patients with gastrointestinal stromal tumors ( GISTs) treated with the receptor tyrosine kinase inhibitor , Imatinib, there w as a substantial decrease in FDG uptak e long before there was evidence of tumor shrinkage in the X-ray computed tomography image.64 Fatty acid metabolism in tumor cells has also been imaged with PET, using 11 C-labeled acetate. 65 In most nor mal tissues, the acetate is oxidized in the TCA cycle and the 11C label is lost as 11 CO2, w hereas in man y tumors, the labeled acetate is used for f atty acid synthesis, resulting in trapping of the label and hence detection in the PET image. 11C-acetate PET was shown to be better than FDG-PET in detecting local recur rent disease in the prostate, or distant metastases, following radical prostatectom y or radiation therapy.66 Changes in choline metabolism can be anal yzed using choline, or choline analogs, labeled with 11C or 18F. In prostate cancer, the uptake of a labeled choline analog was decreased earl y in antiandro gen therapy.67 Elevated amino acid uptake in tumors has been imaged using PET and 11C-labeled amino acids. 61 In glioma, a reduction in methionine uptak e during temozolomide treatment w as shown to predict a f avorable treatment outcome. 68 The MR methods for detecting altered tumor metabolism include the above-mentioned 1H and 31P MRS studies demonstrating changes in choline metabolism. MRS measurements of decreases in the le vels of choline metabolites following treatment ha ve been sho wn to be predictive of response in brain, prostate, and breast cancer in the clinic. 69,70 1H MRS measurements in animal models have demonstrated ele vated tumor lactate le vels, and spectroscopic imaging e xperiments ha ve demonstrated a decrease in tumor lactate concentration follo wing dr ug treatment, consistent with a decrease in tumor glycolytic rate, increased lactate washout, or a decrease in tumor cellularity.71 In a subsequent study, in which spectral editing w as used to suppress the o verlapping lipid resonance, the lactate signal was acquired from the whole

tumor.72 It is not y et clear w hether these MRS measurements of steady-state tumor lactate concentrations ha ve sufficient sensiti vity to be used for treatment response monitoring in the clinic. In general, a lack of sensiti vity has limited the application of MRS to monitoring treatment response in the clinic. The introduction of dissolution dynamic nuclear spin h yperpolarization could , in principle, change this. Intravenous injection of h yperpolarized [1- 13C]pyruvate has been shown, in experiments in mice and rats, to result in rapid labeling of tumor lactate (F igure 4). 73,74 Experiments on isolated tumor cells ha ve demonstrated that this labeling is due primaril y to e xchange of label betw een pyruvate and lactate in a near -equilibrium reaction catalyzed by the enzyme LDH, rather than net chemical conversion of pyruvate to lactate (Figure 5). Although there is no conclusive evidence, this seems likely to be the case in vivo as well since the reaction is widely thought to be near to equilibrium in vivo75 and the mechanism of the enzyme (ordered ternary complex, with binding of the coenzymes before lactate and pyruvate) means that label can exchange between lactate and pyruvate in the absence of dissociation of the coenzymes NAD+ and NADH.76 These animal studies demonstrated that the majority of the lactate w as labeled within the tumor, rather than being labeled in other tissues and then washed into the tumor via the circulation. In effect, the h yperpolarized 13C pyruvate “lights up, ” by exchange, pools of tissue lactate, w hich are often associated with tumors and also with other pathologies that result in tissue h ypoxia or ischemia. P assive e xchange of label into pree xisting metabolite pools probably also e xplains why hyperpolarized [1-13C]pyruvate is effective at labeling alanine in tissues where it is abundant, such as muscle.

Figure 4. Flux of hyperpolarized 13C label between pyruvate and lactate in a subcutaneous murine tumor. A series of 13C magnetic resonance spectra from a 5 mm thick slice through a tumor following intravenous injection of 0.2 mL of 75 mM hyperpolarized [1-13C]pyruvate. Spectra were collected every 2 s for 2 mins starting from 12 s after the beginning of injection. The resonance at 171 ppm is from [1-13C]pyruvate, and the resonance at 183 ppm is from [1-13C]lactate. Reproduced with permission from Day SE et al.74

Hyperpolarized

A

13

C Magnetic Resonance Imaging

385

B

C

Figure 5. Experimental demonstration of 13C label exchange in a tumor cell suspension. Exchange of 13C label between 40 mM [3-13C]pyruvate and 40 mM unlabeled lactate following their addition to the cell suspension. The spectra were collected after 1 min (A) and 40 min (B). The pyruvate and lactate peak integrals are shown as a function of time in (C). The spikes in these traces were due to air bubbles passing through the nuclear magnetic resonance receiver coil. Note that there was no change in the total pyruvate concentration but a decrease in the concentration of [3-13C]pyruvate and an increase in the concentration [3-12C]pyruvate, demonstrating exchange of 13 C label between pyruvate and lactate. The increase in lactate concentration is due to its production from glucose in the medium. Reproduced with permission from Day SE et al.74

Experiments in mice ha ve also demonstrated that the polarized pyruvate experiment can be used for treatment monitoring. In a murine l ymphoma model, treatment with a c ytotoxic dr ug (etoposide) resulted in a decrease in the flux of label from p yruvate to lactate in the tumor.74 This was shown to be due to a number of factors, including a decrease in tumor LDH and lactate concentrations, which probably resulted from a decrease in tumor cellularity. Experiments on isolated tumor cells also showed that there w as a decrease in the N AD(H) pool as a result of pol y(adenosine diphosphate (ADP)ribose) pol ymerase (PARP) acti vation. PARP, w hich is

activated by DNA damage and ADP ribosylates nuclear proteins, uses as its substrate N AD+, substantiall y depleting the dying cell of the coenzyme. 77 In these experiments, the decrease in pyruvate–lactate label flux was demonstrated at 24 hours after dr ug treatment, when there was already an ~20% decrease in tumor volume. Fur ther work is required to deter mine how earl y this decrease in flux can be detected after dr ug treatment and w hether treatment responses can be detected before there is evidence of tumor shrinkage. It will also be important to evaluate this method for detecting treatment response in comparison with other methods that

386

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

are already being used clinically, such as FDG-PET and diffusion-weighted MRI. 78 The latter technique, w hich measures the apparent dif fusion coef ficient (ADC) of tumor w ater, detects treatment response through an increase in water ADC,78 which is thought to be due to a decrease in tumor cellularity in responding tumors. Another h yperpolarized substrate that could potentially be used for treatment monitoring is [5-13C]glutamine. The T 1 for this carbon is reasonab ly long (~16 seconds), and the change in chemical shift w hen glutamine is converted to glutamate is sufficiently large (3.4 ppm) to make detection possible in vivo. We have demonstrated that in a human hepatoma cell line, HepG2, the cell uptak e of glutamine and its net con version to glutamate, in the ir reversible reaction catal yzed b y the intramitochondrial enzyme glutaminase, is suf ficiently rapid to allo w its detection with hyperpolarized [5-13C]glutamine. Since glutamine uptake by hepatomas is ~30 × higher than in surrounding nor mal li ver,79 13C spectroscopic imaging following intravenous injection of [5- 13C]glutamine might be useful clinically for detection of hepatomas, w hich are difficult to detect using current imaging methods. 80 Transfer to the clinic is, in principle, possible since glutamine is already safel y infused in relati vely lar ge amounts into patients. Ho wever, detection of glutamine utilization in vivo is likely to require fur ther improvements in hyperpolarization (currently ~5%) in order to achie ve the required sensitivity, as has been achieved with other substrates, such as p yruvate. Glutamine utilization is closel y cor related with tumor cell proliferation. 79 The nitrogen in the molecule is used for biosynthetic pur poses, and the carbon skeleton is o xidized in the TCA cycle, where it mak es a substantial contribution to cellular ener gy generation. 81 Therefore, measurements of tumor [5- 13C]glutamine utilization might also be used to detect the ef fects of cytostatic drugs, in much the same w ay as the th ymidine analog 3ʹ′-deoxy-3ʹ′-[18F]fluorothymidine (FLT) has been used to detect the ef fects of antiproliferati ve dr ugs with PET .82 FLT is trapped in cells follo wing phosphor ylation b y cytosolic thymidine kinase 1, w hich is upregulated during S phase of the cell c ycle.

CONCLUDING REMARKS The increased SNR obtained using h yperpolarization potentially allo ws impro ved inter ventional catheter tracking, high-resolution angio graphy and mapping of injected substances, quantitati ve perfusion studies and measurements of the metabolism of certain selected substances, and consequently tissue viability. Most of these

parameters are of importance for the diagnosis of cancer and hear t disease and could also be used to detect disease and response to treatment in other or gans in the body. The future success of the technique will depend not only on the polarization that can be achieved but also on instrument development. Some clinical scanners have multinuclear capabilities, but the MR hardware and software needs fur ther de velopment. The pulse sequence used in all repor ted metabolic studies so f ar has been CSI, w hich w as originall y de veloped for proton spectroscopy and in general takes 10 to 15 seconds to collect the data. Ne w pulse sequences under de velopment will allow subsecond metabolite imaging with impro ved image quality. Pyruvate has been used for most of the pub lished studies on metabolic imaging. Ho wever, the h yperpolarization technique allo ws a number of other 13C- or 15 N-labeled molecules to be used , which have exciting potential for impro ving medical diagnosis. These include 15N-labeled choline and [1- 13C]succinate. The latter has also been hyperpolarized using parahydrogen; however, studies on brain tumor cells have provided no evidence, at the cur rent le vels of polarization, that metabolism of this metabolite can be detected during the lifetime of the polarization. 83 The 15N nucleus in choline has an exceptionally long T1 (~2 minutes), and the material has been successfull y polarized to 6%. 10 However, the chemical shift change on phosphorylation is very small (~0.2 ppm), which would make its metabolism dif ficult to detect in vi vo, and it is unclear whether its metabolism is suf ficiently f ast to allo w detection, despite the v ery long lifetime of the polarization. Furthermore, the compound is toxic at the millimolar le vels that w ould be required in the circulation.84 At the present, it w ould seem that h yperpolarized pyruvate has a reasonab le prospect, in the immediate term, of translating to the clinic. Gi ven the v ersatility of DNP for h yperpolarizing a wide range of dif ferent molecules, other endo genous metabolites ma y be shown to be useful in the future, and it is possib le that synthetic probe molecules could be de veloped, as has been the case for hyperpolarized xenon.85 Even if pyruvate were the only molecule to successfully translate to the clinic, it seems lik ely that it will f ind numerous applications in clinical imaging; for treatment response monitoring in tumors, as a general tool for imaging tissue viability, and for detecting the lactate accumulations associated with other tissue patholo gies, such as ischemia and hypoxia.

Hyperpolarized

REFERENCES 1. Hoult DI, Busb y SJW , Gadian DG, et al. Obser vation of tissue metabolites using 31P nuclear magnetic resonance. Nature 1974;252:285–7. 2. Shulman RG, Brown TR, Ugurbil K, et al. Cellular applications of 31P and 13C nuclear magnetic resonance. Science 1979;205:160–6. 3. Radda GK. The use of NMR spectroscop y for the understanding of disease. Science 1986;233:640–5. 4. Jeffrey FMH, Rajagopal A, Malloy CR, Sherry AD. 13C NMR: a simple y et comprehensi ve method for anal ysis of inter mediary metabolism. Trends Biochem Sci 1991;16:5–10. 5. Ardenkjaer-Larsen JH, Fridlund B, Gram A, et al. Increase in signalto-noise ratio of > 10,000 times in liquid-state NMR. Proc Natl Acad Sci U S A 2003;100:10158–63. 6. Krjukov EV, O’Neill JD, Owers-Bradley JR. Br ute force polarization of 129Xe. J Low Temp Phys 2005;140:397–408. 7. MacFall JR, Driehuys B. Hyperpolarized gas imaging. In: Grant DM, Harris RK, editors. Enc yclopedia of magnetic resonance. John Wiley & Sons Ltd; 2002. 8. van Heeswijk RB, Uffmann K, Kurdzesau F, et al. Towards detection of sub-micromolar contrast agent concentration with h yperpolarized 6-lithium. Proc Int Soc Magn Reson Med 2007. 9. Kuhn L T, Bommerich U , Bar gon J . Transfer of parah ydrogeninduced hyperpolarization to19F. J Phys Chem 2006;110:3521–6. 10. Eykyn TR, Re ynolds S, Gabellieri C, Leach MO . Hyper polarised N-15 of choline—potential for obser ving phospholipid metabolism in cancer. Proc Int Soc Magn Reson Med 2007;15:1319. 11. Merritt ME, Har rison C, K ovacs Z, et al. Hyper polarized 89Y offers the potential of direct imaging of metal ions in biological systems by magnetic resonance. J Am Chem Soc 2007;129:12942–3. 12. Pileio G, Consistre M, Car ravetta M, Levitt MH. Long-lived nuclear spin states in the solution NMR of four -spin systems. J Magn Reson 2006;182:353–7. 13. Pileio G, Le vitt MH. J-stabilization of singlet states in the solution NMR of multiple-spin systems. J Magn Reson 2007;187:141–5. 14. Cunningham CH, Chen AP, Albers MJ , et al. Doub le spin-echo sequence for rapid spectroscopic imaging of h yperpolarized 13C. J Magn Reson 2007;187:357–62. 15. Mayer D, Levin YS, Hurd RE, et al. F ast metabolic imaging of systems with sparse spectra: application for h yperpolarized 13C imaging. Magn Reson Med 2006;56:932–7. 16. Leupold J, Wieben O, Mansson S, et al. F ast chemical shift mapping with multiecho balanced SSFP . Magn Reson Mater Ph ys 2006;19:267–735. 17. Leupold J, Mansson S, Wieben O, et al. Of f-resonance behaviour of RARE and TrueFISP in imaging of h yperpolarized 13C. Proc Int Soc Magn Reson Med 2007;15. 18. PCT publication WO 1999/35508 (1998). 19. Wolber J, Ellner F , F ridlund B , et al. Generating highl y polarized nuclear spins in solution using dynamic nuclear polarization. Nucl Instrum Methods Phys Res 2004;526: 173–81. 20. Overhauser A. P olarization of nuclei in metals. Ph ys Re v 1953;92:411–5. 21. Carver TR, Slichter CP. Polarization of nuclear spins in metals. Ph ys Rev 1953;92:212–3. 22. Abragam A. Ov erhauser ef fect in non-metals. Ph ys Re v 1955;98:1729–35. 23. Jeffries CD. Polarization of nuclei b y resonance saturation in paramagnetic crystals. Phys Rev 1957;108:164–5. 24. Abragam A, Goldman M. Nuclear magnetism: order and disorder . Oxford, UK: Clarendon Press; 1982. 25. de Boer WD, Borghini M, Morimoto K, et al. Dynamic polarization of protons, deuterons, and C-13 nuclei—ther mal contact between nuclear spins and an electron spin-spin interaction reservoir. J Low Temp Phys 1974;15:249–67.

13

C Magnetic Resonance Imaging

387

26. de Boer W, Niinikos TO. Dynamic proton polarization in propanediol below 0.5 K. Nucl Instr um Methods 1974;114:495–8. 27. Heckmann J, Meyer W, Radtke E, Reicherz G. Electron spin resonance and its implication on the maximum nuclear polarization of deuterated solid target materials. Phys Rev 2006;74:Art. No. 134418. 28. PCT publications WO 2006/011811 (2004). 29. Comment A, van den Brandt B, Uffmann K, et al. Design and performance of a DNP prepolarizer coupled to a rodent MRI scanner . Concepts Magn Reson 2007;31:255–69. 30. Duijvestijn MJ, Wind RA, Smidt J . Quantitative investigation of the dynamic nuclear polarization effect by fixed paramagnetic centra of abundant and rare spins in solids at room temperature. Ph ysica B&C 1986;138:147–70. 31. Bajaj VS, Hornstein MK, Kreischer KE, et al. 250 GHz CW gyrotron oscillator for dynamic nuclear polarization in biolo gical solid state NMR. J Magn Reson 2007;189:251–79. 32. Hornstein MK, Bajaj VS, Kreischer KE, et al. CW second har monic results at 460 GHz of a gyrotron oscillator—for sensiti vity enhanced NMR. The Joint 30th Inter national Conference on Infrared and Millimeter Waves and 13th International Conference on Terahertz Electronics 2005;2:437–8. 33. Bowers CR, Weitekamp DP. Transformation of symmetrization order to nuclear -spin magnetization b y chemical reaction and nuclear magnetic resonance. Phys Rev Lett 1986;57:2645–8. 34. Bowers CR, Weitekamp DP. P arahydrogen and synthesis allo w dramatically enhanced nuclear alignment. J Am Chem Soc 1987;109, 5541–5542. 35. Pravica MG, Weitekamp DP. Net alignment b y adiabatic transpor t of parahydrogen addition products to high magnetic f ield. Chem Phys Lett 1998;145:255–8. 36. Goldman M, Jóhannesson H, Axelsson O , Karlsson M. Design and implementation of 13C hyper polarization from para-h ydrogen for new MRI contrast agents. Comptes Rendus Chimie 2006;9:357–63. 37. Duckett SB, Sleigh CJ . Applications of the parah ydrogen phenomenon: a chemical perspecti ve. Pro g Nucl Magn Reson Spectrosc 1999;34:71–92. 38. Kuhn LT, Bargon J. Transfer of parahydrogen-induced hyperpolarization to heteronuclei. Top Curr Chem 2007;276:25–68. 39. Bowers CR. Sensitivity enhancement utilizing parahydrogen. In: Grant DM, Harris RK, editors. Encyclopedia of magnetic resoance. Ne w York: John Wiley & Sons Ltd; 2002. p. 750–70. 40. Magnusson P, Johansson E, Månsson S, et al. P assive catheter tracking during inter ventional MRI using h yperpolarized 13C. Magn Reson Med 2007;1140–7. 41. Olsson LE, Chai C-M, Axelsson O, et al. MR coronary angiography in pigs with intraarterial injections of a hyperpolarized 13C substance. Magn Reson Med 2006;55:731–7. 42. Månsson S, Johansson E, Magnusson P , et al. 13C imaging—a ne w diagnostic platform. Eur Radiol 2006;16:57–67. 43. Larsson HB, Fritz-Hanson T, Rostrup E, et al. My ocardial perfusion modeling using MRI. Magn Reson Med 1996;35:716–26. 44. Zhu X, Zhang R, Campagna NF , Zhao M. Assessment of reperfused myocardial inf arction in the h yper-acute phase with dela yed enhancement magnetic resonance imaging. J Cardio vasc Magn Reson 2006;8:461–7. 45. Golman K, Ardenkjer-Larsen JH, Petersson JS, et al. Molecular imaging with endo genous substances. Proc Natl Acad Sci U S A 2003;100:10435–9. 46. Svensson J, Månsson S, Johansson E, et al. Hyper polarized 13C MR angiography using true FISP. Magn Reson Med 2003;50:256–62. 47. Johansson E, Månsson S, Wirestam R, et al. Cerebral perfusion assessment b y bolus tracking using h yperpolarized 13C. Magn Reson Med 2004;51:464–72. 48. Johansson E, Olsson LE, Månsson S, et al. P erfusion assessment with bolus differentiation: a technique applicab le to hyperpolarized tracers. Magn Reson Med 2004;52:1043–51.

388

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

49. Golman K, P etersson JS, Magnusson P , et al. Cardiac metabolism measured noninvasively by hyperpolarized 13C MRI. Magn Reson Med 2008. [In press] 50. Mallet R T. Pyr uvate: metabolic protector of cardiac perfor mance. Proc Soc Exp Biol Med 2000;223:136–48. 51. Panchal AR, Comte B , Huang H, et al. P artitioning of p yruvate between oxidation and anaplerosis in s wine hearts. Am J Physiol Heart Circ Physiol 2000;279:2390–8. 52. Panchal AR, Comte B , Huang H, et al. Acute hiber nation decreases myocardial pyruvate carboxylation and citrate release. Am J Physiol Heart Circ Physiol 2001;281:1613–20. 53. Gatenby RA, Gillies RJ . Why do cancers ha ve high aerobic gl ycolysis? Nat Rev Cancer 2004;4:891–9. 54. Satake O, Kajinami K, Ishikawa Y, et al. Myocardial glucose metabolism assessed b y positron emission tomo graphy and the histopathologic f indings of micro vessels in syndrome X. Circ J 2004;68:220–6. 55. Shaw RJ . Glucose metabolism and cancer . Cur r Opin Cell Biol 2006;18:598–608. 56. Dang CV, Semenza GL. Oncogenic alterations of metabolism. Trends Biochem Sci 1999;24:68–72. 57. Kim J-W, Gao P, Dang CV. Effects of hypoxia on tumor metabolism. Cancer Metastasis Rev 2007;26:291–8. 58. Selak MA, Armour SM, MacK enzie ED, et al. Succinate links TCA cycle dysfunction to onco genesis b y inhibiting HIF- α prolyl hydroxylase. Cancer Cell 2005;7:77–85. 59. Bruick RK, McKnight SL. Oxygen sensing gets a second wind. Science 2002;295:807–8. 60. Glunde K, Jacobs MA, Bhujwalla ZM. Choline metabolism in cancer: implications for diagnosis and therap y. Exper t Re v Mol Diagn 2006;6:821–9. 61. Jager PL, Vaalburg W, Pruim J, et al. Radiolabeled amino acids: basic aspects and clinical applications in oncolo gy. J Nucl Med 2001;42:432–45. 62. Neves AA, Brindle KM. Assessing responses to cancer therapy using molecular imaging. Biochim Biophys Acta 2006;1766:242–61. 63. Brindle K. Ne w approaches for imaging tumour responses to treatment. Nat Rev Cancer 2008;8:1–14. 64. Stroobants S, Goeminne J , See gers M, et al. (18)FDG-P ositron emission tomo graphy for the earl y prediction of response in advanced soft tissue sarcoma treated with imatinib mesylate (Glivec). Eur J Cancer 2003;39: 2012–20. 65. Yoshimoto M, Waki A, Yonekura Y, et al. Characterization of acetate metabolism in tumor cells in relation to cell proliferation: acetate metabolism in tumor cells. Nucl Med Biol 2001;28:117–22. 66. Oyama N, Miller TR, Dehdashti F, et al. C-11-acetate PET imaging of prostate cancer: detection of recur rent disease at PSA relapse. J Nucl Med 2003;44:549–55. 67. DeGrado TR, Coleman RE, Wang SY, et al. Synthesis and evaluation of F-18-labeled choline as an oncolo gic tracer for positron emission tomography: initial f indings in prostate cancer . Cancer Res 2001;61:110–7. 68. Galldiks N, Kracht LW, Burghaus L, et al. Use of C-11-methionine PET to monitor the effects of temozolomide chemotherapy in malignant gliomas. Eur J Nucl Med Mol Imaging 2006;33:516–24. 69. Meisamy S, Bolan PJ, Baker EH, et al. Neoadjuv ant chemotherapy of locally advanced breast cancer: predicting response with in vivo H1 MR spectroscopy—a pilot study. Radiology 2004;233:424–31.

70. Kurhanewicz J , Vigneron DB , Nelson SJ . Three-dimensional magnetic resonance spectroscopic imaging of brain and prostate cancer. Neoplasia 2000;2:166–89. 71. Aboagye EO, Bhujwalla ZM, Shungu DC, Glickson JD. Detection of tumour response to chemotherap y by 1H nuclear magnetic resonance spectroscop y: ef fect of 5-fluorouracil on lactate le vels in radiation-induced f ibrosarcoma I tumours. Cancer Res 1998;58:1063–7. 72. Poptani H, Bansal N , Graham RA, et al. Detecting earl y response to cyclophosphamide treatment of RIF-1 tumors using selecti ve multiple quantum spectroscop y (SelMQC) and dynamic contrast enhanced imaging. NMR Biomed 2003;16:102–11. 73. Golman K, Zandt RI, Lerche M, et al. Metabolic imaging by hyperpolarized 13C magnetic resonance imaging for in vi vo tumor diagnosis. Cancer Res 2006;66:10855–60. 74. Day SE, Kettunen MI, Gallagher FA, et al. Detecting tumor response to treatment using h yperpolarized 13C magnetic resonance imaging and spectroscopy. Nat Med 2007;13:1382–7. 75. Veech RL, Lawson JWR, Cornell NW, Krebs HA. Cytosolic phosphorylation potential. J Biol Chem 1979;254:6538–47. 76. Silverstein E, Bo yer PD. Equilibrium reaction rates and the mechanism of bo vine heart and rabbit muscle lactate deh ydrogenase. J Biol Chem 1964;239:3901–7. 77. Williams SNO, Anthony ML, Brindle KM. Induction of apoptosis in two mammalian cell lines results in increased le vels of fr uctose1,6-bisphosphate and CDP-choline as deter mined b y 31P MRS. Magn Reson Med 1998;40:411–20. 78. Moffat BA, Chenevert TL, Lawrence TS, et al. Functional dif fusion map: a noninvasive MRI biomarker for early stratification of clinical brain tumor response. Proc Natl Acad Sci U S A 2005; 102:5524–9. 79. Bode BP, Souba WW. Modulation of cellular proliferation alters glutamine transpor t and metabolism in human hepatoma cells. Ann Surg 1994;220:411–22. 80. Miller WJ, Baron RL, Dodd GD, Federle MP. Malignancies in patients with cir rhosis: CT sensiti vity and specif icity in 200 consecuti ve transplant patients. Radiology 1994;193:645–50. 81. Reitzer LJ, Wice BM, Kennell D. Evidence that glutamine, not sugar, is the major ener gy source for cultured HeLa cells. J Biol Chem 1979;254:2669–76. 82. Vesselle H, Grierson J, Muzi M, et al. In vi vo validation of 3 ʹ′ deoxy3ʹ′-[F-18]fluorothymidine ([F-18]FLT) as a proliferation imaging tracer in humans: Cor relation of [F-18]FL T uptak e b y positron emission tomography with Ki-67 immunohistochemistry and flow cytometry in human lung tumors. Clin Cancer Res 2002;8:3315–23. 83. Bhattacharya P, Chekmenev EY, Perman WH, et al. Towards hyperpolarized C-13-succinate imaging of brain cancer . J Magn Reson 2007;186:150–5. 84. Robertson S. Choline chloride. CAS 2004;No:67-48-1. 85. Schröder L, Lowery TJ, Hilty C, et al. Molecular imaging using a targeted magnetic resonance h yperpolarized biosensor . Science 2006;314:446–9.

26 MAGNETIC RESONANCE IMAGING AGENTS ELISENDA RODRIGUEZ VARGAS, PHD AND JOHN W. CHEN, MD, PHD

More than 60 million clinical imaging procedures are now performed worldwide each year with magnetic resonance imaging (MRI), and emer ging preclinical and clinical fusion technologies are becoming available. The main advantage of MRI is that no ionizing radiation or radiopharmaceutical agent is used. Instead, MRI detects the magnetic signals from nuclei (mainl y protons). Unlike other tomo graphic modalities, MRI allo ws an y arbitrary plane of imaging. High-resolution images with excellent soft tissue contrast betw een dif ferent tissues can be used to assess anatom y at excellent spatial resolution. Despite these attracti ve parameters, the main obstacle to the use of MRI as a molecular imaging modality is its relati vely lo w sensiti vity to molecular probes compared with nuclear methods, such as positron emission tomography (PET) and single-photon emission computed tomography (SPECT). This chapter reviews k ey MRI molecular agents and co vers cur rent strategies on developing MRI agents not only with high molecular and cellular specif icity but also with improved sensiti vity o ver con ventional MRI agents. In addition, w e also re view the clinicall y promising magnetic nanopar ticle agents. For more in-depth information on magnetic nanopar ticles, see Chapter 33 “Theranostics: Agents for Diagnosis and Therapy,” Chapter 34 “Magnetic Nanopar ticles,” and Chapter 35 “Fluorocarbon Agents for Quantitati ve Multimodal Molecular Imaging and Targeted Therapeutics.” The ph ysical principles of magnetic resonance are based upon the nuclear magnetic resonance (NMR) phenomenon, disco vered b y Purcell and colleagues 1 and Bloch and colleagues 2 in 1946, for w hich the y w ere awarded the 1952 Nobel Prize. In the presence of a strong magnetic f ield, protons rotate with a frequenc y that is dependent on the strength of the magnetic f ield resulting in a net macroscopic magnetization. The net magnetization aligns parallel with the applied magnetic

field and can be perturbed when a radio frequency pulse applied is the same as the protons resonant frequenc y (resonance phenomenon). The component of the magnetization relax es back to its equilibrium v alue with an exponential time constant, T1, the longitudinal (or spinlattice) relaxation time. The predominant mechanism of the T 1 relaxation is through dipole-dipole interactions between the protons nuclei in the sur rounding media. The broad frequenc y range of the media (lattice) will cover the resonance frequenc y of the nuclei and stimulate a transition back to the equilibrium. The time dependence of the magnetization per pendicular to the xy-plane is characterized similarl y by T2, the transv erse (or spin-spin) relaxation time, w hich measures the time for the decay of the transverse magnetization to its equilibrium value of zero. P erturbations to nuclei that cause changes in magnetic resonance (MR) frequenc y will cause the loss of the net transversal magnetization. These perturbations arise from a “spin-spin” e xchange on energy betw een dipolar proton nuclei. The spins will dephase if there are bulk inhomogeneities in the external magnetic field.3,4

PHYSICS OF MRI AGENTS The use of imaging agents has become an integral part of MRI for many applications because the use of exogenous imaging agents allo ws better contrast to be obtained between patholo gic and health y tissues. About 35% of the MRI e xaminations make use of imaging agents, but this percentage is likely to increase further following the development of more ef fective and specif ic contrast media than those cur rently a vailable. Unlik e contrast agents used in X-ra y computed tomo graphy and in nuclear medicine, MRI agents are not directly visualized in the image. Onl y their ef fects are obser ved: increased contrast is af fected b y the ef fect the imaging agent 389

390

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

causes on protons relaxation times and consequentl y on the intensity of the MR signal. In addition to what is covered in this chapter , e xcellent re views are a vailable, which illustrate extensively the relaxivity mechanisms of the MRI agents. 4–7 MRI agents can be di vided into tw o main g roups: positive or ne gative agents, depending on w hether the contrast mechanism predominatel y af fects the T1-weighted or T2-weighted signal of the protons. P ositive imaging agents increase the T1-weighted signal on the images by shortening T1 (as well as T2, although the effect of the T2 shortening is usuall y not apparent on imaging unless v ery high local concentration is achieved, such as in the kidneys) and are predominately paramagnetic comple xes. Ne gative imaging agents decrease the signal on T2-weighted imaging by causing a large shortening of the proton T2 and are exemplified by superparamagnetic iron o xide (SPIO) nanopar ticles. The ability of an agent to af fect T1 and T2 is characterized b y the relaxa vities, r 1 and r 2, respecti vely. These parameters represent the rates 1/T 1 and 1/T 2, respectively, nor malized to concentration, are nor mally expressed as a rate (mM −1s−1) (Eq. 1). (1/Ti)obs = (1/Ti)d + ri[M]

i = 1, 2,

(1)

where (1/T i)obs is the obser ved solvent relaxation rate in the presence of an MRI agent, and (1/T i)d is the (diamagnetic) solv ent relaxation rate in the absence of an MRI agent. The ratio r 2/r1 is an indicator of the relaxometric properties of an imaging agent, and it is used to classify a given type of MRI agent as a T1 or T2 imaging agent. In general, for paramagnetic chelates, r 2/r 1 varies between 1 and 2, and for the superparamagnetic colloids, it can be as large as 50. 8

T1 Imaging Agents In the presence of a strong external magnetic field, the net magnetic moment of paramagnetic metal ions aligns along the magnetic f ield. The molecular motion of these paramagnetic substances can cause a fluctuating local magnetic field around protons as the paramagnets dif fuse nearby. There is a dipolar magnetic interaction betw een the electronic magnetic moment of the paramagnetic atom and the much smaller magnetic moment of the protons, predominately those of the w ater molecules. Random fluctuations in this dipolar magnetic interaction reduce both the longitudinal (T 1) and the transv erse (T 2) relaxation times of the protons. The choice of Gd(III) and Mn(II) as optimal relaxation agents stems from their long

electronic relaxation time and their lar ge magnetic moments.5,7 The total paramagnetic relaxation rate enhancement for the w ater protons (r 1) is due to inner sphere (r 1IS), second-sphere (r 1SS), and outer-sphere (r 1OS) relaxation mechanisms. For small-molecular weight paramagnetic agents, about half of the relaxi vity is due to an inner-sphere effect, where a w ater molecule binds in the inner coordination sphere of the paramagnetic ion and exchanges rapidl y with the bulk w ater protons (inner sphere relaxation). 4 The other half is due to w ater molecules h ydrogen binding to the chelate (second-sphere relaxation) or water molecule diffusion in the outer-sphere environment of the paramagnetic ion (outer -sphere relaxxt ation)9 (Figure 1A). As will be discussed in the ne section, the outer -sphere relaxation mechanism is solel y responsible for determining the relaxivity of SPIO.10 The inner-sphere relaxivity contribution is directly proportional to the molar concentration of the paramagnetic complex, [C], to the number of w ater molecules coordinated to the paramagnetic center , q, and in versely proportional to the sum of the mean residence lifetime, τM, of the coordinated water protons and their relaxation time, T1M: r1 = q[C]/ 55.5(T 1M + τM).

(2)

The Solomon-Bloembergen-Morgan equations illustrate the main parameters that modulate T1M.11–13 At the magnetic f ield strengths typicall y used in MRI (0.5 to 1.5 T, 20 to 60 MHz), the longitudinal relaxation time of the bound water protons T1M is dominated by the molecular reorientation (rotational cor relation) time τR. The slower the paramagnetic complex tumbles, the faster the relaxation rate. For activatable T1 agents, the parameters q and τR are the most amenab le to modulation in response to a biologic event to generate activatable molecular imaging agents, while for chemical exchange saturation transfer (CEST) agents, the parameter τM is important (Figure 1A).4

T1 Imaging Agents, First Generation Although some manganese-containing imaging agents are commerciall y impor tant, the main MRI agents are based on gadolinium (Gd) comple xes. Gadolinium does not have any known physiologic function in mammalians, and its administration as free ions [Gd(III)] is strongl y toxic even at low doses (10 to 20 µmol kg −1). Therefore, to avoid the release of free gadolinium ion in vi vo, it is necessary to use tight chelating str uctures that form very stable comple xes. In par ticular, some pol yaminocarboxylic acids, namely, the linear diethylenetriamine pentaacetic acid (DTP A) and the macroc yclic 1,4,7,

Magnetic Resonance Imaging Agents

10-tetraazacyclododecane-N,Nʹ′,Nʹ′ʹ′,Nʹ′ʹ′ʹ′-tetraacetic acid (DOTA), for m v ery stab le comple xes with Gd(III). 6 As will be shown later in this chapter, using chelating moieties not only allows stable complexes of Gd for in vivo administration, the modif ication of the chelating str ucture is also the primar y means to achie ve increased molecular specificity. Of the six clinicall y appro ved imaging agents used worldwide for intra venous administration, four agents are based upon on Gd(III). The f irst imaging agent appro ved for clinical use w as the anionic Gd(DTP A)2− (Magnevist®; Schering AG, Ger many) that, in more than 10 years of clinical e xperience, has been administered to more than 20 million patients. Other Gd(III)-based

A

B B

imaging agents similar to Magne vist became soon available: Gd(DO TA)− (Dotarem®; GE Health) and Gd(HPDO3A) (Prohance ®; Bracco Imaging, Ital y) (Figure 1B). These first-generation imaging agents have been very useful, and have contributed to the rise of MRI utilization in clinical and preclinical imaging. They distribute mainly into the intra vascular and interstitial space, and therefore are nonspecific.14 As such these agents cannot cross intact blood-brain barrier (BBB) and have become useful to identify lesions that cause disr uption of the BBB (eg, tumors). Two deri vatives of Gd-DTP A with increased molecular affinity ha ve been subsequentl y introduced: Gd(EOBDTPA)2− (Eovist®; Schering AG) and Gd(BOPT A)2− (Multihance®; Bracco Imaging). Both agents are

COO2 2 OOC

N

N

COO2 N 2 OOC

Gd31

Outer sphere

N

2 OOC

N

COO2

COO2

N Gd31

COO2

® Gd-DTPA (MAGNEVIST )

COO2 2 OOC

N

N

COO2

Gd 31

τR

Gd31

T1M

N 2 OOC

OH

N

2 OOC

COO2

N

(H 3 C) 2 NOC

Gd-HPDO3A (PROHANCE ® )

q, τM

N Gd31

N

CON(CH3)2

® Gd-DTPA-BMA (OMNISCAN )

Inner sphere

O COO2

COO2

O

N

N 2 OOC

N

N

2 OOC

Second sphere

COO2

2 OOC

Gd 31

D D

HOOC

N

2 OOC

COO

Gd-BOPTA (MULTIHANCE ® )

C

N

2 OOC

Gd-DOTA (DOTAREM ® )

H2O

391

N

COO2

Gd31 COO2

Gd-EOB-DTPA (EOVIST ® )

HNOC

COOH

CONH N

N H

Ln(III)

kCX

O

O H

H N HOOC

H

N

HNOC

CONH

COOH

H DSI

Dv

H

ppm

ppm

Figure 1. Magnetic resonance imaging (MRI) agents. A, Schematic representation of the relaxometric mechanisms for MRI agents. B, Structures of the clinically approved six Gd(III)-based MRI agents. C, Physical characterization of superparamagnetic iron oxide (SPIO): electron micrograph of hexagonal SPIO (Panels A and B) and molecular model of surface-bound 10-kD dextrans (Panels C) (Modified with permission from Harisinghani et al.73). D, Schematic representation of a chemical exchange saturation transfer (CEST) agent (Ln(III)-DOTAGly) and of a CEST mechanism.

392

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

characterized by an increased lipophilicity due to the introduction of an aromatic substituent on the carbon backbone of the DTP A ligand , resulting in an alteration of the pharmokinetics and the biodistribution of these imaging agents as compared with the parent agent Gd-DTPA.15 Mn(II) (f ive unpaired electrons) has long been used as a relaxation agent. MnCl 2 is the acti ve component of Lumenhance, a commercial contrast agent used for gastrointestinal imaging. 16 Actually, Mn-dip yridoxyl diphosphate, Telescan, has entered the clinical practice and is recommended as a hepatotropic agent. 17

T2 Imaging Agents T2 imaging agents are primaril y based on SPIO (see Chapter 34, “Magnetic Nanoparticles”).18,19 SPIO particles consist of either a Fe3O4 (magnetite) or Fe2O3 (maghemite) core or a mixture with a nanometric size (both for mulas present very similar magnetic proper ties). Superparamagnetism is a phenomenon b y w hich par ticulate materials may e xhibit similar beha vior to paramagnetic materials. Superparamagnetism occurs when the small particles (1 to 10 nm) of unpaired spins are suf ficiently lar ge that the y can be re garded as ther modynamically independent, single-domain par ticles. In the presence of an e xternal magnetic f ield, magnetic spins of these domains are aligned to result in a net positive magnetic moment, which is much g reater than that of a paramagnetic substance. Unlike fer romagnetic materials, w hen the e xternal magnetic field is removed, the magnetic moments are randomly oriented and the net magnetic moment becomes zero, ensuring no self-aggregation between SPIO.20 Typical relaxivities of SPIO nanopar ticles measured at 37 °C and 0.47 T are r 2 of 100 mM −1s−1 and r 1 of 30 mM −1s−1 (r2/r1 > 3 to 4). Se veral parameters af fect

relaxometric beha vior of SPIO nanopar ticles, including particle size and composition, concentration of par ticles within a gi ven imaging v oxel, and data acquisition parameters. The magnetic f ield applied has also a nonlinear influence on the signal. The magnetic core of the SPIO sizes is generally between 4 and 10 nm coated with a biocompatible coating material, thus per mitting safe intravenous administration and ready biode gradability. Because of its small size, SPIO nanopar ticles easil y degrade into paramagnetic for ms of iron. The iron then enters the plasma iron pool and is subsequentl y incor porated into red cell production and other natural uses of iron. The iron will ultimatel y secrete from the body as the body’s iron stores tur n o ver. The amount of iron o xide required for clinical MRI is small compared with ph ysiologic iron stores. Furthermore, extensive toxicity studies in animals have not found acute to xicity or chronic injur y at doses greater than 100 times the clinical effective dose.21,22

T2 Imaging Agents, First Generation Table 1 illustrates the properties and the applications for several commercial SPIO agents. 23,24 The physical properties of the ne w generation of SPIO , monocr ystalline iron oxide nanoparticles (MION), and cross-link ed iron oxide (CLIO) that are being used as the platfor m for target-specific MRI are summarized below. MION

MION differs from other iron oxide preparations in their monocrystallinity. While MION nanopar ticles onl y contains one central iron o xide and despite their very small size (2.8 ± 9 nm), MION still e xhibits superparamagnetic proper ties. The small size of MION

Table 1. COMMERCIALLY AVAILABLE IRON OXIDE NANOPARTICLES AGENTS Agent

Classification

Trade Name-Company

Status

Size

Target/Use

Approved Approved Approved Approved Approved Approved (Eu) Phase 3 (Us) Approved (therapy) Phase 2 (imaging) Phase 3

3.5 µm

GI lumen

Ferumoxsil

Oral SPIO

Ferristene Ferumoxides

Oral SPIO SSPIO

Ferucarbotran

SSPIO

Gastromark–AMAG Luminirem–Guerbet Abdoscan–Amersham Feridex–Berlex Endorem–Guerbet Resovist–Schering

Ferumoxytol

USPIO

Feraheme–AMAG

Ferumoxtran

USPIO

Feruglose

USPIO

Combidex–AMAG Sinerem–Guerbet Clariscan–Nycomed

VSOP-C184

VSPIO

VSOP-C184–Ferropharm

Phase 3 (discontinued) Phase 1–2

300 nm 80–150 nm 62 nm

GI lumen Liver/spleen

17–31 nm

Liver/spleen Perfusion, MRA Iron replacement, MRA

20–40 nm

Lymph nodes, MRA

20 nm 7 nm

Perfusion MRA MRA

GI = gastrointestinal; MRA = magnetic resonance angiography; SPIO = superparamagnetic iron oxide agents include oral (large) SPIO agents; SSPIO = standard SPIO agents; USPIO = ultrasmall SPIO agents; VSPIO = very small PIO agents.

Magnetic Resonance Imaging Agents

allows it to easil y pass through capillar y endothelium 25 (Figure 1C). In contrast to Gd-DTP A, w here relati vely high local concentrations are needed to signif icantly reduce proton relaxation times, MION pro vides higher magnetic susceptibility (tw o to three orders of magnitude) than gadolinium, allo wing in vi vo detection at concentration as low as 1 µg Fe/g tissue. CLIO

CLIO nanopar ticles consist of MION nanopar ticles reacted with epichlorohydrin, a cross-linking agent which connects the par tly “free floating” de xtran chain around the iron core and ammonia. 26 The iron in the resulting CLIO becomes completel y caged , and as the amino groups of the de xtran chains become functionalized and nucleophilic, the y can be used for fur ther conjugation chemistry—this base compound is refer red to as CLIONH2. Man y chemical modif ications to the CLIO ha ve been pub lished and some of their biolo gic application will be discussed later in the chapter.

CEST Agents A new class of imaging agents has been recentl y introduced based on CEST. In the initial experiments, it was also refer red to as saturation transfer or magnetization transfer.27 CEST mechanism is based on the chemical exchange of a nucleus in the NMR e xperiment from one site to a chemically different site. Proton signals of exchangeable hydrogen atoms are onl y visible at their expected resonance frequenc y if e xchange is slo w on the timescale of the NMR measurement (exchange rate constant (1/ τM or k ex) < frequenc y separation betw een the proton signals of water and the protons of the CEST agent ( Δω)). If this condition is satisf ied when a presaturation pulse is applied to the CEST protons, the bulk magnetization will be reduced for the CEST agent causing a reduction in the signal intensity of the w ater protons (Figure 1D). Since the signal intensity of water protons is reduced, CEST agents act as ne gative imaging agents on the MR images. A number of sugars, metabolites, amino acids, and other small diamagnetic molecules possessing exchangeable protons w ere e xplored as suitab le CEST agents. However, the main dra wback of using diamagnetic molecules is that the concentration of the agent required is quite high. 27 A new generation of CEST agents ha ve been de veloped using paramagnetic lanthanide comple xes, referred

393

to as P ARACEST agents. 27–29 As a consequence of unpaired electrons, PARACEST agents induce lar ge Δω values. Ho wever, the rate of w ater e xchange in typical paramagnetic comple xes discussed abo ve e xceeds that allowed by the slo w exchange condition, thus e xcluding traditional T1 imaging agents as CEST agents. Recentl y, Zhang and colleagues 30 showed that a par ticular useful source of highly shifted exchangeable protons, ~50 ppm, can be provided by the slowly exchanging water protons bound to a paramagnetic Eu(III)-chelate. An alter native source for highly Δω may be provided by the amide protons of the paramagnetic chelate (ie, DO TAM-Gly, a tetraglycineamide derivate of DO TA). The rate of w ater exchange for comple xes with tetramide deri vatives of DOTA is slo w enough to allo w the w ater-bound resonance to be observed as a separate resonance.31 This class of agents sho w g reat promise as pH sensing agents and will be discussed later in this chapter. It is possible to design highly sensitive CEST agents by mo ving be yond the small molecular comple xes toward supramolecular str uctures. An optimal nanosystem for developing highly sensitive CEST agents is represented by liposomes. In such systems, the resonance of the w ater protons trapped within the liposome is shifted from that of the b ulk w ater, gi ving rise to tw o distinct w ater protons signals in the NMR spectr um. This new type of assembly, known as LIPOCEST agent, can help visualize tumors b y making use of the biodistribution properties of liposomes. 31

SMART PARAMAGNETIC IMAGING AGENTS MRI agents are relatively insensitive, requiring a minimum threshold of 0.01 to 10 mM for adequate detection during in vivo applications. For comparison, in vivo PET, SPECT, and fluorescence imaging typicall y have detection sensitivity thresholds of picomolar concentrations and are more practical for detecting intracellular tar gets. To generate detectable contrast in MR images, high local concentrations of the imaging agent, as well as high efficiency (high relaxivity values), at the target site are required. Several approaches ha ve been de veloped to improve the agent sensiti vity to gether with a relati ve nonspecificity of the f irst generation MRI agents. Large pub lications of tissue-specif ic based on superparamagnetic nanopar ticles that tar get l ymph nodes, li ver, atherosclerotic plaques or accumulate in the cells of reticuloendothelial system (RES) have been reported14 (see ne xt section). Another strate gy is to

394

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

conjugate the paramagnetic comple xes to antibodies. The f irst repor ts on the use of Gd-DTP A-labeled antibodies appeared in the mid 1980s and described the attachment of fe w chelates per antibody as, for instance, contrast based on biotin-(strept)avidin.32 Currently, MRI agent design is focused on developing agents that can be responsive to a change in physiologic en vironment or acti vity. These “smar t imaging agents” possess the ability to alter their str ucture in response to changes in the local en vironment, usuall y resulting in an increase in relaxi vity.33,34 As has been presented in the last section the most amenable parameters to modulate relaxi vity v alues w ere the number of bound water molecules (q), the rotational tumbling time (τR), and the proton exchange rate (τM). Main targets of this new class of environment-sensing, smart MRI agent are enzymes, pH, and metabolites. Table 2 summarizes this class of agents.

Agents That Are Catalyzed by Enzymes Enzymes represent an impor tant class of potential imaging biomark ers because the y are in volved in nearl y all facets of the biochemical interactions in ph ysiologic and pathologic conditions. Other important advantages of targeting enzymes are (1) high reaction rates: a low enzyme concentration can be suf ficient to modify enough MRI agents to result in a large change in the MR signal and (2) reaction specif icity: enzymes ha ve a high de gree of specificity with respect to both the identities of their substrates and products so resultant changes in MR signal could be confidently attributed to the specific biomarker.

Activation by Enzyme Cleavage

This strategy is based on the incor poration of a masking group that renders the imaging agents relatively MR silent. These masking g roups are carefull y chosen to act as enzyme substrates, and in the presence of a par ticular enzyme, they are cleaved to yield an increase in relaxivity. Enzymatic activation by cleavage was first introduced in a pioneering work by Moats and colleagues 35 in 1997. They designed a gadolinium-based comple x consisting of a galactopyranose moiety positioned in the ninth coordination position of the gadolinium complex. In the presence of β-galactosidase, this b locking sugar g roup is remo ved by enzymatic cleavage, and the T1 relaxivity of the imaging agent increased 20% because of the improved access of the water molecules to the metal after enzymatic clea vage. A later chemical modification called EgadMe introduced a α-methyl group to one ethylenic carbon on the sugar linkage arm of the tetraazacarboxylic macrocycle. This change resulted in a 200% increase of relaxivity.36 To test the ability for detecting in vivo transcription and translation, Louie and colleagues36 microinjected one of the cells of Xenopus embryos at the two-cell stage with a deoxyribonucleic acid construct car rying the lacZ gene after both b lastomeres had been injected with EgadMe. LacZ is the gene encoding β-galactosidase. EgadMe per mits MRI detection of lacZ gene expression just on the side of the embr yo engineered to e xpress β-galactosidase (F igure 2A). The MR image of a li ve embryo correlated well with the image of the same embryo after f ixation and staining with X-gal. In addition to modifying the number of protons that can access the paramagnetic metal center , changes in the rotational tumb ling time ( τR) can also modify the

Table 2. SMART PARAMAGNETIC AGENTS Agent

Target

Parameter changed

Reference

Enzyme-sensitive agents EgadMe Gd-DTPA-(lys)3-HSA Gd-DTPA-FA Gd-5-HT-DOTA

β-galactosidase TAFI lipase Myeloperoxidase

q τR Solubility/q τR

Moats et al.,35 Louie et al36 Hanaoka et al.38 Himmelreich et al.39 Bogdanov et al.,40 Chen et al.,41,42 Bradley et al.,43 Heinecke,44 Querol et al.45

pH-sensitive agents Gd-DOTA-4-AmP5− Gd-PAMAM-EPTPA Yb-DOTAM-Gly

pH pH pH

τM τR CEST

Zhang et al.61 Laus et al.63 Aime et al.29

Agents sensitive to metabolites [Yb(MBDO3AM)]3+ Eu-DOTA-tetraamide-bis(phenylboronate) Perfluorocarbon-Eu3+

Lactate Glucose Fibrin

CEST CEST CEST

Aime et al.68 Zhang et al.69 Winter et al.71

Gd-bis-5-HT-DTPA

CEST = chemical exchange saturation transfer; DTPA = diethylenetriamine pentaacetic acid; DOTA = 1,4,7,10-tetraazacyclododecaneN,Nʹ′,Nʹ′ʹ′,Nʹ′ʹ′ʹ′-tetraacetic acid; EPTPA = ethylenepropylenetriamine-N,N,Nʹ′,Nʹ′ʹ′,Nʹ′ʹ′-pentaacetic acid; HSA = human serum albumin; PAMAM = polyamidoamine; TAFI = thrombin-activatable fibrinolysis inhibitor.

Magnetic Resonance Imaging Agents

395

B B

A HO

OH

R

O

O HO

NH3

NH3

H

O

O

N H

H

OH

O

O

H N

N H

NH O Gd-DTPA H3 N

R

O O

TAFI

O

Lysine

NH

O

H N

N H

O

1 5 R

2 I

H

N

O

O O O

OH

5, 6, 7, 8 I

C C

R

O TAFI Lysine

O

O

Gd-DTPA

1, 2, 3, 4

E

O

N H

2 6 Compound

Gd3+

O

Gd-DTPA H3 N

H3 N

-Gal Gd3+

O

H N

N NH H O Gd-DTPA H3 N 3 7

H N

N

N O O Gd O OO O O H H

R

O

O O

TAFI Lysine

O

N NH H

O O

Gd-DTPA

4 8

H D C

E

Figure 2. Activation by enzyme cleavage. A, Schematic representation of the activation of EgadMe after cleavage of the sugar residue by β-galactosidase allowing the coordination of a water molecule to the Gd(III) metal. On the bottom, MR image of a live Xenopus embryo injected in the left side at the two-cell stage with the lacZ gene and subsequent enzyme expression just on the left side of the embryo. Regions of high signal intensity are found in the bright stripe of endoderm (E), regions of the head (H), and ventrally, including two distinct spots (red arrows) found just ventral to the cement gland (C). Bright-field image of same embryo fixed and stained with X-gal correlating with the regions of high intensity in the MR image (Reproduced with permission from Louie et al.36). B, Activation of the Gd(III) complex after the cleavage of the masking group by thrombin-activatable fibrinolysis inhibitor. Enzyme activation releases the shielding group and allows human serum albumin binding (Reproduced with permission from Nivorozhkin et al.37).

relaxometric properties of the imaging agent. In particular, the binding of the agents to a macromolecule can substantially slow the molecular rotation of the Gd 3+-complex, resulting in an additional increase in T1 relaxivity. This phenomenon is known as receptor-induced magnetization enhancement (RIME). RIME agents ha ve been reported b y Lauf fer and colleagues 37 who prepared a Gd(III) chelate containing a phosphoric ester sensiti ve to the attack of the ser um alkaline phosphatase. The hydrolysis e xposes a h ydrophobic moiety capab le of binding to human serum albumin (HSA). Upon binding, there is an increase in the T1 relaxivity as a consequence of the lengthening of τR. Similarly, a l ysine-containing ligand acting as a masking g roup can be clea ved b y thrombin-activatable f ibrinolysis inhibitor , also kno wn as carbo xypeptidase B (F igure 2B). The con version involved enzymatic remo val from the Gd 3+-complex masking g roups to e xpose binding moieties with HSA

affinity, again resulting in an increase in rotational tumbling time. Recentl y, a no vel β-galactosidase-activated MRI agent with an albumin-binding moiety based on the RIME approach has been also pub lished. In the presence of β-galactosidase, the cleavage of a galactop yranose moiety exposes an HSA-binding domain, gi ving rise to a 57% increase in the r 1 relaxivity.38 The rotational tumbling time of an imaging agent can also be changed either by polymerizing monomeric agents or by degradation of a pol ymer. These strategies will be discussed in the next section. Solubility of the imaging agent has been e xploited to develop real “switch on/off ” agents. Water insolubility of Gd-DTPA can be achie ved b y conjugating h ydrophobic moieties, such as long aliphatic chains. These aliphatic moieties can be cleaved by lipases to increase the solubility of the clea ved product. Aime and colleagues 39 developed and tested a modif ied Gd-DTPA complex with two

396

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

long fatty acid chains for conditional activation by lipase. As expected, the r 1 relaxation rate of the inacti ve, insoluble Gd-DTPA-FA complex was 0 mM −1s−1. The addition of an esterase resulted in an increase of r1 to 4.7 mM−1s−1. An in vitro study w as car ried out in dendritic cells, w hich express high le vels of lipoprotein lipase The phagoc ytic uptake of the Gd-complex w as tested b y incubation for 2 to 24 hours sho wing that the maximum r 1 relaxation rate was reached within 9 hours of incubation. Due to the toxicity studies, no lar ger concentrations than 20 mM were tested. In addition, the insolubility of the inacti ve compound renders intravenous injections impossible.

substrate molecules (*A) w ould then self-pol ymerize (oligomerize) into the lar ger, higher-relaxivity paramagnetic oligomers (Eq. 4): 2AH + [E, H 2 O2] → 2[*A] + 2H 2O + E,

(3)

n[*A] → [A]n.

(4)

Among the biolo gic substrates, use for pero xidases in vivo (ie, Cl −, tyrosine, NO 2−), phenolic compounds are attractive because it can act as the electron donor in the above reaction and become oligomers as the result. Bo gdanov and colleagues40 reported in 2002 the first MR agent capable of acting as an electron donor substrate for horseradish peroxidase by conjugating h ydroxytyramine to GdDOTA. Relaxivity values increased more than 100% after the addition of the peroxidase. Analysis of reaction products by size-exclusion high-performance liquid chromatography and mass spectrometry confirmed condensation of oxidized substrates through the phenolic radical moieties. Moreover, nanograms of peroxidase were detectable by MRI. Horseradish peroxidase, a plant enzyme, is not present in animals and humans. Chen and colleagues42 later reported modifications using serotonin moieties to achie ve sensitivity t o a biologically rele vant enzyme, m yeloperoxidase (MPO) (Figure 3A). MPO is one of the most ab undant

Activation by Enzyme Modification

Instead of clea vage to e xpose the acti ve compound , the enzyme modifies the parent compound into a different molecule. This class of enzymatic acti vation is best illustrated by peroxidase, which amplif ies the T1-weighted signal b y causing the parent compound to polymerize, resulting in an increase in the rotational cor relation time of the acti vated compound compared with the parent compound. 40,41 Peroxidase (E) could catalyze the reduction of hydrogen pero xide using a lo w-relaxivity paramagnetic substrate (AH) as a donor of electrons (Eq. 3). Oxidized

A

C OH H N

H N O

CO

HN

OC N O OC

N Gd(III) O N N O

CONH

HNOC NH N H

HO

O

5-HT-DOTA(Gd)

N HOOC

N

N

COOH COOH Gd(III) bis-5-HT-DTPA(Gd)

OH

S

1

R1~9 mM

1

R1~ 4.5 mM

1

MPO Radical formation and oligomerization

B B

S

1

Protein Protein binding

5-HT-DOTA(Gd)

bis-5-HT-DTPA(Gd)

R1~ 25 mM

1

S

1

Figure 3. Agents activatable by enzyme modification (polymerization). A, Myeloperoxidase (MPO) sensitive agents. B, Intravenous injection of these agents in a mouse model with one flank implanted with human MPO demonstrated about 2X increase in contrast-to-noiseratio. C, Agent mechanism: without MPO, the agents have similar relaxivity as those of Gd-diethylenetriamine pentaacetic acid and Gd-1,4,7, 10-tetraazacyclododecane-N,Nʹ′,Nʹ′ʹ′,Nʹ′ʹ′ʹ′-tetraacetic acid, but in the presence of MPO, the agent is radicalized resulting in oligomers and can bind to proteins to increase the relaxivity and tissue retention (Modified with permission from Chen et al.41).

Magnetic Resonance Imaging Agents

enzymes secreted b y inflammator y cells, 43 including neutrophils, macrophages, and micro glia. It is important in many patholo gic diseases, including rheumatoid ar thritis, atherosclerosis, multiple sclerosis (MS), and Alzheimer’s disease, to name a fe w. MPO can generate highl y reactive molecular moieties, such as h ypochlorite, tyrosyl radicals, and aldehydes and can cause local damage that fur ther activate the inflammatory cascade.44 This activation approach was verified in vivo by an implant model using Matrigel ™ to embed human MPO in the flanks of mice (F igure 3B). 41 Unlike pre viously discussed smart agents, the deli very of the MPO sensor in vivo w as possib le via intra venous injections, important for translational studies. Matrigel ™ is a basement membrane extract obtained from mouse sarcoma and has a unique proper ty of being in the aqueous phase at lo w temperature (~5°C) but becoming a gel at body temperature. It allo ws a per meable and dynamic reser voir within the interstitium, to allo w the deli very of agent while keeping the enzyme trapped in the implant. MPO sensitive agents demonstrated 60 to 100% increase in contrast-to-noise-ratio (CNR) upon activation by MPO. Physicochemical characterizations re vealed that MPO radicalizes this class of agents, and the agents can: (1) oligomerize as discussed abo ve and (2) bind to proteins via co valent bond for mation with aromatic amino acids present in the protein 41,45 (Figure 3C). These properties not onl y increase high relaxi vity of the acti vated probe, confer ring higher sensiti vity to the detection of inflammation, but also prolong pharmacokinetics in vivo, allowing confirmation of MPO activity and inflammation on delayed images in areas with prolonged enhancement. Despite the binding to proteins, the acti vated agents are washed out of the sites of inflammation within 6 hours after administration, likely digested and released b y proteases that are present in sites of inflammation. Several in vivo applications of this class of acti vatable agents are presented here. Myocardial Infarction The specificity of the MPO sensor w as v erified in mouse m yocardial inf arction model.46 Ischemic injur y to the m yocardium causes recruitment of neutrophils and macrophages that secrete MPO. In wild-type mice, there w as signif icant prolonged enhancement lasting se veral hours in the infarcted myocardium when imaged with the MPO sensor. However, when imaged with con ventional nonspecific Gd-DTPA, by 1 hour after injection, the CNR has already retur ned to baseline. Fur thermore, in MPO knockout mice without MPO expression, there was very little enhancement in the infarct. In heterozygous MPOdeficient mice with inter mediate MPO e xpression,

397

there w as cor respondingly inter mediate enhancement (Figure 4A). These findings confirm that the MPO sensor is highl y specif ic to MPO , but not to other pero xidases or enzymes. Furthermore, MPO imaging was able to track the inflammator y state of an inf arct o vertime and to follow treatment response from statins. Atherosclerosis Atherosclerotic plaque for mation is the leading cause of morbidity and mor tality in de veloped countries. Plaques become rele vant to a patient’ s health when they destabilize and become “vulnerab le” to r upturing, leading to thromboembolic e vents. An imaging technique capable of detecting “vulnerable” plaques will enable risk stratification and focus treatment strategies prior to the onset of an acute e vent. There is a well-documented correlation between inflammation, o xidative stress, and pathogenesis of atherosclerosis in humans. 47 Advanced human atherosclerotic plaque contains high numbers of macrophages e xpressing and acti vely secreting MPO .47,48 MPO is capable of generating other highl y reactive molecular species that par ticipate in co valent modif ication (including o xidation) of lo w-density lipoprotein. 49 MPO activity has been linked to the activation of matrix metalloproteinase-7 proenzyme (matrilysin),50 potentially inducing plaque rupture. MPO also has been found to modify highdensity lipoprotein, altering its interaction with sca venger receptors and re versing its “protecti ve” role. 51,52 A recent study in more than 600 patients estab lished that a single MPO measurement in plasma could predict the risk of major adv erse cardiac e vents within the subsequent 6 months.53 The MPO sensor has been used in a rabbit model of atherosclerosis to identify the areas on atherosclerotic plaques high in MPO acti vity.54 In areas of increased MPO activity, there was a two-fold increase in the CNR at 2 hours after agent administrated compared with nor mal vessel w all or images obtained b y con ventional imaging with Gd-DTPA (Figure 4B). The areas most often identified are the shoulders of the plaques, w hich are the high-stress areas. By identifying the plaques with elevated MPO activity, this technique could identify those plaques that are highly inflamed and prone to r upture for early and focused treatment. Demyelinating Diseases Inflammatory demyelinating plaques are the pathologic hallmark of active MS and often precede clinical manifestations. Nonin vasive earl y detection of active plaques would thus be cr ucial in establishing presymptomatic diagnosis and could lead to earl y preventive treatment strate gies. Cur rently, acti ve disease is inferred from contrast enhancement identif ied on MRI. However, it is widely recognized that these MRI techniques have limitations because contrast enhancement reflects breakdown in the BBB with leakage of paramagnetic

398

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

50

E

CNR

D 25

DNA

protein MPO

155 bp

GAPDH

∗

MPO

+/+

MPO +/+

+/–

***

20

MPO imaging CNRDiff (CNR2hr-CNRBaseline)

macrophages

+/+ –/–

–/–

B DTPA(Gd)

+/– –/–

***

***

15

10

5

0

MPO

gad Control

MPO-gad Control

gad CH-fed

MPO-gad CH-fed

C Conventional imaging

MPO imaging

MPO

0.0075

40

3

Volume/lesion (cm )

Number of lesions

50

30 20 10

0.0050

0.0025

0.0000

0 MPO-Gd

DTPA(Gd)

MPO-Gd

DTPA(Gd)

Figure 4. Examples of myeloperoxidase (MPO) imaging. A, In a myocardial infarction mouse model, the MPO agent was able to differentiate between different levels of MPO expression (figure courtesy of Matthias Nahrendorf, MD). B, In a rabbit model of atherosclerosis, MPO imaging accurately identified areas of elevated MPO secretion, with an increase in contrast-to-noise-ratio of 2X compared with different controls (figure courtesy of John W. Chen, MD, PhD (Center for Molecular Imaging Research), Brian Rutt, PhD (Robarts Research Institute), and John Ronald (Robarts Research Institute)). C, In a mouse experimental autoimmune encephalomyelitis model for multiple sclerosis, MPO imaging revealed more and smaller lesions at an earlier time point than imaging with the nonspecific Gd-diethylenetriamine pentaacetic acid.

Magnetic Resonance Imaging Agents

chelates rather than acti ve inflammation, and the tw o may not al ways cor respond. In par ticular, MS lesions at all stages demonstrate some BBB breakdo wn,55 and lesions can remain enhanced 1 to 13 w eeks after the onset of clinical symptoms. 55,56 MPO is e xpressed in acti ve MS plaques.57 Individuals with higher MPO e xpression ha ve increased susceptibility to MS. 58,59 Using murine e xperimental autoimmune encephalom yelitis as a model of MS, Chen and colleagues 60 demonstrated that a MPO imaging sensor can detect and conf irm more, smaller , and earlier active inflammatory lesions in li ving mice by in vivo MRI and that MPO e xpression cor responded with areas of inflammatory cell inf iltration and dem yelination. Higher MPO acti vity as detected b y MPO imaging, biochemical assays, and histopatholo gic anal yses cor related with increased clinical disease se verity (F igure 4C). This approach could be used in longitudinal studies to identify active demyelinating plaques and to more accuratel y track disease course following treatment in clinical trials.

Smart CEST Agents pH-Sensitive Agents

An important motivation for accurate measurement of pH in vivo is in tumor imaging. Tumors tend to have a more acidic pH (typicall y pH 6.8 to 6.9) than healthy tissue (pH 7.4). Se veral methods to har ness pH-responsive imaging agent rely on hydratation and rotational correlation time changes. Another novel method involves the use of PARACEST compounds. Sherry and colleagues 61 in 1999 introduced a gadolinium-based pH-sensiti ve imaging agent, GdDOTA-4-AmP5−. The pH dependence of the relaxivity of this compound is associated with changes in the structure of the second hydration shell. The hydrogen bonding network created b y protonation of the phosphonates on the side ar ms of this comple x pro vides a pathw ay for exchange of the bound w ater protons with protons of bulk w ater. The relaxi vity of this comple x increases between pH 4 and 6, suggesting the prototropic exchange is maximized after the 3 to 4 phosphonate groups are protonated , then decreases up to pH 8.5, remains constant up to pH 10.5 and increases again above pH 10.5 because of shor tening of τM catalyzed by OH− ions. Recentl y, it has been demonstrated that this complex can be successfull y used in vi vo for mapping renal and systemic pH. 62 MRI agents can also e xperience a change in rotational cor relation time in response to a v ariation of pH. Three dif ferent generations of pol yamidoamine

399

(PAMAM) dendrimers ha ve been conjugated with ethylenepropylenetriamine-N,N,Nʹ′,Nʹ′ʹ′,Nʹ′ʹ′-pentaacetic acid.63 The pH dependence of the global rotational correlation time can be related to the protonation of the tertiary amine g roups in the P AMAM sk eleton b y increasing the repulsion betw een the positi vely charged nitrogen atoms, w hich leads to an e xpanded and more rigid dendrimeric str ucture at lo wer pH. The proton relaxivity increases b y 60% as its rigidity increases and thus a slo wer global rotation occurs w hen the pH becomes more acid. Another e xample is based on tw o water-soluble endohedral gadofullerenes, Gd@C60(OH)x and Gd@C 60[C(COOH)2]10. These comple xes possess high relaxi vities through a pH-dependent agg regation phenomenon64: between pH 3 to 12, as the pH decreases, the aggregation increases in size, increasing its rotational dynamics and consequently the T1 relaxivity. As has been discussed in a pre vious section, DOTAtetramide comple xes of lanthanides are characterized b y their slo w w ater e xchange rate, a proper ty that mak es them amenab le to CEST e xperiments.29 Aime and colleagues65 described the CEST properties of a series of lanthanide(III) comple xes (Ln = Eu, Dy , Ho, Er , Tm, Yb) with the macrocyclic DO TAM-Gly ligand a tetraglycineamide derivative of DO TA. These complexes possess two pools of exchangeable protons represented by the coordinated water and the amide protons. Among the metals studied, Yb-DOTAM-Gly showed the most interesting CEST proper ties w hen its amide N-H resonance (16 ppm upfield H2O signal) is irradiated. Up to 70% suppression of the w ater signal is obtained at pH 8. As the exchange of amide protons is base catal yzed, YbDOTAM-Gly is an ef ficient pH-responsi ve probe in the 5.5 to 8.1 pH range. Moreo ver, a ratiometric method has been developed, removing the dependence of the observed pH responsiveness from the absolute concentration of the paramagnetic agent. Using dual-imaging agents, a mixture of Eu-DOTAM-Gly and Yb-DOTAM-Gly was selectively irradiated. Irradiation of the e xchanging amide NH protons of the Yb-DOTAM-Gly and the metal-coordinated water proton of Eu-DO TAM-Gly generated a calibration curve (of the ratio of the Yb/Eu CEST ef fect vs pH). This allowed a concentration-independent method of pH measurement for this pair of coadministered imaging agents as the two sets of mobile protons have different pH dependencies. In a pioneering w ork, v an Zijl and colleagues 66 demonstrated that due to its lo w exchange rate betw een the hydrogen atoms of w ater and the amide, the amide protons of mobile proteins and peptides could be used as pH sensors. This e xchange depends on the h ydrogen

400

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

exchange rates, is base catal yzed for pH v alues abo ve 5 and is propor tional to the h ydroxyl ion concentration and exponentially proportional to pH. Gilad and colleagues 67 designed a no vel prototype CEST-based repor ter gene with promising applications in noninvasive cell tracking and the study of gene deli very and gene e xpression. L ysine rich-protein (LRP) reporter has been chosen as an MR-CEST repor ter owing to the rapid e xchange of the amide protons in polypeptides, such as poly-L-lysine (kex polylysine = 403 s−1 and k ex endogenous proteins = 28 s −1). In vitro cells expressing LRP present a signal-intensity change of 8.2% from control cells, conf irming LRP as a potential reporter for cell tracking. As a consequence of the f ast exchange of the amide protons in the LRP , CEST effect from the LRP can be dif ferentiated in vi vo from other amines present in multiple endo genous proteins in rat brain. Moreover, LRP is strongly pH-dependent because the amide proton e xchange is base catal yzed (the rate is reduced b y a f actor of ~10 per pH unit) so ma y be exploited in vivo to report the pH.

CEST Agents Sensitive to Metabolites

CEST agents can also be useful to detect metabolites that are present at high concentrations. Aime and colleagues68 studied the capability of heptadentate ligands to interact with anionic str uctures to monitor lactate concentration. In the presence of lactate, the anion binds the Ln(III) ion by replacing the metal-bound w ater molecules through the formation of ter nary adducts, resulting in changes of the resonance amide protons. To detect glucose concentration, the Eu-DO TA-tetraamide comple x has been conjugated with bis(phen ylboronate) ar ms because boronic acids can bind selecti vely and re versibly with structures containing cis-diols.69 Such binding altered the water exchange between the Eu 3+-bound water molecule and bulk w ater, and this interaction w as detectab le b y CEST imaging. This system was found to be sensiti ve to changes in glucose concentration over the range of physiologic interest (5 to 10 mM). Fibrin, a k ey component in thrombus for mation, can also be detected using MRI agents. Lanza and colleagues70 had pre viously described perfluorocarbon nanopar ticles using gadolinium chelates on the surface to detect fibrin by MRI. P ARACEST with Eu 3+ complexes functionalized with a phospholipid moiety ha ve also been incor porated onto the surface of perfluorocarbon nanoparticles.71 Detection of the f ibrin b y P ARACEST nanopar ticles with a diameter of 294 nm has been assayed in plasma clots.

Finally, gl ycogen le vels can be detected indirectl y through the water signal after partial saturation of the large number of OH − groups on the pol ymer.72 The potential of the glycoCEST as glycogen reporter agent has been studied in a mouse li ver during glucagon-initiated breakdo wn of gl ycogen. Gl ycoCEST images illustrate depletion of glycogen with time and may be useful to infor m about the glycogen localization in the liver.

IRON OXIDE NANOPARTICLES FOR MOLECULAR IMAGING Iron oxide nanopar ticles are used as repor ters for man y physiologic processes and ha ve impor tant biolo gic and clinical applications, which are described below.

Lymph Node Imaging The presence and location of lymph node metastasis are extremely important in tumor staging. However, current clinical imaging diagnosis of l ymph node metastases is based on imperfect size criteria, and fur thermore, micrometastases cannot be detected b y con ventional imaging methods. Metastatic nodes v ary in diameter , and signal intensities of metastatic nodes do not dif fer from those of nor mal nodes. Harisinghani and colxide leagues73 in a lar ge clinical trial used iron o nanoparticles for metastatic l ymph node detection in humans. After intravenous administration, the nanoparticles (~30 nm) are slowly extravasated from the vascular space into the interstitial space, from w hich they are transported to lymph nodes by way of lymphatic vessels. Within the l ymph nodes, these super paramagnetic nanoparticles are inter nalized b y macrophages, and these intracellular iron-containing par ticles cause changes in magnetic proper ties detectab le b y MRI (Figure 5A). In nor mal l ymph nodes, the MR signal intensity decreases after the administration of the nanoparticles. In contradistinction, in metastatic nodes, the lack of a full y functional RES does not allo w the uptake of the iron o xide nanoparticles, and these nodes remain unchanged in signal intensity after nanopar ticle administration. The utilization of iron o xide nanopar ticles allowed the detection of l ymph node metastases in 80 patients with patholo gically proven prostate cancer with 95.7% sensitivity. The power of this approach lies in that, somewhat unexpectedly, even very small metastases, less than 2 mm in diameter , can be identif ied within nor mal-sized l ymph nodes (F igure 5B). This application of magnetic nanopar ticles holds g reat

Magnetic Resonance Imaging Agents

A

401

B

Figure 5. Lymphotropic imaging with iron oxide nanoparticles (Reproduced with permission from Harisinghani et al.73). A, Mechanism of lymphotropic imaging. B, Examples illustrating a normal pelvic lymph node (top), malignant lymph node (middle), and a lymph node harboring micrometastasis (bottom).

promises to revolutionize how lymph node metastasis is diagnosed and is cur rently in phase 3 clinical trial.

Inflammation Imaging The role of MRI for detection of macrophage phagocytic activity is e volving. Several in vitro and in vivo studies have demonstrated the feasibility and the potential clinical applications (transplant rejection and identif ication of atherosclerotic plaques) of macrophage-specif ic MRI following intravenous administration of ultrasmall SPIO in animals and humans. 74,75 The obser vation that macrophages and other phagocytic cells efficiently internalize SPIO has recentl y led to their e valuation as an MRI agent for the diagnosis of inflammatory and degenerative disorders associated with high macrophage phagocytic acti vity.76,77 These applications of macrophage imaging in the preclinical and the clinical 78–84 settings include MS to track neuroinflammation, human carotid atherosclerosis to assess plaque stability,85 insulitis to identify pancreatic inflammation and type I diabetes,86,87 and peripheral neuropathy to evaluate injury and subsequent healing. 88 Utilizing a highthroughput screening approach, Weissleder and colleagues89 also discovered two new agents with dif ferent macrophage af finity. CLIO-bentri w as preferentiall y internalized into resting macrophages, w hereas CLIOgly w as f avored b y acti vated macrophages. These specific agents that tar get subpopulations of macrophages are potentiall y po werful tools to discriminate betw een active and dormant disease.

Liver and Spleen Imaging After intravenous administration of SPIOs (80 to 150 nm), that is, AMI 125, they are cleared from the b lood within minutes, rapidl y accumulated in the reticuloendothe lial cells of liver and spleen, and can be detected b y imaging as decreased T2-weighted signal intensity in healthy liver and spleen. In the absence of nor mal RES cells due to pathology, such as malignanc y, the use of SPIO can differentiate between healthy and diseased tissues in the liver and the spleen. 90,91

Targeted Nanoparticle Imaging Agents A large number of e xperimental agents have been developed and are co vered in detail in Chapter 34 “Magnetic Nanoparticles.”

CONCLUSIONS Highly specific molecular and cellular MRI agents represent a noninvasive, nonionizing powerful tool to visualize physiologic and pathologic conditions. They have tremendous potential for improving our understanding of molecular processes in volved in a v ariety of impor tant clinical diseases. Functionalized iron oxide nanoparticles, because of the ease of conjugation, ha ve under gone substantial growth in the number of agents and biologic applications. Enzymatic acti vatable gadolinium-based agents, such as the MPO sensitive agents, represent an emer ging class of MR agents that resemb le clinically approved MRI agents

402

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

but offer increased molecular specificity and sensitivity to report on enzymatic acti vity involved in man y diseases. CEST and P ARACEST agents represent an impor tant third class of no vel molecular sensing agents, most notably in the area of pH sensing, impor tant to renal and tumor pathoph ysiology. The strate gies and applications described in this chapter represent onl y the tip of the iceberg of the future that is to come.

REFERENCES 1. Purcell EM, Torrey HC, Pound RV. Resonance absor ption by nuclear magnetic moments in a solid. Phys Rev 1946;69:37–8. 2. Bloch F , Hansen WW, P ackard M. Nuclear induction. Ph ys Re v 1946;69:127. 3. Rink PA. Magnetic resonance in medicine: the basic te xtbook of the European magnetic resonance forum. Oxford, UK: Blackwell Scientific Publications; 1993. 4. Merbach AE, Toth E. The chemistry of the contrast agents in medical magnetic resonance imaging. Wiley; 2001. 5. Lauffer RB. Paramagnetic metal complexes as water proton relaxation agents for NMR imaging: theor y and design. Chem Re v 1987;87:901–27. 6. Caravan P , Ellison JJ , McMur ry TJ, Lauf fer RB . Gadolinium(III) chelates as MRI contrast agents: structure, dynamics, and applications. Chem Rev 1999;99:2293–352. 7. Aime S, Botta M, Fasano M, Terreno E. Lanthanide (III) chelates for NMR biomedical applications. Chem Soc Rev 1998;27:19–29. 8. Coroiu I. Relaxi vities of dif ferent super paramagnetic par ticles for application in NMR tomo graphy. J Magn Magn Mater 1999;201:449–52. 9. Chen JW, Belford RL, Clarkson RB . Second-spehre and outer -sphere proton relaxation of paramagnetic complexes: from EPR to NMRD. J Phys Chem 1998;102:2117–30. 10. Gillis P , K oenig SH. Transverse relaxation of solv ent protons induced by magnetized spheres: application to fer ritin, erythrocytes, and magnetite. Magn Reson Med 1987;5:323–45. 11. Solomon I. Relaxation process in a system of tw o spins. Ph ys Rev 1955;99:559–65. 12. Bloembergen N, Purcell EM, Pound RV. Relaxation effects in nuclear magnetic resonance absorption. Phys Rev 1948;73:679–712. 13. Bloembergen N. Proton relaxation times in paramagnetic solutions. J Chem Phys 1957;27:572–3. 14. Weinmann HJ , Eber t W, Missel witz B , Schmitt-W illich H. Tissuespecific MR contrast agents. Eur J Radiol 2003;46:33–44. 15. Tweedle MF, Wedeking P, Kumar K. Biodistribution of radiolabeled , formulated gadopentetate, gadoteridol, gadoterate, and gadodiamide in mice and rats. Invest Radiol 1995;30:372–80. 16. Small WC, DeSimone-Macchi D, Parker JR, et al. A multisite phase III study of the safety and ef ficacy of a new manganese chloridebased gastrointestinal contrast agent for MRI of the abdomen and pelvis. J Magn Reson Imaging 1999;10:15–24. 17. Pomeroy OH, Wendland M, Wagner S, et al. Magnetic resonance imaging of acute m yocardial ischemia using a manganese chelate, MnDPDP. Invest Radiol 1989;24:531–6. 18. Weissleder R, Elizondo G, Wittenberg J, et al. Ultrasmall super paramagnetic iron o xide: characterization of a ne w class of contrast agents for MR imaging. Radiology 1990;175:489–93. 19. Jung CW, Jacobs P. Physical and chemical proper ties of super paramagnetic iron oxide MR contrast agents: fer umoxides, ferumoxtran, ferumoxsil. Magn Reson Imaging 1995;13:661–74. 20. Muller RN. Encyclopedia of nuclear magnetic resonance. Chichester, UK: John Wiley & Sons; 1995.

21. Schwertmann U, Cornell RM. Iron oxides in the laboratory, preparation and characterization. Cambridge, UK: Wiley-VCH; 2003. 22. Weissleder R, Stark DD, Engelstad BL, et al. Superparamagnetic iron oxide: phar macokinetics and to xicity. AJR Am J Roentgenol 1989;152:167–73. 23. Wang YX, Hussain SM, Krestin GP . Super paramagnetic iron o xide contrast agents: ph ysicochemical characteristics and applications in MR imaging. Eur Radiol 2001;11:2319–31. 24. Thorek DL, Chen AK, Czupr yna J, Tsourkas A. Super paramagnetic iron o xide nanopar ticle probes for molecular imaging. Ann Biomed Eng 2006;34:23–38. 25. Shen T, Weissleder R, P apisov M, et al. Monocr ystalline iron o xide nanocompounds (MION): ph ysicochemical proper ties. Magn Reson Med 1993;29:599–604. 26. Wunderbaldinger P , Josephson L, Weissleder R. Crosslink ed iron oxides (CLIO): a ne w platfor m for the de velopment of tar geted MR contrast agents. Acad Radiol 2002;9 Suppl 2:S304–6. 27. Woods M, Woessner DE, Sher ry AD. P aramagnetic lanthanide complexes as PARACEST agents for medical imaging. Chem Soc Rev 2006;35:500–11. 28. Aime S, Geninatti Crich S, Gianolio E, et al. High sensiti vity lanthanide(III) based probes for MR-medical imaging. Coord Chem Rev 2006;250:1562–79. 29. Aime S, Barge A, Botta M, et al. Direct NMR spectroscopic observation of a lanthanide-coordinated w ater molecule whose exchange rate is dependent on the confomation of the comple xes. Angew Chem Int Ed Engl 1998;37:2673–5. 30. Zhang S, Winter P, Wu K, Sherry AD. A novel europium(III)-based MRI contrast agent. J Am Chem Soc 2001;123:1517–8. 31. Aime S, Delli Castelli D , Terreno E. Highl y sensitive MRI chemical exchange saturation transfer agents using liposomes. Angew Chem Int Ed Engl 2005;44:5513–5. 32. Geninatti Crich S, Bar ge A, Battistini E, et al. Magnetic resonance imaging visualization of tar geted cells b y the inter nalization of supramolecular adducts for med between avidin and biotin ylated Gd3+ chelates. J Biol Inorg Chem 2005;10:78–86. 33. Yoo B, Pagel MD. An overview of responsi ve MRI contrast agents for molecular imaging. Front Biosci 2008;13:1733–52. 34. Bottrill M, Kw ok L, Long NJ . Lanthanides in magnetic resonance imaging. Chem Soc Rev 2006;35:557–71. 35. Moats R, Scott DM, F raser SE, Meade TJ. A “smart” magnetic resonance imaging agent that repor ts on specif ic enzymatic acti vity. Angew Chem Int Ed Engl 1997;36:725–8. 36. Louie AY, Huber MM, Ahrens ET, et al. In vivo visualization of gene expression using magnetic resonance imaging. Nat Biotechnol 2000;18:321–5. 37. Nivorozhkin AL, K olodziej AF, Cara van P, et al. Enzyme-acti vated Gd(3+) magnetic resonance imaging contrast agents with a prominent receptor-induced magnetization enhancement. Angew Chem Int Ed Engl 2001;40:2903–6. 38. Hanaoka K, Kikuchi K, Terai T, et al. A Gd(3+)-based magnetic resonance imaging contrast agent sensitive to beta-galactosidase activity utilizing a receptor-induced magnetization enhancement (rime) phenomenon. Chemistry 2007. 39. Himmelreich U, Aime S, Hieronymus T, et al. A responsive MRI contrast agent to monitor functional cell status. Neuroimage 2006;32:1142–9. 40. Bogdanov A Jr, Matuszewski L, Bremer C, et al. Oligomerization of paramagnetic substrates result in signal amplif ication and can be used for MR imaging of molecular tar gets. Mol Imaging 2002;1:16–23. 41. Chen JW, Querol Sans M, Bo gdanov A Jr, Weissleder R. Imaging of myeloperoxidase in mice by using novel amplifiable paramagnetic substrates. Radiology 2006;240:473–81. 42. Chen JW, Pham W, Weissleder R, Bogdanov A Jr. Human myeloperoxidase: a potential tar get for molecular MR imaging in atherosclerosis. Magn Reson Med 2004;52:1021–8.

Magnetic Resonance Imaging Agents

43. Bradley PP, Christensen RD , Rothstein G. Cellular and e xtracellular myeloperoxidase in pyogenic inflammation. Blood 1982;60:618–22. 44. Heinecke JW. P athways for o xidation of lo w density lipoprotein b y myeloperoxidase: tyrosyl radical, reacti ve aldehydes, hypochlorous acid and molecular chlorine. Biof actors 1997;6:145–55. 45. Querol M, Chen JW, Bogdanov AA Jr. A paramagnetic contrast agent with m yeloperoxidase-sensing proper ties. Or g Biomol Chem 2006;4:1887–95. 46. Nahrendorf M, Sosnovik D, Chen JW, et al. An activatable MR imaging agent follows myeloperoxidase activity in healing inf arcts and detects ator vastatin-mediated attenuation of ischemia reperfusion injury non-invasively. Circulation. 2008;117(9):1153–1160. 47. Libby P. Inflammation in atherosclerosis. Nature 2002;420:868–74. 48. Naruko T, Ueda M, Haze K, et al. Neutrophil inf iltration of culprit lesions in acute coronar y syndromes. Circulation 2002;106: 2894–900. 49. Heinecke JW. P athways for o xidation of lo w density lipoprotein b y myeloperoxidase: tyrosyl radical, reacti ve aldehydes, hypochlorous acid and molecular chlorine. Biof actors 1997;6:145–55. 50. Fu X, Kassim SY, Parks WC, Heinecke JW. Hypochlorous acid o xygenates the cysteine switch domain of pro-matrilysin (MMP-7). A mechanism for matrix metalloproteinase acti vation and atherosclerotic plaque r upture b y m yeloperoxidase. J Biol Chem 2001;276:41279–87. 51. Marsche G, Hammer A, Oskolkova O, et al. Hypochlorite-modified high density lipoprotein, a high af finity ligand to sca venger receptor class B , type I, impairs high density lipoprotein-dependent selective lipid uptak e and reverse cholesterol transport. J Biol Chem 2002;277:32172–9. 52. Bergt C, P ennathur S, Fu X, et al. The m yeloperoxidase product hypochlorous acid o xidizes HDL in the human ar tery w all and impairs ABCA1-dependent cholesterol transpor t. Proc Natl Acad Sci U S A 2004;101:13032–7. 53. Brennan ML, P enn MS, Van Lente F , et al. Pro gnostic v alue of myeloperoxidase in patients with chest pain. N Engl J Med 2003;349:1595–604. 54. Ronald J, Chen JW, Rogers K, et al. In vivo identification of “active” infalmmation in e xperimental atherosclerosis via m yeloperoxidase MR imaging. Joint Molecular Imaging Conference; Pro vidence, RI. 2007. 55. Cotton F, Weiner HL, Jolesz FA, Guttmann CR. MRI contrast uptak e in new lesions in relapsing-remitting MS followed at weekly intervals. Neurology 2003;60:640–6. 56. Bruck W, Bitsch A, Kolenda H, et al. Inflammator y central ner vous system demyelination: correlation of magnetic resonance imaging findings with lesion pathology. Ann Neurol 1997;42:783–93. 57. Nagra RM, Becher B , Tourtellotte WW, et al. Immunohistochemical and genetic evidence of myeloperoxidase involvement in multiple sclerosis. J Neuroimmunol 1997;78:97–107. 58. Chataway J, Sawcer S, F eakes R, et al. A screen of candidates from peaks of linkage: e vidence for the in volvement of m yeloperoxidase in multiple sclerosis. J Neuroimmunol 1999;98:208–13. 59. Zakrzewska-Pniewska B, Styczynska M, P odlecka A, et al. Association of apolipoprotein E and m yeloperoxidase genotypes to clinical course of f amilial and sporadic multiple sclerosis. Mult Scler 2004;10:266–71. 60. Chen JW, Breckwoldt MO, Aikawa E, et al. Myeloperoxidase-targeted imaging of acti ve inflammator y lesions in multiple sclerosis. Brain. 2008;131(4):1123–1133. 61. Zhang S, Wu K, Sherry AD. A novel pH-sensitive MRI contrast agent. Angew Chem Int Ed Engl 1999;38:3192–94. 62. Raghunand N, Howison C, Sher ry AD, et al. Renal and systemic pH imaging b y contrast-enhanced MRI. Magn Reson Med 2003;49:249–57. 63. Laus S, Sour A, Ruloff R, et al. Rotational dynamics account for pHdependent relaxivities of PAMAM dendrimeric, Gd-based potential MRI contrast agents. Chemistr y 2005;11:3064–76.

403

64. Toth E, Bolskar RD , Borel A, et al. Water-soluble gadofullerenes: toward high-relaxivity, pH-responsive MRI contrast agents. J Am Chem Soc 2005;127:799–805. 65. Aime S, Barge A, Delli Castelli D, et al. Paramagnetic lanthanide(III) complexes as pH-sensitive chemical exchange saturation transfer (CEST) contrast agents for MRI applications. Magn Reson Med 2002;47:639–48. 66. Zhou J, Payen JF, Wilson DA, et al. Using the amide proton signals of intracellular proteins and peptides to detect pH ef fects in MRI. Nat Med 2003;9:1085–90. 67. Gilad AA, McMahon MT , Walczak P, et al. Artificial repor ter gene providing MRI contrast based on proton e xchange. Nat Biotechnol 2007;25:217–9. 68. Aime S, Delli Castelli D , Fedeli F, et al. A paramagnetic MRI-CEST agent responsi ve to lactate concentration. J Am Chem Soc 2002;124:9364–5. 69. Zhang S, Trokowski R, Sher ry AD. A paramagnetic CEST agent for imaging glucose by MRI. J Am Chem Soc 2003;125:15288–9. 70. Winter PM, Car uthers SD, Yu X, et al. Impro ved molecular imaging contrast agent for detection of human thrombus. Magn Reson Med 2003;50:411–6. 71. Winter PM, Cai K, Chen J , et al. Targeted PARACEST nanoparticle contrast agent for the detection of f ibrin. Magn Reson Med 2006;56:1384–8. 72. van Zijl PC, Jones CK, Ren J, et al. MRI detection of glycogen in vivo by using chemical e xchange saturation transfer imaging (gl ycoCEST). Proc Natl Acad Sci U S A 2007;104:4359–64. 73. Harisinghani MG, Barentsz J , Hahn PF, et al. Nonin vasive detection of clinically occult l ymph-node metastases in prostate cancer . N Engl J Med 2003;348:2491–9. 74. Kanno S, Wu YJ, Lee PC, et al. Macrophage accumulation associated with rat cardiac allo graft rejection detected b y magnetic resonance imaging with ultrasmall superparamagnetic iron oxide particles. Circulation 2001;104:934–8. 75. Kooi ME, Cappendijk VC, Cleutjens KB, et al. Accumulation of ultrasmall super paramagnetic par ticles of iron o xide in human atherosclerotic plaques can be detected b y in vi vo magnetic resonance imaging. Circulation 2003;107:2453–8. 76. Enochs WS, Harsh G, Hochberg F, Weissleder R. Improved delineation of human brain tumors on MR images using a long-circulating, superparamagnetic iron o xide agent. J Magn Reson Imaging 1999;9:228–32. 77. Dousset V, Delalande C, Ballarino L, et al. In vi vo macrophage activity imaging in the central nervous system detected by magnetic resonance. Magn Reson Med 1999;41:329–33. 78. Berger C, Hiestand P , Kindler-Baumann D, et al. Analysis of lesion development during acute inflammation and remission in a rat model of experimental autoimmune encephalomyelitis by visualization of macrophage infiltration, demyelination and blood-brain barrier damage. NMR in biomedicine 2006;19:101–7. 79. Brochet B , Deloire MS, Touil T, et al. Earl y macrophage MRI of inflammatory lesions predicts lesion se verity and disease de velopment in relapsing EAE. NeuroImage 2006;32:266–74. 80. Dousset V, Brochet B , Deloire MS, et al. MR imaging of relapsing multiple sclerosis patients using ultra-small-par ticle iron o xide and compared with gadolinium. AJNR Am J Neuroradiol 2006;27:1000–5. 81. Dousset V, Gomez C, Petry KG, et al. Dose and scanning delay using USPIO for central ner vous system macrophage imaging. Magma 1999;8:185–9. 82. Floris S, Blezer EL, Schreibelt G, et al. Blood-brain bar rier per meability and monoc yte inf iltration in e xperimental aller gic encephalomyelitis: a quantitati ve MRI study . Brain 2004;127 (Pt 3):616–27. 83. Oweida AJ, Dunn EA, Foster PJ. Cellular imaging at 1.5 T: detecting cells in neuroinflammation using acti ve labeling with super paramagnetic iron oxide. Mol Imaging 2004;3:85–95.

404

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

84. Rausch M, Hiestand P , Baumann D , et al. MRI-based monitoring of inflammation and tissue damage in acute and chronic relapsing EAE. Magn Reson Med 2003;50:309–14. 85. Trivedi RA, Malla warachi C, U-King-Im JM, et al. Identifying inflamed carotid plaques using in vi vo USPIO-enhanced MR imaging to label plaque macrophages. Arterioscler Thromb Vasc Biol 2006;26:1601–6. 86. Denis MC, Mahmood U , Benoist C, et al. Imaging inflammation of the pancreatic islets in type 1 diabetes. Proc Natl Acad Sci U S A 2004;101:12634–9. 87. Turvey SE, Sw art E, Denis MC, et al. Nonin vasive imaging of pancreatic inflammation and its reversal in type 1 diabetes. J Clin Invest 2005;115:2454–61.

88. Bendszus M, Stoll G. Caught in the act: in vi vo mapping of macrophage inf iltration in ner ve injur y b y magnetic resonance imaging. J Neurosci 2003;23:10892–6. 89. Weissleder R, K elly K, Sun EY , et al. Cell-specif ic tar geting of nanoparticles by multivalent attachment of small molecules. Nat Biotechnol 2005;23:1418–23. 90. Vogl TJ, Hammerstingl R, Schw arz W, et al. Super paramagnetic iron oxide—enhanced v ersus gadolinium-enhanced MR imaging for differential diagnosis of focal li ver lesions. Radiolo gy 1996;198:881–7. 91. Weissleder R, Stark DD, Rummeny EJ, et al. Splenic lymphoma: ferrite-enhanced MR imaging in rats. Radiolo gy 1988;166:423–30.

27 OPTICAL IMAGING AGENTS SCOTT A. HILDERBRAND, PHD

In the past decade, there has been a rapid increase in the development of ne w optical probes for in vi vo imaging applications. This has been f acilitated by recent impro vements in optical imaging instr umentation and b y ne w advances in the design of luminescent species with absorption and emission in the f ar-red or near infrared (NIR). Optical imaging in deep tissues is made possib le b y the relative increase in optical transparenc y of tissue from approximately 650 to 900 nm 1 (Figure 1). The enhanced penetration for both e xcitation and emission light in this wavelength range allows for imaging features several centimeters deep in tissue via transillumination of the subject. The use of fluorescent and luminescent probes, w hich absorb and emit in the NIR, results in impro ved signal-tonoise ratios because the y are spectrall y isolated from the autofluorescence of tissue components such as collagen and fluorescence signals arising from dietar y factors such as chlorophyll. As a result of their synthetic flexibility and desirable optical characteristics, most reporters in common use are or ganic dy es, w hich have optimal e xcitation and emission in the f ar-red or NIR. Quantum dots (QDs), which have a broad excitation spectrum and far-red to NIR emission, are also used commonly; however, they are optimally excited in the ultraviolet (UV) to visible where their extinction coefficients are signif icantly larger. In addition to organic dyes and QDs, a v ariety of new materials such as nanodiamonds, metal nanoclusters, carbon dots, and carbon nanotubes are recei ving attention for their NIR luminescence properties. A detailed perspective of the different luminescent compounds emitting in the f ar-red and NIR with potential for incorporation into optical imaging agents will be pre sented. Using cur rently available optical reporters, there are multiple strategies that may be used for the design of ne w imaging agents. These approaches may be di vided into three general classes of probes: nontargeted, targeted, and acti vatable. In this chapter , a detailed examination of the tar geting systems and acti vation

mechanisms of these NIR optical imaging probes will be presented.

ORGANIC DYES Organic fluorophores are the most widel y de veloped NIR repor ters used for the design of optical probes. Their use has increased dramaticall y since the synthesis of water-soluble conjugatable NIR dyes in the late 1980s and earl y 1990s. 2,3 Some of the f irst dy es to recei ve widespread use, such as Cy5, Cy5.5, and Cy7, are based on the carboc yanine platform (see F igure 1). 2,4 Cyanine dyes have been known since the earl y 1900s, and the first in-depth studies of their NIR fluorescence emission were detailed in 1950. 5 Cyanine dy es are bright fluorophores typically characterized b y extinction coefficients of 200,000 M −1cm−1 or more, quantum yields approaching 30%, and Stok es shifts of 20 to 30 nm. By variation of the dy e str ucture, carboc yanine dy es with fluorescence emission spanning the far-red and NIR can be prepared. Conjugatab le v ersions of these dy es with emission between 650 and 850 nm (see F igure 1) ha ve been repor ted.2,6 A wide range of ine xpensive, readil y synthesized dy es based on the carboc yanine scaf fold have been designed. F or e xample, the CyTE dy es are prepared by the nucleophilic attack of an alkyl thiol such as mercaptopropionic acid on a chloride modified dye to give a bioconjugatable fluorophore in near quantitative yield.6 Other dyes using aryl thiols and aryl alcohols as nucleophiles have also been used to generate ine xpensive conjugatab le dy es (see F igure 1). 7,8 The de velopment of synthesis routes to asymmetric functionalized cyanine dyes has also received considerable attention in the past decade. 9–11 These asymmetric carbo xylic acid derivatized c yanine dy es are prepared in a multistage process in w hich a sulfonated indolenium or benzindolenium salt is f irst condensed with a malonaldeh yde 405

406

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

C

Figure 1. Electromagnetic spectrum showing the optical imaging window from 650 to 900 nm where light absorption by tissue is minimal (A), examples of several far-red and near infrared fluorophores (B), and absorption (solid line) and emission (dashed line) traces of CyTE-777 (C).

or glutaconaldehyde derivative. The resulting inter mediate is then combined with a second carbo xylic acid modified indolenium to generate the asymmetric dy e.11 Further improvements in cyanine dyes for their application to in vivo imaging may come from recent reports on the preparation of c yanine dy es with e xtremely long Stokes shifts (see Figure 1).12 These dyes are prepared by the nucleophilic attack of primary and secondary amines on chloride containing dy e precursors. The ability to couple traditional cyanine dyes with the new dyes, which have a Stok es shift of up to 150 nm, ma y ultimatel y allow for de velopment of ne w classes of single e xcitation, dual emission optical imaging probes.

A variety of other dye scaffolds have also been investigated for use as fluorophores suitable for in vivo imaging applications. F or e xample, ne w conjugatab le por phyrin and phthalocyanine derivatives with far-red or NIR absorption and emission have been reported.13,14 Since these new classes of por phyrin and phthaloc yanine deri vatives are expected to ha ve signif icant singlet o xygen quantum yields, they may ultimately find use as combined diagnostic and photodynamic therap y agents. 15 Recently, a ne w silicon-containing water-soluble phthalocyanine derivative with absorption at 689 nm and emission at 700 nm that is 27 times more photostab le than tetrameth ylrhodamine was also repor ted.16 This and other ne w NIR dy es with

Optical Imaging Agents

exceptional photostability may prove useful in applications where photobleaching is a concer n. A new class of conjugated porphyrin arrays was recently reported.17 These dyes are composed of multiple por phyrin units that are link ed into linear ar rays via alk yne moieties at their meso positions. The absor ption and emission w avelengths can be tuned by altering the number of porphyrin monomers in the ar rays. With this approach, dy es with fluorescence emission from 710 to 880 nm ha ve been prepared with quantum yields between 6 and 22%. Although they are not bioconjugatable, these por phyrin arrays have potential for incorporation into larger nanoscaffolds for imaging applications.18 Phenoxazine dyes have also received attention as potential far-red and NIR imaging agents. Carboxylic acid derivatized pheno xazines w ere recentl y repor ted with absorption and emission in the f ar-red.19 Sulfonated phenoxazine dy es designed for bioconjugation to the free amine of the chromophore ha ve been used for preparation of fluorogenic enzyme activatable probes. 20 A wide range of commerciall y a vailable f ar-red and NIR conjugatab le dyes are now available with functional groups such as succinimidyl esters, isothioc yanates, hydrazides, maleimides, and amines.21–26 Recently, efforts have been focused on modif ication of nanopar ticle scaf folds with or ganic dy es (F igure 2). The incor poration of fluorophores into these scaf folds has enab led the preparation of par ticles, w hich can deliver hundreds or thousands of copies of the fluorescent reporters to the site of interest, potentiall y resulting in significant increases in signal-to-noise ratios. In addition, the dy e-loaded nanopar ticles often benef it from improved b lood half-li ves. Longer in vi vo circulation times enable the imaging agents constr ucted with these fluorescent nanopar ticle components suf ficient time to reach and accumulate at the desired site for imaging. A v ariety of or ganic dy es ha ve been incor porated onto the surf ace and interiors of silica nanopar ticles. In one example, a core-shell approach w as tak en enab ling the

A

B

407

development of highl y luminescent par ticles, with good photostability and the ability to incorporate fluorophores with emission in the visible and NIR. 27 In another example, the close proximity of the fluorophores to each other in the interior of silica nanopar ticles w as le veraged to prepare fluorescence resonance ener gy transfer (FRET)based particles with long Stokes shifts.28 Far-red and NIR emitting fluorophores have been encapsulated into pol ystyrene or late x nanoparticles and have been used for in vivo imaging applications. 29 Polymersomes, amphiphilic diblock copolymer-based vesicles, were used to incorporate nonconjugatab le por phyrin ar rays resulting in strongly emissive nanoscale platfor ms for imaging. 18 NIR fluorophores are no w widel y appended to the polymer coatings of super paramagnetic iron o xide nanoparticles to yield materials suitab le for multimodal fluorescence and magnetic resonance imaging. 30 More recently, methods for covalent attachment of hundreds of copies of NIR fluorophores to the surf aces of bacteriophage par ticles ha ve been de veloped.6 This approach yields a fle xible bright platfor m that can be easil y adapted for a variety of in vivo imaging applications. 31

QDS QDs are fluorescent inor ganic semiconductor nanocr ystals with unique photoph ysical characteristics. The highly tunable optical proper ties of QDs are the result of quantum confinement ef fects. Quantum conf inement occurs w hen the diameter of the QD par ticle becomes similar to that of the e xciton, w hich is an e xcited state electron generated upon absorption of light b y the par ticle. When the e xcited state electron retur ns to the g round state, a photon of an energy that is dependent on the diameter of the nanocr ystal is generated. For example, small (approximately 2 nm) core diameter CdSe QDs emit in the b lue whereas larger particles displa y pro gressively longer w avelength emission into the NIR. All QDs ha ve broad absor ption spectra

C

D

Figure 2. Schematic of different dye labeled nanoparticles: A, particles that are uniformly doped with near infrared (NIR) emitting fluorophores. B, particles with a core containing a high concentration of NIR fluorophores surrounded by an undoped shell. C, particles with external fluorophore labeling, and (D) liposomes or polymersomes loaded with NIR emitting dyes.

408

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

characterized b y a pro gressively decreasing e xtinction coefficient as the w avelength increases. As a result of this, their optimal e xcitation wavelength is often in the UV or visible; however, they can still be e xcited in the f ar-red or NIR where the e xtinction coefficients are lo wer. Although the QD cores are v ery small, the unmodif ied cr ystals are not suitable for biologic imaging. The small size of the particles results in a high surf ace area to volume ratio, magnifying the ef fects of surf ace defects on the cr ystals and contributing to a reacti ve and unstab le surface. To protect the emissi ve cores, par ticles are frequentl y coated with a shell of zinc sulf ide. This coating procedure stabilizes the core and leads to dramatic increases in chemical stability , photostability, and luminescence quantum yields. 32,33 To render the particles water-soluble for biologic applications, further modif ication is necessar y to install char ged or amphiphilic groups on the surface of the particles.34,35 As a result of their inorganic nature, QDs are signif icantly more photostable than or ganic dyes. Due to their photostability and tunable emission well into the NIR, QDs are recei ving increasing interest as optical repor ters for de velopment of NIR optical contrast agents. In addition, QDs ha ve broad excitation spectra and nar row emission spectra. The broad excitation spectra allo w for the possibility of multiple xed imaging e xperiments w here multiple QDs with dif ferent emission wavelengths can be imaged simultaneously with a single excitation source, often in the visib le. When visible excitation light is used as a result of its poor tissue penetration ability , the in vi vo use of QDs is limited to epifluorescence techniques; however, if they are excited in the NIR, transillumination or fluorescence molecular tomography (FMT) methodolo gies are possib le. Ev en though QDs and or ganic fluorophores cur rently are the predominant luminescent species used in the preparation of optical imaging agents, a v ariety of ne w luminescent species are beginning to receive attention as potential emissive reporters.

NANODIAMONDS Fluorescent nanodiamonds are chemicall y iner t, contain no to xic elements, are highl y photostab le, and ma y be easily surf ace modif ied with carbo xylic acid g roups for potential bioconjugation. 36 The nanodiamonds are rendered fluorescent b y ir radiating diamonds containing 100 ppm nitrogen with a 3 MeV proton beam followed by annealing at 800°C to generate negatively charged, emissive nitrogen-vacancy centers.37,38 The nanodiamonds are excited at 560 nm, sho w emission centered at 700 nm, have quantum yields approaching unity , and are similar in brightness to w avelength matched CdSe QDs. 37,38 The ability to perfor m intracellular measurements using

the nanodiamonds w as demonstrated with intracellular tracking e xperiments of indi vidual nanocr ystals in li ve HeLa cells. 38 Due to their stability and iner t nature, nanodiamonds show promise for future in vi vo imaging applications; ho wever, their relati vely shor t-wavelength excitation at 560 nm and the inability to tune their emission w avelengths, as is possib le with QDs and or ganic fluorophores, may limit their utility.

EMISSIVE SILVER AND GOLD NANOCLUSTERS Gold and silv er nanoclusters displa y fluorescence emission via a mechanism similar to the one obser ved for semiconductor QDs, in w hich the nanocluster cores are small (< 2 nm) and generate discrete quantum conf ined electronic transitions. As with QDs, the emission w avelengths are tunab le into the NIR b y modulation of the metal cluster size. Until recently the utility of these metallonanoclusters was limited b y their lo w quantum yields, often less than 0.3% for a v ariety of gold clusters. 39–41 New synthesis methods ha ve been de veloped, ho wever, that result in dramatically improved quantum efficiencies. Particles with quantum yields of 15 and 10% with emission at 760 and 866 nm, respecti vely, in aqueous media have now been repor ted.42 These p articles are prepared by the slow reduction of Au salts in aqueous pol y(amidoamine) dendrimer solutions where alteration of the dendrimer to Au ratio determines the nanocluster size. Similar dendrimer-stabilized gold nanoclusters have been targeted to folate receptor e xpressing human epithelial carcinoma KB cells b y conjugation of folate to the peripher y of the dendrimers.43 NIR luminescent silv er clusters w ere recently prepared by reduction of silver nitrate in the presence of single strand cytosine DNA 12mers.43 These silver particles are very bright with a reported extinction coefficient of 320,000 M −1cm−1 at 650 nm, a quantum yield of 17%, and emission at 700 nm in aqueous solution. Unlike QDs, single silver nanocluster particles do not display any blinking effects. To date, there are no repor ts of in vi vo imaging with gold or silv er nanoclusters. Ne vertheless, they are a promising new platform for the development of future optical imaging agents.

CARBON NANOPARTICLES AND NANOTUBES The preparation of luminescent quantum-sized carbon nanoparticles has recentl y been repor ted in the literature. Carbon nanopar ticles can be prepared b y laser ab lation of a carbon tar get. The par ticles for med

Optical Imaging Agents

by this method are initiall y nonemissi ve and must be treated with nitric acid followed by a f inal coating with diamino polyethylene glycol polymers to yield luminescent particles.44 The approximately 5 nm diameter particles prepared b y this procedure are w ater-soluble and display a broad range of emission w avelengths from 400 to 700 nm. The obser ved quantum yields of these particles range from 4 to 10% and depend on the e xcitation w avelength used. 44 These luminescent carbon nanoparticles have been used for 2-photon microscop y after internalization into MCF-7 breast cancer cells.45 In a similar approach, luminescent carbon nanopar ticles were prepared b y treatment of carbon agglomerates from candle soot with nitric acid. 46 This process introduces hydroxyl and carboxylic acid groups on the particle surf aces, rendering them w ater-soluble and providing sites for possib le bioconjugation. The par ticles, which average less than 2 nm in diameter , display a range of emission w avelengths from the b lue to red. Particles with discrete emission w avelengths ma y be separated b y SDS-P AGE chromato graphy.46 The quantum yields of these par ticles, however, are all less than 2%, w hich is similar to the quantum ef ficiencies reported for single-w alled carbon nanotube fragments (SWNTs).47 The str ucture and e xact composition of these small carbon nanopar ticles are still unclear . Further research is needed to elucidate their str ucture and the mechanism for their luminescence. As with luminescent carbon nanopar ticles, SWNTs are receiving attention for possib le repor ters in bioimaging applications. SWNTs have multiple emission peaks in the NIR with emission bands between 900 and 1600 nm.48 The imaging of Drosophila melanogaster (fruit fly) larvae with water-soluble SWNTs was recently reported with excitation at 658 or 785 nm and emission obser ved in the NIR.49 However, as with luminescent carbon nanopar ticles, improvements in the quantum yields of SWNTs are necessary before more widespread use of these materials can be expected.

NONTARGETED AGENTS Currently, there are onl y tw o NIR dy es appro ved for human use in the clinic, indoc yanine g reen (ICG) and methylene blue, both of which are small molecules. The application of these and other small molecule dy es for imaging is predicated on their e xtravasation and/or accumulation in tissues, such as those of the urinar y tract, hepatobiliary system, or tumor tissue as the result of their natural phar macokinetic distribution in vi vo. Some recent research focusing on the use of nontar-

409

geted fluorescence-based imaging probes has centered on the use of dye labeled nanoparticle scaffolds. One of the more ef fective uses of this approach has been for lymph node mapping. In one study , different-sized farred and NIR fluorophore loaded polystyrene nanoparticles were investigated for their ability to map sentinel lymph nodes. 29 NIR emitting QDs have also been used to delineate the sentinel lymph nodes in mouse and pig models.50 In addition, multimodal iron oxide nanoparticles loaded with Cy5.5 w ere in vestigated for their ability to delineate brain tumor mar gins intraoperatively b y optical spectroscop y.51 Although there ha ve been several examples of imaging agents with no specific targeting functionality, their use is limited b y the pharmacokinetics of the materials w hich compose the probes.

TARGETED AGENTS Specific tar geting of optical imaging agents of fers significant improvements over nontargeted probes. The coupling of optical repor ters with biolo gic tar geting molecules enab les imaging of biomark ers, w hich are overexpressed in the diseased state of interest. By using this approach, the ability to specif ically tar get and image virtually any disease is possible, provided effective and selecti ve tar geting materials are a vailable. Targeted probes are generally prepared by the coupling of the tar geting functionality to one or more emissi ve reporter species via h ydrophobic interactions or chemical bond for ming reactions generating amide, disulfide, or ether linkages. Targeted probes can tak e one of the three common for ms: 1:1 fluorophore to targeting agent constructs, fluorescent reporters labeled with se veral tar geting g roups, or lar ge tar geting moieties labeled with multiple fluorophores (Figure 3). As a result of their simple o verall design, allowing the use of a v ariety of optical repor ters, tar geted NIR imaging agents are cur rently the most di verse class of imaging probes in common use. Fur thermore, targeted probes are often combined with other imaging modalities to enhance their utility . Targeted imaging probes can be divided into several general subgroups based on the class of tar geting functionalities used. The major classes of tar geting agents used include peptides, antibodies, and small molecules. More recentl y, there has been increased interest in the de velopment of probes using aptamers for tar geting, and other researchers ha ve be gun to e xplore the possibility of using fluorescently labeled cells for imaging and cellular migration studies in vivo.

410

A

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

B

C

Figure 3. Possible strategies for targeted imaging probes: A, 1:1 coupling of a luminescent reporter to a targeting group, B, a single optical reporter labeled with multiple targeting groups, and C, a single targeting group modified with multiple optical reporters.

PEPTIDES The nonin vasive visualization of the αvβ3 integrin cell adhesion molecule in vi vo has been thoroughl y studied using radionuclide imaging strate gies with the RGD peptide tar geting motif, w hich preferentiall y binds the integrin. Only recently, the RGD peptide has become the focus for development of tumor targeted diagnostic optical imaging agents. In one earl y study, a c yclic RGD peptide w as coupled to a Cy5.5 fluorophore, generating a small molecule αvβ3 integrin targeted imaging probe. 52 Even though the probe has a lo w molecular w eight and is e xpected to have a short blood half-life, it was possible to successfully image subcutaneous U87MG gliob lastoma tumors. More recently, improved optical imaging systems have been used to visualize similar RGD-Cy5.5 conjugates in tumors with high and low expression levels of αvβ3 integrins by FMT.53 In efforts to improve the tar geting specif icity of the shor t RGD peptide, Cy5.5 conjugation strate gies that enab le coupling of one fluorophore to multiple c yclic RGD peptide motifs w ere developed.54–56 One of these approaches uses a c yclic decapeptide platfor m (RAFT) onto w hich four RGD peptides and one Cy5.5 fluorophore ma y be attached. Direct comparisons betw een tar geting of the RAFT-based probe and a 1:1 RGD-Cy5.5 constr uct demonstrate enhanced tar geting and impro ved signal to noise ratios with the RAFT probe. 54,55 This and other multivalent strate gies ha ve become common approaches for improving the deli very and specif icity of optical imaging probes. The multivalent targeting approach has been further modif ied b y incor poration of multiple c yclic RGD peptides and Cy5.5 fluorophores onto nanopar ticle scaffolds. In one e xample, up to 27 c yclic RGD peptides and 7 Cy5.5 fluorophores were conjugated to 28 nm diameter amino dextran cross-linked iron o xide (CLIO) nanopar ticles by disulfide and amide linkages, respectively.57 Strong

accumulation of these par ticles w as obser ved in BT -20 tumors. A similar multi valent approach w as tak en using QDs as the nanopar ticle scaffold.58 With the QD system, both cyclic RGD peptides and 1,4,7,10-tetraazacyclododecane-N,Nʹ′,Nʹ′ʹ′,Nʹ′ʹ′ʹ′-tetraacetic acid (DO TA) chelators w ere appended to the surf ace of the QDs. In this case, optical imaging w as perfor med b y monitoring emission of the QDs between 695 and 770 nm. This QD probe can be multimodal. When radioactive 64Cu is chelated b y the DO TA groups on the surf ace of the QDs, the probe also can be used for positron emission tomography.58 In addition to applications in tumor imaging, tar geting strategies have been used to design probes for cardiovascular disease. One method for imaging cardiovascular disease is based on thrombosis induced cross-linking of a tar geted probe into clots. This occurs when thrombin is acti vated and be gins to con vert f ibrinogen into f ibrin. The fibrin polymerizes and becomes further cross-linked by activated blood coagulation f actor XIII. A specif ic protein substrate sequence, NQEQVSPLTLLK, is cross-link ed into the clot b y transglutaminase factor XIII. Fluorescence-based probes for the clotting cascade can be prepared b y conjugation of NIR fluorescent repor ters to the f actor XIII peptide sequence (F igure 4). 59 The probe is highl y specif ic for incorporation into acti ve thrombi, signif icantly less cross-linking is obser ved, ho wever, in older thrombi where acti vated f actor XIII acti vity is e xpected to decrease.60 Probes that are not cross-link ed to thrombi have also been designed. One of these approaches is based on the gl ycoprotein IIb/IIIc binding sequence. Glycoprotein IIb/IIIc is a surf ace receptor that is closely involved in regulation of platelet adhesion in the clotting process.61 The gl ycoproteins under go a confor mational change upon platelet activation, increasing the affinity of the receptor for f ibrinogen. The bound f ibrinogen can then f acilitate interactions with adjacent platelets. P eptide inhibitors have been developed against glycoprotein IIb/IIIc.62 The conjugation of Cy5.5 to linear and branched versions of these inhibitors generates fluorescence-based probes with high af finity to clots. 63 In another e xample, a peptide sequence, CREKA, w as identified b y in vi vo phage screening that reco gnizes clotted plasma proteins. Once identified, the peptide and Cy7 fluorophores were covalently linked to 50 nm superparamagnetic iron o xide nanopar ticles for imaging. 64 Using this strate gy, it w as possible to load in e xcess of 8,000 peptides per particle for enhanced targeting. The preparation of tar geted probes is not limited to the coupling of dy es or fluorescent nanopar ticles to shor t peptide sequences. Larger peptides and proteins may also

Optical Imaging Agents

Figure 4. Cross-linking of a factor XIII-based optical probe into an active thrombus.

be used for the directed deli very of fluorescence-based reporters. In one e xample, the angio genesis inhibitor , endostatin, a 20 kDa C-ter minal protein fragment derived from type XVIII collagen, w as co valently coupled to Cy5.5, and its tumor homing abilities w ere in vestigated.65,66 After intraperitoneal injection of the probe, an increase in fluorescence signal, w hich peak ed 42 hours postinjection, was observed in mice bearing LLC tumors. Similarly, probes tar geting the epider mal g rowth f actor receptor (EGFR), w hich is o verexpressed in a v ariety of solid tumor types, ha ve been designed. These probes are prepared b y conjugation of Cy5.5 to epider mal g rowth factor, a 6 kDa protein that pla ys impor tant roles in the regulation of cell g rowth, proliferation, and dif ferentiation.67 This probe successfull y homes to MD A-MB-468 tumors, which have high EGFR levels, whereas little or no accumulation of the probe, as monitored b y NIR fluorescence imaging, w as obser ved in MD A-MB-435 tumors that do not e xpress EGFR. In another interesting approach, Cy5.5 was conjugated to anne xin V to prepare a probe for specif ic monitoring of apoptosis in the tumor environment.68 The feasibility of nanopar ticles as generic scaf folds for development of tar geted imaging probes w as investigated. In one study, a series of peptide targeted QD probes were generated from w ater-soluble mercaptoacetic acid69 These coated QDs with emission at 550 or 625 nm. probes w ere tar geted for the lung, tumor b lood v essels, and tumor lymphatic vessels. The lung targeted QD imaging agent used a peptide sequence known to bind to membrane dipeptidases that are e xpressed on endothelial cells lining lung b lood v essels. The other tw o peptides used bind specifically to tumor blood vessels or lymphatic vessels in certain tumors. All three probes were prepared by a simple ligand e xchange reaction betw een mercaptoacetic acid-coated QDs and thiol modif ied peptide sequences. 69 The probes were shown to home to their intended tar gets in vivo by fluorescence microscop y of frozen tissue sections after administration of the probes. Multi valent peptide-based probes ha ve also been designed around fluorophore labeled iron o xide nanopar ticles.70–72 In one

411

of these systems, tar geted in vi vo imaging with par ticles loaded with Cy5.5 w as combined with siRN A deli very and gene silencing in tumors. 71 Over the past se veral y ears, bacteriophage displa y technology and in vi vo selection techniques ha ve become an important means of identifying ne w peptide targets for development into fluorescence-based imaging probes. F or example, peptide sequences that reco gnize colorectal cancer, prostate cancer , and clotted plasma proteins for targeting and tumor accumulation ha ve been de veloped. Once identif ied, these peptide sequences typicall y ha ve been conjugated to a fluorophore or fluorophore modif ied nanoparticles before in vi vo imaging e xperiments w ere undertaken.64,73,74 In the last few years, however, it was discovered that the bacteriophage par ticles themselv es ma y serve as de facto multivalent peptide targeted fluorescencebased imaging probes. The M13 bacteriophage par ticle, which is used frequentl y for phage displa y screening and has randomized peptide libraries displayed on the pIII coat proteins, is approximately 800 nm long by 6.6 nm in diameter. The phage particles also contain approximately 2,700 copies of the pVIII coat protein co vering the bulk of the particle. These coat proteins ha ve their amino ter mini exposed to the solvent for potential bioconjugation. Once a particular phage clone that is specif ic for the imaging target of interest is identif ied, se veral hundred copies of a NIR fluorophore ma y be conjugated to the coat proteins generating a fluorescent peptide-tar geted phage par ticle (Figure 5). In one early example, bacteriophage containing the peptide sequence, VHSPNKK, tar geted to v ascular cellular adhesion molecule-1 (VCAM-1) was labeled with the NIR fluorophore CyTE-777. 6 Proof of principle microscopy studies demonstrated the accumulation of the

Figure 5. A targeted bacteriophage-based imaging agent labeled with near infrared fluorophores.

412

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

imaging agent in murine cardiac endothelial cells e xpressing VCAM-1.6 In another example, a chimeric phage w as designed for tumor imaging. This phage clone contains an RGD motif on the pIII coat protein for inte grin targeting, and the pVIII coat proteins were modified to display streptavidin binding sequences. 75 Red-orange fluorescent streptavidin-coated nanospheres or luminescent QDs coated with streptavidin could be bound to the pVIII coat proteins. These phage par ticles retained their tar geting abilities and w ere inter nalized w hen incubated with αv integrin expressing KS1767 cells. 75 In vivo imaging with NIR fluorescent-peptide labeled phage is also possib le. Phage displa y w as used to identify a peptide tar geting secreted protein acidic and rich in c ysteine (SPARC).31 These SP ARC tar geted bacteriophage par ticles w ere labeled with a NIR emitting fluorophore and w ere shown to target LLC tumors in vivo.31 Other unique peptide-based tar geting approaches have recentl y recei ved attention. In one system, a pH-selective membrane insertion peptide was used to target acidic tissue. In this system, the pHLIP peptide w as used for homing to acidic microen vironments in tumors, kidneys, and inflammator y sites in vi vo.76 The peptide sequence has three states: an unfolded water-soluble form, a membrane surf ace bound state, and an α-helix conformation inserted across the cell membrane. At physiological pH, the peptide e xists primarily in the w ater-soluble conformation, e xplaining its lack of af finity for health y tissue; ho wever, at pH less than 7.0, the equilibrium is shifted toward membrane inser tion. Covalent attachment of NIR emitting dy es such as Cy5.5 or Alexa Fluor 750 was perfor med to prepare probes for in vi vo imaging. 76 When injected into mice bearing CRL-2116 murine breast adenocarcinoma tumors, the probe selecti vely accumulated at the tumor site with a signal-to-noise ratio of approximately f ive. Imaging of acidic kidne y en vironments and inflammation in an ar thritis model w as also demonstrated.

inserted into f ibronectin by alter native splicing during tissue remodeling and angio genesis, tw o hallmarks of advanced stage plaque. The antibodies w ere either labeled with Cy5 or Cy7 at antibody-dy e ratios of 1:5 and 1:2, respectively.78 Following injection of the probes into atherosclerotic apolipoprotein E-null (ApoE) mice accumulation of fluorescent signal w as obser ved in excised aortas cor responding to atherosclerotic re gions with increased e xpression of ED-B .78 HER-2 o verexpression in tumors has also been visualized b y fluorophore labeled antibodies. In this study , antibodies against HER-2/neu antigen were labeled with 2.0 to 2.5 Cy5.5 fluorophores per antibody prior to injection into mice bearing HER-2/neu o verexpressing SK-BR-3 tumors. Imaging e xperiments indicated homing of the dye labeled antibody to the tumor tissue with a signalto-noise ratio of appro ximately tw o.79 Another HER2/neu targeted probe based on trastuzumab labeled with an 111In γ-emitter and a NIR dy e for multimodal imaging was repor ted.80 Rheumatoid ar thritis imaging with labeled antibodies using an antigen-induced ar thritis mouse model system has also been demonstrated. 81 To image, Cy5.5 labeled antibodies against F4/80 antigen, which is e xpressed on macrophages that home in on inflamed joints in the antigen-induced ar thritis mouse model system, w ere prepared. The probe sho wed accumulation in inflamed knee joints of the mouse model and to a lesser e xtent in nonarthritic joints. 81 The localization of the probes w as fur ther conf irmed by histology. In a similar f ashion, NIR fluorophore labeled anti-EGFR antibodies were used to image several tumor cell lines.82,83 Antibody-based imaging strategies are not limited to fluorophore-antibody constr ucts. NIR luminescent QDs have been labeled with multiple copies of antibodies for tar geting and imaging of tumors. F or example, anti-PSMA antibody labeled QDs ha ve been used to image PSMA-positi ve human prostate cancers in Balb/c mice. 35

ANTIBODIES

SMALL MOLECULES

As with peptide-based methods, tar geted probes ha ve been prepared b y coupling of antibodies or antibody fragments to NIR reporters. In one early example, a NIR cyanine dy e with emission at 804 nm w as conjugated to anti-MUC1 antibodies for targeting and visualization of microcancers via an endoscope system. 77 Antibody probes for imaging atherosclerosis ha ve also been designed. To prepare the tar geted imaging agent, an antibody against the e xtra-domain B (ED-B) of fibronectin w as prepared. The ED-B is preferentiall y

As with other tar geting strate gies, the goal is to prepare probes with high specificity for targets of biologic interest. One approach is through small molecule screening. In a recent series of e xperiments, a librar y of nearl y 150 fluorophore-labeled iron oxide nanoparticles was created with different small molecules appended to the par ticle surfaces.84 Screening of this librar y for binding against a panel of dif ferent biorele vant cell types yielded se veral hits for potential de velopment into small molecule-based optical and magnetic resonance-based multimodal imaging

Optical Imaging Agents

agents. In the study , par ticles with demonstrated specificity to endothelial cells, activated macrophages, and pancreatic cancer cells w ere identif ied. The viability of one particle preparation for in vi vo targeting was verified. In this e xperiment, CLIO-isoatoic-Cy5.5 labeled par ticles were sho wn to accumulate in PcCa-2 pancreatic tumors implanted in a mouse model system. 84 It is not necessary, however, to run selectivity screens to prepare ne w small-molecule tar geted optical probes. For example, the folate receptor is a w ell-known surface receptor and is overexpressed in many tumor types including o varian cancer .85 A probe for tar geting the folate receptor w as prepared b y conjugation of a NIR fluorophore to folic acid via a h ydrophilic 2,2 ʹ′-(ethylenedioxy)-bis(ethylamine) spacer (F igure 6). 86 This probe was shown to tar get specif ically the folate receptors displayed on OVCAR3 cells.86 Other known binding pairs such as the asialo glycoprotein receptor that binds to galactose also can be used to develop small molecule optical imaging probes. 87 Small molecule constr ucts for imaging bacterial infection are also possib le. In one system, a ne w squaric acid-based dye was prepared and encapsulated in a rotaxane.88 This results in signif icant impro vements of the photostability and solubility of the core squaraine NIR fluorophore. To this ne w dy e, a bis-dipicol ylamine (DPA), zinc-chelating motif w as appended. These Zn 2+ DPA g roups ha ve a demonstrated selecti vity for the anionic surf aces of bacterial cells o ver the zwitterionic surface charges on health y animal cells. 89 In vivo imaging e xperiments show that pre-labeled Staphylococcus aureus and Escherichia coli can be imaged monitoring fluorescence at 670 nm in a mouse model system with the probe.88 In addition to probes that bind to cell surf aces or surface receptors, small molecule imaging agents that become chemically cross-linked into their desired target can be prepared. F or the de velopment of the sk eletal system in v ertebrates, h ydroxyapatite deposition b y

Figure 6.

413

osteoblasts is necessar y. Hydro xyapatite deposition is also a kno wn hallmark in atherosclerosis and cer tain cancers. Bisphosphonates ha ve similar chemical str ucture to h ydroxyapatite lattice and thus can be incor porated into growing hydroxyapatite deposits. To prepare a probe to visualize this osteob lastic activity, a NIR fluorophore was covalently attached to the primary amine of the bisphosphonate deri vative, palmidronate. 90 NIR optical imaging with the probe in 7-week-old nude mice indicated fluorescence in the bone follo wing IV injection of the probe. 90 Palmidronate-based bisphosphonate NIR constr ucts w ere used in lar ge-animal imaging experiments and sho w promise for use in diagnostic mammography imaging. 91 In a similar line of research, another bisphosphonate containing probe w as used for the visualization of the osteob lastic activity of valvular myofibroblasts in early stage aortic valve stenosis in an ApoE mouse model. 92

APTAMERS Targeted NIR imaging probes based on aptamer nucleic acid ligands have not received as much attention as the other targeted imaging methodologies. Several targeted aptamer probes have been prepared incor porating dyes, which emit in the visib le spectr um.93–95 More recently, QD-based aptamer -targeted probes for imaging and therapy, via delivery of attached doxorubicin molecules, were repor ted.96 Constructs using aptamer -targeting strategies for imaging bacterial spores with farred emitting QD conjugates ha ve been repor ted.97 To constr uct the bacteria imaging probe, an aptamer specif ic for Bacillus thuringiensis spores w as prepared b y systematic evolution of ligands b y exponential enrichment techniques and w as conjugated to QDs. This imaging agent w as subsequentl y used as a fluorescence-based B. thuringiensis spores.97 assay for detection of Although there has been little or no examination of NIR emitting aptamer conjugates for in vi vo imaging, the

Structure of the small molecule-based folate receptor targeted probe for tumor imaging.

414

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

previous demonstration of animal imaging with radiolabeled aptamers indicates that such an imaging strate gy is feasible.94

activatable optical imaging agents and describe se veral examples of each.

RESONANCE ENERGY TRANSFER AND STATIC QUENCHING

CELLULAR MIGRATION Imaging of cellular homing and tar geting in vi vo is a relatively new optical imaging paradigm. In this approach, cells are labeled e x vi vo and then injected into a mouse model. F rom there, the innate mig ration and tar geting tendencies of the labeled cells can be monitored in vi vo by optical spectroscopy. In one e xample, Cy5.5-tat was used to label T lymphocytes ex vivo.98 These labeled cells were then administered IV to rats with e xperimental autoimmune encephalomyelitis, a model for the inflammation process in multiple sclerosis. The ability of the labeledT lymphocytes to target and localize to m yelin in the brain w as subsequently monitored by NIR fluorescence spectroscop y.98 The in vi vo imaging and quantif ication of leuk ocyte immune responses can also be monitored using a similar system. In a model for cytotoxic T lymphocyte (CTL) immunotherapy, CTLs, which are kno wn to home in on hemagglutinin antigen, w ere labeled in vitro with the succinimidyl ester of a NIR fluorophore.99 FMT following IV injection of the labeled CTLs into mice bearing CT44 tumors that e xpress hemagglutinin antigen demonstrated migration of the NIR emitting CTLs to the tumor and adjacent lymph nodes.99

Activatable imaging agents based on resonance ener gy transfer and static quenching are the most common class of acti vatable imaging probes. These tw o mechanisms are often, but not al ways, used in tandem to generate fluorogenic probes based on enzyme activity. In resonance energy transfer -based quenching, w hich operates on principles similar to FRET , ener gy transfer occurs

A

B

ACTIVATABLE AGENTS Vast impro vements in signal to noise can be achie ved with activatable probes that are not possible with targeted imaging strate gies. Often, it is possib le to combine targeting and acti vation schemes in one imaging agent to realize further enhancements in signal and g reater selectivity for the desired tar get. As with tar geted probes, many acti vatable probes can be adapted for use with a variety of nanopar ticle platforms in order to tak e advantage of the inherent properties of the nanoscale materials. Due to the fle xibility af forded b y the photoph ysics of luminescent and fluorescent species, a v ariety of optical characteristics may be used for development into activatable probes. These proper ties include intensity shifts, wavelength shifts, chemiluminescene activation, and fluorescence lifetime changes. Most cur rent acti vatable probes can be g rouped into one of se veral general approaches based on the mechanisms used for their activation. Activation schemes ma y be based on static quenching, resonance energy transfer and/or FRET, PET signaling, chemical modif ication of the fluorophore, or chemiluminescence activation (Figure 7). The following sections will detail these strate gies for preparation of

C

D

E

Figure 7. Mechanistic strategies for the design of activatable imaging agents going from their quenched states (left) to their activated, fluorescent forms (right). A, Static quenching-based probes where physical contact of the dyes results in fluorescence quenching. B, resonance energy transfer-based quenching due to spectral overlap of the fluorophores. C, photoinduced electron transfer quenching that is disrupted upon analyte binding. D, chemical modification of the fluorophore, resulting in altered fluorescence emission, and (E) chemiluminescence-based activation.

Optical Imaging Agents

between dye molecules that are in close pro ximity. The net result of these energy transfer processes is relaxation of the e xcited states of the fluorophores back to the ground state without emission of a photon. A common strategy for design of enzyme acti vatable probes involves close placement of a high number of lik e dy e molecules, typically NIR c yanine dyes. Due to the relatively small Stok es shifts of this class of dy es, there is significant spectral o verlap between the emission band from one dye and the absorption band of an adjacent dye of the same type. It is this spectral o verlap that allo ws resonance ener gy transfer to occur from the emission band of one dy e back to the absor ption band of the adjacent dy e molecule, resulting in quenching of fluorescence emission. Unlike resonance energy transfer, which is an excited state process, static quenching occurs in the ground state. Static quenching requires ph ysical contact of the dy e molecules. This quenching mechanism typicall y occurs when fluorophores for m g round state comple xes via π interactions betw een the molecules. The for mation of these π-stacked comple xes often leads to v ery ef ficient quenching of fluorescence emission. Since this is a ground state process, dramatic changes in the absor ption spectra of the dy es are often obser ved. Probes in w hich their nonactivated states undergo resonance energy transfer and/or static quenching are typicall y activated via the incorporation of enzyme clea vable sequences that upon cleavage release the indi vidual dy e molecules from the probe, restoring their fluorescence emission. In one early example of an activatable probe, a polylysine g raft copol ymer (PGC) consisting of a pol ylysine polymer with appended metho xypolyethylene gl ycol (MPEG) pol ymers w as modif ied with a high density of NIR fluorophores.100 On average, the probe consisted of 92 MPEG pol ymers, 11 Cy5.5 fluorophores, and 44 free

415

lysine side chains per pol ylysine unit. 100 The close proximity of the Cy5.5 dy es in the probe resulted in a 15-fold quenching of the fluorescence emission. This is likely the result of a combination of resonance ener gy transfer and static quenching mechanisms, although the relative contribution of each quenching mechanism is unknown. The probe w as designed so that enzymes with Lys-Lys specificity such as trypsin could cleave the polylysine backbone, abro gating the dy e-dye interactions and restoring the fluorescence signal. In test e xperiments, the probe was incubated with tr ypsin and approximately 95% of the original fluorescence signal could be recovered, corresponding to a 12-fold activation. Strong activation of the probe was also obser ved upon incubation with LX-1 cancer cells, indicating that proteases from the cancer cells were capable of activating the probe. In vivo imaging with the acti vatable probe is also possib le. In mouse models implanted with LX-1 tumors, tumors as small as 300 µm in diameter could be detected , and the probe signal in the tumors could be monitored b y fluorescence reflectance imaging up to 96 hours post probe injection. 100 Improvements on this fluoro genic probe strate gy were quickly developed. Since a variety of proteases are known to be upre gulated during patho genesis, better specificity should be possib le if the probe can be targeted to specif ic proteases. To accomplish this, the design of the original acti vatable probe was modif ied to insert an enzyme specif ic cleavage sequence for cathepsin D betw een the NIR fluorophore and the PGC backbone (Figure 8). 101 Cathepsin D is an aspar tic protease, known to ha ve signif icantly increased e xpression in breast tumors. The probe is prepared b y star ting with PGC containing appro ximately 55 free amines per polymer. To this on average, 42 copies of the cathepsin D specific peptide sequence could be appended. Finally, up to 24 Cy5.5 fluorophores w ere conjugated to the amino

Figure 8. Schematic representation for the activation of a polylysine graft copolymer-based probe after enzymatic cleavage of the polymerattached near infrared fluorophores.

416

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

termini of the cathepsin D peptides to generate the f inal probe. The probe sho wed g reater than 99% quenching 101 when compared with the free fluorophore. Up to 60-fold activation of the probe could be obser ved after activation in in vitro e xperiments.102 In vi vo imaging experiments with the probe performed with mice bearing cathepsin D positi ve tumors demonstrated signalto-noise ratios betw een the tumor and the nontar get tissue of up to 22.8. 102 Using this highly flexible approach that relies on resonance ener gy transfer and static quenching mechanisms, acti vatable probes for other enzymes such as cathepsin K, Caspase-1, and MMP-2 have been reported.103–105 With these imaging agents, the specificity of the probe is only limited by the specificity of the particular enzyme cleavable sequence used. Recently, a small molecule enzyme acti vatable probe for cathepsin S w as prepared. This probe mak es use of a small, branched lysine peptide on to which four cathepsin S specif ic clea vage sequences can be conjugated. F our NIR fluorophores are then attached to the amino ter mini of the cleavage sequences via dif ferent length PEG spacers to prepare the intact probe.106 These probes are discrete single molecules with def ined molecular w eights unlik e the pre vious PGC-based enzyme acti vatable probes. Therefore, these branched lysine-based agents have a uniform, well-defined orientation of the indi vidual peptides and dyes. In PBS with 20% DMSO the probes on average show strong quenching of their fluorescence emission. 106 In addition clear , pronounced absor ption bands appro ximately 70 nm to the b lue of the parent dy e absorption at 780 nm are obser ved. These absor ption bands are indicative of strong static quenching. The relative contributions of the static and dynamic quenching mechanisms were fur ther investigated with the probes. Disr uption of the probes static quenching is possible by dissolution of the probe into DMSO, which more effectively solvates the NIR fluorophores of the probes. In DMSO , the shor twavelength absorption bands due to static quenching disappear. Under these conditions, in w hich an approximate 2-fold decrease in emission intensity is observed, the only probe quenching mechanism should be resonance ener gy transfer based. 106 This 2-fold quenching is in contrast to the 38-fold quenching obser ved from the same probe in PBS with 20% DMSO w here both static and resonance energy transfer quenching occur .106 From these e xperiments, it is clear that the predominate quenching mechanism in the branched lysine probes is based on static quenching. Initial screening studies against cathepsin S in PBS buffer, where the static quenching is e xpected to be maximized, show a g reater than 70-fold acti vation in the NIR fluorescence signal from the probe. 106

Systems that rel y e xclusively on either static quenching or resonance ener gy transfer ma y also be prepared. In one e xample, static quenching betw een far-red or NIR fluorophores and tr yptophan was investigated.107 Since the absorption spectrum of tryptophan does not overlap with the absor ption or emission from the fluorophores used in this study , static quenching is the only possible mechanism by which the fluorophore emission can be attenuated. In the best systems, up to 99% quenching of the fluorophore emission b y a tr yptophan residue conjugated directl y to the dy e w as observed. 107 Even w hen a gl ycine tripeptide w as inserted betw een the fluorophore and the tr yptophan residue, up to 12-fold quenching w as still obser ved.107 Another system, designed around resonance ener gy transfer quenching, used a Cy5-quencher pair , w here the quencher dye has an absorption spectrum that overlaps ef ficiently with the emission from the Cy5. This system allo ws for ef ficient resonance ener gy transfer between the donor (Cy5) and the acceptor quencher molecule. The dy es w ere tethered to each other with β-lactamase clea vage sequence resulting in an essentially nonfluorescent quenched probe. 108 Treatment of the probe with β-lactamase results in a 57-fold increase in signal as the probe is cleaved, and the Cy5 is allowed to diffuse away from the quencher dye.108 Activatable probes appended to nanopar ticle scaffolds may also be prepared. In one e xample, Cy5.5 was covalently link ed to an iron o xide nanopar ticle via a protease cleavable polyarginine sequence. 109 The probe also contained Cy7 fluorophores conjugated to the particle surf ace via a nonenzymaticall y clea vable link er. This system displayed an up to 8.5-fold activation of the Cy5.5 signal upon treatment of the nanopar ticles with trypsin.109 Since it is not cleavable, the Cy7 signal from the probe could potentiall y be used for monitoring the probe distribution, independent of enzyme acti vation.

PET Probes using PET as an acti vation scheme ha ve received considerab le attention in recent y ears.110–112 This acti vation strate gy is par ticularly w ell suited for development of probes for small molecule anal ytes. Design strategies rely on preparing a probe that in the absence of anal yte undergoes a PET process attenuating the fluorescence emission of the fluorescent reporter. PET-based analyte detection systems typically involve placement of an electron donor , such as an aliphatic amine, in close proximity to a fluorophore. In the absence of the anal yte, an electron from the amine

Optical Imaging Agents

is ab le to transfer to the fluorophore after it has been excited. This electron transfer process pre vents deca y of the excited fluorophore back to its g round state and emission of a lower ener gy photon (fluorescence). Binding of the anal yte or chemical reacti vity at the amine alters the ener gy le vels of the amine-centered molecular orbitals, decreasing the probability of electron transfer from the amine to the fluorophore, and restores the nati ve fluorescence emission of the dy e. Although this activation strategy has been the focus of significant interest in the past decade for de velopment of analyte sensors, relati vely little w ork has been conducted for the de velopment of PET -based sensors for biologically rele vant anal ytes that ha ve their fluorescence emission in the far-red to NIR. This is in part due to the decreased efficiency of the PET process between fluorophore and donor as the emission w avelength of the fluorophore increases. Longer w avelength fluorophores emit lo wer ener gy radiation, and thus the energy gap betw een the e xcited state and the g round state of the fluorophore is smaller . This increases the difficulty in matching an electron donor with the dye to ensure efficient PET in the absence anal yte. Although more difficult, it is still possib le to design PET -based analyte sensors that operate in the NIR. In one of the earliest NIR fluoro genic sensors based on a PET mechanism, a bis(2-aminopheno xy)ethaneN,N,N´,N´-tetraacetic acid (B APTA) chelating g roup, which is selective for Ca 2+, was appended of a NIR c yanine fluorophore. 113 The probe has absor ption and emission maxima at 766 and 782 nm, respecti vely. With an extinction coefficient of 200,000 M−1cm−1 and a quantum yield of 12% in the Ca 2+ bound state, this is a bright NIR probe for calcium ions. In the absence of calcium, the anilinic nitrogen atoms on the B APTA chelator ser ve as PET donors to quench the fluorescence emission of the probe. However, this quenching is not very efficient, and only a 4-fold increase in fluorescence signal is obser ved upon coordination of the B APTA moiety to Ca 2+.113 Despite this relati vely w eak tur n on signal, the K d for Ca2+ binding is 240 nM, w hich is similar to e xpected intracellular calcium ion concentrations. 113 More recently, a NIR probe (DIPCY) was developed for detection of Zn 2+. This probe consists of a c yanine dye, which has a dipicolylethylenediamine zinc coordination motif attached to the core of the fluorophore via secondary amine. 114 By joining the chelating g roup to the dye with an amine that is directl y attached to the conjugated methine backbone of the dy e, a long Stok es shift dye is generated. In the absence of Zn2+, the DIPCY

417

probe absorbs at 627 nm, has emission at 758 nm, has an extinction coefficient of 70,000 M−1cm−1, and has a quantum yield of 2% in aqueous buffer.114 After binding Zn2+, with a K d of 98 nM, the absor ption shifts to 671 nm, emission is at 765 nm, the extinction coefficient increases to 85,000 M −1cm−1, and the quantum yield remains approximately 2%. Even though this probe does not display any effective PET-based quenching, it is still useful as a NIR probe for zinc. Because the absor ption wavelength of the dy e shifts from 627 to 671 nm upon Zn 2+ coordination, it can be used as a ratiometric NIR probe for zinc ions in which the ratio of the fluorescence emission at 760 nm using an e xcitation wavelength of 671 or 627 nm is monitored. 114 The PET quenching process is not necessaril y limited to probes, w hich sho w less than 4-fold acti vation upon analyte binding. An alternate approach for preparation of a NIR Zn 2+ sensor w as recentl y detailed that makes use of a trisulfonated porphyrin fluorophore bearing two DPA zinc chelating moieties. 115 The probe has strong emission at 648 nm and a w eaker emission band at 715 nm in aqueous buffer. The Kd for binding of Zn 2+ ion to for m a highl y fluorescent 1:1 comple x is 12 nM. After binding Zn 2+, the quantum yield of the probe increases from 0.4 to 4.6%, a g reater than 10-fold increase.115 This magnitude of fluorescence response should be sufficient for potential in vi vo imaging applications. In addition to its fluorescence sensing properties, the probe has a dual use. When the por phyrin macrocycle is metallated with Mn3+, the system becomes nonfluorescent but can be used as a magnetic resonance active probe for Zn 2+.115 It is possib le to design PET -based NIR probes for analytes other than metal ions. F or e xample, o-phenylenediamines are w ell-known motifs that react with nitric oxide (NO) in the presence of O2 to form a triazole species. This scheme has been used to prepare PET-based probes for NO with emission in the visible.111 By attachment of a similar o-phenylenediamine moiety to a NIR c yanine dye, it was possible to prepare a fluorogenic NIR sensor for NO .116 In this case, the diamine serves as a PET donor that is well matched to the carbocyanine dy e. Upon reaction of the w ater-soluble probe (DAC-S) with NO in aqueous media, the diamine is converted to a triazole, which no longer undergoes efficient PET with the fluorophore, thus resulting in a 14-fold increase in fluorescence emission (F igure 9). 116 This probe w as fur ther used to monitor NO generation ex vivo in mouse kidne ys perfused with the NO donor compound NOC13.117

418

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 9. Activation of a photoinduced electron transfer-quenched nitric oxide probe, DAC-S, resulting in generation of a fluorescent triazole species.

CHEMICAL MODIFICATION Activation strate gies need not rel y on dy e-dye or dye-quencher interactions. It is possible to design systems in which the chemical and therefore electronic structure of the dye itself is modified. As shown in Figure 10, there are three general classes of unsaturated h ydrocarbon compounds, aromatics (characterized b y 4 n + 2 π electrons), polyenes (characterized by alternating π bond orders), and polymethine species (typified by n ± 1π electrons, equal π bond orders, and long-w avelength absorption). Events or chemical reactions that modify the electronics of an organic species to make the dye more or less polymethinelike can be used to de velop acti vatable imaging agents. This is exemplified by fluorescein (see F igure 10). When in acidic solution, the pheno xy g roup of fluorescein becomes protonated, generating a neutral dye, which has a more polyene- or aromatic-lik e electronic str ucture with weak fluorescence emission and an absor ption maximum at appro ximately 450 nm. When fluorescein is deprotonated in basic media, the xanthene core of the dy e has a negative charge and n ± 1π electrons. The absor ption of the deprotonated fluorescein red-shifts to 490 nm and the quantum yield rises to 93%. 118 Approaches that rel y on this type of electronic modification of the fluorophore can often achie ve w ell o ver 100-fold acti vation. Due to its potential for extremely strong fluorescence activation, this approach is be ginning to recei ve more attention as a mechanism for de velopment of acti vatable NIR optical imaging agents. As a result of its pH-dependent fluorescence response, fluorescein has been used as a pH sensor. Imaging of pH in biolo gic systems is desirab le and is associated with disease states such as cancer , c ystic f ibrosis, asthma, and a variety of renal conditions. 119–121 However, because of its visib le emission and decrease in fluorescence as the pH decreases, fluorescein is not a viab le probe monitoring pH in vi vo. NIR pH sensiti ve dy es,

however, have been known since 1940s.122 More recently, several w ater-soluble pH-responsi ve c yanine dy es ha ve been reported. Cyanine dyes can be made sensitive to pH by dealkylation of one or both of the dy es’ indole nitrogens. When dealkylated, protonation of the indole nitrogens generates a positi ve char ge on the chromophore, resulting in long-wavelength polymethine-like absorption and NIR fluorescence emission. If the pH is raised , the indole nitro gen is deprotonated , lea ving a neutral more polyene-like dy e with shor t-wavelength absor ption and little or no fluorescence emission (Figure 11). In addition to confer ring pH responsi ve proper ties to c yanine dyes, the pK as of these dy es can be modulated b y appending different electron withdra wing g roups on the methine backbone of the dy e.123 With this approach, the pK a of a series of Cy5 deri vatives can be modulated betw een 5.7 and 6.3, which should be optimal for imaging acidic environments associated with tumors, c ystic f ibrosis, and renal conditions. 123 To demonstrate the potential of pH responsive cyanine dyes, one dy e with a pK a of 6.3 w as used to visualize mouse urine (pH 6.78) w hile concomitantly showing no fluorescence emission in whole mouse blood.123 Longer w avelength pH sensiti ve c yanine dy es based on an ICG dye core can also be prepared. 124 These dyes have been repor ted to ha ve pK a values of appro ximately 7.2, w hich may have applications in pH imaging of events near ph ysiological pH. 124 It is also possib le to prepare w ater-soluble pH dy es with signif icantly lo wer pKas.125 These are based on a c yanine dye which undergoes a keto-hydroxy equilibrium. When deprotonated, the dyes e xist in the k eto for m, with shor t-wavelength absorption. Protonation of the dy es yields the h ydroxy form, in which the typical cyanine dye optical properties are manifest. Dy es, based on the k eto-hydroxy equilibrium, ha ve pK a values of appro ximately 4.5. 125 Thus, these fluorophores ma y f ind utility in imaging endocytosis into the acidic endosome/l ysosome system where pH can range from 4.5 to 6.5. 126

Optical Imaging Agents

A

419

B

C

Figure 10. Mechanistic strategy for imaging agents based on chemical modification of the fluorophore. A, The three classes of unsaturated hydrocarbons in which compounds with a polymethine-like structure are typified by long-wavelength absorption and often strong fluorescence emission. B, an activation strategy for probes based on chemical modification where alteration of the unsaturated organic structure can shift the electronics of the compound, making it more (or less) polymethine-like. C, deprotonation of fluorescein shifts the structure from a weakly fluorescent species to a highly fluorescent polymethine-like form with n-1 electrons in the conjugated π-system.

A

B

Figure 11. Fluorescence switching of a pH-responsive imaging probe with a pKa of 5.7 where protonation of the fluorophore generates a dye with cyanine-like absorption and emission properties (A). Optical spectra of the pH probe (B) showing absorption of the dye at pH 10 (solid line), absorption at pH 3 (dashed line), and fluorescence emission at pH 3 (dotted line).

Probes based on chemical activation are not limited to deprotonation-protonation e vents. Imaging agents for enzyme activity may also be prepared. In one e xample, a water-soluble nile blue derivative was modif ied on one of the chromophore amine groups with amino acid residues.20 This modif ication changes an amine (w hich is a good donor) into an amide (poor donor), thus resulting in quenching of the long-w avelength emission of the nile blue. Enzymatic clea vage of the amide refor ms the free nile b lue deri vative and results in an increase in fluorescence emission at 675 nm. 20 Naphthofluorescein also has been used as a dy e scaf fold for preparation

of enzyme acti vatable probes. A probe for alkaline phosphatase was recently repor ted by treatment of naphthofluorescein with POCl3 to generate naphthofluoresceindiphosphate. The probe is acti vated b y incubation with alkaline phosphatase, resulting in an increase in fluorescence emission at 660 nm. 127 Fluorogenic probes based on a chemical modif ication strategy have been particularly useful for the design of imaging agents that are sensiti ve to reacti ve oxygen species (R OS). R OS such as supero xide radical anion, hydroxy radical, h ypochlorite, h ydrogen pero xide, NO , and peroxynitrite have myriad pathogenic and physiological roles in biolo gy. Naphthofluorescein-based probes for h ydrogen pero xide were recentl y prepared b y reaction of sulfon yl chloride deri vatives with naphthofluorescein to generate the sulfonate ester modif ied dyes.128 These dyes show good selectivity for hydrogen peroxide over other R OS.128 More recently, an alter native activation scheme for h ydrogen pero xide sensiti ve imaging agents w as proposed. 129 In this system, the pheno xy moieties of resor ufin or fluorescein deri vatives w ere replaced by an ar yl boronate g roup, rendering the dy es essentially nonfluorescent. Reaction of the probes with hydrogen pero xide restores the fluorescence emission with good selecti vity for h ydrogen pero xide over other biologic ROS.129 Sensors for imaging additional oxidants have also been developed. Hypochlorite is a highl y reacti ve R OS implicated in the pathogenesis of atherosclerosis and other inflammatory states. Fluorescein deri vatives modif ied with p-aminophenols w ere repor ted to be sensiti ve for hypochlorous acid (HOCl) and to a lesser e xtent hydroxy radical.130 Recently, a f ar-red emitting HOCl probe based

420

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

on a similar activation strategy was developed. This imaging agent uses a disulfonated naphthofluorescein scaf fold. The sulfonaphthofluorescein has impro ved water solubility and further red shifted emission at 676 nm than naphthofluorescein.131 Upon reaction of the sulfonaphtho aminophenyl fluorescein (SN APF) probe with h ypochlorous acid , the p-aminophenyl ether is clea ved, re generating free sulfonaphthofluorescein (F igure 12). 131 Unlike the fluoresceinbased HOCl probes, SN APF is acti vated e xclusively b y hypochlorite. In vitro, SN APF has been used to identify HOCl produced b y m yeloperoxidase containing human macrophages. It is also ab le to visualize HOCl in fresh human endar terectomies and in vi vo using a mouse peritonitis model system.131 Recent interest in imaging thiols has spur red the development of thiol specif ic probes. Thiols such as glutathione and homoc ysteine play key roles in maintaining redo x homeostasis. Chemical modif ication schemes for thiol acti vatable probes ha ve been reported using visible NBD fluorophores and a far-red emitting hemic yanine deri vative.132,133 The acti vation strategy used b y these probes is based on thiol mediated cleavage of an aryl sulfonamide protecting group, which alters the fluorophore optical proper ties. The merocyanine-based probe sho ws e xcellent selecti vity for thiols such as L-c ysteine, glutathione, and dithiothreitol.133

CHEMILUMINESCENCE Chemiluminescence and bioluminescence are processes through w hich photons are released as the result of a chemical reaction, w here no e xcitation light source is necessary. Typically chemiluminescent species have visible emission, but systems can be tuned to gi ve f ar-red and

Figure 12.

NIR emission. In one example, bioluminescent Renilla reniformis luciferase w as conjugated to the surf ace of QDs.134 The probe is acti vated in the presence of the luciferase substrate, coelenterazine. The reaction of coelenterazine with the luciferase results in bioluminescence emission at 480 nm. Due to the broad excitation spectrum for QDs, this ener gy can be absorbed b y the QDs via a FRET process and emission from the QD is obser ved.134 This process is compatib le with QDs that ha ve emission bands at 605, 655, 705, and 800 nm. 134 The system holds promise for in vivo imaging applications due to dramatically improved signal to noise ratios, w hich result from the lack of need for an e xternal light source for probe excitation. Chemiluminescence may also be used as an acti vation scheme for imaging the biolo gically relevant analyte h ydrogen pero xide. The chemiluminescent probe consists of pol ymeric nanoparticles that contain pero xalate esters and are doped with a f ar-red emitting pentacene dy e. In the presence of h ydrogen pero xide, the peroxalates react to for m high ener gy dioxetanediones, which proceed to chemically excite the polymer embedded pentacene dy es, and emission at 630 nm is observed.135 This system is capab le of visualizing hydrogen peroxide in low µM concentrations in a model system where hydrogen peroxide and the par ticles were co-injected intramuscularly into mice. It is also possible to image endogenous hydrogen peroxide via this peroxalate ester -based chemiluminescent acti vation scheme in a mouse model w here inflammation is induced via LPS injection into the peritoneal cavity of the mouse.135 As with the bioluminescent QD system, no e xternal illumination source is necessar y for in vi vo imaging, resulting in the potential for dramaticall y impro ved signal-to-noise levels.

Chemical activation of sulfonaphthoaminophenyl fluorescein after hypochlorite mediated cleavage of the p-aminophenol moiety.

Optical Imaging Agents

SUMMARY Over the past 10 y ears, the di versity of optical imaging agents has expanded tremendously. This is in part due to the rapid adv ances in the design of ne w luminescent reporter molecules, such as improved NIR emitting, conjugatable small molecule fluorophores. Recent y ears have also seen an increase in the use of more photostable semiconductor QDs as components in a range of targeted and acti vatable imaging agents. With these and other new optical reporters, the breadth of innovative targeting and activation strategies for preparation of NIR optical imaging probes that enab le visualization of the molecular e vents associated with disease in vi vo ha ve been expanded greatly. Future improvements such as brighter, more photostab le or ganic fluorophores, materials with emission greater than 850 nm (enab ling development of new optical imaging channels), and the use of new luminescent materials will likely spur further advances in the design of more efficient optical imaging agents. Furthermore, probes incor porating these ne wer luminescent materials as components of systems relying of resonance energy transfer, static quenching, PET, chemical modification, or chemiluminescence activation schemes should result in more sensiti ve imaging agents. The combination of these acti vation strategies with tar geting g roups and nanopar ticle delivery v ectors is lik ely to of fer further enhancements in specif icity and signal-to-noise ratios. Improved optical materials with ne w photophysical characteristics ma y also enab le the de velopment of activation strate gies such as those based on shifts in luminescence lifetimes after target or analyte binding. In summary, the future of optical imaging agent de velopment is bright. Advances in the basic chemistr y for the design of ne w optical imaging probes ma y ultimatel y contribute to the de velopment of ne w materials for earlier and improved disease diagnosis in a clinical setting.

REFERENCES 1. Weissleder R, Ntziachristos V. Shedding light onto live molecular targets. Nat Med 2003;9:123–8. 2. Mujumdar RB, Ernst LA, Mujumdar SR, et al. Cyanine dy e labeling reagents: sulfoindocyanine succinimidyl esters. Bioconjug Chem 1993;4:105–11. 3. Mujumdar RB, Ernst LA, Mujumdar SR, Waggoner AS. Cyanine dye labeling reagents containing isothioc yanate g roups. Cytometr y 1989;10:11–9. 4. Mujumdar SR, Mujumdar RB , Grant CM, Waggoner AS. Cyaninelabeling reagents: sulfobenzindocyanine succinimidyl esters. Bioconjug Chem 1996;7:356–62. 5. Hofer LJE, Grabenstetter RJ , Wiig EO. The fluorescence of c yanine and related dy es in the monomeric state. J Am Chem Soc 1950;72:203–9.

421

6. Hilderbrand SA, Kelly KA, Weissleder R, Tung C-H. Monofunctional near-infrared fluorochromes for imaging applications. Bioconjug Chem 2005;16:1275–81. 7. Flanagan JHJ , Khan SH, Menchen S, et al. Functionalized tricarbocyanine dyes as near -infrared fluorescent probes for biomolecules. Bioconjug Chem 1997;8:751–6. 8. Strekowski L, Mason CJ, Lee H, et al. Synthesis of water-soluble nearinfrared cyanine dyes functionalized with [(succinimido)o xy]carbonyl group. J Heterocycl Chem 2003;40:913–6. 9. Narayanan N , P atonay G. A ne w method for the synthesis of heptamethine c yanine dy es: synthesis of ne w near -infrared fluorescent labels. J Org Chem 1995;60:2391–5. 10. Pham W, Lai W-F, Weissleder R, Tung C-H. High efficiency synthesis of a bioconjugatab le near -infrared fluorochrome. Bioconjug Chem 2003;14:1048–51. 11. Lin Y, Weissleder R, Tung C-H. No vel near -infrared c yanine fluorochromes: synthesis, proper ties, and bioconjugation. Bioconjug Chem 2002;13:605–10. 12. Peng X, Song F, Lu E, et al. Heptamethine c yanine dyes with a large stokes shift and strong fluorescence: a paradigm for e xcitatedstate intramolecular char ge transfer . J Am Chem Soc 2005;127:4170–1. 13. Borbas KE, Mroz P, Hamblin MR, Lindsey JS. Bioconjugatable porphyrins bearing a compact s wallowtail motif for w ater solubility. Bioconjug Chem 2006;17:638–53. 14. Sibrian-Vazquez M, Or tiz J, Nesterova IV, et al. Synthesis and properties of cell-tar geted Zn(II)-phthaloc yanine-peptide conjugates. Bioconjug Chem 2007;18:410–20. 15. McCarthy JR, P erez JM, Br uckner C, Weissleder R. P olymeric nanoparticle preparation that eradicates tumors. Nano Lett 2005;5:2552–6. 16. Peng X, Drane y DR, Volcheck WM, et al. Phthaloc yanine dye as an extremely photostable and highly fluorescent near-infrared labeling reagent. Proc SPIE Int Society Opt Eng 2006;6097:60970E/160970E/12. 17. Duncan TV, Susumu K, Sinks LE, Therien MJ . Exceptional near infrared fluorescence quantum yields and excited-state absorptivity of highl y conjugated por phyrin ar rays. J Am Chem Soc 2006;128:9000–1. 18. Ghoroghchian PP, Frail PR, Susumu K, et al. Near -infrared-emissive polymersomes: self-assemb led soft matter for in vi vo optical imaging. Proc Natl Acad Sci U S A 2005;102:2922–7. 19. Frade VHJ, Goncalves MST, Moura JCVP. Synthesis and fluorescence proper ties of side-chain carbo xylated 5,9diaminobenzo[a]phenoxazinium salts. Tet Lett 2005; 46:4949–52. 20. Ho N-H, Weissleder R, Tung C-H. Development of water-soluble farred fluoro genic dy es for enzyme sensing. Tetrahedron 2006;62:578–85. 21. Molecular Probes Inc, Eugene, OR, U S A. Available at: http://probes.invitrogen.com. 22. LI-COR Biosciences, Lincoln, NE, U S A. Available at: http:// www.licor.com. 23. VisEn Medical Inc, Woburn, MA, U S A. Available at: http:// www.visenmedical.com. 24. GE Healthcare Bio-Sciences Cor p, Piscataway, NJ, U S A. Available at: http://www.gelifesciences.com. 25. Dyomics GmbH, Jena, Ger many. Available at: http:// www.dyomics.com. 26. ATTO-TEC GmbH, Siegen, Germany. Available at: http://atto-tec.com. 27. Ow H, Larson DR, Siri vastava M, et al. Bright and stab le core-shell fluorescent silica nanoparticles. Nano Lett 2005;5:113–7. 28. Wang L, Tan W. Multicolor FRET silica nanoparticles by single wavelength excitation. Nano Lett 2006;6:84–8. 29. Nakajima M, Takeda M, Kobayashi M, et al. Nano-sized fluorescent particles as new tracers for sentinel node detection: e xperimental

422

30.

31. 32.

33.

34. 35.

36.

37.

38.

39. 40. 41.

42.

43.

44.

45. 46. 47.

48.

49.

50.

51.

52.

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

model for decision of appropriate size and wavelength. Cancer Sci 2005;96:353–6. Josephson L, Kircher MF , Mahmood U, et al. Near -infrared fluorescent nanoparticles as combined MR/optical imaging probes. Bioconjug Chem 2002;13:554–60. Kelly KA, Waterman P, Weissleder R. In vivo imaging of moleculary targeted phage. Neoplasia 2006;8:1011–8. Hines MA, Guy ot-Sionnest P . Synthesis and characterization of strongly luminescing ZnS-capped CdSe nanocr ystals. J Phys Chem 1996;100:468–71. Manna L, Scher EC, Li L-S, Alivisatos AP. Epitaxial growth and photochemical annealing of g raded Cd/ZnS shells on colloidal CdSe nanorods. J Am Chem Soc 2002;124:7136–45. Chan WCW, Nie S. Quantum dot bioconjugates for ultrasensiti ve nonisotopic detection. Science 1998;281:2016–8. Gao X, Cui Y, Levenson RM, et al. In vivo cancer targeting and imaging with semiconductor quantum dots. Nat Biotechnol 2004;22:969–76. Kong X, Huang LCL, Liau S-CV , et al. P olylysine-coated diamond nanocrystals for maldi-T OF mass anal ysis of DN A oligonucleotides. Anal Chem 2005;77:4273–7. Yu S-J, Kang M-W , Chang H-C, et al. Bright fluorescent nanodiamonds: no photobleaching and low cytotoxicity. J Am Chem Soc 2005;127:17604–5. Fu C-C, Lee H-Y, Chen K, et al. Characterization and application of single fluorescent nanodiamonds as cellular biomark ers. Proc Natl Acad Sci U S A 2007;104:727–32. Huang T, Murray RB. Visible luminescence of water-soluble monolayerprotected gold clusters. J Phys Chem B 2001;105:12498–502. Wang G, Huang T, Murray RB, et al. Near-IR luminescence of monolayer protected metal clusters. J Am Chem Soc 2005;127:812–3. Zhang J , Fu Y, Lak owicz JR. Luminescent images of single gold nanoparticles and their labeling on silica beads. Opt Express 2007;15:13415–20. Zheng J, Zhang C, Dickson RM. Highl y fluorescent, w ater-soluble, size-tunable gold quantum dots. Ph ys Re v Lett 2004;93:0774021–4. Shi X, Wang S, Meshinchi S, et al. Dendrimer -entrapped gold nanoparticles as a platform for cancer-cell targeting and imaging. Small 2007;3:1245–52. Sun Y-P, Zhou B , Lin Y, et al. Quantum-sized carbon dots for bright and colorful photoluminescence. J Am Chem Soc 2006;128:7756–7. Cao L, Wang X, Meziani MJ , et al. Carbon dots for multiphoton bioimaging. J Am Chem Soc 2007;129:11318–9. Liu H, Ye T, Mao C. Fluorescent carbon nanopar ticles derived from candle soot. Angew Chem Int Ed Engl 2007;46:6473–5. Xu X, Ra y R, Gu Y, et al. Electrophoretic anal ysis and purif ication of fluorescent single-walled carbon nanotube fragments. J Am Chem Soc 2004;126:12736–7. O’Connell MJ , Bachilo SM, Huf fman CB , et al. Band gap fluorescence from indi vidual single-w alled carbon nanotubes. Science 2002;297:593–6. Leeuw TK, Reith RM, Simonette RA, et al. Single-w alled carbon nanotubes in the intact or ganism: near-IR imaging and biocompatibility studies in drosophila. Nano Lett 2007;7:2650–4. Kim S, Lim Y-T, Soltesz EG, et al. Near -infrared fluorescent type II quantum dots for sentinel l ymph node mapping. Nat Biotechnol 2004;22:93–7. Kircher MF, Mahmood U, King RS, et al. A multimodal nanoparticle for pre-operati ve magnetic resonance imaging and intraoperative optical brain tumor delineation. Cancer Res 2003;63:8122–5. Chen X, Conti PS, Moats RA. In vi vo near -infrared fluorescence imaging of inte grin αvβ3 in brain tumor x enographs. Cancer Res 2004;64:8009–14.

53. von Wallbrunn A, Holtke C, Zuhlsdorf M, et al. In vivo imaging of integrin αvβ3 expression using fluorescence-mediated tomography. Eur J Nucl Med Mol Imaging 2007;34:745–54. 54. Garanger E, Botur yn D, Jin Z, et al. Ne w multifunctional molecular conjugate v ector for tar geting, imaging, and therap y of tumors. Mol Ther 2005;12:1168–75. 55. Jin Z-H, Josserand V, Foillard S, et al. In vivo optical imaging of integrin αvβ3 in mice using multi valent or monovalent cRGD targeting vectors. Mol Cancer 2007;6:41. 56. Cheng Z, Wu Y, Xiong Z, et al. Near -infrared fluorescent RGD peptides for optical imaging of inte grin αvβ3 expression in li ving mice. Bioconjug Chem 2005;16:1433–41. 57. Monet X, Monet-Abou K, Re ynolds F, et al. Nanopar ticle imaging of integrins on tumor cells. Neoplasia 2006;8:214–22. 58. Cai W, Chen K, Li Z-B, et al. Dual-function probe for PET and near infrared fluorescence imaging of tumor v asculature. J Nucl Med 2007;48:1862–70. 59. Tung C-H, Ho N-H, Zeng Q, et al. No vel f actor XIII probes for blood coagulation imaging. Chembiochem 2003;4:897–9. 60. Jaffer FA, Tung C-H, Wykrzykowska JJ, et al. Molecular imaging of factor XIIIa acti vity in thrombosis using a no vel, near -infrared fluorescent contrast agent that covalently links to thrombi. Circulation 2004;110:170–6. 61. Lefkovits J, Plow EF, Topol EJ. Platelet glycoprotein IIb/IIIa receptors in cardio vascular medicine. Ne w Engl J Med 1995;332: 1553–9. 62. Hayashi Y, Katada J , Sato Y, et al. Disco very and str ucture-activity relationship studies of a no vel and specif ic peptide motif, pro-XX-X-asp-X as a platelet f ibrinogen receptor antagonist. Bioor g Med Chem 1998;6:355–64. 63. Tung C-H, Quinti L, Jaffer FA, Weissleder R. A branched fluorescent peptide probe for imaging of acti vated platelets. Mol Phar m 2005;2:92–5. 64. Simberg D , Duza T, P ark JH, et al. Biomimetic amplif ication o f nanoparticle homing to tumors. Proc Natl Acad Sci U S A 2007;104:932–6. 65. Citrin D, Lee AK, Scott T, et al. In vi vo tumor imaging in mice with near-infrared labeled endostatin. Mol Cancer Ther 2004;3:481–8. 66. Citrin D, Scott T, Sproull M, et al. In vivo tumor imaging using a nearinfrared-labeled endostatin molecule. Int J Radiat Oncol Biol Phys 2004;58:536–41. 67. Ke S, Wen X, Gurf inkel M, et al. Near -infrared optical imaging of epidermal g rowth f actor receptor in breast cancer x enografts. Cancer Res 2003;63:7870–5. 68. Schnellenberger EA, Bogdanov AJ, Petrovsky A, et al. Optical imaging of apoptosis as a biomark er of tumor response to chemotherapy. Neoplasia 2003;5:187–92. 69. Åkerman ME, Chan WCW, Laakkonen P, et al. Nanocrystal targeting in vivo. Proc Natl Acad Sci U S A 2002;99:12617–21. 70. Moore A, Medarova Z, Potthast A, Dai G. In vivo targeting of underglycosylated MUC-1 tumor antigen using a multimodal imaging probe. Cancer Res 2004;64:1821–7. 71. Medarova Z, Pham W, Farrar C, et al. In vivo imaging of siRNA delivery and silencing in tumors. Nat Med 2007;13:372–7. 72. Kelly KA, Allport JR, Tsourkas A, et al. Detection of v ascular adhesion molecule-1 e xpression using a no vel multimodal nanopar ticle. Circ Res 2005;96:327–36. 73. Kelly KA, Alencar H, Funovics M, et al. Detection of in vasive colon cancer using a novel, targeted, library-derived fluroescent peptide. Cancer Res 2004;64:6247–51. 74. Newton JR, Kelly KA, Mahmood U, et al. In vi vo selection of phage for the optical imaging of PC-3 human prostate carcinoma in mice. Neoplasia 2006;8:772–80. 75. Chen L, Zurita AJ, Ardelt PU, et al. Design and validation of a bifunctional ligand displa y system for receptor tar geting. Chem Biol 2004;11:1081–91.

Optical Imaging Agents

76. Andreev OA, Dupuy AD, Segala M, et al. Mechanism and uses of a membrane peptide that tar gets tumors and other acidic tissues in vivo. Proc Natl Acad Sci U S A 2007;104:7893–8. 77. Tadatsu M, Ito S, Mugur uma N , et al. A ne w fluorescent-labeling agent and labeled antibody for diagnosing microcancers. Bioor g Med Chem 2003;11:3289–94. 78. Matter CM, Schuler PK, Alessi P, et al. Molecular imaging of atherosclerotic plaques using a human antibody against the e xtradomain B of f ibronectin. Circ Res 2004;95:1225–33. 79. Hilger I, Leistner Y, Berndt A, et al. Near-infrared fluorescence imaging of HER-2 protein o ver-expression in tumor cells. Eur Radiol 2004;14:1124–9. 80. Sampath L, Kw on S, K e S, et al. Dual-labeled trastuzumab-based imaging agent for the detection of human epider mal growth factor receptor 2 o ver-expression in breast cancer . J Nucl Med 2007;48:1501–10. 81. Hansch A, Frey O, Sauner D, et al. In vi vo imaging of e xperimental arthritis with near -infrared fluorescence. Arthritis Rheum 2004;50:961–7. 82. Hsu ER, Anslyn EV, Dhar mawardhane S, et al. A f ar-red florescent contrast agent to image epider mal growth factor receptor expression. Photochem Photobiol 2004;79:272–9. 83. Rosenthal EL, Kulbersch BD, King T, et al. Use of fluorescent labeled anti-epidermal growth factor receptor antibody to image head and neck squamous cell carcinoma x enografts. Mol Cancer Ther 2007;6:1230–8. 84. Weissleder R, K elly KA, Sun EY , et al. Cell-specif ic tar geting of nanoparticles by multivalent attachment of small molecules. Nat Biotechnol 2005;23:1418–23. 85. Wang S, Lee RJ , Mathais CJ , et al. Synthesis, purif iciation and tumor cell uptake of 67Ga-deferoxamine-folate, a potential radiopharmaceutical for tumor imaging. Bioconjug Chem 1996;7:56–62. 86. Tung C-H, Lin Y, Moon WK, Weissleder R. A receptor-targeted nearinfrared fluorescence probe for in vi vo tumor imaging. Chembiochem 2002;3:784–6. 87. Vera DR, Hall DJ , Hoh CK, et al. Cy5.5-DTP A-galactosyl-dextran: a fluorescent probe for in vivo measurement of receptor biochemistry. Nucl Med Biol 2005;32:687–93. 88. Johnson JR, Fu N , Arunkumar E, et al. Squaraine rotaxanes: superior substitutes for Cy-5 in molecular probes for near-infrared fluorescence cell imaging. Angew Chem Int Ed Engl 2007;46:5528–31. 89. Leevy WM, Gammon ST, Jiang ST, et al. Optical imaging of bacterial infection in li ving mice using a fluoresent near -infrared molecular probe. J Am Chem Soc 2006;128:16476–7. 90. Zaheer A, Lenkinski RE, Mahmood A, et al. In vi vo near-infrared fluorescence imaging of osteob lastic acti vity. Nat Biotechnol 2001;10:1148–54. 91. Bhushan KR, Tanaka E, Frangioni JV. Synthesis of conjugatable bisphosphonates for molecular imaging of lar ge animals. Angew Chem Int Ed Engl 2007;46:7969–71. 92. Aikawa E, Nahrendorf M, Sosno vik D, et al. Multimodality molecular imaging identifies proteolytic and osteogenic activities in early aortic valve disease. Circulation 2007;115:377–86. 93. Farokhzad OC, Jon S, Khademhosseini A, et al. Nanopar ticleaptamer bioconjugates: a ne w approach for tar geting prostrate cancer cells. Cancer Res 2004;64:7668–72. 94. Hicke BJ , Stephens AW, Gould T, et al. Tumor tar geting b y an aptamer. J Nucl Med 2006;47:668–78. 95. Herr JK, Smith JE, Medley CD, et al. Aptamer-conjugated nanoparticles for selecti ve collection and detection of cancer cells. Anal Chem 2006;78:2918–24. 96. Bagalkot V, Zhang L, Le vy-Nissenbaum E, et al. Quantum dotaptamer conjugates for synchronous cancer imaging, therap y, and sensing of drug delivery based on bi-fluorescence resonance energy transfer. Nano Lett 2007;7:3065–70.

423

97. Ikanovic M, Rudzinski WE, Br uno JG, et al. Fluorescence assa y based on aptamer-quantum dot binding to Bacillus thuringiensis spores. J Fluoresc 2007;17:193–9. 98. Berger C, Gremlich H-U , Schmidt P, et al. In vi vo monitoring the fate of Cy5.5-tat labeled T l ymphocytes b y quantitati ve near infrared fluorescence imaging during acute brain inflammation in a rat model of experimental autoimmune encephalomyelitis. J Immunol Methods 2007;323:65–77. 99. Swirski FK, Ber ger CR, F igueirdeo J-L, et al. A near-infrared cell tracker reagent for multiscopic in vivo imaging and quantification of leukocyte immune responses. PLoS ONE 2007;2:e1075. 100. Weissleder R, Tung C-H, Mahmood U , Bo gdanov AJ. In vi vo imaging of tumors with protease-acti vated near -infrared fluorescent probes. Nat Biotechnol 1999;17:375–8. 101. Tung C-H, Bredo w S, Mahmood U , Weissleder R. Preparation of a cathepsin D sensiti ve near-infrared fluorescence probe for imaging. Bioconjug Chem 1999;10:892–6. 102. Tung C-H, Mahmood U , Bredow S, Weissleder R. In vi vo imaging of proteolytic enzyme acti vity using a no vel molecular repor ter. Cancer Res 2000;60:4953–8. 103. Bremer C, Tung C-H, Weissleder R. In vivo molecular target assessment of matrix metalloproteinase inhibition. Nat Med 2001;7:743–8. 104. Messerli SM, Prabhakar S, Tang Y, et al. A novel method for imaging apoptosis using a caspase-1 near -infrared fluorescent probe. Neoplasia 2004;6:95–105. 105. Jaffer FA, Kim D-E, Quinti L, et al. Optical visualization of cathepsin K activity in atherosclerosis with a no vel, protease-activatable fluorescence sensor. Circulation 2007;115:2292–8. 106. Galande AK, Hilderbrand SA, Weissleder R, Tung C-H. Enzyme-targeted fluorescent imaging probes on a multiple antigenic peptide core. J Med Chem 2006;49:4715–20. 107. Marme N, Knemeyer J-P, Sauer M, Wolfrum J. Inter- and intramolecular fluorescence quenching of organic dyes by tryptophan. Bioconjug Chem 2003;14:1133–9. 108. Xing B, Khanamiryan A, Rao J. Cell-permeable near-infrared fluorogenic substrates for imaging B-lactamase acti vity. J Am Chem Soc 2005;127:4158–9. 109. Kircher MF, Weissleder R, Josephson L. A dual fluorochrome probe for imaging proteases. Bioconjug Chem 2004;15:242–8. 110. Nolan EM, Burdette SC, Harvey JH, et al. Synthesis and characterizationof zinc sensors based on a monosubstituted fluorescein platform. Inorg Chem 2004;43:2624–35. 111. Kojima H, Nakatsubo N, Kikuchi K, et al. Detection and imaging of nitric o xide with no vel fluorescent indicators: diaminofluoresceins. Anal Chem 1998;70:2446–53. 112. Nolan EM, Lippard SJ. A “turn-on” fluorescent sensor for the selective detection of mercuric ion in aqueous media. J Am Chem Soc 2003;125:14270–1. 113. Ozmen B, Akkaya EU. Infrared fluorescence sensing of submicromolar calcium: pushing the limits of photoinduced electron transfer. Tet Lett 2000;41:9185–8. 114. Kiyose K, K ojima H, Urano Y, Nagano T. Development of a ratiometric fluorescent zinc ion probe in near-infrared region, based on tricarbocyanine chromophore. J Am Chem Soc 2006;128:6548–9. 115. Zhang X-A, Lovejoy KS, Jasanoff A, Lippard SJ. Water-soluble porphyrins as a dual-function molecular imaging platfor m for MRI and fluorescence zinc sensing. Proc Natl Acad Sci U S A 2007;104:10780–5. 116. Sasaki E, Kojima H, Nishimatsu H, et al. Highly sensitive near-infrared fluorescent probes for nitric o xide and their application to isolated organs. J Am Chem Soc 2005;127:3684–5. 117. Hrabie JA, Klose JR, Wink DA, Keefer LK. New nitric oxide-releasing zwitterions derived from polyamines. J Org Chem 1993;58:1472–6. 118. Velapoldi RA, Tonnesen HH. Cor rected emission spectra a nd quantum yields for a series of fluorescent compounds in the visible spectral region. J Fluoresc 2004;14:465–72.

424

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

119. Gillies RJ , Raghunand N , Garcia-Mar tin ML, Gatenb y RA. A review of pH measur ment methods and applications in cancers. IEEE Eng Med Biol Mag 2004;23:57–64. 120. Song Y, Salinas D , Nielson D W, Verkman AS. Hyperacidity of secreted fluid from submucosal glands in earl y c ystic f ibrosis. Am J Physiol Cell Physiol 2006;290:741–9. 121. Ricciardolo FLM, Gaston B , Hunt J. Acid stress in the patholo gy of asthma. J Allergy Clin Immunol 2004;113:610–9. 122. Hamer FM. Bases of which methincyanines are the quaternary salts. J Chem Soc 1940;62:799–808. 123. Hilderbrand SA, Weissleder R. Optimized pH-responsi ve c yanine fluorochromes for detection of acidic en vironments. Chem Commun (Camb) 2007;2747–9. 124. Zhang Z, Achilefu S. Design, synthesis and e valuation of near infrared fluorescent pH indicators in a ph ysiologically rele vant range. Chem Commun (Camb) 2005;5887–9. 125. Strekowski L, Mason CJ , Lee H, et al. Water-soluble pH-sensitive 2,6-bis(substituted eth ylidene)-cyclohexanone/ hydroxy c yanine dy es that absorb in the visib le/near-infrared regions. J Heterocycl Chem 2004;41:227–32. 126. Vieira OV, Bothelo RJ , Grinstein S. Phagosome maturation: aging gracefully. Biochem J 2002;366:689–704. 127. Sarpara GH, Hu SJ , Palmer DA, et al. A new long-wavelength fluorogenic substrate for alkaline phosphatase: synthesis and characterisation. Anal Commun 1999;36:19–20.

128. Xu K, Tang B, Huang H, et al. Strong red fluorescent probes suitable for detecting h ydrogen pero xide generated b y mice peritoneal macrophages. Chem Commun 2005;5974–6. 129. Miller EW, Tulyathan O, Isacoff EY, Chang CJ. Molecular imaging of hydrogen peroxide produced for cell signaling. Nat Chem Biol 2007;3:263–7. 130. Setsukinai K, Urano Y, Kakinuma K, et al. De velopment of no vel fluorescence probes that can reliab ly detec reacti ve o xygen species and distinguish specif ic species. J Biol Chem 2003;278:3170–5. 131. Shepherd J, Hilderbrand SA, Waterman P, et al. A fluorescent probe for the detection of m yleoperoxidase activity in atherosclerosisassociated macrophages. Chem Biol 2007;14:1221–31. 132. Jiang W, Fu Q, Fan H, et al. A highly selective fluorescent probe for thiophenols. Angew Chem Int Ed Engl 2007;46:8445–8. 133. Bouffard J, Kim Y, Swager TM, et al. A highly selective fluorescent probe for thiol bioimaging. Or g Lett 2007. [DOI: 10.1021/ol702539v] 134. So M-K, Xu C, Loening AM, et al. Self-illuminating quantum dot conjugates for in vivo imaging. Nat Biotechnol 2006;24:339–43. 135. Lee D, Khaja S, Velasquez-Castano JC, et al. In vi vo imaging of hydrogen pero xide with chemiluminescent nanopar ticles. Nat Mater 2007;6:765–9.

28 ULTRASOUND CONTRAST AGENTS MARK A. BORDEN, PHD, SHENGPING QIN, PHD, AND KATHERINE W. FERRARA, PHD

Due to its safety , por tability, and lo w cost, ultrasound imaging is widel y used for clinical applications spanning cardiology, peripheral vascular disease, obstetrics, kidney diseases and cancer. Ultrasound molecular imaging is becoming increasingl y popular for small-animal studies, par ticularly those in volving multiple imaging modalities. Ultrasonic b lood echoes are ~tw o orders of magnitude smaller than tissue echoes due to the relatively small acoustic impedance dif ference between red blood cells and plasma. To detect small b lood v essels and receptors within these v essels, ultrasonic contrast agents have been engineered, and each can be characterized as a biocolloid—a colloidal par ticle made from biocompatible materials. Se veral types of biocolloids have been used as ultrasound contrast agents, including gas-liquid emulsions (microbubb les), liquid-liquid emulsions (nanodrops), liposomes, and other par ticles. The de gree of acoustic backscatter depends on the intrinsic properties of the biocolloid. The compressibility of the biocolloid and the density dif ference between the biocolloid and sur rounding tissue contribute to the acoustic backscatter . Nonlinear ef fects and resonance can also contribute considerably to the echo response. One advantage of the biocolloid is its payload capacity, making it amenab le to multimodality imaging or the dual pur poses of imaging and therap y (ie, “point and shoot”). Another is the surface area for targeting. Biocolloids range in size between 10 nm and 10 µm in diameter and therefore can be engineered to present many ligands. Multiple ligand-receptor interactions can lead to f irm adhesion, even in the face of hydrodynamic forces acting to dislodge the biocolloid. This chapter will introduce the biocolloids that are being used as contrast agents for ultrasound molecular imaging, with the str uctures and their typical sizes summarized in F igure 1. Emphasis is placed on the most

popular and theoreticall y superior echo-contrast agent, the microbubb le. Ph ysicochemistry is co vered because an understanding of str ucture-property relationships is the key to engineering and inter preting the perfor mance of tar geted ultrasound contrast agents. Interactions between the microbubble and ultrasound de vice are discussed in the context of methods used to exploit them for the purposes of molecular imaging. Details of ultrasound instrumentation, w hich is used to e xcite and detect microbubbles in vivo, are covered in Chapter 15, “Ultrasound”: Applications to Molecular Imaging. We start by summarizing the proper ties of solid and liquid nanopar ticles, and follo w this discussion with a focus on microbubbles, which are favored for their ability to compress and expand with the passing ultrasound wave and produce distinct acoustic signatures that can be detected with high f idelity.

Gas Pockets

PEG Brush

GAS

Lipid Monolayer

Bubbicle (0.1−10 m)

Microbubble (0.1−10 m)

Perfluorocarbon Liquid

Water

Liposome (10−100 nm)

Nanodrop (10−100 nm) Figure 1. Cartoon representation of ultrasound contrast agents ranging from micron-sized bubbles to liposomes and nanodroplets with diameters as small as tens of nanometers.

425

426

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

SOLID AND LIQUID NANOPARTICLES The commonly used ultrasound contrast agents include a class of submicron par ticles (nanoparticles) that are primarily composed of a solid or liquid. Solid and liquid nanoparticles tend to be less echogenic than gas bubbles (microbubbles) because the y are incompressib le and do not oscillate strongl y with the passing acoustic w ave. However, solid and liquid par ticles are stab le at submicron diameters and therefore can ha ve adv antageous pharmacokinetic properties. Nanoparticles can persist in circulation for hours and can be passi vely tar geted to tumors through the enhanced per meability and retention effect, thus enhancing their reach into the tumor microenvironment.

inclusion of MRI contrast agents into the fluorocarbon liquid nanopar ticles has made these biocolloids useful as multimodal agents (see Chapter 15, “Ultrasound”). 4 Another fascinating imaging application of fluorocarbon nanopar ticles has been the introduction of the “phase-shift colloid ,” w here the liquid droplet v aporizes into a gas microbubb le due to ther mal activation.5 The main benefit of such an agent is the combination of the high stability of the liquid-liquid emulsion with the high echo genicity of the gas-liquid emulsion. Problems, however, include lack of surfactant coverage after area expansion from the droplet to the bubb le, leading to possib le coalescence. When stabilized with a pol ymeric coating, fluorocarbon nanopar ticles can extravasate within tumors, with acti vation onl y at the desired site. 6

Liquid Fluorocarbon Nanoparticles Nanodrops are a type of liquid-liquid emulsion.The emulsion is for med b y mechanical diminution of the fluorocarbon liquid into an aqueous phase. Droplet size can be controlled by sonication and extrusion techniques, as w ell as surf actant f ilm cur vature, to gi ve diameters ranging between 10 and 1000 nm. The liquid fluorocarbon is dispersed in the for m of small par ticles (nanodrops) within the aqueous phase. A surfactant layer on the fluorocarbon surf ace helps stabilize the tin y droplets from aggregating, coalescing, or coarsening. The surf actant layer also ser ves to passi vate the surf ace from immunogenic and thrombo genic effects and pro vides a platfor m on which to attach tar geting ligands and molecular contrast agents for other imaging modalities, such as magnetic resonance imaging (MRI). The fluorocarbon core can serve as a reservoir for hydrophobic drugs. Several dif ferent perfluorocarbon emulsions ha ve been described as echo-contrast agents. The liquid fluorocarbon phase is often highl y hydrophobic, making the particles stable against dissolution. Surface tension at the fluid-fluid interface drives the par ticle to adopt a spherical shape and limits its defor mability. Thus, from a fluid mechanics point of vie w, the liquid droplets act v ery much like solid spheres. In general, liquid fluorocarbon nanopar ticles scatter energy according to typical Ra yleigh scattering theor y.1 For liquid or solid par ticles, the incompressibility of the interior phase prevents significant oscillation in the ultrasound f ield. While individual fluorocarbon nanopar ticles in suspension are poorl y reflecting, their agg regates can be echogenic.2 Nanodrops therefore are suitable for applications involving high-frequency ultrasound and abundant target epitopes (to allo w high accumulation b y ligandreceptor interactions), such as f ibrin in thrombi. 3 The

Liposomes A liposome is a v esicle for med b y a lipid bila yer membrane, enclosing an aqueous core. A car toon of an echogenic liposome is shown in Figure 1. The interior and exterior compartments are separated b y a semipermeable, hydrophobic membrane. Liposomes are fabricated by lipid self assembly and postproduction processing. They can be unilamellar or multilamellar , and their size can be controlled by sonication and e xtrusion to range from ~20 nm to > 10 µm diameter. Typically, large multilamellar vesicles are formed by self assembly during f ilm hydration. Highpower sonication with a cell disr upter at kHz frequencies creates localized shear stresses that break up these lar ger aggregates into smaller , unilamellar v esicles. Further size refinement can be achie ved b y forcing the liposomes through w ell-defined, microscale pores (e xtrusion). The virtually unlimited librar y of available lipids yields a v ast array of physicochemical properties, ranging from per meability to charge density to e xpression of specif ic ligands. The interior space can be loaded with other imaging agents (eg, fluorescent compounds) or with hydrophilic drugs for therapeutic applications. Echogenic liposome for mulations ha ve been described.7,8 The mechanism of echo contrast appears to be the backscatter from entrapped pock ets of air within the liposomes that form during rehydration of the freezedried liposomes. 9 Similar to nanodrops, liposomes could be used as molecular imaging contrast agents in applications in volving high-frequenc y ultrasound (e g, intravascular) and copious target epitopes. The adv antages of liposomes as ultrasound agents include their f avorable phar macokinetics, with a circulation half-life that can be e xtended over se veral da ys (Figure 2A).10 In addition, when the particle diameter is

Ultrasound Contrast Agents

A

B

Figure 2. Positron emission tomography images obtained over 90 min after injection of liposomes or microbubbles. A, 18Ffluorodipalmitin image of liposomes stably circulating. Reproduced with permission from Marik J et al.10 B, 18F-fluorodipalmitin image of liposomes coated with the lipo-PEG-peptide CRPPR adherent to the heart muscle. Reproduced with permission from Zhang H et al.11

small (~100 nm or less), v ascular targeting can be v ery rapid and efficient11 (Figure 2B).

Other Particles Other echo genic par ticles ha ve been found to produce detectable backscatter for ultrasound imaging. These include amorphous solid particles that contain gas pockets in their pores and f issures (so called “bubbicles”). 12 Similar to liposomes, gas pock ets formed during reh ydration may account for the detectab le acoustic backscatter . The formation of these gas pock ets depends strongl y on the size and shape of surf ace features on the par ticle. Such particles can be submicrometer in size, and the gas pockets can be highl y stable due to pinning of the interf acial contact lines within the solid crevices. Silica nanoparticles have also been tested as ultrasound contrast agents. 13

THE ECHOGENIC MICROBUBBLE The microbubb le is an ideal ultrasound contrast agent because it is e xtremely echo genic, as w ell as being biocompatible, multifunctional, and economical. Microbubbles are gas spheres betw een 0.1 and 10 µm in diameter and are much smaller than the w avelength of diagnostic ultrasound, which is typically 100 to 1000 µm. The gas core has a low density and is highly compressible,

427

allowing it to shrink and e xpand with the passage of an acoustic wave. The microbubble increases and decreases in diameter at a rapid v elocity giving rise to a strong and unique echo (Figure 3). It is a fortuitous coincidence that the natural reaction time of a microbubble to a rapid pressure v ariation is on the order of microseconds. Thus, microbubbles resonate at frequencies typicall y used in ultrasound imaging, and resonance can be e xploited to generate a v ery strong echo. Due to this high de gree of backscatter in comparison to plasma and b lood cells, a clinical ultrasound system is capable of detecting the signature from a single microbubble—a volume on the order of a femtoliter. Since microbubbles are too small to be resolv ed by current ultrasound instr uments, microbubb le imaging techniques are designed to dif ferentiate their echoes from those arising from tissue based on changes in the echo spectrum and the response to changes in the amplitude or frequency of the wave. This provides an opportunity to detect a single microbubble adherent to a vascular target within the space of a single v oxel and to distinguish the microbubble from the sur rounding tissue. When a dilute microbubble suspension is driven by a harmonic pulse produced b y a transducer element with a low peak ne gative ultrasound pressure (PNP), microbubble oscillation can be nearly sinusoidal with an amplitude and shape w hich is a linear multiple of the transmitted wave (Figure 4A, B). Increasing the pressure of the driving pulse or introducing a nearb y boundar y alters the oscillatory dynamics (Figure 4C, D). With a higher ultrasonic driving pressure, the rate of collapse is very rapid— faster than the rate of e xpansion—with liquid iner tia acting on the gas bubb le to increase the rate of collapse. The nonlinear dynamics and asymmetrical prof ile result in echoes that contain harmonic multiples of the transmitted pulse. While tissue can also produce echoes that are rich in har monic frequencies, substantial dif ferences in the pressure and frequencies associated with har monic spectra allow microbubble and tissue echoes to be dif ferentiated. The advent of nonlinear imaging techniques has brought microbubbles to the forefront of ultrasound imaging agents. Pulsing schemes that exploit these nonlinear features involve trains of transmitted pulses with the amplitude scaled or the phase of the pulse changed betw een pulses. By summing the retur ned echoes in a strate gic manner, tissue echoes are cancelled and microbubb le echoes sum coherently (Figures 5 and 6A, B); here, the CPS™ strategy of Siemens Medical Solutions is described. By imaging the intensity of the microbubb le echoes, the relati ve vascular density within various regions can be compared.

428

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

C

D

E

F

G

5 mm

H

0

1

2

3

4

5

Figure 3. Two-dimensional optical frame images (A)–(G) and streak image (H) showing oscillation and fragmentation of a lipidshelled microbubble, where fragmentation occurs during compression. Bubble has an initial radius of 1.5 µm, as shown in (A), expanding and contracting in the subsequent images. The streak image in (H), shows the diameter of the bubble as a function of time, and dashed lines indicate the times at which the two-dimensional images were acquired relative to the streak image. Reproduced with permission from Chomas JE et al.81

B

A 1.15 Bubble expansion ratio

Acoustic pressure (kPa)

50

25

0

−25

−50 0

2

4

6 Time (µs)

8

10

1.1 1.05 1 0.95 0.9 0

2

4

10

8

10

6 Bubble expansion ratio

500 Acoustic pressure (kPa)

8

D

C

250

0

−250

−500 0

6 Time (µs)

2

4

6 Time (µs)

8

10

5 4 3 2 1 0 0

2

4

6 Time (µs)

Figure 4. Effect of pulse driving pressure on bubble radial oscillations. A 1.5 radius bubble oscillates under insonation of a 5-cycle pulse with a center frequency of 1 MHz. A, and B, peak negative pressure (PNP) = 50 kPa; C, and D, PNP = 500 kPa, with driving pressure shown in (A), (C) and expansion ratio in (B), (D).

In addition, a high ultrasound pressure will result in the fragmentation of microbubbles (Figure 6D, E). Following their fragmentation and disappearance, the ref ill of ne w agents into the v asculature can be detected and the rate quantified (Figure 6C, F), allo wing real-time images of vascular function (Figure 6B, C).

DOSE AND PHARMACOLOGY OF MICROBUBBLES Little is currently known about the proper dose andpharmacology of targeted microbubbles for use in molecular imaging. Such data has been pub lished for b lood pool

Ultrasound Contrast Agents

A

B

100

100

50

PNP = 100kPa (180°) Pulse (kPa)

Pulse (kPa)

PNP = 50kPa (0°)

0 −50 −100 0

0.5

1 1.5 Time (µs)

2

50 0 −50 −100 0

2.5

0.5

Summation of Echoes (kPa)

bubble tissue

2 0 −2 0.5

1 1.5 Time (µs)

1 1.5 Time (µs)

2

2.5

D Echo (kPa)

Echo (kPa)

C

0

429

2

2.5

bubble tissue

2 0 −2 0

0.5

1 1.5 Time (µs)

2

2.5

E bubble (2c+d) tissue (2c+d)

2 0 −2 0

0.5

1

1.5

2

2.5

Time (µs)

Figure 5. Illustration of a contrast pulse sequence (CPS) imaging strategy, where three scaled pulses are transmitted with scaling factors of 1/2, - 1, 1/2 and the three returned echoes are summed. Shown is a one-cycle transmitted pulse with a center frequency of 2.4 MHz, with PNP of (A) 50 kPa and (B) 100 kPa. The corresponding echoes from a 1-micron bubble and tissue are shown in (C) and (D), where the echoes from the first and third pulses in the sequence would be expected to be similar. E, Summation of 2 times echo in (C) plus echo in (D). The linear echoes from tissue are cancelled while nonlinear echoes from the bubble are acquired.

contrast agents approved for echocardiography, which is described belo w. Ho wever, tar geted microbubb les ha ve inherently dif ferent and unique surf ace chemistries (Figure 7A), o wing to the tar geting ligands, and therefore could e xhibit distinctly different pharmacokinetics. Clearly, immunogenicity of the targeted microbubbles is a key concern. The f irst F ood and Dr ug Administration (FD A)approved microbubble contrast agent for ultrasound imaging was Albunex® (GE Healthcare Systems), which has an air core encased b y an albumin shell. So-called “secondgeneration” microbubb le contrast agents w ere de veloped shortly thereafter to contain a fluorinated gas core, w hich significantly increased the stability in b lood as described below. Optison™ (GE Healthcare Systems) is no w an FDA-approved protein-shelled microbubble contrast agent,

which contains a perfluoropropane (perflutren) gas core. The indicated use is in patients with suboptimal echocardiograms to opacify the left v entricle and to impro ve the delineation of the left ventricular endocardial borders. Dose information for Optison and other commercial agents is given in Table 1. F ollowing injection, most of the gas is eliminated through the lungs in the f irst 10 minutes, with a recovery of 96 ± 23% (mean ± SD) and a pulmonary elimination peak at 30 to 40 seconds after administration and half-life of 1.3 ± 0.69 minutes. The protein shell is believed to be handled through nor mal metabolic routes for human serum albumin, which includes degradation by proteases in the li ver. Def inity® (Lantheus Medical Imaging) w as the first phospholipid-shelled, fluorocarbon-gas f illed agent to receive FD A appro val. Gas elimination routes and shell metabolism likely are similar to Optison.

430

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

D

B

E

C

F

Figure 6. Ultrasound images and acquisition strategy for the estimation of flow rate. A, B-mode image of a Met-1 tumor158 in a mouse model. B, Corresponding image of microbubble density (color) overlaid on the B-mode image, where multipulse (CPS) imaging was used to detect the presence of the microbubbles. C, Corresponding flow rate image, with color indicating the time required for microbubbles to refill a voxel. D–F, Methodology for the acquisition of replenishment images. D, Microbubbles fill a region. E, A destruction pulse removes all circulating agents. F, Microbubbles refill the region and their signal is detected.

A

Composition Xphospholipid Xemulsifier XCharge TTm

H2

e0

s surface tension

Echogenicity visco-elasticity Stability

gas permeability & solubility

Pdomain

s

P/A

Rshell, L

Microstructure

Loading & Release

buckling

hs, Gs

phase separation

x1 x2

shedding

R0

Adomain 2

folding

brush density z

charge density

Immunogenicity & Half-life

B

m layers polycation

lipid monolayer

x1 x2

n layers polyanion

Construction

Figure 7. Microbubble basics: Design and pharmacokinetics. A, Schematic representation showing principles of the rational design of a microbubble for molecular imaging. The composition, architecture, microstructure, and construction can all be engineered to change the physicochemical properties, as discussed in the text. The physicochemical properties, in turn, control the performance, such as echogenicity, stability, and immunogenicity. B, 90-minute maximum intensity projection positron emission tomography (PET) 18F-fluorodipalmitin image of microbubble pharmacokinetics, where the microbubbles circulate for a short interval and then accumulate in the liver and spleen.

Ultrasound Contrast Agents

431

Table 1. REPORTED DOSES FOR COMMERCIALLY AVAILABLE MICROBUBBLE CONTRAST AGENTS Formulation Optison Definity Imagent

Shell/Gas

Concentration (mL- 1)

Mean Diameter (µm)

Recommended Dose (µL/kg)

Albumin/C3F8 Lipid/C3F8 Lipid/C6F14

5.0–8.0 × 108 1.2 × 1010 5.9–13.7 × 108

3.0–4.5 1.1–3.3 6§

6* 10* 6*

Maximum Dose (µL) 10.0† 1.3‡ Single dose only

*

Bolus intravenous injection into peripheral vein.

†

Bolus administrations within 10 min; maximum of 8.7 mL in any one patient study.

‡

Infusion: diluted in 50 mL saline and administered up to 10.0 mL/min.

§

Based on volume-weight, all other mean diameters expressed as number-weight.

Note that commercially available ultrasound contrast agents are polydisperse in size, thereby making ambiguous the classification by average diameter.

Microbubbles exhibit similar mechanical proper ties to erythrocytes as they circulate in the blood. They have a similar size and defor mability and therefore tend to mig rate with red blood cells to the vessel center in parabolic, steady flow.14 Intravital microscop y studies ha ve conf irmed that microbubbles have similar v elocities to red b lood cells in arterioles, venules, and capillaries. 15–17 When they become lodged in a blood vessel, they simply dissolve, deform, and become dislodged without significant vascular effects, as is observed for nor mal leukocyte plugging. 17,18 Safety issues regarding ultrasound contrast agents are co vered at the end of this chapter. Positron emission tomography (PET) imaging of a lipid-shelled agent demonstrated that circulation was short lived and accumulation in the li ver and spleen is significant, with some variation according to the model system. PET imaging of microbubb le biodistribution in a rat model is shown in Figure 7B,19 and that of a mouse model is shown by Willmann and colleagues. 20 Despite the biocompatibility of the materials used to f abricate microb ubbles (e g, proteins, lipids, and biopolymers), microbubb les tend to be decorated with immunological mark ers present in the b lood (ie, they become opsonized) and eliminated from circulation via the reticuloendothelial system (RES). Early studies with protein-coated microbubb les sho wed clearance b y macrophages.21,22 Consistent with the location of macrophages in dif ferent animal models, roughl y 60% of radio-labeled Albunex w as found to accumulate in the liver (Kupffer cells) of rats, w hereas 90% accumulated in the lungs of pigs. 22 Phagocytic accumulation can provide a means of imaging ph ysiologic processes related to immune function in vi vo.23 Opsonization and leukocyte attachment of microb ubbles can also be used to image vascular events, such as ischemia/reperfusion24 and atherosclerosis. 25,26 Lipid-coated microbubbles can be engineered to limit clearance b y phagoc ytosis. F or e xample, SonoV ue® (Bracco Diagnostics Inc.) w as found to evade li ver

clearance,27 whereas Sonazoid™ (GE Healthcare) w as found to be e xclusively phagocytosed by Kupffer cells. 28 As with liposomes, the mechanism of clearance depends on the surf ace chemistry (as well as diameter). F or example, anionic microbubb les exhibit different pharmacology than cationic microbubbles.24,29

PHYSICOCHEMICAL PROPERTIES OF MICROBUBBLES Rational design principles can be applied to the formulation of superior molecular imaging contrast agents. Figure 7A shows a schematic diag ram detailing some of the means b y w hich a microbubb le ma y be engineered. The physicochemical properties can be controlled through concepts in biocolloid engineering design, including composition, microstructure, architecture, and constr uction. In tur n, these ph ysicochemical properties affect the final performance of the microbubble in a molecular imaging study.

Microbubble Fabrication Various methods have been used to fabricate microbubbles for ultrasound imaging. The most popular method has been emulsification by entrainment of a gaseous hood into the aqueous phase b y mechanical agitation of the g as-liquid interface. Methods of diminution include shaking (amalgamation) and sonication. These techniques rely on stochastic e vents that produce a pol ydispersed size distribution, generally ranging betw een submicrometer to tens of micrometers in diameter . Size fractionation techniques can be emplo yed, which are based on buo yancy.30 Newer techniques have been developed to produce monodisperse microbubbles. These microfluidic methods include flo w focusing,31,32 T-junctions,33 jetting,34 and electroh ydrodynamic atomization.35

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Diminution techniques pro vide rapid and costeffective microbubble generation that can be done at the bedside. Microfluidic technolo gies will need to sho w similar robustness and ease of preparation in generating a suf ficient microbubb le dose. Ho wever, the potential gains in control over the microbubble surface chemistry and size could signif icantly enhance quantif ication in molecular imaging studies. Several methods ha ve been described to encapsulate gas in a pol ymer shell. Dispersion and ionic gelation ha ve been used to create alignate-shelled microbubb les.36 Organic solvents have been used to dissolv e and disperse the pol ymer, w hich is then resuspended to for m hollo w polymer capsules. 37–39 A technique involving par tial f ilming of surf actant-coated microbubb les with nanopar ticles was recently described b y Schmidt and Roessling. 40 Polymerization at the air -liquid interf ace during agitation of an acidic medium w as used by Cavalieri and colleagues. 41 Each of these methods has produced microbubb les with enhanced stability. However, chain entanglement and covalent bonds inherent in the pol ymer shells severely dampen the oscillation of the gas core,42,43 thus reducing echogenicity prior to shell r upture.

Microbubble Stability A clean microbubble is inherently unstable owing to surface tension ( σ) of the gas-liquid interf ace (~72 mN/m for an air-water interface). Any gas-liquid or liquid-liquid interface will, b y def inition, e xhibit a surf ace tension owing to disr uption of cohesive intermolecular forces. A force balance over the cur ved surface reveals a net pressure that is g reater on the conca ve side (ie, inside the microbubble). The o verpressure inside the microbubb le (ΔP) was given by Young and Laplace 44:

2σ ΔP = Pb − Pa = , R

(1)

where Pb is the total pressure inside the bubb le, Pa is the ambient pressure, and R is the bubb le radius. F or a microbubble, the o verpressure is on the order of an atmosphere. According to Henr y’s Law, the overpressure increases the local solubility of the gas at the microbubble surf ace, thus creating a chemical potential g radient over which gas diffuses into the sur roundings. Thus, surface tension drives the dissolution of the microbubble. Epstein and Plesset deri ved an ordinar y differential equation for the transport of gaseous species into the surrounding medium.45,46 The model was formulated by taking a mass balance over the microbubble and coupling it

to the dif fusion equation to ar rive at the follo wing expression for the microbubb le radius ( R) as a function of time ( t):

⎛ 2 σ shell 1+ − ⎜ Pa R dR LDw ⎜ − = 4σ shell dt R ⎜ ⎜ 1 + 3 P R ⎝ a

⎞ f ⎟ ⎟ , ⎟ ⎟ ⎠

(2)

where L is Ostwald’s coefficient, Dw is the gas diffusivity in water, Rshell is the resistance of the shell to gas per meation in Eq. (3), σshell is the surf ace tension of the shell, and f is the ratio of the gas concentration in the bulk medium versus that at saturation. This model neglects the time to de velop the concentration boundar y la yer and assumes a perfectly spherical geometry for a microbubble dissolving in an isotropic medium. The Epstein-Plesset equation predicts that a free air microbubble will completely dissolve within a second in saturated water (Figure 8). One approach to increase stability has been to use h ydrophobic gases, such as perfluorocarbons, w hich ha ve w ater per meation resistances (L−1Dw−1) that are se veral orders of magnitude higher than air.47,48 The water permeation resistance of n-C 4F10, for example, is over 100 fold greater than that of air. The molecular w eight, M, and other proper ties of rele vant gases in saline are summarized in Table 2. The molecular weight and solubility of air are appro ximated by that of N2, the principal component of air , and the molecular weight and solubility of N2 are shown in parentheses. The diffusivity v alues of n-C 3F8 and n-C 4F10 are obtained

Microbubble Dissolution Free Air

3

Free PFB Bubble Radius [ m]

432

Shelled PFB 2

f f

1

f

0

1

1

0 10 4 10 310 210

1

100 101 102 103 104 105 106 Time [s]

Figure 8. Radius versus time predicted for microbubbles of different composition dissolving in a static, isotropic medium. Changing the gas content from air [free air] to perfluorobutane for an unshelled microbubble [free PFB] increases the lifetime in a saturated medium, but the gas still dissolves within a minute due to surface tension. Adding a solid phospholipid shell [shelled PFB] significantly enhances the predicted lifetime, even in a completely degassed medium.

Ultrasound Contrast Agents

433

Table 2. GAS PARAMETERS USED IN THE MODELING OF MICROBUBBLE DISSOLUTION TIME Gas Air (N2) n-C3F8 n-C4F10

Molecular Weight, M (g/mol)

Solubility, L × 103 (cm3/cm3)

(28) 188 238

(15) 4.6 0.51

from the Stok es-Einstein approximation, which is based on the assumption that the molecular w eight is proportional to Rm2, w here Rm is the molecular radius. The solubility of n-C3F8 and n-C4F10 in water at 20°C is more than five times larger than that in water at body temperature (37°C). 49 Although the use of n-C 4F10 or n-C 3F8 can increase microbubble lifetime b y an order of magnitude or more (see F igure 8), the surf ace tension ef fect drives complete microbubble dissolution within a minute, which is far too short for a molecular imaging study. Encapsulation is therefore required to stabilize the microbubble. For appropriate stability on the shelf and in vivo, the microbubb le shell must both eliminate surf ace tension and impar t a signif icant per meation resistance. Interestingly, this can be achieved with surfactants, such as lipids below their main phase transition temperature, due to jamming of the molecules into a kinetically trapped configuration that is disordered but has solid-lik e character , such as high viscosity .50,51 Removing the o verpressure eliminates the dri ving force for dissolution in saturated media. This allo ws long-ter m storage of microbubb les, which are stable for months in a sealed vial. In addition to eliminating surf ace tension, the shell ma y contribute a resistance to gas lea ving the core, as modeled b y Borden and Longo52:

dR L − = (1 − f ). R dt + Rshell Dw

(3)

Gas permeation through the shell and diffusion in the surrounding medium are modeled here as resistances in series, analo gous to electrical circuits. The shell resistance is a function of the per meating gas species and the shell composition. In convective flow, where the diffusive boundary la yer becomes thin, the shell resistance becomes the dominant term.

Shell Materials Early ultrasound contrast agents w ere coated with an adsorbed layer of albumin protein.53 The albumin-coated microbubbles Albunex® and Optison ™ (GE Healthcare)

Diffusivity, Dw × 105 (cm2/s) 20 7.7 6.9

were the f irst commerciall y a vailable, FD A-approved contrast agents. More recentl y, protein-shelled microbubbles have been functionalized to carry targeting ligands54 and therapeutic pa yloads.55,56 Albumin shells tend to be rigid and less stab le to ultrasound,57 however, and introduce the typical immuno genicity issues associated with animal-derived materials. Phospholipid shells are most commonl y used for ultrasound molecular imaging. Se veral phospholipidbased ultrasound contrast agents are commercially available worldwide. Lipid-stabilized microbubb les are easy to manuf acture, biocompatib le, and echo genic. Once a gas particle is entrained in a suspension of lipid v esicles and micelles, the h ydrophobic ef fect dri ves adsor ption and orientation of the lipid molecules at the gas liquid interface to minimize surface tension. The lipid shell has a similar str ucture to a Langmuir monola yer at high compression, with lipid head g roups oriented outw ard, except that it is completel y self enclosed in a spherical geometry. A broad librar y of different lipids is available to provide stability and functionality. Most formulations for targeted microbubbles consist of three components: a matrix lipid, an emulsifying lipid , and a tar geted lipid. The matrix lipid stabilizes the shell b y providing cohesion and is often chosen to be below the main phase transition temperature. The emulsifying lipid usuall y contains a polymeric group, such as polyethylene glycol (PEG), that aids in lipid adsor ption and assembly.58 The brush also inhibits coalescence and passi vates the surface. The br ush is often for med by PEG 2000 to 5000 Da; shor ter PEGs (e g, 1000) are unab le to stabilize the microbubble. Up to 20 mol% PEGylated lipid can be incorporated.51 On the tar geting lipid, a pol ymer spacer is necessary to extend the ligand past the br ush.59 Lipid composition can ha ve a dramatic ef fect on microbubble proper ties. Longer chains pro vide more cohesion through enhanced van der Waals and hydrophobic interactions, which increase the shear viscosity 60 and decrease the gas per meability.61 Longer chains also change the mechanism and kinetics of lipid collapse and shedding from vesiculation to fracture and folding.62 This is evidenced by morphological changes in the microbubble during static dissolution 52 and in the destr uction kinetics during acoustic pulsing. 63

434

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Surface Microstructure The pol ycrystalline str ucture of the lipid microbubb le shell was initially shown in pioneering work by Kim and colleagues.60 Microscopy and spectroscop y e vidences indicate that the shell consists of multiple phases, often exhibiting the characteristics of an ordered phase dispersed into a disordered phase. 51,64 The ordered phase tends to be populated with the matrix lipid , whereas the disordered phase tends to be enriched with the emulsifier. Microstructure depends on both composition and processing conditions. Increasing PEG lipid concentration in the shell leads to an increase in the area fraction of the disordered phase.51 Heating and cooling changes the morphology of the ordered phase. 51,60,64 The ordered phase can be melted by heating above the main phase transition temperature of the matrix lipid. Cooling through the transition at different rates leads to differences in domain density, shape, and size. For example, slow annealing can yield v ery large domains on the shell. The domains are not rigid plates; they can bend to accommodate the spherical surf ace. Microstructure af fects man y of the same ph ysicochemical properties as composition. Increasing the defect density produces similar results to using shor ter chain lipids. The gas permeability increases61,65 and the surface shear viscosity decreases. 60

Surface Architecture and Ligand Chemistry Microbubble stability is enhanced by the incorporation of a brush layer of PEGylated lipids, with concentrations of five to nine molar percent typicall y reported. Ligands are incorporated on the distal end of a pol ymer, w here the polymer length can be chosen to extend beyond the brush layer to efficiently bind to their target. Architecture refers to the inter nal str ucture of the br ush la yer on the microbubble shell. Architecture therefore depends on both composition and microstructure (eg, the local composition and dimensions of the ordered and disordered phases). Bimodal br ushes can be used. A longer spacer ar m than the surrounding brush can be used to increase the ligand availability and therefore adhesion strength. 59 Alternatively, a shorter spacer arm can be used to decrease ligand availability, and therefore immunogenicity.66,67 The architecture of the tar geting ligand depends on details of the ligand molecule and the linking chemistr y. Small ligands (order 100 Da) attached to the distal end of the PEG chain will e xhibit dynamics go verned by the thermal motion of the pol ymer spacer .68 Larger ligands (order 1000 Da), however, will significantly change polymer dynamics.69

Small molecule ligands can be directl y attached to lipids, and the ligand-lipid conjugate is then purif ied prior to the incor poration in microbubbles. The methods for lipo-PEG peptide synthesis ha ve been proposed for many years, and the details of such strategies are recently described.11 Peptides can be synthesized manuall y b y standard fluoren ylmethoxycarbonyl (FMOC) chemistr y protocols on solid phase. 70 PEG is coupled onto a peptidyl resin, and Fmoc-Lys(Fmoc)-OH and stearic acid are coupled in sequence. Lipo-PEG peptides are clea ved from the resin and purif ied with HPLC. Large proteins (including antibodies) are often attached through covalent or noncovalent chemistry after the contrast agent is for med due to the harsh conditions that accompan y microbubb le creation. The relati vely large diameter of a microbubble necessitates the use of a large number of ligands (100,000 per microbubb le) to insure adequate co verage. Biotin-strepta vidin approaches, in w hich biotin is attached to the lipid or PEG molecule, ha ve been used most commonl y71,72 due to their simplicity . A common ligand linkage motif is biotin-avidin-biotin. Avidin contains multiple binding pockets and therefore can bind to multiple underl ying biotinylated lipids. The a vidin molecule has a mass of ~60 kDa and is se veral times lar ger than the underl ying PEG. Furthermore, antibody molecules often have a mass of ~120 kDa. Thus, the architecture can be vie wed as a scaffolding str uctured with a thin pol ymer cushion and bulky protein outer layer. To use avidin-biotin linkage, an excess of streptavidin is incubated with the microbubbles, followed by centrifugation to remove excess streptavidin. The biotintylated ligand is then added in a final step. The disadvantage of this approach is the immuno genicity of the resulting surface, rendering this approach to be inappropriate for translational studies. A protein-bearing surface, such as one with a vidin and antibody , will lik ely have exposed nucleophilic groups, such as hydroxyls and amines, that can bind to the unstable thioester bond on the complement protein C3b . Binding of C3b to the surf ace of the microbubb le not onl y changes the ligand binding properties but also marks the microbubb le for clearance by the RES and proceeds with acti vation of the complement system and the inflammator y response. Complement acti vation is of course undesired for molecular imaging, which is primarily used for diagnostic purposes of the unaltered physiology. Thus, it is important to shield the ligand by covering it with a methoxy-terminated PEG overbrush. The ligand can be re vealed b y ultrasound oscillation and radiation force, as discussed later. Covalent attachment strate gies ha ve also been implemented. The advantages and disadvantages of carboxylic acid-amine approaches, in which a carboxylated

Ultrasound Contrast Agents

lipid deri vative is incor porated into the microbubb le shell and reacts with an amine on the ligand , w ere described in detail by Klibanov.71 Potential side products of this reaction include the for mation of N-ac ylisourea, which bind to the microbubb le shell. 71 A more ef ficient strategy for postlabeling with antibodies has been developed using a maleimide-thiol approach, in w hich a maleimide-PEG-lipid is incor porated in the monola yer shell and reacts with a thiol-containing tar geting ligand.71,72 Maleimide reacti ve g roups are more stab le in aqueous buffers than an NHS buf fer, side reactions are reduced, and a g reater percentage of the ligand binds to the target site.

Surface Construction Construction refers to the addition of shell components after formation and stabilization of the initial lipid-coated microbubble. Therefore, constr uction adds another la yer of architectural comple xity. In addition to a vidin-biotin linkage chemistr y, other materials can be constr ucted onto the lipid monola yer shell. F or e xample, clustered polymeric for ms of ligands ha ve been used to enhance adhesion in high-shear flow.73 Another means of construction involves the deposition of oppositely-charged polyelectrolytes as in la yer-by-layer assembly. Such layering can significantly change the properties of the shell. Multilayer shells were shown to enhance the stability of the gas core against dissolution. 74 Further, multilayer shells w ere sho wn to increase the number of plasmid DN A molecules that could be loaded per microbubble.75 Interestingly, multila yers w ere sho wn to slightly dampen the oscillation of the microbubb le (compared to just a lipid-coated agent of similar gas-core diameter) although the damping ef fect disappeared after the first few cycles.75

BIOMEDICAL PERFORMANCE OF MICROBUBBLES Insonified Microbubbles Sophisticated e xperimental systems, w hich incor porate microscopes, custom strobe lights or lasers and high frame-rate cameras, ha ve been de veloped to measure microbubble dynamics during oscillation at me gaHertz frequencies.57,76–80 High spatial and temporal resolution is necessary due to the small size of the microbubb le and the rapid oscillations the y incur during vibration in the ultrasound f ield. Optical obser vation provides infor mation on the dynamic motion of ultrasound microbubb les

435

during insonation that is una vailable with traditional methods of anal yzing the recei ved echoes from contrast agents or modeling of bubb le motion. Diffraction effects associated with the gas-liquid interf ace are used to characterize the dynamics of the gas core; fluorescent probes are inserted within the lipid shell to follo w translation of the shell material and its association with the gas bubb le over time. Very high-speed cameras with a shutter duration of picoseconds to nanoseconds are used to capture two-dimensional images of the microbubb le oscillation; typically, the number of recorded images is limited and the time duration a vailable for recording is less than 1 ms.79,81 In addition, “streak” cameras continuously record the oscillation of a single line across the microbubb le diameter although again over a time window limited to a small fraction of a second. 81 An alternative approach has been to use a v ery bright strobe with a shor t duration, such as that produced b y a high duty c ycle laser, to illuminate the sample w hile recording obser vations with a slower camera. 57,78 This approach has pro ven to be successful in imaging the oscillation of microbubbles within blood v essels and in recording e vents that require lar ge numbers of pulses within a single position.

Gas Dissolution During and After Insonation Dissolution kinetics change when the shell properties are altered during insonation, as sho wn through microscopy. Prior to insonation, dissolution is not apparent on typical optical timescales of observation (seconds). One example of the altered dynamics is sho wn in F igure 9A, w here a single one-c ycle ultrasound pulse with a peak ne gative pressure of 240 kP a and center frequenc y of 2.25 MHz occurred at the time sho wn b y the ar row. The lipidshelled agent decreases in diameter at the time of the pulse (the time inter val from 90–120 ms); ho wever, the diameter remains constant after the completion of the pulse (dashed lines) for the remaining 5 s of obser vation. In this case, the lipid-shelled microbubble contains C4F10, where unencapsulated gas bubbles of C 4F10 are predicted to dissolv e into the sur rounding liquid within 5 s. By comparison, the diameter of the albumin-shelled agents continues to decrease after the end of the ultrasound pulse (solid lines). Fur ther, w hen a train of ultrasound pulses (peak negative pressure of 240 kP a and center frequency of 2.25 MHz) is directed to lipid-shelled microbubb les (Figure 9B), a decrease in diameter of ~0.11 µm is observed with each pulse, with the diameter again remaining constant betw een pulses. Thus, the lipidshelled agents sho w a small, rapid decrease in diameter

436

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

on the order of that predicted b unshelled gas bubble.

A

y equations for an

Theoretical Predictions of Microbubble Oscillation The radial motion of a single ultrasound contrast agent under insonation can be captured by the Rayleigh-Plesset equation that assumes the liquid is incompressible and infinite.82–85 To apply the Rayleigh-Plesset equation, one usually assumes that the gas in the bubb le has unifor m pressure and obeys the polytropic gas law, which requires that the velocity of the bubble wall is small relative to the speed of sound in the surrounding medium.86 Under these assumptions, the equation of motion for the bubb le wall has the form:

B 3

radius (mm)

2.5

3κ ⎡ 3 2 1 ⎛ 2σ ⎞ ⎛ R0 ⎞ ⎢ RR + R = p0 + ⎜ ρ ⎢ ⎝ 2 R ⎟⎠ ⎜⎝ R ⎟⎠ ⎣

2

⎤ 2σ 4η R − − p0 − pi (t ) ⎥ , − R R ⎥ ⎦

1.5

1 0.0

50.0

100.0 150.0 time (s)

200.0

250.0

Figure 9. Microbubble dissolution after insonation. A, Radius versus time obtained from 30 frames per second images of phospholipid-shelled and albumin-shelled agents (arrow indicates time of ultrasound pulse). All bubbles are insonified with a single-cycle pulse of 240 kPa transmission pressure at time = 90 ms, signified by the vertical arrow along the abscissa. Albumin-shelled agents (solid lines) decrease in radius due to static diffusion. Lipid-shelled agents (dashed lines) decrease in diameter with each pulse but remain intact between pulses. Reproduced with permission from Chomas JE et al.100 B, Change in lipid-shelled microbubble diameter over a train of pulses, demonstrating decreased diameter with each pulse (mechanism of acoustically-driven diffusion). Each bubble is insonified by one single-cycle pulse every 15 s. No decrease in radius is observed between pulses. The decrease in radius observed immediately after insonation is due to acoustically-driven diffusion. Reproduced with permission from Chomas JE et al.100

coincident with the ultrasound pulse and then remain unchanged for an extended period that is greater than the time required for static dissolution of the remaining gas into the sur rounding liquid. The magnitude of this step 63 change per c ycle is a function of the coating lipid. Therefore, the concept of “shell rupture” often thought to be associated with insonation of shelled microbubb les is not an accurate description of lipid-shelled microbubb le behavior. As opposed to the stability of lipid-shelled agents after insonation, albumin-shelled agents e xhibit static dissolution after insonation, with a dissolution rate

(4)

where R0 is the bubble radius at equilibrium, R and R represent, respectively, the f irst- and second-order time derivatives of the bubb le radius R, p0 is the h ydrostatic pressure, pi (t ) is the incident ultrasound pressure in the liquid at an inf inite distance from the microbubb le, κ is the polytropic exponent, and ρ, σ, and η are the density, surface tension, and viscosity of the bulk fluid , respectively. The bubble will behave isothermally (ie, κ ≈ 1) if the ther mal dif fusion length in the gas is g reater than the bubb le radius, w hereas it will beha ve adiabaticall y (ie, κ ≈ γ, the specif ic heat ratio of the gas within the bubble) if the thermal diffusion length in the gas is much smaller than the bubb le radius and the bubb le radius is much less than the w avelength of sound in the bubb le.87 In the framework of the linearized theory, the bubble resonance frequenc y can be obtained b y the w ell-known results of Minnaert88:

1 ω0 = R0

1 2

⎡ 3κ ⎛ 2 σ ⎞ 2 σ ⎤ − ⎢ ⎜ p0 + ⎥ . ⎟ R0 ⎠ ρR0 ⎥⎦ ⎢⎣ ρ ⎝

(5)

The pressure of emitted ultrasound at distance r from the bubble center is as follows:

ρR p( r ) = ( 2 R 2 + RR). r

(6)

Ultrasound Contrast Agents

When the incident pressure w ave is increased , the ratio of the v elocity of the bubb le w all to the sound speed in the liquid ( R /c) approaches unity, and sound radiation becomes impor tant. A number of Ra yleighPlesset deri vatives ha ve been proposed , par ticularly modifications that account for sound radiation, including the K eller equation, the Her ring equation and the Gilmore equation. 89–99

A

C

D

E

Observations of Microbubble Oscillation

F

G

Figure 10. Oscillation and destruction of polymer-shelled agent BG1135. Still images (A)–(C) depict the agent before, during, and after exposure to a 2-cycle, 2.25 MHz, 1.4 MPa ultrasound pulse. The streak image (D) shows one line of sight through the agent versus time, with the acquisition times of the still images indicated. The agent is observed to acquire a shell defect and subsequently to eject a new fragment some distance from the original agent. Images (E)–(G) show destruction of another polymer-shelled bubble in response to a 2-cycle, 2.25 MHz, 1.2 MPa pulse. Reproduced with permission from Bloch SH et al.43

6

A

B

BR14

BG1135

5 Expansion ratio

Under limited conditions, oscillation is nearl y symmetrical. The center frequenc y, pressure, and phase of the transmitted pulse alter the oscillation of lipid-shelled microbubbles.100,101 Decreasing the center frequenc y and increasing the peak negative pressure act to increase the maximum e xpansion and the rate of microb ubble collapse. When relative expansion, def ined as the ratio of maximum to initial diameter , e xceeds a threshold (~3), the bubb le is unstab le and is frequentl y obser ved to fragment into smaller daughter bubb les. These smaller gas bubb les often recombine with subsequent ultrasonic c ycles, and c ycles of fragmentation and fusion are obser ved with long transmitted pulses. The oscillation and fragmentation of a lipid-shelled microbubble are shown in Figure 3, combining a set of two-dimensional frame images of e xpansion and contraction with a continuous “streak” recording of a single line through the center of the microb ubble. At a time near 2 µs and betw een images D and E, the microb ubble fragments, with a set of small bubbles observed after this time point. As a basis of comparison, BG1135 (Bracco Research S. A., Gene va, Switzerland) is a pol ymershelled, air -filled microsphere with a rigid , 100 nm thick shell. Expansion and contraction of these microbubbles is not e vident until the shell r uptures (Figure 10). Upon insonation with sufficient ultrasound pressure, the microbubble suddenly ejects a gas bubble. Gas bubb le e xtrusion, ejection, and displacement b y microns are obser ved on a timescale of microseconds100, still a relati vely lo w v elocity compared with the wall motion during oscillation. 81 A summary of the oscillation of lipid and pol ymer-shelled microbubb les is provided in F igure 11. The lipid-shelled microbubbles (see Figure 11A) insonified under the same conditions e xpand with relati ve e xpansion deter mined b y their initial radius, and fragmentation depends on expansion. P olymer-shelled microbubb les (see F igure 11B) insonif ied at a lo w pressure do not e xpand, whereas those insonif ied at a higher pressure e xpand

B

437

4 3 2 1 0

0

5 Initial radius (mm)

10

0

5 Initial radius (mm)

10

Figure 11. Expansion ratio (rmax/r0) vs. initial radius (r0) in response to a 2-cycle, 2.25 MHz pulse for lipid-shelled agent BR14 (A), (square, 180 kPa; diamonds, 360 kPa; triangles, 920 kPa) and polymer-shelled agent BG1135 (B), (circles, 660 kPa; triangles, 1.2 MPa; squares, 1.4 MPa). Closed symbols indicate bubbles intact after insonation; open symbols indicate fragmented bubbles. Reproduced with permission from Bloch SH et al.43

and fragment. We have also repor ted ejection of a gas bubble through a shell defect in Optison agents, w hich have a semirigid albumin shell. 57 We observed that the resulting gas bubble moved away from the shell (traveling several microns in milliseconds), and that the shell collapsed after the gas bubble was ejected.

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

The shell changes the b ubble’s mechanical proper ties including resonance, viscous damping, and scattering properties. de Jong and colleagues 102–105 treated the bubble shell as an elastic solid layer and theoretically studied acoustic attenuation, backscatter , and nonlinear oscillation, w hich w as v alidated b y e xperimental results. Church106 derived a Ra yleigh-Plesset-like equation describing the dynamics of an encapsulated gas bubb le, assuming that the coating material is a la yer of incompressible solid elastic material. Hof f and colleagues 107 developed a model that included viscous and elastic properties of the shell to describe pol ymeric microbubb le behavior. The resonance frequenc y of a pol ymeric gas bubble was described as follows: 1 2

1 ⎡ 3κ ⎛ 2 σ ⎞ 2 σ 12Gε 0 ⎤ ω0 = − + ⎢ ⎜ p0 + ⎥ , ⎟ R0 ⎢⎣ ρ ⎝ R0 ⎠ ρR0 ρR0 ⎥⎦

(7)

where G is the shear modulus of the shell and ε0 is the initial shell thickness at equilibrium. Equation (7) shows that the resonance frequency tends to decrease as bubble size increases and to increase as the modulus of the shell rigidity increases (F igure 12A). The resonant nature of the unshelled microbubble can also be recognized by the narrow peak in the scattering cross section, which is def ined as the po wer scattered per unit v olume. The scattering cross section of a shelled microbubble is predicted to peak at a higher frequenc y than the unshelled microbubb le; ho wever, the peak is substantially broadened (F igure 12B). Church’ s studies106 indicate that for bubb les with a diameter in the range of ultrasound contrast agents ( R0 ≤ 10 µm), the damping ef fects of the bubb le shell are dominated by viscous, compared with ther mal mechanisms, and that the attenuation coef ficient in a bubb ly liquid decreases as either the rigidity or the viscosity of the bubble shell increases. The oscillation of microbubb les in small b lood vessels at target sites is of g reat interest for applications in molecular imaging and ultrasound-enhanced dr ug and gene delivery. Microbubble oscillation in small vessels is substantially dif ferent from that predicted b y the Rayleigh-Plesset equation. Thus, there ha ve been increasing ef forts to model microbubb le oscillation in small vessels.108–115 The models can be summarized into two areas: linear appro ximation and direct numerical simulation based on the Na vier-Stokes equation. Microbubble oscillation in these systems is quite complex.

A

10 9

Resonance frequency [MHz]

The Effect of the Microbubble Shell and Constraining Vessels

8 7 6

Polymer shell No shell

5

12.9 MPa

4

10.6 MPa

3 2 1 0

0

1

2

3

4 5 6 Diameter [mm]

7

8

9

10

B 104

Scattering cross–section [mm2]

438

103 102 101

8 mm

4 mm

8 mm

4 mm

0

10

No shell

Polymer shell

102 1 102 2 5 10

106 Frequency [Hz]

107

Figure 12. Changes in microbubble properties with diameter and materials. A, Calculated resonance frequency as function of particle diameter. The solid lines show the values found for the bubble encapsulated in a polymer shell, whereas the dashed line shows values calculated for free air bubbles. The two curves for the polymeric microbubbles correspond to the shear modulus G = 10.6 MPa, shear viscosity µ = 0.39 Pa s and G = 12.9 MPa, µ = 0.49 Pa s. Reproduced with permission from Hoff L et al.107 B, Scattering cross section as function of frequency. Calculated for polymer-encapsulated air bubbles (solid lines) and for air bubbles without shells (dashed lines) with diameters 4 and 8 µm. The two curves for the polymeric microbubbles correspond to the shear modulus G = 10.6 MPa, shear viscosity µ = 0.39 Pa s and G = 12.9 MPa, µ = 0.49 Pa s. Reproduced with permission from Hoff L et al.107

The bubb le oscillation not onl y depends on the shell material proper ties and the acoustic parameters (e g, pressure, frequenc y, and pulse length) but also on the size and mechanical proper ties of the v essel. The linear oscillation frequency of a bubble decreases within small rigid vessels113,116 and increases within small compliant vessels.112 In small v essels, bubble oscillation is asymmetrical, and e xpansion is reduced for rigid v essels. Within small compliant v essels, bubb le oscillation increases the pressure across the v essel w all and therefore could enhance vascular permeability.111 For 0.5

Ultrasound Contrast Agents

MPa (or lar ger) and 1 MHz ultrasound pulses, bubb le oscillation induces a lar ge circumferential stress within the vessel wall that may exceed the vessel strength. The induced stress within the v essel w all has a stronger dependence on insonation frequenc y than suggested b y mechanical index (MI = P * f−1/2).111 Experiments ha ve sho wn results consistent with those predicted b y theoretical models. Compared with bubble oscillation within inf inite liquids or large vessels, bubble maximum e xpansion within small v essels is decreased while the lifetime is increased.78,117 The threshold of bubble collapse has been found to depend not only on the applied pressure amplitude b ut also on the v essel size, with an increase for smaller silica vessels.114 Ex vivo studies have shown that during insonation, small bubbles tend to fuse into lar ger b ubbles ( ≥ 10 µm), and these fused bubb les can displace the v essel w all up to a fe w microns.118 Vessel wall deflection increases with increasing initial bubb le size and decreasing v essel diameter . Optical obser vation has sho wn that phagoc ytosed microbubbles e xperience viscous damping within the cytoplasm and yet remain acoustically active and capable of lar ge v olumetric oscillations during an acoustic pulse.118 Phagocytosed microbubb les produce an echo with a higher mean frequenc y than free microbubb les in response to a rarefaction-first, single-cycle pulse.118

DELINEATING ADHERENT MICROBUBBLES Microbubbles adherent to a v essel surf ace ha ve been shown to oscillate asymmetricall y and with a lo wer volumetric increase and decrease compared with free microbubbles.119 Still, g reater echo har monic ener gy is produced by adherent microbubb les than sur rounding tissue. Adherent and free microbubb les can be distinguished based on their pulse-to-pulse motion, since adherent microbubbles move at a velocity that is comparable to tissue. Therefore, signal processing methods can be de veloped to distinguish between bound and free microbubb les based on the individual pulse echo and the echoes obtained over a pulse train. 119

EXAMPLES OF MOLECULAR TARGETING Microbubbles are tar geted ef fectively to v ascular receptors accessible to the luminal space, including those associated with inflammation, thrombus, and angio genesis. Proof of concept w as demonstrated b y Klibano v and colleagues,73 where targeted microbubbles incorporating a biotinylated shell component (0.15–7.5 mol%) adhered to

439

avidin-coated petri dishes, remaining adherent to the dish under tangential flo w rates up to 0.6 m/s. Flo wing microbubbles and their cargo can adhere to the surf ace of a vessel with applied ultrasound radiation pressure. 119,120 We briefl y describe e xamples of tar geted microbubb le imaging here; further examples are provided in Chapter 40, “Protein Engineering for Molecular Imaging”.

Inflammation Changes in endothelial receptors occur rapidl y after the onset of an inflammatory stimulus, and these receptors have been effectively targeted with microbubble contrast agents. VCAM-1 and ICAM-1 have been shown to be upregulated in sites of atherosclerotic lesion for mation.121 P-selectin is available on the endothelial surf ace shor tly after ischemic events.122 Leukocytes are recruited to the site of inflammation and also serve as early inflammatory targets. Nonspecific interaction betw een microbubb les and leukocytes was exploited in early studies of targeted imaging of inflammation.123 Subsequent studies have used antibody tar geting of P-selectin, VCAM, and ICAM for studies of induced inflammation, acute cardiac transplant rejection, and atherosclerosis. 124,125 Microbubbles w ere retained in the mouse cremaster muscle and kidne y with induced inflammation compared to controls in nonstimulated nor mal muscle and in P-selectin-def icient mice. 124 Acoustically-reflective liposomes tar geted to ICAM-1 qualitatively showed acoustic image enhancement in vi vo using transv ascular and IVUS. 126 Increasingly, sophisticated ligand systems have been shown to enable capture of microbubbles over a wide range of flow rates,73,127,128 facilitating rapid and f irm capture.

Thrombus The development of contrast agents that enhance the detection of blood clots that are associated with stroke, myocardial inf arction, and deep-v ein thrombosis has been an important goal. In vitro tar geting of thrombi w as performed by Lanza and colleagues 3 using f ibrin-targeted nanoparticles. Using antif ibrinogen-targeted echo genic liposomes, thrombi w ere visib le with epicardial and transthoracic ultrasound.129

Angiogenesis Many receptors ha ve recentl y been sho wn to be upre gulated on angio genic and metastatic endothelial cells. 130 Integrins have been widely evaluated for targeting of imaging agents, drugs, and particles to the tumor endothelium,

440

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

with αvβ3 perhaps recei ving the g reatest attention. 131,132 Targeting of microbubb les to glial tumors’ integrins via echistatin has been demonstrated 133 and to a Matrigel model system in a study b y Stie ger.134 The peptide ar ginine-arginine-leucine w as sho wn to selecti vely tar get microbubbles to angiogenic tumor vasculature by a mechanism that remains unkno wn.135 Antibodies to VEGF-R2 have been attached to microbubbles and used to selectively image malignant tumors in a study b y Willmann and colleagues136 and Lee and colleagues. 137 Dual imaging with ligands directed to both VEGF-R2 and the αvβ3 integrin increased the tar geted agent signal compared with the singly labeled microbubbles.20

SAFETY The complement proteins are an inte gral par t of the innate immune system, w hich is the main line of defense in detecting and removing foreign pathogens in the body. The immunity pathways converge on the complement protein C3 that is con verted to the fragments C3b and C3a by C3 convertase.138 Normally, these fragments are de graded and rec ycled b y complement-control elements to maintain homeostasis. The complement system is acti vated when C3b binds to the surf ace of a foreign particle. Immobilized C3b can be recognized by phagocytic cells, and it can interact with the other complement proteins to stimulate humoral immunity and form the membrane attack comple x. The C3b fragment is a stick y molecule; it contains an unstab le thioester bond that binds to an ar ray of nucleophilic g roups.139 This poses a signif icant challenge for engineering therapeutic devices for injection or transplantation. Surf ace passivation with meth yl-terminated pol y(ethylene glycol) (mPEG) chains has been a major adv ance in biomaterials science. On liposomes, for e xample, the mPEG br ush has been sho wn to signif icantly increase particle circulation half-life. 140 Interestingly, the prolonged lifetimes of stericall y protected par ticles in the circulation ma y not result directly from reduced protein adsor ption. 141–145 Opsonization of stericall y protected par ticles often does occur as exemplified by complement activation by long-circulating liposomes. 146 For e xample, Do xil® (Ortho Biotech) has been shown to produce signif icant complement activation in human ser um in vitro, where incubation of Do xil increased complement protein complex SC5b-9 le vels 100 to 200% o ver control in 7 of 10 different normal human sera.146 This result supports the notion that the ne gative char ge associated with the phosphate g roup in DSPE–mPEG2000 ma y

play a critical role in complement activation. The reason that stealth liposomes remain long circulating w hile they are associated with complement acti vation is not full y kno wn. One h ypothesis is that inaccessib le complement f ixation on PEG-bearing liposomes prevents ligation to complement receptors. 147 For liposomes, methylation of the phosphate o xygen of phospholipid-mPEG conjugate, and hence the remo val of the ne gative char ge, totall y pre vented complement activation.144 For microbubbles, there is also substantial evidence that charged lipids play a role in complement activation. Preferential complement attachment has been sho wn to lipid-shelled microbubbles with a net ne gative charge29 and those exposing biotin or RGD.66,67 Complement C3 has also been sho wn to bind to albumin-encapsulated microbubbles, mediating adherence of the microb ubble to the v ascular endothelium. 26 Adverse reactions to ultrasound contrast agents ha ve been rare in human studies; these have typically been transient and mild. 148 A small number of serious reactions have been reported, including severe hypotension, bradycardia, anaphylactic shock, and f atal outcome in patients under going contrast echocardio graphy. While these obser vations must be vie wed in the light of millions of e xaminations, at this writing the European Medicine Agency (EMEA) and United States FDA have recently taken steps to limit the use of microbubbles or recommend cardiac monitoring in a small subset of patients. 148,149 At the cur rent time, based on ne w repor ts150 these changes are not expected to substantiall y influence the f ield of ultrasound contrast imaging.

Safety of Targeted Agents Decorating microbubbles with targeting ligands—such as proteins, peptides, or metabolites—could fur ther present chemical g roups that bind to C3b and trigger immune activation. The same holds for other colloidal constructs used as imaging contrast agents. Clearl y, avoidance of an immunogenic response is desirable for targeted contrast agents, not just to minimize hypersensitivity reactions b ut also to enab le long enough circulation persistence for accumulation at the tar get. Ideally, the tar geting ligand w ould be hidden from the milieu until the contrast agent reaches the tar get site, where it is e xposed for binding and through multiple ligand-receptor interactions results in adhesion. The unique properties of microbubbles (expansion and contraction in response to an ultrasound f ield) have been shown to facilitate a stealth ligand. 66,67

Ultrasound Contrast Agents

Safety of Insonified Contrast Agents Ultrasound contrast imaging, thus far used clinically only without a tar geting ligand, has been widel y shown to be safe and ef ficacious, with no e vidence of ca vitationrelated biological effects in humans. 151 In vitro, traveling shock waves, fluid shear waves, and liquid jets have been observed, impinging on w alls or the cell monolayer.80,152,153 Such effects are assumed to be e xploitable for microbubb le-induced dr ug and gene deli very in vivo.154–156 These effects are a function of frequency, with a dependence on frequenc y that is g reater than that predicted b y the mechanical inde x and also increase with transmitted pressure.157 Direct observations of microbubble oscillation within small blood vessels and models for constrained microbubb le oscillation ha ve been repor ted only recently, where the e xpansion of microbubb les was reported to be diminished within v essels with a diameter on the order of 15 µm or less. 78,111

CHALLENGES AND LIMITATIONS Microbubbles, w hich are the most echo genic of ultrasound contrast agents, are large and therefore constrained to the intravascular space. This limits the type of receptor molecules that can be tar geted and restricts molecular imaging to endothelial cell surface phenotypes. However, there are se veral impor tant v ascular mark ers that are readily accessible to microbubble targeting, and methods are underw ay to impro ve the echo genicity and signal detection of nanopar ticle contrast agents. Challenges for ultrasound molecular imaging include enhancing tar get accumulation and impro ving the delineation of signals from free and bound microbubb les. More work needs to be done to better understand the in vi vo fate of tar geted microbubbles and the associated bioef fects when insonified b y ultrasound. It should also be mentioned that through microbubb le-induced v essel per meabilization, ultrasound may provide a means of facilitating other molecular imaging modalities b y allowing molecular probes to reach their extravascular target.

CONCLUSIONS AND FUTURE DIRECTIONS Ultrasound molecular imaging is an emerging field that is enabling real-time, in vi vo imaging of se veral v ascular disease processes through longitudinal studies. Research on the fundamental ph ysicochemistry and acoustic response of microbubb les has led to re volutionary advances in the rational design of microbubbles and

441

ultrasound systems that better detect their signature echoes. Commercial systems are cur rently available, and ultrasound no w presents a practical, economical, and effective means of molecular imaging. Yet the f ield currently is restricted to animal models of human disease. Several new microbubble constructs, targeting strategies, comprehensive scanner systems and imaging protocols are currently under de velopment to e xpand this technology to the clinical arena.

REFERENCES 1. Couture O, Bevan PD, Cherin E, et al. In vestigating perfluorohexane particles with high-frequenc y ultrasound. Ultrasound Med Biol 2006;32:73–82. 2. Lanza GM, Wallace KD, Scott MJ, et al. A novel site-targeted ultrasonic contrast agent with broad biomedical application. Circulation 1996;94:3334–40. 3. Lanza GM, Wallace KD, Fischer SE, et al. High-frequency ultrasonic detection of thrombi with a tar geted contrast system. Ultrasound Med Biol 1997;23:863–70. 4. Wickline SA, Lanza GM. Nanotechnolo gy for molecular imaging and targeted therapy. Circulation 2003;107:1092–5. 5. Correas JM, Qua y SD . EchoGen(TM) emulsion: a ne w ultrasound contrast agent based on phase shift colloids. Clin Radiol 1996;51:11–4. 6. Rapoport N, Gao ZG, Kennedy A. Multifunctional nanoparticles for combining ultrasonic tumor imaging and targeted chemotherapy. J Natl Cancer Inst 2007;99:1095–106. 7. Alkan-Onyuksel H, Demos SM, Lanza GM, et al. De velopment of inherently echogenic liposomes as an ultrasonic contrast agent. J Pharm Sci 1996;85:486–90. 8. Huang SL, Hamilton AJ, Nagaraj A, et al. Impro ving ultrasound reflectivity and stability of echo genic liposomal dispersions for use as tar geted ultrasound contrast agents. J Pharm Sci 2001;90:1917–26. 9. Huang SL, Hamilton AJ, Pozharski E, et al. Physical correlates of the ultrasonic reflecti vity of lipid dispersions suitab le as diagnostic contrast agents. Ultrasound Med Biol 2002;28:339–48. 10. Marik J, Tartis MS, Zhang H, et al. Long-circulating liposomes radiolabeled with [18F]fluorodipalmitin ([18F]FDP). Nucl Med Biol 2007;34:165–71. 11. Zhang H, K usunose J, Kheirolomoom A, et al. Dynamic imaging of arginine-rich heart-targeted vehicles in a mouse model. Biomaterials 2008;29:1976–88. 12. Parker KJ, Tuthill TA, Ler ner RM, Violante MR. A par ticulate contrast agent with potential for ultrasound imaging of li ver. Ultrasound Med Biol 1987;13:555–66. 13. Liu J, Levine AL, Mattoon JS, et al. Nanopar ticles as image enhancing agents for ultrasonography. Phys Med Biol 2006;51:2179–89. 14. Goldsmith H, Mason SG. Axial mig ration of par ticles in P oiseuille flow. Nature 1961;190:1095–6. 15. Feinstein SB, Shah PM, Bing RJ , et al. Microbubb le dynamics visualized in the intact capillar y circulation. J Am Coll Cardiol 1984;4:595–600. 16. Keller MW, Se gal SS, Kaul S, Duling B . The behavior of sonicated albumin microbubb les within the microcirculation—a basis for their use during m yocardial contrast echocardio graphy. Circ Res 1989;65:458–67. 17. Lindner JR, Song J , Jayaweera AR, et al. Micro vascular rheology of definity microbubbles after intra-ar terial and intra venous administration. J Am Soc Echocardiogr 2002;15:396–403.

442

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

18. Braide M, Rasmussen H, Albrektsson A, Bagge U . Micro vascular behavior and ef fects of Sonazoid microbubb les in the cremaster muscle of rats after local administration. J Ultrasound Med 2006;25:883–90. 19. Tartis M, K D, Zheng H, et al. Dynamic microPET imaging of ultrasound contrast agents and lipid deli very. J Control Release 2008. [In press] 20. Willmann JK, L A, Paulmurugan R, et al. Dual-targeted contrast agent for ultrasonic assessment of tumor angio genesis in vi vo. Radiology 2008. [In press] 21. Killam AL, Mehlhaff PM, Zavorskas PA, et al. Tissue distribution of I-125-labeled albumin in rats, and whole blood and exhaled elimination kinetics of octafluoropropane in anesthetized canines, following intravenous administration of OPTISON (R) (FS069). Int J Toxicol 1999;18:49–63. 22. Walday P, Tolleshaug H, Gjoen T, et al. Biodistributions of air -filled albumin microspheres in rats and pigs. Biochem J 1994;299:437–43. 23. Lindner JR, Da yton PA, Coggins MP, et al. Nonin vasive imaging of inflammation by ultrasound detection of phagocytosed microbubbles. Circulation 2000;102:531–8. 24. Lindner JR, Coggins MP, Kaul S, et al. Microbubble persistence in the microcirculation during ischemia/reperfusion and inflammation is caused by integrin- and complement-mediated adherence to activated leukocytes. Circulation 2000;101:668–75. 25. Xie F, Tsutsui JM, Anderson D, et al. Detection of albumin microbubble adherence to aor tic endothelium at different stages in the atherosclerotic process. J Am Coll Cardiol 2007;49:A167. 26. Anderson DR, Tsutsui JM, Xie F, et al. The role of complement in the adherence of microbubbles to dysfunctional ar terial endothelium and atherosclerotic plaque. Cardiovasc Res 2007;73:597–606. 27. Yanagisawa K, Moriyasu F, Miyahara T, et al. Phagoc ytosis of ultrasound contrast agent microbubb les by Kupffer cells. Ultrasound Med Biol 2007;33:318–25. 28. Kindberg GM, Tolleshaug H, Roos N , Skotland T. Hepatic clearance of Sonazoid perfluorobutane microbubb les by Kupffer cells does not reduce the ability of li ver to phagocytose or degrade albumin microspheres. Cell Tissue Res 2003;312:49–54. 29. Fisher NG, Christiansen JP, Klibanov A, et al. Influence of microbubble surf ace char ge on capillar y transit and m yocardial contrast enhancement. J Am Coll Cardiol 2002;40:811–9. 30. Kvale S, Jakobsen HA, Asbjornsen OA, Omtveit T. Size fractionation of gas-filled microspheres by flotation. Sep Technol 1996;6:219–26. 31. Ganan-Calvo AM, Gordillo JM. P erfectly monodisperse microb ubbling by capillary flow focusing. Phys Rev Lett 2001;87. 32. Talu E, Lozano MM, P owell RL, et al. Long-ter m stability b y lipid coating monodisperse microbubb les for med b y a flo w-focusing device. Langmuir 2006;22:9487–90. 33. Xu JH, Li SW , Wang YJ, Luo GS. Controllab le gas-liquid phase flow patterns and monodisperse microbubbles in a microfluidic T-junction device. Appl Phys Lett 2006;88. 34. Bohmer MR, Schroeders R, Steenbakk ers JAM, et al. Preparation of monodisperse polymer particles and capsules b y ink-jet printing. Colloids Surf A Physicochem Eng Asp 2006;289:96–104. 35. Farook U, Stride E, Edirisinghe MJ , Moaleji R. Microbubb ling b y co-axial electrohydrodynamic atomization. Med Biol Eng Comput 2007;45:781–9. 36. Wheatley MA, Schrope B, Shen P. Contrast agents for diagnostic ultrasound—development and evaluation of polymer-coated microbubbles. Biomaterials 1990;11:713–7. 37. Schneider M, Arditi M, Bar rau MB , et al. Br1—a ne w ultrasonographic contrast agent based on sulfur he xafluoride-filled microbubbles. Invest Radiol 1995;30:451–7. 38. Bjerknes K, Braenden JU , Braenden JE, et al. Air-filled pol ymeric microcapsules from emulsions containing dif ferent or ganic phases. J Microencapsul 2001;18:159–71. 39. Narayan P, Wheatley MA. Preparation and characterization of hollo w microcapsules for use as ultrasound contrast agents. P olym Eng Sci 1999;39:2242–55.

40. Schmidt W, Roessling G. No vel manuf acturing process of hollo w pol ymer microspheres. Chem Eng Sci 2006; 61:4973–81. 41. Cavalieri F, El Hamassi A, Chiessi E, P aradossi G. Stable polymeric microballoons as multifunctional device for biomedical uses: synthesis and characterization. Langmuir 2005;21:8758–64. 42. Leong-Poi H, Song J, Rim SJ, et al. Influence of microbubble shell properties on ultrasound signal: implications for lo w-power perfusion imaging. J Am Soc Echocardiogr 2002;15:1269–76. 43. Bloch SH, Wan M, Da yton PA, Ferrara KW. Optical obser vation of lipid- and pol ymer-shelled ultrasound microb ubble contrast agents. Appl Phys Lett 2004;84:631–3. 44. Hunter RJ. Foundations of colloid science. Vol I. Oxford: Oxford University Press;1986. 45. Duncan PB , Needham D . Test of the Epstein-Plesset model for gas microparticle dissolution in aqueous media: effect of surface tension and gas undersaturation in solution. Langmuir 2004;20:2567–78. 46. Epstein PS, Plesset MS. On the stability of gas bubb les in liquid-gas solutions. J Chem Phys 1950;18:1505–9. 47. Kabalnov A, Klein D, Pelura T, et al. Dissolution of multicomponent microbubbles in the bloodstream: 1. Theory. Ultrasound Med Biol 1998;24:739–49. 48. Kabalnov A, Bradley J, Flaim S, et al. Dissolution of multicomponent microbubbles in the bloodstream: 2. Experiment. Ultrasound Med Biol 1998;24:751–60. 49. Kabalnov AS, Makarov KN, Shcherbakova OV. Solubility of fluorocarbons in w ater as a k ey parameter deter mining fluorocarbon emulsion stability. J Fluor Chem 1990;50:271–84. 50. Crane JM, Hall SB. Rapid compression transfor ms interfacial monolayers of pulmonary surfactant. Biophys J 2001;80:1863–72. 51. Borden MA, Martinez GV, Ricker J, et al. Lateral phase separation in lipid-coated microbubbles. Langmuir 2006;22:4291–7. 52. Borden MA, Longo ML. Dissolution beha vior of lipid monola yercoated, air-filled microbubbles: effect of lipid h ydrophobic chain length. Langmuir 2002;18:9225–33. 53. Feinstein SB, Cheirif J, Tencate FJ, et al. Safety and efficacy of a new transpulmonary ultrasound contrast agent—initial multicenter clinical-results. J Am Coll Cardiol 1990;16:316–24. 54. Korpanty G, Grayburn PA, Shohet RV, Brekken RA. Targeting vascular endothelium with a vidin microbubbles. Ultrasound Med Biol 2005;31:1279–83. 55. Frenkel PA, Chen SY, Thai T, et al. DN A-loaded albumin microbubbles enhance ultrasound-mediated transfection in vitro. Ultrasound Med Biol 2002;28:817–22. 56. Lentacker I, De Geest BG, Vandenbroucke RE, et al. Ultrasoundresponsive pol ymer-coated microb ubbles that bind and protect DNA. Langmuir 2006;22:7273–8. 57. Dayton PA, Morgan KE, Klibano v AL, et al. Optical and acoustical observations of the effects of ultrasound on contrast agents. IEEE Trans Ultrason Ferroelectr Freq Control 1999;46:220–32. 58. Klibanov AL. Ultrasound contrast agents: de velopment of the field and current status. Top Curr Chem 2002;222:73–106. 59. Kim DH, Klibano v AL, Needham D . The influence of tiered la yers of surface-grafted pol y(ethylene gl ycol) on receptor -ligand-mediated adhesion between phospholipid monola yer-stabilized microbubbles and coated class beads. Langmuir 2000;16:2808–17. 60. Kim DH, Costello MJ, Duncan PB, Needham D. Mechanical properties and microstructure of polycrystalline phospholipid monolayer shells: novel solid microparticles. Langmuir 2003;19:8455–66. 61. Borden MA, Longo ML. Oxygen per meability of fully condensed lipid monolayers. J Phys Chem B 2004;108:6009–16. 62. Pu G, Borden MA, Longo ML. Collapse and shedding transitions in binary lipid monola yers coating microbubb les. Langmuir 2006;22:2993–9. 63. Borden MA, Kr use DE, Cask ey CF, et al. Influence of lipid shell physicochemical proper ties on ultrasound-induced microbubb le destruction. IEEE Trans Ultrason F erroelectr F req Control 2005;52:1992–2002.

Ultrasound Contrast Agents

64. Borden MA, Pu G, Runner GJ , Longo ML. Surf ace phase beha vior and microstr ucture of lipid/PEG-emulsif ier monola yer-coated microbubbles. Colloids Surf B Biointerf aces 2004;35:209–23. 65. Pu G, Longo ML, Borden MA. Ef fect of microstructure on molecular o xygen per meability through condensed phospholipid monolayers. J Am Chem Soc 2005;127:6524–5. 66. Borden MA, Sarantos MR, Stie ger SM, et al. Ultrasound radiation force modulates ligand a vailability on tar geted contrast agents. Mol Imaging 2006;5:139–47. 67. Borden MA, Zhang H, Gillies RJ , et al. A stimulus-responsive contrast agent for ultrasound molecular imaging. Biomaterials 2007. [In press] 68. Jeppesen C, Wong JY, Kuhl TL, et al. Impact of polymer tether length on multiple ligand-receptor bond for mation. Science 2001;293:465–8. 69. Lin JJ, Silas JA, Bermudez H, et al. The effect of polymer chain length and surface density on the adhesiveness of functionalized polymersomes. Langmuir 2004;20:5493–500. 70. Chan WC, White PD. Fmoc solid phase peptide synthesis: a practical approach. New York: Oxford University Press; 2000. 71. Klibanov AL. Ligand-car rying gas-f illed microbubb les: ultrasound contrast agents for tar geted molecular imaging. Bioconjug Chem 2005;16:9–17. 72. Lanza GM, Wickline SA. Targeted ultrasonic contrast agents for molecular imaging and therapy. Prog Cardiovasc Dis 2001;44:13–31. 73. Klibanov AL, Rychak JJ, Yang WC, et al. Targeted ultrasound contrast agent for molecular imaging of inflammation in high-shear flo w. Contrast Media Mol Imaging 2006;1:259–66. 74. Shchukin DG, K ohler K, Mohw ald H, Sukhor ukov GB . Gas-f illed polyelectrolyte capsules. Angew Chem Int Ed Engl 2005;44:3310–4. 75. Borden MA, Caskey CF, Little E, et al. DN A and polylysine adsorption and multila yer constr uction onto cationic lipid-coated microbubbles. Langmuir 2007;23:9401–8. 76. Klibanov AL, F errara KW, Hughes MS, et al. Direct video microscopic obser vation of the dynamic ef fects of medical ultrasound on ultrasound contrast microspheres. In vest Radiol 1998;33:863–70. 77. Dayton PA, Morgan KE, Klibanov ALS, et al. A preliminary evaluation of the ef fects of primar y and secondar y radiation forces on acoustic contrast agents. IEEE Trans Ultrason F erroelectr F req Control 1997;44:1264–77. 78. Caskey CF, Stieger SM, Qin S, et al. Direct obser vations of ultrasound microbubble contrast agent interaction with the micro vessel wall. J Acoust Soc Am 2007;122:1191–200. 79. de Jong N, Frinking PJA, Bouakaz A, et al. Optical imaging of contrast agent microbubbles in an ultrasound f ield with a 100-MHz camera. Ultrasound Med Biol 2000;26:487–92. 80. Prentice P, Cuschieri A, Dholakia K, et al. Membrane disr uption by optically controlled microbubb le ca vitation. Nat Ph ys 2005;1:107–10. 81. Chomas JE, Dayton PA, May D, et al. Optical obser vation of contrast agent destruction. Appl Phys Lett 2000;77:1056–8. 82. Rayleigh L. On the pressure developed in a liquid during the collapse of a spherical cavity. Philos Mag 1917;34:94–8. 83. Vokurka K. On Rayleigh model of a freely oscillating bubble. 1. Basic relations. Czech J Phys 1985;35:28–40. 84. Plesset MS, Prosperetti A. Bubble dynamics and cavitation. Annu Rev Fluid Mech 1977;9:145–85. 86. Putterman S, Ev ans PG, Vazquez G, Weninger K. Ca vitation science—is there a simple theor y of sonoluminescence? Nature 2001;409:782–3. 87. Prosperetti A. Bubble dynamics—a re view and some recent results. Appl Sci Res 1982;38:145–64. 88. Miller DL. Ultrasonic-detection of resonant ca vitation bubb les in a flow tube b y their 2nd-har monic emissions. Ultrasonics 1981;19:217–24.

443

89. Prosperetti A, Lezzi A. Bubble dynamics in a compressib le liquid. 1. 1st-order theory. J Fluid Mech 1986;168:457–78. 90. Prosperetti A. The equation of bubb le dynamics in a compressib le liquid. Phys Fluids 1987;30:3626–8. 91. Lezzi A, Prosperetti A. Bubb le dynamics in a compressib le liquid. 2. 2nd-order theory. J Fluid Mech 1987;185:289–321. 92. Keller JB , K olodner II. Damping of underw ater e xplosion bubb le oscillations. J Appl Phys 1956;27:1152–61. 93. Keller JB , Miksis M. Bubb le oscillations of lar ge-amplitude. J Acoust Soc Am 1980;68:628–33. 94. Prosperetti A, Crum LA, Commander KW. Nonlinear bubble dynamics. J Acoust Soc Am 1988;83:502–14. 95. Trilling L. The collapse and rebound of a gas bubb le. J Appl Phys 1952;23:14–7. 96. Brenner CE. Cavitation and bubble dynamics. Oxford: Oxford University;1995. 97. Gilmore FR. The growth or collapse of a spherical bubble in a viscous compressible liquid. In: Hydrodynamics laborator y. P asadena (CA): California Institute Technology;1952. 98. Barber BP, Hiller RA, Lofstedt R, et al. Def ining the unkno wns of sonoluminescence. Phys Rep 1997;281:65–143. 99. Lofstedt R, Weninger K, Putter man S, Barber BP. Sonoluminescing bubbles and mass dif fusion. Ph ys Re v E 1995;51: 4400–10. 100. Chomas JE, Da yton P, Allen J, et al. Mechanisms of contrast agent destruction. IEEE Trans Ultrason F erroelectr F req Control 2001;48:232–48. 101. Morgan KE, Allen JS, Dayton PA, et al. Experimental and theoretical e valuation of microb ubble beha vior: ef fect of transmitted phase and bubble size. IEEE Trans Ultrason Ferroelectr Freq Control 2000;47:1494–509. 102. Dejong N, Cornet R, Lancee CT. Higher harmonics of vibrating gasfilled microspheres. 1. Simulations. Ultrasonics 1994;32:447–53. 103. Dejong N, Cornet R, Lancee CT. Higher har monics of vibrating gasfilled microspheres. 2. Measurements. Ultrasonics 1994;32:455–9. 104. Dejong N , Hof f L. Ultrasound scattering proper ties of Albunex microspheres. Ultrasonics 1993;31:175–81. 105. Dejong N , Hof f L, Sk otland T, Bom N . Absorption and scatter of encapsulated gas f illed microspheres—theoretical considerations and some measurements. Ultrasonics 1992;30:95–103. 106. Church CC. The effects of an elastic solid-surface layer on the radial pulsations of gas-bubbles. J Acoust Soc Am 1995;97:1510–21. 107. Hoff L , Sontum PC, Ho vem JM. Oscillations of pol ymeric microbubbles: effect of the encapsulating shell. J Acoust Soc Am 2000;107:2272–80. 108. Ory E, Yuan H, Prosperetti A, et al. Growth and collapse of a vapor bubble in a narrow tube. Phys Fluids 2000;12:1268–77. 109. Yuan H, Oguz HN, Prosperetti A. Growth and collapse of a v apor bubble in a small tube. Int J Heat Mass Transf 1999; 42:3643–57. 110. Hu YT, Qin SP, Hu T, et al. Asymmetric oscillation of cavitation bubbles in a micro vessel and its implications upon mechanisms of clinical v essel injur y in shock-w ave lithotripsy. Int J Nonlinear Mech 2005;40:341–50. 111. Qin SP, Ferrara KW. Acoustic response of compliab le microvessels containing ultrasound contrast agents. Ph ys Med Biol 2006;51:5065–88. 112. Qin SP, Ferrara KW. The natural frequency of nonlinear oscillation of ultrasound contrast agents in micro vessels. Ultrasound Med Biol 2007;33:1140–8. 113. Sassaroli E, Hyn ynen K. Resonance frequenc y of microbubb les in small b lood v essels: a numerical study . Ph ys Med Biol 2005;50:5293–305. 114. Sassaroli E, Hyn ynen K. On the impact of v essel size on the threshold of bubble collapse. Appl Phys Lett 2006;89:123901. 115. Ye T, Bull JL. Microbubb le expansion in a fle xible tube. J Biomech Eng 2006;128:554–63.

444

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

116. Oguz HN , Prosperetti A. The natural frequenc y of oscillation of gas bubbles in tubes. J Acoust Soc Am 1998;103:3301–8. 117. Caskey CF, Kruse DE, Dayton PA, et al. Microbubb le oscillation in tubes with diameters of 12, 25, and 195 microns. Appl Phys Lett 2006;88:033902. 118. Dayton PA, Chomas JE, Lum AFH, et al. Optical and acoustical dynamics of microbubble contrast agents inside neutrophils. Biophys J 2001;80:1547–56. 119. Zhao S, Kr use DE, F errara KW, Dayton PA. Selecti ve imaging of adherent tar geted ultrasound contrast agents. Ph ys Med Biol 2007;52:2055–72. 120. Lum AFH, Borden MA, Dayton PA, et al. Ultrasound radiation force enables tar geted deposition of model dr ug car riers loaded on microbubbles. J Control Release 2006;111:128–34. 121. Nakashima Y, Raines EW, Plump AS, et al. Upregulation of VCAM1 and ICAM-1 at atherosclerosis-prone sites on the endothelium in the apoE-def icient mouse. Arterioscler Thromb Vasc Biol 1998;18:842–51. 122. Jian-Guo G, Ming C, Kuo-Chen C. P-selectin cell adhesion molecule in inflammation, thrombosis, cancer g rowth and metastasis. Cur r Med Chem 2004;11:2153–60. 123. Christiansen JP, Leong-Poi H, Klibanov AL, et al. Noninvasive imaging of myocardial reperfusion injury using leukocyte-targeted contrast echocardiography. Circulation 2002;105:1764–7. 124. Lindner JR, Song J , Christiansen J, et al. Ultrasound assessment of inflammation and renal tissue injur y with microb ubbles targeted to P-selectin. Circulation 2001;104:2107–12. 125. Weller GER, Lu E, Csikari MM, et al. Ultrasound imaging of acute cardiac transplant rejection with microb ubbles targeted to intercellular adhesion molecule-1. Circulation 2003;108:218–24. 126. Hamilton AJ, Huang SL, Warnick D, et al. Intra vascular ultrasound molecular Imaging of atheroma components in vi vo. J Am Coll Cardiol 2004;43:453–60. 127. Kaufmann BA, Lewis C, Xie A, et al. Detection of recent myocardial ischaemia by molecular imaging of P-selectin with tar geted contrast echocardiography. Eur Heart J 2007;28:2011–7. 128. Kaufmann B A, Sanders JM, Da vis C, et al. Molecular imaging of inflammation in atherosclerosis with tar geted ultrasound detection of vascular cell adhesion molecule-1. Circulation 2007;116:276–84. 129. Hamilton A, Huang SL, Warnick D, et al. Left v entricular thrombus enhancement after intra venous injection of echo genic immunoliposomes—studies in a ne w e xperimental model. Circulation 2002;105:2772–8. 130. Ruoslahti E. Vascular zip codes in angio genesis and metastasis. Biochem Soc Trans 2004;32:397–402. 131. Varner JA, Cheresh D A. Integrins and cancer . Cur r Opin Cell Biol 1996;8:724–30. 132. Ruoslahti E. The RGD stor y: a personal account. Matrix Biol 2003;22:459–65. 133. Ellegala DB, Poi HL, Carpenter JE, et al. Imaging tumor angio genesis with contrast ultrasound and microbubb les targeted to αvβ3. Circulation 2003;108:336–41. 134. Stieger SM, Dayton PA, Borden MA, et al. Imaging of angio genesis using Cadence™ contrast pulse sequencing and tar geted contrast agents. Contrast Media Mol Imaging 2008;3:9–18. 135. Weller GER, Wong MKK, Modzelewski RA, et al. Ultrasonic imaging of tumor angio genesis using contrast microb ubbles tar geted via the tumor -binding peptide ar ginine-arginine-leucine. Cancer Res 2005;65:533–9. 136. Willmann JK, Paulmurugan R, Chen K, et al. Ultrasonic imaging of tumor angiogenesis with contrast microbubbles targeted to vascular endothelial g rowth f actor receptor 2 in mice. Radiolo gy 2008;246:508–18. 137. Lee DJ, Lyshchik A, Huamani J, et al. Relationship between retention of a vascular endothelial growth factor receptor 2 (VEGFR2)-targeted ultrasonographic contrast agent and the le vel of VEGFR2

138.

139.

140.

141.

142.

143.

144.

145.

146.

147.

148.

149. 150. 151. 152. 153.

154.

155.

156.

157.

158.

expression in an in vi vo breast cancer model. J Ultrasound Med 2008;27:855–66. Sahu A, Lambris JD . Str ucture and biolo gy of complement protein C3, a connecting link betw een innate and acquired immunity . Immunol Rev 2001;180:35–48. Janssen BJC, Huizinga EG, Raaijmak ers HCA, et al. Str uctures of complement component C3 provide insights into the function and evolution of immunity. Nature 2005;437:505–11. Klibanov AL, Mar uyama K, Torchilin VP, Huang L. Amphipathic polyethyleneglycols ef fectively prolong the circulation time of liposomes. FEBS Lett 1990;268:235–7. Moghimi SM, Muir IS, Illum L, et al. Coating par ticles with a block co-polymer (polo xamine-908) suppresses opsonization but permits the acti vity of dysopsonins in the ser um. Biochim Bioph ys Acta 1993;1179:157–65. Price ME, Cornelius RM, Brash JL. Protein adsorption to polyethylene glycol modified liposomes from fibrinogen solution and from plasma. Biochim Biophys Acta 2001;1512:191–205. Moghimi SM, Szebeni J . Stealth liposomes and long circulating nanoparticles: critical issues in phar macokinetics, opsonization and protein-binding properties. Prog Lipid Res 2003;42:463–78. Moghimi SM, Hamad I, Andresen TL, et al. Meth ylation of the phosphate oxygen moiety of phospholipid-methoxy(polyethylene glycol) conjugate prevents PEGylated liposome-mediated complement activation and anaphylatoxin production. FASEB J 2006;20:2591–3. Moghimi SM, Hamad I, Bunger R, et al. Activation of the human complement system b y cholesterol-rich and pe gylated liposomes—modulation of cholesterol-rich liposome-mediated complement acti vation b y ele vated ser um LDL and HDL le vels. J Liposome Res 2006;16:167–74. Szebeni J, Baranyi L, Savay S, et al. Role of complement activation in hypersensitivity reactions to Do xil and HYNIC PEG liposomes: experimental and clinical studies. J Liposome Res 2002;12:165–72. Wilkinson BJ, Peterson PK, Quie PG. Cryptic peptidoglycan and the antiphagocytic ef fect of the Staph ylococcus aureus capsule: model for the antiphagocytic effect of bacterial cell surf ace polymers. Infect Immun 1979;23:502–8. Correas JM, Bridal L, Lesa vre A, et al. Ultrasound contrast agents: properties, principles of action, tolerance, and ar tifacts. Eur Radiol 2001;11:1316–28. Torzilli G. Adverse ef fects associated with SonoV ue use. Exper t Opin Drug Saf 2005;4:399–401. Available at: http://www.radiopharm .com/ News_03132008 .html. Quaia E. Microbubb le ultrasound contrast agents: an update. Eur Radiol 2007;17:1995–2008. Lauterborn W, Ohl CD . Ca vitation b ubble dynamics. Ultrason Sonochem 1997;4:65–75. Wolfrum B, Mettin R, K urz T, Lauterborn W. Observations of pressure-wave-excited contrast agent b ubbles in the vicinity of cells. Appl Phys Lett 2002;81:5060–2. Chen S, Ding J-H, Bek eredjian R, et al. Ef ficient gene deli very to pancreatic islets with ultrasonic microbubble destruction technology. Proc Natl Acad Sci U S A 2006;103:8469–74. Choi JJ, Pernot M, Brown TR, et al. Spatio-temporal analysis of molecular deli very through the b lood-brain bar rier using focused ultrasound. Phys Med Biol 2007;52:5509–30. Hynynen K, McDannold N, Sheikov NA, et al. Local and re versible blood-brain barrier disruption by noninvasive focused ultrasound at frequencies suitab le for trans-skull sonications. Neuroimage 2005;24:12–20. Stieger SM, Caskey CF, Adamson RH, et al. Enhancement of vascular permeability with lo w-frequency contrast-enhanced ultrasound in the chorioallantoic membrane model. Radiology 2007;243:112–21. Guy CT, Cardif f RD, Muller WJ. Induction of mammar y-tumors b y expression of pol yomavirus middle T-oncogene—a transgenic mouse model for metastatic disease. Mol Cell Biol 1992;12:954–61.

29 MULTIMODALITY AGENTS WEIBO CAI, PHD, AND XIAOYUAN CHEN, PHD

In this chapter , w e will f irst briefl y describe the typical imaging agents used for each modality. Among all the (molecular) imaging modalities, no single modality is perfect or sufficient to obtain all the necessar y information needed to answer a par ticular question. Combination of dif ferent modalities can yield syner gistic advantages over using an y single modality alone. Thus, it is critical to de velop multimodality agents that can be simultaneousl y detected by two or more imaging modalities. Herein, we will summarize the current state-of-the-art multimodality agents reported in the literature, most of which are based on certain nanoparticles. Lastly, we will discuss the future directions for the development of multimodality agents.

The f ield of molecular imaging, recentl y def ined by the Society of Nuclear Medicine (SNM) to be “the visualization, characterization, and measurement of biolo gical processes at the molecular and the cellular le vels in humans and other li ving systems,”1 has flourished o ver the last decade. Molecular imaging techniques include positron emission tomo graphy (PET), single-photon emission computed tomo graphy (SPECT), molecular magnetic resonance imaging (mMRI), magnetic resonance spectroscopy, targeted ultrasound, optical bioluminescence, and optical fluorescence. 2,3 Many h ybrid systems that combine tw o or more of these imaging modalities are also commerciall y available, and cer tain other systems are under active development.4–6 A molecular imaging agent, defined as “a probe used to visualize, characterize, and measure biolo gical processes in living systems,”1 is typically composed of an imaging label, a car rier that contains (a) tar geting ligand(s) or is a tar geting ligand, and a linker between the carrier and the label (F igure 1). The label enab les its detection by an imaging instrument, and the targeting ligand pro vides molecular specif icity. In some cases, the sole purpose of the link er is to connect the label and the targeting ligand, whereas in many cases, it can also help

optimizing the phar macokinetics of the imaging agent. Both endogenous molecules and exogenous agents can be molecular imaging agents, and in some cases, not all three components are present. Man y f actors need to be optimized simultaneously to obtain the best imaging outcome. In this chapter , w e will f irst briefl y describe the typical imaging agents used for each modality alone and then summarize the cur rent state-of-the-ar t of multimodality agents.

TYPICAL IMAGING AGENTS FOR EACH MODALITY Molecular imaging can be cate gorized o verall into radionuclide-based imaging (PET and SPECT) and nonradionuclide-based imaging (MRI, ultrasound imaging, and optical imaging). In this section, w e will give a brief overview of the commonly used imaging agents for each modality.

Label

Linker

Carrier

Target

PET SPECT MRI Ultrasound Optical

Length Flexibility Hydrophilicity Charge

Cell Virus Particle Antibody Protein Peptide Small Molecule

Figure 1. A molecular imaging agent is typically composed of an imaging label, a carrier that is/contains a targeting ligand and a linker between the carrier and the label.

445

446

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

PET PET, a nuclear medicine imaging technique that produces a three-dimensional (3-D) image or map of functional processes in the body, was f irst developed in the mid-1970s.7 The most widel y used PET isotopes for imaging applications include 11C (T 1/2: 20 min), 18F (T1/2: 110 min), and 64Cu (T 1/2: 12.8 h). A v ariety of synthons have been developed for 11C-labeling of small molecules, with the most popular ones being 11 C-methyl iodide (F igure 2) and 11C-methyl triflate. 8,9 The meth ylation reaction is usuall y quite simple and straightforward to yield man y biolo gically interesting radiopharmaceuticals. 18 F-labeling of bioacti ve molecules is generall y achieved through three types of functional g roups: carboxylic acid g roup, amino g roup, and sulfh ydryl g roup. 18 F-labeling via the carboxylic acid group is not commonly used. Only two 18F-containing amines that react with carboxylate, 1-[4-( 18F-fluoromethyl)benzoyl]aminobutane-4amine,10 and 4- 18F-fluorophenylhydrazine,11 have been reported. The majority of 18F-labeling reagents react with the amino g roup. N-succinimidyl 4- 18F-fluorobenzoate is probably the most commonl y used acti ve ester for 18Flabeling via amide bond formation (see Figure 2).12–15 Fluoroacylation based on 2- 18F-fluoropropionate (see F igure 2),16 reductive amination using 4- 18F-fluorobenzaldehyde (18F-FBA, see F igure 2), 17 oxime for mation using 18FFBA,18,19 imidation reaction using 3- 18F-fluoro-5-nitrobenzimidate,20 photochemical conjugation using 4-azidophenacyl-18F-fluoride,21 and alk ylation reactions using 4- 18F-fluorophenacyl bromide ha ve also been reported.20 One concer n of peptide/protein labeling using these reagents is the possible interference with biolo gic activity: modif ication of one or more l ysines located at or near the acti ve site of the peptide/protein could reduce its receptor binding af finity. Site-specif ic labeling, pro vided

Figure 2.

Representative

11

the site has been rationall y chosen, has much less disturbance to the biologic activity/binding affinity. Several thiol-reactive 18F-synthons ha ve been described ,22–24 each bearing a maleimide g roup for thiol-specif ic Michael addition. Ho wever, in vi vo PET imaging data was onl y available for another synthon of this type, N-[2-(4-18F-fluorobenzamido)ethyl]maleimide (18F-FBEM, see Figure 2).25 64 Cu has recentl y become increasingl y more popular and significant research has been de voted to the development of ligands that can stably chelate 64Cu.26,27 The most extensively used class of chelators for 64Cu are the macrocyclic pol yaminocarboxylates, such as 1,4,7,10-tetraazacyclododecane-N,Nʹ′,Nʹ′ʹ′,Nʹ′ʹ′ʹ′-tetraacetic acid (DO TA, see Figure 2) 28–37 and 1,4,8,11-tetraazac yclododecane1,4,8,11-tetraacetic acid (TETA, see F igure 2). 38,39 These systems are superior to ac yclic (ie, not c yclic) chelating agents due to the g reater geometric constraint incor porated into the macroc yclic ligand , w hich enhances the kinetic iner tness and ther modynamic stability of their 64 Cu comple xes. Cross-bridged ligands, such as 4,11bis(carboxymethyl)-1,4,8,11-tetraazabicyclo[6.6.2]hexadecane (CB-TE2A, see F igure 2), ha ve also been reported for 64Cu-labeling of peptides. 40 The chelators described above can also be used for another PET isotope of copper, 61Cu (T1/2: 3.4 h). 13 N Less commonl y used PET isotopes include (T1/2: 10 min), 15O (T1/2: 2 min), 68Ga (T1/2: 68 min), 76Br (T1/2: 16.2 h), 86Y (T1/2: 14.7 min), 89Zr (T1/2: 3.3 d), 94mTc (T1/2: 53 min), and 124I (T 1/2: 4.2 d). 13NH3 and H 215O can be used for quantitati ve evaluation of m yocardial and cerebral b lood flo w.41,42 The most commonl y used macro68 Ga are DO TA and cyclic chelating agents for 1,4,7-triazacyclononane-N,Nʹ′,Nʹ′ʹ′-triacetic acid (NO TA, see Figure 2). 43,44 The stability constant (ie, the equilibrium constant for the equilibrium betw een a metal ion surrounded b y w ater molecules and the same ion surrounded b y other ligands in a ligand displacement

C/18F-labeling agents and chelators for radiometal labeling.

Multimodality Agents

reaction) of Ga-NO TA has been deter mined to be TA, 1030.98,45 significantly higher than that of Ga-DO which is 10 21.33.46 86Y and 89Zr, both typicall y incor porated through bifunctional chelators, ha ve not been w ell documented in the literature due to the limited a vailability.47–50 76Br,51 94mTc,52 and 124I 53,54 have also been used for PET imaging applications.

SPECT SPECT imaging detects γ-rays.55,56 A collimator is used to only allow the emitted γ photon to travel along certain directions to reach the detector , which ensures that the position on the detector accurately represents the source of the γ-ray. Because of the use of lead collimators to define the angle of incidence, SPECT imaging has a very low detection efficiency (< 10 −4 times the emitted number of γ-rays).57 The major adv antage of SPECT imaging is that it can be used for simultaneous imaging of multiple radionuclides since the γ-rays emitted from different radioisotopes can be dif ferentiated based on the energy.58 Thus, SPECT can potentially allow simultaneous detection of multiple biolo gic events with multiple isotopes, which is not possible with PET. Common radioisotopes used for SPECT imaging are 99m Tc (T1/2: 6.0 h), 111In (T1/2: 2.8 d), 123I (T1/2: 13.2 h), and 131 I (T1/2: 8.0 d). 99mTc-labeling has been extensively studied and man y excellent reviews have been pub lished.59–62 One of the characteristics of Tc is its rich and diverse redox chemistry, where the oxidation state can v ary between +1 and +7. Moreo ver, the coordination geometries of Tc are also different at dif ferent oxidation states. Man y different Tc cores ha ve been studied for planar γ camera or SPECT imaging, such as the “naked” Tc atom, [Tc = O]3+, N]2+, [Tc(CO) 3]+, and [Tc] [O = Tc = O] +, [Tc 6-hydrazinopyridine-3-carboxylic acid ([Tc]HYNIC) core (Figure 3). For 111In-labeling, the most widely used chelators are DOTA and dieth yltriaminepentaacetic acid (DTPA).63,64 123I is typicall y used for imaging onl y, whereas 131I can be used for both imaging and therapeutic applications.65–67

imaging are the absence of radiation and higher spatial resolution (usuall y sub-millimeter le vel). The major disadvantage of MRI is its inherent low sensitivity, which can be par tially compensated for b y w orking at higher magnetic f ields (4.7 to 14 T in small-animal models), acquiring data for a much longer time period , and using exogenous contrast agents. Many exogenous agents can enhance the MR contrast by selectively shortening either the T1 (longitudinal) or T2 (transverse) relaxation time. The MR image can be weighted to detect differences in either T1 or T2 by adjusting certain parameters during data acquisition. Traditionally, gadolinium (Gd) chelates have been used to increase the T 1 contrast.69 Drawbacks of lo w-molecular w eight MR contrast agents, such as Gd-DO TA or Gd-DTP A (Figure 4A), are their nonspecif icity, rapid renal clearance, and lo w relaxivity. Other MR contrast agents with significantly higher relaxi vities, such as paramagnetic Gd-containing liposomes/micelles, ha ve been repor ted.70 Recently, F eCo/single-graphitic-shell nanopar ticles that are soluble and stable in aqueous solutions have also been prepared (Figure 4A).71 These nanoparticles exhibit ultrahigh saturation magnetization and r 1/r2 relaxivities. Preliminary in vivo experiments have demonstrated longlasting positive-contrast enhancement for vascular MRI in rabbits.71 Iron o xide (IO) nanopar ticles are the most widel y used nanopar ticle-based MR contrast agents (F igure 4A).72 This type of nanopar ticle usuall y consists of a crystalline IO core sur rounded b y a pol ymer coating, such as de xtran or pol y(ethylene gl ycol) (PEG). The presence of thousands of iron atoms in each particle gives

MRI MRI is a nonin vasive diagnostic technique based on the interaction of protons (or other nuclei) with each other and with sur rounding molecules in a tissue of interest.68 Different tissues ha ve different relaxation times that can result in endogenous magnetic resonance (MR) contrast. The major adv antages of MRI o ver radionuclide-based

447

Figure 3.

Common Tc cores used for

99m

Tc-labeling.

448

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A N

O 2

N

O O

HO

O

N

O

Gd31

OH

2O

O

O2 H2O

Gd-DTPA

B

FeCo

Iron Oxide

OH O

HO OH CO 2 OH 2 CO22

O

CH3

CH

CO22 CO22

b -gal

CH2OH 2 O2C

CH3

CH

2

O2C N

N

N

N

N N

N

N

Activatable MR Contrast Agent Figure 4.

Common agents used for enhancing the magnetic resonance contrast.

very high T2 relaxivity.73 Nontargeted IO nanopar ticles have been used for liver,74,75 spleen,76,77 and lymph node imaging.78,79 These nanopar ticles can also nonspecif ically accumulate in the tumor tissue due to the presence of leaky vasculature (the extent of extravasation depends on the porosity of the angio genic tumor v essels) and macrophage uptak e.80–82 In central ner vous system diseases, the role of macrophages in pathologic tissue alterations has led to the use of IO nanopar ticles for imaging of strok e,83 multiple sclerosis, 84 brain tumors, 85 and carotid atherosclerotic plaques. 86 IO nanopar ticles can be detected at quite lo w concentration and singlecell/single-particle detection has been repor ted.87–89 Thus, IO nanopar ticles have recently been used to label cells and track their biodistribution and migration in vivo with MRI.90–94 Activatable MR contrast agents ha ve been repor ted. The first example of this class of agents iscomposed of Gd bound by a macrocylic ligand modified with an appended galactose group, which is attached through a β-galactosidase (β-gal)-cleavable linker.95 The water access of Gd 3+, blocked by the galactose g roup, can be opened b y β-gal, which leads to enhanced MR contrast. Subsequentl y, a methyl group was introduced on the link er that resulted in a 55% relaxi vity increase upon clea vage (F igure 4B). 96 This modified agent has been tested for in vivo imaging of gene expression. A few other strategies to improve the MR contrast ha ve been studied. A pero xidase-based system,

which can o xidize the Gd 3+-containing monomeric construct, thus leading to self-pol ymerization to for m large, paramagnetic pol ymers with increased relaxi vity, has been repor ted.97 An engineered human transfer rin receptor that lacks the iron-regulatory region has been used to shuttle transfer rin-conjugated IO nanopar ticles into transfected cells, w hich resulted in appro ximately a f ivefold increase in signal. 98 In li ving mice, tumors deri ved from such transfected cells showed detectable MR contrast enhancement compared with tumors deri ved from control transfected cells. Recent de velopments that ma y re volutionize the MRI field include hyperpolarization, chemical exchange saturation transfer (CEST) agents, and paramagnetic CEST (PARACEST) agents. At a magnetic field of 1.5 T, the polarization at body temperature is onl y about 5 × 10−6 and 1 × 10−6 for 1H and 13C, respecti vely. To improve the sensiti vity of MRI, h yperpolarization techniques have been de veloped, where a nonther mal equilibrium le vel of polarization is created , and the population dif ference betw een the tw o possib le ener gy states for a spin 1/2 nucleus ma y be altered b y several orders of magnitude. Consequentl y, visualization of the hyperpolarized nuclei can result in images with high signal-to-noise ratio. 99,100 The f irst in vi vo imaging experiment using hyperpolarized agents was reported in the mid-1990s, w here the hyperpolarized noble gas 3He or 129Xe w as inhaled to generate images of the lung

Multimodality Agents

cavity.101,102 Subsequently, clinical MRI in vestigations, which used h yperpolarized gas to study asthma, 103,104 chronic obstr uctive pulmonar y disease, 105,106 cystic fibrosis,107,108 pediatric chronic lung disease,100 and lung transplant109,110 were repor ted. The recent de velopment of hyperpolarization methods for 13C has opened a ne w field of in vivo application. Two different hyperpolariza13 tion approaches ha ve been used for C-containing organic molecules: parah ydrogen-induced h yperpolarization111,112 and dynamic nuclear polarization. 113 Clinical applications of h yperpolarized 13C compounds include v ascular/angiographic imaging, perfusion mapping, inter ventional applications, and metabolic/ molecular imaging.99,114 Low-molecular w eight compounds with slo wly exchanging –NH or –OH protons can be used to alter tissue contrast via CEST of presaturated spins to bulk w ater.115 Lanthanide ions, such as Dy 3+, Tb 3+, Tm 3+, and Yb3+, can offer additional adv antages for introducing MRI contrast via the CEST mechanism, refer red to as P ARACEST agents.116,117 PARACEST agents has been designed to sense molecular e xchange phenomena in tissue, such as pH, 118 temperature,119 lactate,120 glucose,121,122 arginine,123 zinc,124 and caspase-3. 125,126 However, the modest sensiti vity of PARACEST agents, often requiring a minimum concentration of 1 to 10 mM for adequate detection, has limited the applicability of this approach to the detection of high concentration endogenous molecular targets only.

Ultrasound Because of its safety , lo w cost, ease of use, and wide availability, ultrasonography is the most commonl y used clinical imaging modality .127 High-frequency sound waves are emitted from a transducer placed against the skin, and ultrasound images are obtained based on the sound wave reflected back from the inter nal organs. The contrast of ultrasound is dependent on the sound speed , sound attenuation, backscatter, and imaging algorithm.128

449

Ultrasound contrast agents ha ve been used in the clinic for applications such as b lood pool enhancement, characterization of li ver lesions, and perfusion imaging.129,130 These contrast agents are typicall y small, acoustically active particles ranging from several hundred nanometers to a fe w micrometers in diameter. Microbubbles (Figure 5A) resonate in an ultrasound beam, rapidl y contracting and e xpanding in response to the pressure changes of the sound w ave, thus leading to enhanced ultrasound contrast. 131 Targeting is accomplished either through manipulating the chemical proper ties of the microbubble shell or through conjugation of tar getspecific ligands to the microbubble surface.127,132 As these microbubbles are too lar ge to e xtravasate, the disease process must be characterized b y molecular changes in the vascular compar tment to be imaged. Targeted ultrasound studies of integrin αvβ3 (Figure 5B),133–135 vascular endothelial g rowth f actor receptor -2 (Flk-1/KDR), 136,137 and several other vascular proteins, 138 have been reported using ligand-conjugated microbubb les (1 to 5 µm in diameter).

Optical Imaging Optical imaging is a relati vely low-cost method suitab le primarily for small-animal studies. Bioluminescence imaging (BLI) is based on the e xpression of a lightemitting enzyme (eg, firefly luciferase) in target cells and tissues.139 In the presence of its substrate (eg, D-luciferin), an energy-dependent reaction releases photons that can be detected by an imaging system. BLI has been applied for various applications, such as studying gene-e xpression ficiency,141 patterns,140 measuring gene transfer ef monitoring tumor g rowth and response to therap y,142 investigating protein–protein interactions in vi vo,143,144 and deter mining the location and the proliferation of stem cells. 145 Optical coherence tomo graphy (OCT), an imaging technique with high resolution (typicall y 10 to 15 µm), can allow for real time, cross-sectional imaging

Figure 5. Targeted ultrasound imaging using microbubbles. A, The schematic structure of a microbubble. B, Contrast-enhanced ultrasound images of a rat with brain tumor depicting parametric perfusion data (left) and signal enhancement from integrin αvβ3-targeted microbubbles (right). T: tumor; V: ventricles; M: a periventricular metastasis. Adapted with permission from Ellegala DB et al.133

450

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

through biologic tissues.146 OCT detects the reflection of a low-coherence light source directed into a tissue and determines at what depth the reflections occur red. Gold nanoshells147 and nanocages 148 have been repor ted for OCT imaging in vitro. The surf ace-plasmon resonance properties of these nanopar ticles make them promising both as contrast agents for in vivo optical imaging and as therapeutic agents for photother mal treatment of diseases.149 In fluorescence imaging, e xcitation light illuminates the subject and the emission light is collected at a shifted wavelength.150 The main drawback is that these systems are typically not quantitative and the image information is surface-weighted due to tissue absorption.3 In most cases, significant backg round signal is also obser ved because of tissue autofluorescence. Spectral imaging techniques (where fluorescence signals can be separated based on the emission spectra of dif ferent fluorophores 151) and fluorescence-mediated tomo graphy152,153 can f acilitate accurate inter pretation of the fluorescence imaging data. A number of fluorescent dyes have been studied for in vivo imaging, with the most popular ones being the cyanine dyes (Figure 6A).154–158 Fluorescent proteins, such as the g reen fluorescent protein (Figure 6A), have enabled sophisticated studies of protein function and wide-ranging processes from gene expression to intracellular/intercellular signaling cascades, typicall y through a fusion protein repor ter approach rather than through direct labeling. 159

A

The most widel y studied nanopar ticles for optical imaging are quantum dots (QDs, F igure 6A). 160 QDs are inorganic fluorescent semiconductor nanopar ticles with superior optical proper ties than or ganic fluorophores, such as higher quantum yields, higher molar e xtinction coefficients, better resistance to photob leaching and chemical degradation, wider absorption spectra spanning ultraviolet to near-infrared (NIR; 700 to 900 nm) re gion, longer fluorescence lifetime (> 10 ns), narrower emission spectra (typicall y 20 to 30 nm full width at half maximum), and larger effective Stokes shifts.161–163 Numerous in vitro and cell-based applications have been discovered for QDs. 160,164,165 For in vi vo applications, nontar geted QDs have been used for cell traf ficking,166–168 vasculature imaging, 169,170 sentinel l ymph node (SLN) mapping,171–173 and neural imaging. 174,175 A fe w in vi vo targeted imaging studies using QD-based agents ha ve also been reported.176–178 A series of protease-acti vatable optical imaging probes have been developed (Figure 6B). 179 The mechanism of activation is based on dequenching upon clea vage by proteases, such as cathepsin B ,180 cathepsin D,181 cathepsin S, 182 cathepsin K, 183 MMP-2/MMP-9,184 MMP-7,185 caspase-1,186 caspase-3,187 human immunodeficiency vir us protease, 188 urokinase-type plasmino gen activator,189,190 and thrombin. 191 These probes are typically designed to be maximall y quenched in their nati ve state and brightly fluorescent in their clea ved state, thus

SO3-

SO3-

-O S 3 N

+ N

SO3-

O OH

B

Figure 6.

Commonly used fluorescent agents for optical imaging applications.

Multimodality Agents

leading to enhanced contrast. Since proteases pla y k ey roles in cardio vascular, oncolo gic, neurode generative, and inflammator y diseases, 192 such protease-activatable fluorescence imaging probes can ha ve future clinical applications in earl y diagnosis, dr ug disco very, and monitoring of treatment efficacy.

Computed Tomography Computed tomo graphy (CT) is a medical imaging method, w here digital geometr y processing is used to generate a 3-D image of the inter nals of an object from a large series of tw o-dimensional X-ra y images tak en around a single axis of rotation. 193 CT is not a molecular imaging modality y et due to the lack of tar get-specific imaging agents. Cur rent CT contrast agents are typicall y based on iodine- and Gd-based molecules, w hich ha ve mostly nonspecific distribution and rapid pharmacokinetics.194,195 Iodinated nanoparticles have also been reported as CT contrast agents. 196,197 Recently, detection of macrophages in atherosclerotic plaques of rabbits, following intra venous injection of a contrast agent composed of iodinated nanopar ticles dispersed in surf actant, was achieved with a clinical CT scanner .198 This contrast agent ma y become an impor tant adjunct to the clinical evaluation of coronary arteries with CT. To date, all contrast-enhanced CT imaging is based on nonspecif ic targeting and molecular CT has not been achieved. A polymer-coated Bi2S3 nanoparticle (10 to 50 nm per side, 3 to 4 nm thick) w as recently reported as an injectable CT contrast agent. 199 With more than f ive-fold X-ray absor ption than iodine, long circulation time (> 2 h) in vivo, and an efficacy/safety profile comparable with or better than iodinated contrast agents, these nanopar ticles and their bioconjugates ma y potentially be used for CT imaging of molecular tar gets and patholo gic conditions. Because of the ubiquitous nature of CT in the clinical setting, as well as the increasing use and development of microCT and h ybrid systems that combine PET or SPECT with CT, molecular CT will likely become a reality in the near future.

DUAL-MODALITY AGENTS FOR MR AND OPTICAL IMAGING Every imaging modality has its advantages and disadvantages.2 For example, it is dif ficult to accuratel y quantify fluorescence signal in li ving subjects based on fluorescence imaging alone, par ticularly in deep tissues; MRI has high resolution and good soft-tissue contrast, y et it suffers from v ery lo w sensiti vity. Combination of MRI

451

and optical imaging can offer synergistic advantages over either modality alone. Dual-modality agents that can be detected by both MR and fluorescence imaging have been extensively studied. Since dif ferent car riers (e g, liposomes and dendrimers) can be used to attach v arious imaging labels (e g, Gd chelates, fluorescent dy es, or IO/QD nanoparticles that can be the car rier themselves), there are many different combinations. The most straightforward approach is to chemicall y link Gd chelators, typically DTPA, to fluorescent dyes.

Chemically Linking Gd Chelates to Fluorescent Dyes A lipophilic agent composed of Gd-DTP A, a fluorescent dye, and a 16-carbon alk yl chain for intercalati ve labeling of low-density lipoprotein (LDL) par ticles, w as repor ted for in vivo detection of LDL receptors by MRI and in vitro monitoring of cellular localization b y confocal fluorescence microscopy.200 Uptake of labeled LDL par ticles in subcutaneously implanted B16 melanoma tumors in mice led to a modest decrease in T1 relaxation time of the tumor. In another repor t, a chemoselecti ve reaction w as used to synthesize an agent that contains a peptidic tar geting ligand, Gd-DTPA, and a fluorescent dy e, Oregon Green 488 (Figure 7A).201 However, in vitro and in vi vo behaviors of this agent were not studied. Another synthetic bifunctional agent, containing a Gd chelate and fluorescein, has also been synthesized. 202 The use of fluorescein made it onl y suitable for cell-based assa ys but not for in vi vo imaging, due to the shor t excitation/emission wavelength. Cyanine dyes ha ve been co valently attached to Gd-DTP A-polylysine of an e xtended, uncoiled confor mation for fluorescence imaging of preclinical subcutaneous and or thotopic mammary gland tumors. 203 When a wide-f ield illumination camera system was devised, this agent could allow for intraoperative delineation of the tumor margin. Gadophrin-2, composed of a por phyrin ring and tw o covalently linked Gd-DTPA, was used to label and trace intravenously injected human hematopoietic cells in athymic mice. 204 After intravenous injection, the distribution of gadophrin-2-labeled cells in nude mice w as visualized b y MRI, optical imaging, and fluorescence microscopy. In a recent study, the effect of an MR/optical agent, Gd rhodamine dextran205, on neural stem cells w as investigated.206 Although no significant effect on cell viability was observed, a significant decrease in proliferation was evident in cells that underw ent 24 hours of labeling. Further in vi vo testing will be required to ensure that labeling with Gd-based MR contrast agents, such as the two described above, does not perturb the cells’ function.

452

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

Figure 7. Chemically linking Gd-chelates to fluorescent dyes for dual-modality magnetic resonance (MR)/optical imaging. A, A contrast agent that contains a peptidic targeting ligand, Gd-DTPA, and an organic dye. B, A generation 6 (G6) polyamidoamine (PAMAM) dendrimer-based nanoprobe for MR and fluorescence imaging.

In the abo ve-mentioned repor ts, the ratio of Gd:dy e is typicall y 1:1. Due to the v ery lo w sensiti vity of Gd-based MRI, such composition is not optimal for in vivo imaging applications. A class of pol ymer-based MR/optical agents has been repor ted, w hich contains multiple copies of Gd in each molecule. A pol yamidoamine (P AMAM) dendrimer -based nanoprobe for MR/optical imaging has been synthesized (Figure 7B).207 Fluorescence studies re vealed that Gd-comple xation to the probe had no ef fect on the quantum yield of the dy e. However, increase in the dy e content resulted in par tial quenching. The potential of this nanoprobe as an MR/ optical agent was demonstrated in vivo by efficient visualization of SLNs in mice b y both MR and fluorescence imaging. After establishing the optimal dose, this agent was injected into the mammary glands of normal mice to examine the l ymphatic drainage from the breast using a 3T clinical MR scanner .208 Immediately after MRI, optical imaging and image-guided surgery were performed to compare the tw o imaging modalities. It w as found that 750 nmol of this agent w as needed to easily identify and resect the SLNs under image-guided sur gery. Although external optical imaging f ailed to identify SLNs close to the injection site due to shine through of the intensi ve fluorescence signal, MR lymphography was able to identify all SLNs regardless of their location. In the NIR region, the absorbance of all biomolecules reaches a minimum and pro vides a clear windo w for in vivo optical imaging. 209 Moreover, there is also significantly less tissue autofluorescence in this re gion. Poly(L-glutamic acid) conjugated with Gd-DTP A and an NIR dy e, NIR813 (emission maximum: 813 nm), has been used for SLN mapping in normal and tumor-bearing mice.210 It w as repor ted that there are more than 50

Gd-DTPA units per pol ymer chain. After subcutaneous injection, axiliary and branchial lymph nodes were clearly visualized with both T1-weighted MRI and optical imaging within 3 minutes, e ven at the lo west dose tested (2 µmol Gd/kg; 4.8 nmol of NIR813). After intralingual injection in tumor -bearing mice, both MR and NIR fluorescence (NIRF) imaging identif ied most of the superf icial cer vical l ymph nodes. Histopatholo gic e xamination of the SLNs resected under NIRF imaging guidance revealed micrometastases in all SLNs identif ied.

Liposomes Containing Gd Chelates and Fluorescent Dyes Liposomes are spherical v esicles composed of bila yer membranes, typically containing phospholipid and cholesterol.211,212 A fe w repor ts ha ve described the use of liposomes to car ry v arious agents for dual-modality MR/optical imaging. PEGylated paramagnetic and fluorescent immuno-liposomes carrying anti-E-selectin monoclonal antibody as the tar geting ligand ha ve been constructed (F igure 8). 213 Both MRI and fluorescence microscopy revealed the specif ic association of the liposome with stimulated human umbilical v ein endothelial cells, w hich o verexpress the adhesion molecule E-selectin w hen treated with the pro-inflammator y cytokine tumor necrosis f actor-α (TNF-α). Similar liposomes car rying ar ginine-glycine-aspartic acid (RGD) peptides (integrin αvβ3 antagonists214,215) have also been investigated for in vi vo tumor imaging. 216 Both RGDconjugated liposomes and RAD (a control peptide that does not bind to inte grin αvβ3)-conjugated liposomes give enhanced T1-weighted MR contrast. Using e x vivo

Multimodality Agents

Figure 8. Schematic representation of a PEGylated paramagnetic liposome composed of Gd-DTPA, 1,2-distearoyl-sn-glycero3-phosphocholine (DSPC), cholesterol, polyethylene glycol (PEG), and targeting ligands coupled to the distal end of the PEG chains. Adapted with permission from Mulder WJ et al.213

fluorescence microscop y, it w as found that RGDconjugated liposomes are specifically associated with the activated tumor endothelium w hile RAD-conjugated liposomes are mostl y located in the e xtravascular compartment. Recentl y, this type of liposomal nanopar ticle was also conjugated to angine x, a synthetic angiostatic peptide w hich homes to angio genic endothelium, and tested in activated endothelial cells. 217 Traditionally, Gd-based molecules ha ve been the most widely used MR contrast agents. However, the poor sensitivity of Gd-based MRI is a major dra wback, and whether Gd-based dual-modality MR/optical agents can be useful in the clinic remains uncer tain. The large size of liposomes (usually more than 100 nm in dimeter) can accommodate multiple copies of Gd and multiple copies of the targeting ligand, which can not onl y result in signal amplification but also in target specificity. This family of lipidic vehicles can also potentially be used to deliver therapeutic agents, thus for ming multifunctional nanoplatforms that can inte grate both therapeutic components and multimodality imaging.

QD-Based Agents Multimodality imaging using a small-molecule-based agent is very challenging, and sometimes impossible, due to the limited number of attachment points and the potential interference with its receptor binding af finity. On the other hand, nanoparticles have large surface areas, where multiple functional moieties can be attached for multimodality molecular imaging. 218 Nanotechnology, an interdisciplinary research field involving chemistry, engineering, biolo gy, and medicine, has g reat potential for

453

providing a means for early detection, accurate diagnosis, and personalized treatment of diseases. 219,220 Nanoparticles are typicall y smaller than se veral hundred nanometers, comparable with the size of large biologic molecules such as enzymes, receptors, and antibodies. With the size about tw o to four orders of magnitude smaller than human cells, these nanoparticles can offer unprecedented interactions with biomolecules both on the surface of and inside the cells, which may revolutionize disease diagnosis and treatment. Ov er the last decade, there ha ve been numerous nanotechnolo gy centers estab lished w orldwide.221,222 In the United States alone, more than six billion dollars has been invested in nanotechnology research and more than 60 centers, networks, and facilities, funded by v arious agencies, are no w in operation or soon to open.223 After establishing an interdisciplinary nanotechnology workforce, it is expected that nanotechnology will mature into a clinically useful field in the near future. The most w ell-studied nanopar ticles include QDs, 224,225 IO nanoparticles,72 carbon nanotubes, 226,227 and man y others.219,228 The remainder of this chapter will focus mainly on nanoparticle-based dual-modality imaging agents. Many reports have employed QD as the base for constructing dual-modality MR/optical imaging agents. Polymer-coated Fe2O3 cores overcoated with CdSe-ZnS QD shells, fur ther functionalized with antibodies, ha ve been used to magnetically capture breast cancer cells and view them with fluorescence imaging. 229 Magnetic QDs composed of CdS-F ePt with sizes of around 7 nm w ere prepared in a one-pot synthesis (Figure 9A).230 The same synthetic process ma y also allo w the production of large quantities of various types of other heterostructures at the nanoscale level. Another bifunctional nanocomposite system consisting of IO nanopar ticles and CdSe QDs has been reported.231 In this study, CdSe QDs were grown onto preformed IO cores at high temperature (300 ºC) in the presence of or ganic surf actants, yielding either heterodimers or a homo geneous dispersion of QDs around the IO core with both super paramagnetism and tunable optical emission proper ties. After being coated with a thin silica shell, these nanoparticles can be readily used for bioconjugation through the amino groups on the surface. Stabilization of single QD micelles by a silica shell using “amphiphilic” and “hydrophilic” silica precursors has been repor ted.232 The silica-shelled single CdSe/ZnS QD micelles possess desirab le proper ties, such as high quantum yield in aqueous solution, shar p photoluminescence spectra, absence of agg regation, and high transparency. Moreover, the hydrophobic layer between the QD and the silica shell can be manipulated

454

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

C

Figure 9. Quantum dot (QD)-based dual-modality agents for optical/magnetic resonance imaging. A, Magnetic QDs composed of the CdS-FePt heterostructure. B, A QD encapsulated in a paramagnetic micelle with multiple covalently linked annexin A5 molecules (ligand). C, A QD linked to annexin A5 and multiple copies of Gd-diethyltriaminepentaacetic acid (DTPA) complexes through biotin-streptavidin interaction.

for incorporating hydrophobic paramagnetic substances for dual-modality MR/optical imaging. A series of core/shell CdSe/Zn 1-xMnxS nanopar ticles ha ve been synthesized for MR/optical imaging. 233 The Mn 2+ content, in the range of 0.6 to 6.2%, v aries with the thickness of the shell or the amount of Mn 2+ introduced to the reaction. The quantum yield and Mn2+ concentration in these nanopar ticles w as found to be suf ficient for producing contrast for both modalities at a relati vely low concentration. QD encapsulated in a paramagnetic micelle has been constr ucted, and multiple recombinant human annexin A5 protein molecules w ere covalently coupled to the nanopar ticle for tar geting apoptosis, or programmed cell death (F igure 9B). 234 The specif icity of the annexin A5-conjugated nanopar ticles for apoptotic cells w as demonstrated with both fluorescence microscopy and MRI. In another study , the same nanoparticle was functionalized with co valently linked RGD peptides, and the specif icity w as assessed and confirmed on cultured endothelial cells. 235 A biotin ylated construct consisting of a l ysine wedge with eight Gd-DTPA comple xes attached to the peripher y, to increase its perceptibility in MRI b y increasing the load of Gd-DTP A, w as synthesized. 236 The resulting molecule, along with biotin ylated anne xin A5, w as

conjugated to QD-streptavidin (Figure 9C). The resulting construct was then used for analyzing biologic samples and v ascular str uctures with MRI, at the anatomic level, and with tw o-photon laser scanning microscop y, at the cellular le vel. F or all these abo ve-mentioned QD-based MR/optical imaging agents, in vivo-targeted imaging has not been achie ved with either MRI or optical imaging. In clinical settings, optical imaging is rele vant for tissues close to the surf ace of the skin and tissues accessible by endoscopy, as w ell as during intraoperative visualization. 160 The major roadb locks for clinical translation of QD-based agents are inefficient delivery, potential to xicity, and lack of quantif ication. To date, very few successful examples are available for in vivotargeted imaging using QD-based probes. 176,177 With the development of smaller ,172,237 less-toxic238,239 QDs, and further improvement of the conjugation strate gy, it is certainly possible that QD-based probes may achieve optimal tumor-targeting efficacy with acceptable toxicity prof ile for clinical translation in the near future. Although proof-of-principle studies have been reported for MR/optical imaging using QD-based probes in vitro and e x vi vo, w hether adding MRI susceptibility can offer signif icant adv antages o ver QD-based optical imaging alone remains unclear.

Multimodality Agents

Paramagnetic Nanoparticles Containing Fluorescent Moiety Attaching fluorescent dy es or other fluorescent moieties to magnetic nanopar ticles (either Gd-based or IO nanoparticle-based) is another approach to constr uct dual-modality MR/optical agents. A viral capsid has been conjugated with multiple copies of fluorescein and Gdchelates.240 In this study , it w as deter mined that there were 55 fluoresceins and 360 Gd3+ ions per viral particle. Although this work illustrates the potential for engineering natural protein assemb lies for bionanotechnolo gy applications, in vitro and in vi vo targeting have not been achieved with this agent. A fluorochrome-coupled bacterial magnetic nanopar ticle has been e valuated as a dual241 Bacterial magnetic modality imaging agent. nanoparticles with a diameter of 42 nm, har vested from Magnetospirillum gryphiswaldense, were covalently coupled to a fluorescent dy e DY-676 (e xcitation: 676 nm; emission: 701 nm). Murine macrophages, after incubation with this agent for 3 hours, could be imaged with both a 1.5 T MR scanner and NIRF imaging. Luminescent h ybrid nanopar ticles with a paramagnetic Gd 2O3 core encapsulated within a pol ysiloxane shell, w hich car ries or ganic fluorophores and carbo xylated PEG co valently tethered to the inor ganic network, were tested as MR/optical imaging agents. 242 The major advantage of the Gd2O3 core is that it gives enhanced positive contrast in the MR images, rather than the ne gative

A

455

contrast from IO-based agents. Fluorescence imaging was also achieved due to the presence of or ganic dyes in the pol ysiloxane shell. 157Gd possesses a high neutroncapture cross-section (more than 60 times higher than that of 10B); therefore, these Gd2O3-based particles can be potential alter natives to boron compounds for neutroncapture therapy.243 Functionalized super paramagnetic IO (SPIO) nanoparticles, with a PEGylated phospholipid micelle coating conjugated with a fluorescent dy e and the Tat peptide (for cell membrane penetration 244), ha ve been reported (Figure 10A).245 The size of the coated nanoparticles w as betw een 12 and 14 nm in diameter . These micelle-coated SPIO nanopar ticles offer a v ersatile platform for conjugation of a variety of other functional moieties. In another study , IO nanopar ticles with co valently bound bifunctional PEG pol ymer w ere functionalized with Cy5.5 and chlorotoxin for glioma tumor targeting.246 PEGylated and rhodamine-labeled liposomes loaded with IO nanoparticles have been used for magnetic targeting to solid tumors in potential combination with dual-modality MR/optical imaging. 247 In a x enograft prostate tumor model, human prostate adenocarcinoma tumor model, a magnetic f ield g radient w as applied to the tumor b y external apposition of a magnet. Nonin vasive f ibered confocal fluorescence microscop y was used to track the liposomes in vivo within organs and tumor blood vessels. It was also found that the liposomes preserved the vesicle structure and the content during this process.

B

C

Figure 10. Paramagnetic nanoparticles containing fluorescent moiety. A, An iron oxide (IO) nanoparticle, with a PEGylated phospholipid micelle coating, conjugated with a fluorescent dye and the Tat peptide. B, A nanoprobe composed of multiple fluorescent dyes and multiple IO nanoparticles. C, A heterostructured complex composed of an IO nanoparticle and a single-walled carbon nanotube.

456

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A nanoprobe composed of multiple fluorescent dy es and multiple magnetic nanopar ticles has been constructed (Figure 10B). 248 This “core-satellite” str uctured nanoparticle contains a dy e-doped silica “core” and multiple “satellites” of magnetic nanopar ticles. Rhodamine dy e-doped silica (DySiO 2) nanopar ticles with amino g roups on the surface were prepared with a homo geneous size of 30 nm. Water-soluble IO nanopar ticles, with a diameter of 9 nm, were also synthesized and coated with 2,3-dimercaptosuccinic acid. Conjugation of the DySiO 2 nanoparticles with the IO nanopar ticle w as achie ved through a cross-link er, sulfosuccinimidyl-(4-N-maleimidomethyl)cyclohexane-1carboxylate. The use of this nanoprobe as a dual-modality MR/optical agent has been demonstrated with neurob lastoma cells in vitro. A heterostr uctured comple x composed of an IO nanoparticle and an NIR fluorescent single-w alled carbon nanotube (SWNT) has been reported as a dual-modality imaging agent (F igure 10C). 249 Fe catalyst-grown SWNT was individually dispersed in aqueous solution after encapsulation with oligonucleotides. This complex exhibits distinct NIRF, Raman scattering, and visible/NIR absorbance features cor responding to the v arious nanotube species. Atomic force microscop y and cr yo-transmission electron microscopy revealed that the DN A-encapsulated complex was composed of an appro ximately 3 nm IO par ticle attached to a SWNT on one end. Macrophage cells that engulf this DN A-wrapped comple x w ere imaged using both MR and optical imaging, demonstrating that such multifunctional nanostr uctures can potentiall y be useful for multimodality biomedical imaging. For most of the above-mentioned MR/optical imaging agents, in vivo (targeted) imaging was either not tested or not achieved. Although all these agents could be useful for dual-modality MR/optical imaging, usuall y demonstrated in cell culture, most of them are not ideal/suitab le for in vivo applications. Whether in vivo-targeted imaging can be achieved with these agents needs to be pro ven in future studies. A g roup at Har vard Uni versity/Massachusetts General Hospital has reported a series of in vivo-(targeted) imaging studies using cross-link ed IO (CLIO)-Cy5.5 nanoparticles for dual-modality MR/optical imaging. Therefore, although CLIO-Cy5.5 f alls in the cate gory of paramagnetic nanoparticles containing fluorescent moiety, these reports are discussed in a separate section belo w.

CLIO-Cy5.5 Amino-CLIO w as initiall y conjugated to Cy5.5 either through a disulf ide or a thioether linkage. 250 After subcutaneous injection of the probe, axillar y and brachial

lymph nodes could be detected by both MR and optical imaging. In this study, in vivo targeting was not clearly demonstrated. Subsequently, the uptake of CLIO-Cy5.5 by macrophages in inf arcted myocardium was studied, and an increase in the MR contrast w as seen in the anterolateral walls of the inf arcted mice but not in the sham-operated mice. 251 The uptake of CLIO-Cy5.5 b y macrophages infiltrating the infarcted myocardium was confirmed b y e x vi vo fluorescence microscop y and immunohistochemistry. A similar agent has also been tested in the apolipoprotein-E-def icient (apoE −/−) mouse model for imaging cellular inflammation in vivo during atherosclerosis. 252 In vi vo MRI demonstrated strong contrast enhancement of the plaque by the agent, which was confirmed by ex vivo MRI and fluorescence reflectance imaging. CLIO nanopar ticles ha ve also been loaded with Cy5.5 either via an enzyme-clea vable linker253 or with additional tar geting ligands, such as anti-v ascular cell adhesion molecule (VCAM-1) antibodies 254 or E-selectin-binding peptides. 255 In both studies, in vi vo optical imaging w as car ried out y et nonin vasive MRI was not achie ved, lik ely due to the lo w sensiti vity of MRI. In other repor ts, CLIO-Cy5.5 par ticles w ere linked with VCAM-1 binding peptides, 256,257 bombesin peptides,258 or anne xin V. 259 In these cases, in vi vo MRI was accomplished but in vivo optical imaging was not, likely due to the limited tissue penetration of light. The same CLIO-Cy5.5 par ticles ha ve been used to map endothelial cells, macrophages, proteol ysis, and osteogenesis in aor tic v alves of h ypercholesterolemic apoE-deficient mice. 260 Using a panel of dif ferently functionalized agents described abo ve, this study showed that molecular imaging can nonin vasively detect key cellular events in early aortic valve diseases, including endothelial cell and macrophage acti vation, proteolytic activity, and osteogenesis.

Dual-Modality MR/Optical Imaging in Vivo The above-mentioned studies demonstrated the potential of dual-modality MR/optical imaging in vi vo, but the tw o modalities w ere not equall y ef fective. The less sensiti ve modality was usually used for e x vivo validation of the in vivo results obtained from the more sensiti ve imaging modality. Although only single-modality noninvasive imaging was achieved with these dual-modality agents, w hich does not tak e full adv antage of the nanopar ticle-based approach, the capability of detecting the agent with another imaging modality did pro vide a con venient w ay for ex vivo validation that is signif icantly more reliab le and

Multimodality Agents

advantageous than the single-modality probes. It is not until recently that nonin vasive dual-modality MR/optical imaging was accomplished using a nanoparticle-based approach. Gene silencing using short-interfering ribose nucleic acid (siRNA) has become an attractive approach to probe gene function in mammalian cells. 261–263 A multifunctional probe for in vivo transfer of siRNA and simultaneous imaging of its accumulation in tumors b y both MR and optical imaging has been repor ted (F igure 11A). 264 This probe consists of magnetic nanopar ticles, labeled with Cy5.5, co valently link ed to siRN A molecules specific for either model or therapeutic targets. Additionally, the nanoparticle was modified with a membrane translocation peptide for intracellular deli very.244,265 Tumor accumulation of the multifunctional probe in mouse models was demonstrated by both MR and optical imaging (Figure 11B). This study represents the f irst example of combining nonin vasive multimodality imaging and molecular therapy using a nanoparticle-based approach. Molecular MRI is in its inf ancy.266 Dual-modality mMRI/optical imaging is rare. In man y studies, e x vivo histology w as not car ried out to in vestigate/validate whether the targeted MR/optical agents are tar geting the tumor vasculature, targeting the tumor cells, or accumulating nonspecif ically in the interstitial space. It is lik ely that the feasib le targets reachable by targeted nanoparticles will be mostly vasculature related, as the overall size of these nanopar ticles with surf ace polymer coating and targeting ligands are quite lar ge (> 20 nm in diameter). Newly developed nanoparticles with smaller size (preferably < 10 nm in diameter) and long circulation half-life (at least a fe w hours) ma y allow for e xtravasation from the leaky tumor v asculature to a cer tain extent. Surf ace coating of these nanopar ticles with small molecules or

A

457

peptides should gi ve better tar geting ef ficacy than antibody-conjugated nanopar ticles because of the much larger number of targeting ligands per particle and significantly smaller o verall size (1 to 2 kDa vs 150 kDa in molecular weight). Since the major disadvantage of MRI is its inherent low sensitivity, future development of novel contrast agents with the capability of targeting the cells in addition to the vasculature may dramatically increase the MR/optical signal and f acilitate the biomedical applications of dual-modality MR/optical imaging agents.

DUAL-MODALITY AGENTS FOR PET AND OPTICAL IMAGING Radionuclide-based imaging (PET and SPECT) has much higher sensitivity than MRI and much better tissue penetration than optical imaging and ultrasound. 2 The most important advantage of radionuclide-based imaging over other imaging modalities is the ability to quantitatively measure, at any depth, the radionuclide concentration in v arious or gans over time, 267 which can pro vide invaluable infor mation about the phar macokinetics and the full body distribution of the imaging agents. An 124Ilabeled photosensitizer , meth yl 3-(1 ʹ′-m-iodobenzyloxyethyl)-3-devinylpyropheophorbide-a, was reported as a dual-modality agent with tumor fluorescence and PET imaging capability.268 This “see and treat” strategy can be potentially improved using more tumor -avid and/or target-specific photosensitizers. Due to the dif ficulty in quantifying the fluorescence signal in vi vo and man y other technical challenges that remain to be solv ed, in vi vo imaging of QDs are so f ar mostly qualitative or semiquantitative. Development of a

B

Figure 11. A multifunctional probe for in vivo dual-modality imaging and therapy. A, Schematic illustration of the multifunctional probe consisting of a magnetic nanoparticle labeled with Cy5.5, membrane translocation peptides (MPAP), and short-interfering ribose nucleic acid (siRNA) molecules targeting green fluorescent protein (siGFP). B, In vivo magnetic resonance imaging (MRI) of mice bearing subcutaneous tumors (arrows) before and after treatment. High-intensity optical signal in the tumor confirmed the delivery of the nanoparticle. Adapted with permission from Medarova Z et al.264

458

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

dual-modality agent containing both an NIR QD and a PET isotope will allow for sensitive, accurate assessment of the phar macokinetics and tumor -targeting efficacy of NIR QDs b y PET , w hich ma y g reatly f acilitate future translation of QDs into clinical applications. To mak e QDs more useful for in vi vo imaging and other biomedical applications, QDs need to be effectively, specifically, and reliab ly directed to a specif ic or gan or disease site without alteration. We recentl y repor ted a QD-based probe for both NIRF and PET imaging (F igure 12A and B).30 In this study, cyclic RGD peptides and DOTA chelators were conjugated to a QD (maximum emission 705 nm) for targeted dual-modality PET/NIRF imaging after 64 Cu-labeling. Using this dual-modality PET/NIRF imaging agent, w e quantitati vely e valuated the tumor targeting ef ficacy and found that the majority of the A

dual-modality agent in the tumor w as within the tumor vasculature. This PET/NIRF probe can confer suf ficient tumor contrast detectable by PET at much lower concentration than that required for in vi vo NIRF imaging, 177 thus signif icantly reducing the potential to xicity of cadmium-based QDs that ma y g reatly f acilitate their future biomedical applications. 269,270 Histologic e xamination revealed that DO TA-QD-RGD tar gets primaril y the tumor vasculature via RGD-integrin αvβ3 interaction with little e xtravasation (F igure 12C). Combining PET and optical imaging also o vercomes the tissue penetration limitation of NIRF imaging, allo wing for quantitati ve in vivo targeted imaging in deep tissue, w hich will be cr ucial for future image-guided sur gery through sensiti ve, specific, and real-time intraoperative visualization of the molecular features of normal and diseased processes.

B

5 %ID/g

6h

1h

Optical

5h

0 %ID/g

PET

C

CD31

QD

Merge

Mouse β3

QD

Merge

Figure 12. Dual-modality imaging of integrin α vβ3 in tumor vasculature. A, A schematic structure of the dual-modality positron emission tomography (PET)/near infrared fluorescence (NIRF) probe. B, Optical (after injection of quantum dot (QD)-RGD) and coronal microPET (after injection of 64Cu-DOTA-QD-RGD) images of U87MG tumor-bearing mice. Arrowheads indicate the tumors. C, Excellent overlay between CD31 and QD fluorescence, as well as between murine β3 and QD fluorescence, confirmed that DOTA-QD-RGD mainly targeted integrin αvβ3 on the tumor vasculature. Adapted with permission from Cai W et al.30

Multimodality Agents

SWNTs e xhibit unique size, shape, and ph ysical properties that make them promising candidates for biologic/biomedical applications. 226,227 We recentl y in vestigated the biodistribution of 64Cu-labeled SWNTs in mice by PET, ex vivo biodistribution, and Raman spectroscopy (Figure 13). 32 It w as found that these SWNTs are v ery stable in vivo, and the surface PEG chain length can significantly affect the biodistrib ution and circulation halflife. Ef fectively PEGylated SWNTs e xhibit relati vely long circulation half-life (about 2 h) and lo w uptake by the reticuloendothelial system (RES). Ef ficient targeting of inte grin αvβ3-positive tumor in mice w as achie ved using RGD peptide-conjugated PEGylated SWNTs. The intrinsic Raman signatures of SWNTs w ere used to directly probe the presence of SWNTs in mice tissues and confirm the radionuclide-based imaging results. Recently, the tumor accumulation of this SWNT -based construct in li ving mice has also been nonin vasively 271 imaged using a prototype Raman imaging system. After e valuating the phar macokinetics and tumor targeting efficacy, as well as conf irming the lack of toxicity,272,273 the use of SWNT as a nanoplatfor m for integrated multimodality imaging and molecular therap y is currently being explored. A

C

459

Radiolabeled nanoparticles represent a ne w class of probes that has enor mous potential for clinical applications. Dif ferent from other molecular imaging modalities, w here typicall y the nanopar ticle itself is detected , radionuclide-based imaging detects the radiolabel rather than the nanopar ticle. The nanopar ticle distribution is measured indirectly by assessing the localization of the radionuclide, w hich can pro vide quantitati ve measurement of the tumor-targeting efficacy and pharmacokinetics onl y if the radiolabel on the nanopar ticle is stab le enough under physiologic conditions. However, dissociation of the radionuclide (typicall y metal) from the chelator, and/or the radionuclide-containing pol ymer coating from the nanopar ticle, ma y occur , w hich can cause signif icant dif ference betw een the nanopar ticle distribution and the radionuclide distribution. Direct measurement of the SWNT in v arious tissues using its intrinsic Raman signal, as w ell as rigorous v alidation of the stability of the radiolabel on the nanoparticle, should always be carried out to obtain more reliab le results.32 The biodistribution of SWNTs in animals has also been repor ted using other radiolabeled SWNTs. 274,275 Surprisingly, these SWNTs w ere repor ted to under go either complete or par tial renal clearance in mice, with

B

D

Figure 13. Single-walled carbon nanotubes (SWNTs) for tumor integrin αvβ3 targeting. A, A schematic structure of the functionalized SWNT. The phospholipid (blue segment) binds strongly to the sidewall of the SWNT. The PEG chains render water solubility, and DOTA molecules are used to chelate 64Cu for positron emission tomography (PET) imaging. B, Two-dimensional projection microPET images of U87MG tumor-bearing mice at 8 h postinjection of SWNT-RGD without and with (denoted as “Block”) co-injection of the RGD peptide. Arrowheads indicate the tumors. C, Raman spectra of tissue homogenates confirm the presence of SWNT in the tumor. D, Good agreement between the biodistribution data obtained by PET imaging and ex vivo Raman measurements confirmed the in vivo stability and tumor-targeting efficacy of SWNT-RGD. Adapted with persmission from Liu Z et al.32

460

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

little uptake by the liver or other organs of the RES. These findings defy the general trend of high RES uptak e for nanoparticles and such contradictor y biodistribution results of SWNTs in mice deser ve further investigation/ validation. Studies ha ve shown that typicall y only molecules less than 70 kDa (a fe w nanometers in diameter) undergo renal clearance. 276–278 The SWNTs in these reports are typically more than 200 nm in length, even up to a fe w micrometers. 274,275 It is v ery unlik ely that the SWNTs can be cleared from the kidne y; ho wever, the shedded pol ymer coating containing the radionuclide (which was measured to evaluate the clearance pattern in these reports) will certainly undergo renal excretion.

DUAL-MODALITY AGENTS FOR SPECT AND OPTICAL IMAGING Dual-modality agents for both SPECT and optical imaging have also been reported. The cyclic peptide c(RGDfK) was labeled with both 111In and NIR dye, for γ scintigraphy and continuous-wave imaging of integrin αvβ3-positive tumors in x enografts, respecti vely.279 Twenty-four hours after administration of the dual-modality agent at a dose equi valent to 90 µCi of 111In and 5 nmol of the dye, whole-body γ scintigraphy and optical imaging were conducted. It was found that the tar get-to-background ratios of nuclear and optical imaging were similar for surface regions of interest, consistent with the origin of γ signal and NIR signal from a common tar geted peptide. Fur thermore, an anal ysis of signal-to-noise ratio versus contrast showed greater sensitivity of optical o ver nuclear imaging for subcutaneous tumor tar gets. Fluorescence-mediated tomo graphy is still not a mature technolo gy, thus accurate comparison between the two modalities was not feasible, however, such proof-of-principle study did demonstrate for the f irst time the direct comparison of molecular optical and planar nuclear imaging for surf ace and subsurf ace tumors. In a follow-up study, the same probe was found to enable noninvasive detection of the bound probe to αvβ3-positive tumors in li ving mice with both optical imaging and γ scintigraphy, w here optical imaging pro vided better resolution and sensitive detection of the superficial lesions while γ scintigraphy allowed for more sensiti ve detection of deeper structures.280 In a recent study , a dual-modality imaging agent, 111In-DTPA-Bz-NH-SA-K(IR-783-S-PhCO)-c(CGRRAGGSC)NH2, called DLIA-IL11R α, w as constructed.281 The c yclic peptide c(CGRRA GGSC), which is kno wn to tar get interleukin-11 receptor α-chain (IL-11Rα),282,283 serves as the targeting ligand. Cell-based studies re vealed that the c yclic peptide possessed the

targeting capability to IL-11R α after conjugation of both the optical and the radioisotope labels. Crossvalidation and direct comparison of optical and nuclear imaging of a tumor were achieved using a single injection, and preliminary results showed that the conjugate also had tumor-targeting efficacy in vivo (Figure 14). Compared with the size of either c(RGDfK) or c(CGRRAGGSC), the 111In-DTPA complex and the fluorescent dye are relatively bulky and may significantly affect the receptor binding. The reason w hy the imaging studies were carried out at 24 hours after injection for such a lo wmolecular w eight agent (typicall y small molecules are cleared from the circulation in minutes) w as also unclear . Recognizing such dra wbacks, the in vestigators recentl y used the same approach to label a monoclonal antibody , trastuzumab (Herceptin; Genentech Inc.). 284 Overexpression of the human epidermal growth factor receptor (HER) family has been implicated in many cancer types because of its pivotal role in signaling pathw ays that re gulate cellular proliferation, dif ferentiation, motility , and sur vival.285,286 Trastuzumab is an inhibitor y antibody de veloped against the e xtracellular domain of HER-2.287 In this repor t, (111In-DTPA)n-trastuzumab-(IRDye 800)m was synthesized, and fluorescence confocal microscop y was used to determine the molecular specif icity of (DTP A)n-trastuzumab(IRDye800)m in vitro in SKBr3 (HER-2-positi ve) and MDA-MB-231 (HER-2-ne gative) breast cancer cells. 284 Confocal microscopy revealed that (DTPA)n-trastuzumab(IRDye800)m had signif icantly g reater binding to SKBr3 than to MD A-MB-231 cells, and the binding occur red predominantly around the cell membrane, as HER-2 is a membrane-bound protein. In vi vo SPECT and optical imaging of SKBr3 xenografts injected with ( 111In-DTPA)ntrastuzumab-(IRDye800)m revealed signif icantly more uptake in the tumor re gion than in the contralateral muscle 111 region. This dual-modality agent, ( In-DTPA)ntrastuzumab-(IRDye800)m, may be an ef fective diagnostic biomarker capab le of tracking HER-2 o verexpression in patients with breast cancer . Replacing 111In with a PET isotope may further improve the imaging proper ties of this agent since PET has much higher sensiti vity than SPECT.

DUAL-MODALITY AGENTS FOR PET AND MR IMAGING PET has been routinely used in the clinic for staging and evaluating many types of cancer. The recently developed PET/CT scanner, already being used on a routine basis in clinical oncolo gy, g reatly f acilitated pinpointing the regions of increased acti vity on PET .4,288 However,

Multimodality Agents

461

A

B

White Light

SPECT/CT

Optical

Figure 14. Dual-modality single-photon emission computed tomography (SPECT)/optical imaging. A, Chemical structure of the dualmodality probe. B, SPECT/computed tomography (CT) and optical images of a tumor-bearing mouse at 24 h after probe injection. Reproduced with permission from Wang W et al.281

accurate localization of PET probe uptak e, e ven with PET/CT, can be v ery difficult in some cases due to the absence of identif iable anatomic str uctures, particularly in the abdomen.289–291 MRI has exquisite soft-tissue contrast and combination of PET/MR can have many synergistic ef fects. F or e xample, highl y accurate image registration can of fer the possibility of using the MR image to correct for PET partial volume effect (ie, image blurring introduced by the finite spatial resolution of the imaging system292) and aid in PET image reconstruction. PET/MR also has g reatly reduced radiation e xposure compared to PET/CT. Prototype PET/MR systems ha ve been implemented for small-animal imaging, and preliminary data for human studies has recentl y been presented in various conferences. 6,293 PET/MR, acquired in one measurement, has the potential to become the imaging modality of choice for v arious clinical applications, such as neurolo gic studies, cer tain types of cancer , stroke, and the emer ging f ield of stem cell therap y. The future of PET/MR scanners will greatly benefit from the use of dual-modality PET/MR imaging agents.

We recentl y de veloped an IO nanopar ticle-based probe for PET/MR imaging of tumor inte grin αvβ3 expression (Figure 15A).294 Poly(aspartic acid)-coated IO nanoparticles (PASP-IO) were synthesized, and the surface amino groups were coupled to cyclic RGD peptides for integrin αvβ3 targeting and DO TA chelators for PET 64 imaging (after labeling with Cu), respecti vely. The PASP-IO nanoparticle has a core size of 5 to 7 nm and a hydrodynamic diameter of ~40 nm. Both microPET and T2-weighted MR imaging showed integrin-specific delivery of RGD-P ASP-IO nanopar ticles to the U87MG human glioblastoma tumor (Figure 15B). On the basis of PET imaging, the tumor accumulation of 64Cu-labeled RGD-PASP-IO peak ed at about 4 hours after injection at 10.1 ± 2.1 percentage injected dose per g ram of tissue (%ID/g). In contrast, the nontar geted par ticle 64 Cu-DOTA-IO showed significantly lower tumor uptake at less than 5%ID/g. Blocking e xperiment with unconjugated RGD peptides also signif icantly reduced the tumor uptake of the dual-modality agent, demonstrating receptor specificity in vivo. T2-weighted MRI corroborated the

462

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

vivo imaging w as only achieved with PET (but not with optical imaging or MRI) due to its much better sensitivity and tissue penetration.

CONCLUSION AND FUTURE PERSPECTIVES B

Figure 15. An iron oxide (IO) nanoparticle-based dual-modality positron emission tomography (PET)/magnetic resonanace (MR) agent. A, Schematic illustration of the dual-modality probe. The DOTA chelator enables PET imaging after 64Cu-labeling. B, Coronal PET and T2-weighted MR images of tumor-bearing mice at 4 h after injection of 64Cu-labeled RGD-PASP-IO, PASP-IO, and RGD-PASPIO mixed with unconjugated RGD peptides (denoted as “Block”). Prussian blue staining of the U87MG tumor (integrin αvβ3-positive) tissue slices after scanning is also shown, where blue spots indicate the presence of IO nanoparticles.

PET findings. After in vivo PET and MRI scans, the animals w ere sacrif iced, and Pr ussian b lue staining of the tumor tissue confirmed integrin αvβ3-specific delivery of the RGD-PASP-IO nanopar ticles. Similar to most other nanoparticle-based imaging agents, the RES uptak e of this probe was also quite prominent. Overall, good correlation between the in vitro, in vi vo, and e x vi vo assays demonstrated that the imaging results accurately reflected the biodistribution of the dual-modality probe. This study represents the f irst e xample of in vi vo dual-modality PET/MR imaging using a single agent. In future studies, the constr uction of a fluorescentl y labeled analo gous conjugate, thus for ming a multimodality probe (PET , MR, and NIRF), ma y pro vide e ven more infor mation regarding the molecular mechanisms of diseases. Recently, one such multimodality probe w as reported.295 However, no tar geting ligand w as incor porated, and in

To date, the repor ted multimodality agents are almost exclusively dual-modality MR/optical fluorescence agents. Such an intriguing phenomenon is partially due to the e xploding interest in tw o types of nanopar ticles, IO and QD, for MR and optical imaging, respecti vely.72,162 However, w hether MR/optical combination is the most optimal dual-modality approach is debatable since neither modality is very quantitative. Other dual-modality agents (PET/optical, SPECT/optical, and PET/MR) ha ve been very rare. To the best of our knowledge, there is no report on dual-modality imaging agents that combine ultrasound and another modality although it is cer tainly feasib le. Dual-modality agents that combine radionuclide-based imaging (PET or SPECT), w hich is v ery sensiti ve and highly quantitati ve, and non-radionuclide based approaches, for e xample optical imaging (w hich can significantly facilitate ex vivo validation of the in vi vo data) and MRI (w hich can pro vide high resolution anatomical information), should be of par ticular interest for biomedical applications. One scenario w here a dual-modality approach will be par ticularly useful is that an initial whole-body PET scan can be car ried out to identify the location of tumor(s), and the optical component can subsequently help pinpointing the position during surgery for tumor resection. In one repor t, liposomes w ere labeled with both radionuclides and Gd 3+ for SPECT and MR imaging in vitro. 296 However, in vi vo imaging w as not achieved. PET has signif icantly higher sensiti vity than 267 SPECT (at least an order of magnitude) ; therefore, PET/optical and PET/MR agents deser ve much research effort in the future. Imaging agents that can be detected by more than two modalities are also expected to emerge, and a few abstracts were already presented in the Joint Molecular Imaging Conference held in Pro vidence, Rhode Island in 2007. Various combinations of different imaging modalities can ser ve different purposes depending on the study design. We en vision that a PET/MR/optical agent may f ind the most widespread use for future biomedical/clinical applications since such a combination provides e xtremely high sensiti vity (PET), quantitation capability (PET), e xcellent anatomic infor mation and soft-tissue contrast (MRI), as w ell as a means for e x vivo validation (optical) that itself can also be useful for highly sensitive imaging in certain sites of the human body.

Multimodality Agents

Although not discussed in this chapter , multiplexed imaging within the same modality will also ha ve man y clinical applications. F or SPECT imaging, dif ferent isotopes that emit dif ferent energy γ-rays can be dif ferentiated based on the ener gy.58 However, this approach has not been widel y pursued, and the concur rent use of se veral SPECT isotopes without perturbation of the underlying parent molecules w ould have to be possib le before a clear advantage could be achie ved for SPECT o ver PET in multiple-event imaging. For optical fluorescence imaging, QDs are ideal agents for multiple xing studies. 297 Many other optical imaging agents, such as SWNTs and gold nanoparticles, are also amenable to multiplexing and may find interesting biologic/biomedical applications.149 Most of the dual-modality imaging agents repor ted so f ar are based on cer tain nanopar ticles. The use of molecularly tar geted nanopar ticles af fords man y advantages o ver con ventional small-molecule-based approaches. F irst, hundreds, thousands, or e ven more imaging labels or a combination of labels for dif ferent imaging modalities can be attached to a single nanoparticle, w hich can lead to dramatic signal amplif ication. Second, multiple, potentially different, targeting ligands on the nanopar ticle can pro vide enhanced binding affinity/specificity. Third, the ability to inte grate v arious means to b ypass biolo gic bar riers will gi ve enhanced tar geting ef ficacy. Ultimatel y, the combination of different targeting ligands, imaging labels, therapeutic dr ugs, and man y other agents ma y allo w for effective and controlled delivery of therapeutic agents in patients, which can be nonin vasively and quantitati vely monitored in real time. Several barriers exist for in vivo applications in preclinical animal models and future clinical translation of such multifunctional/multimodality nanoparticles, among which are the biocompatibility, in vi vo kinetics, tar geting ef ficacy, acute and chronic toxicity, the ability to escape the RES, and costeffectiveness. With continuous ef forts by multidisciplinary approaches, the use of such nanoplatfor ms will shed ne w light on molecular diagnostics, molecular therapy, and personalized medicine. Nanoparticles usually suffer from poorer extravasation when compared with small molecules or proteins. 30,177,298 Thus, the majority of nanoparticle-based research is so far limited to vasculature-related diseases. The most promising applications of nanopar ticle-based agents will be in cardiovascular imaging/medicine, w here there is much less biolo gic bar rier for ef ficient deli very of nanopar ticles, and in oncology, where the leaky tumor vasculature can allo w for better tissue penetration than in nor mal organs/tissues. In many of the literature repor ts, it is not

463

clear w hether the nanopar ticle-based imaging or therapeutic agents are actually targeting the vasculature or the (tumor) cells. It is lik ely that some nanopar ticles do not extravasate at all or onl y e xtravasate to spaces in close proximity to the v essels since most of the nanopar ticles used so far are larger than 20 nm in diameter. Much care should be tak en when interpreting the imaging data, target specif icity, and tar geting ef ficacy. Rigorous in vivo/ex vi vo v alidation should al ways be car ried out before any nanoparticle-based multimodality agents enter clinical trials. Nanoparticle-based multimodality imaging, and potentially therap y, will pla y a k ey role in shaping twenty-first centur y patient management. To foster the continued discovery and development of better molecularly targeted multimodality agents, cooperati ve efforts are needed from cellular/molecular biologists to identify and v alidate no vel tar gets, chemists/radiochemists to synthesize and characterize the imaging probes, and engineers/medical physicists/mathematicians to develop high-sensitivity/high-resolution imaging de vices/hybrid instruments and better image reconstruction algorithms. Close par tnership among academic researchers, clinicians, the phar maceutical industr y, the National Institutes of Health, and the F ood and Dr ug Administration is also needed to quickl y appl y the ne wly de veloped multimodality agents to multiple f acets of patient management.

REFERENCES 1. Mankoff D A. A def inition of molecular imaging. J Nucl Med 2007;48:18N, 21N. 2. Massoud TF, Gambhir SS. Molecular imaging in living subjects: seeing fundamental biolo gical processes in a ne w light. Genes De v 2003;17:545–80. 3. Weissleder R, Mahmood U . Molecular imaging. Radiolo gy 2001;219:316–33. 4. Beyer T, Townsend DW, Brun T, et al. A combined PET/CT scanner for clinical oncology. J Nucl Med 2000;41:1369–79. 5. Even-Sapir E, Ler man H, Lie vshitz G, et al. L ymphoscintigraphy for sentinel node mapping using a h ybrid SPECT/CT system. J Nucl Med 2003;44:1413–20. 6. Catana C, Wu Y, Judenhofer MS, et al. Simultaneous acquisition of multislice PET and MR images: initial results with a MRcompatible PET scanner. J Nucl Med 2006;47:1968–76. 7. Phelps ME, Hof fman EJ, Mullani N A, Ter-Pogossian MM. Application of annihilation coincidence detection to transaxial reconstruction tomography. J Nucl Med 1975;16:210–24. 8. Wuest F, Ber ndt M, Kniess T. Carbon-11 labeling chemistr y based upon [ 11C]methyl iodide. Er nst Schering Res F ound Workshop 2007;183–213. 9. Elsinga PH. Radiophar maceutical chemistr y for positron emission tomography. Methods 2002;27:208–17. 10. Shai Y, Kirk KL, Channing MA, et al. 18F-labeled insulin: a prosthetic group methodolo gy for incor poration of a positron emitter into peptides and proteins. Biochemistry 1989;28:4801–6.

464

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

11. Wilbur DS. Radiohalo genation of proteins: an o verview of radionuclides, labeling methods and reagents for conjugate labeling. Bioconjug Chem 1992;3:432–70. 12. Vaidyanathan G, Zalutsk y MR. Impro ved synthesis of N-succinimidyl 4-[ 18F]fluorobenzoate and its application to the labeling of a monoclonal antibody fragment. Bioconjug Chem 1994;5:352–6. 13. Chen X, P ark R, Hou Y, et al. MicroPET imaging of brain tumor angiogenesis with 18F-labeled PEGylated RGD peptide. Eur J Nucl Med Mol Imaging 2004;31:1081–9. 14. Zhang X, Xiong Z, Wu X, et al. Quantitati ve PET imaging of tumor integrin αvβ3 expression with 18F-FRGD2. J Nucl Med 2006; 47:113–21. 15. Cai W, Olafsen T, Zhang X, et al. PET imaging of colorectal cancer in xenograft-bearing mice by use of an 18F-labeled T84.66 anti-carcinoembryonic antigen diabody. J Nucl Med 2007;48:304–10. 16. Haubner R, Kuhnast B, Mang C, et al. [ 18F]Galacto-RGD: synthesis, radiolabeling, metabolic stability , and radiation dose estimates. Bioconjug Chem 2004;15:61–9. 17. Herman LW, Fischman AJ, Tompkins RG, et al. The use of pentafluorophenyl derivatives for the 18F labelling of proteins. Nucl Med Biol 1994;21:1005–10. 18. Poethko T, Schottelius M, Thumshirn G, et al. Two-step methodology for high-yield routine radiohalo genation of peptides: 18F-labeled RGD and octreotide analogs. J Nucl Med 2004;45:892–902. 19. Schottelius M, Poethko T, Herz M, et al. F irst 18F-labeled tracer suitable for routine clinical imaging of sst receptor-expressing tumors using positron emission tomo graphy. Clin Cancer Res 2004; 10:3593–606. 20. Kilbourn MR, Dence CS, Welch MJ, Mathias CJ. Fluorine-18 labeling of proteins. J Nucl Med 1987;28:462–70. 21. Wester HJ, Hamacher K, Stoecklin G. A comparative study of n.c.a. fluorine-18 labeling of proteins via acylation and photochemical conjugation. Nucl Med Biol 1996;23:365–72. 18 22. Shiue CY , Wolf AP, Hainfeld JF . Synthesis of F-labelled N-(p-[18F]fluorophenyl)maleimide and its derivatives for labelling monoclonal antibody with 18F. J Labelled Comp Radiophar m 1998;26:287–9. 18 23. de Br uin B , K uhnast B , Hinnen F , et al. 1-[3-(2-[ F] fluoropyridin-3-yloxy)propyl]pyrrole-2,5-dione: design, synthesis, and radiosynthesis of a ne w [ 18F]fluoropyridine-based maleimide reagent for the labeling of peptides and proteins. Bioconjug Chem 2005;16:406–20. 24. Toyokuni T, Walsh JC, Dominguez A, et al. Synthesis of a ne w heterobifunctional link er, N-[4-(aminooxy)butyl] maleimide, for f acile access to a thiol-reactive 18F-labeling agent. Bioconjug Chem 2003; 14:1253–9. 25. Cai W, Zhang X, Wu Y, Chen X. A thiol-reactive 18F-labeling agent, N-[2-(4-18F-fluorobenzamido)ethyl]maleimide (18F-FBEM), and the synthesis of RGD peptide-based tracer for PET imaging of αvβ3 integrin expression. J Nucl Med 2006;47:1172–80. 26. Ma D, Lu F, Overstreet T, et al. No vel chelating agents for potential clinical applications of copper. Nucl Med Biol 2002;29:91–105. 27. Wadas TJ, Wong EH, Weisman GR, Anderson CJ. Copper chelation chemistry and its role in copper radiophar maceuticals. Cur r Pharm Des 2007;13:3–16. 28. Wu Y, Zhang X, Xiong Z, et al. MicroPET imaging of glioma αv-integrin expression using 64Cu-labeled tetrameric RGD eptide. J Nucl Med 2005;46:1707–18. 29. Cai W, Chen K, He L, et al. Quantitative PET of EGFR expression in xenograft-bearing mice using 64Cu-labeled cetuximab, a chimeric anti-EGFR monoclonal antibody . Eur J Nucl Med Mol Imaging 2007;34:850–8. 30. Cai W, Chen K, Li ZB , et al. Dual-function probe for PET and near infrared fluorescence imaging of tumor v asculature. J Nucl Med 2007;48:1862–70.

31. Hsu AR, Cai W, Veeravagu A, et al. Multimodality molecular imaging of gliob lastoma g rowth inhibition with v asculature-targeting fusion toxin VEGF121/rGel. J Nucl Med 2007;48:445–54. 32. Liu Z, Cai W, He L, et al. In vi vo biodistribution and highly efficient tumour targeting of carbon nanotubes in mice. Nat Nanotechnol 2007;2:47–52. 33. Cai W, Chen K, Mohamedali KA, et al. PET of v ascular endothelial g rowth f actor receptor e xpression. J Nucl Med 2006; 47:2048–56. 34. Cai W, Ebrahimnejad A, Chen K, et al. Quantitative radioimmunoPET imaging of EphA2 in tumour -bearing mice. Eur J Nucl Med Mol Imaging 2007;34:2024–36. 35. Wang H, Cai W, Chen K, et al. A new PET tracer specif ic for vascular endothelial g rowth f actor receptor 2. Eur J Nucl Med Mol Imaging 2007;34:2001–10. 36. Li ZB, Cai W, Cao Q, et al. 64Cu-labeled tetrameric and octameric RGD peptides for small-animal PET of tumor αvβ3 integrin expression. J Nucl Med 2007;48:1162–71. 37. Cai W, Wu Y, Chen K, et al. In vitro and in vivo characterization of 64 Cu-labeled Abegrin, a humanized monoclonal antibody against integrin αvβ3. Cancer Res 2006;66:9673–81. 38. Anderson CJ, Jones LA, Bass LA, et al. Radiotherap y, toxicity and dosimetry of copper -64-TETA-octreotide in tumor -bearing rats. J Nucl Med 1998;39:1944–51. 39. Cutler PD, Schwarz SW, Anderson CJ, et al. Dosimetr y of copper-64labeled monoclonal antibody 1A3 as deter mined by PET imaging of the torso. J Nucl Med 1995;36:2363–71. 40. Sprague JE, Peng Y, Sun X, et al. Preparation and biolo gical evaluation of Copper -64-labeled Tyr3-octreotate using a cross-bridged macrocyclic chelator. Clin Cancer Res 2004;10:8674–82. 41. Hutchins GD. Quantitative evaluation of myocardial blood flow with [13N]ammonia. Cardiology 1997;88:106–15. 42. Ter-Pogossian MM, Hersco vitch P . Radioacti ve o xygen-15 in the study of cerebral b lood flow, blood volume, and oxygen metabolism. Semin Nucl Med 1985;15:377–94. 43. Gabriel M, Decristoforo C, K endler D , et al. 68Ga-DOTA-Tyr3octreotide PET in neuroendocrine tumors: comparison with somatostatin receptor scintig raphy and CT . J Nucl Med 2007; 48:508–18. 44. Henze M, Dimitrakopoulou-Strauss A, Milker-Zabel S, et al. Characterization of 68Ga-DOTA-D-Phe1-Tyr3-octreotide kinetics in patients with meningiomas. J Nucl Med 2005;46:763–9. 45. Clarke ET, Mar tell AE. Stabilities of the iron(III), gallium(III) and indium(III) chelates of N ,N’,N”-triazacyclononanetriacetic acid. Inorg Chim Acta 1991;181:273–80. 46. Clarke ET, Martell AE. Stabilities of trivalent metal ion complexes of the tetraacetate derivatives of 12-, 13- and 14-membered tetraazamacrocycles. Inorg Chim Acta 1991;190:37–46. 47. Perk LR, Visser GW, Vosjan MJ , et al. 89Zr as a PET sur rogate radioisotope for scouting biodistribution of the therapeutic radiometals 90Y and 177Lu in tumor-bearing nude mice after coupling to the inter nalizing antibody cetuximab. J Nucl Med 2005; 46:1898–906. 48. Nagengast WB, de Vries EG, Hospers GA, et al. In vivo VEGF imaging with radiolabeled be vacizumab in a human o varian tumor xenograft. J Nucl Med 2007;48:1313–9. 49. Clifford T, Bos well CA, Biddlecombe GB , et al. Validation of a novel CHX-A ’’ derivative suitab le for peptide conjugation: small animal PET/CT imaging using yttrium-86-CHX-A ’’octreotide. J Med Chem 2006;49:4297–304. 50. McQuade P, Miao Y, Yoo J, et al. Imaging of melanoma using 64Cuand 86Y-DOTA-ReCCMSH(Arg11), a c yclized peptide analo gue of α-MSH. J Med Chem 2005;48:2985–92. 51. Rowland DJ, Tu Z, Xu J , et al. Synthesis and in vi vo evaluation of 2 high-affinity 76Br-labeled sigma2-receptor ligands. J Nucl Med 2006;47:1041–8.

Multimodality Agents

52. Nickles RJ , Nunn AD, Stone CK, Christian BT . Technetium-94mteboroxime: synthesis, dosimetry and initial PET imaging studies. J Nucl Med 1993;34:1058–66. 53. Rault E, Vandenberghe S, Van Holen R, et al. Comparison of image quality of dif ferent iodine isotopes (I-123, I-124, and I-131). Cancer Biother Radiopharm 2007;22:423–30. 54. Divgi CR, P andit-Taskar N, Jungbluth AA, et al. Preoperati ve characterisation of clear -cell renal carcinoma using iodine-124labelled antibody chimeric G250 ( 124I-cG250) and PET in patients with renal masses: a phase I trial. Lancet Oncol 2007; 8:304–10. 55. Peremans K, Cor nelissen B, Van Den Bossche B , et al. A review of small animal imaging planar and pinhole spect Gamma camera imaging. Vet Radiol Ultrasound 2005;46:162–70. 56. Kjaer A. Molecular imaging of cancer using PET and SPECT . Adv Exp Med Biol 2006;587:277–84. 57. Chatziioannou AF. Instrumentation for molecular imaging in preclinical research: Micro-PET and Micro-SPECT . Proc Am Thorac Soc 2005;2:510–11, 533–6. 58. Berman DS, Kiat H, Van Train K, et al. Dual-isotope myocardial perfusion SPECT with rest thallium-201 and stress Tc-99m sestamibi. Cardiol Clin 1994;12:261–70. 59. Liu S, Edwards DS. 99mTc-labeled small peptides as diagnostic radiopharmaceuticals. Chem Rev 1999;99:2235–68. 60. Banerjee S, Pillai MR, Ramamoorthy N. Evolution of Tc-99m in diagnostic radiopharmaceuticals. Semin Nucl Med 2001;31:260–77. 61. Mease RC, Lambert C. Newer methods of labeling diagnostic agents with Tc-99m. Semin Nucl Med 2001;31:278–85. 62. Banerjee SR, Maresca KP, Francesconi L, et al. New directions in the coordination chemistr y of 99mTc: a reflection on technetium core structures and a strate gy for ne w chelate design. Nucl Med Biol 2005;32:1–20. 63. Virgolini I, Traub T, Novotny C, et al. Experience with indium-111 and yttrium-90-labeled somatostatin analo gs. Cur r Phar m Des 2002;8:1781–807. 64. Eberle AN, F roidevaux S. Radiolabeled α-melanocyte-stimulating hormone analo gs for receptor -mediated tar geting of melanoma: from tritium to indium. J Mol Reco gnit 2003;16:248–54. 99m 123 65. Kulkarni PV . Recent de velopments in Tc and Iradiopharmaceuticals for SPECT imaging. Int J Rad Appl Instrum B 1991;18:647–54. 66. Lind P. 131I whole body scintig raphy in th yroid cancer patients. Q J Nucl Med 1999;43:188–94. 67. Woodrum DT, Gauger PG. Role of 131I in the treatment of w ell differentiated thyroid cancer. J Surg Oncol 2005;89:114–21. 68. Pathak AP, Gimi B, Glunde K, et al. Molecular and functional imaging of cancer: adv ances in MRI and MRS. Methods Enzymol 2004;386:3–60. 69. de Roos A, Door nbos J, Baleriaux D , et al. Clinical applications of gadolinium-DTPA in MRI. Magn Reson Annu 1988;113–45. 70. Zhang Z, Nair SA, McMurry TJ. Gadolinium meets medicinal chemistry: MRI contrast agent de velopment. Cur r Med Chem 2005;12:751–78. 71. Seo WS, Lee JH, Sun X, et al. F eCo/graphitic-shell nanocrystals as advanced magnetic-resonance-imaging and near -infrared agents. Nat Mater 2006;5:971–76. 72. Thorek DL, Chen AK, Czupryna J, Tsourkas A. Super paramagnetic iron oxide nanoparticle probes for molecular imaging. Ann Biomed Eng 2006;34:23–38. 73. Bertini I, K owalewski J , Luchinat C, P arigi G. Cross cor relation between the dipole-dipole interaction and the Curie spin relaxation: the ef fect of anisotropic magnetic susceptibility . J Magn Reson 2001;152:103–8. 74. Shamsi K, Balzer T, Saini S, et al. Superparamagnetic iron oxide particles (SH U 555 A): e valuation of ef ficacy in three doses for hepatic MR imaging. Radiology 1998;206:365–71.

465

75. Reimer P, Jahnke N, Fiebich M, et al. Hepatic lesion detection and characterization: v alue of nonenhanced MR imaging, superparamagnetic iron oxide-enhanced MR imaging, and spiral CTROC analysis. Radiology 2000;217:152–8. 76. Weissleder R, Hahn PF , Stark DD , et al. Super paramagnetic iron oxide: enhanced detection of focal splenic tumors with MR imaging. Radiology 1988;169:399–403. 77. Weissleder R, Stark DD, Rummeny EJ, et al. Splenic lymphoma: ferrite-enhanced MR imaging in rats. Radiolo gy 1988;166:423–30. 78. Mack MG, Balzer JO, Straub R, et al. Superparamagnetic iron oxideenhanced MR imaging of head and neck lymph nodes. Radiology 2002;222:239–44. 79. Anzai Y, Piccoli CW , Outw ater EK, et al. Ev aluation of neck and body metastases to nodes with fer umoxtran 10-enhanced MR imaging: phase III safety and ef ficacy study. Radiolo gy 2003; 228:777–88. 80. Zimmer C, Weissleder R, Poss K, et al. MR imaging of phagoc ytosis in experimental gliomas. Radiology 1995;197:533–8. 81. Moore A, Marecos E, Bogdanov A Jr, Weissleder R. Tumoral distribution of long-circulating de xtran-coated iron oxide nanoparticles in a rodent model. Radiology 2000;214:568–74. 82. Harisinghani MG, Barentsz J , Hahn PF, et al. Nonin vasive detection of clinically occult l ymph-node metastases in prostate cancer . N Engl J Med 2003;348:2491–99. 83. Saleh A, Schroeter M, Jonkmanns C, et al. In vi vo MRI of brain inflammation in human ischaemic strok e. Brain 2004;127: 1670–77. 84. Dousset V, Brochet B, Deloire MS, et al. MR imaging of relapsing multiple sclerosis patients using ultra-small-par ticle iron o xide and compared with gadolinium. AJNR Am J Neuroradiol 2006;27: 1000–5. 85. Enochs WS, Harsh G, Hochber g F, Weissleder R. Impro ved delineation of human brain tumors on MR images using a long-circulating, super paramagnetic iron o xide agent. J Magn Reson Imaging 1999;9:228–32. 86. Corot C, P etry KG, Trivedi R, et al. Macrophage imaging in central nervous system and in carotid atherosclerotic plaque using ultrasmall super paramagnetic iron o xide in magnetic resonance imaging. Invest Radiol 2004;39:619–25. 87. Shapiro EM, Skrtic S, Sharer K, et al. MRI detection of single par ticles for cellular imaging. Proc Natl Acad Sci USA 2004; 101:10901–6. 88. Shapiro EM, Sharer K, Skr tic S, Koretsky AP. In vivo detection of single cells by MRI. Magn Reson Med 2006;55:242–9. 89. Heyn C, Ronald JA, Mackenzie LT, et al. In vivo magnetic resonance imaging of single cells in mouse brain with optical v alidation. Magn Reson Med 2006;55:23–9. 90. Stuckey DJ, Carr CA, Martin-Rendon E, et al. Iron particles for noninvasive monitoring of bone mar row stromal cell eng raftment into, and isolation of viable engrafted donor cells from, the heart. Stem Cells 2006;24:1968–75. 91. Arbab AS, Jordan EK, Wilson LB, et al. In vi vo trafficking and targeted delivery of magnetically labeled stem cells. Hum GeneTher 2004;15:351–60. 92. Hill JM, Dick AJ, Raman VK, et al. Serial cardiac magnetic resonance imaging of injected mesench ymal stem cells. Circulation 2003;108:1009–14. 93. de Vries IJ, Lesterhuis WJ, Barentsz JO , et al. Magnetic resonance tracking of dendritic cells in melanoma patients for monitoring of cellular therapy. Nat Biotechnol 2005;23:1407–13. 94. Ahrens ET, Feili-Hariri M, Xu H, et al. Receptor-mediated endocytosis of iron-oxide particles provides efficient labeling of dendritic cells for in vivo MR imaging. Magn Reson Med 2003;49:1006–13. 95. Moats RA, F raser SE, Meade TJ. A “smar t” magnetic resonance imaging agent that repor ts on specif ic enzymic acti vity. Angew Chem Int Ed Engl 1997;36:726–8.

466

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

96. Louie AY, Huber MM, Ahrens ET, et al. In vivo visualization of gene expression using magnetic resonance imaging. Nat Biotechnol 2000;18:321–5. 97. Bogdanov A Jr, Matuszewski L, Bremer C, et al. Oligomerization of paramagnetic substrates result in signal amplif ication and can be used for MR imaging of molecular tar gets. Mol Imaging 2002;1:16–23. 98. Weissleder R, Moore A, Mahmood U , et al. In vi vo magnetic resonance imaging of transgene expression. Nat Med 2000;6:351–5. 99. Golman K, Petersson JS. Metabolic imaging and other applications of hyperpolarized 13C. Acad Radiol 2006;13:932–42. 100. Fain SB , K orosec FR, Holmes JH, et al. Functional lung imaging using h yperpolarized gas MRI. J Magn Reson Imaging 2007;25:910–23. 101. Albert MS, Cates GD, Driehuys B, et al. Biological magnetic resonance imaging using laser-polarized 129Xe. Nature 1994;370:199–201. 102. Middleton H, Black RD, Saam B, et al. MR imaging with hyperpolarized 3He gas. Magn Reson Med 1995;33:271–5. 103. de Lange EE, Altes TA, Patrie JT, et al. Evaluation of asthma with hyperpolarized helium-3 MRI: cor relation with clinical severity and spirometry. Chest 2006;130:1055–62. 104. Holmes JH, Korosec FR, Du J, et al. Imaging of lung ventilation and respiratory dynamics in a single v entilation cycle using hyperpolarized He-3 MRI. J Magn Reson Imaging 2007;26:630–6. 105. Stavngaard T, So gaard LV, Mor tensen J, et al. Hyper polarized 3He MRI and 81mKr SPECT in chronic obstructive pulmonary disease. Eur J Nucl Med Mol Imaging 2005;32:448–57. 106. Parraga G, Ouriadov A, Evans A, et al. Hyper polarized 3He ventilation defects and apparent dif fusion coef ficients in chronic obstructive pulmonar y disease: preliminar y results at 3.0 Tesla. Invest Radiol 2007;42:384–91. 107. McMahon CJ, Dodd JD, Hill C, et al. Hyper polarized 3Helium magnetic resonance ventilation imaging of the lung in c ystic fibrosis: comparison with high resolution CT and spirometr y. Eur Radiol 2006;16:2483–90. 108. van Beek EJ, Hill C, Woodhouse N, et al. Assessment of lung disease in children with c ystic f ibrosis using h yperpolarized 3-Helium MRI: comparison with Shw achman score, Chrispin-Nor man score and spirometry. Eur Radiol 2007;17:1018–24. 109. Zaporozhan J, Ley S, Gast KK, et al. Functional anal ysis in singlelung transplant recipients: a comparative study of high-resolution CT, 3He-MRI, and pulmonar y function tests. Chest 2004;125:173–81. 110. Bink A, Hanisch G, Kar g A, et al. Clinical aspects of the apparent diffusion coefficient in 3He MRI: results in healthy volunteers and patients after lung transplantation. J Magn Reson Imaging 2007;25:1152–8. 111. Golman K, Axelsson O, Johannesson H, et al. Parahydrogen-induced polarization in imaging: subsecond 13C angiography. Magn Reson Med 2001;46:1–5. 112. Goldman M, Johannesson H, Axelsson O, Karlsson M. Hyper polarization of 13C through order transfer from parah ydrogen: a ne w contrast agent for MRI. Magn Reson Imaging 2005;23:153–7. 113. Ardenkjaer-Larsen JH, Fridlund B, Gram A, et al. Increase in signalto-noise ratio of > 10,000 times in liquid-state NMR. Proc Natl Acad Sci U S A 2003;100:10158–63. 114. Mansson S, Johansson E, Magnusson P , et al. 13C imaging-a ne w diagnostic platform. Eur Radiol 2006;16:57–67. 115. Ward KM, Aletras AH, Balaban RS. A new class of contrast agents for MRI based on proton chemical exchange dependent saturation transfer (CEST). J Magn Reson 2000;143:79–87. 116. Zhang S, Merritt M, Woessner DE, et al. PARACEST agents: modulating MRI contrast via w ater proton e xchange. Acc Chem Res 2003;36:783–90. 117. Woods M, Woessner DE, Sherry AD. Paramagnetic lanthanide complexes as PARACEST agents for medical imaging. Chem Soc Rev 2006;35:500–11.

118. Aime S, Barge A, Delli Castelli D, et al. Paramagnetic lanthanide(III) complexes as pH-sensitive chemical exchange saturation transfer (CEST) contrast agents for MRI applications. Magn Reson Med 2002;47:639–48. 119. Zhang S, Mallo y CR, Sher ry AD. MRI ther mometry based on PARACEST agents. J Am Chem Soc 2005;127:17572–3. 120. Aime S, Delli Castelli D, Fedeli F, Terreno E. A paramagnetic MRICEST agent responsi ve to lactate concentration. J Am Chem Soc 2002;124:9364–5. 121. Zhang S, Trokowski R, Sherry AD. A paramagnetic CEST agent for imaging glucose by MRI. J Am Chem Soc 2003;125:15288–9. 122. Trokowski R, Zhang S, Sher ry AD. Cyclen-based phen ylboronate ligands and their Eu 3+ complexes for sensing glucose b y MRI. Bioconjug Chem 2004;15:1431–40. 123. Aime S, Delli Castelli D , Terreno E. Supramolecular adducts between pol y-L-arginine and [TmIIIdotp]: a route to sensitivity-enhanced magnetic resonance imaging-chemical exchange saturation transfer agents. Angew Chem Int Ed Engl 2003;42:4527–9. 124. Trokowski R, Ren J , Kalman FK, Sher ry AD. Selective sensing of zinc ions with a PARACEST contrast agent. Angew Chem Int Ed Engl 2005;44:6920–3. 125. Yoo B , P agel MD . A P ARACEST MRI contrast agent to detect enzyme activity. J Am Chem Soc 2006;128:14032–3. 126. Yoo B, Raam MS, Rosenb lum RM, P agel MD. Enzyme-responsive PARACEST MRI contrast agents: a ne w biomedical imaging approach for studies of the proteasome. Contrast Media Mol Imaging 2007;2:189–98. 127. Bloch SH, Dayton PA, Ferrara KW. Targeted imaging using ultrasound contrast agents. Pro gess and oppor tunities for clinical and research applications. IEEE Eng Med Biol Mag 2004; 23:18–29. 128. Wink MH, Wijkstra H, De La Rosette JJ , Grimber gen CA. Ultrasound imaging and contrast agents: a safe alter native to MRI? Minim Invasive Ther Allied Technol 2006;15:93–100. 129. Cosgrove D. Ultrasound contrast agents: an o verview. Eur J Radiol 2006;60:324–30. 130. Jakobsen JA. Ultrasound contrast agents: clinical applications. Eur Radiol 2001;11:1329–37. 131. Stride E, Saf fari N . Microbubb le ultrasound contrast agents: a review. Proc Inst Mech Eng [H] 2003;217:429–47. 132. Kaufmann BA, Lindner JR. Molecular imaging with tar geted contrast ultrasound. Curr Opin Biotechnol 2007;18:11–6. 133. Ellegala DB, Leong-Poi H, Carpenter JE, et al. Imaging tumor angiogenesis with contrast ultrasound and microb ubbles tar geted to αvβ3. Circulation 2003;108:336–41. 134. Leong-Poi H, Christiansen J , Klibanov AL, et al. Nonin vasive assessment of angio genesis b y ultrasound and microbubb les tar geted to αv-integrins. Circulation 2003;107:455–60. 135. Leong-Poi H, Christiansen J, Heppner P, et al. Assessment of endogenous and therapeutic arteriogenesis by contrast ultrasound molecular imaging of integrin expression. Circulation 2005;111:3248–54. 136. Korpanty G, Carbon JG, Grayburn PA, et al. Monitoring response to anticancer therap y by targeting microbubbles to tumor v asculature. Clin Cancer Res 2007;13:323–30. 137. Willmann JK, Paulmurugan R, Chen K, et al. US imaging of tumor angiogenesis with microb ubbles targeted to v ascular endothelial growth f actor receptor type 2 in mice. Radiolo gy 2008; 246:508–18. 138. Dayton P A, Rychak JJ . Molecular ultrasound imaging using microbubble contrast agents. Front Biosci 2007;12:5124–42. 139. Contag CH, Bachmann MH. Advances in in vi vo bioluminescence imaging of gene e xpression. Annu Re v Biomed Eng 2002; 4:235–60. 140. Contag CH, Spilman SD, Contag PR, et al. Visualizing gene expression in li ving mammals using a bioluminescent repor ter. Photochem Photobiol 1997;66:523–31.

Multimodality Agents

141. Lipshutz GS, Flebbe-Rehw aldt L, Gaensler KM. Ree xpression following readministration of an adeno viral vector in adult mice after initial in utero adeno viral administration. Mol Ther 2000; 2:374–80. 142. Sweeney TJ, Mailander V, Tucker AA, et al. Visualizing the kinetics of tumor-cell clearance in li ving animals. Proc Natl Acad Sci U S A 1999;96:12044–9. 143. Paulmurugan R, Umeza wa Y, Gambhir SS. Nonin vasive imaging of protein-protein interactions in li ving subjects b y using reporter protein complementation and reconstitution strate gies. Proc Natl Acad Sci U S A 2002;99:15608–13. 144. De A, Gambhir SS. Noninvasive imaging of protein-protein interactions from li ve cells and li ving subjects using bioluminescence resonance energy transfer. FASEB J 2005;19:2017–9. 145. Cao YA, Wagers AJ, Beilhack A, et al. Shifting foci of hematopoiesis during reconstitution from single stem cells. Proc Natl Acad Sci USA 2004;101:221–6. 146. Low AF, Tearney GJ, Bouma BE, Jang IK. Technology Insight: optical coherence tomo graphy—current status and future de velopment [quiz 172]. Nat Clin Pract Cardio vasc Med 2006;3:154–62. 147. Loo C, Lo wery A, Halas N , et al. Immunotar geted nanoshells for integrated cancer imaging and therapy. Nano Lett 2005;5:709–11. 148. Cang H, Sun T, Li ZY, et al. Gold nanocages as contrast agents for spectroscopic optical coherence tomo graphy. Opt Lett 2005; 30:3048–50. 149. Hu M, Chen J , Li ZY, et al. Gold nanostr uctures: engineering their plasmonic properties for biomedical applications. Chem Soc Re v 2006;35:1084–94. 150. Ntziachristos V. Fluorescence molecular imaging. Annu Rev Biomed Eng 2006;8:1–33. 151. Mansfield JR, Gossage KW, Hoyt CC, Le venson RM. Autofluorescence removal, multiplexing, and automated analysis methods for in-vivo fluorescence imaging. J Biomed Opt 2005;10:41207. 152. Ntziachristos V, Tung CH, Bremer C, Weissleder R. Fluorescence molecular tomography resolves protease activity in vivo. Nat Med 2002;8:757–60. 153. Montet X, Ntziachristos V, Grimm J, Weissleder R. Tomographic fluorescence mapping of tumor targets. Cancer Res 2005;65:6330–6. 154. Zaheer A, Wheat TE, F rangioni JV. IRDy e78 conjugates for near infrared fluorescence imaging. Mol Imaging 2002;1:354–64. 155. Licha K, Riefke B, Ntziachristos V, et al. Hydrophilic c yanine dyes as contrast agents for near -infrared tumor imaging: synthesis, photophysical properties and spectroscopic in vi vo characterization. Photochem Photobiol 2000;72:392–8. 156. Hsu ER, Anslyn EV , Dhar mawardhane S, et al. A f ar-red fluorescent contrast agent to image epider mal g rowth f actor receptor expression. Photochem Photobiol 2004;79:272–9. 157. Cheng Z, Wu Y, Xiong Z, et al. Near-infrared fluorescent RGD peptides for optical imaging of inte grin αvβ3 expression in li ving mice. Bioconjug Chem 2005;16:1433–41. 158. Wu Y, Cai W, Chen X. Near-infrared fluorescence imaging of tumor integrin αvβ3 expression with Cy7-labeled RGD multimers. Mol Imaging Biol 2006;8:226–36. 159. van Roessel P, Brand AH. Imaging into the future: visualizing gene expression and protein interactions with fluorescent proteins. Nat Cell Biol 2002;4:E15–20. 160. Cai W, Hsu AR, Li ZB, Chen X. Are quantum dots ready for in vi vo imaging in human subjects? Nanoscale Res Lett 2007;2:265–81. 161. Alivisatos P. The use of nanocr ystals in biolo gical detection. Nat Biotechnol 2004;22:47–52. 162. Michalet X, Pinaud FF , Bentolila LA, et al. Quantum dots for li ve cells, in vivo imaging, and diagnostics. Science 2005;307:538–44. 163. Medintz IL, Uy eda HT, Goldman ER, Mattoussi H. Quantum dot bioconjugates for imaging, labelling and sensing. Nat Mater 2005;4:435–46.

467

164. Alivisatos AP, Gu W, Larabell C. Quantum dots as cellular probes. Annu Rev Biomed Eng 2005;7:55–76. 165. Li ZB , Cai W, Chen X. Semiconductor quantum dots for in vi vo imaging. J Nanosci Nanotechnol 2007;7:2567–81. 166. Dubertret B, Skourides P, Norris DJ, et al. In vi vo imaging of quantum dots encapsulated in phospholipid micelles. Science 2002;298:1759–62. 167. Voura EB, Jaiswal JK, Mattoussi H, Simon SM. Tracking metastatic tumor cell e xtravasation with quantum dot nanocr ystals and fluorescence emission-scanning microscopy. Nat Med 2004;10:993–8. 168. Rieger S, K ulkarni RP, Darcy D, et al. Quantum dots are po werful multipurpose vital labeling agents in zebrafish embryos. Dev Dyn 2005;234:670–81. 169. Larson DR, Zipfel WR, Williams RM, et al. Water-soluble quantum dots for multiphoton fluorescence imaging in vi vo. Science 2003;300:1434–6. 170. Stroh M, Zimmer JP, Duda DG, et al. Quantum dots spectrall y distinguish multiple species within the tumor milieu in vi vo. Nat Med 2005;11:678–82. 171. Kim S, Lim YT, Soltesz EG, et al. Near -infrared fluorescent type II quantum dots for sentinel l ymph node mapping. Nat Biotechnol 2004;22:93–7. 172. Zimmer JP, Kim SW, Ohnishi S, et al. Size series of small indium arsenide-zinc selenide core-shell nanocr ystals and their application to in vivo imaging. J Am Chem Soc 2006;128:2526–7. 173. Ballou B, Ernst LA, Andreko S, et al. Sentinel l ymph node imaging using quantum dots in mouse tumor models. Bioconjug Chem 2007;18:389–96. 174. Thorne RG, Nicholson C. In vi vo diffusion analysis with quantum dots and dextrans predicts the width of brain e xtracellular space. Proc Natl Acad Sci USA 2006;103:5567–72. 175. Jackson H, Muhammad O , Daneshvar H, et al. Quantum dots are phagocytized by macrophages and colocalize with experimental gliomas [discussion]. Neurosurgery 2007;60:524–30. 176. Gao X, Cui Y, Le venson RM, et al. in vi vo cancer tar geting and imaging with semiconductor quantum dots. Nat Biotechnol 2004; 22:969–76. 177. Cai W, Shin DW, Chen K, et al. P eptide-labeled near-infrared quantum dots for imaging tumor v asculature in li ving subjects. Nano Lett 2006;6:669–76. 178. Tada H, Higuchi H, Wanatabe TM, Ohuchi N. in vivo real-time tracking of single quantum dots conjugated with monoclonal antiHER2 antibody in tumors of mice. Cancer Res 2007;67:1138–44. 179. Funovics M, Weissleder R, Tung CH. Protease sensors for bioimaging. Anal Bioanal Chem 2003;377:956–63. 180. Weissleder R, Tung CH, Mahmood U, Bogdanov A Jr. In vivo imaging of tumors with protease-acti vated near -infrared fluorescent probes. Nat Biotechnol 1999;17:375–8. 181. Tung CH, Bredo w S, Mahmood U , Weissleder R. Preparation of a cathepsin D sensiti ve near-infrared fluorescence probe for imaging. Bioconjug Chem 1999;10:892–6. 182. Galande AK, Hilderbrand SA, Weissleder R, Tung CH. Enzymetargeted fluorescent imaging probes on a multiple antigenic peptide core. J Med Chem 2006;49:4715–20. 183. Jaffer FA, Kim DE, Quinti L, et al. Optical visualization of cathepsin K activity in atherosclerosis with a novel, protease-activatable fluorescence sensor. Circulation 2007;115:2292–8. 184. Bremer C, Tung CH, Weissleder R. in vivo molecular target assessment of matrix metalloproteinase inhibition. Nat Med 2001;7:743–8. 185. Pham W, Choi Y, Weissleder R, Tung CH. De veloping a peptidebased near -infrared molecular probe for protease sensing. Bioconjug Chem 2004;15:1403–7. 186. Messerli SM, Prabhakar S, Tang Y, et al. A novel method for imaging apoptosis using a caspase-1 near -infrared fluorescent probe. Neoplasia 2004;6:95–105.

468

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

187. Pham W, Weissleder R, Tung CH. An azulene dimer as a near-infrared quencher. Angew Chem Int Ed Engl 2002;41:3659–62, 3519. 188. Shah K, Tung CH, Chang CH, et al. In vivo imaging of HIV protease activity in amplicon v ector-transduced gliomas. Cancer Res 2004;64:273–8. 189. Law B, Curino A, Bugge TH, et al. Design, synthesis, and characterization of urokinase plasmino gen-activator-sensitive near -infrared reporter. Chem Biol 2004;11:99–106. 190. Hsiao JK, Law B, Weissleder R, Tung CH. In-vivo imaging of tumor associated urokinase-type plasmino gen acti vator acti vity. J Biomed Opt 2006;11:34013. 191. Tung CH, Gerszten RE, Jaf fer F A, Weissleder R. A no vel near infrared fluorescence sensor for detection of thrombin acti vation in blood. Chembiochem 2002;3:207–11. 192. Ossovskaya VS, Bunnett NW. Protease-activated receptors: contribution to physiology and disease. Physiol Rev 2004;84:579–621. 193. Mortele KJ, McTavish J, Ros PR. Cur rent techniques of computed tomography. Helical CT, multidetector CT, and 3D reconstruction. Clin Liver Dis 2002;6:29–52. 194. Nelson RC, Chezmar JL, P eterson JE, Ber nardino ME. Contrastenhanced CT of the li ver and spleen: comparison of ionic and nonionic contrast agents. AJR Am J Roentgenol 1989;153:973–6. 195. Remy-Jardin M, Bahepar J , Laf itte JJ, et al. Multi-detector ro w CT angiography of pulmonar y circulation with gadolinium-based contrast agents: prospecti ve evaluation in 60 patients. Radiolo gy 2006;238:1022–35. 196. McIntire GL, Bacon ER, Illig KJ, et al. Time course of nodal enhancement with CT X-ray nanoparticle contrast agents: effect of particle size and chemical structure. Invest Radiol 2000;35:91–6. 197. Wisner ER, Katzberg RW, Koblik PD, et al. Iodinated nanopar ticles for indirect computed tomo graphy lymphography of the craniocervical and thoracic l ymph nodes in nor mal dogs. Acad Radiol 1994;1:377–84. 198. Hyafil F , Cor nily JC, F eig JE, et al. Nonin vasive detection of macrophages using a nanoparticulate contrast agent for computed tomography. Nat Med 2007;13:636–41. 199. Rabin O , Manuel P erez J , Grimm J , et al. An X-ra y computed tomography imaging agent based on long-circulating bismuth sulphide nanoparticles. Nat Mater 2006;5:118–22. 200. Li H, Gray BD, Corbin I, et al. MR and fluorescent imaging of lo wdensity lipoprotein receptors. Acad Radiol 2004;11:1251–9. 201. Dirksen A, Langereis S, de Waal BF, et al. Design and synthesis of a bimodal target-specific contrast agent for angio genesis. Org Lett 2004;6:4857–60. 202. Mishra A, Pfeuf fer J , Mishra R, et al. A ne w class of Gd-based DO3A-eth ylamine-derived tar geted contrast agents for MR and optical imaging. Bioconjug Chem 2006;17:773–80. 203. Uzgiris EE, Sood A, Bove K, et al. A multimodal contrast agent for preoperative MR Imaging and intraoperative tumor margin delineation. Technol Cancer Res Treat 2006;5:301–9. 204. Daldrup-Link HE, Rudelius M, Metz S, et al. Cell tracking with gadophrin-2: a bifunctional contrast agent for MR imaging, optical imaging, and fluorescence microscop y. Eur J Nucl Med Mol Imaging 2004;31:1312–21. 205. Huber MM, Staubli AB, Kustedjo K, et al. Fluorescentl y detectable magnetic resonance imaging agents. Bioconjug Chem 1998; 9:242–9. 206. Brekke C, Mor gan SC, Lo we AS, et al. The in vitro ef fects of a bimodal contrast agent on cellular functions and relaxometr y. NMR Biomed 2007;20:77–89. 207. Talanov VS, Re gino CA, K obayashi H, et al. Dendrimer -based nanoprobe for dual modality magnetic resonance and fluorescence imaging. Nano Lett 2006;6:1459–63. 208. Koyama Y, Talanov VS, Bernardo M, et al. A dendrimer-based nanosized contrast agent dual-labeled for magnetic resonance and

209. 210.

211.

212. 213.

214. 215. 216.

217.

218. 219. 220. 221.

222.

223. 224. 225. 226. 227.

228.

229.

230.

231.

232.

optical fluorescence imaging to localize the sentinel l ymph node in mice. J Magn Reson Imaging 2007;25:866–71. Frangioni JV. in vivo near-infrared fluorescence imaging. Cur r Opin Chem Biol 2003;7:626–34. Melancon MP, Wang Y, Wen X, et al. Development of a macromolecular dual-modality MR-optical imaging for sentinel l ymph node mapping. Invest Radiol 2007;42:569–78. Park JW, Benz CC, Mar tin FJ. Future directions of liposome- and immunoliposome-based cancer therapeutics. Semin Oncol 2004;31:196–205. Gabizon AA, Shmeeda H, Zalipsk y S. Pros and cons of the liposome platform in cancer drug targeting. J Liposome Res 2006;16:175–83. Mulder WJ, Strijkers GJ, Griffioen AW, et al. A liposomal system for contrast-enhanced magnetic resonance imaging of molecular targets. Bioconjug Chem 2004;15:799–806. Cai W, Chen X. Anti-angiogenic cancer therap y based on inte grin αvβ3 antagonism. Anticancer Agents Med Chem 2006;6:407–28. Cai W, Gambhir SS, Chen X. Multimodality tumor imaging tar geting integrin αvβ3. Biotechniques 2005;39:S6–S17. Mulder WJ, Strijkers GJ, Habets JW, et al. MR molecular imaging and fluorescence microscopy for identification of activated tumor endothelium using a bimodal lipidic nanopar ticle. F ASEB J 2005;19:2008–10. Brandwijk RJ, Mulder WJ, Nicolay K, et al. Anginex-conjugated liposomes for targeting of angiogenic endothelial cells. Bioconjug Chem 2007;18:785–90. Cai W, Chen X. Nanoplatfor ms for tar geted molecular imaging in living subjects. Small 2007;3:1840–54. Ferrari M. Cancer nanotechnolo gy: oppor tunities and challenges. Nat Rev Cancer 2005;5:161–71. Sahoo SK, Parveen S, Panda JJ. The present and future of nanotechnology in human health care. Nanomedicine 2007;3:20–31. Horton MA, Khan A. Medical nanotechnolo gy in the UK: a perspective from the London centre for nanotechnolo gy. Nanomedicine 2006;2:42–8. Kawasaki ES, Pla yer A. Nanotechnolo gy, nanomedicine, and the development of new, effective therapies for cancer. Nanomedicine 2005;1:101–9. Thayer AM. Building up nanotech research. Chem Eng Ne ws 2007;85:15–21. Bruchez M Jr, Moronne M, Gin P, et al. Semiconductor nanocrystals as fluorescent biological labels. Science 1998;281:2013–6. Chan WC, Nie S. Quantum dot bioconjugates for ultrasensitive nonisotopic detection. Science 1998;281:2016–8. Balasubramanian K, Burghard M. Chemically functionalized carbon nanotubes. Small 2005;1:180–92. Lacerda L, Bianco A, Prato M, K ostarelos K. Carbon nanotubes as nanomedicines: from to xicology to phar macology. Adv Dr ug Deliv Rev 2006;58:1460–70. Grodzinski P, Silv er M, Molnar LK. Nanotechnolo gy for cancer diagnostics: promises and challenges. Exper t Re v Mol Diagn 2006;6:307–18. Wang D, He J , Rosens weig N , Rosenzw eig Z. Super paramagnetic Fe2O3 beads-CdSe/ZnS quantum dots core-shell nanocomposite particles for cell separation. Nano Lett 2004;4:409–13. Gu H, Zheng R, Zhang X, Xu B . F acile one-pot synthesis of bifunctional heterodimers of nanopar ticles: a conjugate of quantum dot and magnetic nanopar ticles. J Am Chem Soc 2004;126:5664–5. Selvan ST , P atra PK, Ang CY , Ying JY . Synthesis of silica-coated semiconductor and magnetic quantum dots and their use in the imaging of li ve cells. Angew Chem Int Ed Engl 2007;46:2448–52. Bakalova R, Zhele v Z, Aoki I, et al. Silica-shelled single quantum dot micelles as imaging probes with dual or multimodality . Anal Chem 2006;78:5925–32.

Multimodality Agents

233. Wang S, Jarrett BR, Kauzlarich SM, Louie AY. Core/shell quantum dots with high relaxivity and photoluminescence for multimodality imaging. J Am Chem Soc 2007;129:3848–56. 234. van Tilborg GA, Mulder WJ, Chin PT , et al. Annexin A5-conjugated quantum dots with a paramagnetic lipidic coating for the multimodal detection of apoptotic cells. Bioconjug Chem 2006;17:865–8. 235. Mulder WJ, Koole R, Brandwijk RJ, et al. Quantum dots with a paramagnetic coating as a bimodal molecular imaging probe. Nano Lett 2006;6:1–6. 236. Prinzen L, Miserus RJ, Dirksen A, et al. Optical and magnetic resonance imaging of cell death and platelet acti vation using annexin a5-functionalized quantum dots. Nano Lett 2007;7:93–100. 237. Pradhan N, Battaglia DM, Liu Y, Peng X. Efficient, stable, small, and water-soluble doped ZnSe nanocr ystal emitters as non-cadmium biomedical labels. Nano Lett 2007;7:312–7. 238. Kim SW, Zimmer JP , Ohnishi S, et al. Engineering InA SxP1-x/ InP/ZnSe III-V allo yed core/shell quantum dots for the near infrared. J Am Chem Soc 2005;127:10526–32. 239. Pradhan N , P eng X. Ef ficient and color -tunable Mn-doped ZnSe nanocrystal emitters: control of optical perfor mance via g reener synthetic chemistry. J Am Chem Soc 2007;129:3339–47. 240. Anderson EA, Isaacman S, Peabody DS, et al. Viral nanoparticles donning a paramagnetic coat: conjugation of MRI contrast agents to the MS2 capsid. Nano Lett 2006;6:1160–4. 241. Lisy MR, Har tung A, Lang C, et al. Fluorescent bacterial magnetic nanoparticles as bimodal contrast agents. In vest Radiol 2007;42:235–41. 242. Bridot JL, F aure AC, Laurent S, et al. Hybrid gadolinium o xide nanoparticles: multimodal contrast agents for in vi vo imaging. J Am Chem Soc 2007;129:5076–84. 243. Yang W, Barth RF, Adams DM, Soloway AH. Intratumoral deli very of boronated epidermal growth factor for neutron capture therapy of brain tumors. Cancer Res 1997;57:4333–9. 244. El-Andaloussi S, Holm T, Langel U. Cell-penetrating peptides: mechanisms and applications. Curr Pharm Des 2005;11:3597–611. 245. Nitin N, LaConte LE, Zurkiya O , et al. Functionalization and peptide-based delivery of magnetic nanopar ticles as an intracellular MRI contrast agent. J Biol Inor g Chem 2004;9:706–12. 246. Veiseh O , Sun C, Gunn J , et al. Optical and MRI multifunctional nanoprobe for targeting gliomas. Nano Lett 2005;5:1003–8. 247. Martina MS, Fortin JP, Fournier L, et al. Magnetic tar geting of rhodamine-labeled superparamagnetic liposomes to solid tumors: in vivo tracking b y f ibered confocal fluorescence microscop y. Mol Imaging 2007;6:140–6. 248. Lee JH, Jun YW, Yeon SI, et al. Dual-mode nanopar ticle probes for high-performance magnetic resonance and fluorescence imaging of neuroblastoma. Angew Chem Int Ed Engl 2006;45:8160–2. 249. Choi JH, Nguy en FT , Barone PW , et al. Multimodal biomedical imaging with asymmetric single-w alled carbon nanotube/iron oxide nanoparticle complexes. Nano Lett 2007;7:861–7. 250. Josephson L, Kircher MF, Mahmood U, et al. Near-infrared fluorescent nanoparticles as combined MR/optical imaging probes. Bioconjug Chem 2002;13:554–60. 251. Sosnovik DE, Nahrendorf M, Deliolanis N , et al. Fluorescence tomography and magnetic resonance imaging of m yocardial macrophage infiltration in infarcted myocardium in vivo. Circulation 2007;115:1384–91. 252. Jaffer FA, Nahrendorf M, Sosno vik D , et al. Cellular imaging of inflammation in atherosclerosis using magnetofluorescent nanomaterials. Mol Imaging 2006;5:85–92. 253. Kircher MF, Weissleder R, Josephson L. A dual fluorochrome probe for imaging proteases. Bioconjug Chem 2004;15:242–8. 254. Tsourkas A, Shinde-Patil VR, Kelly KA, et al. In vi vo imaging of activated endothelium using an anti-VCAM-1 magnetooptical probe. Bioconjug Chem 2005;16:576–81.

469

255. Funovics M, Montet X, Re ynolds F, et al. Nanopar ticles for the optical imaging of tumor E-selectin. Neoplasia 2005;7:904–11. 256. Kelly KA, Allport JR, Tsourkas A, et al. Detection of vascular adhesion molecule-1 e xpression using a no vel multimodal nanopar ticle. Circ Res 2005;96:327–36. 257. Nahrendorf M, Jaffer FA, Kelly KA, et al. Noninvasive vascular cell adhesion molecule-1 imaging identif ies inflammator y activation of cells in atherosclerosis. Circulation 2006;114:1504–11. 258. Montet X, Weissleder R, Josephson L. Imaging pancreatic cancer with a peptide-nanopar ticle conjugate tar geted to nor mal pancreas. Bioconjug Chem 2006;17:905–11. 259. Sosnovik DE, Schellenber ger EA, Nahrendorf M, et al. Magnetic resonance imaging of cardiomyocyte apoptosis with a novel magneto-optical nanoparticle. Magn Reson Med 2005;54:718–24. 260. Aikawa E, Nahrendorf M, Sosnovik D, et al. Multimodality molecular imaging identifies proteolytic and osteogenic activities in early aortic valve disease. Circulation 2007;115:377–86. 261. Fire A, Xu S, Montgomery MK, et al. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 1998;391:806–11. 262. Stevenson M. Therapeutic potential of RN A interference. N Engl J Med 2004;351:1772–7. 263. Mello CC, Conte D Jr . Re vealing the w orld of RN A interference. Nature 2004;431:338–42. 264. Medarova Z, Pham W, Farrar C, et al. In vivo imaging of siRNA delivery and silencing in tumors. Nat Med 2007;13:372–7. 265. Mae M, Langel U . Cell-penetrating peptides as v ectors for peptide, protein and oligonucleotide deli very. Cur r Opin Phar macol 2006;6:509–14. 266. Sosnovik DE, Weissleder R. Emer ging concepts in molecular MRI. Curr Opin Biotechnol 2007;18:4–10. 267. Gambhir SS. Molecular imaging of cancer with positron emission tomography. Nat Rev Cancer 2002;2:683–93. 268. Pandey SK, Gr yshuk AL, Sajjad M, et al. Multimodality agents for tumor imaging (PET, fluorescence) and photodynamic therapy. A possible “see and treat” approach. JMed Chem 2005;48:6286–95. 269. Kirchner C, Liedl T, Kudera S, et al. Cytotoxicity of colloidal CdSe and CdSe/ZnS nanoparticles. Nano Lett 2005;5:331–8. 270. Derfus AM, Chan WCW, Bhatia SN. Probing the cytotoxicity of semiconductor quantum dots. Nano Lett 2004;4:11–8. 271. Keren S, Zavaleta C, Cheng Z, et al. Noninvasive molecular imaging of small li ving subjects using Raman spectroscop y. Proc Natl Acad Sci U S A 2008;105:5844–9. 272. Liu Z, Davis C, Cai W, et al. Circulation and long-term fate of functionalized, biocompatible single-walled carbon nanotubes in mice probed b y Raman spectroscop y. Proc Natl Acad Sci USA 2008;105:1410–5. 273. Schipper ML, Naka yama-Ratchford N, Davis CR, et al. A pilot toxicology study of single-w alled carbon nanotubes in a small sample of mice. Nat Nanotechnol 2008;3:216–21. 274. Singh R, P antarotto D, Lacerda L, et al. Tissue biodistribution and blood clearance rates of intra venously administered carbon nanotube radiotracers. Proc Natl Acad Sci USA 2006; 103:3357–62. 275. McDevitt MR, Chattopadh yay D, Kappel BJ , et al. Tumor targeting with antibody-functionalized , radiolabeled carbon nanotubes. J Nucl Med 2007;48:1180–9. 276. Caliceti P, Veronese FM. Phar macokinetic and biodistribution properties of pol y(ethylene gl ycol)-protein conjugates. Adv Dr ug Deliv Rev 2003;55:1261–77. 277. Wu AM, Senter PD. Arming antibodies: prospects and challenges for immunoconjugates. Nat Biotechnol 2005;23:1137–46. 278. Soo Choi H, Liu W, Misra P, et al. Renal clearance of quantum dots. Nat Biotechnol 2007;25:1165–70. 279. Houston JP, Ke S, Wang W, et al. Quality anal ysis of in vi vo nearinfrared fluorescence and con ventional gamma images acquired

470

280.

281.

282. 283.

284.

285. 286.

287. 288. 289.

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

using a dual-labeled tumor -targeting probe. J Biomed Opt 2005;10:054010. Li C, Wang W, Wu Q, et al. Dual optical and nuclear imaging in human melanoma x enografts using a single tar geted imaging probe. Nucl Med Biol 2006;33:349–58. Wang W, Ke S, Kwon S, et al. A new optical and nuclear dual-labeled imaging agent targeting interleukin 11 receptor alpha-chain. Bioconjug Chem 2007;18:397–402. Arap W, K olonin MG, Trepel M, et al. Steps to ward mapping the human vasculature by phage display. Nat Med 2002;8:121–7. Zurita AJ, Troncoso P, Cardo-Vila M, et al. Combinatorial screenings in patients: the interleukin-11 receptor alpha as a candidate tar get in the pro gression of human prostate cancer . Cancer Res 2004;64:435–9. Sampath L, Kw on S, K e S, et al. Dual-labeled trastuzumab-based imaging agent for the detection of human epider mal growth factor receptor 2 o verexpression in breast cancer . J Nucl Med 2007;48:1501–10. Cai W, Niu G, Chen X. Multimodality imaging of the HER-kinase axis in cancer. Eur J Nucl Med Mol Imaging 2007;35:186–208. Niu G, Cai W, Chen X. Molecular imaging of human epider mal growth f actor receptor 2 (HER-2) e xpression. F ront Biosci 2008;13:790–805. Harries M, Smith I. The development and clinical use of trastuzumab (Herceptin). Endocr Relat Cancer 2002;9:75–85. Townsend DW, Beyer T. A combined PET/CT scanner: the path to tr ue image fusion. Br J Radiol 2002;75 Spec No:S24–30. Ruf J, Lopez Hanninen E, Oettle H, et al. Detection of recurrent pancreatic cancer: comparison of FDG-PET with CT/MRI. P ancreatology 2005;5:266–72.

290. Pannu HK, Cohade C, Bristo w RE, et al. PET -CT detection of abdominal recur rence of o varian cancer: radiolo gic-surgical correlation. Abdom Imaging 2004;29:398–403. 291. Judenhofer MS, Wehrl HF, Ne wport DF, et al. Simultaneous PET MRI: a new approach for functional and mor phological imaging. Nat Med 2008;14:459–65. 292. Soret M, Bacharach SL, Buvat I. Partial-volume effect in PET tumor imaging. J Nucl Med 2007;48:932–45. 293. Shao Y, Cher ry SR, F arahani K, et al. Simultaneous PET and MR imaging. Phys Med Biol 1997;42:1965–70. 294. Lee HY, Li Z, Chen K, et al. PET/MRI dual-modality tumor imaging using ar ginine-glycine-aspartic (RGD)-conjugated radiolabeled iron oxide nanoparticles. J Nucl Med 2008;49:1371–9. 295. Nahrendorf M, Zhang H, Hembrador S, et al. Nanopar ticle PET-CT imaging of macrophages in inflammator y atherosclerosis. Circulation 2008;117:379–87. 296. Zielhuis SW, Seppenw oolde JH, Mateus VA, et al. Lanthanideloaded liposomes for multimodality imaging and therap y. Cancer Biother Radiopharm 2006;21:520–7. 297. Kobayashi H, Hama Y, K oyama Y, et al. Simultaneous multicolor imaging of f ive dif ferent l ymphatic basins using quantum dots. Nano Lett 2007;7:1711–6. 298. Smith BR, Cheng Z, De A, et al. Real-time intra vital imaging of RGD-quantum dot binding to luminal endothelium in mouse tumor neovasculature. Nano Lett 2008;8:2599–606.

30 “CLICK CHEMISTRY”: APPLICATIONS TO MOLECULAR IMAGING NEAL K. DEVARAJ, PHD, AND RALPH WEISSLEDER, MD, PHD

Advancing the state-of-the-ar t molecular imaging agents will lik ely require ne w chemistries. Specif ically, because imaging agents are often for med by the coupling of molecular entities with distinct but complementar y functionality, the use of more sophisticated and reliab le co valent coupling reactions is desirab le. Recently, a class of reactions has gained tremendous attention in the chemistr y community under the moniker of “click chemistry,” a concept introduced b y K olb and colleagues. 1 Of the man y reactions that f all under the cate gory of “click” reactions, the copper-catalyzed azide-alkyne cycloaddition (CuAAC) has become the paradigm. This reaction, which was independently reported by both Rostovtsev and colleagues2 and Tornoe and colleagues 3 in 2002, has quickl y become one of the most reliab le coupling reactions in chemistr y and has found numerous applications in nearl y e very f ield related to chemistr y from materials science to chemical biology. The pur pose of this chapter is to re view “click chemistry,” specifically CuAACs and relevant applications to molecular imaging. We will give background to the pertinent chemistr y dra wing attention to features that are unique and should be of par ticular interest to those interested in appl ying the reaction to the design of molecular imaging agents. We will then re view the literature of ho w “click chemistry” has been applied to the constr uction of specific molecular imaging agents as w ell as applications that are of interest to the molecular imaging community , and we believe will gain further attention in the future.

CLICK CHEMISTRY Though typicall y confused with a single reaction (CuAAC), “click chemistr y” refers to a philosoph y of how to do chemistry. Kolb and colleagues reasoned that

the cur rent methods in dr ug disco very that relied on laborious chemistr y to create compounds with numerous carbon–carbon bonds and comple x structures were not the most ef ficient manner to disco ver ne w dr ugs. Indeed, the function of dr ugs, not the chemical str ucture, is most impor tant. With this desire of function over structure, to accelerate the process of drug discovery, they noted that “all searches must be restricted to molecules that are easy to mak e.”1 From this requirement, came the idea of “click chemistr y,” a class of efficient and selecti ve reactions that could be used to stitch together function in a f acile manner. The “click” moniker is meant to signify that with the use of these methods, joining molecular pieces is as easy as “clicking” together the tw o pieces of a buckle. 4 The buckle works no matter what is attached to it, as long as its two pieces can reach each other . The components of the buckle can make a connection only with each other. Reactions that f it in this cate gory must fulf ill a number of stringent requirements. The reaction has to produce high yields, generate inof fensive b yproducts that can be easily removed, the reaction conditions have to be simple and insensiti ve to both air and w ater, the starting materials should be readily available, purif ication must be b y a nonchromato graphic method , and finally, the product should be stable under physiological conditions. Numerous reactions w ere identif ied to fulfill these criteria, including 1,3-dipolar c ylcoadditions (of w hich the CuAA C f alls under), Diels–Alder reactions, nucleophilic substitution chemistr y such as ring-opening chemistry of epoxides, carbonyl chemistry not falling under “aldol” chemistry such as thiourea formation, and o xidative additions to carbon–carbon multiple bonds such as epoxidations. 471

472

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

In the initial introduction of “click chemistr y,” the uncatalyzed 1,3 dipolar c ycloaddition between an azide and alkyne was introduced and highlighted as the “cream of the crop” of concer ted click reactions (F igure 1A). This reaction has been known for a long time, first introduced b y A. Michael in 1893 and then studied e xtensively b y Huisgen 5 in the later half of the tw entieth century. The reaction, lik e all “click” reactions, is thermodynamically v ery f avorable. Ho wever, the reaction suffers from two drawbacks: a high activation energy due to a kinetic bar rier and the possibility of for ming tw o products either 1,4 or 1,5 regioisomers. In 2002, the copper catal yzed v ariant of this reaction (CuAA C) w as introduced by both Rostovtsev and colleagues 2 and Tornoe and colleagues 3 (Figure 1B). This reaction occurs between an or ganic azide and a ter minal acetylene. The catalyst activation both lowered the kinetic barrier allowing the reaction to proceed at room temperature and conferred regioselectivity with the 1,4 regioisomer being the only product. Copper-catalyzed azide-alk yne c ycloaddition w as quickly deemed a “click” reaction and has rapidly found utilization in numerous f ields. The CuAA C is surel y responsible for the tremendous popularity of the “click” concept, and many simply associate “click chemistry” to mean triazole formation between an azide and alkyne. In addition to fulf illing all the original “click chemistr y” criteria, it possessed two other very important properties. First, the copper catalyst allowed the reaction to proceed at room temperature, enab ling the reaction’s use in biological settings w here mild reaction conditions are a requirement. Second, and perhaps e ven more impor tant, the reaction is e xtremely chemoselective, meaning that the azides and alkynes are primed to only react with one another and to ignore other functional g roups. This is especially tr ue under ph ysiological conditions of room temperature and aqueous conditions. There are tw o A A +

N

N

R1

–

N

R1

80 – 120 C 12 – 24 h

R2

B

R3

N

R1

N N

+ R3

R2

R3

N

N N R2

+

N N R2

–

N

R1 H

Cu(I)

R1

N

N N

0 – 25 C H

R2

Figure 1. A, Non-catalyzed azide-alkyne cycloaddition. B, Copper catalyzed azide-alkyne cycloaddition.

immediate ramif ications from this chemoselecti vity. First, side product for mation is minimized as the onl y possible products are triazoles. Second , the reaction can be conducted in comple x solutions containing a wide range of functional g roups without consequence. The azides and alkynes are essentially invisible to other functional groups present whether they are impurities or biological macromolecules. This feature of the reaction is clearly responsible for the high f idelity of the reaction.

UNIQUE PROPERTIES OF THE CUAAC What are the adv antages and disadv antages in using CuAAC? What are the proper ties of the reaction that distinguish it from alter native coupling strate gies? The properties of the CuAA C reaction that mak e the method generally appealing include its high yield , tolerance of reaction conditions (pH, solv ent, etc), for mation of a single product, and rate. Ho wever, there are a number of additional features that distinguish this reaction from the vast majority of other coupling schemes and can be considered adv antageous or disadv antageous to imaging applications. 1. Chemoselectivity. The reaction partners, an azide and alkyne, are iner t to reaction with numerous other functional groups under the typical mild reaction conditions. This can be adv antageous if one is interested in immobilizing a comple x molecule containing numerous functional g roups. Because azides and alkynes are not biolo gically present and do not react with biological functionalities, the reaction has often been described as bio-or thogonal and is v ery useful for site-specif ically coupling biomolecules to gether. However, the disadv antage is that the azide or the alkyne must be chemicall y introduced to the molecule. In other w ords, one cannot tak e adv antage of groups that are often present in nature (such as amines, acids, or thiols) because the azide and alkyne are abiotic. 2. Catalyzed b y copper (I). Though catal ysis dramatically increases the rate of the CuAA C reaction, the requirement of copper has dra wbacks. This requirement adds an extra component to the reaction: copper (I) complexes can react with oxygen generating reactive oxygen species and aqueous copper (I) can react with itself and dispropor tionate into copper (II) and copper (0). These problems can damage the sensitive functionalities that one is tr ying to immobilize and can contaminate a surf ace with elemental copper . Fortunately, the reacti vity of copper (I) with o xygen and itself can be modulated b y using commerciall y

“Click Chemistry”: Applications to Molecular Ima ging

available ligands such as tris-triazo ylbenzylamine (TBTA).6 An adv antage of copper (I) catal ysis that has been vir tually ignored is, for a wide v ariety of catalytically active complexes, the copper redox state can be easily switched and catalysis can be turned on and off.7 Coupling catalysts that can be turned on and off via a one electron change are rare, and rarer still are catalysts that switch at mild potentials in aqueous conditions. 3. The resulting 1,2,3 triazole is unique in that the link is an aromatic heteroc ycle. This has tw o impor tant qualities. The link has superior stability to linkages resulting from other couplings, being resistant to both hydrolysis and redox reactions. In addition, the triazole pro vides good electronic coupling betw een organic molecules.8 Thus, for applications where one wants to electronicall y couple tw o molecules, this link may be appropriate.

CONJUGATION CHEMISTRY By f ar, the most common application of “click chemistry” to imaging has been to couple functional molecules to known imaging agents to generate an imaging agent with specif ic proper ties. In these applications, the click reaction is simply used as a coupling reaction. There have been numerous applications of coupling fluorophores using “click chemistr y.” One of the earliest demonstrations involved the bioconjugation of fluorescein deri vatives to an alk yne- or azide-modif ied co wpea mosaic virus.9 The virus consists of 60 identical protein subunits that were each modified with either an azide or analkyne. Despite the demanding requirement of 60 reactions per virus particle, under appropriate conditions, fully derivatized virus could be recovered in 96% yield. Conjugation of fluorophores b y “click chemistr y” has also been applied to acti vity-based protein prof iling.10 This technique uses acti ve site–directed probes with broad tar get selecti vity to visualize the acti vity of several proteins simultaneousl y. P ast methods relied on homogenization of the cell prior to labeling, a requirement because of the bulk and char ge on fluorophore tags that are unab le to per meate throughout the cell in a homogenous fashion. This is not ideal as homogenization removes proteins from their nati ve en vironments and potentially alters their acti vity. To solv e this prob lem, smaller azide containing probes w ere used to label protein acti vity in li ve cells and li ve mice. After labeling in the native environment, the cells or animals could be sacrificed and proteins visualized b y CuAAC to alk ynemodified rhodamine.

473

Copper-catalyzed “click” couplings ha ve been applied to creating radiolabeled imaging agents. A “click” ready 18F-azide labeling agent 2-[ 18F]Fluoro ethylazide was introduced as a mean to increase the fle xibility of an 18F where it can be incor porated on biomolecules and allow labeling in the presence of other reactive functional groups.11 Due to the shor t half-life of 18F, the group screened numerous acetylenes and reaction conditions and found that the reaction could be completed in 15 minutes by either using a reacti ve acetylene such as a propynamide or heating the solution at 80 °C. The fluorine label w as coupled to a prop ynamide modif ied peptide in 15 minutes with 92% radiochemical yield. A clever method to introduce technetium-99m onto biomolecules involves “clicking” either L-propargyl glycine or L-azide alanine to an appropriately functionalized biomolecule.12 The amine of the amino acid and the resulting triazole for m a chelate that strongl y binds the radioactive metal. In this fashion, the authors were able to rapidly create technetium-99m probes of a tumor af fine peptide, sugars, nucleotides, and phospholipids. This is an interesting example as the triazole serves to conjugate two different functionalities and act as a ligand for a metal. Magnetic Resonance Imaging (MRI) probes ha ve been conjugated using “click chemistr y.” More recentl y, Sun and colleagues 13 have modif ied dextran-coated iron oxide nanopar ticles with both azides and alk ynes and shown that these par ticles are amenab le to modif ication via CuAAC. A general method for using CuAAC to functionalize iron o xide nanopar ticles w as sho wn b y White and colleagues 14 who decorated them with both azide organo-phosphates and alk yne carbo xylates. Lin and colleagues15 later e xtended on this w ork b y attaching azide- or alk yne-terminated silanes to magnetic iron oxide nanopar ticles. Azide-terminated nanopar ticles were functionalized using CuAA C with a v ariety of alkynes, including proteins containing alk ynes that were site specifically introduced to the C terminus. Commonly used ligand 1,4,7,10-tetraazadodecaneN,N,N,Ntetraacetic acid (DOTA) was “clicked” to four cyclic peptides, complexed to 111In and used to image αVβ3 integrin expressing tumors in vivo.16 Several groups have reported using triazole formation to successfully modify magnetic nanoparticles, which can be used as magnetic resonance contrast agents. Semiconducting nanocrystals are also being investigated for in vi vo imaging applications for their high quantum yields and photostability . Both ther mal cycloaddition and CuAA C have been used to functionalize fluorescent cadmium selenide nanopar ticles.17 However, in this w ork, the authors conclude that the

474

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

thermal method is superior because the luminescence of the dots is ir reversibly quenched after e xposure to copper. We believe that other copper -free “click” techniques will be amenab le or a copper catal yst that does not react with the quantum dot core will e xpand the applications of “click chemistr y” to quantum dots.

FLUOROGENIC CLICK CHEMISTRY For in vi vo applications, fluoro genic probes of fer a method to lo wer the backg round fluorescence, allo wing for more sensiti ve detection. A number of g roups ha ve created conjugation schemes that tur n-on fluorescence upon “click” reaction. Three basic strate gies ha ve been used. Perhaps the most straightforward is the conjugation of a FRET pair via “click” reaction.18,19 This is a versatile strategy b ut is dif ficult to adapt to a bioconjugation scheme due to the need of introducing one of the two fluorophores onto the biomolecule to be labeled prior to conjugation. A clever strategy is to introduce an azide in a position on a fluorophore so that it acts as an ef ficient quencher, presumab ly via photoinduced electron transfer.20 Upon triazole for mation, the quencher is remo ved and fluorescence tur ns on. This has been applied to coumarins with azides directly conjugated on the position 3 and to anthracenes, w here the azide is separated from the fluorophore core by a methylene spacer. Fluorogenic coumarins ha ve been used to label ne wly synthesized proteins in cells by tagging an acetylene containing amino acid, which competes with methionine for incorporation into proteins. 21 A drawback of this technique is that azides, particularly aromatic azides, are reacti ve and can degrade with time, causing the fluorophore to turn on and increasing the backg round signal. A technique that removes this risk involves attaching an acetylene directly to the fluorophore core. Zhou and F ahrni22 created a water-soluble coumarin modif ied b y acetylene in the position 7. In this state, the coumarin is nearly completely nonemissive. However, after “click” reaction, the electron donating proper ties of the for med triazole ring tur ns increases the quantum yield of the electronically coupled coumarin by approximately 18 times.

the ability to rapidl y create strong-binding enzyme inhibitors. Although “click chemistr y” has most often been used as a coupling reaction, there has been signif icant work done showing that the reaction can be used to develop novel enzyme inhibitors. Because “click chemistry” w as introduced as a w ay to rapidl y create molecules with biolo gical functionality , it should not be surprising that man y g roups ha ve sho wn that “click chemistry” can be used to create highly selective enzyme inhibitors. There have been several strategies used for this endeavor. P erhaps, the most ele gant w as introduced b y Manetsch and colleagues 23 and is ter med in situ click chemistry. This method is a for m of tar get guided synthesis, in this case, the tar get being the acti ve site of the enzyme of interest. The enzyme is e xposed to pairs of azides and alkynes with different functionality and, ideally, already known weak ligands of the enzyme of interest.24 When two ligands come together at the enzyme, the azide and the alk yne are forced to be in close pro ximity to one another (F igure 2). This lowers the kinetic bar rier to the irreversible “click” reaction, and the two weak ligands are joined together, creating a molecule that cooperatively binds to the enzyme in not one but tw o locations. Because of the cooperati vity, these molecules are much stronger inhibitors than the starting molecules. One of the fascinating aspects of this technique is that the enzyme actually catal yzes the for mation of its o wn inhibitor . Azides and alk ynes in the cocktail that are not suited to form an inhibitor do not fuse because triazole products

ENZYME INHIBITORS One of the main goals in molecular imaging is to create imaging agents that tar get specif ic proteins associated with disease. A straightforw ard strate gy for imaging biomarkers is to ha ve a strong-binding enzyme inhibitor and attach the inhibitor to an imaging agent.This requires

Figure 2. Crystal structure of an in situ formed inhibitor bound to the active site of acetylcholinesterase. The white arrow points to the triazole that is formed between the azide and alkyne. Courtesy of Zoran Radic, K. B. Sharpless, and Palmer Taylor.

“Click Chemistry”: Applications to Molecular Ima ging

from the uncatal yzed reaction are ne gligible. Screening for potential inhibitors is done b y mass spectrometr y of the mixture and searching for product for mation. This screening method has the added benef it that mixtures of acetylenes or azides can be screened simultaneousl y for the ability to form inhibitors. This has two advantages: it lowers the amount of enzyme required and increases the throughput. Another impro vement that can be implemented is the use of microfluidics to fur ther reduce the quantity of required enzyme and increase throughput. 25 Using the in situ method, Manetsch and colleagues 23 have disco vered the most potent inhibitors for acetylcholinesterase to date, including the one with a dissociation constant of 33 femtomolar . Although initial w ork focused on the fusion of tw o already kno wn w eak inhibitors, this has been e xtended b y taking kno wn binders and screening their ability to “click” with building b locks that ha ve an unkno wn af finity with the enzyme.26 This produces unique inhibitors that contain functional groups that were previously unknown to interact with the enzyme in an y specif ic manner. In addition to acetylcholinesterase, this strategy has also been used to create potent inhibitors for HIV -1 protease and carbonic anhydrase II.27–29 Although allowing an enzyme to create its o wn inhibitor is ideal, other g roups ha ve chosen to screen small libraries of triazole compounds instead. This

475

has also produced a lar ge number of e xcellent triazole enzyme inhibitors.29–31 A rational design strate gy that is commonl y used to create biolo gically acti ve molecules via “click chemistry” is to use the triazole as either an amide bond or a heterocycle isostere. Man y g roups have assembled triazoles by taking kno wn small molecules with biolo gical activity and by substituting in a triazole.32–38 Oftentimes, these triazoles possess similar biolo gical acti vity. This strategy produces molecules that are syntheticall y less challenging to create and , with appropriate libraries of azides and alk ynes, ma y allo w screening of a lar ge number of analo gs to the parent compound in a highthroughput manner. The ability of the triazole to mimic the str ucture and biolo gical function of an amide is particularly striking. Comparing the crystal structures of HIV protease bound to the amide inhibitor amprena vir and triazole analo gs sho wed that the enzyme interacts with the triazole in a manner nearl y identical to the amide. Triazoles possess both a h ydrogen donor and an acceptor moiety . Another striking e xample is the triazole-based peptidomic of a tyrosine kinase inhibitor .39 The nati ve inhibitor is a c yclic peptide containing tw o peptide bonds. Replacing both peptide bonds with triazoles yields a molecule with a nearly identical disassociation constant. Recentl y, a study w as conducted on the

B

A

O N N

NH

OMe

OMe OMe

N

OMe

Figure 3. Triazoles can mimic the function of amides. The synthetic capsaicin nonivamide (A) contains an amide bond that has previously been shown essential to its activity. Replacement of the amide bond by a triazole via CuAAC (B) creates a molecule with similar biological activity.

476

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

effect of a peptide-triazole point mutation on relati vely simple molecules w hose amide linkage w as critical for biological acitivity.40 As a class of molecules, the investigators chose capsaicinoids, a v anillyl g roup link ed to an aliphatic chain by an amide. The amide group of these molecules is essential for receptor reco gnition; pre viously, only the thiourea g roup was known as a suitab le isostere. Analogs of biolo gically acti ve capsaicinoids with a 1,4 triazole replacing the amide w ere found to be potent acti vators of capsaicinoid receptors (F igure 3). This is, perhaps, the most con vincing experiment of the ability of the triazole ring to mimic the biolo gical function of the amide bond. In this same w ork, the authors also attempted to mimic the function of doub le bonds with 1,4 and 1,5 triazoles but concluded that the higher polarity of the triazole ring had a detrimental ef fect on bioactivity.

LABELING OF BIOLOGICAL MACROMOLECULES The bio-orthogonality of “click chemistry” makes it nearly ideal for labeling in biolo gical systems. One typicall y incorporates an azide or alk yne in a probe that labels the living system. Because free copper is to xic to most li ving cells (see below), CuAAC is used in either fixed or homogenized cell extracts and then reacted with an imaging agent conjugated to an alkyne or azide for visualization. A popular strate gy is to tag cell components b y exposing cells to azide or acetylene containing metabolites. These compounds can often then be incor porated into macromolecular str uctures by slightly promiscuous pre-existing cellular machiner y. F or instance, in metabolic oligosaccharide engineering, cells are g rown in the presence of analo gs of sugar precursors containing groups such as azides. The cells then metabolize and incorporate these sugars nonspecif ically into the cell structure as if they were the native sugar. Though the use of azide sugars w as initiall y introduced to react with phosphines via Staudinger ligation, these azide sugars were also reacti ve via CuAA C. This method has been used to image the gl ycosylation of proteins using a v ariety of azide or acetylene containing saccharide probes, which are then reacted with fluorophores (Figure 4A).41,42 Amino acids can also be visualized by bio-orthogonal noncanonical amino acid tagging.21,43 In this process, cells are e xposed to a noncanonical amino acid , such as homopropargylglycine, w hich contains an acetylene group (F igure 4B). The amino acid is incor porated into proteins in lieu of methionine thanks to the promiscuity of

the methionyl-tRNA synthetase. After labeling, cells are washed, f ixed, and labeled with a fluorophore through CuAAC. This technique has been r un in a pulse-labeling manner to visualize ne wly synthesized proteins in mammalian cells. Modification of DN A via “click chemistr y” has also been sho wn using acetylene-modif ied th ymidine analogs (5-eth ynyl-2-deoxyuridine; F igure 4C). 44 These nucleotide analogs are incorporated into DNA and can be labeled with dy e-azide csonjugates for detection. This technique has been in vestigated as a method to monitor cell proliferation superior to using radioacti ve thymidine or 5-bromo-2-deoxyuridine. While the pre vious methods nonspecif ically label biomolecules with azides, Deiters and colleagues 45 and Deiters and Schultz 46 had sho wn that azide- and alkyne-modified amino acids can be site-specif ically incorporated into proteins and then later modif ied through CuAA C. This po werful method uses e volved tRNA-synthetase pairs that are ab le to incor porate unnatural amino acids in response to the amber TAG codon. Both azide- and acetylene-containing unnatural amino acids ha ve been incor porated site specif ically into proteins and then labeled via CuAA C with molecules such as fluorescent tags. The technique has been shown with yeast cells and f ilamentous phage. 47 An operationall y simpler method for site-specif ic labeling via CuAA C can be accomplished b y taking advantage of the alkylation activity of yeast farnesyl transferases.48,49 The alkylation takes place specif ically at proteins with a C-terminus sequence of CAAX (where A is an aliphatic residue and X is typically either serine or methionine). The cysteine residue of this sequence is alk ylated. The enzymes native subtrate is farnesyl disphosphate, but it can also recognize and attach analogs that contain either azides or alk ynes. By engineering the CAAX motif into the C ter minus of proteins, one can specif ically label a protein of interest and then label this protein through CuAAC.

COPPER-FREE “CLICK CHEMISTRY” IN LIVING SYSTEMS Because of the introduction of “click chemistry,” chemical biolo gists ha ve been interested in using the chemistry to label molecules in living cells and animals. There has been a need for a f aster method to covalently attach molecules in a chemoselecti ve manner. Numerous azide-bearing sugars, lipids, and amino acids ha ve shown to be ab le to incor porate themselv es into

“Click Chemistry”: Applications to Molecular Ima ging

A

477

O

O OAc

OAc N3

HN

HN O AcO AcO

AcO AcO

OAc

O OAc

B

H 2N

H2N

COOH

COOH

C O

HN

HO

O

N O

H

H

H

H OH

H

Figure 4. Azide and acetylene containing metabolite analogs that can be nonspecifically incorporated into biological macromolecules. A, Cell-permeable mannosamine derivative. B, Amino acids homopropargylglycine and ethynylphenylalanine. C, An acetylene containing thymidine analog 5-ethynyl-2'-deoxyuridine.

living molecular architectures. Fur thermore, the bioorthogonality of reaction between an azide and alkyne is one of the essential proper ties of any selective labeling method. Ho wever, the main challenge to using “click chemistry” in living systems is the requirement of copper (I) to catalyze the cycloaddition. Although copper is an essential metal (it is at the acti ve site of numerous essential enzymes), free copper ions are to xic to cells and living systems. Indeed, there is an e xtensive literature on the creation of c ytotoxic copper comple xes as well as DN A-cleaving copper comple xes. Much of this activity is due to the redo x properties of copper and its ability to catal yze the reduction of o xygen-forming reactive oxygen species. For this reason, the body carefully regulates copper once it enters the body.

To circumvent this prob lem, Bertozzi and coworkers noted that strained inter nal acetylenes can also spontaneously react via c ycloaddition with or ganic azides.50 This is due to the ring strain that is released upon “click” reaction. To this end , the y created a fluorophore-tagged cyclooctyne derivative and labeled azide-modified sugars on the surfaces of cells with this copper free strain promoted reaction.51 This initial label showed the feasibility of the idea. Ho wever, simpl y straining the alk yne did not result in a rate that w as comparable to the copper-catalyzed version of the reaction, and the reaction proceeded sluggishl y. To fur ther increase the reacti vity of the c yclooctyne, the y added fluorine atoms adjacent to the inter nal acetylene (Figure 5). 52 It had been sho wn in the past that the

478

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

F F

HOOC O

Figure 5. Left: a difluorinated cyclooctyne (DIFO) that is used to perform copper-free “click chemistry” in living cells and in animals. The acid can be conjugated to fluorophores such as Alexa-Fluor 488. Right: Time-lapse imaging of glycan trafficking using an Alexa Fluor 488 derivative of DIFO. A–H, Chinese hamster ovary cells were incubated with 100 µM N-azidoacetylmannosamine (A–D) or 100 µM peracetylated N-acetylmannosamine as a negative control (E–H) for 3 days and subsequently labeled with 100 µM DIFO-488 at 37°C for 1 min. I–N, Time-lapse imaging of a single cell from the previous experiment over 1 h at 25°C (I–M, N-azidoacetylmannosamine; N, peracetylated N-acetylmannosamine). Reprinted with permission from Baskin JM et al.52

addition of electrone gative fluorines adjacent to acetylenes increased the rate of cycloaddition. This new cyclooctyne that coupled strain and adjacent fluorines works e xtremely w ell, and in a comparati ve e xperiment, Bertozzi has shown that the rate of cycloaddition is comparab le to the copper -catalyzed v ersion of the reaction. Using Alexa Fluor modif ied v ersions of this cyclooctyne, the y ha ve modif ied azido functionalized sugars on the surf ace of Chinese hamster o vary cells in approximately 1 minute and have used the technique to monitor real-time gl ycan trafficking (see F igure 5). Furthermore, the probes w ere used to perfor m in vi vo “click chemistry” in mice.

CONCLUSION Copper-catalyzed azide-alk yne c ycloaddition has become a mainstream coupling reaction and a paradigm of “click chemistr y.” Its widespread use sho ws that it has f illed a v oid that pre viously e xisted in synthetic chemistry. Recent w ork has sho wn that CuAAC is a straightforward and robust method to couple a lar ge

variety of imaging agents to other functional entities. These include azide- and alk yne-containing analogs of metabolic precursors such as monosaccharides, amino acids, and DNA nucleotides. In addition, “click chemistry” is ab le to rapidly generate potent enzyme inhibitors in a modular f ashion, and w e belie ve this will eventually give easier synthetic access to selecti ve imaging agents for a v ariety of disease biomarkers. An emerging area is perfor ming “click chemistr y” in li ve cells and animals. The development of “click” reactions that do not require copper is yet another important step in this regard. In the future, there will be fur ther novel imaging applications of “click chemistr y,” and these applications will g row more sophisticated as w e lear n more about the reaction mechanism and the requirements for catalysis.

ACKNOWLEDGMENTS The authors w ould lik e to ackno wledge the helpful discussions with K. B . Shar pless, M. G. F inn, and Hartmuth Kolb.

“Click Chemistry”: Applications to Molecular Ima ging

REFERENCES 1. Kolb HC, Finn MG, Sharpless KB. Click chemistry: diverse chemical function from a fe w good reactions. Angew Chem Int Ed 2001;40:2004–21. 2. Rostovtsev VV, Green LG, Fokin VV, Sharpless KB. A stepwise Huisgen c ycloaddition process: copper(I)-catal yzed re gioselective “ligation” of azides and ter minal alk ynes. Angew Chem Int Ed 2002;41:2596–9. 3. Tornoe CW , Christensen C, Meldal M. P eptidotriazoles on solid phase: [1,2,3]-triazoles b y regiospecific copper(I)-catalyzed 1,3dipolar cycloadditions of terminal alkynes to azides. J Or g Chem 2002;67:3057–64. 4. Finn MG, K olb HC, Shar pless KB . Click chemistr y—definition and aims. Kagaku Kogyo 2007;60:976–80. 5. Huisgen R. 1,3-dipolar c ycloaddition chemistr y. Ne w York: Wiley; 1984. 6. Chan TR, Hilg raf R, Shar pless KB, Fokin VV. Polytriazoles as copper(I)-stabilizing ligands in catalysis. Org Lett 2004;6:2853–5. 7. Devaraj NK, Dinolfo PH, Chidsey CED, Collman JP. Selective functionalization of independentl y addressed microelectrodes b y electrochemical activation and deacti vation of a coupling catalyst. J Am Chem Soc 2006;128:1794–5. 8. Devaraj NK, Decreau RA, Ebina W, et al. Rate of interf acial electron transfer through the 1,2,3-triazole linkage. J Ph ys Chem B 2006;110:15955–62. 9. Wang Q, Chan TR, Hilgraf R, et al. Bioconjugation by copper(I)-catalyzed azide-alk yne [3+2] c ycloaddition. J Am Chem Soc 2003;125:3192–3. 10. Speers AE, Adam GC, Cravatt BF. Activity-based protein prof iling in vivo using a copper(I)-catalyzed azide-alkyne [3+2] cycloaddition. J Am Chem Soc 2003;125:4686–7. 11. Sirion U, Kim HJ, Lee JH, et al. An efficient F-18 labeling method for PET study: Huisgen 1,3-dipolar c ycloaddition of bioacti ve substances and F-18-labeled compounds. Tetrahedron Lett 2007;48:3953–7. 12. Mindt TL, Str uthers H, Brans L, et al. “Click to chelate”: synthesis and installation of metal chelates into biomolecules in a single step. J Am Chem Soc 2006;128:15096–7. 13. Sun EY, Josephson L, Weissleder R. “Clickable” nanoparticles for targeted imaging. Mol Imaging 2006;5:122–8. 14. White MA, Johnson JA, Koberstein JT, Turro NJ. Toward the syntheses of universal ligands for metal o xide surfaces: controlling surface functionality through click chemistr y. J Am Chem Soc 2006;128:11356–7. 15. Lin PC, Ueng SH, Yu SC, et al. Surf ace modif ication of magnetic nanoparticle via Cu(I)-catal yzed alk yne-azide [2+3] c ycloaddition. Org Lett 2007;9:2131–4. 16. Dijkgraaf I, Rijnders AY, Soede A, et al. Synthesis of DO TA-conjugated multi valent c yclic-RGD peptide dendrimers via 1,3dipolar c ycloaddition and their biolo gical e valuation: implications for tumor tar geting and tumor imaging pur poses. Org Biomol Chem 2007;5:935–44. 17. Binder WH, Sachsenhofer R, Straif CJ , Zirbs R. Surf ace-modified nanoparticles via ther mal and Cu(I)-mediated “click” chemistr y: generation of luminescent CdSe nanopar ticles with polar ligands guiding supramolecular reco gnition. J Mater Chem 2007;17: 2125–32. 18. Viguier RFH, Hulme AN. A sensitized europium comple x generated by micromolar concentrations of copper(I): to ward the detection of copper(I) in biology. J Am Chem Soc 2006;128:11370–1. 19. Zhu L, L ynch VM, Anslyn EV. FRET induced b y an ‘allosteric’ cycloaddition reaction re gulated with e xogenous inhibitor and effectors. Tetrahedron 2004;60:7267–75. 20. Sivakumar K, Xie F , Cash BM, et al. A fluoro genic 1,3-dipolar cycloaddition reaction of 3-azidocoumarins and acetylenes. Or g Lett 2004;6:4603–6.

479

21. Beatty KE, Liu JC, Xie F , et al. Fluorescence visualization of ne wly synthesized proteins in mammalian cells. Angew Chem Int Ed 2006;45:7364–7. 22. Zhou Z, F ahrni CJ. A fluoro genic probe for the copper(I)-catal yzed azide-alkyne ligation reaction: modulation of the fluorescence emission via (3)(n, pi*)-(1)(pi, pi*) in version. J Am Chem Soc 2004;126:8862–3. 23. Manetsch R, Krasinski A, Radic Z, et al. In situ click chemistr y: enzyme inhibitors made to their o wn specif ications. J Am Chem Soc 2004;126:12809–18. 24. Lewis WG, Green LG, Gr ynszpan F, et al. Click chemistr y in situ: acetylcholinesterase as a reaction vessel for the selective assembly of a femtomolar inhibitor from an ar ray of building b locks. Angew Chem Int Ed 2002;41:1053–7. 25. Wang JY, Sui GD , Mocharla VP, et al. Inte grated microfluidics for parallel screening of an in situ click chemistr y librar y. Angew Chem Int Ed 2006;45:5276–81. 26. Krasinski A, Radic Z, Manetsch R, et al. In situ selection of lead compounds by click chemistr y: target-guided optimization of acetylcholinesterase inhibitors. J Am Chem Soc 2005;127:6686–92. 27. Whiting M, Muldoon J, Lin YC, et al. Inhibitors of HIV -1 protease by using in situ click chemistry. Angew Chem Int Ed 2006;45:1435–9. 28. Mocharla VP, Colasson B , Lee L V, et al. In situ click chemistr y: enzyme-generated inhibitors of carbonic anh ydrase II. Angew Chem Int Ed 2005;44:116–20. 29. Lee LV, Mitchell ML, Huang SJ, et al. A potent and highly selective inhibitor of human alpha-1,3-fucosyltransferase via click chemistry. J Am Chem Soc 2003;125:9588–9. 30. Tornoe CW, Sanderson SJ, Mottram JC, et al. Combinatorial library of peptidotriazoles: identif ication of [1,2,3]-triazole inhibitors against a recombinant Leishmania me xicana cysteine protease. J Comb Chem 2004;6:312–24. 31. Xie J, Seto CT . A two stage click-based librar y of protein tyrosine phosphatase inhibitors. Bioorg Med Chem 2007;15:458–73. 32. Alam MS, Kajiki R, Hanatani H, et al. Synthesis and str ucture-activity relationships of 1-phen yl-1H-1,2,3-triazoles as selecti ve insect GABA receptor antagonists. J Agric Food Chem 2006;54:1361–72. 33. Chang KH, Lee L, Chen J , Li WS. Lithocholic acid analo gues, new and potent alpha-2,3-sialyltransferase inhibitors. Chem Commun 2006:629–31. 34. Goeminne A, McNaughton M, Bal G, et al. 1,2,3-Triazolylalkylribitol derivatives as nucleoside hydrolase inhibitors. Bioorg Med Chem Lett 2007;17:2523–6. 35. Kim S, Cho M, Lee T, et al. Design, synthesis, and preliminar y biological e valuation of a no vel triazole analo gue of ceramide. Bioorg Med Chem Lett 2007;17:4584–7. 36. Pirali T, Gatti S, Di Brisco R, et al. Estrogenic analogues synthesized by click chemistry. Chemmedchem 2007;2:437–40. 37. Prazeres VFV, Sanchez-Sixto C, Castedo L, et al. Nanomolar competitive inhibitors of Mycobacterium tuberculosis and Streptomyces coelicolor type II dehydroquinase. Chemmedchem 2007;2:194–207. 38. Suh BC, Jeon HB , Posner GH, Silverman SM. Vitamin D side chain triazole analogs via c ycloaddition ‘click’ chemistry. Tetrahedron Lett 2004;45:4623–5. 39. Bock VD, Speijer D, Hiemstra H, van Maarseveen JH. 1,2,3-Triazoles as peptide bond isosteres: synthesis and biolo gical evaluation of cyclotetrapeptide mimics. Org Biomol Chem 2007;5:971–5. 40. Appendino G, Bachhie ga S, Minassi A, et al. The 1,2,3-triazole ring as a peptido- and olef inomimetic element: disco very of click vanilloids and cannabinoids. Angew Chem Int Ed 2007;46:9312–5. 41. Hsu TL, Hanson SR, Kishikawa K, et al. Alkynyl sugar analogs for the labeling and visualization of gl ycoconjugates in cells. Proc Natl Acad Sci U S A 2007;104:2614–9. 42. Sawa M, Hsu TL, Itoh T, et al. Glycoproteomic probes for fluorescent imaging of fucosylated glycans in vivo. Proc Natl Acad Sci U S A 2006;103:12371–6.

480

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

43. Beatty KE, Xie F , Wang Q, Tirrell D A. Selecti ve dy e-labeling of newly synthesized proteins in bacterial cells. J Am Chem Soc 2005;127:14150–1. 44. Bradford JA, Clarke ST, Buck SB, et al. Analysis of cell cycle blockers using click chemistr y catalyzed EdU detection. Cytometr y A 2007;71A:766. 45. Deiters A, Cropp TA, Mukherji M, et al. Adding amino acids with novel reactivity to the genetic code of Saccharom yces cerevisiae. J Am Chem Soc 2003;125:11782–3. 46. Deiters A, Schultz PG. In vi vo incor poration of an alk yne into proteins in Escherichia coli. Bioor g Med Chem Lett 2005; 15:1521–4. 47. Feng T, Tsao ML, Schultz PG. A phage display system with unnatural amino acids. J Am Chem Soc 2004;126:15962–3. 48. Gauchet C, Labadie GR, P oulter CD . Re gio- and chemoselecti ve covalent immobilization of proteins through unnatural amino acids. J Am Chem Soc 2006;128:9274–5.

49. Labadie GR, Viswanathan R, P oulter CD . F arnesyl diphosphate analogues with ome ga-bioorthogonal azide and alk yne functional groups for protein farnesyl transferase-catalyzed Ligation reactions. J Org Chem 2007;72:9291–7. 50. Wittig G, Krebs A. Zur existenz niedergliedriger cycloalkine .1. Chem Ber 1961;94:3260–75. 51. Agard NJ, Prescher J A, Ber tozzi CR. A strain-promoted [3+2] azidealkyne cycloaddition for covalent modif ication of b lomolecules in living systems. J Am Chem Soc 2004;126:15046–7. 52. Baskin JM, Prescher JA, Laughlin ST, et al. Copper-free click chemistry for dynamic in vi vo imaging. Proc Natl Acad Sci U S A 2007;104:16793–7.

31 THE “ONE-BEAD-ONE-COMPOUND” COMBINATORIAL APPROACH To IDENTIFYING MOLECULAR IMAGING PROBES RUIWU LIU, PHD, OLULANU H. AINA, DVM, PHD, EKAMA ONOFIOK, BS, AND KIT S. LAM, MD, PHD

In the “one-bead-one-compound” (OBOC) combinatorial library method, thousands to millions of compound beads are rapidly generated on solid support with a “split-mix” synthesis approach such that each bead displa ys only one chemical entity. These chemical compounds can be peptides, peptoids, peptidomimetics, small molecules and macroc yclic molecules. These OBOC libraries are then screened in a highthroughput f ashion for binding or functional acti vities, and positive beads are isolated for str ucture elucidation. This enabling technology not onl y f acilitates dr ug discovery but also provides a highly efficient approach to develop imaging agents for various diseases. In this chapter, the application of the OBOC method to disco ver cell surf ace targeting agents and highly specific protease substrates, both of which can be used for cancer imaging, will be discussed.

In 1991, we first reported the “one-bead-one-compound” (OBOC) combinatorial librar y method in w hich lar ge combinatorial peptide-bead libraries (> one million permutations) are rapidl y generated b y a “split-mix” synthesis approach (F igure 1A). 1 The resulting peptide library consists of beads of 90 µm diameter, with each individual bead car rying 10 13 copies of a single peptide entity. The OBOC libraries are then screened against a variety of biolo gical tar gets and the positi ve beads are isolated for structure determination. In the past 16 y ears, various screening assa ys have been de veloped to screen OBOC combinatorial libraries, 2,3 and many investigators around the world have successfully used these techniques in identifying peptide epitopes of specif ic monoclonal antibodies; peptide motifs reco gnized b y major histocompatibility complex molecules; peptide motifs that are substrates for kinases, proteases, and phosphatases; peptide inhibitors against v arious enzymes and cell

signaling proteins; peptide and peptoid motifs against various protein domains and cell surf ace receptors; peptidic artificial enzymes with catalytic activities; and peptides with affinity for specif ic organic dyes. We have recently reviewed the use of phage-displa y and OBOC combinatorial librar y methods to de velop cell surf ace ligands as tar geting agents for cancer therapy and imaging. 4 The phage-displa y peptide librar y method involves the introduction of random DN A segments into the genome of f ilamentous phages, resulting in the displa y of random peptides at the pIII proteins located at one end of the phages. Phages that bind to intact cells can be isolated for DN A sequencing. In the OBOC combinatorial library method, 90 µm resin beads displaying peptides, peptidomimetics, or small molecules on the bead surf ace are incubated with li ve cells (10 to 20 µm in diameter) at 37°C. Beads coated with cells are then isolated for chemical str ucture determination by Edman microsequencing (peptides) or chemical decoding (peptidomimetics or small molecules). A major advantage of the OBOC combinatorial librar y method over phage-displa y librar y methods is that the OBOC method is synthetic; thus, one can easily incor porate D-amino acids, unnatural amino acids, organic moieties, macrocyclic, bic yclic, tur ned, or branched str uctures into the libraries. In contrast, in phage-display libraries, one is restricted to the use of the 20 geneticall y coded natural L-amino acids. As a result, peptides identif ied through phage-displa y methods are subject to proteolytic de gradation, w hereas D-amino acids containing peptides, peptidomimetics, or small molecules identified through the OBOC method are much more

481

482

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B Resin beads

H2N

Cycle 1 (3 compounds) Cycle 2 (9 compounds)

Split

I5

M5 1) Mix

H2N 1) Mix

Cycle 3 (27 compounds)

P5

M5

H2N

2) Split

I5

P5

(20%)

FmocHN (80%) H2N

H2N

2) Split

I5

M5

H2N

P5

1) Swell in water, 24 h 2) Fmoc-OSu (0.2 eq. to resin) DIEA, DCM/ether (55/45), 30 min

Alloc-OSu, DIEA AllocHN DMF, 1 h

AllocHN

1) Swell in water, 24 h 2) Pd(PPH3)4, PhSiH3 DCM, 7 min

(20%)

H2N AllocHN (80%)

C

Figure 1. A, General synthetic scheme for one-bead-one-compound combinatorial library using a “split-mix” synthesis approach. M, I, P represent different building blocks. B, Two approaches for generating topographically segregated bilayer beads, and C, photomicrograph of bilayer beads from (B). Free amines at the inner core of each bead react with bromophenol blue to produce blue color.

stable to proteolysis, a very important consideration for effective in vivo imaging studies. The systematic evolution of ligands by exponential enrichment aptamer technology is another combinatorial library method that can be used to identify ligands against cell surf ace receptors. 5 In this method , a librar y of random single strand oligonucleotides (aptamers) is synthesized with the four standard nucleoside building b locks. Aptamers that bind to intact cells or purif ied cell surface receptors are then isolated and enzymaticall y amplified for subsequent c ycles of panning. This method enab les one to identify high-af finity cell surface-binding aptamers. However, there have been only a few reports6 on the use of aptamers in in vi vo imaging. Compared to shor t peptides (e g, 10-mers), aptamers (30-mers) are signif icantly lar ger and more costl y to synthesize. In addition, to a void degradation by plasma nucleases, aptamers need to be chemicall y modified. OBOC combinatorial peptide librar y methods ha ve also been used successfull y in the identif ication of substrates for proteases, 7–10 protein kinases, 11–14 and phosphatases.15,16 There has been considerab le interest in the development of in vi vo imaging probes for both proteases and kinases. In principle, the substrates identif ied through screening OBOC combinatorial peptide libraries could potentially be de veloped into in vi vo imaging agents for these enzymes. Tumor cells secrete proteol ytic enzymes into the extracellular matrix at the tumor site. P eptide substrate probes that can detect specif ic proteol ytic acti vity

could potentially be used as tumor imaging agents. In contrast, most protein kinases of interest are intracellular , and therefore, cell-per meable peptidomimetic or small molecule substrates will be needed. Our preliminar y data indicate that such substrates can indeed be disco vered through screening OBOC libraries. In principle, 18F labeled protein kinase substrate can be phosphor ylated and retained inside the tumor cells with high kinase activity. To discover in vivo imaging agents for other intracellular cell signaling proteins, one ma y consider screening OBOC combinatorial small molecule and macrocyclic molecules for high affinity and high-specificity ligands against these proteins. Structure deter mination of isolated positi ve OBOC library beads can be achieved by direct microsequencing if the librar y compounds are α-amino acid-containing peptides or peptoids. If the librar y compound is nonsequenceable by Edman degradation, chemical encoding is needed. To facilitate chemical encoding, w e have developed topographically segregated bilayer beads such that the librar y compound is on the bead surf ace w here the cells interact and the coding tags reside on the bead interior.17,18 In this chapter, we will discuss the use of OBOC combinatorial library methods in identif ication and optimization of cell surf ace-binding ligands, w hich can be used as tar geting agents for cancer imaging and therap y. In addition, we will briefly discuss methods of disco vering plasma-stab le peptide substrates, w hich are highl y specific to unique proteases, b y screening D-amino acid containing OBOC combinatorial libraries.

The “One-Bead-One-Compound” Combinatorial Approach to Identifying Molecular Ima ging Probes

OBOC CONCEPT Merrifield f irst described the solid phase peptide synthesis (SPPS) method in 1963, 19 for w hich he was awarded the Nobel prize in 1985. Split-synthesis or splitmix synthesis method1,20,21 is a modification of the SPPS method in w hich the resin is split into equal por tions, placed into separate reaction vessels, and each portion of resin is reacted with a dif ferent chemical entity. Prior to the ne xt coupling c ycle, indi vidual resin por tions are combined, re-split, and reacted with a second set of chemical entities. Using this method, equimolar ratios of all possible peptide permutations are attained. Furka and colleagues20 used this approach to mak e solution phase peptide libraries. Houghten and colleagues 21 used this method to mak e solution phase peptide mixtures. Lam and colleagues 1 recognized that since each resin bead encounters onl y one chemical entity during each coupling cycle and the reaction is dri ven to completion, the resulting bead librar y will contain all possib le permutation of peptides but each bead will display only one peptide entity . As a consequence, a lar ge number of chemical molecules can be synthesized in a shor t amount of time using this approach, and the resulting OBOC library can be screened immediatel y in an ultrarapid f ashion. Literall y millions of pepti des can be synthesized and screened within a week. In our initial report, we succeeded in identifying ligands against streptavidin and a monoclonal antibody (anti- β endorphin), two model target proteins that ha ve been used in screening phage-displa y peptide libraries. Ov er the y ears, technical aspects and applications of this basic method have improved signif icantly with the availability of new solid support, assays, and chemistries, including synthesis of peptoid , peptidomimetic, small molecule, and macrocyclic OBOC libraries. No vel bilayer beads ha ve also been de veloped for ef ficient and reliab le chemical encoding.22 These OBOC libraries are not commerciall y available so f ar but can be easil y made b y experienced chemists.

GENERAL CONSIDERATIONS FOR OBOC LIBRARY SYNTHESIS OBOC Library Design The length of the amino acid chain in the peptide librar y and the number of amino acids used in each coupling cycle deter mine the per mutation or di versity that can exist within the librar y. F or e xample, a 5-mer peptide library synthesized with 20 different amino acids in each

483

position has 20 5 or 3.2 × 106 permutations. It has been statistically estimated that w hile screening a librar y with less than 3 × 106 permutations, one can easily cover 95% of all possib le permutations (with a 99% CI) b y screening 10 7 beads. However, it is not necessar y to screen all possible per mutations in a biolo gical system since most target molecules have only three to f ive contact residues that are essential for specif ic interaction. The remaining residues are considered spacers. Considering this and the fact that 1 mL (settled beads) of an OBOC peptide library (90 µm bead) contains appro ximately 7.5 × 105 beads, one can comfor tably screen 1 mL of the librar y in each experiment and repeat this several times. For small molecule or peptidomimetic OBOC libraries, w e generall y incorporate three-point di versities into a f ixed chemical scaffold (a chemical molecule with multiple functional groups).17,23 The scaffold can be premade23,24 or for med in situ on the resin as thebuilding blocks are incorporated (unpublished data). Nonpeptidic macroc yclic libraries containing heteroc yclic building b locks ha ve been less explored in the OBOC library method but have the potential as a useful source for compounds that interact with protein surfaces.

Selecting Resin for OBOC Library For OBOC libraries used in on-bead binding assa ys (e g, whole cell binding as described belo w), TentaGel resin (Rapp Polymere, Tübingen, Germany) is typically used as solid support due to its proper ties: it is nonsticky, uniform in size and distribution, and e xhibits high s welling in a wide range of organic solvents and water. OBOC libraries made on TentaGel resin have been successfully used in the identification of cell surface-binding ligands, using wholecell binding assays (see below). The OBOC library method has also been used successfully in the identification of protease substrates and inhibitors 7–10; here, highly porous resins such as PEGA beads (Calbiochem-No vabiochem, San Die go) w ere used for librar y constr uction since it allowed the enzyme to gain access to the bead interior. The OBOC method has also been adapted to solution phase assays. An elegant and potentiall y po werful approach to OBOC librar y screening is the in situ solution phase releasable assay,25–27 in which the compound-bead libraries are immobilized within a thin la yer of agar , followed by release of compounds from each bead into the immediate vicinity surrounding each bead, for solution phase assay in the semisolid matrix. An ideal resin for such an application must be mechanicall y durable, chemically stable, of uniform size, shape, and substitution, and enab le the compounds (both hydrophilic and hydrophobic) to dif fuse out

484

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

of the bead readil y in aqueous en vironments. Neither TentaGel nor PEGA beads meet these requirements, and thus we have recently developed a novel shell-core bilayer bead in which a hydrogel is grafted onto the TentaGel bead as an outer shell (5 to 90 µm thick), via free radical surface-initiated polymerization (unpublished data).

OBOC Library Compounds Have Inbuilt Handles for Tagging One advantage of the OBOC method is that each librar y compound is tethered to the solid suppor t via a link er such as pol yethylene gl ycol. When positi ve compounds are identified via screening against biological cancer targets, the linker can be used as a convenient handle to link the cancer targeting ligand to an optical dye, a radiometal chelate or a c ytotoxic drug (see below). In each indi vidual probe development study, we have to determine what type and length of spacer w ould be appropriate for minimal interference. A spacer that is too shor t may interfere with the receptor binding to the ligand , likewise a spacer that is too long ma y wrap around the ligand and also interfere with binding.

OBOC LIBRARY SYNTHESIS Synthesis of OBOC Peptide Libraries OBOC peptide libraries can be readil y synthesized on solid phase suppor t using a split-mix synthesis method (Figure 1A). Beads are split into reaction vessels and a single amino acid is coupled in each v essel. Beads are then combined and re-split into reaction v essels for a second coupling c ycle. This process is continued until the last cycle of coupling is completed. F or OBOC peptide libraries consisting of sequenceable α-amino acids (including 20 eukaryotic, v arious amino acid deri vatives, and unnatural amino acids), synthesis is straightforw ard using a standard SPPS method and using 9-fluoren ylmethoxycarbonyl (Fmoc)/ tert-butyl (t-Bu) chemistr y.3 The coupling reagent pair N-hydroxybenzotriazole (HOBt)/ N, Nʹ′-diisopropylcarbodiimide (DIC) is preferab le because the reaction is mild and usuall y gi ves cleaner results. For dif ficult couplings, 2-( 1H-9-azabenzotriazole-1-yl)1,1,3,3-tetramethyluronium hexafluorophosphate (HATU), 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU), or benzotriazole-1-yloxy-tris-pyrrolidino-phosphonium he xa-fluoro-phosphate (PyBOP) in the presence of N,N-diisopropylethylamine (DIEA) may be used. P ositive peptide beads isolated via

screening can be microsequenced directl y with Edman chemistry. F or peptides without a ter minal amino g roup (eg, cyclic peptides using the N-terminal amino g roup for cyclization), branched peptides, or peptides with one or more nonsequenceable building blocks (eg, β- and γ-amino acids), the OBOC librar y synthesis requires modif ication of the synthetic strate gy and chemical encoding (see below).

Synthesis of Encoded OBOC Peptidomimetic and Small Molecule Libraries Synthesis of OBOC peptidomimetics and small molecule libraries requires a dif ferent strate gy and chemical encoding method. Chemical coding tags are added to the bead to record the synthetic histor y of each compound bead during the coupling steps. The coding tag can then be decoded b y either Edman microsequencing or mass spectrometry (MS). To avoid coding tag interference during screening, we typically use bilayer beads for OBOC library synthesis, w here the coding tags are within the bead interior. We have developed two simple but robust methods to prepare bila yer beads. The f irst method is called the par tial amine-protection (P AP) bila yer approach.17 In this method , the TentaGel resin bead is first swollen in water and then vigorously mixed with an amine-derivatizing reagent [e g, N-(9-fluorenylmethoxycarbonyloxy) succinimide (Fmoc-OSu)] in a mixture of dichloromethane (DCM) and dieth yl ether (55/45, v/v). The outer layer of the bead is exposed to the derivatizing reagent, whereas the bead interior remains in w ater with negligible exposure to the derivatizing reagent, thus confining deri vatization to the outer la yer of the bead (Figure 1B, top). The thickness of the outer layer (Fmocprotection percentage) of beads can be controlled b y the amount of Fmoc-OSu used (e g, 0.2 eqv . to resin total loading). Another approach is called par tial allocdeprotection (PAD) bilayer method.18 In this method, the amino g roups of the TentaGel beads are f irst fully protected by alloc and then thoroughl y swollen in w ater, followed b y deprotection with a palladium reagent [(Pd(PPh3)4] in DCM for a pre-defined, limited amount of time. This results in beads with a deprotected outer la yer (free N-ter mini) and an alloc-protected inner core (Figure 1B , bottom). The thickness of the outer la yer (alloc-deprotection percentage) of beads is dependent on the duration of deprotection. It is impor tant to point out that the “bilayer step” for both methods can be used at any stage of the librar y synthesis and can be repeated multiple times. With bila yer beads on hand , we ha ve

The “One-Bead-One-Compound” Combinatorial Approach to Identifying Molecular Ima ging Probes

successfully de veloped se veral encoding methods for synthesis of nonsequenceab le peptide, peptidomimetic, and small molecule OBOC libraries. 17,24,28 In our encoding systems, the librar y compounds are displayed on the outer layer of the bead, whereas the coding tags reside on the bead interior . Such a conf iguration f acilitates the encoding process and also allo ws us to minimize interference by the coding tags during screening. The coding tag can then be decoded by either Edman microsequencing or MS. An e xample of a peptide-encoded peptidomimetic library is shown in Figure 2. In our encoding strategy, the reduction of the NO 2 group, the removal of protecting g roups Fmoc and N-[1-(4,4-dimeth yl-2,6dioxocyclohex-1-ylidene)ethyl] (Dde) of NH 2 [using 2% hydrazine in N,N-dimethylformamide (DMF)], and the acylation of both coding and testing ar ms are achie ved simultaneously. This g reatly reduces the number of synthetic steps, and therefore, reduces possib le by-products. For decoding, positive beads are isolated and subjected to automatic microsequencing. An advantage of microsequencing as a decoding method is the nonrequirement of clea vage and retrie val of coding tags. Ho wever, microsequencing has the follo wing disadv antages: (1) Edman sequencing is time consuming and e xpensive, compared to MS anal ysis; (2) microsequencing is not as readily available to most laboratories as MS.

STRUCTURE DETERMINATION OF OBOC LIBRARY Edman Sequencing As mentioned abo ve, for OBOC peptide libraries that are comprised of α-amino acids or their deri vatives and peptoids (N-substituted oligo glycine), an automatic

H2N

Couple X1 and X2 Split/mix synthesis

H2NX2X1

protein sequencer (e g, Procise 494, P erkin-Elmer/ Applied Biosystems) is routinel y used to deter mine the amino acid sequence via our de veloped sequencing method.29 Microsequencing can also be applied for str ucture deter mination to a v ariety of peptide-encoded OBOC libraries including nonsequenceable peptide, peptidomimetic and small molecule libraries.

MS Decoding The common MS decoding methods for OBOC peptide libraries are “ladder sequencing,”30 “ladder synthesis,”31 and the very recently reported “partial Edman degradation” (PED) method.32 The ladder sequencing method is restricted to libraries with sequenceab le peptides or peptoids. Unlik e our encoding strate gy that emplo ys coding tags within the inner core of the beads, the “ladder synthesis” method repor ted in the original ar ticle presented peptide ladders on the bead surf ace, w hich interferes with biolo gical screening. We recentl y reported an impro ved “ladder synthesis” approach that applies both “ladder synthesis” and bila yer bead concepts to encode OBOC nonsequenceab le peptide and peptidomimetic libraries. 18 The general synthetic scheme and MS-based decoding strate gy of such peptide libraries is sho wn in F igure 3 (using an OBOC tetra-peptide librar y as an e xample). Prior to librar y synthesis, we assembled a clea vable linker (CL) inside the beads, which can be cleaved with cyanogen bromide in the presence of acids such as trifluoroacetic acid. During library construction, the PAD “bilayer step” was used prior to each Fmoc-amino acid coupling c ycle. As a result, w e w ere ab le to remo ve the alloc protecting group, layer by layer, from the bead surf ace to the bead

1) Swell in water, 24 h 2) Alloc-OSu (0.2 eq.), DIEA, DCM/ether (55/45), 30 min Bilayer beads preparation

AllocHNX2X1

1) Fmoc-Phe(4-NO2), HOBt, DIC 2) 20% Piperidine 3) Boc-K(Dde), HOBt, DIC 4) Pd(PPh3)4, PhSiH3

H2NX2X1

TentaGel resin

H2NX2X1

Testing molecule FmocHN

COOH

FmocHN

R1COHN COHNX2X1

NO2

BocHN-K-Phe-X2X1 Dde NO2

HOBt, DIC

NO2 BocHN-K-Phe-X2X1

1) 2% NH2NH2/DMF 2) R1COOH, HOBt, DIC 3) SnCl2·2H2O, DMF

COHNX2X1

NHCOR

2 4) R2COCl, DIEA; or R2COOH, HOBt, DIC 5) TFA cocktail deprotection H2N-K-Phe-X2X1

Dde NO2

COR1 NHCOR2 Peptide coding tag

Figure 2.

485

Synthetic scheme of a peptide-encoded one-bead-one-compound peptidomimetic library.

486

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

interior. Upon completion of the librar y constr uction, each peptide bead carried a complete library compound (X4X3X2X1-) on the outer la yer, and four tr uncated ladder members (X 4X3X2X1-, X4X3X2-, X4X3-, and X 4-) within the bead interior (F igure 3). The major advantages of the ne wly developed method are (i) onl y a single building b lock is used during each coupling step thereby eliminating problems caused by the differential coupling rates of tw o dif ferent building b locks (as in the initial “ladder synthesis” method); (ii) this method can be utilized to synthesize peptides comprised of both sequenceab le and nonsequenceab le building blocks; and (iii) the undesirab le interference of ladder tags during biological screening can be a voided (unlike in the pre viously reported ladder synthesis method). In the PED method , Edman de gradation chemistr y w as used to for m a ladder of tr uncated peptide coding tags from the N-ter minal (X 4X3X2X1-, X 3X2X1-, X 2X1-, and X1-) (Figure 4). The Pei group has successfully applied our PAP bilayer approach to ward synthesis of a c yclic peptide OBOC librar y, in w hich peptide coding tags reside within the inner core of the beads, and are decoded using the PED method.32 One limitation of this method is that the peptide coding tag has to be comprised of α-amino acids.

1) Swell in water, 24 h 2) Alloc-OSu (0.2 eq.) DIEA, DMF, 1 h

5) Couple Fmoc-A(Br) 6) De-Fmoc 7) Couple Fmoc-Ebes 8) De-Fmoc 9) Alloc-OSu, DIEA

H2N (80%)

PAD

Fmoc-X1

(20%)

(20%) PAD

(20%) Fmoc-X1-CL

3) Couple Fmoc-X1 4) Combine and mix beads

7 min

(60%) AllocHN-CL

(20%)

1) PAD; De-Fmoc; Couple Fmoc-X3 2) PAD; De-Fmoc; Couple Fmoc-X4

Cell surf ace receptors are e xcellent biological targets for polyethylene gl ycol imaging of disease processes, especially cancer. Target specificity is greatly enhanced by differential e xpression of cell surf ace receptors betw een cancer and nor mal tissues; it is ideal that the tar geting agent be specific for its cognate receptor on the cancer cell, with little to no binding in non-malignant tissue such as liver, spleen, and bone marrow. Peptides, peptidomimetics, small molecules, or macroc yclic molecules are thus becoming increasingly popular alternatives to antibody targeting as the latter tends to bind nonspecif ically to the reticuloendothelial system. To disco ver peptide, peptidomimetic, or small molecule tar geting agents, one ma y use purif ied receptors in conjunction with an enzymelinked colorimetric assa y.33 Alternatively, “w hole” li ving cells may be used as probes to screen the OBOC libraries (Figure 5). 34 The use of “li ve” cells in librar y screening is convenient and attracti ve because (i) cloning, e xpression, and purif ication of the receptors are not needed , (ii) the receptor is more likely to be in its native state, and (iii) one

AllocHN

AllocHN

AllocHN-CL

AllocHN -Ebes-A(Br)-Arg(Pbf)-Met

1) De-Fmoc 2) Couple Fmoc-X2

Fmoc-X1

(20%) Fmoc-X1-CL H2N-CL (20%) (40%) AllocHN-CL

X4X3X2X1

(20%) X4X3X2X1-CL (20%) X4X3X2-CL

3) De-Fmoc; Side chain deprotection

General Considerations

1) Couple Fmoc-Met, HOBt/DIC 2) 20% Piperidne (De-Fmoc) 3) Couple Fmoc-Arg(Pbf) 4) De-Fmoc

(20%) AllocHN

H2N

1) Swell in water, 24 h 2) Pd(PPh3)4, PhSiH3 DCM, 7 min

DISCOVERY OF TARGETING AGENTS AGAINST CELL SURFACE RECEPTORS

Fmoc-X2X1 Fmoc-X2X1-CL Fmoc-X2-CL AllocHN-CL

CNBr, TFA

MALDI-TOF MS

Release

Decoding

X4

X2

X3

X1

M/Z

(20%) X4X3-CL X4-CL (20%)

HN

Br

NH2 NH

Ebes:

H2N

O

O

O

H N O

OH

CL: cleavable linker

H2N

O

O

O

H N O

O N H

N H

H N

O OH

O S

Figure 3. General synthetic scheme and MS-based decoding strategy of Edman-unsequenceable peptide libraries. To facilitate the interpretation of the mass spectra, a bromine-containing amino acid is incorporated into the cleavable linker so that the coding peak (doublet) can be readily identified.

The “One-Bead-One-Compound” Combinatorial Approach to Identifying Molecular Ima ging Probes

H2N

“Split-mix” peptide library synthesis

H2N-X4X3X2X1

1) 5:1 PITC/Nic-OSu 2) TFA

Nic-X4X3X2X1 repeat 3x

Nic-X4X3X2X1-CL Nic-X3X2X1-CL Nic-X2X1-CL

Nic-X4X3X2X1 Nic-X4X3X2X1-CL H2N-X3X2X1-CL

H2N-X4X3X2X1-CL

H2N-CL

487

CNBr, TFA

MALDI-TOF MS

Release

Decoding

X1

X2

X3

X4

M/Z

Nic-X1-CL Figure 4. Partial Edman degradation (PED) decoding of bead-bound peptide. PITC = phenylthioisocyanate; Nic-OSu = N-hydroxysuccinimidyl nicotinate.

A

B

C

D

E

Figure 5. Screening methods for the identification of ligands against cell surface receptors: A, enzyme-linked colorimetric assays using soluble cell surface receptors as the screening probe, B, whole-cell binding assay using live suspension cells as the screening probe, C, whole-cell binding assay using live adherent cells as the screening probe, and D/E, dual-color whole-cell binding assays.

can disco ver no vel ligands against unkno wn receptors. However, since there are numerous receptors and biolo gical molecules on the cell surface, one will need to develop methods to eliminate all nonspecif ic-ligands. To discover ligands against known receptors, one can take advantage of the nati ve ligands, using their w ell-studied str uctures as

templates for OBOC combinatorial librar y design. F or example, -ar ginine-glycine-aspartic- (-RGD-) is a w ellknown binding motif recognized by a number of inte grins such as αvβ3, αIIbβ3, αvβ5, and α5β1; one could design cyclic peptide libraries of dif ferent ring sizes, containing the RGD motif or RGD-related sequences. One ma y

488

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

include se veral basic residues such as ar ginine, homoarginine, lysine, and ornithine in the “R” position and use acidic residues such as aspar tate and glutamate at the “D” position. The flanking residues can be either D- or Lamino acids. Cyclization can be achie ved b y disulf ide bond or β-lactam for mation. F or receptors without an y known ligands or for or phan receptors, one ma y screen a number of random linear , c yclic, tur ned, or branched libraries. To f acilitate selection of the appropriate librar y prior to large-scale screening, we have recently developed a novel “rainbow” bead librar y approach 35 in which small samples of each color -coded librar y beads (eight colors) are combined and screened simultaneousl y against li ving cells. Based on the bead color , w e can easil y identify libraries that generate the most number of positi ve beads. A lar ger sample of those specif ic libraries can then be screened under highl y stringent conditions for high-af finity binding. For example, we prepared six RGD-containing cyclic peptide libraries of v arious ring sizes and labeled each of these libraries with a dif ferent color. We noticed that the c yclic librar y with cxRGDxxc motif yielded the highest number of positive beads when αvβ3 integrin-transfected K562 cells were used in library screening. From our experience with whole-cell binding assays, the number of tr ue positive binding beads detected in a specific OBOC library screening depends greatly on both the cell line and the librar y used. Therefore, in a primary screen for a specific cell line, we often screen both linear and cyclic peptide libraries with various lengths and ring sizes and select the most promising libraries for higher stringency screening.

Binding Specificity Since there are an enormous number of proteins or macromolecules on the surf ace of each cell and these cells are exposed to o ver 100,000 compound beads, it is e xpected that some nonspecif ic binding can occur . One such “nonspecific” binding interaction can result from charge-charge interaction such as that obser ved in nonspecif ic cell interactions with polycations such as polylysine. It is therefore not surprising that during our previous studies, we isolated several beads displaying highly positively charged ligands. We have since limited (but not eliminated) the number and amount of basic building b locks (such as ar ginine and lysine) in the library design. To eliminate compound beads that bind to both cancer and normal cells alike, we perform sequential screenings, f irst with the cancer cells and subsequently with their nonmalignant counter parts. Cancer cell-bound beads are f irst isolated, stripped of cells using guanidine HCl, rec ycled, and then incubated with nor mal

cells. Only those beads that bind to the cancer cells but not normal cells are isolated for structure determination. Alternatively, we have devised a dual color screening strate gy (Figure 5 D/E),36 which emplo ys Calcin AM labeling of malignant cells, and a counterstain of the normal cells such that beads which bind cells of a distinct stain are selected for str ucture deter mination. To ensure specif icity of the isolated ligands for a par ticular receptor, one ma y use a subtraction assa y, in w hich cell binding is perfor med before and after addition of a kno wn competing ligand or antibody against the receptor of interest.

Lead Optimization and Screening Stringency Initial ligands identif ied through screening di verse OBOC libraries often e xhibit lo w af finity. To optimize such ligand leads, w e often perfor m simple str ucturefunction relationship studies, such as an “alanine scan, ” deletion studies, and D-amino acid replacements, to determine the critical residues or secondar y str uctures required for binding. In an alanine scan, each amino acid of the parent peptide is replaced with alanine and the binding proper ties of each of these analo gues are determined. In deletion studies, the peptide can be tr uncated separately from the N-ter minus or from the C-ter minus, one amino acid at a time. To determine the importance of the chirality of each residue, each amino acid can be replaced with its enantiomer . Based on such studies, a motif of residues critical for ligand-receptor interaction can be deter mined and used as a template to design focused OBOC libraries for fur ther optimization. Another v ariation of this type of focused librar y is the one where the acti ve motif is f ixed, while extending the amino or carbo xyl ter minus of the peptide with a sequence of random residues. This approach allo ws one to probe for additional contact residues adjacent to the initial ligand binding site, w hich fur ther optimize binding. Additional pharmacophores can also be probed by generating a branch of random sequence, at the middle of the peptide using the ε-amino group of L- or D-l ysine as the branching residue. Alternatively, one ma y use nitrophenylalanine as the branching residue. 17 The nitro group can be readil y converted to an amino g roup with SnCl2 reduction prior to attachment of additional building b locks. Another approach for optimization is to develop homolo g libraries in w hich building b locks found in the lead compounds at specif ic positions are incorporated into the librar y in higher amounts (concentration) at those positions. As a result, the librar y will be biased toward analogues related to the lead compounds. 2

The “One-Bead-One-Compound” Combinatorial Approach to Identifying Molecular Ima ging Probes

Many of these methods ha ve already been successfull y applied in optimization of our l ymphoma and o varian cancer targeting ligands.34,37 To distinguish high-af finity beads from moderateaffinity beads, one ma y screen such libraries under higher stringenc y. The three general approaches for increasing screening stringenc y are: (i) decrease the surface substitution of the beads prior to librar y construction such that ligand density on the bead surface is lo wer,38 (ii) incor porate a competing ligand in the incubation medium, 37 and (iii) shor ten the incubation time for binding. These approaches can be used either alone or in combination. In our e xperience, screening focused libraries under high stringenc y has repeatedly led to the de velopment of compounds 10 to 1,000 fold more potent than the initial leads 37 as determined b y a major drop in half maximal (50%) inhibitory concentration (IC 50) in a competiti ve cell adhesion assa y to immobilized peptide deri ved from native ligand (see section on LLP2A belo w).

NOVEL CELL SURFACE TARGETING LIGANDS IDENTIFIED THROUGH SCREENING OBOC LIBRARIES Over the past decade, w e ha ve screened numerous OBOC libraries against v arious cancer cell lines: primary tumor cells isolated from leuk emia, l ymphoma, and transition-cell carcinoma patients and cells isolated from mammar y tumor tissue of transgenic mice. Some of the ligands are unique to a specif ic cell line, whereas others are common to multiple cancer cell types. We often identify ligands with w ell kno wn or pre viously reported motifs such as the RGD motif, reco gnized by cell surface integrins. However, the receptor identities of the majority of our ligands are still unkno wn. One approach to determine the receptor identity is to evaluate the ability of a series of antireceptor antibodies to b lock cell binding to the compound bead. In one of our studies, we identif ied a cDGXGXXc peptide motif to w hich the receptor was unknown. When we searched a protein database for natural proteins that may contain this motif, we found (amongst man y) collagen I. We therefore reasoned that if the peptide ligand resemb led collagen I, the cells ma y be binding through inte grin receptors. Therefore, we systematically performed competitive cell adhesion assay using peptide-coated beads and a panel of anti-integrin receptor antibodies. Using this method , we disco vered that cDGXGXXc w as a no vel binding motif for α3β1 integrin.34 A second approach is to screen cell lines transfected with a single species of inte grin or

489

receptor, subsequently comparing differential binding of the transfected and parent cell lines with the compound beads. A third approach is to de velop the ligand into a photoaffinity or a chemoaf finity labeling probe thereb y cross-linking it to its receptor , w hich is ultimatel y tagged for isolation and sequence identif ication. Interestingly, many but not all the ligands that w e have isolated from screening OBOC peptide libraries were found to bind to matrix receptors and cell adhesion molecules such as inte grins. This f inding is not sur prising as cell adhesion molecules are present at relati vely high concentrations on the surf ace of most cells and tend to bind to shor t peptide motifs present in the e xtracellular matrix. Table 1 summarizes published cancer cell surface ligands identif ied b y the OBOC combinatorial librar y method. A limited number of these ligands ha ve been used successfull y for in vi vo optical and PET imaging.4,37,39,40

dGXG Peptides for Alpha-3 Integrin Imaging This section recapitulates methods b y which cancer cell surface targeting ligands can be identif ied and optimized through screening random OBOC combinatorial libraries and their further development into optical and PET imaging agents. CaO V-3 (o varian adenocarcinoma cell line) was f irst screened with a random linear peptide librar y (XXXXXXX, w herein X = 19 L-natural amino acids except c ysteine), however, we f ailed to identify an y ligands with specif ic and signif icant binding. CaO V-3, SKOV-3, ES-2, and O VCAR-3 ovarian cancer cells w ere then screened with se veral random c yclic libraries (eg, cXnc and cBnc, wherein B = 19 L-natural amino acids but with lo wer le vels of R or K). A number of positive beads containing a fe w distinct motifs w ere identif ied: (D/N)GR-, -DGXG-, -WDD-, and -LDI-. Of the 90 sequenced hits, nine of the peptides w ere re-synthesized to conf irm binding to these cell lines. Re-binding results sho wed that peptides with a cDGXGXXc motif bound with higher af finity than the others. We subsequently perfor med str ucture-activity relationship (SAR) studies (e g, alanine scan) on f ive of the sequenced hits including the cDGXGXXc peptides. We deter mined that the critical residues required for cell binding were DGXG, and a free N-terminus and peptide constraint (cyclization) 34 were also required. Two secondar y libraries (cXGXGXXc; 100% and 20% surf ace substitution) were then designed and synthesized based on the SAR data, wherein X represents 45 different amino acids that include both natural and unnatural amino acids. To dif ferentiate

490

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 1. CANCER TARGETING LIGANDS IDENTIFIED BY ONE-BEAD-ONE-COMPOUND COMBINATORIAL LIBRARY METHODS Ligands to Specific Tumor Cell Surface Receptors

Sequence

In vivo Imaging

α3-integrin ovarian adenocarcinoma (CaOV-3, ES-2, SKOV-3, OVCAR-3); metastatic breast cancer (MDA-MB 231), glioblastoma (A172)34; and melanoma

cd/DGX*GXXc (X* = Cha, Chg, HCit, Cit, F, Y, Nle, M, L or I )

Aina et al.4,40

α3-integrin NSCLC49

cNGXGXXc

––

α4β1 integrin Raji B-cell lymphoma

FSIpLDI, sppLDIn, eapLDId, fypLDFf, QSYpLDF, cLDYWDc, cWDLDHHc

––

α4β1 integrin Jurkat T-cell lymphoma41, review50

cLDIXXc, cXLDI/V/Fc, cXXLDIc, cWDXXXc, XXXpLDI/F/V, xLDFpXXX, xxxxp-Nle-DIxxxx

––

α4β1 integrin Jurkat T-cell lymphoma37,51,52

“LLP2A” peptidomimetic and a series of analogues

α4β1 integrin bronchioloalveolar H1650 carcinoma53

c-Nle-DXXXXc, cX-Nle-DXXXXc

––

α4β1 integrin ES-2 ovarian adenocarcinoma

c-Nle-D-Chg-YMc, cSD-Nle-D-Chg-c, c-Nle-DVDEc, c-Nle-DWEEc, wdinp-Nle-DIgsfn, yminp-Nle-DIdnhh, vqgp-Nle-DIafvl, vgnvp-Nle-DIgqea, wsrip-Nle-DIqeps, vswap-Nle-DIgspd

––

α4β1 integrin SKOV-3 ovarian adenocarcinoma

cLDI-Chg-Hyp-Yc, c-Nle-D-Chg-NDFc, c-Nle-D-Nle-PhgDc, cDEL-Nle-EWc

––

α6β1 integrin DU145 prostate cancer cell line54,55

kmviywkaG, kikmviswkG, yiknrkhhG, kGGrhykfG, LNNIVSVNGRHX, DNRIRLQAKXX

––

WEHI-231 murine lymphoma cell line,33 review50

lwxxpewi, wGeyixvx, kwxGpxw, XWYD/T/V

––

WEHI-279 murine lymphoma cell line,33 review50

xGrfxswx, xtxGmxkx, RWID, RWFD

––

Peng et al.37,39

Single letter representation for amino acid according to standard convention, except for those amino acids without single letter representation: Cha = cyclohexylalanine; Chg = α-cyclohexylglycine; Cit = citrulline; HCit = homocitrulline; Hyp = hydroxyproline; Nle = norleucine; Phg = phenylglycine.

D-isomers from L-isomers, 10% norleucine (Nle) w as added to the bead interior w hen a D-isomer w as chosen for that position. These ne w “secondar y” libraries w ere screened with CaO V-3, O VCAR-3, ES-2, and SK OV-3 cell lines to identify ligands with higher af finity for these cells. We noticed that while it took about 24 to 48 hours to observe cell binding with the initial random libraries, w e observed binding within 45 minutes of incubation of cells with the focused cXGXGXXc libraries. Approximately 45 hits were isolated and sequenced, and seven new motifs emerged.4

Development of LLP2A, a High-Affinity and High-Specificity Lymphoma Targeting Ligand Through screening random linear and c yclic peptide libraries, we previously discovered LDI, LDV, and LDF as specif ic ligands against malignant T-cells but not

normal peripheral b lood mononuclear cells. 41 Through database search and antibody blocking studies, we subsequently deter mined the nati ve receptor for these ligands to be α4β1 integrin and its use as an e xcellent therapeutic and imaging tar get for both T- and B-l ymphoma. Based on a pre vious repor t that a 4-(( Nʹ′-2methylphenyl)ureido)-phenylacetyl N-terminal cap at LDV g reatly enhanced α4β1 integrin binding, 42 we designed se veral peptidomimetic OBOC libraries capped b y v arious ureido deri vatives and screened these libraries under high stringent conditions. Based on these studies, prefer red building blocks at the LDV tripeptide positions w ere deter mined and a highl y focused encoded OBOC peptidomimetic librar y with 1560 per mutations w as designed and prepared 37 (Figure 6). When screened under v ery high stringenc y in the presence of 500 µM of a competing α4β1 antagonist (BIO-1211), 42 12 positive beads w ere isolated for chemical decoding and 6 of them had identical chemical str uctures. We named this no vel tar geting

The “One-Bead-One-Compound” Combinatorial Approach to Identifying Molecular Ima ging Probes

491

COOH

(

H N

)

O N H

O

X2X3X4

O

N H

N H

Coding tag

N

H2N

O

X3 position

H2N

COOH

COOH

H2N

COOH

NH2

O

NH

Focused library

X2 position

O

H N

N H

O

N H

L D V

X1

O

H N

LLP2A

X4 position

H2N

COOH H2N

COOH H2N

COOH H2N

COOH H2N

H2N

COOH H2N

COOH H2N

COOH H2N

COOH H2N

H2N

COOH

COOH

COOH H2N

COOH

H2N

COOH

H2N

COOH

COOH

COOH H2N

H2N

COOH

H2N

H2N

COOH H N

COOH

COOH

COOH

COOH

H2N

COOH

COOH

H2N

OH H2N

COOH H2N

COOH

N

HN S COOH

H2N H N

COOH

Cl

I

O H2N

COOH

OH H2N

H N

H2N

COOH

I

COOH O

H2N

H2N

H2N

H2N

COOH

COOH

H2N

COOH

COOH

N O

O H2N

COOH

H2N

H2N

COOH

COOH

H2N

COOH

O

Figure 6. Chemical composition and structure of the highly focused encoded one-bead-one-compound peptidomimetic library from which LLP2A was identified as well as its structure. X1 position contains 4-aminophenylacetic acid and 2-(4-aminophenyl)propanoic acid.

agent LLP2A (see F igure 6). LLP2A is resistant to proteolysis, stab le in human plasma at 37°C for 18 days, binds to both T- and B-malignant lymphoid cells, and has an IC50 of 2 pM in a cell adherence assay using immobilized CS-1 f ibronectin peptide. 37

DEVELOPMENT OF IMAGING AND PHOTOAFFINITY OR CHEMOAFFINITY LABELING PROBES FROM LIGANDS ISOLATED FROM OBOC LIBRARIES Design of Probes During on-bead w hole-cell binding assa ys, OBOC library compounds are tethered to the beads via a long

polyethylene gl ycol link er. Therefore, carbo xyl ter mini of the positi ve ligands isolated through such screenings can be con veniently used as handles for tagging: Cy5.5 or Alexa 680 (near infrared fluorescent dy es) for optical imaging, 1,4,7,10-tetraazac yclododecane-1,4,7,10tetraacetic acid (DO TA) or h ydrazinonicotinamide (Hynic) for radioimaging, 4-benzo yl-L-phenylalanine (Bpa) or azido g roups for photoaf finity labeling, and L-3,4-dihydroxyphenylalanine (DOPA) for chemoaf finity labeling. The ligand can also be biotinylated and then linked to various secondary streptavidin conjugates. The general scheme for synthesis of these conjugates is shown in F igure 7. To minimize the interference of ligand binding by the tag, two hydrophilic linkers (ie, Ebes linker,43 see F igure 3) can be inser ted betw een the tag and the ligand.

492

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

1) 20% Piperidine 2) Fmoc-R, HOBt, DIC 3) 20% Piperidine

FmocHN

R

H N

4) Fmoc-Ebes, HOBt, DIC 5) 20% Piperidine 6) Fmoc-Ebes, HOBt, DIC 7) 20% Piperidine

Rink resin

O

H2N

O

NH O

O

H N

N H

O

O

H N

O

O Ligand synthesis

Cleavage and purification

O

N H

Ligand

R

B: Ligand-Bpa

NH O

O

H N

N H

O

O

H N

O

H2N

A: Ligand-biotin

O

O

O

O

O

C: Ligand-azido

D: Ligand-DOPA OH

O

R

S

N H

O

R

H

R

R

OH

N3

H HN

NH O

F: Ligand-Cy5.5

E: Ligand-DOTA

O

O

R

N H

O COOH N N

HOOC

G: Ligand-Hynic

N

R

N H N

N

R O

N COOH

N

N H

NH2

S O O

O O S O

N H

S O O O

O S O O

Figure 7. Synthesis of ligand-biotin (A), ligand-Bpa (B), ligand-azido (C), ligand-DOPA (D), ligand-DOTA (E), ligand-Cy5.5 (F), and ligand-Hynic (G).

In Vivo Optical and PET Imaging In recent y ears, w e ha ve repor ted the successful use of ligands developed by the OBOC library method for in vivo tumor imaging. F or e xample, cdG-HoCit-GPQc (O A02) peptide, an α-3 inte grin ligand for ES-2 human o varian clear cell carcinoma, w hen directly linked to either Cy 5.5 or Alexa-680 to for m a smaller uni valent conjugate, or biotinylated follo wed b y comple xation with strepta vidinCy5.5 to for m a lar ger tetravalent probe, has been used to image mouse x enografts with high specif icity.40 The smaller (~2.6 kD) uni valent OA02-Cy5.5/Alexa 680 probe showed a rapid uptak e in the tumor (within 5–15 min postinjection) with a peak in intensity around 60 to 90 minutes and a signal that w as sustained over several hours. In contrast, the lar ger (~60 kD) tetra valent O A02-biotinStreptavidin-Cy5.5 conjugate accumulated betw een 30 minutes postinjection and peak ed at 6 to 24 hours.

Clearance via the kidne ys w as also more rapid for the smaller OA02-Cy5.5 probe. No uptake was observed in the lungs, liver, spleen, heart, or muscle. Some background signal was observed in the gastrointestinal tract pre-injection and postinjection when ingested contents were not cleared. We have also perfor med PET imaging studies with OA02 using 64Cu and DOTA as a radiometal chelator.4 Ex vivo biodistribution data from mice ( n = 5) euthanized 2 hours postinjection with 9 to 13µCi [64Cu]-labeled OA02DOTA radioconjugate sho wed uptak e (%ID/g ± SD) in urine (1.75 ± 0.25), bladder (0.063 ± 0.022), and kidneys (0.049 ± 0.0018). Uptak e b y the ES-2 tumor at 2 hours was 0.0100 ± 0.0008 (negative control Raji tumor showed an uptake value of 0.0044 ± 0.0009). Some retention was observed in the liver (0.036 ± 0.0061) and small intestine (0.011 ± 0.0015). Muscle, hear t, spleen, b lood, and ovaries had uptak e values between 0.001 and 0.005 with a SD of 0.0003 to 0.001. The high radiouptak e

The “One-Bead-One-Compound” Combinatorial Approach to Identifying Molecular Ima ging Probes

observed in the liver and gastrointestinal tract was probably due to chelation instability and subsequent variations in sequestration and metabolism of free, unchelated 64Cu in these organs compared to others. Similarly, LLP2A, a high-af finity (IC 50 = 2 pM) peptidomimetic ligand for acti vated α4β1 integrin can image both T- and B-l ymphoma x enografts with high specificity using biotin/strepta vidin-Alexa 680 or Cy5.5 as optical tags 37,39 and 111In-DOTA or 64Cu-DOTA as radioactive tags for in vi vo imaging (manuscript submitted). However, we have noticed that renal uptak e was as high as tumor uptake in these studies. F igure 8 shows the in vivo and e x vivo near infrared imaging result of nude mice bearing Molt-4 l ymphoma and A549 non-small cell lung cancer (a negative control) xenografts 24 hours after

a A

IV injection of biotin-LLP2A/strepta vidin-Alexa 680 conjugate. Uptak e of the LLP2A optical probe b y the Molt-4 lymphoma was highly specific. Little or no uptake was obser ved in the A549 lung cancer and li ver. Renal uptake of both LLP2A-strepta vidin-Alexa 680 conjugate and streptavidin-Alexa 680 alone was, however, very high. It has been repor ted that 1,4,8,11-tetraazabic yclo[6.6.2]hexadecane-4,11-diacetic acid (CB-TE2A) is a superior chelate for 64Cu with no leakage of the radiometal out of the chelate. 44,45 Recent PET imaging studies in our laboratory with [ 64Cu]-labeled LLP2A-CB-TE2A conjugate did indeed show that tumor radiouptake was high but liver radiouptake was very low, confirming very low leakage of [64Cu] out of the radioconjugate (unpublished data). Equally impor tant is that renal uptak e of the ne w

c C

b B

Molt-4 (+)

0

A549 (-)

SA680 alone

493

7500

LLP2ASA680

SA680 alone

LLP2ASA680

SA680 alone

LLP2ASA680

Bladder

Lymph node

Muscle

Intestine

Lung

Kidney

Spleen

A549 tumor

Molt-4 tumor

Liver

Heart

d D

LLP2ASA680 E e

f F

SA680 alone

g G

Figure 8. In vivo and ex vivo near-infra red imaging of nude mice bearing Molt-4 T-lymphoma (left shoulder) and A549 non-small cell lung cancer (right shoulder) xenografts. Near-infra red images (B, E, & G) were obtained 24 h after injection of LLP2A-biotin/Streptavidin-Alexa 680 conjugate (LLP2A-SA680) (D & E) or Streptavidin-Alexa 680 (SA680) alone (F & G) via tail vein. Specific uptake of optical probe into Molt-4 tumor, but not lung cancer was evident. However, renal uptake was observed with both conjugates. Minimal liver uptake was observed. A, D, and F are white light images. C is an overlap image of A and B.

494

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

radioconjugate was also v ery low. This latter obser vation, together with the imaging data obtained with the LLP2A optical and radoimaging probes mentioned above, strongly suggest that renal uptake of the LLP2A conjugates was due to DOTA, Cy5.5, or strepta vidin but not LLP2A (manuscript in preparation).

bond between amino acid P 1 and P 1ʹ′. Proteases generally only cleave peptides with an L-amino acid at the P 1 site. To be highl y specif ic, a protease peptide substrate requires highly specif ic cleavage at the P 1-P1ʹ′ site, and in addition, the rest of the substrate must be highly resistant to cleavage by an y other proteases, for e xample, the hundreds of proteases present in the blood. To achieve this, we generated fluorescent-quenched OBOC random D-amino acid peptide libraries on PEGA beads (F igure 9A), w hich contain a limited number of L-amino acids 48: Y(NO 2)-xxXXXxxK(Abz)-bead, Y(NO2)-xxxXXxx-K(Abz)-bead, and Y(NO2)-xxxXxxx-K(Abz)-bead, w herein Y(NO2) = 3nitrotyrosine, Abz = 2-aminobenzoic acid , x = D-amino acids, and X = L-amino acids. In the presence of nitrotyrosine (the quencher), Abz (fluorescent probe) is quenched and therefore the majority of the beads do not fluoresce. Upon cleavage by a protease, depending on the clea vage site, a por tion of the peptides on the positi ve beads is cleaved, rendering a loss of quencher -fluorophore proximity within each bead as the quencher is released into the medium and thus generating a fluorescent signal at the bead (see Figure 9B). Upon microsequencing with Edman degradation, one can often deter mine the precise proteolytic cleavage site on the peptide. To eliminate peptide beads susceptib le to clea vage by plasma proteases, w e f irst screened the complete librar y with w hole plasma, follo wed b y remo val of fluorescent beads (indicative of proteolytic cleavage) manually under a fluorescent microscope or b y the use of a fluorescent bead sorter (COPAS, Biometrica Inc). The remaining beads were then incubated with the protease of interest. F or tumor imaging, urokinase and MMP-2 are good choices as these proteases are known to be present at high level in some cancers. To develop on-demand CLs for radioimmunotherap y, we incubated plasma-prescreened libraries with TNKase (Genetech Inc.), a tissue plamino gen activator. Using this screening strategy, we were able to identify highly specific

DISCOVERY OF HIGH-SPECIFICITY AND HIGH-EFFICIENCY PROTEASE SUBSTRATES AS CANCER IMAGING PROBES Tung and colleagues46 first reported the use of fluorescentquenched oligomeric protease peptide substrates as in vivo optical imaging probes for cancer . Jiang and colleagues later de veloped an ele gant approach to image protease activity in vi vo by linking fluorescent-labeled oligo-ar ginine and quencher -labeled oligo-glutamic acid with a matrix metalloproteinase-2 (MMP-2) substrate PLGLAG.47 Upon cleavage of the protease substrate at the tumor site, the fluorescent-labeled oligo-ar ginine is no longer blocked by the oligo-glutamic acid , leading to cellular uptake of the fluorescent probe at the tumor site. Both methods, although very useful, are still plagued by a major problem: the presence of a large number of circulatory proteases. The majority of injected peptide substrates containing all L-amino acids will be cleaved in circulation prior to reaching the tumor site. To o vercome this prob lem, w e have developed a method to de velop novel plasma-stable peptide substrates for specif ic proteases. Although our strategy w as initiall y de veloped for the disco very of on-demand cleavable linkers (CLs) for therapeutic radioimmunoconjugates,48 it can be readily applied to the development of protease imaging agents for cancer. By convention, the proteolytic site is defined by P3-P2P1-P1ʹ′-P2ʹ′-P3ʹ′, of w hich the clea vage site is the peptide A

BB

(A)

1) Dde-Lys(Fmoc) HOBt, DIC 2) 20% Piperidine

O DdeHN

H2N

O

1) Boc-Abz, HOBt, DIC 2) 2% hydrazine, DMF

H2N

N H

Peptide synthesis

N H

HN

PEGA resin

O

H2N H N

H2N-xxxXXxx O

O N H

HN O NHBoc

1) Fmoc-Tyr(3-NO2) HOBt, DIC 2) 20% Piperidine

NHBoc H2N

3) TFA cocktail side chain deprotection

O

H N

xxxXXxx N H

O

O NO2

OH

N H

HN O NH2

Figure 9. Discovery of highly specific protease substrates with fluorescent-quenched one-bead-one-compound random D-amino acid containing peptide libraries. A, cleavage of fluorescent-quenched peptide library results in fluorescent beads and B, photomicrographs of fluorescent beads as a result of specific cleavage by added protease.

The “One-Bead-One-Compound” Combinatorial Approach to Identifying Molecular Ima ging Probes

TNKase substrates that are stab le in human plasma. As expected, the number of fluorescent beads resulting from treatment of a random OBOC librar y with plasma w as directly proportional to the L-amino acid content in the peptide librar y. As an alter native to using D-amino acids as flanking residues for the P 1-P1ʹ′ cleavage site, one may use other building b locks. F or e xample one ma y use Nalkylated glycines to generate h ybrid peptide-peptoid substrate libraries or or ganic molecules to generate peptidomimetic libraries, both of w hich could potentiall y yield highly specific protease substrates.

FUTURE DIRECTIONS It is clear from the above discussion that the OBOC combinatorial librar y method is a po werful technology that can facilitate the discovery of cell surf ace targeting ligands and protease substrates, which are useful for tumor imaging and therap y. Tumor-specific tar geting ligands linked to a metal chelator can be used as imaging agents when loaded with 111In or 64Cu or as therapeutic agents when loaded with 90Y or 67Cu. The highl y specif ic MMP-2 link ers can be used to de velop ef fective prodrugs, w hich can be acti vated at the tumor site. Using these imaging-therapy parallel platforms, physicians will be able to practice personalized medicine by using these imaging agents as tools to select ef fective therapies for specific patients and to monitor their treatment response. With the recent adv ances in the chemistr y, encoding, and screening of OBOC combinatorial libraries, we are no longer restricted to peptide or peptoid libraries. Small molecule, peptidomimetic, macroc yclic, photos witchable, glyco-organic, bic yclic, tric yclic, and branched libraries can no w be readil y synthesized on bila yer beads and encoded with various strategies. Small molecule substrates for a number of disease-associated enzymes could, in principle, be discovered using this approach. We have already succeeded in identifying peptidomimetic substrates for specific protein kinases. These compounds sho w g reat promise as ef fective in vi vo imaging probes for these enzymes. Using in situ releasab le assa ys,25 hundreds of thousands of dif ferent small molecules can each be released from indi vidual OBOC librar y beads and monitored for specific cellular uptake and retention via intrinsic fluorescent signals or via a radioacti ve tag. Some of these molecules ma y ha ve specif ic af finity to disease-rele vant cellular targets and therefore ma y prove useful as in vi vo imaging agents. In addition to screening OBOC libraries with living cells, one ma y also de velop screening assa ys using whole or ganisms, such as zebra f ish, to deter mine biodistribution of the fluorescent librar y compounds

495

within the li ving organism, in a high throughput manner . The future use of OBOC combinatorial library methods to develop novel imaging agents is both promising and bright.

ACKNOWLEDGMENTS The authors would like to acknowledge the financial support from NIH g rants R33 CA89706, NCDDG U19CA113298, R01CA15483, 1R21CA135345, and Califor nia Cancer Research Program, Contract No. 00-00764V-20133.

REFERENCES 1. Lam KS, Salmon SE, Hersh EM, et al. A new type of synthetic peptide librar y for identifying ligand-binding acti vity. Nature 1991; 354:82–4. 2. Lam KS, Lebl M, Krchnak V. The “one-bead-one-compound” combinatorial library method. Chem Rev 1997;97:411–48. 3. Lam KS, Lehman AL, Song A, et al. Synthesis and screening of “onebead one-compound” combinatorial peptide libraries. Methods Enzymol 2003;369:298–322. 4. Aina OH, Liu R, Sutcliffe JL, et al. From combinatorial chemistry to cancer-targeting peptides. Mol Pharm 2007;4:631–51. 5. Lupold SE, Hicke BJ, Lin Y, Coffey DS. Identif ication and characterization of nuclease-stabilized RN A molecules that bind human prostate cancer cells via the prostate-specif ic membrane antigen. Cancer Res 2002;62:4029–33. 6. Hicke BJ, Stephens AW, Gould T, et al. Tumor targeting by an aptamer. J Nucl Med 2006;47:668–78. 7. Juskowiak GL, Stachel SJ , Tivitmahaisoon P, Van Vranken DL. Fluorogenic peptide sequences—transfor mation of shor t peptides into fluorophores under ambient photoo xidative conditions. J Am Chem Soc 2004;126:550–6. 8. Meldal M, Svendsen I, Breddam K, Auzanneau FI. Portion-mixing peptide libraries of quenched fluoro genic substrates for complete subsite mapping of endoprotease specificity. Proc Natl Acad Sci U S A 1994;91:3314–8. 9. Meldal M. The one-bead two-compound assay for solid phase screening of combinatorial libraries. Biopolymers 2002;66:93–100. 10. Olivos HJ, Bachhawat-Sikder K, Kodadek T. Quantum dots as a visual aid for screening bead-bound combinatorial libraries. Chembiochem 2003;4:1242–5. 11. Wu J, Ma QN, Lam KS. Identifying substrate motifs of protein kinases by a random library approach. Biochemistry 1994;33:14825–33. 12. Wu JJ , Afar DE, Phan H, et al. Reco gnition of multiple substrate motifs by the c-ABL protein tyrosine kinase. Comb Chem High Throughput Screen 2002;5:83–91. 13. Lam KS, Wu J, Lou Q. Identification and characterization of a no vel synthetic peptide substrate specific for Src-family protein tyrosine kinases. Int J Pept Protein Res 1995;45:587–92. 14. Lou Q, Leftwich ME, Lam KS. Identif ication of GIYWHHY as a novel peptide substrate for human p60c-src protein tyrosine kinase. Bioorg Med Chem 1996;4:677–82. 15. Cheung YW, Abell C, Balasubramanian S. A combinatorial approach to identying protein tyrosine phosphatase substrates from a phosphotyrosine peptide library. J Am Chem Soc 1997;119:9568–9. 16. Garaud M, Pei D. Substrate prof iling of protein tyrosine phosphatase PTP1B by screening a combinatorial peptide librar y. J Am Chem Soc 2007;129:5366–7. 17. Liu R, Marik J, Lam KS. A novel peptide-based encoding system for “one-bead one-compound” peptidomimetic and small molecule combinatorial libraries. J Am Chem Soc 2002;124:7678–80.

496

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

18. Wang X, P eng L, Liu R, et al. P artial alloc-deprotection approach for ladder synthesis of “one-bead one-compound” combinatorial libraries. J Comb Chem 2005;7:197–209. 19. Merrifield RB. Solid phase peptide synthesis. I. The synthesis of a tetrapeptide. J Am Chem Soc 1963;85:2149–54. 20. Furka A, Sebestyen F, Asgedom M, Dibo G. General method for rapid synthesis of multicomponent peptide mixtures. Int J P ept Protein Res 1991;37:487–93. 21. Houghten RA, Pinilla C, Blondelle SE, et al. Generation and use of synthetic peptide combinatorial libraries for basic research and drug discovery. Nature 1991;354:84–6. 22. Liu R, Wang X, Song A, et al. Development and applications of topologically se gregated bila yer beads in one-bead one-compound combinatorial libraries. QSAR Comb Sci 2005;24:1127–40. 23. Dixon SM, Milinkevich KA, Fujii J, et al. A spiroisoxazolinoprolinebased amino acid scaf fold for solid phase and one-bead-onecompound library synthesis. J Comb Chem 2007;9:143–57. 24. Song A, Zhang J, Lebrilla CB, Lam KS. A novel and rapid encoding method based on mass spectrometr y for “one-bead-one-compound” small molecule combinatorial libraries. J Am Chem Soc 2003;125:6180–8. 25. Salmon SE, Liu-Ste vens RH, Zhao Y, et al. High-v olume cellular screening for anticancer agents with combinatorial chemical libraries: a new methodology. Mol Divers 1996;2:57–63. 26. Jayawickreme CK, Sauls H, Bolio N , et al. Use of a cell-based , lawn format assa y to rapidl y screen a 442,368 bead-based peptide library. J Pharmacol Toxicol Methods 1999;42:189–97. 27. Silen JL, Lu AT, Solas DW, et al. Screening for no vel antimicrobials from encoded combinatorial libraries by using a two-dimensional agar format. Antimicrob Agents Chemother 1998;42:1447–53. 28. Wang X, Zhang J , Song A, et al. Encoding method for OBOC small molecule libraries using a biphasic approach for ladder -synthesis of coding tags. J Am Chem Soc 2004;126:5740–9. 29. Liu R, Lam KS. Automatic Edman microsequencing of peptides containing multiple unnatural amino acids. Anal Biochem 2001;295:9–16. 30. Chait BT, Wang R, Bea vis RC, K ent SB. Protein ladder sequencing. Science 1993;262:89–92. 31. Youngquist RS, Fuentes GR, Lace y MP, Keough T. Generation and screening of combinatorial peptide libraries designed for rapid sequencing b y mass spectrometr y. J Am Chem Soc 1995; 117:3900–6. 32. Joo SH, Xiao Q, Ling Y, et al. High-throughput sequence deter mination of cyclic peptide library members by partial Edman degradation/mass spectrometry. J Am Chem Soc 2006;128:13000–9. 33. Lam KS, Lou Q, Zhao ZG, et al. Idiotype specific peptides bind to the surface immunoglobulins of tw o murine B-cell l ymphoma lines, inducing signal transduction. Biomed Pept Proteins Nucleic Acids 1995;1:205–10. 34. Aina OH, Marik J , Liu R, et al. Identif ication of no vel tar geting peptides for human o varian cancer cells using “one-bead onecompound” combinatorial libraries. Mol Cancer Ther 2005; 4:806–13. 35. Luo J, Zhang H, Xiao W, et al. Rainbow beads: a color coding method to f acilitate high-throughput screening and optimization of one-bead one-compound combinatorial libraries. J Comb Chem 2008;10:599–604. 36. Lam KS, Wade S, Abdul-Latif F, Lebl M. Application of a dual color detection scheme in the screening of a random combinatorial peptide library. J Immunol Methods 1995;180:219–23.

37. Peng L, Liu R, Marik J , et al. Combinatorial chemistr y identif ies high-affinity peptidomimetics against α4β1 integrin for in vi vo tumor imaging. Nat Chem Biol 2006;2:381–9. 38. Wang X, P eng L, Liu R, et al. Applications of topolo gically se gregated bila yer beads in ‘one-bead one-compound’ combinatorial libraries. J Pept Res 2005;65:130–8. 39. Peng L, Liu R, Andrei M, et al. In vivo optical imaging of human lymphoma xenograft using a librar y-derived peptidomimetic against α4β1 integrin. Mol Cancer Ther 2008;7:432–7. 40. Aina OH, Marik J, Gandour-Edwards R, Lam KS. Near-infrared optical imaging of o varian cancer x enografts with no vel α3-Integrin binding peptide “OA02.” Mol Imaging 2005;4:439–47. 41. Park SI, Manat R, Vikstrom B , et al. The use of one-bead one-compound combinatorial librar y method to identify peptide ligands for α4β1 integrin receptor in non-Hodgkin’ s l ymphoma. Lett Pept Sci 2002;8:171–8. 42. Lin K, Ateeq HS, Hsiung SH, et al. Selective, tight-binding inhibitors of inte grin α4β1 that inhibit aller gic airw ay responses. J Med Chem 1999;42:920–34. 43. Song A, Wang X, Zhang J, et al. Synthesis of hydrophilic and flexible linkers for peptide deri vatization in solid phase. Bioor g Med Chem Lett 2004;14:161–5. 44. Boswell CA, Sun X, Niu W, et al. Comparati ve in vi vo stability of copper-64-labeled cross-bridged and conventional tetraazamacrocyclic complexes. J Med Chem 2004;47:1465–74. 45. Garrison JC, Rold TL, Sieckman GL, et al. In vi vo e valuation and small-animal PET/CT of a prostate cancer mouse model using 64Cu bombesin analogs: side-by-side comparison of the CB-TE2A and DOTA chelation systems. J Nucl Med 2007;48:1327–37. 46. Tung CH, Mahmood U, Bredow S, Weissleder R. In vi vo imaging of proteolytic enzyme activity using a novel molecular reporter. Cancer Res 2000;60:4953–8. 47. Jiang T, Olson ES, Nguyen QT, et al. Tumor imaging by means of proteolytic activation of cell-penetrating peptides. Proc Natl Acad Sci U S A 2004;101:17867–72. 48. Kumaresan PR, Natarajan A, Song A, et al. Development of tissue plasminogen activator specific “on demand cleavable” (odc) linkers for radioimmunotherapy by screening one-bead-one-compound combinatorial peptide libraries. Bioconjug Chem 2007;18:175–82. 49. Lau D, Guo L, Liu R, et al. P eptide ligands targeting integrin α4β1 in non-small cell lung cancer. Lung Cancer 2006;52:291–7. 50. Aina OH, Sroka TC, Chen ML, Lam KS. Therapeutic cancer tar geting peptides. Biopolymers 2002;66:184–99. 51. Liu R, Peng L, Han H, Lam KS. Str ucture-activity relationship studies of a series of peptidomimetic ligands for α4β1 integrin on Jurkat T-leukemia cells. Biopolymers 2006;84:595–604. 52. Carpenter RD, Andrei M, Lau EY, et al. Highly potent, water soluble benzimidazole antagonist for activated α4β1 integrin. J Med Chem 2007;50:5863–7. 53. Mikawa M, Wang H, Guo L, et al. No vel peptide ligands for inte grin α4β1 overexpressed in cancer cells. Mol Cancer Ther 2004; 3:1329–34. 54. Pennington ME, Lam KS, Cress AE. The use of a combinatorial library method to isolate human tumor cell adhesion peptides. Mol Divers 1996;2:19–28. 55. DeRoock IB , P ennington ME, Sroka TC, et al. Synthetic peptides inhibit adhesion of human tumor cells to extracellular matrix proteins. Cancer Res 2001;61:3308–13.

32 CHEMICAL BIOLOGY APPROACHES MOLECULAR IMAGING

TO

STANLEY SHAW, MD, PHD

Molecular imaging relies on a specif ic reco gnition event between an imaging probe and its cellular tar get. This chapter focuses on small organic molecules, either as targeting moieties to direct nanopar ticles to specif ic cell types or as critical components of generalizab le, isotope-free imaging probes. Historicall y, man y small molecules have been shown to bind their protein targets with high affinity and specif icity and have been incorporated into imaging probes (e g, PET -based agents, reviewed in other chapters). Serendipity has pla yed a significant role in the discovery of many of these small molecules, however, limiting wider application of this approach. In recent y ears, adv ances in synthetic and bioconjugation chemistr y, high-throughput screening and anal ytic methods ha ve con verged to enab le the more systematic application of small molecules as probes of biolo gic processes. In this conte xt, the ter m chemical biology refers to the use of small molecules and other chemical tools in a systematic (rather than ad hoc) manner to study biologic problems. This chapter will highlight recent developments in chemical biology that promise to enab le new advances in molecular imaging.

NOVEL CHEMICAL MATTER: NATURAL PRODUCTS, DRUG-LIKE MOLECULES, AND DIVERSITY-ORIENTED SYNTHESIS Natural Products The ability of natural product small molecules (ie, isolated from li ving organisms) to bind protein tar gets or modulate their function has opened se veral important new fields of biology. For instance, ingestion of the corn lily Veratrum californicum causes cyclopia (a single eye,

or holoprosencephaly) in lambs; this led to the identif ication of the steroidal alkaloid c yclopamine as the causative terato gen.1,2 Subsequent w ork sho wed that cyclopamine inhibits the Hedgehog signaling pathway3 4 by binding to Smoothened (Smo). Furthermore, cyclopamine b locks the e xcessive hedgeho g signaling and abnor mal cell g rowth associated with onco genic mutations af fecting the Shh pathw ay,5 identifying the hedgehog pathw ay as a tar get of mechanism-based cancer therap y, and leading to in vestigations of cyclopamine and other hedgeho g inhibitors as anticancer agents. Cyclopamine is just one e xample among many natural products that have catalyzed major advances in biological understanding and medical therap y, including cyclosporin (which led to the discovery of calcineurin),6 rapamycin (mT OR),7,8 and trapo xin (histone deacetylases [HDACs]).9 In general, natural products target key pathways or re gulatory nodes and cause compelling phenotypes in vitro or in vi vo. They are str ucturally complex molecules with multiple stereocenters and bind their targets with very high affinity (Figure 1). Despite these f avorable proper ties, se veral signif icant limitations ha ve pre vented the more widespread pursuit of natural products as imaging probes or therapeutics. Their disco very is often serendipitous; purif ication of individual compounds from extracts is challenging; and obtaining sufficient quantities can be dif ficult because of the intricate schemes required for their synthesis.

Drug-like Molecules Because of the above limitations of natural product chemistry, simpler, “dr ug-like” molecules are commonl y used as a source of chemical per turbations, particularly in the 497

498

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 1. Comparison of the relative structural complexity and ease of synthesis among drug-like, natural product, and diversity-oriented synthesis molecules.

pharmaceutical industry. These compounds are much easier to synthesize in large quantities using a limited number of steps (enab ling lar ge scale high-throughput screening efforts), but at the cost of being much simpler str uctures; for instance, these compounds typicall y ha ve no or fe w stereocenters and are relati vely flat (see F igure 1). Ov er time, empirical principles w ere developed to guide their design such that certain structural features became associated with inhibiting cer tain classes of tar gets (eg, G-protein coupled receptors) or a f avorable oral absor ption profile (eg, Lipinski’s Rule of 5).10 However, these principles ma y ha ve biased synthetic ef forts to ward cer tain types of str uctures and left other re gions of “chemical space” relatively unexplored.

Known Bioactive Compounds The protein tar gets for most of the compounds in druglike libraries are not kno wn. Because of this, commercially a vailable “kno wn bioacti ve” collections are also commonly screened. These compounds include tool compounds of known mechanism (eg, kinase inhibitors) or actual appro ved drugs. Because of their functional annotation, known bioactive libraries can provide a relatively rapid w ay to identify candidate proteins or pathways that pla y a role in the biolo gical prob lem under investigation.11

Diversity-Oriented Synthesis (DOS) A newer synthetic strategy, termed diversity-oriented synthesis (DOS), seeks to combine the best features of natural products and drug-like molecules.12 In contrast to traditional retrosynthetic analysis (which devises a series of steps to synthesize a unique molecule), DOS chooses

building blocks and reaction conditions to generate a library of diverse compounds with a modest number of steps (eg, approximately three to f ive). Despite their relatively rapid synthesis, DOS libraries are designed to create molecules that are natural product lik e in their complexity, possess multiple stereocenters, are nonplanar, and also include signif icant skeletal and appendage diversity.13–15 (see Figure 1) As the number of a vailable DOS compounds grows, they offer the promise of novel, complex chemical compositions that are, nonetheless, available in numbers and quantities that approach the scale of simpler dr ug-like compounds and can therefore be subjected to large-scale screening. All of the abo ve types of compounds are viab le candidates for screening projects or librar y initiati ves directed at the disco very of ne w imaging probes, with their own advantages and disadv antages depending on the specif ic application. However, the str uctural f ilters that have guided the synthesis of dr ug-like compounds are based on a set of considerations (e g, oral bioa vailability, cell permeability) that may only partially overlap with the considerations inherent in designing an imaging probe. Thus, efforts that use no vel DOS compounds may be particularly exciting going forward.

INCREASED AVAILABILITY OF HIGH-THROUGHPUT SCREENING AND ASSOCIATED ANALYTIC TOOLS Recent y ears ha ve seen high-throughput screening evolve from being the exclusive province of pharmaceutical and biotechnolo gy companies to become a scientific tool increasingly available in academic settings. In addition to a g rowing number of high-throughput screening centers at uni versities and medical schools,

Chemical Biology Approaches to Molecular Ima ging

the NIH Roadmap has funded an intramural screening center (The National Chemical Genomics Center [NCGC]), as well as a consortium of extramural centers that comprise the Molecular Libraries Screening Center Network (). Collecti vely, these and other centers ha ve enabled the application of systematic chemical per turbation to a much broader range of biological problems and diseases than when screening was largely confined to industrial settings. Large-scale chemical screens require robust e xperimental practices, data preprocessing, and ana lysis to distinguish experimental noise from actual screening “positives.”60 Considerations include potential technical artifacts (such as position effects across the wells of a 384well plate), the need for replicate measurements, and assessments of the statistical signif icance of a small molecule’s effect (taking into account multiple h ypothesis testing).16,17 Two publicly accessible resources annotate small molecules using a combination of literature references, structural descriptors, and acquired screening data: PubChem (), administered by the NIH Molecular Libraries Roadmap Initiati ve and ChemBank (), created b y the Chemical Biolo gy Pro gram at the Broad Institute, and w hich also includes planning and disco very tools to facilitate analysis of screening data.

NEW MODES OF TARGET-BASED AND PHENOTYPE-BASED SCREENING Small-Molecule Microarrays The most common paradigm in phar maceutical compound screening is target based. Typically, basic science studies will implicate a protein in disease, and screens will then be designed for compounds that ph ysically interact with the protein or modify its activity or expression in vitro. Recentl y, there is a g rowing literature on the use of small molecule microar rays to rapidl y identify small molecules that bind to a gi ven protein target. In this approach, small molecules are immobilized onto a glass slide at high density (eg, up to 11,000 per slide). These microar rays are then probed with the tar get protein of interest; binding of the target protein to an immobilized small molecule is visualized using a fluorescently labeled primar y or secondar y antibody directed ag ainst t he p rotein ( Figure 2 A). 18,19 This approach ma y be par ticularly w ell suited to identify small molecules that act as tar geting moieties for

499

molecular imaging probes. F or instance, the same molecular handle can be used to immobilize the small molecule to the microar ray and subsequentl y to the surface of the imaging nanoparticle. Thus, the aspect of the small molecule that is displa yed should be reasonably similar in both cases, increasing the likelihood that a binding interaction on the small-molecule microar ray will be recapitulated on a nanopar ticle. Recent studies ha ve sho wn the utility of smallmolecule microarrays created using a v ariety of conjugation chemistries (F igure 2A). F or instance, isocyanate chemistr y is v ery fle xible in its ability to covalently capture a wide v ariety of nucleophilic functional groups, including amines, sulfhydryls, alcohols, and carboxylic acids.20–22 This flexibility allows a wide v ariety of compounds from man y dif ferent sources to be immobilized and screened; it also enables the display of small molecules bearing multiple functional g roups in v arying orientations. In contr ast, fluorous microarrays use the very specific, noncovalent fluorous interaction to immobilize fluorous-tagged small molecules onto a fluoroalk ylsilane surf ace.23–25 An advantage of fluorous microarrays is their ability to display small molecules in a controlled, uniform orientation. Small-molecule microar ray platfor ms ha ve allowed the discovery of novel small-molecule ligands to a v ariety of proteins, including human IgG, 26 transcription factors,27 or that inhibit only one function of a multi-function protein. 28 These and other tar get-based approaches identify small molecules that bind a protein of interest. Complementary techniques, such as Surface Plasmon Resonance, can conf irm a biochemical interaction, and deter mine quantitative parameters for the binding. 19 These smallmolecule candidates can then be conjugated to nanoparticles and tested for their ability to confer tar geting properties.

Cell-Based Phenotypic Screens The chief advantage of a phenotype-based screen is the opportunity to discover new probes that cause interesting biolo gic phenotypes, without requiring a priori knowledge of the tar get proteins that mediate the phenotype. In this respect, phenotype-based screens hearken back to the phenotypic obser vations that spark ed interest in man y impor tant natural products, but in a more systematic and high-throughput manner. Conceptually, phenotypic screens are v ery analo gous to forward genetic screens; instead of a genetic mutation,

500

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

Figure 2. Emerging small-molecule screening methods. A, Small-molecule microarrays. The inset depicts isocyanate- and fluorousbased immobilization methods. Adapted from Duffner JL et al.19 B, Cell-based, phenotypic screens. SM = small molecule; NP = nanoparticle.

however, cells are per turbed using small molecules or some other librar y of probes. This unbiased approach can open ne w a venues of biolo gy, b y re vealing no vel connections betw een small molecules, proteins, and phenotypes. A disadv antage of phenotypic screens, however, is the frequent need to e xpend considerab le effort to identify the targets of the “positives” or “hits” from the screen. This problem of tar get identif ication, discussed fur ther below, can be daunting and is lik ely responsible for the relati vely paucity of phenotypic screening in the phar maceutical industr y. However, as methods for tar get identif ication become more de veloped, the opportunities for biologic discovery have attracted growing interest in phenotypic screening, particularly within academic centers. A typical approach is to plate cells within the wells of a 384-well plate, then add a perturbagen to each well (eg, a small molecule, RN Ai, or a nanopar ticle), and after some period of incubation, some kind of cellular measurement is perfor med (F igure 2B). A v ariety of measurement modalities can be used , including antibody-based in-plate Western b lotting,29 immunofluorescence, luminescence, fluorescent indicator dyes, and gene expression.30 High-content screens use automated microscopy to capture an image of each w ell and then automatically analyze morphologic parameters to quantitate a rich ar ray of cellular phenotypes. 31,32 In the realm of small molecules, phenotypic screening has achieved many successes; a snapshot of acti vity at one screening center circa 2006 has been described by Tolliday and colleagues. 33 An early success w as the compound monastrol, w hich interferes with microtubule

spindle for mation through a no vel mechanism. 34 The mitotic kinesin proteins w ere identif ied as a ne w class of antimitotic drug targets. This vignette illustrates how a phenotypic chemical screen, much lik e a forw ard genetic screen, can re veal no vel roles for proteins in important biological processes.

TARGET IDENTIFICATION The process of identifying the protein target of a probe (eg, a small molecule or an imaging agent) is refer red to as target identification. Target identif ication is commonly required w hen a small molecule is disco vered based on its phenotype, as opposed to through a targetbased approach. A traditional technique is biochemicalaffinity-based purif ication, in w hich the small molecule is immobilized to a solid suppor t, such as agarose beads; cellular e xtracts are passed o ver the beads, and proteins that bind preferentiall y to the immobilized small molecule (as opposed to binding equally w ell to control beads) should be enriched for the protein target. While this approach has successfully identified protein targets of small molecules (especially with natural products of high af finity9), it is, nonetheless, challenging to e xecute successfull y for man y small molecules. This may be in part because the small molecule must remain bound to its protein tar get throughout the af finity purif ication, and man y small molecules possess of f rates that preclude this. Consistent with this, se veral g roups ha ve successfull y executed af finity purif ications w hen the small molecule becomes co valently attached to its tar get, such as

Chemical Biology Approaches to Molecular Ima ging

Control

[13C]Arg

Tryptic digest Mass spec Relative intensity

by using photo-acti vatable benzophenones 35 or b y the opening of an epo xide ring on the small molecule to form a covalent adduct. 36 Genomic-scale anal yses can also help re veal the target of a small molecule. F or instance, Giae ver and colleagues37 used haploinsuf ficiency prof iling in the y east Saccharomyces cer evisiae to identify the proteins that interact with small molecules; the rationale i s t hat, if the gene dosage of a small molecule’s target is decreased from tw o copies to one, the y east should be more sensiti ve to the small molecule. By treating the complete set of heterozygous yeast deletion strains with 10 dif ferent small molecules, they identified both established and novel smallmolecule-protein interactions. 38 More broadl y, genome-wide gene e xpression anal ysis can help identify the proteins modulated b y small molecules. F or instance, Gene Set Enrichment Analysis (GSEA) can determine w hether the changes in gene e xpression induced by a small molecule are statisticall y similar to the gene e xpression patter ns associated with cer tain curated proteins or pathw ays.39,40 Furthermore, the Connectivity Map tool can query whether the changes in gene e xpression induced b y a no vel small molecule (of unknown mechanism) are statistically similar to the expression changes associated with hundreds of small molecules of kno wn mechanism, including FD Aapproved dr ugs.41 This method w as recentl y used to annotate inhibitors of andro gen receptor acti vation (discovered through a phenotypic screen) as heat-shock protein (hsp)90 inhibitors. 42 A p romising a pproach t o t arget i dentification, termed Stable Isotopically Labeled Amino Acids in Cell Culture (SILAC), uses hea vy-isotope incor poration into cellular proteins to increase the signalto-noise ratio of af finity purif ication.43–45 Originally developed to identify protein–protein interactions, this technique is also adaptab le to identify protein par tners of small molecules. In this approach (F igure 3), one cell culture is grown under standard conditions (control culture), whereas another culture is g rown in the presence of [ 13C]-arginine (causing arginine residues to have a molecular w eight that is 6 Da g reater). Cellular extracts are prepared from both control and [ 13C]-arginine cultures. The control e xtract is incubated with unmodified beads; the [ 13C]-arginine e xtract is incubated with beads conjugated to the probe (e g, a small molecule). The eluates from the control and probeconjugated beads are then combined in a 1:1 ratio and subjected to tr yptic digest and mass spectrometr y.

501

Specific Non-specific

m/z

Figure 3. Outline of Stable Isotopically Labeled Amino Acids in Cell Culture (SILAC)-based affinity purification to identify targets of a small molecule. Adapted from Blagoev B et al.44

Arginine-containing peptides will be detected as doublet peaks separated by 6 Da, corresponding to peptides isolated from the cont rol and [ 13C]-arginine extracts. P eptides from proteins that bind nonspecif ically to both control and probe-conjugated beads will be represented as doub lets of equal intensity , whereas peptides that bind specifically to the probe-conjugated bead will show a larger signal at the m/z + 6 peak. This technique sho ws signs of representing a signif icant advance in the area of target identification and may help relie ve a major bottleneck in phenotypic screening.

APPLICATIONS OF CHEMICAL BIOLOGY TO MOLECULAR IMAGING The above advances in chemical biology can impact molecular imaging in se veral impor tant w ays. F irst, novel small-molecule libraries can be a source for novel imaging probes. Although antibodies and peptides ha ve been used as targeting moieties, small-molecule ligands would be adv antageous in ter ms of cost, stability , decreased risk of antigenicity , and increased v alency (ie, ability to conjugate more small molecules per nanoparticle). Fur thermore, phenotype-dri ven screens can be a po werful v ehicle for the disco very of no vel imaging probes (e g, to distinguish cellular states or

502

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 4. Overview of library approaches to discovery of new molecular imaging probes. Small-molecule libraries may first be screened (either in target- or phenotype-based screens); these will identify small molecules that are imaging probes themselves or can act as targeting ligands when coupled to a nanoparticle. Alternatively, a library of small molecule–nanoparticle conjugates can be synthesized and screened.

parse cellular hetero geneity); for man y impor tant unmet imaging challenges, a candidate-tar get approach (ie, choosing a single-cell surf ace receptor and designing a probe against it) is unlik ely to accomplish subtle phenotypic distinctions with adequate target-to-background ratios. Figure 4 illustrates how small-molecule libraries can be screened to identify no vel imaging probes. One approach is to synthesize libraries of nanopar ticles, each decorated by a unique small molecule, and then screen these nanopar ticle libraries. Another is to f irst screen the small-molecule librar y for promising compounds and then to convert the “hits” from these screens into imaging probes, either as tar geting ligands for a nanoparticle or as unador ned small molecules. The following sections will describe recent pro gress along these paths.

SYNTHESIS OF SMALL MOLECULE–NANOPARTICLE LIBRARIES Efficient creation of a librar y of small molecule– nanoparticle conjugates raises se veral considerations. First, the coupling chemistry should be compatible with aqueous en vironments to a void nanopar ticle precipitation. Second , the conjugation and subsequent purification methods should be amenable to at least moderate-throughput protocols (e g, perfor ming reactions in a 96-w ell plate). Third, a wide v ariety of compounds should be accessible to conjugation through a variety of functional g roups.

Conjugation Chemistries 46 Weissleder and colleagues have repor ted a 146-member librar y in w hich a v ariety of commercially a vailable small molecules are conjugated to magneto-fluorescent nanoparticles. The starting material nanopar ticle consists of a magnetic nanopar ticle where surf ace de xtrans w ere cross-link ed with epichlorhydrin ( CLIO) and reacted with amm onia to yield primar y amines (CLIO-NH 2). F ollowing attachment of FITC moieties, these primar y amines were then conjugated to small molecules bearing a variety of functional g roups: anhy drides, amines, hydroxyls, carbo xylic acids, thiols, and epo xides (Figure 5A). 46,47 Since many small molecules may possess more than one functional g roup, a general conjugation approach would use a single functional group, without modifying other parts of the molecules. One solution is the use of the Huisgen 1,3-dipolar cycloaddition, perhaps the most famous of the “click” chemistry reactions.48 In this reaction, a ter minal azide g roup reacts with a ter minal alkyne g roup on another molecule in the presence of Cu(I), resulting in the formation of the 1,4-triazole adduct with high re gioselectivity. Sun and colleagues 49 have demonstrated the feasibility of appl ying this reaction to nanopar ticle imaging probes b y introducing either azide or alkyne moieties onto the dextran surface of magnetic nanoparticles. To accomplish this, they first converted the amines on CLIO-NH 2 to carboxylic acids by reacting with succinic anh ydride. CLIO-COOH w as conjugated to propargylamine by EDC/NHS coupling to

Chemical Biology Approaches to Molecular Ima ging

O

A R2

O O

R1

H N

R1

O

O

H N

R-OH

O

H N

R

R O

EDC/NHS H N

R-SH

R

O

SIA/SOCI2

R-CO2H

H N

O

EDC/NHS

NH2

O

H N

R-NH2

S

S

R

O

SPDP O

OH

H N

R

R

O

O

B B

H2N

503

N3

N H

N3

R Cu(I)

CO2H H2N

N H

N N N

R

O

O N H

R

N3

Cu(I)

N H

N

N N R

Figure 5. Conjugation strategies for coupling small molecules to iron oxide nanoparticles (CLIO). A, Conjugation to amine-functionalized CLIO using multiple different functional groups. Adapted from Sun EY et al.47 B, Conjugation to carboxyl-functionalized CLIO using Huisgen 1,3-dipolar cycloaddition. Adapted from Sun EY et al.49

form alkyne-substituted CLIO; similarl y, CLIO-COOH was conjugated to 3-azido prop ylamine b y EDC/NHS coupling to for m azido-substituted CLIO (F igure 5B). Alkyne- or azido- substituted CLIO were then “clicked” with biotin, fluorochromes, and natural products bearing the complementar y alk yne or azide moiety in the presence of Cu(I); the reactions proceeded nearl y quantitatively, regardless of whether the CLIO nanoparticle bore the alkyne or azide moiety.49

SCREENING OF SMALL MOLECULE–NANOPARTICLE LIBRARIES The 146-member small molecule–nanopar ticle librar y synthesized b y Weissleder and colleagues 46 was subjected to se veral phenotype-based in vitro screens to test if the conjugated small molecules could modulate the localization and uptak e of the nanopar ticles (the starting nanoparticle is typically taken up primarily by monocytes/ macrophages). Each of these nanopar ticles

possesses a small number of FITC moieties conjugated to its surf ace to enable its use in imaging applications. Cell lines w ere plated in the w ells of a 96-w ell plate, and a different small molecule–nanoparticle conjugate was added to each w ell. F ollowing incubation and washing, nanopar ticle uptak e w as quantitated b y an immunoassay for nanopar ticle-borne FITC present in each well. For instance, they screened for nanoparticle conjugates that were specif ic for the activation state of human macrophages, whereas the starting nanoparticle was tak en up equall y w ell b y resting and acti vated macrophages; the y identif ied a small-molecule modification that led to uptake specifically in activated macrophages but not in resting macrophages, as well as a distinct small molecule that led to uptake specifically in resting macrophages but not in acti vated macrophages46 (Figure 6A). They also identif ied a small molecule– nanopar ticle conjugate that confer red preferential uptak e into pancreatic adenocarcinoma cells in vitro; this f inding was conf irmed in vivo when

504

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Activated

Resting GMCSF

oxLDL

LPS

A control NP

O

O

O

O O

O

O

H2N

CO2H

B O

control NP

O N H

C

O

O O

control NP N H

O

Figure 6. Small molecule–nanoparticle conjugates with novel targeting properties. A, Immunofluorescence showing uptake of small molecule–nanoparticle conjugates specifically into either resting or activated macrophages (by GMCSF, oxidized LDL, or lipopolysaccharide). Control NP = CLIO-NH2, which is taken up equally well by all macrophages; other rows are CLIO-NH2 conjugated to the depicted small molecules. oxLDL = oxidized LDL; LPS = lipopolysaccharide. B, Increased uptake of small molecule–nanoparticle conjugate (CLIO-isatoic) into bilateral pancreatic adenocarcinoma xenografts in nude mice, relative to uptake of control NP. Left panel: Cy3.5 channel image to visualize uptake of control nanoparticle (CLIO-Cy3.5); Right panel: Cy5.5 channel image to visualize uptake of CLIO-isatoic-Cy5.5 (structure of isatoic anhydride is shown). C, Fluorescence micrographs of tumor explants following nanoparticle injection. Left panel: Cy3.5 channel image to visualize uptake of control nanoparticle (CLIO-Cy3.5); Right panel: Cy5.5 channel image to visualize uptake of CLIO-isatoic-Cy5.5 (structure of isatoic anhydride is shown). Adapted from Weissleder R et al.46

the intra venously injected probe localized to subcutaneous tumor xenografts in nude mice46 (Figure 6B). These f indings vi vidly illustrate the po wer of phenotypic screens to disco ver imaging probes with medically useful discriminating properties; a target-oriented approach toward the same tar geting goals would be limited b y the lack of suitab le candidate tar gets. Subsequent experiments will be required to identify the cellular mechanisms that are mediating this selecti ve targeting and may also shed light on the basic biolo gical m echanisms u nderlying t hese d iseased c ellular states. Future screening ef forts will undoubtedl y benef it from a number of other cell-based measurements. F or instance, automated microscop y and image anal ysis

could assess nanopar ticle uptak e, subcellular localization, and other quantitative morphologic phenotypes. Such a screen could re veal insights not onl y into nanoparticle tar geting b ut also pro vide a vie w into pleiotropic effects that a nanopar ticle may have upon a cell, some of them potentially toxic. Hogemann and colleagues50 have also demonstrated the feasibility of screening for the cellular uptak e of super paramagnetic nanoparticles using magnetic resonance imaging. In this approach, the uptak e of nanopar ticles into cells results in an increased iron concentration in the well, which decreases the T 2 relaxation time; the authors demonstrated that the T2 times observed were reproducible and related closel y to the iron concentration (and thus the extent of nanoparticle internalization) in each well.

Chemical Biology Approaches to Molecular Ima ging

SYNTHESIS AND SCREENING OF SMALL-MOLECULE IMAGING PROBES Rosania and colleagues51 used a combinatorial approach to synthesize a librar y of 574 compounds based on styryl dyes, which are fluorescent, lipophilic cations; of these, 276 were fluorescent, with emission wavelengths ranging from approximately 450 to 700 nm. Strikingly, when the fluorescent compounds were screened in a human melanoma cell line, 119 of the 276 fluorescent compounds sho wed a distincti ve cellular localization pattern. Most of the compounds localized to the mitochondria; other sites included the endoplasmic reticulum, cytoplasm, nuclei, and nucleoli.51 A related library of 855 styryl compounds yielded eight small molecules with strong nuclear staining; one compound’ s fluorescent yield increases up to 13-fold upon binding to double-stranded DNA, suggesting that it acts as a fluorescent sensor for doub le-stranded DNA.52 Still, other compounds using the styryl scaffold act as RN A sensors.53 Most interestingl y, neutral v ersions of these styryl libraries yielded small molecules that bind β-amyloid plaques in vitro (ie, in human Alzheimer’s disease brain sections) 54 but can also penetrate the blood-brain barrier in vivo to target amyloid plaques in murine models of Alzheimer’s disease. 55

GENERAL APPROACHES TO CELL SURFACE LABELING As a complement to design probes that are tak en up by specif ic cell types, a v ery promising a venue of research is the direct, specif ic chemical labeling of cell-surface proteins. The considerations in volved in site-specifically labeling li ving cells ha ve been reviewed recently56 and include (1) labels that are sufficiently small so as to not disrupt cellular phenomena; (2) the generalizability of a labeling approach to incorporate multiple imaging modalities; (3) the e xistence of or thogonal chemistries to allo w multiple xed imaging in a single e xperiment; (4) persistent labeling to allow for extended dynamic observations. Even mainstays such as g reen fluorescent protein (GFP) or other expressed repor ter genes are not ideal labeling agents as the y are relati vely lar ge (e g, GFP is comprised of 238 amino acids) and limited to a single imaging modality. Ting and colleagues ha ve pioneered an approach to labeling mammalian cell surf ace proteins that tak es adv antage of Escherichia coli (E. coli )

505

peptide ligases and their unique shor t acceptor peptides. F or instance, the surf ace protein to be labeled can be e xpressed as a fusion protein with a 15 amino acid reco gnition sequence or acceptor peptide for E. coli biotin ligase (BirA). This peptide sequence is absent from mammalian genomes, allowing highly specific labeling of the tar geted surf ace protein upon the addition of BirA activity to the tissue culture medium. This system has been used to conjugate surface proteins with biotin (the endo genous substrate for BirA), which can then interact with streptavidin conjugated to a quantum dot57 (Figure 7A). Two modifications of this procedure h ave f urther i ncreased i ts g enerality. The same g roup found that BirA can a lso couple a ketone isostere o f b iotin t o i ts a cceptor p eptide, w hich c an then be used to conjugate a wide v ariety of hydrazideor h ydroxylamine-functionalized probes. 58 Furthermore, the use of E. coli lipoic acid ligase (LplA) allows conjugation of an alk yl azide (an unnatural substrate) to its cognate acceptor peptide; the alkyl azide can then react with imaging probes containing a c yclooctyne moiety.59 The BirA and LplA systems are or thogonal, allowing simultaneous live-cell labeling of two surface 59 proteins (one using BirA, the other using LplA). Taken together, this body of work provides a generalizable approach to co valently label specif ic cell surf ace proteins with a v ariety of imaging probes and enab le live-cell imaging with e xcellent tar get-to-background ratios and minimal physiologic disruption due to small imaging probes.

CONCLUSIONS Chemical biolo gy is increasingl y being used to interrogate biolo gical prob lems and mak e no vel connections between small molecules, genes, and phenotypes. At the same time, molecular imaging is being called upon to perfor m increasingl y subtle phenotypic distinctions. The emerging ability to synthesize libraries of novel complex structures and rapidly screen them for desired cellular phenotypes appears w ell suited to catalyze no vel solutions to impor tant needs in molecular imaging. Notwithstanding the successes discussed here, the application of chemical biology to imaging is still in its infancy. A critical challenge for the next several years will be w hether chemical approaches can signif icantly impact clinical molecular imaging, such as pro viding imaging sur rogates for genotype-phenotype cor relations, disease status, or treatment response.

506

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

O HN H

O

A

HN NH H H

SA NH H

HN H

NH H S

S CO2H

S

O

SA

Biotin BirA

H N

O

N

O H

B

O H

H

H

S H2N CO2H

S

S

O

H

H

N H

BirA

O

C

N3 7

O HO

NH

N N N

O NH 7

N3 7

LplA

D

Cy3 (LAP–LDLR)

Alexa Fluor 488 (AP–EphA3)

Alexa 488/Cy3 ratio 3.0 1.5

Figure 7. Site-specific labeling of cell surface proteins. A, Conjugation of biotin to the biotin ligase (BirA) acceptor peptide (red) by BirA; interaction with a streptavidin-quantum dot conjugate then links a quantum dot to the cell surface protein (blue). Adapted from Howarth M et al.57 B, Conjugation of a ketone isostere of biotin to the BirA acceptor peptide (red) by BirA; the ketone can then react with a hydrazide to link an imaging probe (green) to the cell surface protein (blue). Adapted from Chen I et al.58 C, Conjugation of an alkyl azide to the lipoic acid ligase (LplA) acceptor peptide (orange); the azide can then react with a cyclooctyne to link an imaging probe (green) to the cell surface protein (blue). Adapted from Fernandez-Suarez M et al.59 D, Orthogonal labeling by the BirA and LplA systems in the same cells. Left panel: LDL-receptor labeled with Cy3 using LplA; Middle panel: EphA3 labeled with Alexa Fluor 488 using BirA/biotin; Right panel: intensity ratios of Alexa 488 and Cy3 signals. Adapted from Fernandez-Suarez M et al.59

REFERENCES 1. Binns W, James LF, Shupe JL, Everett G. A congenital cyclopian-type malformation in lambs induced b y maternal ingestion of a range plant, veratrum californicum. Am J Vet Res 1963;24:1164–75. 2. Keeler RF, Binns W. Teratogenic compounds of Veratrum californicum (Durand). V. Comparison of c yclopian effects of steroidal alkaloids from the plant and str ucturally related compounds from other sources. Teratology 1968;1:5–10.

3. Cooper MK, P orter JA, Young KE, Beach y PA. Teratogen-mediated inhibition of tar get tissue response to Shh signaling. Science 1998;280:1603–7. 4. Chen JK, Taipale J, Cooper MK, Beachy PA. Inhibition of Hedgehog signaling by direct binding of cyclopamine to Smoothened. Genes Dev 2002;16:2743–8. 5. Taipale J, Chen JK, Cooper MK, et al. Effects of oncogenic mutations in Smoothened and patched can be re versed b y c yclopamine. Nature 2000;406:1005–9.

Chemical Biology Approaches to Molecular Ima ging

6. Liu J, Farmer JDJ, Lane WS, et al. Calcineurin is a common target of cyclophilin-cyclosporin A and FKBP-FK506 complexes. Cell 1991;66:807–15. 7. Brown EJ, Albers MW, Shin TB, et al. A mammalian protein targeted by G1-arresting rapamycin-receptor complex. Nature 1994;369:756–8. 8. Sabatini DM, Erdjument-Bromage H, Lui M, Tempst P, Snyder SH. RAFT1: a mammalian protein that binds to FKBP12 in a rapamycin-dependent fashion and is homolo gous to y east TORs. Cell 1994;78:35–43. 9. Taunton J, Hassig CA, Schreiber SL. A mammalian histone deacetylase related to the y east transcriptional re gulator Rpd3p. Science 1996;272:408–11. 10. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Experimental and computational approaches to estimate solubility and per meability in drug discovery and development settings. Adv Drug Deliv Rev 2001;46:3–26. 11. Root DE, Flaher ty SP, Kelley BP, Stockwell BR. Biolo gical mechanism prof iling using an annotated compound librar y. Chem Biol 2003;10:881–92. 12. Schreiber SL. Target-oriented and di versity-oriented organic synthesis in drug discovery. Science 2000;287:1964–9. 13. Burke MD, Schreiber SL. A planning strate gy for di versity-oriented synthesis. Angew Chem Int Ed Engl 2004;43:46–58. 14. Tan DS. Di versity-oriented synthesis: e xploring the intersections between chemistry and biology. Nat Chem Biol 2005;1:74–84. 15. Bender A, Fergus S, Galloway WR, et al. Diversity oriented synthesis: a challenge for synthetic chemists. Er nst Schering Res F ound Workshop 2006:47–60. 16. Eli Lilly & Company, NIH Chemical Genomics Center . Assay Guidance Manual Version 4.1. 2005. 17. Malo N , Hanle y J A, Cer quozzi S, P elletier J , Nadon R. Statistical practice in high-throughput screening data anal ysis. Nat Biotechnol 2006;24:167–75. 18. MacBeath G, Koehler AN, Schreiber SL. Printing small molecules as microarrays and detecting protein-ligand interactions en masse. J Am Chem Soc 121:7967–8. 19. Duffner JL, Clemons PA, Koehler AN. A pipeline for ligand disco very using small-molecule microarrays. Curr Opin Chem Biol 2007;11:74–82. 20. Bradner JE, McPherson OM, Mazitschek R, et al. A robust smallmolecule microar ray platfor m for screening cell l ysates. Chem Biol 2006;13:493–504. 21. Kurosu M, Mowers WA. Small-molecule microarrays: development of novel linkers and an efficient detection method for bound proteins. Bioorg Med Chem Lett 2006;16:3392–5. 22. Bradner JE, McPherson OM, Koehler AN. A method for the covalent capture and screening of diverse small molecules in a microar ray format. Nat Protoc 2006;1:2344–52. 23. Ko KS, Jaipuri F A, Pohl NL. Fluorous-based carboh ydrate microarrays. J Am Chem Soc 2005;127:13162–3. 24. Vegas AJ, Bradner JE, Tang W, et al. Fluorous-based small-molecule microar rays for the disco very of histone deacetylase inhibitors. Angew Chem Int Ed Engl 2007;46:7960–4. 25. Nicholson RL, Ladlo w ML, Spring DR. Fluorous tagged small molecule microarrays. Chem Commun (Camb) 2007:3906–8. 26. Uttamchandani M, Walsh DP, Khersonsky SM, et al. Microar rays of tagged combinatorial triazine libraries in the disco very of smallmolecule ligands of human IgG. J Comb Chem 2004;6:862–8. 27. Koehler AN, Shamji AF, Schreiber SL. Discovery of an inhibitor of a transcription factor using small molecule microar rays and di versity-oriented synthesis. J Am Chem Soc 2003;125:8420–1. 28. Kuruvilla FG, Shamji AF, Sternson SM, et al. Dissecting glucose signalling with di versity-oriented synthesis and small-molecule microarrays. Nature 2002;416:653–7. 29. Koeller KM, Haggarty SJ, Perkins BD, et al. Chemical genetic modifier screens: small molecule trichostatin suppressors as probes of

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45. 46.

47.

48.

49. 50.

507

intracellular histone and tubulin acetylation. Chem Biol 2003; 10:397–410. Stegmaier K, Ross KN , Cola vito SA, et al. Gene e xpression-based high-throughput screening(GE-HTS) and application to leuk emia differentiation. Nat Genet 2004;36:257–63. Yarrow JC, P erlman ZE, Westwood NJ , Mitchison TJ. A highthroughput cell mig ration assa y using scratch w ound healing, a comparison of image-based readout methods. BMC Biotechnol 2004;4:21. Perlman ZE, Mitchison TJ, Ma yer TU. High-content screening and profiling of drug activity in an automated centrosome-duplication assay. Chembiochem 2005;6:145–51. Tolliday N, Clemons PA, Ferraiolo P, et al. Small molecules, big players: the National Cancer Institute’s Initiative for Chemical Genetics. Cancer Res 2006;66:8935–42. Mayer TU, Kapoor TM, Haggarty SJ, et al. Small molecule inhibitor of mitotic spindle bipolarity identif ied in a phenotype-based screen. Science 1999;286:971–4. Smukste I, Bhalala O , Persico M, Stockw ell BR. Using small molecules to o vercome dr ug resistance induced b y a viral onco gene. Cancer Cell 2006;9:133–46. Sin N, Meng L, Wang MQ, et al. The anti-angiogenic agent fumagillin covalently binds and inhibits the methionine aminopeptidase, MetAP-2. Proc Natl Acad Sci U S A 1997;94:6099–103. Giaever G, Shoemak er DD, Jones TW, et al. Genomic prof iling of drug sensiti vities via induced haploinsuf ficiency. Nat Genet 1999;21:278–83. Giaever G, Flaher ty P, K umm J , et al Chemo genomic prof iling: identifying the functional interactions of small molecules in yeast. Proc Natl Acad Sci U S A 2004;101:793–8. Mootha VK, Lindg ren CM, Eriksson KF , et al. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet 2003;34:267–73. Subramanian A, Tamayo P, Mootha VK, et al. Gene set enrichment analysis: a kno wledge-based approach for inter preting genomewide e xpression prof iles. Proc Natl Acad Sci U S A 2005; 102:15545–50. Lamb J, Crawford ED, Peck D, et al. The Connectivity Map: using gene-expression signatures to connect small molecules, genes, and disease. Science 2006;313:1929–35. Hieronymus H, Lamb J , Ross KN , et al. Gene e xpression signaturebased chemical genomic prediction identif ies a no vel class of HSP90 pathway modulators. Cancer Cell 2006;10:321–30. Ong SE, Blagoev B, Kratchmarova I, et al. Stable isotope labeling by amino acids in cell culture, SILA C, as a simple and accurate approach to e xpression proteomics. Mol Cell Proteomics 2002; 1:376–86. Blagoev B, Kratchmarova I, Ong SE, et al. A proteomics strate gy to elucidate functional protein-protein interactions applied to EGF signaling. Nat Biotechnol 2003;21:315–8. Ong SE, Mann M. A practical recipe for stab le isotope labeling b y amino acids in cell culture (SILAC). Nat Protoc 2006;1:2650–60. Weissleder R, K elly K, Sun EY , Shtatland T, Josephson L. Cellspecific tar geting of nanopar ticles b y multi valent attachment of small molecules. Nat Biotechnol 2005;23:1418–23. Sun EY , Josephson L, K elly KA, Weissleder R. De velopment of nanoparticle libraries for biosensing. Bioconjug Chem 2006; 17:109–13. Kolb HC, Finn MG, Sharpless KB. Click Chemistry: Diverse Chemical Function from a F ew Good Reactions. Angew Chem Int Ed Engl 2001;40:2004–21. Sun EY, Josephson L, Weissleder R. “Clickable” nanoparticles for targeted imaging. Mol Imaging 2006;5:122–8. Hogemann D , Ntziachristos V, Josephson L, Weissleder R. High throughput magnetic resonance imaging for e valuating tar geted nanoparticle probes. Bioconjug Chem 2002;13:116–21.

508

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

51. Rosania GR, Lee JW , Ding L, Yoon HS, Chang YT. Combinatorial approach to or ganelle-targeted fluorescent librar y based on the styryl scaffold. J Am Chem Soc 2003;125:1130–1. 52. Lee JW, Jung M, Rosania GR, Chang YT. Development of no vel cellpermeable DNA sensiti ve dy es using combinatorial synthesis and cell-based screening. Chem Commun (Camb) 2003;1852–3. 53. Li Q, Kim Y, Namm J, et al. RNA-selective, live cell imaging probes for studying nuclear structure and function. Chem Biol 2006; 13:615–23. 54. Li Q, Lee JS, Ha C, et al. Solid-phase synthesis of styr yl dyes and their application as am yloid sensors. Angew Chem Int Ed Engl 2004;43:6331–5. 55. Li Q, Min J , Ahn YH, et al. Styr yl-based compounds as potential in vivo imaging agents for beta-am yloid plaques. Chembiochem 2007;8:1679–87.

56. Chen I, Ting AY. Site-specif ic labeling of proteins with small molecules in live cells. Curr Opin Biotechnol 2005;16:35–40. 57. Howarth M, Takao K, Hayashi Y, Ting AY. Targeting quantum dots to surface proteins in living cells with biotin ligase. Proc Natl Acad Sci U S A 2005;102:7583–8. 58. Chen I, Howarth M, Lin W, Ting AY. Site-specific labeling of cell surface proteins with bioph ysical probes using biotin ligase. Nat Methods 2005;2:99–104. 59. Fernandez-Suarez M, Bar uah H, Mar tinez-Hernandez L, et al. Redirecting lipoic acid ligase for cell surf ace protein labeling with small-molecule probes. Nat Biotechnol 2007;25:1483–7. 60. Seiler KP, George GA, Happ MP, et al. ChemBank: a small-molecule screening and cheminfor matics resource database. Nucleic Acids Res 2008:36(database issue): D351-9. Epub 2007 Oct 18.

33 THERANOSTICS: AGENTS FOR DIAGNOSIS AND THERAPY JASON R. MCCARTHY, PHD

The combination of diagnostic and therapeutic entities into one dr ug deli very v ehicle yields agents that are capable of the simultaneous diagnosis and treatment of disease. These theranostic agents ha ve unique applications as they not only promote the diagnosis and therapy of disease but also allo w for feedback mechanisms to determine the localization, release, and therapeutic ef ficacy of treatments. Theranostic nanoagents further allo w for the incor poration of multiple functionalities within one particle. These agents modify the pharmacokinetics of the incor porated moieties, allo wing for the creation of drug-release systems based upon the local environment or material composition. Finally, theranostic agents allo w for detailed studies of the biodistribution of therapeutics in vi vo using numerous imaging modalities. Although there are a number of adv antages to the use of theranostic agents, there are also se veral challenges. F or e xample, multifunctional platfor ms will only be applicable in cer tain circumstances as there is no need to administer a diagnostic moiety every time a patient receives a therapeutic drug. In addition, there is often a signif icant mismatch betw een the dose required for diagnosis and therap y, with relevant therapeutic doses often being higher. Although this field is still in its inf ancy, it is clear that theranostics of fer unique capabilities and their applications require further study. This chapter reviews the different types of theranostic agents and pro vides a summar y of the adv antages and challenges of each type. Specific examples of the different platforms will be dis cussed. In addition, the rele vant chemistry involved in the synthesis and tar geting of these agents will be reviewed.

THERANOSTIC ANTIBODIES Antibodies have been labeled with a v ariety of imaging agents primaril y to deter mine the phar macokinetics of the antibody or to detect disease. Clinicall y, this paradigm is impor tant, especiall y in the deter mination of receptor expression in cancers, as the presence of specific receptors dictates the treatment options a vailable. Therapeutic antibodies, such as trastuzumab and cetuximab, are useful as an adjuv ant therap y in the treatment of cancers o verexpressing members of the epider mal growth factor receptor f amily (Her2/ neu and EGFR) as they specifically target their respective receptors. Antibody targeted imaging agents are discussed in detail (see Chapter 25, “Targeted Antibodies and Peptides”). The chemistries in volved in the functionalization of antibodies are well established.1 Often, they are labeled b y reaction of succinimidyl esters or anh ydrides of the imaging ligand, nonspecif ically labeling an y nonsterically hindered free amines. Alternately, free amines can be converted to sulfh ydryl g roups b y reaction with Traut’s reagent (2-iminothiolane) or N-succinimidyl-S-acetylthioacetate. These thiols, as w ell as nati ve free thiols within the antibody , can then be reacted with maleimide- or iodoacetyl-containing ligands. Sampath and colleagues 2 have reported the synthesis of a dual-labeled trastuzumab incorporating both a γ-emitter and a near -infrared (NIR) fluorophore. The antibody was initiall y reacted with dieth ylenetriaminepentaacetic acid (DTPA) dianhydride, a chelator for 111In, followed by conjugation of IRDy e 800. In vitro anal ysis of the antibody conjugate in cells e xpressing high le vels of Her2 (SKBr3) showed a significant binding of the construct versus cells e xpressing low levels of Her2 (MD A-MB-231).

509

510

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Importantly, this binding could be abro gated by pretreatment of the SKBr3 cells with trastuzumab . The dual modality nature of the conjugate w as sho wn in vi vo in mice inoculated with SKBr3 tumors. Fluorescence reflectance imaging (FRI) clearl y shows signif icant accumulation of the agent within the tumor 48 hours after administration (Figure 1). As was seen in the in vitro studies, blocking of the Her2 receptors b y pretreatment with trastuzumab (200-fold e xcess) reduced the fluorescence signal from the tumor . Single photon emission computed tomography (SPECT)/computed tomography (CT) images of the mice also sho wed increased uptak e of the agent in the tumor in relation to the adjacent muscle (tumor to muscle ratio = 2.7). These results were also correlated ex vivo and histolo gically, with both the fluorescent and radiochemical signals colocalizing with the diseased tissues. Cetuximab (Erbitux) binds with high affinity to EGFR and causes cell-cycle arrest in the G1 phase. In doing so, it decreases angiogenesis and cellular adhesion and inhibits matrix metalloproteinases (MMPs) responsib le for metastases.3 Two simple conjugates of the antibody have recently been repor ted, with one featuring a radionuclide chelate 4 and the other a fluorophore. 5 The conjugation of 1,4,7,10tetraazadodecane-N,Nʹ′,Nʹ′ʹ′,Nʹ′ʹ′ʹ′-tetraacetic acid (DOTA) to the antibody w as accomplished b y activation of DO TA with N-hydroxysulfosuccinimide (sulfo-NHS) in the presence of 1-eth yl-3-(3-dimethylaminopropyl)carbodiimide (EDC) follo wed b y reaction with cetuximab . F ollowing conjugation, the chelator w as labeled with 64Cu. The avidity of this agent for se veral xenograph tumor models

A

C

B 1 2

1 2

3

3

F

E

D

was ascertained and cor related with w estern blot analysis of receptor e xpression. As e xpected, tumors e xpressing high le vels of EGFR (U87MG and PC-3) displa yed similarly high accumulation of the agent, as determined by planar positron emission tomo graphy (PET) imaging, whereas those with lo w e xpression (SW620 and MD AMB-435) showed minimal uptake. Similar conjugates featuring 99mTc,6 89Zr,7 and 68Ga8 have also been reported. Optically labeled cetuximab has also been investigated, with the intent of using the agents intraoperatively, allo wing for dif ferentiation of health y and diseased tissues.5,9 In one particular embodiment, Hama and colleagues9 synthesized an acti vatable conjugate, based upon a tw o-part scheme in w hich a biotin-modif ied cetuximab is injected and allo wed to localize in areas overexpressing EGFR. This is follo wed b y injection of neutravidin-modified BODIPY -FL, w hich binds selectively to the antibody and “tur ns on.” The ability to visualize cancer foci w as tested using this strate gy in vi vo. Athymic nude mice w ere inter peritoneally injected with A431 cells 10 days prior to administration of the biotinylated cetuximab. Aggregated tumors were clearly visible by FRI in the e xposed peritoneal cavity, whereas smaller foci were visible upon closer examination. Barrett and colleagues 10 have also repor ted on the utility of antibody cocktails in identifying distinct EGFR subtypes in vi vo using fluorescence imaging. In these experiments, cetuximab and trastuzamab w ere modif ied with NIR dy es Cy5.5 and Cy7, respecti vely, and w ere injected into nude mice bearing A431 (EGFR o verexpressing) and NIH3T3/HER2 + (Her2 o verexpressing) tumors. Twenty-four hours after injection of the antibodies, the mice were imaged using FRI. The different tumor types w ere readil y identif ied using spectral unmixing, illustrating that it is possib le to detect and dif ferentiate receptor expression using optical imaging, a feat which is not possible with radionuclide imaging.

1 2

1 2

1 2

3

3

3

Figure 1. SKBr3-luc cancer cells were inoculated into left flank of athymic nu/nu mice. A, Bioluminescence image overlaid on whitelight image shows the presence of SKBr3-luc tumor xenograft near left flank of nu/nu mouse. B–F, White-light image (B) of mouse taken before fluorescence imaging of the trastuzumab conjugate (C) 200fold molar excess of trastuzumab followed by injection of antibody conjugate (D) control, not targeted conjugate (E) or equivalent dose of IRDye 800CW (F) 48 h after administration. Reproduced with permission from Sampath et al.2

LIGHT-ACTIVATED THERAPEUTICS Light-activated therapies, including photodynamic therapy (PDT), have been used for the treatment of numerous cancers, age-related macular de generation, and der matological conditions. 11 The under pinning mechanism of action is that they generate cytotoxic singlet oxygen upon irradiation with the appropriate w avelength of light and are thus innocuous in its absence. As light can be focused at the site of interest, adjacent healthy tissues, even those that may contain the agent, are spared. The mechanism of the therapeutic ef fect is v aried and depends on the PDT

Theranostics: Agents for Diagnosis and Therapy

agent used. Direct cell killing b y the light-acti vated agents is a result of mitochondrial damage caused b y reactive oxygen species and results in both apoptotic and necrotic pathways. Loss of vascular patency as a result of therapy is also seen as an indirect method b y which cells are killed.12 A subtype of light-acti vated therapies are those that are excited by and emit near-infrared light (near-infrared light activated therapeutics, NILAT) and are thus par ticularly useful in imaging applications. Two different strategies are cur rently being pursued for their utility in theranostic applications: (a) the use of a single NILAT for both diagnosis and therap y and (b) decoupled use of separate diagnostic and therapeutic moieties on the same delivery v ehicle. A number of por phyrinic macroc ycles have sho wn adv antageous photoph ysical proper ties and have been e xploited to deter mine the agent localization prior to light therapy.13 In the cases of Barrett’s esophagus and nonmuscle invasive bladder cancer, NILAT are often administered 24 to 48 hours prior to therap y to allow for localization in neoplastic tissues. On the day of the treatment, the agent is detected using fluorescence endoscopy, which f acilitates light deli very to the diseased tissues. The ability of these molecules to act as fluorophores is discussed in more detail in Chapter 27, “Optical Imaging Agents.” The intrinsic theranostic proper ties of NILAT have recently been advanced by the development of constructs that are nonfluorescent and nonphototoxic until activated by localization within cer tain en vironments or b y the presence of certain enzymes. In one recent nanopar ticulate system, the NILA T, meso-tetraphenylporpholactol, is encapsulated within pol y (lactic-co-gl ycolic acid) nanoparticles.14 Upon encapsulation, the e xcited states of the chromophore are quenched , rendering it nonfluorescent and nonphototoxic. When the par ticles are incubated with lipid or cells, the NILA T is released and regains its e xcited state photoph ysical proper ties. The fluorescence and phototo xicity of the par ticles sho wed time-dependent acti vation, w hich w as used to treat a murine model of cancer . Following light treatment, all mice used in the study sho wed complete eradication of the tumors, with no residual-disease presence 1 month after therapy. In one other e xample, Zheng and colleagues 15 have synthesized a peptide-based constr uct containing a NILAT, p yropheophorbide a, a quencher moiety , b lack hole quencher 3, and an enzyme clea vable peptide substrate recognized by matrix metalloproteinase 7 (MMP7). MMPs are an impor tant class of enzymes, as the y are upregulated in a number of diseases, such as cancer ,

511

atherosclerosis, and pulmonary disorders. Upon synthesis of the constr uct, the e xcited states of the NILA T are quenched b y fluorescence resonance ener gy transfer (FRET) to the quencher molecule. When treated with the enzyme, the peptide sequence is cleaved, and the authors noted a 12-fold increase in fluorescence and a 19-fold increase in singlet o xygen generation. This effect can be ablated in the presence of an MMP inhibitor . This theranostic agent w as also tested in vi vo in mice bearing KB tumors. Within 3 hours of injection, the agent w as localized and acti vated within the tumor , as deter mined by FRI. F ollowing therapeutic ir radiation, the tumor regressed in size and was not present after 30 days. One of the main drawbacks of using NILATs for both imaging and therap y is that singlet o xygen is generated concurrent with fluorescence emission. This can result in extraneous photosensiti vity in nontar get tissues due to imperfect tar geting. To circumv ent this, nanopar ticulate constructs incorporating NILAT and another imaging moiety ha ve been constr ucted. One e xample w as recentl y described b y us, in w hich a potent NILA T, 5-(4-carboxyphenyl)-10,15,20-triphenyl-2,3-dihydroxychlorin (TPC), was conjugated to a super paramagnetic iron oxide nanoparticle coated with epichlorohydrin cross-linked dextran (F igure 2A). 16 In addition, AlexaFluor 750 (AF750) was also conjugated to the par ticle to enable NIR fluorescence imaging of the localization of the agent at a w avelength spectrally distinct from the NILA T. These particles are expected to be useful in the treatment of atherosclerotic vascular disease as the y show a high a vidity for acti vated macrophages (as opposed to all macrophages or resting macrophages), which are abundant in vulnerable, inflamed atherosclerotic lesions. The cellular uptak e and phototo xicity of the par ticles were sho wn in RA W 264.7 murine macrophages. Upon incubation, the cells inter nalized the par ticles in a timedependent f ashion, as deter mined b y flo w c ytometry and fluorescence microscopy. When irradiated at the therapeutic wavelength (650 nm), a dose-dependent phototo xicity w as obser ved. Ongoing in vi vo e xperiments in apolipoprotein E def icient mice that spontaneously develop atherosclerotic lesions ha ve sho wn that the agent preferentially accumulates within atheromata, as determined by intravital fluorescence microscop y (IVFM, F igure 2B). Light therapy of the lesion results in massi ve macrophage death within 24 hours of therap y. At 1 w eek post-therapy, reinjection of the agent results in minimal uptake within the plaque, hinting at reduced macrophage content (Figure 2C), whereas at 3 w eeks post-therapy, histologic examination of the lesion sho ws decreased inflammation and a thick ened fibrous cap.

512

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

B B

A A H2N

NH2

H2N

Iron oxide core

H2N

NH2

H2N

NH2

500 mm Crosslinked dextran

C C

NILAT

NH2 AlexaFluor 750

H2N H2N H2N

500 mm

Figure 2. A, Schematic representation of NILAT nanoagent. IVFM image before (B) and 1 week post-therapy (C) showing decreased NILAT uptake in treated plaque.

A similar strate gy w as also used to image and treat brain tumors. Magnetic nanopar ticles and light-acti vated therapeutic molecules (Photofrin) w ere incor porated within pol yacrylamide nanopar ticles functionalized with polyethylene gl ycol (PEG) and a peptide-based tar geting moiety, F3. 17 This peptide homes to the nucleus of angiogenic endothelial cells in tumor v asculature. The particles were also labeled with AlexaFluor 594 to allo w for the determination of cellular uptak e and subcellular localization. In vitro, the cells e xhibited signif icant uptak e in MDA-MB-435 breast cancer cells and resulted in o ver 90% cell death w hen ir radiated at the therapeutic w avelength (630 nm). When injected into rats bearing gliomas, both targeted and untargeted particles accumulated within the tumor within 10 minutes, as deter mined by magnetic resonance imaging (MRI), with the untar geted par ticles washing out o ver the course of the ne xt 2 hours. Ov er this same time, the tar geted par ticles continued to accumulate. Therapeutic efficacy was determined in rats treated with the targeted particles, the untargeted particles, and the NILAT, Photofrin, and monitored b y dif fusion w eighted MRI. 8 da ys after the therap y, Photofrin and untar geted control groups showed a 25% change in dif fusion values, whereas the group receiving the targeted particles showed a 40% change. This increase in dif fusion also cor related with sur vival time, with the control g roups e xhibiting a median survival time of 13 days and the F3-targeted particle group surviving 33 days, with two animals disease free 6 months after therapy.

ULTRASOUND AGENTS AND DRUG DELIVERY Microbubbles used as contrast agents in ultrasound imaging are often composed of a lipid shell encapsulating a water-insoluble gas. These lipospheres are also capab le of

incorporating h ydrophobic dr ug moieties and can selectively deliver their payloads upon high pressure insonation (Figure 3A). The main drawback of this system is that the size of the bubb les often precludes e xtravasation, limiting therapeutic ef ficacy, but can be utilized for the local delivery of therapeutic pa yloads. The synthesis and utility of these microbubb les is fur ther discussed in Chapter 28, “Ultrasound Contrast Agents.” 18 have described the Tartis and colleagues incorporation of paclitax el into dr ug deli very v ehicles called acousticall y acti ve lipospheres (AAL), w hich are microbubbles sur rounded b y a shell of oil and lipid (Figure 3B). These AAL w ere also functionalized with cyclic RGD (ar ginine–glycine–aspartic acid) ligands to impart the ability to tar get tumor vasculature (αvβ3). Control bubbles containing Vybrant DiI were also synthesized. Drug delivery with AAL is acti vated by fragmentation of the microbubb les with the appropriate ultrasound pulse sequence. To test the in vitro c ytotoxicity of the agent, it was plated with A375m cells and insonated. Compared with conventional Taxol therapy, the AAL or targeted AAL displayed similar cytotoxicity, but was significantly greater than controls. Lastly, the in vivo binding ability of the particles w as tested in a chorioallantoic membrane model, using the fluorescently labeled bubbles. Following 1 minute of insonation, the dy e clearl y e xtravasates from the vasculature as the fluorescence signal becomes diffuse. This targeted deli very strate gy has the potential to reduce the overall amount of chemotherapeutic that needs to be given, as the microbubbles localize and can be imaged at the site of interest and selectively fragmented upon acti vation with the appropriate ultrasound pulse sequence. Perfluorocarbon (PFC) nanoparticles consist of a PFC liquid core encapsulated within lipid emulsions (see also Chapter 35, “Fluorocarbon Agents for Quantitative Multimodal Molecular Imaging and Targeted Therapeutics”). Crowder and colleagues 19 have described the synthesis

Theranostics: Agents for Diagnosis and Therapy

of PFC nanopar ticles incor porating fluorescently conjugated phospholipids (fluorescein or rhodamine) and targeted b y a thiolated peptidomimetic vitronectin antagonist of the αvβ3 integrin. As compared to the untargeted par ticles, the tar geted agents readil y bound to and w ere inter nalized b y αvβ3-expressing human C32 melanoma cells. Sonif ication of both agents for 5 minutes resulted in signif icant fluorescence increase throughout the c ytoplasm, with the most dramatic ef fect for the tar geted par ticles (F igure 3C). Similarly, Da yton and colleagues 20 have described the synthesis of fluorescentl y labeled paclitax el loaded PFC nanoparticles. Cell viability in cells treated with increasing concentrations of par ticles was compared with cells receiving just ultrasound or par ticles and no ultrasound. In the absence of ultrasound , the toxicity of the par ticles was comparab le to cells treated with ultrasound alone, whereas up to a 9-fold increase in to xicity was observed for cells treated with the nanoagent and insonif ication.

NANOMATERIALS Nanoparticles can be localized to sites of interest via two methods, namel y passi ve and acti ve tar geting. Both methods tak e adv antage of the f act that nanopar ticles have innately longer blood half-lives when functionalized with ligands, such as PEG, that prevent opsonization and

513

rapid clearance. Passive targeting relies upon the vascular permeability of inflamed tissues, such as those present in tumors.21,22 Long-circulating nanopar ticles are ab le to extravasate from leaky, damaged vessels and access target tissues, enabling the imaging of the disease and the delivery of therapeutic drugs. Also advantageous is the lack of lymphatic drainage in these inflamed tissues, allowing for the accumulation of nanoagents. The ability to functionalize the surface of the nanomaterial also allows for cell or tissue-specif ic targeting. This has been achie ved using antibodies, peptides, aptamers, and small molecules. 23 Most impor tantly, multiple copies of the tar geting ligands can be displa yed on the par ticle surface, allowing it to multi valently bind to a number of receptors simultaneously, thus increasing the binding affinity of the agent for its target.24 The utility of these targeted particles fur ther increases with the addition of other ligands, such as those used for imaging and therap y. Multifunctional nanoagents, or those bearing more than one type of ligand or functionality, are the main goal of nanomedicine, allowing for the creation of theranostic nanoagents that are able to simultaneously diagnose and treat diseases, monitor the deli very of therapeutic molecules, and determine treatment ef ficacy. The synthesis and utility of nanoagents used in molecular imaging are also reviewed in Chapter 26, “Magnetic Resonance Imaging Agents” and Chapter 34, “Magnetic Nanoparticles.”

A

1

2

B B

3

4

C Taxol Phospholipid cRGD peptide Water-inspluble gas

Figure 3. Ultrasound microbubbles. A, Targeted imaging and therapy with microbubbles: (1–2) Targeted microbubbles traverse the vasculature, localizing at sites presenting the appropriate receptor. (3) Accumulation of the contrast agent can be detected by ultrasonography. (4) High pressure insonation ruptures the liposome, releasing the cargo. B, Schematic representation of a representative therapeutic microbubble construct. C, Ultrasound and targeting augment lipophilic delivery to C32 cells using ultrasound nanoemulsions. (top left) Confocal micrographs under normal conditions for nontargeted cells show minimal nonspecific internalization. (top right) αvβ3-integrin targeted (T) cells show specific targeting of nanoparticles and delivery of lipids throughout the cell. After ultrasound insonification for 5 min, marked enhancement of lipid delivery for both nontargeted (bottom left) and targeted (bottom right) cells was observed. Adapted with permission from Crowder KC et al.19

514

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

resulting par ticles retain their MR acti vity and can also act as photother mal therapy agents w hen irradiated with the appropriate wavelength of NIR light. Importantly, the ability to change the relati ve dimensions of the core radius and shell thickness of the nanopar ticles allows for the optical resonance w avelengths to be tuned. This is especially important for in vivo applications where wavelengths in the NIR re gion of the electromagnetic spectrum penetrate most deepl y into tissues. To elicit targeting, the par ticles w ere appended with anti-Her2/neu antibodies. Incubation of Her2 – (H520 lung cancer) and Her2+ (SKBR3 breast cancer) cells with the targeted agent, followed by T2-weighted MRI showed significant dark ening of the Her2 + cells, illustrating uptake of the par ticles versus the Her2 – cells. This result was further illustrated b y treating the cells with 800 nm light using a femtosecond-pulse laser. Although some cell death w as obser ved at higher light doses in the H520 cells, signif icant therapeutic effect was seen at lo w-light doses in the SKBR3 cells (F igure 4B). Gold nanoshells are also useful in molecular imaging as the y increase the backscattering of light in modalities such as optical coherence tomo graphy (OCT). Gobin and colleagues30 have reported the synthesis of gold nanoshells that have been PEGylated to enhance circulation times and decrease the immune response in vi vo. Injection of the nanoparticle preparations into tumor bearing mice (CT26 murine colon carcinoma) resulted in increased contrast within the tumor by OCT versus mice receiving only saline injections or healthy tissues. The mice also received a therapeutic dose of light and were monitored for several weeks post-therapy. The treated mice displayed complete regression of the tumors, w hereas the control mice sho wed significant tumor growth, with mice receiving PBS and laser irradiation surviving an a verage of 14 da ys, the untreated

Nanoparticles for Thermoablation When tissues are heated abo ve 46°C, nor mal cellular processes cease, causing e xtensive cell death. This process, termed thermoablation, has been investigated for the treatment of cancers. 25 Researchers have synthesized materials that are capab le of tar geting disease and that can act as susceptors, materials adept at con verting electromagnetic energy into heat. To this end, much attention has focused on iron o xide nanopar ticles26–29 and gold nanoshells or nanorods. 30–36 The main dif ferences between the tw o par ticles is the for m of ener gy that is used to acti vate the ther moablative proper ties, with the magnetic material requiring alter nating magnetic f ields (AMF) for hysteresis heating and the gold nanomaterials absorbing NIR energy. One e xample of a tar geted iron o xide preparation involves the modification of iron oxide–impregnated dextran beads with 111In-labeled chimeric L6 (ChL6) antibodies.26,28,37 ChL6 targets an integral membrane glycoprotein highly expressed on human breast, colon, ovary, and lung carcinomas. The authors were able to show that the probe accumulated signif icantly in vi vo in a murine tumor model and that upon placement within an AMF, the growth rate of the tumors in treated mice w as slowed versus control or untreated mice. Thermoablation w as confirmed histologically after the animals were sacrificed, revealing areas of necrosis within the tumor. Magnetic nanoparticles have also ser ved as an magnetic resonance (MR) repor ter in the synthesis of gold nanoshells. In one report, Kim and colleagues33 described the synthesis of gold nanoshells from silica core particles (Figure 4A). The silica is initiall y coated with magnetic nanoparticles and seed gold par ticles, which is follo wed by the g rowth of gold shells around the silica core. The

No nanoshells

B

A

Non-specific antibody

Anti-HER2

Scatter-based imaging Silica core Iron oxide

Therapy

Gold shell Anti-Her2 Antibody

Silver stain

Figure 4. A, Schematic representation of silica-cored gold nanoshell encapsulating iron oxide nanoparticles. B, Combined imaging and therapy of SKBr3 breast cancer cells using HER2-targeted nanoshells. Scatter-based darkfield imaging expression (top row), cell viability assessed via calcein staining (middle row), and silver stain assessment of nanoshell binding (bottom row). Cytotoxicity was observed in cells treated with a NIR-emitting laser following exposure and imaging of cells targeted with anti-HER2 nanoshells only. Adapted with permission from Loo C et al.34

Theranostics: Agents for Diagnosis and Therapy

control group surviving 10 da ys. The majority of mice in the photother mal therap y g roup (80%), ho wever, w ere alive after several weeks.

Delivery of Small Molecule Chemotherapeutics Drug deli very v ehicles, such as liposomes, polymeric micelles, and inorganic nanoparticles, have been investigated for the site-specif ic localization of s mall-molecule chemotherapeutics. These nanoagents ha ve man y adv antages o ver systemic deli very of their small-molecule counterparts, such as longer blood half-lives, reduction of systemic to xicity, and increased stability of the dr ug entity due to decreased contact with the biolo gical milieu.38 In controlled-release strate gies, dissociation of the drug from the nanovehicle has the potential to maint ain therapeutic dr ug levels. Whether delivered by passive or active targeting, the use of nanopar ticles may also allow for a decrease in the total amount of dr ug need as it can be released mainl y at the site of interest. Depending on the nanoparticle formulation used, therapeutic or reporter molecules can be encapsulated within the par ticle core, associated with the par ticle due to electrostatic interactions or conjugated to the par ticle surface. Encapsulation within Nanoparticles

Encapsulation within nanopar ticulate car riers alters a drug’s pharmacodynamics, reducing toxicity and increasing stability.39 Due to the ability to control the par ticle size during their synthesis, the agents ha ve increased pharmacokinetic prof iles, allo wing for e xtravasation from leaky vasculature, as well as enhanced intracellular uptake through endoc ytosis. The surf ace of the par ticle can be modif ied to include a number of ligands, including those used for targeting, imaging, and therapy, as well as other polymers, including PEG. In addition to the conjugation of therapeutic moieties to the par ticle, they may also be adsorbed on the surf ace or encapsulated within the nanoparticle core. Liposomes ha ve been used for the deli very of a number of imaging moieties, such as fluorescent dy es, radionuclides,40,41 and MR 42 and CT 43,44 contrast agents. Depending on the for mulation, the repor ter is included within either the lipid bilayer or the aqueous core. Often, radionuclides, such as 111 In, and paramagnetic metal ion contrast agents, such as gadolinium, are chelated b y amphiphilic polymers that associate with the h ydrophobic por tion of the liposome. 45,46 This increases the

515

loading efficiency of the nanovehicle and enables greater signal from each liposome. Nanopar ticulate imaging agents, such as iron o xide nanopar ticles or quantum dots, ha ve also been incor porated within liposomes to increase their stability under physiological conditions. 47 One of the more commonl y used c ytotoxic agents in the treatment of a v ariety of cancers is cisplatin. Ramachandran and colleagues48 have reported the encapsulation of cisplatin within the aqueous core of fluorescently labeled liposomes. The fluorophore, Dansyl, w as covalently conjugated to the phospholipid and incor porated within the lipid bilayer. Incubation of the resulting liposomes with A2780 cells, a cisplatin-sensiti ve human ovarian cancer cell line, resulted in endoc ytic uptake of the particles, as deter mined by fluorescence microscopy. Incubation of the cells with the agent for 10 hours resulted in massi ve cell death, compared with control liposomes without cisplatin. Another g roup has created temperature-sensiti ve liposomes encapsulating do xorubicin (Do x) and manganese sulf ate (Mn). 49 Using specif ic MRI techniques, the release of Mn from the liposome after h yperthermia can be obser ved, allo wing for estimation of local Do x concentrations. Rats bearing FSA-1 fibrosarcoma tumors were injected with the liposomes and were then subject to hyperthermia via a 18-gauge catheter through w hich heated water was passed. F ollowing therapy, the tumors of the treated mice e xhibit signif icant T1 brightening, indicating release of the Mn from the liposomes (Figure 5). The authors w ere then ab le to estimate the amount of Do x released from the liposomes using the ratio of the encapsulated Mn to Dox. In rats that exhibited the highest amount of tumoral Do x, the therapeutic response was the most pronounced , whereas the control groups or those with lo wer Do x concentrat ions sho wed poor antitumor effects. Nanoparticulate dr ug deli very systems based upon biodegradable, synthetic polymers have also received significant attention. 50 Release of dr ugs from pol ymeric nanoparticles occurs b y se veral dif ferent mechanisms, including diffusion, erosion of the par ticle surface,51 and particle s welling.50 Several acti vation schemes are also used, including pH52,53 or temperature-induced changes in polymer degradation.54,55 Nasongkla and colleagues 56 have repor ted the synthesis of tar geted polymeric micelles incor porating iron oxide nanopar ticles for MRI and the chemotherapeutic doxorubicin (DOX), which is released from the par ticles through a pH-dependent mechanism upon cellular internalization. The nanopar ticle matrix consists of poly(ethylene gl ycol)-block-poly(D,L-lactide) and w as

516

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A A

B B

C C

1 cm

Figure 5. Tumor drug distribution after administration of doxorubicin- and manganese-containing lysolipid-based temperature-sensitive liposomes and hyperthermia by three different schedules. (A–C) Axial pelvic magnetic resonance images show rats bearing flank fibrosarcomas (top left). LTSLs administered during steady-state hyperthermia resulted in peripheral enhancement (liposome content release; white) at the edge of the tumor (A); LTSLs administered before hyperthermia resulted in central enhancement (B); and LTSLs administered in two equal doses, half before hyperthermia and the remainder after steady-state hyperthermia was reached, resulted in uniform enhancement (C). Reproduced with permission from Ponce AM et al.49

targeted to αvβ3 integrin by conjugation of c yclic RGD moieties to the h ydrophilic PEG coating. The tar geting ability of the conjugates was ascertained in SLK endothelial cells overexpressing αvβ3 using T2-weighted MRI. As compared to control cells, or those recei ving untargeted particles, the RGD-tar geted agent displa yed signif icant T2 darkening. This uptake was further confirmed by flow cytometry and fluorescence microscop y, utilizing the innate fluorescence of the DOX. The therapeutic efficacy of the nanoagent w as deter mined b y incubation of the SLK cells with free DOX, versus a variety of targeted or untargeted formulations. Although free DOX showed the largest inhibition of cell g rowth, the nanoagents with 16% of the surface PEGs modified with targeting ligands exhibited comparable cytotoxicity. Interestingly, preincubation of the cells with the cRGD peptide abro gated the cell killing potential of the par ticles. Functionalization of the Nanoparticle Surface

A complementary strategy to the one discussed above has been to decorate the surf ace of much smaller nanopar ticles with diagnostic and therapeu tic molecules. Although drug loading is lo wer, these materials are often less than 100 nm in diameter and are easier to tar get to sites of interest. Jain and colleagues 57 have described the synthe sis of water-dispersible oleic acid–pluronic coated iron o xide nanoparticles capab le of incor porating w ater-insoluble Dox. The magnetic nanopar ticles w ere initiall y coated with h ydrophobic oleic acid , allo wing for a core into which the drug would partition. Pluronic was then added to coat THE nanoparticle and allow for its suspension in aqueous solution. Addition of Dox to a suspension of the particles resulted in a for mulation with 8.2 wt% dr ug per particle at 82% encapsulation ef ficiency. The in vitro cytotoxicity of the resulting preparation w as tested in

MCF-7 breast cancer and PC3 prostate cancer cells. Versus free Do x, the par ticles sho wed a slightl y lo wer toxicity, most lik ely due to the slo w release of the d rug from the preparation (40% release within 5 d ays). Interestingly, the intracellular distribution of the par ticles differs from that of the free Do x, as deter mined b y confocal microscopy. F ree Do x immediatel y localizes to the nucleus, w hereas the par ticles are mainl y seen in the cytoplasm 2 hours after incubation. At 24 and 48 hours after treatment, the dr ug is seen localized in the nucleus, fur ther showing the slo w release nature of the nanoparticles. Whereas the magnetic properties of the particles were secondary to the drug delivery ability of the above preparation, Alexiou and colleagues 58 have used the super paramagnetic behavior of the magnetic par ticles to target and image the tumoral distribution of the iron o xide. In this study, mitoxantrone (MIT) was adsorbed onto the surf ace of starch-modified nanoparticles due to the interaction of the positi vely char ged dr ug with the ne gatively char ged starch phosphates. Mito xantrone has been used to treat breast carcinoma, non-Hodgkin lymphoma, and a number of other solid tumors and inhibits DN A and RN A synthesis b y intercalating in DN A molecules, causing strand breaks. Although MIT is readily associated with the particles at pH 7.4, it dissociates under physiological conditions (~60 min). The efficacy of the resulting preparation w as deter mined in vi vo in rabbits, using an electromagnet to localize the par ticles to the tumor . Animals recei ving treatment with the par ticle intraarterially under magnetic tar geting displa yed increased accumulation of the agent within the tumor , as w as observed histologically and by MRI. In fact, these animals displayed complete tumoral re gression, with signif icant decreases in tumor size within 14 days. In contrast, regression was observed only in animals treated with MIT alone at the highest doses.

Theranostics: Agents for Diagnosis and Therapy

Similar to magnetic nanopar ticles, semiconductor quantum dot (QDs) have well-established chemistries for surface modification to elicit targeted delivery to sites of interest. More recentl y, these par ticles ha ve also been used to deli ver chemotherapeutics. Using an elaborate strategy, Bagalk ot and colleagues 59 have de veloped aptamer-targeted QDs capab le of sensing dr ug delivery. The system is based upon a bi-FRET reporter system and is comprised a QD core, an aptamer ligand tar geting prostate specif ic membrane antigen (PSMA), and Do x. The delivery system is synthesized by conjugation of the aptamer to the carbo xylated QD core, follo wed by incubation with Do x. The resulting agent displa ys fluorescence quenching of the QD via FRET to Do x, whereas the Do x fluorescence is quenched b y intercalation into the aptamer (Figure 6A). The targeting specif icity of the nanoagent w as e xamined in PSMA e xpressing LNCaP and PSMA-negative PC3 prostate adenocarcinoma cells. As expected, the PC3 cells displayed negligible uptake of the par ticles, w hereas fluorescence from both the Dox and the QD w ere obser ved in the LNCaP cells (Figure 6C, D). Once preferential uptake was determined, the authors e xamined the c ytotoxicity of the agent in

517

both cell lines. Incubation of the cells with free Do x resulted in similar cell death of PC3 and LNCaP cells, whereas incubation with the tar geted conjugate caused significantly more cell death in the LNCaP calls than in the PC3 cells (52.5 ± 1.6% vs 77.2 ± 3.1%, respectively). Dendrimers, and in par ticular pol y(amidoamine) (PAMAM), ha ve also been used to deli ver dr ugs,60,61 radionuclides,62 MRI agents, 63–65 and other ligands. Recently, Majoros and colleagues66,67 and Thomas and colleagues68 have described the synthe sis and utility of targeted theranostic nanoagents based upon a generation 5 (G5) PAMAM dendrimer. The G5 dendrimer contains 110 free amines, w hich are par tially acetylated to neutralize the cationic character of the pol ymer and decrease possible deleterious nonspecific interactions when used in vitro or in vi vo. To this w as added fluorescein isothioc yanate (FITC) for fluorescence imaging and folic acid (F A) for targeting. Methotrexate (MTX), a potent chemotherapeutic used in the treatment of a number of malignancies, was then conjugated to the dendrimer after modif ication of the dr ug with F A.67,68 This w as follo wed b y reaction of the dendrimer with gl ycidol to neutralize the residual amine g roups. In vitro e xamination of the resulting CC

A A Dox

QD

QD

QD QD-Apt: “ON”

B B

QD

“OFF”

QD

QD-Apt(Dox): “OFF”

D D

Target cancer cell

Drug release

Iysosome Nucleus

QD

QD: “ON”

Dox: “ON”

Figure 6. A, Schematic illustration of QD-Apt(Dox) Bi-FRET system. In the first step, the CdSe/ZnS core-shell QDs are surface functionalized with the A10 PSMA aptamer. The intercalation of Dox within the A10 PSMA aptamer on the surface of QDs results in the formation of the QD-Apt(Dox) and quenching of both QD and Dox fluorescence through a Bi-FRET mechanism: the fluorescence of the QD is quenched by Dox, whereas simultaneously the fluorescence of Dox is quenched by intercalation within the A10 PSMA aptamer resulting in the “off” state. B, Schematic illustration of specific uptake of QD-Apt(Dox) conjugates into target cancer cell through PSMA mediate endocytosis. The release of Dox from the QD-Apt(Dox) conjugates induces the recovery of fluorescence from both QD and Dox (“on” state) thereby sensing the intracellular delivery of Dox and enabling the synchronous fluorescent localization and killing of cancer cells. C, Confocal laser scanning microscopy images of PSMA expressing LNCaP cells after incubation with 100 nM QD-Apt-(Dox) conjugates for 0.5 h at 37ºC, washed two times with PBS buffer and further incubated at 37ºC for 0 h and 1.5 h. Dox and QD are shown in red and green, respectively, and the lower right images of each panel represents the overlay of Dox and QD fluorescent. The scale bar is 20 µm. Reproduced with permission from Bagalkot V et al.59

Table 1. THERANOSTIC AGENTS Type

Therapeutic Moiety

Diagnostic Moiety

Antibody molecule

EGFR

Perk et al.7

99m

Antibody molecule

EGFR

Schechter et al.6

68

Antibody molecule

EGFR

Velikyan et al.8

BODIPY FL

Antibody molecule

EGFR

Rosenthal et al.5

Trastuzumab and Cetuximab

Cy 5.5 and Cy 7

Antibody molecule

HER2/neu (erbB2) and EGFR

Barrett et al.10

Chlorin e6

Chlorin e6

Enzyme cleavable polymer

Cancer

Choi et al.77

Porpholactol

Porpholactol

Polymeric nanoparticle Cancer

IRDye 800 CW and Cy 5.5 and 111In 89

Zr Tc

Ga

Thermal ablation

References

HER2/neu (erbB2) Sampath et al.2 HER2/neu (erbB2) Xu et al.76

Cetuximab

Light activated therapy

Target/Application

Antibody molecule Antibody molecule

Antibody based Trastuzumab systems

111

Vehicle

In

McCarthy et al.14

Tetrahydroxyphenylchlorin Tetrahydroxyphenylchlorin Polymeric nanoparticle Macrophages

Konan et al.78,79

Tetraphenylchlorin

MFNP (AF750)

MNP

McCarthy et al.16

Photofrin

MFNP (AF594)

Polymeric nanoparticle Cancer

Reddy et al.17

Gold nanoshell

OCT

Gold nanoshell

Cancer

Gobin et al.,30 O’Neal et al.35

MNP

Gold nanoshell

Cancer

Loo et al.34 Ji et al.80

Gold nanorod

OCT

Gold nanorod

Cancer

Huang et al.32

Iron oxide nanoparticle

MNP

MNP

Cancer

Sonvico et al.29

111

MNP

Cancer

DeNardo et al.26,37

Gentamicin

Gd-DTPA

Hydrogel

Drug release

Weissleder et al.81

Paclitaxel

AAL

AAL

Cancer

Tartis et al.18

PFC nanoparticle and Vibrant DiI

PFC nanoparticle

Cancer

Crowder et al.,19 Dayton et al.20

Cisplatin

Dansyl

Liposome

Cancer

Ramachandran et al.48

Doxorubicin

MnSO4

Liposome

Cancer

Ponce et al.49

MNP

Polymeric micelle

Cancer

Nasongkla et al.56

MNP

MNP

Cancer

Jain et al.57

Quantum dots and Doxorubicin

Quantum dot

Cancer

Bagalkot et al.59

MIT

MNP

MNP

Cancer

Alexiou et al.58

Methotrexate or paclitaxel

Fluorescein

PAMAM dendrimer

Cancer

Majoros et al.,66,67 Thomas et al.68

siRNA

64

Cationic cyclodextrin

Luciferase

Bartlett et al.72

Quantum dots

Cationic liposomes

Lamin A/C and T-cad genes

Chen et al.73

Quantum dots

Quantum dots

EGFP

Derfus et al.74

MFNP (Cy 5.5)

MNP

EGFP and surviving

Medarova et al.75

In

Cancer

Self-contained systems Chemotherapy

RNA interference

Cu

EGFP = enhanced green fluorescent protein; EGFR = epidermal growth factor receptor; MFNP = magnetofluorescent nanoparticle; MIT = mitoxantrone; MNP = magnetic nanoparticle; OCT = optical coherence tomography; PFC = perfluorocarbon.

Theranostics: Agents for Diagnosis and Therapy

nanoagent showed a strong, dose-dependent uptak e in FA receptor expressing KB cells, w hich could be ab lated by preincubation with FA, as deter mined by flow cytometry and fluorescence microscop y. This uptak e w as not observed for particles that were not targeted with FA. The antiproliferative ef fect of the nanoagent w as also e xamined. Unfortunately, at shor t incubation times (up to 4 h), the conjugate sho wed a little ef fect. Longer incubation times (1 to 3 d) of the dendrimer w ere required to sho w cytotoxicity with both time- and dose-dependent g rowth inhibition. Majoros and colleagues 66 expanded upon the abo ve work by conjugating paclitaxel to the dendrimer instead of MTX. Taxol was modified with succinic anhydride to allow for the creation of an ester bond, followed by activation with N-hydroxysuccinimide and reaction with the free amines of the dendrimer . The ester linkage is impor tant as it is readily cleaved by intracellular esterases, allo wing for the controlled release of the dr ug. Incubation of the resultant particles with KB cells showed a dose-dependent uptake of the agent. Cytotoxicity was determined after incubation of KB cells with the par ticle, washing, and further incubation for an additional 72 hours. At concentrations as lo w as 50 nM, the nanoagent sho wed signif icant to xicity to FA+ KB cells, with little cell death in FA– KB cells.

RNA Interference Small interfering RN A (siRN A) are doub le-stranded RNA molecules approximately 20 to 25 nucleotides long that are involved in RNA interference (RNAi), knocking down the e xpression of a specif ic gene. This class of compound is impor tant therapeuticall y as it has the potential to tur n off the acti vity of genes in cer tain diseases.69,70 One of the main prob lems with the in vi vo delivery of the siRN A is the circulation half-life of the unmodified agent (~6 min).71 To circumv ent this, complexes of the siRN A have been developed, using car riers such as cationic proteins and pol ymers. Attempts to target these comple xes with antibodies, peptides, aptamers, and small molecules ha ve also been made, but lack the ability to visualize localization. Transferrin (Tf)-tar geted siRN A nanopar ticles ha ve been synthesized b y Bar tlett and colleagues 72 using cyclodextrin-containing pol ycations. To the siRN A w as also conjugated DOTA for the chelation of 64Cu and subsequent imaging b y micro-PET/CT . The siRN A used knocked do wn luciferase e xpression and thus could be used to measure delivery efficiency. When injected intravenously into tumor -bearing mice, the nanopar ticles exhibited extremely short blood half-lives, comparable to

64

519

Cu-labeled siRNA (2.4 min), and w ere cleared through the reticuloendothelial system (RES), as is often observed for nanopar ticles. Interestingly, the Tf-targeted and untar geted par ticles displa yed similar tissue distributions, as deter mined b y micro-PET/CT ; y et at 1 day postinjection, the animals treated with the Tf-targeted particles had a 50% lo wer increase in luciferase acti vity versus the untargeted particle, as determined by bioluminescence imaging. Two fur ther strate gies to image the deli very of siRNA ha ve been de veloped b y Chen and colleagues 73 and Derfus and colleagues. 74 Addition of optically active QDs to traditional transfection agents, such as cationic liposomes, results in siRNA complexes that are not modified in any manner, reducing the lik elihood of potential damage to the therapeutic moiety. Delivery of liposomes bearing g reen QDs and Lamin A/C siRN A into human 3T3 fibroblasts was monitored by flow cytometry, gating on the fluorescence of the QD. Although the overall fluorescence increase for cells transfected with the QD-containing liposomes w as modest, compared with control cells, there w as a broad distribution of fluorescence intensities. Using fluorescence assisted cell sor ting (FACS), 10% of the brightest and dimmest cells were collected and re-plated to conduct protein anal ysis. After 72 hours further growth, the cells were analyzed by western blot or immunostaining. In the cells bearing the highest degree of fluorescence, the silencing of Lamin w as pronounced, with appro ximately 90% knockdo wn, whereas the least fluorescent cells sho wed ne gligible silencing. Unsorted cells sho wed 20 to 30% do wnregulation when treated with siRNA alone. One of the main advantages of using QDs is that the y allow for multiple xing. To illustrate this, 3T3 cells w ere incubated with the pre viously described Lamin siRNA complexes, as well as a liposome complex incor porating an orange QD and siRN A for T-cad, which plays a functional role in cell–cell interactions. Again, the por tion of the cells with high and lo w accumulations (8%) of both par ticles were separated b y FACS, with the most fluorescent par ticles e xhibiting a 96% knockdo wn of Lamin and a 98% knockdo wn of T-cad. The ability to multiple x using QDs allo ws for the tracking and sor ting of multiple knockdo wn within once cell population. One of the main dra wbacks of the abo ve approach is that it is not amenable to in vivo utility as the liposomes are readily inter nalized b y macrophages and cleared through the RES, likely due to their large size. To circumvent this, a multifunctional nanoagent was synthesized by conjugation of siRNA duplexes and tar geting moieties to a PEGylated QD core (F igure 7A). 74 The siRNA used w as designed to

520

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

B A F3 peptide siRNA

Quantum dot

C

+ F3/siRNA QDS

PEG

Figure 7. A, Schematic representation of F3-targeted QD. EGFP-expressing HeLa cells were treated with F3-targeted siRNA-QDs or lipofectamine (control) for 4 h and then washed with cell media. B, When assayed for green fluorescence 48 h later, no knockdown was observed in the control cells. C, Fluorescence imaging of cells incubated with theranostic QDs (red dots) showed a reduced green fluorescence. Reproduced with permission from Derfus AM et al.74

knock down enhanced g reen fluorescent protein (EGFP), whereas the par ticles w ere tar geted with the F3 peptide, which homes to tumor vasculature. This strategy allows for the creation of nanoagents that ha ve a smaller overall size, potentially evading uptake by the RES. Preliminary in vitro experiments were performed in HeLa cells transfected with EGFP. Uptake of the F3-tar geted particles was determined by flow c ytometry, compared to control par ticles (PEGylated QDs, control peptide-tar geted QDs and untar geted QDs modif ied with siRN A), and sho wed appro ximately 100-fold greater uptake (Figure 7B, C). This uptake could readily be inhibited b y preincubation of the cells with the free F3 peptide. The role of the siRN A conjugation chemistry on the ability of the nanoagent to silence EGFP w as also investigated. When the siRN A was conjugated to the particles by a clea vable disulf ide bond, the cells e xhibited greater knockdown versus an amide bond linkage, as determined by flow cytometry. One of the main dra wbacks of this technique is that the tar geted QD conjugates cannot deliver siRNA to promote knockdo wn in the absence of a transfection agent due to entrapment within endosomes, as determined by fluorescence microscopy. Endosome escape of the par ticles can be af fected by incubation of the cells with cationic liposomes and is v erified by knockdown of EGFP. An analo gous siRN A deli very constr uct has also been repor ted by Medarova and colleagues, 75 using magnetofluorescent nanoparticles with similar results.

CONCLUSIONS Theranostic agents ha ve the potential to signif icantly enhance the diagnosis and treatment of numerous diseases by of fering the ability to co-deli ver therapeutic entities with diagnostic imaging agents. This paradigm also allows for the deter mination of the localization, release,

and therapeutic ef ficacy of the requisite treatment. Numerous strategies have been developed for the synthesis of these agents, yet a number of key questions still must be answered: w hen should theranostic agents be used , as opposed to the appropriate diagnostic or therapeutic moiety? Can the difference between conventional imaging and therapeutic doses be reconciled? Although many examples have been reported, further advances in the clinical formulation of multifunctional theranostic agents are expected to enhance diagnostic and therapeutic medicine.

REFERENCES 1. Harlow E, Lane D. Antibodies: a laboratory manual. New York: Cold Spring Harbor Laboratory Press; 1988. 2. Sampath L, Kw on S, K e S, et al. Dual-labeled trastuzumab-based imaging agent for the detection of human epider mal growth factor receptor 2 o verexpression in breast cancer. J Nucl Med 2007; 48:1501–10. 3. Mendelsohn J, Baselga J. Status of epider mal growth factor receptor antagonists in the biolo gy and treatment of cancer . J Clin Oncol 2003;21:2787–99. 4. Cai W, Chen K, He L, et al. Quantitative PET of EGFR expression in xenograft-bearing mice using 64Cu-labeled cetuximab, a chimeric anti-EGFR monoclonal antibody . Eur J Nucl Med Mol Imaging 2007;34:850–8. 5. Rosenthal EL, Kulbersh BD, King T, et al. Use of fluorescent labeled anti-epidermal growth factor receptor antibody to image head and neck squamous cell carcinoma x enografts. Mol Cancer Ther 2007;6:1230–8. 6. Schechter NR, Wendt RE, Yang DJ , et al. Radiation dosimetr y of 99mTc-labeled C225 in patients with squamous cell carcinoma of the head and neck. J Nucl Med 2004;45:1683–7. 7. Perk LR, Visser GW, Vosjan MJ , et al. (89)Zr as a PET sur rogate radioisotope for scouting biodistribution of the therapeutic radiometals (90)Y and (177)Lu in tumor -bearing nude mice after coupling to the inter nalizing antibody cetuximab . J Nucl Med 2005;46:1898–906. 8. Velikyan I, Sundberg AL, Lindhe O, et al. Preparation and evaluation of (68)Ga-DOTA-hEGF for visualization of EGFR e xpression in malignant tumors. J Nucl Med 2005;46:1881–8.

Theranostics: Agents for Diagnosis and Therapy

9. Hama Y, Urano Y, Koyama Y, et al. Activatable fluorescent molecular imaging of peritoneal metastases follo wing pre-tar geting with a biotinylated monoclonal antibody. Cancer Res 2007;67:3809–17. 10. Barrett T, Koyama Y, Hama Y, et al. In vi vo diagnosis of epider mal growth factor receptor expression using molecular imaging with a cocktail of opticall y labeled monoclonal antibodies. Clin Cancer Res 2007;13:6639–48. 11. Vrouenraets MB, Visser GW, Snow GB, van Dongen GA. Basic principles, applications in oncolo gy and improved selectivity of photodynamic therapy. Anticancer Res 2003;23:505–22. 12. MacDonald IJ, Dougherty TJ. Basic principles of photodynamic therapy. J Porphyr Phthalocyanines 2001;5:105–29. 13. Brown SB, Brown EA, Walker I. The present and future role of photodynamic therap y in cancer treatment. Lancet Oncol 2004; 5:497–508. 14. McCarthy JR, P erez JM, Br uckner C, Weissleder R. P olymeric nanoparticle preparation that eradicates tumors. Nano Lett 2005; 5:2552–6. 15. Zheng G, Chen J, Stefflova K, et al. Photodynamic molecular beacon as an activatable photosensitizer based on protease-controlled singlet oxygen quenching and acti vation. Proc Natl Acad Sci USA 2007;104:8989–94. 16. McCarthy JR, Jaffer FA, Weissleder R. A macrophage-targeted theranostic nanopar ticle for biomedical applications. Small 2006; 2:983–7. 17. Reddy GR, Bhojani MS, McCon ville P , et al. Vascular tar geted nanoparticles for imaging and treatment of brain tumors. Clin Cancer Res 2006;12:6677–86. 18. Tartis MS, McCallan J , Lum AF, et al. Therapeutic effects of paclitaxel-containing ultrasound contrast agents. Ultrasound Med Biol 2006;32:1771–80. 19. Crowder KC, Hughes MS, Marsh JN , et al. Sonic acti vation of molecularly-targeted nanoparticles accelerates transmembrane lipid delivery to cancer cells through contact-mediated mechanisms: implications for enhanced local dr ug delivery. Ultrasound Med Biol 2005; 31:1693–1700. 20. Dayton PA, Zhao S, Bloch SH, et al. Application of ultrasound to selectively localize nanodroplets for targeted imaging and therapy. Mol Imaging 2006;5:160–74. 21. Maeda H. The enhanced per meability and retention (EPR) ef fect in tumor vasculature: the key role of tumor -selective macromolecular drug targeting. Adv Enzyme Regul 2001;41:189–207. 22. Maeda H, Seymour LW, Miyamoto Y. Conjugates of anticancer agents and polymers: advantages of macromolecular therapeutics in vivo. Bioconjug Chem 1992;3:351–62. 23. McCarthy JR, Kelly KK, Sun EY, Weissleder R. Targeted delivery of multifunctional magnetic nanopar ticles. Nanomedicine 2007; 2:153–67. 24. Montet X, Funovics M, Montet-Abou K, et al. Multi valent effects of RGD peptides obtained b y nanopar ticle displa y. J Med Chem 2006;49:6087–93. 25. Gordon RT, Hines JR, Gordon D . Intracellular h yperthermia. A biophysical approach to cancer treatment via intracellular temperature and biophysical alterations. Med Hypotheses 1979;5:83–102. 26. DeNardo SJ, DeNardo GL, Miers LA, et al. De velopment of tumor targeting bioprobes ((111)In-chimeric L6 monoclonal antibody nanoparticles) for alter nating magnetic f ield cancer therap y. Clin Cancer Res 2005;11:S7087–92. 27. Hilger I, Hiergeist R, Hergt R, et al. Thermal ablation of tumors using magnetic nanoparticles: an in vivo feasibility study. Invest Radiol 2002;37:580–6. 28. Ivkov R, DeNardo SJ, Daum W, et al. Application of high amplitude alternating magnetic f ields for heat induction of nanopar ticles localized in cancer. Clin Cancer Res 2005;11:S7093–103. 29. Sonvico F, Mor net S, Vasseur S, et al. F olate-conjugated iron o xide nanoparticles for solid tumor tar geting as potential specif ic

30.

31.

32.

33.

34. 35.

36.

37.

38. 39. 40.

41.

42.

43.

44.

45. 46. 47.

48.

49.

50.

51.

521

magnetic h yperthermia mediators: synthesis, ph ysicochemical characterization, and in vitro experiments. Bioconjug Chem 2005; 16:1181–8. Gobin AM, Lee MH, Halas NJ , et al. Near -infrared resonant nanoshells for combined optical imaging and photothermal cancer therapy. Nano Lett 2007;7:1929–34. Hirsch LR, Stafford RJ, Bankson JA, et al. Nanoshell-mediated nearinfrared ther mal therap y of tumors under magnetic resonance guidance. Proc Natl Acad Sci USA 2003;100:13549–54. Huang X, El-Sa yed IH, Qian W, El-Sayed MA. Cancer cell imaging and photother mal therap y in the near -infrared re gion b y using gold nanorods. J Am Chem Soc 2006;128:2115–20. Kim J, Park S, Lee JE, et al. Designed f abrication of multifunctional magnetic gold nanoshells and their application to magnetic resonance imaging and photother mal therap y. Angew Chem Int Ed Engl 2006;45:7754–8. Loo C, Lowery A, Halas N, et al. Immunotargeted nanoshells for integrated cancer imaging and therapy. Nano Lett 2005;5:709–11. O’Neal DP, Hirsch LR, Halas NJ, et al. Photo-thermal tumor ablation in mice using near infrared-absorbing nanopar ticles. Cancer Lett 2004;209:171–6. Pissuwan D, Valenzuela SM, Cor tie MB. Therapeutic possibilities of plasmonically heated gold nanopar ticles. Trends Biotechnol 2006;24:62–7. DeNardo SJ, DeNardo GL, Natarajan A, et al. Thermal dosimetry predictive of efficacy of 111In-ChL6 nanoparticle AMF-induced thermoablative therapy for human breast cancer in mice. J Nucl Med 2007;48:437–44. Langer R. Drug delivery and targeting. Nature 1998;392:5–10. Goyal P, Goyal K, Vijaya Kumar SG, et al. Liposomal dr ug delivery systems—clinical applications. Acta Pharm 2005;55:1–25. Andreopoulos D, Kasi LP, Asimacopoulos PJ, et al. Selective in vitro labeling of w hite b lood cells using 99mTc-labeled liposomes. Nucl Med Biol 2002;29:185–90. Harrington KJ, Mohammadtaghi S, Uster PS, et al. Effective targeting of solid tumors in patients with locall y advanced cancers by radiolabeled pegylated liposomes. Clin Cancer Res 2001;7:243–54. Mulder WJ, Strijkers GJ, van Tilborg GA, et al. Lipid-based nanoparticles for contrast-enhanced MRI and molecular imaging. NMR Biomed 2006;19:142–64. Sachse A, Leike JU, Rossling GL, et al. Preparation and evaluation of lyophilized iopromide-car rying liposomes for li ver tumor detection. Invest Radiol 1993;28:838–44. Sachse A, Leike JU, Schneider T, et al. Biodistribution and computed tomography b lood-pool imaging proper ties of pol yethylene gl ycol-coated iopromide-car rying liposomes. In vest Radiol 1997; 32:44–50. Torchilin VP. P olymeric contrast agents for medical imaging. Cur r Pharm Biotechnol 2000;1:183–215. Torchilin VP. Recent advances with liposomes as pharmaceutical carriers. Nat Rev Drug Discov 2005;4:145–60. Al-Jamal WT, Kostarelos K. Liposome-nanoparticle hybrids for multimodal diagnostic and therapeutic applications. Nanomed 2007;2:85–98. Ramachandran S, Quist AP, Kumar S, Lal R. Cisplatin nanoliposomes for cancer therap y: AFM and fluorescence imag ing of cisplatin encapsulation, stability , cellular uptak e, and to xicity. Langmuir 2006;22:8156–62. Ponce AM, Viglianti BL, Yu D, et al. Magnetic resonance imaging of temperature-sensitive liposome release: dr ug dose painting and antitumor effects. J Natl Cancer Inst 2007;99:53–63. Pridgen EM, Langer R, F arokhzad OC. Biode gradable, pol ymeric nanoparticle deli very systems for cancer therap y. Nanomed 2007;2:669–80. Siepmann J, Gopferich A. Mathematical modeling of bioerodible, polymeric drug delivery systems. Adv Drug Deliv Rev 2001;48:229–47.

522

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

52. Heffernan MJ, Murthy N. Polyketal nanoparticles: a new pH-sensitive biodegradable dr ug deli very v ehicle. Bioconjug Chem 2005; 16:1340–2. 53. Seow WY, Xue JM, Yang YY. Targeted and intracellular deli very of paclitaxel using multifunctional polymeric micelles. Biomaterials 2007;28:1730–40. 54. Choi SH, Lee SH, Park TG. Temperature-sensitive pluronic/poly(ethylenimine) nanocapsules for ther mally triggered disr uption of intracellular endosomal compar tment. Biomacromolecules 2006; 7:1864–70. 55. Han HD, Choi MS, Hwang T, et al. Hyperthermia-induced antitumor activity of thermosensitive polymer modified temperature-sensitive liposomes. J Phar m Sci 2006;95:1909–17. 56. Nasongkla N, Bey E, Ren J, et al. Multifunctional polymeric micelles as cancer -targeted, MRI-ultrasensiti ve dr ug deli very systems. Nano Lett 2006;6:2427–30. 57. Jain TK, Morales MA, Sahoo SK, et al. Iron o xide nanoparticles for sustained deli very of anticancer agents. Mol Phar m 2005; 2:194–205. 58. Alexiou C, Arnold W, Klein RJ, et al. Locore gional cancer treatment with magnetic drug targeting. Cancer Res 2000;60:6641–8. 59. Bagalkot V, Zhang L, Le vy-Nissenbaum E, et al. Quantum dot-aptamer conjugates for synchronous cancer imaging, therapy, and sensing of drug delivery based on bi-fluorescence resonance energy transfer. Nano Lett 2007;7:3065–70. 60. Kojima C, Kono K, Maruyama K, Takagishi T. Synthesis of polyamidoamine dendrimers having poly(ethylene glycol) grafts and their ability to encapsulate anticancer dr ugs. Bioconjug Chem 2000;11:910–7. 61. Patri AK, Majoros IJ , Baker JR. Dendritic pol ymer macromolecular carriers for drug delivery. Curr Opin Chem Biol 2002;6:466–71. 62. Kobayashi H, Wu C, Kim MK, et al. Ev aluation of the in vivo biodistribution of indium-111 and yttrium-88 labeled dendrimer -1B4MDTPA and its conjugation with anti-T ac monoclonal antibody . Bioconjug Chem 1999;10:103–11. 63. Kobayashi H, Brechbiel MW. Nano-sized MRI contrast agents with dendrimer cores. Adv Drug Deliv Rev 2005;57:2271–86. 64. Xu H, Regino CA, Koyama Y, et al. Preparation and preliminar y evaluation of a biotin-targeted, lectin-targeted dendrimer-based probe for dual-modality magnetic resonance and fluorescence imaging. Bioconjug Chem 2007:18:1474–82. 65. Xu R, Wang Y, Wang X, et al. In vi vo e valuation of a PAMAMcystamine-(Gd-DO3A) conjugate as a biodegradable macromolecular MRI contrast agent. Exp Biol Med (Ma ywood) 2007; 232:1081–9. 66. Majoros IJ, Myc A, Thomas T, et al. PAMAM dendrimer-based multifunctional conjugate for cancer therap y: synthesis, characterization, and functionality. Biomacromolecules 2006;7:572–9.

67. Majoros IJ , Thomas TP, Mehta CB , Bak er JRJ . P oly(amidoamine) dendrimer-based multifunctional engineered nanodevice for cancer therapy. J Med Chem 2005;48:5892–9. 68. Thomas TP, Majoros IJ, Kotlyar A, et al. Targeting and inhibition of cell growth by an engineered dendritic nanode vice. J Med Chem 2005;48:3729–35. 69. Dykxhoorn DM, P alliser D , Lieber man J . The silent treatment: siRNAs as small molecule dr ugs. Gene Ther 2006;13:541–52. 70. Uprichard SL. The therapeutic potential of RN A interference. FEBS Lett 2005;579:5996–6007. 71. Soutschek J, Akinc A, Bramlage B, et al. Therapeutic silencing of an endogenous gene b y systemic administration of modif ied siRNAs. Nature 2004;432:173–8. 72. Bartlett DW, Su H, Hildebrandt IJ, et al. Impact of tumor-specific targeting on the biodistribution and efficacy of siRNA nanoparticles measured by multimodality in vi vo imaging. Proc Natl Acad Sci USA 2007;104:15549–54. 73. Chen AA, Derfus AM, Khetani SR, Bhatia SN. Quantum dots to monitor RN Ai deli very and impro ve gene silencing. Nucleic Acids Res 2005;33:e190. 74. Derfus AM, Chen AA, Min DH, et al. Targeted quantum dot conjugates for siRNA delivery. Bioconjug Chem 2007;18:1391–6. 75. Medarova Z, Pham W, Farrar C, et al. In vivo imaging of siRNA delivery and silencing in tumors. Nat Med 2007;13:372–7. 76. Xu H, Baidoo K, Gunn AJ, et al. Design, synthesis, and characterization of a dual modality positron emission tomo graphy and fluorescence imaging agent for monoclonal antibody tumor -targeted imaging. J Med Chem 2007;50:4759–65. 77. Choi Y, Weissleder R, Tung CH. Selective antitumor effect of novel protease-mediated photodynamic agent. Cancer Res 2006;66:7225–9. 78. Konan YN, Ber ton M, Gur ny R, Allémann E. Enhanced photodynamic acti vity of meso-tetra(4-h ydroxyphenyl)porphyrin b y incorporation into sub-200 nm nanopar ticles. Eur J Phar m Sci 2003;18:241–9. 79. Konan YN, Cerny R, Favet J, et al. Preparation and characterization of sterile sub-200 nm meso-tetra(4-hydroxyphenyl)porphyrin-loaded nanoparticles for photodynamic therap y. Eur J Phar m Biophar m 2003;55:115–24. 80. Ji X, Shao R, Elliott AM, et al. Bifunctional gold nanoshells with a superparamagnetic iron o xide-silica core suitab le for both MR imaging and photother mal therap y. J Ph ys Chem C 2007 111:6245–51. 81. Weissleder R, Poss K, Wilkinson R, et al. Quantitation of slo w drug release from an implantable and degradable gentamicin conjugate by in vi vo magnetic resonance imaging. Antimicrob Agents Chemother 1995;39:839–45.

34 MAGNETIC NANOPARTICLES ANDREW TSOURKAS, PHD AND LEE JOSEPHSON, PHD

Developments in material sciences such as the de velopment of nanometer-sized magnetic nanoparticles and micron-sized magnetic par ticles are yielding major advances in di verse areas of biolo gy and medicine. One of the most impor tant of these is the application of magnetic materials as magnetic resonance (MR) imaging agents. Magnetic nanoparticles embody a series of properties including high detectability b y MR (nanomolar), biodegradability, and lack of to xicity that mak e them valuable tools in biolo gical and medical research, and practical as diagnostic drugs. Here, we describe the physical and biolo gical proper ties of these materials and review many of their important applications.

COMPOSITION AND MAGNETIC CHARACTERISTICS Superparamagnetic iron o xide (SPIO) nanopar ticles typically used for imaging consist of tw o basic components, a cr ystalline magnetic iron o xide core and a hydrophilic surface coating. The iron o xide core is composed of a mixture of two magnetic iron oxides, magnetite (Fe3O4) and maghemite ( γFe2O3), both of w hich possess similar cubic close pack ed o xygens in an in verse spinel crystallographic str uctures. The mixture is refer red to as nonstoichiometric magnetite, and a general formula that permits mixtures can be used: (FeO)1 − n(Fe2O3)n. For magnetite n = 0, for maghemite, n = 1. In f act, magnetite can be converted to maghemite under oxidative conditions or maghemite to magnetite under reducing conditions. Magnetite and maghemite also e xhibit similar magnetic properties although maghemite’ s saturation magnetization is slightly lower. At 300 K, magnetite e xhibits a saturation magnetization of 92 emu/g, whereas maghemite exhibits a saturation magnetization of 78 emu/g. 1 The iron oxides used as MR imaging agents and for most other biological applications are superparamagnetic.2

They have a high, positive magnetic susceptibility, but in the absence of a magnetic f ield, they exhibit essentially no magnetic remanence (no per manent magnetic moment).3–5 The lack of remanence occurs w hen the applied magnetic f ield is remo ved because ther mal energy at about 300 K randomizes orientation of the single magnetic moment present within a super paramagnetic cr ystal. Super paramagnetic materials sho w magnetic saturation at room temperature; their magnetic moment reaches a plateau as the applied f ield is increased. Essential to the beha vior of SPIOs are the strong interactions betw een the lar ge numbers of unpaired electrons within the in verse spinel iron o xide crystal structure, interactions that are so strong they force unpaired electrons to align as a single magnetic moment. When SPIOs are digested with acid , the crystal structure and super paramagnetism are destro yed. A paramagnetic iron salt results, and here unpaired electrons are uncoupled (ie, the y act independentl y of each other). Hence, very strong magnetic f ields produce onl y a v ery par tial alignment of unpaired electron spins, a w eak paramagnetism. The con version of iron o xides from super paramagnetic to paramagnetic materials occurs in biolo gical systems as the f irst chemical step in the de gradation and utilization of SPIO. At physiological temperatures, the iron in homosiderin, fer ritin, hemoglobin, and other naturally occur ring compounds are paramagnetic, and thus far less magnetic than the iron in SPIO .3–5 Magnetic properties of materials are strongly temperature dependent, where at lower temperatures, electron spins are more easily aligned leading to stronger magnetic properties. At 300 K, crystals of nonstoichiometric magnetite between about 2 and 15 nm in diameter are super paramagnetic. At lower temperatures or with larger crystals, superparamagnetic iron oxides become ferromagnetic, that is, they e xhibit magnetic saturation but retain a magnetic moment when the field is removed (magnetic remanence). 523

524

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Although a wide v ariety of materials can be used to confer magnetism on magnetic par ticles and nanopar ticles, the magnetic proper ties of super paramagnetic iron oxides mak e them w ell suited for use as MR imaging agents for three reasons. F irst, super paramagnetismbased particles are strongly magnetic in an applied f ield but nonmagnetic when the f ield is removed. This lack of magnetic attraction between SPIO nanoparticles prevents magnetically induced particle aggregation. Second, SPIO nanoparticles exhibit a higher potency (relaxivity) as MR imaging agents than paramagnetic metal chelates or ferromagnetic iron o xide–based magnetic par ticles.3–5 Finally, the to xicity and metabolism of SPIO are w ell understood, and the y are accepted as components of drugs b y re gulatory agencies throughout the w orld (discussed below).

SYNTHESIS OF SPIO SPIO nanopar ticles can be prepared using a wide range of synthetic approaches that result in nanopar ticles with distinct sizes, physical properties, coatings, and size distribution prof iles. Con ventionally, magnetite nanopar ticles are prepared b y adding a base (e g, NaOH, ammonium h ydroxide) to an aqueous mixture of fer ric (Fe3+) and ferrous (Fe2+) chloride in an oxygen-free environment. When conducted directl y in bulk solution, this approach i s k nown a s c oprecipitation a nd i s o ften referred to as the Massart method. The synthetic reaction may be written as:

2 Fe 3+ + Fe 2+ + 8OH − → Fe 3O4 + 4 H 2 O The ionic strength, pH, and concentration of the fer ric and ferrous salts can be adjusted to control the size of the nanoparticles. Often coprecepitation is carried out directly in the presence of a surf ace coating material, such as de xtran or other pol yols. Surf ace coatings are required before the use of iron oxide nanoparticles in biomedical applications to pre vent nanoparticle agglomeration and destabilization, improve solubility, and modulate nanoparticle biodistr ubution. The v arious surf ace coatings that have been used on iron o xide nanoparticles are discussed in more detail below. In general, synthesizing SPIO nanopar ticles is relatively simple. Ho wever, it has pro ven to be quite challenging to de velop a synthetic approach that can produce a monodisperse population of iron o xide colloids of controllable size at the large scales that are necessary for industrial production. For example, traditional coprecipitation methods that can be conducted directly in bulk solution can be used to generate

large quantities of iron o xide nanopar ticles with relatively good control o ver the mean iron o xide core size (2 to 15 nm), but this procedure often results in nanoparticles with a broad size distribution (typicall y greater than ±25% of mean). 6 Much narrower size distributions can be obtained w hen microemulsion techniques are used (size v ariation typically within ±5% of mean), but purif ication (ie, removal of surfactants) can be quite challenging and much smaller quantities of nanoparticles are synthesized. 7,8 Other less utilized techniques include ultrasound ir radiation (or sonochemical synthesis),9 sol–gel synthesis,10 layer-by-layer deposition, 11,12 and spra y and laser p yrolysis. 13–16 Although some of these latter methods can provide narrower size distributions and/or more complex core/shell structures, they often require adv anced equipment and expertise. One par ticular promising approach for the ultra-large-scale synthesis of monodisperse iron o xide colloids relies on the ther mal decomposition of iron salts.17,18 It has been repor ted that iron o xide nanopar ticles with continuously tunable sizes (6 to 30 nm) and narrow s ize d istributions ( ~2.5%) c an b e p repared v ia thermal decomposition in a single reaction.18,19 The particle size is controlled simpl y b y perfor ming ther mal decomposition of iron salts in solv ents with dif ferent boiling points 19 or b y v arying the reaction time. 18 This approach has garnered a great deal of excitement because the synthetic process is ine xpensive and uses nonto xic iron salts as reactants. Aside from dif ferences in size and distribution, iron oxide colloids prepared by different synthetic routes may also exhibit large differences in their saturation magnetization. These differences have been attributed to different degrees of str uctural disorder.20,21 It has also been suggested that the creation of antiphase boundaries and the existence of a magneticall y dead la yer at the iron o xide surface could contrib ute to v ariances in saturation magnetization.22,23 Accordingly, iron oxide colloids have been reported to e xhibit saturation magnetization v alues anywhere from 10 to 60 emug −1. Interestingly, independent of the preparation route, the saturation magnetization is lower than that of magnetite and maghemite, 90 emug −1 and 78 emug −1, respectively. Furthermore, the saturation magnetization is generall y found to decrease with a decrease in par ticle size. 20,24 It has been deter mined that the reduction in saturation magnetization is due to surface spin disorder, whereby specific spins are oriented in random directions at the surf ace.20,25,26 High surface curvature may also encourage disordered cr ystal orientation and can even lead to a magnetically dead layer.20,24,27

Magnetic Nanoparticles

SURFACE COATINGS In general, SPIO nanopar ticles that are to be used for biomedical applications must be coated with surface complexing agents. Surf ace modif ication is necessar y to prevent nanopar ticle agglomeration, reduce to xicity, and alter pharmacokinetics and biodistribution. It is also often desirable to select a surf ace coating that will impar t additional functionality by allowing the attachment of targeting ligands and/or therapeutics. Iron oxide colloids either can be for med directly in the presence of a surf ace coating material or modif ied afterwards. Although both approaches can be used to generate a monodisperse suspension of nanopar ticles, it may be assumed that more control o ver the physical and chemical properties can be achieved when the nanoparticles are modif ied with a surface coating after for mation of the iron oxide core. Of course, this approach generally requires an additional purif ication step before surf ace complexation. The functional groups that are known to bind to the surface iron oxide nanoparticles include phosphates, sulphates, and carbo xylates.28 Of these, carboxylate is the strongest binder and is therefore often incorporated into coating materials to maintain stable dispersions. The nature of the materials used to coat iron o xide nanoparticles can typically be classified as either organic (monomers and pol ymers) or inor ganic (metals and oxides). Examples of or ganic monomers include citric acid and gluconic acid. 29–31 These molecules typicall y provide a thin orga-nophilic shell that is able to prevent particle agglomeration through electrostatic repulsions. It has been speculated that the free alcohol present on gluconic acid could potentially allow for the coupling of biological agents; ho wever, because surf ace comple xes of gluconic acids are weaker than citric acid, mixtures of the two are likely necessary. Phosphonate and phosphate surfactants can also be used to obtain stab le dispersions of iron oxide nanoparticles. It was suggested that these ligands form a bilayer on the iron o xide surface. Monodisperse iron o xide nanopar ticles have also been stabilized by long-chain f atty acid monomers such as oleic acid although the ability to use fatty acids is highly dependent on the chain length and w ettability.32,33 Using a different surface coating scheme, fatty acids have also been used to create bilayers on iron oxide nanoparticles.32,34,35 A major limitation of this method , however, is the destabilization of the bilayer following dilution.32,36 Polymeric coatings represent the most common class of surface coatings used to impro ve the biocompatibility and stability of iron oxide nanoparticles. Some examples

525

include de xtran, carbo xymethylated de xtran, pol yvinyl alcohol (PVA), starches, chitosan, poly(methyl methacrylate) (PMMA), poly(ethylene glycol) (PEG), poly(lacticco-glycolic acid) (PLGA), pol yvinylpyrrolidone (PVP), and pol y(acrylic acid) (P AA).22,37 The stability of the nanoparticles can be even further improved by cross-linking the pol ymeric coating once it has been comple xed with the iron oxide core.38 Interestingly, it is not only the polymer utilized that effect the pharmacokinetics and biodistribution of the iron o xide nanopar ticle but also surface density of the pol ymer and its char ge.39,40 The polymer utilized will also dictate w hether tar geting agents can be coupled to the iron o xide nanoparticle. For example, dextran can be functionalized by oxidation with periodate to for m cross-links with the amino g roups of proteins. Alternatively, de xtran can be functionalized with amines via treatment with epichloroh ydrin and ammonia.38,41 These amines can subsequently be used for conjugations to biological and/or drug molecules. As an alter native to organic coating materials, man y groups have investigated the idea of coating iron o xide colloids with inorganic materials such as gold42–45 and silica.46–50 These materials can be used to for m coatings as thin as 1 to 2 nm on the iron oxide nanoparticles; they can confer an e xtremely high stability, and the y also provide a conducive surface for subsequent functionalization. For example, the gold surf ace readily binds to any biological molecules that ha ve a free thiol, w hereas the silanol groups on silica surf ace can react with alcohols and silane coupling agents.49 It should be noted that one complication that is encountered with the use of gold is continued agglomeration of the iron o xide nanopar ticles. Typically, ionic capping ligands must be used to coat the gold surf ace and pre vent the for mation of nanopar ticle clusters. In general, re gardless of the material used to coat iron oxide colloids, there is a ne gative effect on its magnetic proper ties. It has been suggested that the surf ace materials cause spin pinning, w hich gives rise to a noncollinear spin str ucture.26,51,52 There is also e vidence that the crystallinity may decrease with increasing polymer concentration.53

SPIO CLASSIFICATIONS There are man y for ms of SPIO in academic and clinical use, which have made it necessar y to create broad classifications to dispel any confusion. Accordingly, SPIOs are typically def ined b y their o verall h ydrated diameter (including their biocompatib le coating) and di vided into

526

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

three categories, oral SPIO at 300 nm to 3.5 µm, regular, polydisperse SPIO (PSPIO) at 50 to 150 nm, and ultrasmall SPIO (USPIO) at < 50 nm.A subcategory of USPIO known as MION is also often identif ied and refers to USPIO that possess an iron o xide core that is monocr ystalline in nature (ie, monocr ystaline iron oxide nanoparticles). A prevalent derivatization of MION, consisting of a chemically cross-link ed and aminated pol ysaccharide shell, is referred to as CLIO (cross-linked iron oxide). Aside from the general classif ications described above, SPIOs are also commonl y refer red by their trade names as well as other generic names. Furthermore, there can be multiple trade names associated with an y gi ven SPIO, due to mark eting in dif ferent re gions across the globe. For example, Ferumoxsil or AMI-121 is a ~400 nm FDA-approved oral-SPIO that is also kno wn as GastroMARK (USA) and Lumirem (EU , Brazil). F erumoxtran or AMI-227 is a 20 to 40 nm de xtran-coated USPIO that is also known as Sinerem (EU) and Combidex (USA) and is currently in late-phase clinical trials (phase 3). Table 1 contains a listing of various SPIOs, their alternate names, surface coating, and size.

MECHANISM OF SPIO CONTRAST In contrast to other imaging modalities (ie, X-ra y, nuclear, and ultrasound), the effect of MR contrast agents is not seen directly on the image, but rather it is their ef fect on proton relaxation, nor mally sur rounding w ater protons, that is

observed. In other words, it is the change in relaxation rate of water protons in the presence of SPIO that is detectab le by MR and is responsib le for enhancing image contrast. The measure used to def ine how well MR contrast agents change the relaxation rate of water protons is termed relaxivity. Relaxivity is represented quantitatively as:

⎛ 1 ⎞ Δ ⎜ ⎟ ⎝ T1 ⎠ R1 = [M ]

⎛ 1 ⎞ Δ ⎜ ⎟ ⎝ T 2 ⎠ R2 = , [M ]

where T 1 and T 2 refer to the relaxation time of protons in the longitudinal and transverse plane, respectively, and [M] is the concentration of contrast agent. SPIOs are typicall y used to shor ten the T 2 relaxation times, w hich tends to reduce the signal intensity, ie, darken the image. As a result, iron o xide nanopar ticles are often refer red to as ne gative contrast agents. However, despite the f act that the majority of studies take advantage of the T 2 -weighted contrast generated by SPIO, several examinations have shown that SPIO may also be used as T 1 contrast agents.54 In these cases, the SPIOs are used to shor ten the T1 relaxation time of surrounding water protons, which tends to increase the signal intensity creating a “positi ve” contrast. The limited use of SPIO as a T1 imaging agent likely stems from the sensitivity of this T1 contrast to magnetic f ield strength and the degree of particle aggregation and/or compartmentalization.54 For e xample, it has been obser ved that w hen iron o xide nanopar ticles are compar tmentalized into

Table 1. NAMES AND CHARACTERISTICS OF COMMON MAGNETIC NANOPARTICLES Classification Oral SPIO

PSPIO

USPIO

Agent

US Adopted Name

Trade Name

Coating Material

Hydrodynamic Diameter

Half-life*

AMI-121

Ferumoxsil

Lumirem Gastromark

Silicon

300 nm

––

OMP

Ferristene

Abdoscan

Sulphonated styrenedivinylbenzene

3.5 µm

––

AMI-25

Ferumoxide

Endorem Feridex IV

Dextran

80–180 nm

2

SHU-555A

Ferucarbotran or Ferrixan

Resovist Cliavist

Carboxydextran

60–62 nm

2.4–3.6

7 nm

0.6–1.3

VSOP-C184 NC100150 SHU-555C

–– Feruglose ––

Code 7228

Ferumoxytol

AMI-227

Ferumoxtran-10

––

Citrate

Clariscan

Pegylated starch

20 nm

6

Supravist

Carboxydextran

21 nm

6

Carboxymethyl dextran

30 nm

10–14

Dextran

15–40 nm

24–36

–– Sinerem Combidex

MION

––

––

Dextran

30–50 nm

––

CLIO

––

––

Cross-linked dextran

30–50 nm

––

*In human hours. AMI-121, AMI-25, AMI-227, and Code 7228 are distributed by AMAG Pharmaceuticals/Guerbet; OMP and NC100150 are distributed by GE-Healthcare; SHU-555A and SHU-555C are distributed by Schering; VSOP-C184 is distributed by Ferropharm. Table is modified from Wang and colleagues113 and Corot and colleagues.121

Magnetic Nanoparticles

endosomes/lysosomes of K upffer cells, the T 1-weighted contrast is reduced w hereas the T 2 -weighted contrast is enhanced. The ability of SPIO to shor ten the T 1 and T 2 proton relaxation times has been widel y used to enhance the contrast of MR images. The discussion below focuses on the T1 and T 2 relaxation processes using spin echo imaging to explain many of the issues, but the high magnetic susceptibility of SPIO can be exploited to generate contrast with g radient echo pulse sequences as w ell.55–57 In fact, the T 2 -weighted gradient echo sequences are generally the most sensiti ve MR methods for detecting the presence of SPIO in tissues. A recent and e xciting development e xploits the high susceptibility of SPIO , which causes protons near the magnetic par ticles to be amenable to off-resonance excitation.

Effects of SPIO on T 1 and T 2 in Spin Echo Imaging The effects on T 1 arise when water molecules contact the surface of iron oxide, whereas the effects on T 2 arise when protons pass through volumes of microscopic of magnetic field inhomo geneity around the super paramagnetic iron oxide. Whether super paramagnetic iron o xides are exploited as brightening agents on spin echo–based T 1weighted images or dark ening agents on T 2 spin echo images depends principally on three variables: (1) the ratio of R2 to R1 of a preparation in vitro, (2) the ratio of R2 to R1 particles assume in various biological compartments, tissues, and organs in vivo, and (3) MR instrument settings.

R2/R1 for Different Preparations In Vitro The ratio of R2 (spin–spin relaxivity) to R1 (spin–lattice relaxivity), or value of R2/R1, as determined in vitro varies greatly for superparamagnetic iron oxide particles and nanoparticles.3,4 Table 2 pro vides the v alues of R2 and R1 for three super paramagnetic iron o xide preparations that have been used clinicall y, as w ell as v alues for the paramagnetic chelate, Gd-DTP A. Super paramagnetic iron oxides can be highly selective in their effects on R2 (eg, Ferumoxsil), or lik e Combide x/MION have values of R2/R1 that approach that of the classical T 1 brightening agent, GdDPTA. The value of R2/R1 reflects the size of iron o xide particles in solution; a solution featuring a small number of larger par ticles e xerts stronger ef fects on T 2 . Values of R2/R1 not only vary with the intrinsic particle size but also reflect the agg regation or dispersion of magnetic par ticles when they react with molecular targets in vitro, which is the basis of magnetic relaxation s witch biosensors (discussed below).58,59 Note that the absolute value of R1 for MION is

527

about 4 times higher than Gd-DTP A, but Gd-DTP A is a more effective brightening agent because of its relatively greater effect on R1 (R2/R1 = 1.27). Ferumoxsil, a highly selective T 2 altering agent ( R2/R1 = 22.5), dark ens T 2 weighted images but has virtually no effect on T1-weighted images. MION, with its intermediate value of R2/R1 of 2.3, can be a brightening or dark ening agent, depending on its distribution knocks down luciferase expression and instr ument settings, see below.

R2/R1 in Biological Systems The value R2/R1 in vitro is a reasonab ly good predictor of a brightening or dark ening ef fect of super paramagnetic iron oxide knocks down luciferase expression, provided that the distribution particles are not greatly altered by the biolo gical system. F or e xample, brightening or darkening preparations of superparamagnetic iron oxides maintain their in vitro beha vior w hen used to pro vide contrast for the lumen of the gastrointestinal (GI) tract or the lumen of major blood vessels. However, when SPIOs are cleared or processed b y a biological system, the in vitro R2/R1 no longer applies. One of the most striking examples of this occurs when injected superparamagnetic iron oxides accumulate within phagol ysozomes found in Kupffer cells of the liver. Hepatic accumulation of superparamagnetic iron o xides causes them to become highl y T 2 selective agents, regardless of the value of R1 they exhibit in vitro. Thus, when in b lood, MION/Combidex produces a strong hepatic v ascular phase T 1 brightening because the blood volume of the liver is approximately 20% of tissue volume. Then, when in hepatic phagolysozomes, MION acts as a T 2 darkening agent. 60–62 Significant changes from the R2/R1 in vitro also occur w hen superparamagnetic materials accumulate in cells, whether macrophages or other cells, in l ymph nodes, spleen, or other organs.

Instrumental Settings (Pulse Sequence) With spin echo imaging, when time to echo (TE) is shortened, the T 1 brightening ef fects of super paramagnetic iron o xides becomes g reater. This results from the weighting of signal intensity and TR and TE parameters in the basic spin echo signal intensity equation. 63,64

Imaging SPIO with Off-resonance Excitation The high R2/R1 values that occur when SPIOs are present in cells, w hether macrophages in li ver, spleen, or l ymph nodes after injection, or cells loaded e x vi vo, yield a

528

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

−1

Table 2. RELAXIVITIES OF SUPERPARAMAGNETIC IRON OXIDES (MM × SECONDS)

0.47 T, 20 MHz

Comment

R1

R2

R2/R1

MION/Ferumoxtran Combidex

Monodisperse USPIO

22.7

53

2.3

Ferumoxsil/Gastromark

Polydisperse SPIO

3.2

72

22.5

Ferumoxides/Feridex IV

Polydisperse SPIO

23.7

107

4.5

Gd-DTPA/Magnevist

Paramagnetic chelate

Particle

4.5

5.7

1.27

Table is modified from Jung and colleagues.3

darkening on T 2 -weighted spin echo or g radient echo images. Unfor tunately, the image dark ening (signal loss) can originate from sources other than SPIO such as calcification or air. Techniques for obtaining positi ve contrast with SPIO use an of f-resonance e xcitation of protons, a class of protons that e xist because the high magnetic susceptibility of SPIO causes protons in the vicinity of the SPIO to resonant at something other than Lar mour frequency def ined b y the f ield of the MR magnet. Recent advances in off-resonance methods include off-resonance contrast angio graphy (ORCA) 65 or in version reco very with on-resonant water suppression (IRON) used for MR angiography.66 IRON angio graphy generates a positi ve signal using an of f-resonance e xcitation, w hereas the signal from on-resonant protons from the backg round is suppressed. This leads to a very high contrast between the lumen of major v essels and the surrounding tissue, which surpasses that from more conventional T 1 -weighted MRA. This noninvasive technique has the potential to further improve the delineation of vascular anatomy and possibly the identif ication of vascular pathology.

TOXICITY Because superparamagnetic iron oxides use a form of iron to enhance proton relaxation, the to xicity of this metal must be carefull y considered. F ortunately, the to xicity of iron is w ell understood because of the profound role it plays in human health and its long use in the treatment of iron-based anemia. 67–70 A normal male adult has about 4,000 mg of total body iron, with about 2,500 mg as hemoglobin. A considerable amount is present as myoglobin (the red storage protein in muscle) and about 400 mg is present as the iron storage proteins in the li ver, principally hemosiderin and fer ritin. Hepatic iron transits to bone mar row via transfer rin and is incor porated with red b lood cells. Females ha ve less iron, being smaller and ha ving lo wer hematocrits because of menstr uation. Humans ha ve no mechanism for excreting iron but lose it very slowly as iron-bearing cells are lost or de graded. Examples of cell

loss include menstruation, degraded red blood cell heme in feces, and epithelial cell loss from the skin or intestine. This profound inability to excrete iron, which is not found in rodents, can lead to the storage of abnor mally lar ge amounts of iron and a v ariety of iron storage–related diseases. In these iron overload states, hepatic stores increase far above the nor mal 400 mg, reaching se veral grams per liver. F or MR contrast agents, iron o verload to xicity is avoided by administering a small iron dose, in relation to the total body iron stores. Thus, for hepatic contrast, a dose of roughly 50 mg/person of iron is used (Feridex IV, Resovist), whereas roughly 200 mg/person is used for imaging lymph nodes (Combide x, Sinerem). 71,72 The iron from superparamagnetic iron o xides is metabolized o ver a period of weeks with iron joining nor mal body iron stores and being used to synthesize hemoglobin.73 Animals tolerate extremely high doses of superparamagnetic iron oxides with v arious compositions. 71,73,74 When humans ha ve severe anemia, anemia that w ould be poorl y treated with oral iron, various injectable forms of iron are administered at far higher doses than used for MR contrast agents. Thus, with injected for ms of iron, to xicity is related to the total body iron burden placed on the recipient and not to an intrinsic toxicity of the highl y nontoxic forms of iron that have been developed for anemia treatment or as MR contrast agents. Iron is not the onl y component to be considered wi th super paramagnetic iron o xides. Additional potential sources of toxicity include the polymeric coating, surface chemistr y, and other chemical features of the preparation. A number of studies are in vestigating the toxicity ef fects in detail and b y using high-throughput screening techniques.75

METABOLISM AND DEGRADATION The well-understood metabolism and demonstrated path of utilization of the iron in superparamagnetic iron oxides is cr ucial to their use as MR imaging agents and not a characteristic of man y ne wer, no vel, and perhaps more interesting nanomaterials. Unfortunately, there are a large

Magnetic Nanoparticles

number of materials that are not metabolized or utilized and that ha ve se vere deleterious ef fects after long-ter m residence in human cells and tissues. These include materials used in industrial settings (silica, carbon/coal, metal oxide ores, asbestos), as w ell as materials used medicinally (thorast (ThO 2), polyvinylpyrolidine).76–81 A recent and tragic example of the consequences of the long-term retention of a nonto xic material not nor mally found in humans can be found with Gd-chelate–based MR contrast agents. When Gd-chelates are administered to patients with kidney failure, the slow renal elimination permits Gd de-chelation and tissue retention of the Gd metal. The resulting nephro genic systemic f ibrosis, or thickening of or gans, has no kno wn cure and can be fatal.82–84 The FDA recently ordered a black box warning be added to all f ive approved Gd-chelates regarding their use in patients with kidne y problems. A wide v ariety of materials have been used in animal imaging such as carbon nanotubes,85 gadolinium oxide,86 and silica (SiO2).87–89 A popular fluorescent nanoparticle for animal studies is cadmium (Cd)-based quantum dots. 90,91 However, because of the unfor tunate e xposure of batter y workers, Cd is a known cancer-causing agent in humans, the highest danger classif ication for carcino gens.92 Cd, like mercury and lead, is a highly toxic, poorly eliminated heavy metal that should not be disposed of b y dilution into normal solid or liquid waste. When administered b y intra venous injection, both PSPIO and USPIO are predominantly cleared from circulation by macrophages that reside in or gans rich in reticuloendothelial cells. 71,73 Liver endothelial cells have also been found to be in volved in the uptak e and degradation of some USPIOs (ie, NC100150). 93 This dif ference in uptake is lik ely mediated b y the composition of the nanoparticle coating. In general, it has been deter mined that USPIO and SPIO are inter nalized b y receptor mediated endocytosis and are metabolized via the lysosomal pathway.94 Recently, the distribution and elimination of de xtran-coated USPIO follo wing uptak e w as closel y evaluated by using 14C-labeled (on dextran) ferumoxtran10 and 59Fe-labeled fer umoxtran-10.71 It was found that after uptake by macrophages the de xtran coating undergoes a pro gressive degradation and is eliminated almost exlusively in urine (89%). The rest is e xcreted in feces. The iron contained in fer umoxtran-10 is incor porated into the body’s iron store and is progressively found in red blood cells (hemoglobin). Like endogenous iron, it is eliminated very slowly and predominantl y via the feces. Dextran-coated PSPIO appears to under go similar metabolic de gradation and clearance. 73 Interestingly, it w as recently found in rats that SPIO with similar coatings but

529

different sizes exhibited similar rates of liver clearance (based on R2* values), but SPIO with different coatings exhibited significantly different rates.95 The order of the half-lives w ere de xtran (fer umoxide and fer umoxtran) < carbo xydextran (SHU 555A) < o xidized starch (NC100150).

PASSIVE TARGETING/BIODISTRIBUTION The biodistribution of SPIO is predominantly dictated by its hydrodynamic diameter although, the composition, charge, density, and thickness of the surf ace coating ha ve also been found to alter opsonization and influence SPIO pharmacokinetics.96 The disparate biodistribution patter ns observed for different types of SPIO have allowed for imaging of a wide v ariety of biolo gical systems e ven without explicit molecular specif icity. Examples of ho w dif ferent types of SPIO rely on natural physiological processes to delineate disease states are summarized below.

Oral SPIO Oral SPIOs are typically used to image the gastrointestinal tract and are administered either by the oral or rectal route. The tw o oral SPIO agents that ha ve been most widely evaluated are AMI-121 and OMP, both of w hich have been appro ved by the FD A for clinical use. AMI121 consists of an iron oxide core ~10 nm in diameter with a siloxane coating.97 The hydrodynamic diameter of the entire par ticle is ~300 nm. OMP has an iron o xide core ~50 nm in diameter and possesses a pol ystyrene coating.98 The h ydrodynamic diameter of OMP is 3.5 µm. Both siloxane and polystyrene are nonbiodegradable and insoluble, which help prevent the ingested iron from being absorbed. The coatings also help pre vent par ticle aggregation, w hich f acilitates the homo geneous distribution of oral SPIO in the bowel. Oral SPIO agents are generall y used to impro ve the def inition of or gan boundaries such as the uter us and l ymph nodes and to help delineate nor mal and patholo gic str ucture in the bowel.99,100 Oral SPIOs also lead to improved delineation of the head and tail of the pancreas, anterior margins of the kidneys, and para-aortic region.97

PSPIO PSPIO agents are typically administered intravenously and are small enough to traverse the capillary beds in the lung, brain, hear t, and kidney. The circulation of PSPIO , however, is shor t-lived as the par ticles are rapidl y

530

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

removed b y phagoc ytic cells of the reticuloendothelial system (RES). Consequentl y, the T 2 -signal of the li ver and spleen is rapidly reduced after intravenous injection. Because malignant neoplasms in these or gans are devoid of phagocytic cells, the y show up as areas of bright signal intensity creating a shar p contrast betw een nor mal and diseased tissue. As a consequence, PSPIO has been commonly used to impro ve the detection of hepatic and splenic tumors. 101–107 With contrast enhancement, the mean threshold size for detecting hepatic tumors decreased from ~1 cm to less than 2 mm in diameter.108,109 PSPIO has also pro ven to be capab le of dif ferentiating between benign and malignant lesions because of the presence of phagoc ytic cells in h yperplastic nodules.103 The PSPIO-enhanced MR imaging has shown the ability to distinguish betw een benign and malignant lesions with a > 88% sensiti vity and specif icity for metastatic lesions. 102 However, it should be noted that there are some malignant lesions that may contain phagocytes—such as well-differentiated hepatocellular carcinomas—in w hich case there ma y be no dif ferential uptake compared with normal liver parenchyma. Currently, two for ms of PSPIO ha ve been clinicall y approved for contrast-enhanced MR imaging, AMI-25 and SHU 555A. AMI-25 is a de xtran-coated iron o xide nanoparticle with an iron core that is 3 to 6 nm in diameter and hydrodynamic diameter of 80 to 150 nm. AMI25 is administered via slo w infusion, w hich lasts up to 30 minutes, and the plasma half-life ofAMI-25 is roughly 2 hours. 110,111 Uptake b y hepatic RES K upffer cells accounts for 80% of the injected dose,73,74,101,112 while approximately 5 to 10% of theAMI-25 particles are taken up b y the spleen. 110,113,114 Peak concentrations of iron were achie ved in the spleen at 2 hours and in the li ver 4 hours after injection. 73 T 2 - and T 2 *-weighted images are typically taken 30 minutes to ~6 hours after infusion and the signal intensity returns to normal within 7 days.115 The obser ved half-life of iron originating from AMI-25 was 3 days for the liver and 4 days for the spleen. 73 SHU 555A possesses a carbo xydextran coating with an iron oxide core diameter of 4.2 nm and a hydrodynamic diameter of 62 nm. SHU 555A is administered as a bolus injection because unlike AMI-25, it does not result in side effects lik e h ypotension and lumbar pain. 110,114,116 The 117,118 plasma half-life of SHU 555A is 2.4 to 3.6 hours. The small size of SHU 555A, and thus stronger T 1-relaxivity, combined with the ability to perform bolus injections allows SHU 555A to be used in dynamic T 1-weighted imaging e xaminations and MR angio graphy.113,116 Dynamic imaging of the li ver during the perfusion phase sho ws hypervascular hepatocellular carcinomas as

hyperintense re gions and has thus been used to signif icantly impro ve the dif ferentiation betw een benign and malignant tumors. 116 MR angio graphy has been used to enhance the visibility of the por tal venous system, w hich may improve planning of liver resections.119

USPIO USPIO particles are usually defined as having a hydrated particle diameter of less than 50 nm and a plasma halflife that is signif icantly longer than PSPIO.120 For example, the plasma half-life for USPIO has been repor ted to range anywhere from 80 minutes to more than 24 hours, compared with only ~2 to 4 hours for PSPIO.121 The large variability in USPIO circulation time lar gely has to do with the composition of the surface coating that is used to stabilize the particle.22,120,122 In general, the extended circulation time for USPIO is a consequence of the particles not immediately being recognized by phagocytic cells of the RES. The small size and prolongation of the plasma half-life also enables this agent to cross the capillary wall and ha ve more widespread tissue distribution. Twentyfour hours after administration, USPIO par ticles can be found in the lymph nodes, bone marrow, liver, and spleen. This is in contrast to PSPIO, which is found almost exclusively in the li ver and spleen. USPIO par ticles transmigrate the capillar y wall by means of v esicular transpor t and through interendothelial junctions. Upon gaining access to the interstitium, USPIOs are cleared by draining lymphatic v essels and are transpor ted to l ymph nodes, which sho w up as areas of reduced signal intensity on T 2 /T 2 *-weighted images. 123 It has been repor ted that 24 hours after administration of de xtran-coated USPIO , 0.79% of the injected dose (3.62% of injected dose per gram of tissue) is found in l ymph nodes. 120 If metastases cause disturbances in node flo w or architecture, USPIOs lose accessibility to the node, w hich subsequentl y appears as h yperintense. The dif ferential uptak e of USPIO has made them quite suitab le for T 2 /T 2 *weighted MR lymphography. With the help of USPIO, the sensitivity of detecting lymph node metastases increased from 45.4 to 100% with a specif icity of 95.7%. 124 Aside from MR l ymphography, the small size and long circulation time of USPIO also enab les impro ved accessibility to bone marrow compared with PSPIO. Twenty-four hours after the administration of one specific type of USPIO, 2.91% of injected dose per g ram of tissue was found in bone marrow.120 This was sufficient to reduce signal intensity of bone mar row on T 2 /T 2 *weighted images.125 Retention of USPIO in bone marrow is a consequence of phagoc ytosis b y macrophages.

Magnetic Nanoparticles

USPIOs are not tak en up as ef ficiently in neoplastic marrow infiltrates, which possess fewer macrophages, these lesions remain hyperintense. Therefore, USPIO can be used to dif ferentiate betw een nor mal and neoplastic marrow.126 This mechanism for contrast enhancement is similar to imaging hepatic tumors with PSPIO particles as discussed abo ve. In addition, USPIO are capab le of differentiating between hypercellular growths and tumor infiltratates in marrow.127 Although USPIOs exhibit some unique features that allow them to generate a signal enhancement of lymph nodes and bone mar row, they can still be used for RESspecific imaging of the li ver and spleen. 116,128,129 Biodistribution studies showed that 66.1% of the injected dose was found in the liver 24 hours after injection and 5.15% was in the spleen. 120 In general, USPIO e xperienced reduced phagocytosis and consequently a reduced signal loss compared with PSPIO; ho wever, this discrepanc y could be reduced with increasing dose. 128,129 By exploiting the unique biodistribution patterns and macrophage uptake of USPIO, a number of clinical applications ha ve emer ged w here USPIOs are used to help delineate various pathologic conditions, including inflammation, strok e, transplant rejection, and cancer , among others. Examples of these applications are covered in more detail in other chapters. In addition to taking adv antage of the biodistribution patter ns of USPIO , many groups have also used USPIO as a blood pool agent. This is possible because of the long b lood half-life and T 1 shortening effects of USPIO.121 These blood pool characteristics provide a long time windo w for data acquisition. Specif ic applications have included MR angiography,130–134 cerebral blood v olume imaging, 135,136 and dynamic contrastenhanced imaging of benign and malignant tumors, that is, permeability imaging.137–140

ACTIVE TARGETING Although the passi ve tar geting strate gies described abo ve rely on natural phar macokinetic and phagoc ytic mechanisms to dictate SPIO localization, acti ve tar geting uses SPIO particles that have been functionalized with targeting ligands to promote selecti ve binding to biomark ers on target tissues. The premise of active targeting is that the acquisition of infor mation re garding the molecular prof ile of cells will allo w ph ysicians to estab lish a more accurate diagnosis and tailor therapeutic treatments for indi vidual patients. Insight into the e xpression prof iles of patholo gic tissues could also allow for the early identification of molecular abnor malities before the onset of visib le disease. Furthermore, periodic imaging of disease biomarkers could

531

allow physicians to monitor drug efficacy over the course of treatment. Many studies have already reported the ability of active targeting strategies to identify biomark ers of cancer, cardiovascular disease, apoptosis, and other patholo gical conditions. Various examples are outlined below.

Cancer Imaging Over the past fe w decades, an abundance of infor mation has been gar nered re garding the biolo gical prof iles of cancer. It is no w accepted that the onset and pro gression of cancer can be characterized b y a m yriad of v ariant molecular processes. These unique molecular prof iles have recently been e xploited as tar gets for MR imaging agents. To date, many biomarkers of cancer have already been evaluated as tar gets for SPIO . One e xample is the transferrin receptor (TfR). 141 Many studies ha ve found that elevated levels of TfR are often found on cancer cells compared with their nor mal counter parts. In breast cancer, the TfR has been found to be up to four - to f ive-fold higher in malignant cells w hen compared with nonneoplastic cells. 141–143 Increased TfR e xpression has also been found on b ladder transitio nal cell carc inomas, prostate cancer, gliomas, lung adenocarcinomas, chronic lymphocytic leukemia, non-Hodgkins lymphoma, and on peripheral b lood mononuclear cells from l ymphoma or leukemic tumors. 141,144,145 Accordingly, w hen human transferrin proteins coupled to USPIO nanoparticles were injected into tumor-bearing mice, a 40% change in signal intensity w as obser ved in T 2 -weighted images. Onl y a 10% reduction in signal was observed with human serum albumin–labeled USPIO nanopar ticles.146 One drawback of using TfR as a tar get is that it is also e xpressed on a limited number of normal tissues including the basal epidermis, the endocrine pancreas, hepatoc ytes, K upffer cells, testis, and pituitary.147 This likely leads to sequestration of some USPIO from circulation. Another well-studied target that has been used for cancer imaging is the folate receptor . The folate receptor is a highly selective tumor mark er that has been found to be overexpressed in many types of tumors including ovarian, endometrial, colorectal, breast, lung, renal cell carcinomas, brain metastases derived from epithelial cancers, and neuroendocrine cancers. 148 Conversely, the folate receptor is generally absent in most nor mal tissues with the e xceptions of choroids ple xus, placenta, and lo w levels in lung, thyroid, and kidne y. When USPIO w ere tethered with folate and administered intra venously into mouse models with nasopharyngeal epidermal carcinoma xenografts, MR imaging sho wed an a verage intensity decrease of 38% within the tumor.149

532

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Another cancer mark er recently targeted with SPIO is the underglycosylated MUC-1 tumor antigen (uMUC1).150 This antigen is one of the early hallmarks of tumorigenesis and is o verexpressed in man y epithelial cell adenocarcinomas including breast, pancreatic, colorectal, lung, prostate, and gastric cancers. The specif ic imaging probe used to tar get the uMUC-1 consisted of CLIO modified with th e u MUC-1–specific peptide EPPT . Upon intra venous administration, a 46.5% reduction in the average T 2 relaxation rates associated with the tumor was observed. It was further shown that the CLIO–EPPT conjugate could be used to monitor therapeutic response and evaluate differential treatment outcome. The number of cancer biomark ers that has been targeted with SPIO continues to e xpand. Other e xamples of biomarkers that have been successfully targeted with SPIO conjugates include c-MET on hepatocellular carcinomas using micron-sized SPIO labeled with anti-C-MET antibodies,151 αvβ3 integrins on angio genic v essels using USPIO conjugated to Arg-Gly-Asp- (RGD)-peptides,152,153 epidermal g rowth f actor receptors (EGFR) on esophageal squamous cell carcinomas using USPIO labeled with antiEGFR antibodies,154 HER2/neu receptor on fibroblast cells using herceptin–USPIO conjugates,155 and CD20 receptors on l ymphoma cells using USPIO par ticles labeled with anti-CD20 antibodies.156 In some cases, tar geting ligands ha ve been used to direct SPIO accumulation within tumors e ven with little knowledge of the tar get. F or e xample, an antibody (ie, L6)-labeled MION w as able to bind an unkno wn ligand expressed in intracranial tumors that were grown in nude rats.157 In this case, the ligand w as previously identif ied by immunizing mice with human non-small cell lung carcinomas and then h ybridizing their spleen cells with mouse myeloma cells. 158 It was only known that the L6 antibody def ined carboh ydrate antigens. This antibody also reacted with most carcinomas of the breast and colon. In a similar study , an antibody (ie, A7)-bound USPIO w as ab le to bind an unkno wn gl ycoprotein expressed in colorectal cancers. 159,160 An alternative active targeting strategy that has been adopted to improve the image contrast of tumors involves targeting SPIO to the health y cells that sur round the tumor in tar get tissues. This approach allo ws for malignant cells to appear hyperintense in T 2 -weighted images. In one study that used this approach, tumors in the pancreas w ere ef fectively visualized b y tar geting cholec ystokininA (CCKA) receptors that are onl y e xpressed in healthy acinar cells. When MION w ere labeled with CCK, a hor mone produced b y the small intestine that

binds to CCKA receptors, it generated a signif icant reduction in the T 2 relaxation of healthy pancreatic tissue leading to the clear delineation of pancreatic tumors. 161 More recentl y, malignancies in the pancreas ha ve been detected through tar geting of the bombesin (BN) receptor.162 Bombesin receptors are absent in human pancreatic ductal adenocarcinomas (PDAC). CLIO–BN conjugates were found to accumulate strictly in healthy pancreatic tissues, developing sufficient contrast between cancerous a nd health y tissues. The r esults sho wed reduced T 2 signal of the nor mal pancreas and enhanced the visualization of an implanted tumor in rats. The ability to identify lesions in the liver has also been in vestigated b y acti vely tar geting the asialo glycoprotein(ASG) receptor on healthy hepatocytes by coating USPIO particles with arabinogalactan (AG), a galactosecontaining polysaccharide.161,163,164 ASG receptors have a high af finity for the ter minal galactose g roups. In vi vo MR imaging experiments showed that targeting USPIO to hepatoc ytes rather than rel ying on the nonspecific uptake by Kupffer cells allo wed for a considerab le dose reduction, increased tumor-liver contrast, and potentially allows distinction of ASG-positive (benign) and ASGnegative (malignant) tumors. 163 Although man y e xamples of acti ve targeting of cancer cells with SPIO have been presented, options for molecular tar gets ha ve generall y been limited to receptors that are highl y overexpressed on tumor cells because of the intrinsicall y lo w sensiti vity of SPIOenhanced MR imaging. As a result, several groups have introduced amplif ication strate gies to amplify the coincident signal to impro ve the lo wer detection limit. One commonl y used approach to amplify the SPIO signal involves specifically targeting receptors that are rapidly inter nalized (eg, transfer rin receptor and folate receptor). An internalizing receptor is a prefer red target because follo wing SPIO endoc ytosis, the receptor can be rec ycled back to the cell surf ace and recr uits additional SPIO into the cell. Therefore, a single receptor can be effectively used to localize a large number of SPIOs within each tar get cell. Another approach borrowed from positron emission tomo graphy (PET) studies uses staggered injection of two populations of probes.165 The f irst probe tar gets a specif ic ligand , whereas the second (ie, SPIO) onl y has an af finity for the first, not for any endogenous marker. This approach allows for impro ved contrast because the initial targeting agent typically possesses several high affinity sites for the SPIO probe, effectively allowing multiple SPIOs to bind each cell ligand.

Magnetic Nanoparticles

Imaging of Cardiovascular Disease Similar to cancer imaging, SPIO-enhanced molecular imaging has also taken great strides in improving the contrast of various cardiovascular diseases. Imaging of molecular signatures of cardiovascular disease is expected to provide a natural adjunct to personalized medicine b y improving diagnosis and risk stratif ication as w ell as helping to tailor drug selection.166 One particular promising application in volves identifying patients w ho harbor high-risk atherosclerotic plaques that may cause myocardial inf arction or strok e. P otential candidates for SPIO targeting include VCAM-1 (vascular adhesion molecule1), P-selectin, and E-selectin. VCAM-1 is of par ticular interest because it selectively binds to classes of leukocytes found predominantl y during the early stages of inflammation and, in many cases, before the onset of visible disease. Detection of VCAM-1 has recently been performed with anti-VCAM-1–tar geted CLIO , as w ell as with CLIO labeled with VCAM-1–specific peptides.167–169 In the latter case, phage displa y was used to identify a VCAM-1–specific peptide that w as rapidl y internalized upon binding to endothelial cells. As a result of the intracellular trapping of the peptides and the recruitment of additional targeting agents because of receptor rec ycling, the VCAM-1–specific peptide w as reported to e xhibit 12-fold higher tar get-to-background ratios compared with anti–VCAM-1 antibodies. F ollowing intra venous injection of the CLIO–peptide conjugates, clear decreases in MR signal intensity could be observed at sites of atherosclerotic plaques in cholesterol-fed apolipoprotein E (apoE −/−) mice. Imaging of thrombi has recentl y been accomplished using USPIO coupled to c yclic RGD peptides. 170 The RGD peptide competes with fibrinogen for binding to the platelet inte grin gl ycoprotein IIb/IIIa. It w as found that RGD–USPIO results in better thrombus visualization than nontargeted USPIO; however, the ability to visualize the clot was highly dependent on the spatial resolution of the image. In vivo, an in-plane resolution of less than 0.2 × 0.2 mm2 was required for good visualization after contrast enhancement, but if the resolution w as increased much fur ther a signif icant reduction in the signal-tonoise ratio w as obser ved. P otential applications of RGD–USPIO include the ability to diagnose acute coronary syndromes and to guide and assess the effect of glycoprotein IIb/IIIa antagonists or other upstream platelet inhibitors.166 Myocardial inf arcts ha ve been imaged with MION using antimyosin Fab (R11D10) as a targeting molecule.171 Antimyosin is immunospecif ic to necrotizing m yocytes.

533

1 hour after intravenous injection into rats, MR images of excised hear ts sho wed a mark ed decrease in the signal intensity of the inf arct relati ve to the sur rounding myocardium. No decrease in cardiac signal intensity w as observed w hen unconjugated USPIO w as administered intravenously.

Apoptosis Apoptosis, or programmed cell death, is necessary to control many aspects of normal physiology in humans including embr yonic de velopment, homeostasis, aging, and immunity. The dysre gulation of apoptosis can lead to inappropriate cell loss or patholo gical cell accumulation, culminating in a v ariety of diseases such as cancer , neurodegeneration, acute m yocardial inf arction, strok e, and immunodifficiency. Accordingly, se veral apoptosistargeting agents ha ve recentl y been de veloped. In one approach, USPIO w as conjugated to the C 2 domain of synaptotagmin I, w hich binds to anionic phospholipids including phosphatidylserine. 172 Phosphatidylserine is a phospolipid that is kno wn to redistribute from the inner to the outer leaflet of the plasma membrane early in apoptosis. After treatment of solid tumors with chemotherapeutics, a reduction in signal intensity w as obser ved in those regions of the tumor that contained large numbers of apoptotic cells. An alternative targeting agent that is cur rently being explored to image apoptosis is Annexin V, w hich possesses a high specificity and affinity for phosphatidylserine.173 Although n ot y et v alidated i n v ivo, Annexin V–targeted CLIO allo wed for the identif ication of cell suspensions containing apoptotic cells in vitro.

Other Targeting Applications Although active targeting of USPIO has generally been aimed towards cardiovascular and cancer imaging, other pathological conditions ha ve also been in vestigated. For example, in one earl y study , it w as found that MION labeled with human polyclonal immunoglobulin could be used to identify sites of inflammation in an animal model of myositis.174 Another application that has recently been explored involves imaging amyloid-β with MION labeled with the A β1-40 peptide, w hich is kno wn for it high affinity to amyloid-β. It was found that intraarterial injection of USPIO-A β1-40 with mannitol, to transientl y open the b lood–brain bar rier, enab led the detection of many plaques via MR. 175 Of course, these are just a fe w of the potential applications for acti vely tar geted SPIO.

534

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

With the continued adv ancement in cellular and molecular biolo gy, especiall y in the areas of genomics, proteomics, and metabolomics, the number of applications for actively targeted SPIO is unbounded.

CELL TRACKING A promising ne w direction for SPIO is their use in tracking the mig ration and biodistribution of cells in vivo. The ability to track cells is e xpected to ha ve a significant impact in man y clinical applications, most notably stem cell therapies and tissue repair . Cell tracking typically involves labeling cells e x vivo and subsequently implanting or systemicall y administering them into li ving subjects. It has been w ell documented that many cell types can be labeled by simple incubation with SPIO including monoc ytes, macrophages, cancer cells, and schw ann cells. 176–180 SPIO can be inter nalized b y phagocytosis, receptor-mediated endocytosis, or fluid phase–mediated e ndocytosis. L abeling e fficiency i s dependent on both SPIO size and surf ace coating. In general, it has been found that the lar ger PSPIO nanoparticles and char ged surf ace coatings (anionic or cationic) are preferab le to the smaller USPIO par ticles and neutral surf ace coatings (e g, de xtran), respectively.176,178,179,181–183 Interestingly, it has recentl y been sho wn that a number of cell types can e ven take up micron-sized particles, including hematopoietic and mesench ymal stem cells, hepatoc ytes, f ibroblasts, and macrophages. 184–188 This is potentially advantageous because cells labeled with just one or more micron-sized par ticles can be detected by MR because of the large iron content of each particle.186,188 Furthermore, it has been repor ted that uptake of micron-sized par ticles does not alter cell viability, proliferation, colon y-forming ability, or dif ferentiation capacity.184,187 Although simple incubation with SPIO can lead to cell labeling, many studies have now shown that the extent and rate of uptake can be profoundly improved if SPIO is complexed with transfections agents, cell-penetrating peptides (eg, HIV-tat and poly-arginine), or protamine sulfate.189 In general, these agents also allow many cells that do not naturally endocytose SPIO to be labeled. 190,191 For example, only umbilical cord b lood cells but not peripheral blood cells were labeled with SPIO after simple incubation; ho wever, cationic transfection agents (ie, Lipofectin, Life Technologies, Gibco BRL) allowed both cell types to be tracked in vivo.191 Other commercially available transfection agents that ha ve been used to improve cell labeling include Superfect (Qiagen), a

low-generation heat-acti vated dendrimer; Lipofectamine Plus (Invitrogen Life Technologies), a liposomal agent; and pol y-L-lysine (PLL), a pol yamine.190 These agents were successfully used to label human mesenchymal stem cells (MSCs), mouse l ymphocytes, rat oligodendroc yte progenitor CG-4 cells, and human cer vical carcinoma cells with no apparent effect on cell viability or proliferation, although this is lik ely dose dependent. For example, one study found that proliferation can be substantiall y limited for MSCs labeled with 100 µg Fe/mL or greater.192 Furthermore, it should be noted that there is some e vidence t hat S PIO–lipofectamine a nd S PIO–PLL c omplexes may be at least par tially absorbed to the surf ace of cells as opposed to inter nalized.189,193 Similar to transfection agents, cell-penetrating peptides have also been used to ef ficiently shuttle SPIO into many cell types. 181,194–198 For example, when the HIV-tat peptide was covalently link ed to the de xtran coating on USPIO nanoparticles, the nanoparticles were internalized into l ymphocytes o ver 100-fold more ef ficiently than unmodified particles.194 It has also been sho wn that incorporation of tat-peptide derivatized USPIO into progenitor cells does not af fect cell viability , differentiation, or proliferation. 195 Furthermore, w hen T cells w ere loaded with USPIO-tat peptide conjugates, cell labeling did not interfere with their nor mal activation or homing properties.199,200 Because the majority of pol ycationic transfection agents an d c ell-penetrating pe ptides h ave n ot b een approved by the US FDA, the naturally occurring polycationic peptide protamine sulphate has recently been introduced as a viab le FD A-approved alter native for cell labeling with SPIO .201 These peptides can be nonco valently complexed with SPIO through electrostatic interactions or covalently attached using chemical cross-linking agents.201,202 Both adherent cells (MSCs, monocytes, and macrophages) and cells grown in suspension (hematopoietic stem cells, T cells) w ere found to be ef ficiently labeled with the protamine–SPIO complex with no shortor long-term toxicity, changes in function, differentiation capacity, or phenotype compared with unlabeled cells.189,201 It has also been reported that protamine–SPIO is less likely to precipitate and absorb to the cell surf ace and more lik ely to inter nalize than SPIO–lipofectamine and SPIO–PLL complexes.189,193 Cell surface receptor–mediated endocytosis of SPIO is another a venue through w hich cells can be labeled. Recently, the transferrin receptor has been exploited as an efficient intracellular deli very de vice for oligodencrocytes.203 This approach of fers the unique possibility of being able to replenish the suppl y of inter nalized SPIO

Magnetic Nanoparticles

through active targeting of the transfer rin receptor using transferrin- or anti–transferrin receptor-bound SPIO. Furthermore, cells that do not naturally express high levels of transferrin can be easil y engineered to constituti vely overexpress an unregulated form of the transferrin receptor.204 It has already been shown that transferrin receptor transgene e xpression can be visualized b y MR in vi vo using transferrin-labeled USPIO.205 The major applications of magneticall y labeled cells will likely be to aid in the design and administration of cellular-based repair, replacement, and treatment strate gies.189 These and other applications are discussed in Chapter 44, “Cell Voyeurism Using Magnetic Resonance Imaging.”

MAGNETIC RELAXATION SWITCHES An interesting characteristic of SPIO is that it becomes more efficient at dephasing the spins of surrounding water protons, that is, reduced T 2 relaxation times, w hen selfassembled into lar ger comple xes. This mechanism of switching from a high T 2 relaxivity to a low T 2 relaxivity during SPIO self-assembly (or vice versa during SPIO disassembly) is generall y refer red to as magnetic relaxation switching (MRSW). Recentl y, this phenomenon has been exploited as a tool for detecting biomolecules in homo geneous assays. Ligands that have been used to trigger SPIO self-assembly (or disassemb ly) include oligonucleotides (ON), enzymes, proteins, enantiomers, and vir uses. A unique advantage of MRSW o ver many alternative detection technologies is that changes in T 2 can easily be measured in turbid media, whole-cell lysates, and potentially in vivo. The feasibility of MRSW was first evaluated with the detection of nucleic acids in solution.58,206 In these studies, two unique CLIO–ON conjugates were designed to recognize adjacent sites on nucleic acid tar gets. Thus, upon hybridization to complementar y tar gets, the CLIO–ON conjugate pairs were brought into close pro ximity, which resulted in a detectab le reduction in T 2 relaxation time. RNA was detected in both pure samples and directl y in cell l ysates. Fur thermore, MRSW w as reversible, w hich was demonstrated with ther mal c ycling betw een ON annealing and melting temperatures, and e xhibited single base specificity. Sensitivity measurement showed a lower detection limit of 500 attomoles. One promising application of this SPIO-based nucleic acid sensor in volves the high-throughput detection of telomerase acti vity, a marker for genetic instability. In a recent study , CLIO–ON conjugates w ere designed to h ybridize to the 30 bp telomeric repeats (TTAGGG) that are found at the ends of linear eukaryotic

535

chromosomes.207 Bound CLIO–ON conjugates w ere arranged linearly along the repeats. The sensitivity of this MRSW method was reported to be in the range of 10 to 100 attomoles of telomerase-synthesi zed D NA. This astounding lower detection limit is most likely due to the presence of multiple adjacent binding sites within the telomere, creating a multi-CLIO comple x, and augmenting the MRSW effect. In addition to detecting nucleic acid h ybridization, MRSW has also been used to screen for DN A cleaving and meth ylating enzymes. 208 Specifically, tw o unique CLIO–ON conjugates w ere designed to self-h ybridize and thus for m multi-CLIO self-assemb led comple xes bridged b y doub le-stranded DN A. In the presence of sequence-specific restriction enzymes, the DN A bridge was clea ved causing the CLIO to disperse and the T2 relaxation time to increase. The ability of each restriction enzyme to clea ve methylated DNA was easily screened for by methylating the DN A bridges before e xposure to the enzymes. The MRSW technique has also been used to detect proteins,58,209,210 bacteria,211 and viruses.212 Protein assays generally require the use of pol yclonal antibodies to direct multiple SPIO to wards each antigen, unless the antigen itself is multi valent. Alternatively, bacteria and viruses can be tar geted by SPIOs that ha ve been labeled with monoclonal and/or pol yclonal antibodies. When CLIO nanopar ticles w ere labeled with surf ace-specific antibodies raised against Mycobacterium avium spp., Paratuberculosis (MAP), adeno virus-5, and her pes simplex virus-1 and exposed to the respective bacterial or viral target, a lar ge reduction in T 2 relaxation time w as observed. The developed method w as sho wn to specif ically detect adeno virus-5 and her pes simple x vir us-1 at concentrations as lo w as f ive viral par ticles per 10 µL without the need for e xtensive sample preparation. As few as 15.5 colony forming units were also detected. Another application of MRSW with possib le pharmacologic impact is the detection of enantiomeric impurities.213 D and L-enantiomers can dif fer profoundl y in pharmacological acti vity, and thus there is an ur gent demand for methods to rapidly evaluate enantioselective syntheses and detect enantiomeric impurities. In a proofof-concept study, CLIO nanopar ticles were labeled with D-Phe (ie, the impurity) and self-assemb led in the presence of stereoselecti ve antibodies o wing to the bi valent nature of antibodies. The subsequent addition of D-Phe impurities caused the CLIO to disperse and the T 2 relaxation time to increase. It w as found that D-Phe enantiomeric impurities could be detected e ven in the presence of enantiomeric excesses of 99.998%.

536

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A recent adv ancement of the MRSW technique involves their use as continuous biosensors for anal ytes214 and solub le biomark ers215 in potentiall y implantab le devices. These de vices generall y consist of SPIO molecular conjugates enclosed within a semi-per meable chamber. The pores allow the analyte or biomarker to enter but retain the SPIO within the chamber. The dispersion and self-assembly of the SPIO in the presence of the tar get molecule can then be monitored continuousl y b y MR. Proof of principle has already been estab lished with biosensors for glucose and hCG-β. Although MRSW has not y et been used for in vi vo applications, one potential application is the detection of myeloperoxidase (MPO) acti vity. My eloperoxidase is an enzyme associated with atherosclerosis and inflammation. In a recent study, it was found that CLIO labeled with serotonin could be used to detect MPO, where serotonin crosslinking resulted in a decrease in the T 2 signal in a concentration-dependent manner.216 It should be noted that although MRSW holds g reat promise for in vi vo applications, a signif icant hindrance to the use of MRSW techniques in living subjects is the possibility of the nonspecific aggregation of SPIO. Any accumulation of SPIO that is not the result of tar get binding ma y relay a f alse decrease in signal. Therefore, precautions must be taken when making conclusions from these types of experiments.

REFERENCES 1. Cullity BD . Introduction to magnetic materials. Reading (MA): Addison-Wesley; 1972. 2. Bean CP, Li vingston JD . Super paramagnetism. J Appl Ph ys Suppl 1959;10:S120–9. 3. Jung CW, Jacobs P. Physical and chemical-proper ties of super paramagnetic iron-oxide Mr contrast agents—fer umoxides, ferumoxtran, ferumoxsil. Magn Reson Imaging 1995;13:661–74. 4. Josephson L, Le wis J, Jacobs P, et al. The effects of iron o xides on proton relaxivity. Magn Reson Imaging 1988;6:647–53. 5. Josephson L, Bigler J , White D . The magnetic proper ties of some materials af fecting MR images. Magn Reson Med 1991; 22:204–8; discussion 213–5. 6. Shen T, Weissleder R, P apisov M, et al. Monocr ystalline iron-oxide nanocompounds (MION)—ph ysicochemical proper ties. Magn Reson Med 1993;29:599–604. 7. Lopez-Perez JA, Lopez-Quintela MA, Mira J, Rivas J. Preparation of magnetic fluids with par ticles obtained in microemulsions. IEEE Trans Magn 1997;33:4359–62. 8. Feltin N, Pileni MP. New technique for synthesizing iron ferrite magnetic nanosized particles. Langmuir 1997;13:3927–33. 9. Suslick KS, Fang MM, Hyeon T. Sonochemical synthesis of iron colloids. J Am Chem Soc 1996;118:11960–1. 10. Prakash A, McCormick, AV, Zachariah MR. Aero-sol-gel synthesis of nanoporous iron-oxide particles: a potential o xidizer for nanoenergetic materials. Chem Mater 2004;16:1466–71. 11. Caruso F, Susha AS, Giersig M, Mohw ald H. Magnetic core-shell particles: preparation of magnetite multilayers on polymer latex microspheres. Adv Mater 1999;11:950–3.

12. Caruso F, Spasova M, Susha A, et al. Magnetic nanocomposite particles and hollow spheres constructed by a sequential layering approach. Chem Mater 2001;13:109–16. 13. Choa YH, Yang JK, Yang WJ, Auh KH. Synthesis and characterization of isolated iron o xide nanopar ticle dispersed in MgO matrix. J Magn Magn Mater 2003;266:20–7. 14. Grimes CA, Qian D, Dickey EC, et al. Laser p yrolysis fabrication of ferromagnetic gamma( ʹ′)-Fe4N and F eC nanopar ticles. J Appl Phys 2000;87:5642–4. 15. Bomati-Miguel, O, Tartaj P, Morales MP, et al. Core-shell iron-iron oxide nanoparticles synthesized by laser-induced pyrolysis. Small 2006;2:1476–83. 16. Veintemillas-Verdaguer S, Morales MP, Serna CJ. Effect of the oxidation conditions on the maghemites produced b y laser p yrolysis. Appl Organomet Chem 2001;15:365–72. 17. Rockenberger J, Scher EC, Alivisatos AP. A new nonhydrolytic singleprecursor approach to surfactant-capped nanocrystals of transition metal oxides. J Am Chem Soc 1999;121:11595–6. 18. Yu WW, Falkner JC, Yavuz CT, Colvin VL. Synthesis of monodisperse iron o xide nanocr ystals b y ther mal decomposition of iron carboxylate salts. Chem Commun (Camb) 2004;2306–7. 19. Park J, An K, Hwang Y, et al. Ultra-large-scale syntheses of monodisperse nanocrystals. Nat Mater 2004;3:891–5. 20. Morales MP, Veintemillas-Verdaguer S, Montero MI, et al. Surf ace and internal spin canting in gamma-Fe2O3 nanoparticles. Chem Mater 1999;11:3058–64. 21. Morales MP, Bomati-Miguel O, de Alejo RP, et al. Contrast agents for MRI based on iron o xide nanopar ticles prepared b y laser pyrolysis. J Magn Magn Mater 2003;266:102–9. 22. Gupta AK, Gupta M. Synthesis and surface engineering of iron oxide nanoparticles for biomedical applications. Biomaterials 2005; 26:3995–4021. 23. Zhou WL, Wang KY, O’Connor CJ, Tang J. Granular growth of Fe3O4 thin f ilms and its antiphase boundaries prepared by pulsed laser deposition. J Appl Phys 2001;89:7398–400. 24. Millan A, Ur tizberea A, Silv a NJO , et al. Surf ace ef fects in maghemite nanopar ticles. J Magn Magn Mater 2007; 312:L5–L9. 25. Iglesias O, Labarta A. Role of surface disorder on the magnetic properties and hysteresis of nanopar ticles. Physica B Condens Matter 2004;343:286–92. 26. Coey JMD. Noncollinear spin arrangement in ultraf ine ferrimagnetic crystallites. Phys Rev Lett 1971;27:1140–2. 27. Sato T, Iijima T, Seki M, Inagaki N . Magnetic-properties of ultrafine ferrite particles. J Magn Magn Mater 1987;65:252–6. 28. Cornell RM, Scher tmann U. The iron o xides: str ucture, proper ties, reactions, occurence and uses. Weinheim, Ger many: VCH Publishers; 1996. 29. Wagner S, Schnorr J, Pilgrimm H, et al. Monomer-coated very small superparamagnetic iron o xide par ticles as contrast medium for magnetic resonance imaging—Preclinical in vi vo characterization. Invest Radiol 2002;37:167–77. 30. Fauconnier N, Bee A, Roger J, Pons JN. Adsorption of gluconic and citric acids on maghemite par ticles in aqueous medium. In: Trends in colloid and interface science X. Berlin/Heidelberg, Germany: Springer; 1996. 31. Fauconnier N, Bee A, Roger J, Pons JN. Synthesis of aqueous magnetic liquids by surface complexation of maghemite nanoparticles. J Mol Liq 1999;83:233–42. 32. Khalafalla SE, Reimers GW. Preparation of dilution-stable aqueous magnetic fluids. IEEE Trans Magn 1980;16:178–83. 33. Tadmor R, Rosens weig RE, F rey J, Klein J. Resolving the puzzle of ferrofluid dispersants. Langmuir 2000;16:9117–20. 34. Shen LF, Stacho wiak A, Fateen SEK, et al. Str ucture of alkanoic acid stabilized magnetic fluids. A small-angle neutron and light scattering analysis. Langmuir 2001;17:288–99.

Magnetic Nanoparticles

35. Shimoiizaka J , Nakatsuka K, Fujita T, K ounosu A. Sink-float separators using per manent-magnets and w ater based magnetic fluid. IEEE Trans Magn 1980;16:368–71. 36. Shen LF, Stacho wiak A, Hatton TA, Laibinis PE. P olymerization of olefin-terminated surf actant bila yers on magnetic fluid nanopar ticles. Langmuir 2000;16:9907–11. 37. Thorek DL, Chen AK, Czupr yna J, Tsourkas A. Super paramagnetic iron oxide nanopar ticle probes for molecular imaging. Ann Biomed Eng 2006;34:23–38. 38. Palmacci S, Josephson L. U.S., 1993. 39. Clement O, Siauve N, Cuenod CA, Frija G. Liver imaging with ferumoxides (feride x): fundamentals, contro versies, and practical aspects. Top Magn Reson Imaging 1998;9:167–82. 40. De Gennes PG. Confor mations of polymers attached to an interf ace. Macromolecules 1980;13:1069–75. 41. Hogemann D, Josephson L, Weissleder R, Basilion JP. Improvement of MRI probes to allo w ef ficient detection of gene e xpression. Bioconjug Chem 2000;11:941–6. 42. Chen M, Yamamuro S, Farrell D, Majetich SA. Gold-coated iron nanoparticles for biomedical applications. J Appl Ph ys 2003; 93:7551–3. 43. Lin J , Zhou WL, K umbhar A, et al. Gold-coated iron (F e@Au) nanoparticles: synthesis, characterization, and magnetic fieldinduced self-assembly. J Solid State Chem 2001;159:26–31. 44. Carpenter EE. Iron nanopar ticles as potential magnetic car riers. J Magn Magn Mater 2001;225:17–20. 45. Zhou WL, Carpenter EE, Lin J , et al. Nanostr uctures of gold coated iron core-shell nanoparticles and the nanobands assemb led under magnetic field. Eur Phys J D 2001;16:289–92. 46. Santra S, Tapec R, Theodoropoulou N, et al. Synthesis and characterization of silica-coated iron o xide nanoparticles in microemulsion: the effect of nonionic surfactants. Langmuir 2001;17:2900–6. 47. Tartaj P, Gonzalez-Carreno T, Serna CJ. Single-step nanoengineering of silica coated maghemite hollow spheres with tunable magnetic properties. Adv Mater 2001;13:1620–4. 48. Tartaj P, Gonzalez-Car reno T, Ser na CJ. Synthesis of nanomagnets dispersed in colloidal silica cages with applications in chemical separation. Langmuir 2002;18:4556–8. 49. Ulman A. Formation and structure of self-assembled monolayers. Chem Rev 1996;96:1533–54. 50. Mulvaney P, Liz-Marzan LM, Giersig M, Ung T. Silica encapsulation of quantum dots and metal clusters. J Mater Chem 2000; 10:1259–70. 51. Morup S. Magnetic h yperfine splitting in mossbauer -spectra of micro-crystals. J Magn Magn Mater 1983;37:39–50. 52. Davies KJ, Wells S, Charles SW. The effect of temperature and oleate adsorption on the g rowth of maghemite par ticles. J Magn Magn Mater 1993;122:24–8. 53. Lee J, Isobe T, Senna M. Preparation of ultraf ine Fe3O4 particles by precipitation in the presence of PV A at high pH. J Colloid Interface Sci 1996;177:490–4. 54. Bjornerud A, Johansson L. The utility of super paramagnetic contrast agents in MRI: theoretical consideration and applications in the cardiovascular system. NMR Biomed 2004;17:465–77. 55. Majumdar S, Zoghbi SS, Gore JC. The influence of pulse sequence on the relaxation ef fects of super paramagnetic iron-o xide contrast agents. Magn Reson Med 1989;10:289–301. 56. Fretz CJ, Elizondo G, Weissleder R, et al. Super paramagnetic iron oxide-enhanced MR imaging: pulse sequence optimization for detection of liver cancer. Radiology 1989;172:393–7. 57. Anzai Y, Prince MR. Iron o xide-enhanced MR l ymphography: the evaluation of cer vical lymph node metastases in head and neck cancer. J Magn Reson Imaging 1997;7:75–81. 58. Perez JM, Josephson L, O’Loughlin T, et al. Magnetic relaxation switches capable of sensing molecular interactions. Nat Biotechnol 2002;20:816–20.

537

59. Perez JM, Josephson L, Weissleder R. Use of magnetic nanopar ticles as nanosensors to probe for molecular interactions. Chembiochem 2004;5:261–4. 60. Small WC, Nelson RC, Ber nardino ME. Dual contrast enhancement of both T 1- and T 2 -weighted sequences using ultrasmall superparamagnetic iron oxide. Magn Reson Imaging 1993;11:645–54. 61. Mayo-Smith WW, Saini S, Slater G, et al. MR contrast material for vascular enhancement: v alue of super paramagnetic iron o xide. AJR Am J Roentgenol 1996;166:73–7. 62. Oswald P, Clement O, Chambon C, et al. Liver positive enhancement after injection of superparamagnetic nanoparticles: respective role of circulating and uptak en particles. Magn Reson Imaging 1997; 15:1025–31. 63. Rogers JM, Le wis J, Josephson L. Visualization of superior mesenteric lymph nodes by the combined oral and intravenous administration of the ultrasmall superparamagnetic iron oxide, AMI-227. Magn Reson Imaging 1994;12:1161–5. 64. Rogers J, Lewis J, Josephson L. Use of AMI-227 as an oral MR contrast agent. Magn Reson Imaging 1994;12:631–9. 65. Edelman RR, Storey P, Dunkle E, et al. Gadolinium-enhanced off-resonance contrast angiography. Magn Reson Med 2007;57:475–84. 66. Stuber M, Gilson WD, Schar M, et al. P ositive contrast visualization of iron o xide-labeled stem cells using in version-recovery with ON-resonant water suppression (IRON). Magn Reson Med 2007; 58:1072–7. 67. Fleming RE, Bacon BR. Orchestration of iron homeostasis. N Engl J Med 2005;352:1741–4. 68. Andrews NC. Disorders of iron metabolism. N Engl J Med 1999; 341:1986–95. 69. Callender ST . Treatment of iron def iciency. Clin Haematol 1982; 11:327–38. 70. Burns DL, Pomposelli JJ. Toxicity of parenteral iron dextran therapy. Kidney Int Suppl 1999;69:S119–24. 71. Bourrinet P, Bengele HH, Bonnemain B, et al. Preclinical safety and pharmacokinetic profile of ferumoxtran-10, an ultrasmall superparamagnetic iron o xide magnetic resonance contrast agent. Invest Radiol 2006;41:313–24. 72. Bonnemain B . Super paramagnetic agents in magnetic resonance imaging: physicochemical characteristics and clinical applications. A review. J Drug Target 1998;6:167–74. 73. Weissleder R, Stark DD, Engelstad BL, et al. Superparamagnetic iron oxide: phar macokinetics and to xicity. AJR Am J Roentgenol 1989;152:167–73. 74. Bacon BR, Stark DD , Park CH, et al. F errite par ticles—a new magnetic-resonance-imaging contrast agent—Lack of acute or chronic hepatotoxicity after intra venous administration. J Lab Clin Med 1987;110:164–71. 75. Shaw S, Westly E, Pittet M, et al. Systemic prof iling of nanomaterial biologic activity. Under review 2007. 76. Lipshutz GS, Brennan TV, Warren RS. Thorotrast-induced liver neoplasia: a collective review. J Am Coll Surg 2002;195:713–8. 77. Khan AA. Thorotrast-associated li ver cancer . Am J Gastroenterol 1985;80:699–703. 78. Stover BJ . Ef fects of Thorotrast in humans. Health Ph ys 1983;44 Suppl 1:253–7. 79. Kuo TT, H u S , H uang C L, e t a l. C utaneous i nvolvement i n polyvinylpyrrolidone storage disease: a clinicopathologic study of five patients, including tw o patients with se vere anemia. Am J Surg Pathol 1997;21:1361–7. 80. Bubis JJ, Cohen S, Dinbar J, et al. Storage of polyvinylpyrrolidone mimicking a congenital mucolipid storage disease in a patient with Munchausen’ s syndrome. Isr J Med Sci 1975; 11:999–1004. 81. Dunn P, Kuo T, Shih LY, et al. Bone mar row failure and myelofibrosis in a case of PVP storage disease. Am J Hematol 1998; 57:68–71.

538

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

82. Grobner T, Prischl FC. Gadolinium and nephro genic systemic fibrosis. Kidney Int 2007;72:260–4. 83. Broome DR, Girguis MS, Baron PW, et al. Gadodiamide-associated nephrogenic systemic f ibrosis: w hy radiolo gists should be concerned. AJR Am J Roentgenol 2007;188:586–92. 84. High WA, Ayers RA, Chandler J , et al. Gadolinium is detectable within the tissue of patients with nephrogenic systemic fibrosis. J Am Acad Dermatol 2007;56:21–6. 85. Choi JH, Nguy en FT , Barone PW , et al. Multimodal biomedical imaging with asymmetric single-w alled carbon nanotube/iron oxide nanoparticle complexes. Nano Lett 2007;7:861–7. 86. Vuu K, Xie J, McDonald MA, et al. Gadolinium-rhodamine nanoparticles for cell labeling and tracking via magnetic resonance and optical imaging. Bioconjug Chem 2005;16:995–9. 87. Lee JH, Jun YW, Yeon SI, et al. Dual-mode nanopar ticle probes for high-performance magnetic resonance and fluorescence imaging of neuroblastoma. Angew Chem Int Ed Engl 2006;45:8160–2. 88. Lu CW , Hung Y, Hsiao JK, et al. Bifunctional magnetic silica nanoparticles for highly efficient human stem cell labeling. Nano Lett 2007;7:149–54. 89. Rieter WJ, Kim JS, Taylor KM, et al. Hybrid silica nanopar ticles for multimodal imaging. Angew Chem Int Ed Engl 2007;46:3680–2. 90. Gao X, Cui Y, Le venson RM, et al. In vi vo cancer tar geting and imaging with semiconductor quantum dots. Nat Biotechnol 2004; 22:969–76. 91. Kim S, Lim YT, Soltesz EG, et al. Near -infrared fluorescent type II quantum dots for sentinel l ymph node mapping. Nat Biotechnol 2004;22:93–7. 92. Klaassen CD. Casarett and Doull’s Toxicology: the basic science of poisons. New York, NY: McGraw-Hill Health Professions Di vision; 1996. 93. Briley-Saebo K, Bjornerud A, Grant D, et al. Hepatic cellular distribution and degradation of iron oxide nanoparticles following single intra venous injection in rats: implications for magnetic resonance imaging. Cell Tissue Res 2004;316:315–23. 94. Schulze E, Ferrucci JT Jr, Poss K, et al. Cellular uptake and trafficking of a prototypical magnetic iron o xide label in vitro. In vest Radiol 1995;30:604–10. 95. Briley-Saebo KC, Johansson LO , Hustvedt SO, et al. Clearance of iron oxide particles in rat liver: effect of hydrated particle size and coating material on li ver metabolism. In vest Radiol 2006; 41:560–71. 96. Chouly C, P ouliquen D, Lucet I, et al. De velopment of super paramagnetic nanopar ticles for MRI: ef fect of par ticle size, char ge and surf ace nature on biodistribution. J Microencapsul 1996; 13:245–55. 97. Hahn PF, Stark DD , Le wis JM, et al. F irst clinical trial of a ne w superparamagnetic iron o xide for use as an oral gastrointestinal contrast agent in MR imaging. Radiology 1990;175:695–700. 98. Bach-Gansmo T. Ferrimagnetic susceptibility contrast agents. Acta Radiol Suppl 1993;387:1–30. 99. Heusler RCH, Wight E, Marincek B . Oral super paramagnetic contrast agent (ferumoxsil): tolerance and efficacy in MR imaging of gynecologic diseases. J Magn Reson Imaging 1995;5:385–91. 100. Johnson WK, Stoupis C, Torres GM, et al. Super paramagnetic iron oxide (SPIO) as an oral contrast agent in gastrointestinal (GI) magnetic resonance imaging (MRI): comparison with state-ofthe-art computed tomo graphy (CT). Magn Reson Imaging 1996; 14:43–9. 101. Ferrucci JT, Stark DD. Iron oxide-enhanced MR imaging of the liver and spleen: re view of the f irst 5 y ears. AJR Am J Roentgenol 1990;155:943–50. 102. Vogl TJ, Hammerstingl R, Schwarz W, et al. Superparamagnetic iron oxide—enhanced v ersus gadolinium-enhanced MR imaging for differential diagnosis of focal li ver lesions. Radiolo gy 1996; 198:881–7.

103. Kawamori Y, Matsui O , Kado ya M, et al. Dif ferentiation of hepatocellular carcinomas from h yperplastic nodules induced in rat li ver with fer rite-enhanced MR imaging. Radiolo gy 1992; 183:65–72. 104. Ros PR, Freeny PC, Harms SE, et al. Hepatic MR imaging with ferumoxides: a multicenter clinical trial of the safety and efficacy in t he d etection o f f ocal h epatic l esions. R adiology 1 995; 196:481–8. 105. Weissleder R, Hahn PF , Stark DD , et al. Super paramagnetic iron oxide: enhanced detection of focal splenic tumors with MR imaging. Radiology 1988;169:399–403. 106. Caramella D, Jin XN, Mascalchi M, et al. Liver and spleen enhancement after intra venous injection of carbo xydextran magnetite: effect of dose, dela y of imaging, and f ield strength in an e x vivo model. MAGMA 1996;4:225–30. 107. Namkung S, Zech CJ , Helmberger T, et al. Super paramagnetic iron oxide (SPIO)-enhanced liver MRI with ferucarbotran: efficacy for characterization of focal li ver lesions. J Magn Reson Imaging 2007;25:755–65. 108. Kawamura Y, Endo K, Watanabe Y, et al. Use of magnetite par ticles as a contrast agent for MR imaging of the li ver. Radiology 1990; 174:357–60. 109. Tsang YM, Stark DD , Chen MC, et al. Hepatic micrometastases in the rat: ferrite-enhanced MR imaging. Radiology 1988;167:21–4. 110. Stark DD, Weissleder R, Elizondo G, et al. Super paramagnetic iron oxide: clinical application as a contrast agent for MR imaging of the liver. Radiology 1988;168:297–301. 111. Majumdar S, Zo ghbi SS, Gore JC. Phar macokinetics of superparamagnetic iron-o xide MR contrast agents in the rat. Invest Radiol 1990;25:771–7. 112. Saini S, Stark DD, Hahn PF, et al. Ferrite particles: a superparamagnetic MR contrast agent for the reticuloendothelial system. Radiology 1987;162:211–6. 113. Wang YX, Hussain SM, Krestin GP . Super paramagnetic iron o xide contrast agents: ph ysicochemical characteristics and applications in MR imaging. Eur Radiol 2001;11:2319–31. 114. Reimer P, Rummen y EJ , Daldr up HE, et al. Clinical results with Resovist: a phase 2 clinical trial. Radiolo gy 1995;195:489–96. 115. Gandon Y, Heautot JF, Brunet F, et al. Superparamagnetic iron oxide: clinical time-response study. Eur J Radiol 1991;12:195–200. 116. Reimer P, Tombach B. Hepatic MRI with SPIO: detection and characterization of focal liver lesions. Eur Radiol 1998;8:1198–204. 117. Reimer P , Balzer T. F erucarbotran (Reso vist): a ne w clinicall y approved RES-specif ic contrast agent for contrast-enhanced MRI of the liver: properties, clinical development, and applications. Eur Radiol 2003;13:1266–76. 118. Hamm B, Staks T, Taupitz M, et al. Contrast-enhanced MR imaging of liver and spleen: f irst experience in humans with a ne w superparamagnetic iron oxide. J Magn Reson Imaging 1994;4:659–68. 119. Reimer P , Marx C, Rummen y EJ , et al. SPIO-enhanced 2D-TOF MR angiography of the por tal venous system: results of an intraindi vidual comparison. J Magn Reson Imaging 1997; 7:945–9. 120. Weissleder R, Elizondo G, Wittenberg J, et al. Ultrasmall superparamagnetic iron o xide: characterization of a ne w class of contrast agents for MR imaging. Radiology 1990;175:489–93. 121. Corot C, Rober t P, Idee JM, P ort M. Recent adv ances in iron o xide nanocrystal technology for medical imaging. Adv Drug Deliv Rev 2006;58:1471–504. 122. Briley-Saebo K. Uppsala University (Uppsala); 2004. 123. Weissleder R, Elizondo G, Wittenberg J, et al. Ultrasmall superparamagnetic iron-oxide—an intravenous contrast agent for assessing lymph-nodes with MR imaging. Radiology 1990;175:494–8. 124. Harisinghani MG, Barentsz J, Hahn PF, et al. Noninvasive detection of clinicall y occult l ymph-node metastases in prostate cancer . N Engl J Med 2003;348:2491–5.

Magnetic Nanoparticles

125. Seneterre E, Weissleder R, Jaramillo D , et al. Bone mar row: ultrasmall superparamagnetic iron oxide for MR imaging. Radiology 1991;179:529–33. 126. Bush CH, Mladinich CR, Montgomer y WJ. Evaluation of an ultrasmall superparamagnetic iron oxide in MRI in a bone tumor model in rabbits. J Magn Reson Imaging 1997;7:579–84. 127. Daldrup-Link HE, Rummen y EJ, Ihssen B , et al. Iron-o xideenhanced MR imaging of bone mar row in patients with nonHodgkin’s lymphoma: differentiation between tumor infiltration and hypercellular bone marrow. Eur Radiol 2002;12:1557–66. 128. Mergo PJ, Engelken JD, Helmberger T, Ros PR. MRI in focal liver disease: a comparison of small and ultra-small superparamagnetic iron o xide as hepatic contrast agents. J Magn Reson Imaging 1998;8:1073–8. 129. Bremer C, Allkemper T, Baermig J, Reimer P. RES-specific imaging of the liver and spleen with iron oxide particles designed for blood pool MR-angiography. J Magn Reson Imaging 1999;10:461–7. 130. Anzai Y, Prince MR, Chene vert TL, et al. MR angio graphy with an ultrasmall super paramagnetic iron o xide b lood pool agent. J Magn Reson Imaging 1997;7:209–14. 131. Knollmann FD, Bock JC, Teltenkotter S, et al. Ev aluation of por tal MR angio graphy using super paramagnetic iron o xide. J Magn Reson Imaging 1997;7:191–6. 132. Allkemper T, Bremer C, Matuszewski L, et al. Contrast-enhanced blood-pool MR angiography with optimized iron oxides: effect of size and dose on v ascular contrast enhancement in rabbits. Radiology 2002;223:432–8. 133. Schmitz SA, Taupitz M, Wagner S, et al. Magnetic resonance imaging of atherosclerotic plaques using superparamagnetic iron oxide particles. J Magn Reson Imaging 2001;14:355–61. 134. Wacker FK, Wendt M, Eber t W, et al. Use of a b lood-pool contrast agent for MR-guided vascular procedures: feasibility of ultrasmall superparamagnetic iron o xide par ticles. Acad Radiol 2002; 9:1251–4. 135. Zimmer C, Weissleder R, O’Connor D, et al. Cerebral iron oxide distribution: in vi vo mapping with MR imaging. Radiolo gy 1995; 196:521–7. 136. Simonsen CZ, Oster gaard L, Vestergaard-Poulsen P, et al. CBF and CBV measurements by USPIO bolus tracking: reproducibility and comparison with Gd-based values. J Magn Reson Imaging 1999; 9:342–7. 137. Rydland J, BjOrnerud A, Haugen O, et al. New intravascular contrast agent applied to dynamic contrast enhanced MR imaging of human breast cancer. Acta Radiol 2003;44:275–83. 138. Daldrup-Link HE, Rydland J , Helbich TH, et al. Quantif ication of breast tumor microvascular permeability with feruglose-enhanced MR imaging: initial phase II multicenter trial. Radiolo gy 2003; 229:885–92. 139. Daldrup-Link HE, Kaiser A, Helbich T, et al. Macromolecular contrast medium (feruglose) versus small molecular contrast medium (gadopentetate) enhanced magnetic resonance imaging: dif ferentiation of benign and malignant breast lesions. Acad Radiol 2003; 10:1237–46. 140. Turetschek K, Rober ts TP, Flo yd E, et al. Tumor micro vascular characterization using ultrasmall superparamagnetic iron oxide particles (USPIO) in an e xperimental breast cancer model. J Magn Reson Imaging 2001;13:882–8. 141. Hogemann-Savellano D , Bos E, Blondet C, et al. The transfer rin receptor: a potential molecular imaging marker for human cancer. Neoplasia 2003;5:495–506. 142. Tonik SE, Shindelman JE, Sussman HH. Transferrin receptor is inversely correlated with estrogen receptor in breast cancer. Breast Cancer Res Treat 1986;7:71–6. 143. Walker RA, Da y SJ. Transferrin receptor e xpression in non-malignant and malignant human breast tissue. J P athol 1986; 148:217–24.

539

144. Daniels TR, Delgado T, Rodriguez JA, et al. The transferrin receptor part I: Biolo gy and tar geting with c ytotoxic antibodies for the treatment of cancer. Clin Immunol 2006;121:144–58. 145. Keer HN, Kozlowski JM, Tsai YC, et al. Ele vated transferrin receptor content in human prostate cancer cell lines assessed in vitro and in vivo. J Urol 1990;143:381–5. 146. Kresse M, Wagner S, Pfefferer D, et al. Targeting of ultrasmall superparamagnetic iron oxide (USPIO) particles to tumor cells in vi vo by using transfer rin receptor pathw ays. Magn Reson Med 1998; 40:236–42. 147. Gatter KC, Bro wn G, Trowbridge IS, et al. Transferrin receptors in human tissues: their distribution and possib le clinical rele vance. J Clin Pathol 1983;36:539–45. 148. Sudimack J, Lee RJ . Targeted dr ug delivery via the folate receptor . Adv Drug Deliv Rev 2000;41:147–62. 149. Choi H, Choi SR, Zhou R, et al. Iron oxide nanoparticles as magnetic resonance contrast agent for tumor imaging via folate receptor targeted delivery. Acad Radiol 2004;11:996–1004. 150. Medarova Z, Pham W, Kim Y, et al. In vi vo imaging of tumor response to therapy using a dual-modality imaging strate gy. Int J Cancer 2006;118:2796–802. 151. Towner RA, Smith N, Tesiram YA, et al. In vivo detection of c-MET expression in a rat hepatocarcino genesis model using molecularl y targeted magnetic resonance imaging. Mol Imaging 2007;6:18–29. 152. Montet X, Montet-Abou K, Reynolds F, et al. Nanopar ticle imaging of integrins on tumor cells. Neoplasia 2006;8:214–22. 153. Zhang C, Jugold M, Woenne EC, et al. Specif ic tar geting of tumor angiogenesis b y RGD-conjugated ultrasmall super paramagnetic iron oxide particles using a clinical 1.5-T magnetic resonance scanner. Cancer Res 2007;67:1555–62. 154. Suwa T, Ozawa S, Ueda M, et al. Magnetic resonance imaging of esophageal squamous cell carcinoma using magnetite par ticles coated with anti-epidermal growth factor receptor antibody. Int J Cancer 1998;75:626–34. 155. Huh YM, Jun YW, Song HT, et al. In vivo magnetic resonance detection of cancer b y using multifunctional magnetic nanocr ystals. J Am Chem Soc 2005;127:12387–91. 156. Baio G, Fabbi M, de Totero D, et al. Magnetic resonance imaging at 1.5 T with immunospecific contrast agent in vitro and in vivo in a xenotransplant model. MAGMA 2006;19:313–20. 157. Remsen LG, McCor mick CI, Roman-Goldstein S, et al. MR of carcinoma-specific monoclonal antibody conjugated to monocr ystalline iron o xide nanopar ticles: the potential for nonin vasive diagnosis. AJNR Am J Neuroradiol 1996;17:411–8. 158. Hellstrom I, Hor n D, Linsley P, et al. Monoclonal mouse antibodies raised against human lung carcinoma. Cancer Res 1986; 46:3917–23. 159. Kitamura K, Takahashi T, Yamaguchi T, et al. Immunochemical characterization of the antigen reco gnized by the murine monoclonal antibody A7 against human colorectal cancer. Tohoku J Exp Med 1989;157:83–93. 160. Otsuji E, K uriu Y, Okamoto K, et al. Monoclonal antibody A7 coupled to magnetic par ticles as a contrast enhancing agent for magnetic resonance imaging of human colorectal carcinoma. Cancer Immunol Immunother 2006;55:728–33. 161. Reimer P, Weissleder R, Shen T, et al. P ancreatic receptors: initial feasibility studies with a targeted contrast agent for MR imaging. Radiology 1994;193:527–31. 162. Montet X, Weissleder R, Josephson L. Imaging pancreatic cancer with a peptide-nanopar ticle conjugate tar geted to nor mal pancreas. Bioconjug Chem 2006;17:905–11. 163. Reimer P, Weissleder R, Lee AS, et al. Receptor imaging: application to MR imaging of liver cancer. Radiology 1990;177:729–34. 164. Tanimoto A, Kuribayashi S. Hepatocyte-targeted MR contrast agents: contrast enhanced detection of li ver cancer in dif fusely damaged liver. Magn Reson Med Sci 2005;4:53–60.

540

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

165. Artemov D , Mori N , Ok ollie B , Bhujw alla ZM. MR molecular imaging of the Her -2/neu receptor in breast cancer cells using targeted iron o xide nanopar ticles. Magn Reson Med 2003; 49:403–8. 166. Jaffer FA, Libby P, Weissleder R. Molecular imaging of cardiovascular disease. Circulation 2007;116:1052–61. 167. Kelly KA, Allport JR, Tsourkas A, et al. Detection of vascular adhesion molecule-1 e xpression using a no vel multimodal nanopar ticle. Circ Res 2005;96:327–36. 168. Nahrendorf M, Jaffer FA, Kelly KA, et al. Noninvasive vascular cell adhesion molecule-1 imaging identif ies inflammator y activation of cells in atherosclerosis. Circulation 2006;114:1504–11. 169. Tsourkas A, Shinde-P atil VR, Kelly KA, et al. In vi vo imaging of activated endothelium using an anti-VCAM-1 magnetooptical probe. Bioconjug Chem 2005;16:576–81. 170. Johansson LO, Bjornerud A, Ahlstrom HK, et al. A targeted contrast agent for magnetic resonance imaging of thrombus: implications of spatial resolution. J Magn Reson Imaging 2001;13:615–8. 171. Weissleder R, Lee AS, Kha w B A, et al. Antimyosin-labeled monocrystalline iron oxide allows detection of myocardial infarct: MR antibody imaging. Radiology 1992;182:381–5. 172. Zhao M, Beauregard DA, Loizou L, et al. Non-invasive detection of apoptosis using magnetic resonance imaging and a tar geted contrast agent. Nat Med 2001;7:1241–4. 173. Schellenberger EA, Bo gdanov A Jr , Ho gemann D , et al. Annexin V-CLIO: a nanopar ticle for detecting apoptosis b y MRI. Mol Imaging 2002;1:102–7. 174. Weissleder R, Lee AS, F ischman AJ, et al. P olyclonal human immunoglobulin G labeled with pol ymeric iron o xide: antibody MR imaging. Radiology 1991;181:245–9. 175. Wadghiri YZ, Sigurdsson EM, Sado wski M, et al. Detection of Alzheimer’s amyloid in transgenic mice using magnetic resonance microimaging. Magn Reson Med 2003;50:293–302. 176. Matuszewski L, P ersigehl T, Wall A, et al. Cell tagging with clinically approved iron oxides: feasibility and effect of lipofection, particle size, and surf ace coating on labeling ef ficiency. Radiology 2005;235:155–61. 177. Dunning MD, Lakatos A, Loizou L, et al. Super paramagnetic iron oxide-labeled Schw ann cells and olf actory ensheathing cells can be traced in vivo by magnetic resonance imaging and retain functional properties after transplantation into the CNS. J Neurosci 2004;24:9799–810. 178. Metz S, Bonater ra G, Rudelius M, et al. Capacity of human monocytes to phagoc ytose approved iron o xide MR contrast agents in vitro. Eur Radiol 2004;14:1851–8. 179. Wilhelm C, Billotey C, Roger J, et al. Intracellular uptake of anionic superparamagnetic nanopar ticles as a function of their surf ace coating. Biomaterials 2003;24:1001–11. 180. Weissleder R, Cheng HC, Bogdanova A, Bogdanov A Jr. Magnetically labeled cells can be detected by MR imaging. J Magn Reson Imaging 1997;7:258–63. 181. Song M, Moon WK, Kim Y, et al. Labeling ef ficacy of super paramagnetic iron o xide nanopar ticles to human neural stem cells: comparison of fer umoxides, monocr ystalline iron o xide, crosslinked iron o xide (CLIO)-NH2 and tat-CLIO . K orean J Radiol 2007;8:365–71. 182. Engberink RDO, van der P ol SM, Dopp EA, et al. Comparison of SPIO and USPIO for in vitro labeling of human monoc ytes: MR detection and cell function. Radiology 2007;243:467–74. 183. Bulte JW, Douglas T, Witwer B , et al. Magnetodendrimers allo w endosomal magnetic labeling and in vi vo tracking of stem cells. Nat Biotechnol 2001;19:1141–7. 184. Hinds KA, Hill JM, Shapiro EM, et al. Highl y efficient endosomal labeling of progenitor and stem cells with large magnetic particles allows magnetic resonance imaging of single cells. Blood 2003;102:867–72.

185. Shapiro EM, Skrtic S, Koretsky AP. Sizing it up: cellular MRI using micron-sized iron o xide par ticles. Magn Reson Med 2005; 53:329–38. 186. Shapiro EM, Skrtic S, Sharer K, et al. MRI detection of sin gle particles for cellular imaging. Proc Natl Acad Sci U S A 2004; 101:10901–6. 187. Hill JM, Dick AJ, Raman VK, et al. Serial cardiac magnetic resonance imaging of injected mesench ymal stem cells. Circulation 2003;108:1009–14. 188. Williams JB , Ye Q, Hitchens TK, et al. MRI detection of macrophages labeled using micrometer-sized iron oxide particles. J Magn Reson Imaging 2007;25:1210–8. 189. Frank J A, Anderson SA, Kalsih H, et al. Methods for magnetically labeling stem and other cells for detection b y in vivo magnetic resonance imaging. Cytotherapy 2004;6:621–5. 190. Frank JA, Miller BR, Arbab AS, et al. Clinically applicable labeling of mammalian and stem cells b y combining super paramagnetic iron oxides and transfection agents. Radiology 2003;228:480–7. 191. Daldrup-Link HE, Rudelius M, Oostendor p RA, et al. Comparison of iron oxide labeling properties of hematopoietic progenitor cells from umbilical cord b lood and from peripheral b lood for subsequent in vi vo tracking in a x enotransplant mouse model XXX. Acad Radiol 2005;12:502–10. 192. Bos C, Delmas Y, Desmouliere A, et al. In vivo MR imaging of intravascularly injected magneticall y labeled mesench ymal stem cells in rat kidney and liver. Radiology 2004;233:781–9. 193. Montet-Abou K, Montet X, Weissleder R, Josephson L. Cell internalization of magnetic nanopar ticles using transfection agents. Mol Imaging 2007;6:1–9. 194. Josephson L, Tung CH, Moore A, Weissleder R. High-ef ficiency intracellular magnetic labeling with novel superparamagnetic-Tat peptide conjugates. Bioconjug Chem 1999;10:186–91. 195. Lewin M, Carlesso N, Tung CH, et al. Tat peptide-derivatized magnetic nanoparticles allow in vivo tracking and recovery of progenitor cells. Nat Biotechnol 2000;18:410–14. 196. Zhao M, Kircher MF, Josephson L, Weissleder R. Differential conjugation of tat peptide to super paramagnetic nanopar ticles and its effect on cellular uptake. Bioconjug Chem 2002;13:840–4. 197. Nitin N, LaConte LE, Zurkiya O , et al. Functionalization and peptide-based delivery of magnetic nanopar ticles as an intracellular MRI contrast agent. J Biol Inor g Chem 2004;9:706–12. 198. Kaufman CL, Williams M, Ryle LM, et al. Super paramagnetic iron oxide par ticles transacti vator protein-fluorescein isothioc yanate particle labeling for in vivo magnetic resonance imaging detection of cell mig ration: uptak e and durability . Transplantation 2003; 76:1043–6. 199. Dodd CH, Hsu HC, Chu WJ, et al. Nor mal T-cell response and in vivo magnetic resonance imaging of T cells loaded with HIV transactivator-peptide-derived superparamagnetic nanoparticles. J Immunol Methods 2001;256:89–105. 200. Kircher MF, Allport JR, Gra ves EE, et al. In vi vo high resolution three-dimensional imaging of antigen-specif ic c ytotoxic T-lymphocyte trafficking to tumors. Cancer Res 2003;63:6838–46. 201. Arbab AS, Yocum GT, Kalish H, et al. Ef ficient magnetic cell labeling with protamine sulfate complexed to ferumoxides for cellular MRI. Blood 2004;104:1217–23. 202. Reynolds F, Weissleder R, Josephson L. Protamine as an ef ficient membrane-translocating peptide. Bioconjug Chem 2005; 16:1240–5. 203. Bulte JW, Zhang S, van Gelderen P, et al. Neurotransplantation of magnetically labeled oligodendroc yte pro genitors: magnetic resonance tracking of cell migration and myelination. Proc Natl Acad Sci U S A 1999;96:15256–61. 204. Moore A, Basilion JP, Chiocca EA, Weissleder R. Measuring transferrin receptor gene e xpression by NMR imaging. Biochim Biophys Acta 1998;1402:239–49.

Magnetic Nanoparticles

205. Weissleder R, Moore A, Mahmood U, et al. In vi vo magnetic resonance imaging of transgene expression. Nat Med 2000;6:351–5. 206. Josephson L, Tung CH, Moore A, Weissleder R. Magnetic nanosensors for the detection of oligonucleotide sequences. Angew Chem Int Ed Engl 2001;40:3204–6. 207. Grimm J, Perez JM, Josephson L, Weissleder R. Novel nanosensors for rapid analysis of telomerase activity. Cancer Res 2004;64:639–43. 208. Perez JM, O’Loughin T, Simeone FJ , et al. DN A-based magnetic nanoparticle assembly acts as a magnetic relaxation nanos witch allowing screening of DN A-cleaving agents. J Am Chem Soc 2002;124:2856–7. 209. Sun EY, Josephson L, K elly KA, Weissleder R. De velopment of nanoparticle libraries for biosensing. Bioconjug Chem 2006; 17:109–13. 210. Kim GY, Josephson L, Langer R, Cima MJ . Magnetic relaxation switch detection of human chorionic gonadotrophin. Bioconjug Chem 2007;18:2024–8.

541

211. Kaittanis C, Naser SA, P erez JM. One-step, nanopar ticle-mediated bacterial detection with magnetic relaxation. Nano Lett 2007; 7:380–3. 212. Perez JM, Simeone FJ, Saeki Y, et al. Viral-induced self-assembly of magnetic nanopar ticles allows the detection of viral par ticles in biological media. J Am Chem Soc 2003;125:10192–3. 213. Tsourkas A, Hofstetter O , Hofstetter H, et al. Magnetic relaxation switch immunosensors detect enantiomeric impurities. Angew Chem Int Ed Engl 2004;43:2395–9. 214. Sun EY, Weissleder R, Josephson L. Continuous analyte sensing with magnetic nanoswitches. Small 2006;2:1144–7. 215. Daniel KD, Kim GY, Vassiliou CC, et al. Multi-reser voir device for detecting a soluble cancer biomarker. Lab Chip 2007;7:1288–93. 216. Perez J, Simeone F, Tsourkas A, et al. Peroxidase substrate nanosensors for MR imaging. Nano Lett 2004;4:119–22.

35 FLUOROCARBON AGENTS FOR QUANTITATIVE MULTIMODAL MOLECULAR IMAGING AND TARGETED THERAPEUTICS SAMUEL A. WICKLINE, MD, RALPH P. MASON, PHD, SHELTON D. CARUTHERS, PHD, JUNJIE CHEN, PHD, PATRICK M. WINTER, PHD, MICHAEL S. HUGHES, PHD, AND GREGORY M. LANZA, MD, PHD

19

F-Nuclear magnetic resonance (NMR) has been widel y exploited for both spectroscopic studies and increasingl y for magnetic resonance imaging (MRI). The 19F-atom exhibits high NMR sensiti vity, while there is essentiall y no backg round signal in the body . There are numerous drugs in clinical use that include a fluorine atom allowing direct investigation of pharmacokinetics, such as the cancer chemotherapeutics, 5-fluorouracil (5-FU) and gemcitabine, anesthetics and psychoacti ve dr ugs, such as fluoxetine, and the cholesterol-lo wering dr ug, ator vastatin. Furthermore, the unique sensiti vity of the fluorine atom to its microen vironment has prompted the design and the development of many reporter molecules to probe such diverse aspects of physiology as pO2, pH, metal-ion concentrations, v ascular flo w, and v olume. Recent reviews of fluorine NMR and its applications to probing physiology and phar macology in vi vo of fer perspecti ve on its use in medicine. 1–5 The use of fluorinated phar maceuticals has increased substantially since the late 1950s due to favorable effects on absorption, distribution, metabolism, and e xcretion (ADME) and the introduction of safe laboratory methods for generating fluorinated compounds from highl y reacti ve elemental F2.6 Fluorine is the most electrone gative element in nature with high ionization potential and lo w polarizability.7,8 It forms a C-F bond of g reat stability due to the o verlapping of orbitals and as such is ther mally and chemically very stable and biologically inert. The dense electron cloud

surrounding such compounds protects them from interaction with other materials, which is a feature that has been used to increase stability of F-substituted compounds b y b locking reactive sites. Compounds in this class are both lipophobic and h ydrophobic simultaneousl y, y et 19F-substituents also can increase binding affinity in certain cases.6 Perfluorocarbons (PFCs, perfluorochemicals, or fluorocarbons) are synthetic or ganic compounds in which all or most of the h ydrogen atoms ha ve been replaced with fluorine atoms. This substitution yields compounds that of fer unique perfor mance capabilities and excellent safety prof iles. The use of perfluorocarbons has been e xplored for other medical applications, including percutaneous transluminal cardiac angioplasty, par tial liquid lung v entilation,9–11 gastrointestinal X-ray contrast agent,12,13 and blood substitutes.7,14–18 Fluosol® from Green Cross w as approved for intravascular use in the United States, for e xample, and contains perfluorodecalin, one of the prefer red PFCs. Nanoscale micelles and surf actant-stabilized nanodroplets of PFCs can be created b y “self-assemb ly” of molecules to produce agents that can be used for clinical cellular and molecular imaging. 14,19,20 The design of targeted nanoscale molecular imaging agents must accomplish a long circulating half-life, sensiti ve and selecti ve binding to the epitope of interest, prominent contrastto-noise enhancement, acceptab le toxicity, ease of clinical use, and applicability with standard commercially available

Grant Support: CA119342, CA126608A–SAIRP (UT Southwestern Small Animal Imaging Resource), HL073646, Philips Medical Systems. 542

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

imaging systems. The use of nanopar ticles or tar geted nanoemulsions as car riers has the adv antage of selectively delivering large payloads of imaging or therapeutic agents to the tissue site. In this chapter , we review the use of selfassembling liquid PFC nanoparticles and other small molecule 19F-reporters for multimodal molecular imaging and drug delivery in cancer and cardiovascular disease.

DESIGN AND FORMULATION A broad array of nanomaterials is available for use as contrast agents for molecular imaging and drug delivery. Size considerations dictate both mechanisms and rates of clearance, as w ell as the access to m olecular tar gets. F or example, for agents that are > 5 0 nm, intravascular targets appear most appropriate, w hereas smaller par ticles may penetrate the intact or the leaky vascular endothelium by endothelial per meability and retention (EPR) mechanisms and thus directly target tissues. “Functionalization” (or preparation of the par ticle surface for binding tar geting ligands or dr ug-delivery ligands) can be achie ved by various chemical means, such as pro vision of reacti ve moieties on the surface, through avidin-biotin interactions or electrostatic interactions (e g, deo xyribonucleic acid binding to cationic lipids in par ticle membranes). To improve the par ticle stability and to per mit adequate circulation times, some types of particles (such as liposomes and/or pol ymers) may require surf ace component crosslinking to enhance str uctural integrity or incorporation of polyethylene glycol to a void immediate sequestration b y the reticuloendothelial system (RES). Others, such as the PFC nanoparticles, are highly stable by virtue of the inert chemistry of their components, as described abo ve.

543

PFC material and natural or synthetic lipid surf actants is driven through a ceramic frit at high pressure (~10,000 psi) to yield a stable oil-in-water-type emulsion. Different perfluorocarbons can be used for the core, including perfluorodichlorooctane, perfluorodecalin, perfluoro 15crown-5 ether (CE), and most commonl y, perfluorooctyl bromide (PFOB or perflubron). PFC nanopar ticles are 98% perfluorocarbon b y v olume, w hich equates for PFOB (1.98 g/mL, 498 daltons) to appro ximately 100 M concentration of fluorine within a nanoparticle. A typical formulation might comprise 20% (v/v) PFOB , 2% (w/v) of a surf actant comixture (~70 mol% lecithin), 1.7% (w/v) gl ycerin, and w ater for the balance. The nominal size of the resulting par ticles ranges from 200 to 400 nm depending on the specif ic materials used. The PFC core materials are sur rounded b y a lipid monolayer that can be used to contain v arious agents for imaging or therapeutic action. Multifunctional acti vity can be realized b y incor porating combinations of one or more targeting ligands, imaging agents, and/or dr ugs into the for mulation simultaneousl y. Materials can be co valently or noncovalently linked to the par ticle surface, dissolved in the coating (e g, lipophilic dr ugs deposited in lipid membrane layers) or car ried in the par ticle interiors for cellular deposition and acti vation. For production of a paramagnetic particle, for example, 30 mol% gadolinium (Gd) dieth ylene-triamine-pentaacetic acid-bis-oleate might be added to car ry a high pa yload of the lanthanide (~100,000 to 200,000 Gd atoms) for MRI. Other additives that can be formulated directly for targeting might be, for example, a 0.1 mol% PEG-phosphatidylethanolamine coupled covalently to a binding ligand. 22–24

Targeting Composition The production of PFC nanopar ticles (F igure 1) uses a microfluidization process.14,21 Basically, a mixture of neat

A variety of different types of tar geting ligands can be used, including antibodies or antibody fragments, small peptides, peptidomimetics, pol ysaccharides, and

Figure 1. Scanning electron microscope of nanoparticle-targeted thrombus (ligand: antifibrin antibody). Left: Untreated. Middle: Treated (lower power). Right: Treated (higher power). (Reprinted with permission from Lanza GM and Wickline SA28).

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

aptamers. All these elements can be comple xed with particles b y functionalizing the par ticle surf aces b y including standard linking g roups. Common methods for associating tar geting ligands with par ticles include (1) noncovalent avidin-biotin interactions, (2) co valent complexation via reacti ve g roups, or (3) nonspecif ic surface adsor ption. Avidin-biotin interactions are extremely useful, high-af finity, nonco valent tar geting systems that have been incorporated into many biologic and analytic systems and selected in vi vo applications. Additionally, avidin, with four independent biotin-binding sites, pro vides signal amplif ication and impro ves detection sensitivity. Avidin-biotin interactions may be used to create a “one-step” system b y perfor ming the avidin-biotin conjugations in vitro prior to injection. Investigators ha ve used this approach to successfull y target vascular epitopes in vivo.25,26 For in vi vo use, tar geting ligands are preferab ly attached chemically to the contrast agent b y a v ariety of methods depending upon the nature of the particle surface. Conjugations may be performed before or after the particle is created depending upon the ligand used and its tolerance to the chemical processing conditions required. Direct chemical conjugation of ligands to proteinaceous agents often tak es adv antage of numerous amino g roups (e g, lysine) inherently present within the surface. Alternatively, functionally active chemical groups, such as pyridyldithiopropionate, maleimide, and amino or aldeh yde ma y be incorporated into the surf ace as chemical “hooks” for ligand conjugation after the par ticles are formed. Multiple copies of ligands can be incor porated depending on the size of the par ticle, w hich ser ves to enhance avidity and tar get detectability b y reducing the particle dissociation rate, thus better securing the agent at the intended site. Typically, 20 to 40 monoclonal antibodies or 200 to 400 small-molecule ligands can be attached to the surf ace of the par ticle. These ligands would be anchored b y phospholipid (e g, phosphatidylethanolamine) anchors that can be comple xed to the ligand via inter mediate bifunctional reacti ve g roups as mentioned abo ve. Specif icity is confer red by the targeting ligand itself, and generall y should be in the nanomolar range, although high-a vidity agents ma y in part overcome this limitation by multivalent interactions. To ensure high ligand-binding inte grity and maximize targeted-particle a vidity, fle xible pol ymer spacer ar ms, eg, polyethylene glycol or simple caproate bridges, can be inserted betw een an acti vated surf ace functional g roup and the targeting ligand. These extensions can be 10 nm or longer and minimize interference of ligand binding by particle surface interactions. Regardless of targeting, it is

clear that most such agents will e xhibit a modicum of nonspecific targeting related either to nonspecif ic ligand attachment, the EPR effect, or sequestration in immature vasculature (angiogenesis).

Absorption, Distribution, Metabolism, Excretion (ADME) Once the par ticles bind to the tar get, the y stick and sta y with high a vidity because of multi valent interactions, due to the incorporation of many ligands (~200 small-molecule ligands per par ticle) on their surf ace.27,28 There is no detectable change in tissue morphology since only a single layer of particles can bind to the target (ie, they do not continue to accumulate and for m a “clot” since the y do not bind to each other). Ultimatel y, the lipids are rec ycled through plasma car riers, the perfluorocarbon is e xhaled through the lung (due to its high v apor pressure relative to mass), and other components are de graded at the site (proteins). Particles that do not bind are remo ved conveniently b y the li ver and the spleen. We ha ve detected persistent binding of paramagnetic v ersions of these particles b y MRI after 24 hours, but the signal f ades subsequently, after peaking locally by 4 to 8 hours. The particle size of the emulsion predicts the v ascular half-life and biodistribution. Phar macokinetic analysis (F igure 2) indicates that the clearance half-life is around 3 to 6 hours depending on species. 29 The tissue distribution (F igure 3) reflects the e xpected acti vity of the macrophage phagoc ytic system, w here the spleen and the liver remove the majority of the par ticles after

5

10

3 Model C1 A0 e t B0 e T1/2 309 136 min

4.5 4

%ID/mL

544

VD 380

66 mL

CI 0.68

0.12 mL/min

80

100

t

3.5 3 2.5 2 1.5 1 0

20

40

60

120

Time Post Injection (min) Figure 2. Distribution and clearance from circulation of αvβ3-integrin targeted 111In nanoparticles (NP) with ~10 111In/NP. Percent injected dose (ID) in blood versus time is presented for one animal over the first 2 hours. (Reprinted with permission from Hu G et al.29)

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

Figure 3. The biodistribution of perfluorocarbon nanoparticles in rabbits injected with nanoparticle emulsion at dosages of 0.25 mL/kg, 0.5 mL/kg, and 1.0 mL/kg (n = 3/dose). Tissue perfluorocarbon content was measured directly by gas chromatography. (Reprinted with permission from Hu G et al.29)

intravenous injection. 29 Ultimately, the optimal par ticle size of the emulsion should result in a v ascular persistence that is clinically efficacious while limiting deep-tissue retention, which for most intra venous applications is approximately 200 to 400 nm.

Safety Natural phospholipids are the most commonl y used surfactants due to their biocompatibility , w hich has been established o ver the past 40 y ears as surf actants of injectable f at emulsions for parenteral nutrition. The biocompatibility of liquid fluorocarbons is w ell documented.16 Even at lar ge doses, most fluorocarbons are innocuous and ph ysiologically inacti ve. No to xicity, carcinogenicity, mutagenicity or teratogenic effects have been reported for pure fluorocarbons within the 460 to 520 MW range. PFCs have tissue half-life residencies ranging from 4 da ys for PFOB up to 65 da ys for perfluorotriprop ylamine, and are not metabolized, but rather slowly reintroduced to the circulation in dissolved form by lipid carriers and e xpelled through the lungs. Increased pulmonar y residual volumes with b lood transfusion le vel dosages of PFC emulsions ha ve been repor ted in rabbits, s wine, and macaque but not in mouse, dog, and human. 7 The acute toxicity of properly prepared PFC emulsions is very low. Typically, the LD50 of PFCs is in the range of 30 to 41 g PFC/kg body weight. This represents an approximate 10-fold safety margin over the doses typically used in blood substitute applications and a g reater than 200-fold safety margin at the projected doses for use as a tar geted contrast agent. Clinical side ef fects obser ved thus f ar have been characterized by a “flu-like” reaction (headache,

545

nausea, fever, etc). These reactions are generall y mild and transient, with complete resolution within 12 hours. The mechanism of this “flu-lik e” reaction appears to be related to the natural clearance (phagocytosis) of the emulsion particles from the b loodstream. The phagocytic macrophages are acti vated in this process and release products of the arachidonic acid cascade, such as prostaglandins and cytokines. It has also been shown that plasma cytokine levels are directly related to the PFC dose. Ho wever, the projected clinical dosing for this agent used for dr ug delivery or imaging is anticipated to be approximately 200-fold less than for those emulsions in which these clinical side effects were observed.

ULTRASOUND MOLECULAR IMAGING Ultrasound has man y adv antages re garding the benign imaging energies used (compressional waves) and its flexibility, high throughput, low cost, and excellent patient tolerance. However, it is more operator dependent than other tomographic methods and cannot image all areas of the body. Clinically available ultrasound contrast agents comprise shell-stabilized gas-f illed microbubb les and man y have been developed that can be tar geted to vascular epitopes.30 In the re gimen of smaller par ticles, acoustically active emulsion nanopar ticles for both imaging and therapy and reflecti ve liposomes for imaging ha ve recei ved the most attention. 21,27,28,31–33 Although ultrasound is exquisitely sensitive for detecting microbubbles, it is less so for liquid nanoparticles because of the size dependency (r6) and relati ve incompressibility of liquid par ticles, which eliminates the use of available harmonic resonancebased imaging techniques typicall y applied to microbubble detection. A balancing consideration is the unique potential for precisely depositing large amounts of highly focused energy in a con venient manner that could f acilitate nanopar ticle-based imaging and therapeutics with exogenous ultrasound acti vation as has been described recently by several groups.34–36 The advent of mathematical models to characterize the fundamental scattering behavior from ne wer classes of nanopar ticles (eg, emulsions) raises the potential for extracting more quantitative information from the reflected signals. 37–41

Mechanisms of Ultrasonic Contrast Enhancement Ultrasound contrast agents are designed to alter the absorption, reflection, or refraction of sound to enhance differentiation of the signal from that of the sur rounding sample. For this pur pose, gas-f illed microbubbles were developed

546

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

initially as nontar geted agents to transientl y enhance the blood-pool signal, w hich is otherwise onl y w eakly echogenic. These agents operate b y se veral mechanisms that include a high intrinsic reflecti vity due to a lar ge acoustic impedance mismatch betw een the contained gas and the sur rounding fluid medium, augmentation of scattered signal due to acoustic resonance phenomena, and nonlinear elastic beha vior as compared with sur rounding tissue elements. The scattering cross section of nanoparticles is lower than that for microbubb les (nontargeted or tar geted) per individual scatterer because the nanopar ticle scatterers are smaller b y 10 × in diameter (100 × in surf ace area) than are bubbles and do not e xhibit traditional nonlinear bubble-like beha viors (resonance and higher/lo wer harmonics). Ne vertheless, earl y w ork b y g roups, such as Robert Mattre y’s, actually paved the w ay for perfluorocarbon nanopar ticles as tar geted ultrasound contrast agents by showing in earl y g roundbreaking experiments that if signif icant concentrations of nanopar ticles w ere injected into the b lood stream, substantial and sustained blood contrast could be generated. 42 Unfortunately, the nanoparticle concentrations required to produce b loodpool contrast necessitated too lar ge a total injection volume to be clinically practical for ultrasound imaging. Our g roup introduced the f irst repor ted tar geted ultrasound contrast agents for molecular imaging in 1995, and these agents ha ve continued to build promise for ultrasound tissue characterization of pathology.27,21,33,43–45 The f irst example of this class of agents, a nongaseous liquid PFC nanopar ticle emulsion, has proven useful in the diagnosis of earl y cardio vascular disease and cancer .21,33,39,41,43–46 Subsequent w ork b y other groups has demonstrated potential for targeted gascontaining liposomes31,32,47–50 and microbubbles as highly sensitive molecular detectors for certain classes of vascular biomarkers.30,51–53 As mentioned, liquid perfluorocarbon nanopar ticles were the f irst vehicles developed for site-directed imaging with ultrasound. Nongaseous nanopar ticles, unlik e the microbubbles, are weak scatterers that are not generally effective as blood-pool contrast agents at clinical frequencies. Importantly, this feature is useful since there is no competing background signal (~30 dB below blood at clinical frequencies) to obscure the identif ication of specific binding. 54 The long circulating half-life of PFC nanoparticles and their immunity to destr uction by ultrasound at clinical and higher imaging powers and frequencies renders them suitab le agents for molecular imaging and drug delivery. The mechanism of action of liquid perfluorocarbon nanoparticles as contrast agents has been modeled using

acoustic transmission line approaches that account for the material properties and the geometry of the particles.37,38,55 Such physical properties as the speed of sound through the liquid and the material compressibility def ine their intrinsic acoustic impedance, w hich dif fers signif icantly from that of the sur rounding targeted tissues. The amplitude of reflected plane waves impinging on the interface is a function of the la yer thickness and acoustic impedance mismatch between the perfluorocarbon and the substrate and surrounding media and is gi ven b y R(k) = R 12 + T12 T12 R23e2ikd/(1 – R21R23e2ikd), where k is the wave number in the host medium, d is the thickness of the contrast layer, Rxy and T xy represent the reflection and the transmission coefficients at interf ace of media x and y , and subscripts 1, 2, and 3 represent the host medium, perfluorocarbon layer, and substrate, respectively. Although their small size yields a “per unit” backscatter cross section that is considerab ly reduced as compared with the micron-sized bubb le contrast agents (eg, ~1000-fold smaller in v olume), when they bind and concentrate on the surf ace of a cell or tissue, these par ticles create a local acoustic impedance mismatch that produces a strong ultrasound signal without a concomitant increase in the background signal. Indeed after targeting, the magnitude of the reflected ultrasound w ave ma y exceed the native reflectivity of some materials by 20 dB or more, depending on the physical properties of the substrate to which the nanoparticles bind.27,56 The signal augmentation upon specif ic binding to intrinsicall y bright objects is less than for poorly reflective substrates, but this is also the case for microbubb les. Also, no “shadowing” (attenuation) ar tifact is created b y binding as can be observed for high concentrations of microbubb les in circulating blood or at tissue sites. The lack of backg round signal and the need to achie ve a threshold concentration of locall y bound par ticles ser ves to enhance specif icity for detection of molecular epitopes since some de gree of nonspecific binding always can be expected for any contrast agent. Unlike the case for tar geted microbubb les that can manifest very low numbers of binding e vents per cell and exhibit e xtreme sensitivity such that e ven single bubb les can be readil y detected,57 these liquid PFC nanopar ticles behave more as an ensemb le of scatterers in repor ting the concentrations of scarce molecular epitopes. Fur ther work from our lab has conf irmed that liquid nanopar ticles are acting neither lik e bubbles to create acoustic backscatter , nor are they susceptible to disr uption by ultrasound as are microbubble agents, such as Optison® or Definity®. On the basis of measurements of ph ysical proper ties of PFC nanoparticles, we have detected no resonance phenomena at an y frequenc y or po wer, and the acoustic attenuation

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

profile of these agents is not characteristic of bubb le behavior.39,54 Also, Soman and colleagues58 proved directly that ultrasound ener gy applied to human umbilical v ein endothelial cells e xposed to Def inity in culture at clinical powers and frequencies causes marked cell membrane disruption and death b y cavitation, whereas similar insonif ication parameters applied to cells e xposed to PFC nanoparticles did not alter appearance, membrane parameters, or viability . Accordingly, the mechanism of contrast enhancement and the response to ultrasound is entirely different for these agents as compared with microbubb les. Furthermore, because they are not destroyed by cavitation in a sound f ield, with resulting disr uptive effects on living tissues, their safety prof ile is benign.

547

ranging from 13.96 ± 1.5 dB at 27°C to 18.16 ± 1.5 dB at 47°C (Figure 5, top panel). The increase in ultrasonic contrast enhancement measured w as w ell described b y a simple, acoustic transmission line model with temperature-dependent material proper ties f actoring into the intrinsic acoustic impedance prof iles (see F igure 5, bottom panel). These results suggest that temperaturedependent changes in acoustic backscatter from these particles may be used to discriminate tissues targeted with site-specific nanopar ticles from sur rounding nor mal soft tissues. Fur thermore, if radio frequenc y (r f ) detection schemes are suf ficiently sensiti ve (see belo w), the backscatter augmentation-deca y c ycle that w ould occur upon mild transient local tissue heating and cooling would serve as a unique ultrasonic molecular nanobeacon.

Unique Signatures: Temperature Dependence of Scattering

Imaging Applications Fibrin: A sine qua non of the disr upted plaque is f ibrin deposition. Not onl y is f ibrin deposition one of the earliest signs of plaque rupture or erosion but along with intraplaque hemor rhage, it also for ms a considerab le part of the core of growing lesions. The diagnosis of disrupted plaque by detection of small deposits of f ibrin in erosions or microfractures could allo w characterization 20 Reflection Enhancement (dB)

Molecular imaging contrast agents specif ically detect the biochemical “signatures” of disease before anatomic manifestations are apparent. What is lacking for molecular imaging with ultrasound is a unique signal that is clearl y differentiable from backg round, as is the case for nuclear imaging. Because the acoustic proper ties of perfluorocarbons are known to vary with temperature (speed of sound decreases significantly at higher temperatures), 56 we have shown that temperature can be used to augment the magnitude of enhancement impar ted b y tar geted nanopar ticles. As an example, Figure 4 shows acoustic backscatter from human plasma clots tar geted b y f ibrin-binding nanoparticles measured at temperatures ranging from 27° to 47°C in 5°C increments. Ultrasonic contrast enhancement from bound nanopar ticles e xhibited a temperature-dependent ultrasonic signal enhancement

Experiment

18.1 16.4

15

17.0

15.0 13.9

10

27

32 37 42 Temperature (C)

47

20 Reflection Enhancement (dB)

Control

Treated

32

37 42 Transition from 27 C to

47

Figure 4. Experimental fibrin “dot-clots” on cellulose substrate treated with fibrin-targeted nanoparticles and exposed to heat. Difference in enhancement from 27°C at specific temperatures for sitetargeted human plasma clot. Darker grays represent greater changes in enhancement (Reprinted with permission from Hall CS et al.56)

Theory 15

10

15.9

27

16.8

17.7

18.5

32 37 42 Temperature (C)

19.3

47

Figure 5. Top panel contains experimental results for the reflection enhancement of human fibrin clot over control at specific temperatures. Bottom panel contains matched predictions from theoretic backscatter model. (Reprinted with permission from Hall CS et al.56)

548

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

of a potential “culprit” lesion before a high-g rade stenosis has been for med that is detectab le b y cardiac catheterization. The possibility of nanopar ticle-targeted fibrin imaging with either ultrasound or paramagnetic MR contrast agents was first demonstrated by Lanza and colleagues21 as earl y as 1996. In this case, the ligand comprised an antibody fragment highly specific for certain cross-link ed f ibrin peptide domains. 59 For ultrasound imaging, thrombi for med in situ in canine carotid arteries w ere detectab le within 30 minutes with commercially available 7.5-MHz linear ar ray imaging transducers (Figure 6).

particles on medial smooth muscle cells that manifestTF expression in response to injur y. In this case, the enhancement after intra venous nanopar ticle deli very is first obser ved after 15 to 30 minutes, reaching plateau values after 3 to 4 hours, and generall y remains detectable at the site for more than 24 hours. These data demonstrate the potential of this tar geted agent to bind and stick to epitopes in the v ascular media well beyond the lumen of the ar tery.

Tissue Factor

Angiogenesis is a key feature of tumor pathology and is required for tumor g rowth and metastasis. 63,64 The ability to detect earl y e vidence of accelerated v ascular growth, or the so-called “angio genic s witch” that triggers rapid tumor expansion, would be useful for staging, metastasis detection and early intervention, and therapy follow up. Challenges for site-tar geted nanopar ticle ultrasound contrast agents for detection of sparse epitopes e xpressed on neo vasculature concer n their detection in the presence of local bright echoes returned from the surrounding tissue, especially highly reflective (specular) vascular surfaces. To enhance the sensitivity for detection of neovasculartargeted PFC nanoparticles, we have developed new signal

Tissue f actor (TF) is a prothrombotic transmembrane glycoprotein e xpressed within plaques that is upre gulated after v ascular injur y or stent placement and that contributes as a mito gen to restenosis after angioplasty.60,61 With the use of tar geted PFC nanopar ticles, TF upregulation in the vascular wall can been detected in vivo by molecular imaging with ultrasound. 44,62 Figure 7 illustrates nanopar ticles tar geted to v ascular smooth muscle cell TF induced by balloon injury with the use of anti-TF monoclonal antibody (mAb) ligands. The ultrasound readout from a 30-MHz intra vascular ultrasound catheter readily demonstrates the deposition of 250 nm A

Angiogenesis Imaging in Pre-cancerous Lesions Using Novel Signal Processing Schemes

B

Thrombus Before Targeted Biotinylated Contrast

Thrombus After Targeted Biotinylated Contrast

Figure 6. Acoustic enhancement of canine femoral artery thrombus, targeted with biotinylated antifibrin antibody, before (A) and after (B) exposure to biotinylated perfluorocarbon emulsion. The acute arterial thrombus (left panel) is poorly visualized with a 7.5-MHz linear array, focused transducer. The transmural electrode used to create the clot (a) and the wall boundaries of the femoral artery (f) are clearly delineated (A). After exposure to the biotinylated emulsion, the thrombus is easily visualized. The anode (A) produces an ultrasonic shadowing effect in the midportion of the contrast-enhanced thrombus. (Reprinted with permission from Lanza GM et al.21)

Figure 7. Left: Tissue factor (TF)-targeted nanoparticles bound to vascular smooth muscle cells (SMCs) in vitro (lower left), showing competitive inhibition (specificity) of binding in lower right panel after first blocking with free monoclonal antibody (mAb) to TF. Middle: TF-targeted particles binding to medial SMCs induced to express TF after angioplasty balloon injury: bright enhanced regions due to nanoparticle binding (left panel, arrows). Right: TF epitopes in medial SMCs by immunocytochemistry (brown stain). (Reproduced with permission from Lanza GM et al.44,45)

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

αvβ3-integrin was used, which exhibits nM-binding affinities for αvβ3 as compared with µM for other related integrins (e g, GP-IIbIIIa). After injecting 0.5 cc/kg of αvβ3-integrin tar geted nanopar ticles, rf backscatter data were acquired from the ears with a 30 MHz VEVO-660 rodent imager. Rf data w ere processed for signal ener gy (a conventional detector) and entrop y (infor mation theoretic receiver). Only the Tr animals exhibited a progressive and sustained change in entrop y with nanoparticle targeting, as compared with a small but opposite-signed change in the wild type (WT). Also, traditional signal detectors (not shown) were relatively insensitive to the nanopar ticles after binding i n Tr animals. These data suggest that the nanoparticles bind specifically to αvβ3-integrin sites in Tr ears, but not in WT ears since the “sign” of the tw o responses is opposite. Also, the binding kinetics are v ery similar to pre vious obser vations of ultrasound contrast enhancement both in vitro and in vi vo with dif ferent ligands and substrates, 46,47,75–77 indicating the robustness of the targeting process. Because the targeted ears are very thin (~300 to 500 microns), these data pro vide a reasonably stringent test case for ultrasound molecular imaging of small and dif fuse tar gets. Additionally, these same ultrasound recei ver algorithms ha ve been applied clinically to patients with Duchenne’ s muscular dystroph y (without nanopar ticle tar geting) to detect inflammator y changes in affected muscle groups.40 Cancer Imaging with Ultrasound Information Theoretic Receivers

Angiogenesis in tw o additional models of o vert cancer has been imaged with the use of tar geted nanopar ticles revealing enhanced sensitivity of the information theoretic receivers v ersus conventional ultrasound signal process-

Change in % Area Enhanced

receivers (ie, mathematical operations that reduce an entire rf waveform or a portion of it to a single number) based on information-theoretic quantities, such as Shannon entrop y (H) or its counter part for continuous signal ( Hc).39,41,65,66 These receivers appear to be sensitive to diffuse, low amplitude features of the signal that often are obscured b y noise or else lost in large specular echoes and, hence, not usually perceivable by a human obser ver.65,67–70 Although entropybased techniques ha ve a long histor y in image processing for image enhancement and postprocessing of reconstructed images, the approach we have developed differs in that entrop y is used directl y as the quantity def ining the pixel values in the image. To illustrate the po wer of these detectors o ver traditional signal processing methods for re gistering nanoparticle binding to neo vasculature, w e performed preliminar y e xperiments in the K14-HPV16 transgenic (Tr) mouse tumor model that e xpresses human papillomavirus type-16 (HPV-16) oncogenes under control of the k eratin-14 (K-14) promoter .41 These mice e xpress the HPV-16 E6 and E7 oncogenes in the basal cells of their squamous epithelia 71; in the FVB/n strain backg round, they spontaneously develop epidermal squamous cell cancers (SCC) in a multistage f ashion.72 Their skin de velops highly angiogenic focal dysplasias between 3 and 6 months of age. By 1 year, 50% of these lesions progress to invasive SCCs, accompanied by the upregulation of proangiogenic factors, such as v ascular endothelial g rowth f actor73 and basic f ibroblast growth factor,74 and by the e xpression of geted αvβ3-integrins on endothelial cells that can be tar with ligand-directed PFC nanoparticles. In F igure 8, w e illustrate the tar geting and the quantification of binding to the neo vasculature of these pre-cancerous lesions. To tar get the PFC par ticles, a small-molecule RGD vitronectin antagonist against the

549

10

Tr

WT

0 15 30 45 60 Time Post-Injecton (min)

0 15 30 45 60 Time Post-Injecton (min)

5 0 –5

–10

Figure 8. “Entropy imaging” of neovasculature in mice ears with targeted nanoparticles. Left: αvβ3-integrin targeted nanoparticles injected into transgenic K14-HPV16 (Tr) or wild type (WT) mice and ears imaged for 60 minutes with 30 MHz rodent system. Left Mid: Ultrasound (US): One mouse ear is marked on either side with a red arrow. Colorized entropy values (Hc) are overlaid on the regular video image. After 60 minutes the Tr ear exhibits marked changes in entropy (purple) in the central vascular region, whereas the WT ear (turquoise) remains stable. Right Mid: Immunohistology stain for vasculature—markedly expanded (red stained) in Tr ear and associated with dysplasia. Right: Time course for Tr and WT mice indicates marked augmentation of Hc for angiogenic ears that bind nanoparticles but a slight decrease for WT. (Reprinted with permission from Hughes MS et al.41)

550

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

DHc Mean (bits/symbol relaitve to 0 min)

ing. One model, the MD A 435 tumors implanted in athymic nude mice, stimulates the de velopment of neovasculature expressing αvβ3-integrins on endothelial cells around the tumor rim but in relati vely sparse concentrations elsewhere. After injection of αvβ3-integrin targeted particles, the change in entrop y is obser ved as par ticles accumulate at the sites of neo vasculature.39 Another model, the VX2 tumor x enograft implanted subcutaneously in rabbits, also manifests a progressive increase in entropy after binding of αvβ3-integrin particles (Figure 9). It is interesting that in all of these models, the progressive change in entrop y after tar geting of the neo vasculature is relatively similar in magnitude and time course despite the dif ferences in animal models or imaging equipment. This observation may reflect the fact that most tissue is comprised of scatterers ha ving a re gular, easily described str ucture, w hereas tumor neo vasculature exhibits a tor tuous and con voluted architecture that is more complicated to describe. The potential implication is that the infor mation theoretic recei vers, entrop y in particular, are sensiti ve to such architectural dif ferences 0.25 Targeted (N–9) Control (N–6)

0.2 0.15 0.1 0.05 0

0

20 40 60 80 100 120 140 Time Post-Injection (min)

Figure 9. Time-dependent ultrasound enhancement from entropy (Hc) images of rabbits injected with αvβ3-integrin targeted or nontargeted (control) nanoparticles. Vertical axis: Hc-thresholded pixel values relative to Hc value at zero minutes.

whose fundamental complexity is revealed after the binding of nanoparticles.41 It is already clear that this speculation may be the case as was mentioned above for detection of tissue architectural features (ie, comple xity) in Duchenne’s muscular dystroph y even without nanopar ticle binding.40

Quantification To date the signals deri ved from con ventional ultrasound contrast agents, such as microb ubbles, have not been able to predict a quantitati ve relationship between the concentration of contrast agent and the received signal magnitudes. Although the e xtreme sensiti vity to even single bubbles could be useful for molecular imaging of sparse epitopes in vi vo as a screening test, the existence of a simple relationship betw een a local concentration of targeted microbubbles and received signal intensity (or other parameter) appears dif ficult to envision. Contrast agents used in other imaging modalities, such as nuclear , positron emission tomo graphy (PET), and MRI, bear a demonstrab le monotonic relationship between agent concentration and signal magnitude. The significance of modeling such relationships for ultrasound contrast agents resides in the potential use of molecular imaging constr ucts for longitudinal e valuation of target epitope concentration after therap y. Or, at the least, reliable semiquantitative estimates of trends in epitope concentration could be indicative of therapeutic efficacy, thereby allowing these agents to ser ve as surrogate monitoring tools for clinical management. To delineate the relationship betw een the magnitude of contrast enhancement of targeted surfaces and the density (and concentration) of tar geted perfluorocarbon (PFC) nanopar ticles, we e valuated backscatter from tw o dramatically dif ferent model substrates (F igure 10):

8 mm

8 mm

No Treatment Targeted Oil

NonTargeted PFOB

Targeted Targeted PFOB Oil/PFOB (50-50 mixture)

Figure 10. “En face” cell surface imaging of ultrasound backscatter from smooth muscle cells in culture under selected treatment conditions. Brightness indicates magnitude of backscatter from cell surfaces exposed to (from left to right): buffer solution, targeted but nonreflective nanoparticles (oil based), nontargeted but reflective (PFOB based) particles, a 50-50 mixture of targeted oil and PFOB particles, and 100% targeted PFOB nanoparticles. (white = −10 dB, black = −28 dB). (Reprinted with permission from Marsh JN et al.55)

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

(1) flat, thin monola yer avidin-coated agar disks tar geted with biotinylated PFC nanoparticles and (2) thicker layers of cultured smooth muscle cells expressing the transmembrane gl ycoprotein “TF” tar geted with anti-TF antibody bearing PFC nanopar ticles. Nonreflecti ve oil-based (no PFC) nanoparticles of the same size w ere used as a control-targeted emulsion. 55 The relationship betw een ultrasound reflecti vity enhancement and bound PFC content (measured by gas chromatography) varied in a curvilinear but monotonic f ashion and exhibited an apparent asymptote of approximately +16 dB and +9 dB enhancement for agar and cell samples, respecti vely, at the maximum PFC concentrations (~150 µg and ~1000 µg PFOB for agar and cell samples, respectively). Despite the dif ferences in the substrate to which the particles bind, the resultant quantitative binding prof iles sho wn in F igure 11 w ere both described by curves of the for m y = A(1 – e−Bx) with simA

Agar

Reflectivity Enhancement (dB)

20

15

5 Experiment y = 15.9* (1

0 0

B

20

40

60 80 100 mg PFOB

120

e–0.005x)

140

160

Smooth Muscle Cells

12 Reflectivity Enhancement (dB)

ilar le vels of statistical signif icance. Samples tar geted with onl y oil-based nanopar ticles e xhibited backscatter values that w ere identical to untreated samples (< 1 dB difference), conf irming the specif icity of the enhancement due to the PFC substrate itself. The specif icity of enhancement also w as conf irmed in competiti ve binding experiments with mixtures of low and high contrast particles that produced inter mediate le vels of contrast enhancement. These data illustrate the potential for tar geted PFC nanoparticles to augment acoustic reflecti vity monotonically from surf aces after binding to a selected molecular epitope. These observations suggest that substantial contrast enhancement with liquid perfluorocarbon nanoparticles can be realized e ven in cases of par tial surf ace coverage (as might be encountered w hen tar geting sparsely populated epitopes) or w hen targeting surf aces with locally irregular topography. Furthermore, it may be possible to assess the quantity of bound cellular epitopes through acoustic means. 19

10

10 8 6 4 2 Experiment y = 9.35* (1

0 0

200

400 600 mg PFOB

800

e–0.00022x)

1000

1200

Figure 11. Experimentally observed backscatter enhancement vs. bound [PFOB] for (A) agar phantoms and (B) smooth muscle cells in culture exposed to PFOB nanoparticle mixtures (data from Figure 10), along with curve fits of the form y = A(1 − e−Bx). (Reprinted with permission from Marsh JN et al.55)

551

F-NMR AND MOLECULAR MRI

MRI offers several advantages over the other modalities, such as high resolution, high anatomic contrast, high signal-to-noise (SNR), wide-spread clinical a vailability, and lack of ionizing radiation. Ho wever, the comparatively modest MR contrast enhancement achievable with targeted contrast agents for molecular imaging necessitates the delivery of higher pa yloads of contrast materials, w hich can be pro vided b y novel nanotechnologies. Because molecular epitopes of interest ma y reside on or inside of cells in very sparse quantities at low nanomolar or picomolar concentrations, considerab le amplif ication of the local contrast ef fect might be achie ved by incorporating large amounts of paramagnetic or other relaxing agents as the payload.78 In the case of T1-weighted imaging, the surf ace of the par ticle can be decorated with numerous copies of Gd chelates (100,000+) to achie ve the micromolar concentrations required per v oxel. F or PFC par ticles, the oppor tunity to use a unique MRI/MRS signature emanating from its fluorine ( 19F) core can be adv antageous.79 The ability to decorate the extensive surf ace with high pa yloads of “P araCEST” relaxing agents also renders this approach feasib le.80 Recently, quantitati ve approaches ha ve been described for molecularl y tar geted emulsions that allo w the computation of concentration of bound nanopar ticles under cer tain circumstances, based on either the 1H- or 19 F-signals.78,79

552

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Mechanisms of Contrast Enhancement Paramagnetic Nanoparticulate Agents

The amount of MRI enhancement achie ved with paramagnetic PFC nanoparticles depends upon the number of nanoparticles bound and the paramagnetic pa yload of each particle. One way to improve the MRI enhancement is to increase the spin-lattice (r1) relaxi vity per nanoparticle by attaching more Gd agents to each indi vidual particle, w hich shor tens the T1-time in the vicinity of the bound Gd. In this w ay, the relaxi vity at 1.5 T can be increased from 180,000 (s*mM) −1 (10 mole%) to 540,000 (s*mM) −1 (40 mole%), in ter ms of particle con81 centration, without reaching a saturation plateau. Increasing the pa yload, ho wever, can lead to impaired particle stability and might also interfere with the binding affinity of the targeting ligands. Therefore, the paramagnetic chelate is usuall y limited to 30 to 40 mole% of the total surfactant, yielding 100,000+ Gd atoms per particle. Another avenue of increasing nanoparticle relaxivity is to maximize the T1-shortening effect of the paramagnetic chelate. The relaxivity of a paramagnetic contrast agent depends upon the ef fective correlation time ( τc), a composite function of the rotational correlation time (τr), the electron spin-relaxation time ( τs), and the w aterexchange time ( τM), with the o verall effect being dominated b y the f astest process. The rotational cor relation time is largely independent of particle size over the range from 50 to 400 nm 82; therefore, ef forts to increase the relaxivity by altering particle size are ineffective.83 However, manipulation of the w ater-exchange time could lead to signif icant increases in the relaxi vity of paramagnetic nanopar ticles. F or e xample, w e ha ve demonstrated that the position of the Gd relati ve to the particle surf ace is impor tant for maximizing the exchange with free water molecules to augment r1 relaxivity.84 For these studies, tw o formulations were created: a Gd-DTPA-BOA, or bis-oleate constr uct, w hich positions the Gd in contact with the lipidic surf ace of the particle, and Gd-DTPA-PE, or phosphatidylethanolamine, that moves the Gd away from the par ticle surface. The PE construct in this case demonstrates the higher r1 relaxivity when measured in neat solutions. However, increasing the relaxi vity of paramagnetic nanoparticles w ould pro vide no practical benef it if the improved chelate interferes with the tar geting ef ficacy. Therefore, T 1-shortening must be e valuated with the nanoparticles bound to a biolo gic tar get, such as f ibrin incorporated into thrombus (F igure 12). Relaxation maps collected at 1.5 T sho w higher R 1 values at the clot

Figure 12. Scanning electron microscope images of (A) untreated fibrin clot and (B) and clot after application of fibrin-targeted PFC nanoparticles. A, Fibrin fibrils (arrows) form a tight weave in these acellular clots. B, Targeted nanoparticles densely bind to the exposed fibrin epitopes. (Reprinted with permission from Flacke S et al.81)

Figure 13. Color-coded maps of R1 relaxation rates from human plasma clots imaged on a clinical 1.5 T scanner. Fibrintargeted Gd-DTPA-BOA nanoparticles (left) induced significant T1 shortening at the clot surface (white arrow) relative to the clot interior. Gd-DTPA-PE nanoparticles, however, produced a 72% larger change in R1 compared with Gd-DTPA-BOA. (Reprinted with permission from Winter PM et al.84)

surface, demonstrating the paramagnetic influence of the fibrin-targeted nanoparticles (Figure 13). Compared with the clot interior , the surf ace la yer of Gd-DTP A-BOA nanoparticles increased R 1 at the clot mar gin b y 48% (ΔR1 = 0.45 ± 0.02 1/s), w hile the Gd-DTP A-PE agent increased R1 by 72% (ΔR1 = 0.77 ± 0.02 1/s). No signif icant differences were obser ved between Gd-DTPA-BOA and Gd-DTPA-PE nanoparticles in terms of the R1 values obtained inside the clot or in the saline sur rounding the clot. The increased R 1 observed for Gd-DTPA-PE treated clots was found to be a direct result of the improved relaxivity of this chelate and not caused b y any differences in nanoparticle binding characteristics. Alternative constructs with improved relaxivity relative to Gd-DTP A-BOA can be prepared b y changing the chelating moiety . Gd-MeO-DO TA-PE e xhibits greater par ticulate relaxi vities (1,470,000 s*mM −1) than Gd-DTPA-BOA (1,260,000 (s*mM) −1) The MeODOTA structure is a cyclic chelate, which may provide improved stability and relaxi vity compared with linear

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

chelates, such as DTP A.85 The impro ved relaxi vity likely reflects improved water exchange due to placing the Gd-MeO-DOTA complex farther out from the lipid surface. Adding a small spacer molecule (trigl ycine or caproate) to the DOTA construct enhances relaxivity of the agent by another 20%, presumab ly due to the e ven better water-exchange features. The DOTA chelate also provides more stable binding to the paramagnetic metal. Pre vious studies ha ve shown that DTP A transmetallation increases propor tionately with the loss of metal coordinate bonds through the coupling of lipophilic link ers.86 Macrocyclic DO TA-based chelates are w ell known to retain Gd much more a vidly than DTPA87 and therefore ma y be the prefer red avenue for molecular imaging applications gi ven the e xtended biologic half-life of tar geted contrast agents. The use of MeO-DOTA preser ves the coordination of Gd to the chelate w hile pro viding a reacti ve isothioc yanate side group for reaction with the ter minal amines of the PE anchor. 19

F Small-Molecule Reporters

NMR is a par ticularly rich phenomenon pro viding information through multiple parameters including chemical shift, relaxation processes (R 1 and R 2), scalar coupling, and chemical e xchange, and each of these parameters has been e xploited for specif ic 19F-NMR reporter molecules. The simplest concept of NMR is that of chemical shift, and 19F is exceptionally sensitive to molecular and microen vironmental changes. Fluorine NMR has a v ery large chemical shift range (~300 ppm) allowing multiple agents to be examined simultaneously with minimal danger of signal o verlap. NMR signal can be quantitati ve, so that the inte gral of a signal is directl y propor tional to the amount of material being interrogated. Detection sensitivity is governed by numerous parameters including the v olume of inter rogation, the required spatial resolution, and relaxation properties of the molecule and its tendency to accumulate or disperse from a re gion of interest. For preclinical investigations, there is a rich assortment of repor ter molecules, which have been designed specifically to e xploit fluorine chemical shift, coupling, or relaxation to re veal physiological parameters. Agents ma y be considered in tw o primar y classes: active and passi ve. Active agents typicall y f all into three cate gories: (1) a ph ysical interaction, for e xample, perfluorocarbons, w hich e xhibit e xceptional gas solubility and re veal o xygen tension (pO 2) based on modification of relaxation parameters 5; (2) re versible

553

trapping/binding of specif ic entities, such as ions, specifically, for e xample, H + (pH),88 and metal ions (Ca2+, Mg 2+)3; and (3) an ir reversible chemical interaction modifying str ucture of the repor ter substrate, as revealed by a change in chemical shift. These are represented by gene repor ter molecules, w here substrates are clea ved b y specif ic enzyme acti vity and h ypoxia agents, which are modif ied by reductases and trapped. By comparison, passi ve agents simpl y occup y, and hence, re veal a space, compar tment, or v olume, for example, tumor blood volume. To date, the most widespread applications ha ve considered o xygen tension, pH, and calcium ion concentrations. NMR oximetry exploits the paramagnetic influence of dissolved oxygen on the 19F-NMR spin-lattice relaxation rate (R 1) of a perfluorocarbon. 5 The solubility of gas, notably oxygen, in PFCs occurs as an ideal gas-liquid mixture, and thus, R 1 varies linearly with pO 2.5,89,90 The solubility of o xygen is much higher in PFCs than water, and thus, PFCs can act as molecular amplif iers. Several reviews have considered the relative sensitivity of different PFCs and individual resonances from complex molecules.91–94 R1 of PFCs is sensitive to temperature, and magnetic f ield, but impor tantly, it is essentially unresponsive to pH, CO2, charged paramagnetic ions, mixing with b lood, or emulsif ication.95–97 For the PFC emulsion of perfluorotributylamine (Oxypherol), it has been sho wn that calibration cur ves obtained in solution are valid in vivo.98 Many PFCs (eg, perfluorotributylamine, perflubron (for merly perfluorooctyl bromide; PFOB) and Therox™ (F44-E)) ha ve several 19F-NMR resonances, which can be exploited to provide additional infor mation in spectroscopic studies, but complicate effective imaging.99–101 PFCs with a single resonance pro vide optimal SNR ratio and simplify imaging: tw o agents hexafluorobenzene102 and perfluoro-15-crown-5-ether (15C5) 103 have found extensive use. The perfluoro-15-cro wn-5-ether has been emulsified for systemic administration to animals, whereas HFB is highl y v olatile and does not for m effective emulsions. Nonetheless, it has been used extensively based on direct injection into tissues. Many in vestigations based on PFCs ha ve used emulsions to probe tumor o xygenation. Uptak e and deposition of PFC emulsions in tumors are highly variable and hetero geneous with most signal occur ring in well-perfused re gions.104,105 Indeed, pO 2 values measured soon after intra venous infusion, but follo wing vascular clearance (typicall y, 2 da ys), are generall y high, approaching ar terial pO 2.104 Most PFC emulsion becomes sequestered in the RES. While this is ideal for

554

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

measuring pO 2 in the li ver, spleen, and e ven bone marrow, the associated hepatome galy is less satisf actory for studies focusing on tumors. Thus, effective use of PFC emulsions for tissues other than the RES will require the development of both tar geting the tissue of interest and detar geting the RES (stealth qualities), an active area of research. Direct injection of repor ter molecules into tissue allows immediate inter rogation of any region of interest. This approach w as used b y Berkowitz and colleagues 106 to examine pO2 in the eye and Duong and colleagues107 in the v entricular ca vity of the brain. Direct injection has been widely used by Zhao and colleagues 5 in tumors and is now f inding application in other tissues. 91 The sensitivity of R 1 to oxygen is attributab le in par t to the v ery long spin-lattice relaxation times found in perfluorocarbons, which can lead to long imaging acquisition times, particularly for relaxation measurements required for pO2 determinations. Echo planar imaging accelerates data acquisition as used b y Dardzinski and colleagues 103 and developed b y Bark er and colleagues 108 using the socalled FREDOM (Fluorocarbon Relaxometry using Echo planar imaging for Dynamic Oxygen Mapping). 5 FREDOM typically provides 50 to 150 indi vidual pO 2 measurements across a tumor simultaneousl y in about 6.5 minutes with a precision of 1 to 3 tor r in relati vely hypoxic regions based on 50 µL injected dose. pO 2 heterogeneity is ob vious in most tumors w hen rats breathe air and dif ferential response is seen with inter ventions. FREDOM has been used to examine the effects of vascular targeting agents,109,110 vasoactive agents,111 and hyperoxic gases 5,111–120 on tumors. Measurements are consistent with sequential deter minations made using electrodes121,122 and f iber-optic probes (FO XY™ and y OxyLite®).113,123 Repeat measurements are highl reproducible and generall y quite stab le in tumors under baseline conditions. Results are also consistent with hypoxia estimates using the histolo gic marker pimonidazole.115 Most significantly, estimates of pO 2 and modulation of tumor h ypoxia are found to be consistent with modified tumor response to ir radiation.112,124 19 F-NMR ion indicators ha ve been based on three strategies: (1) de velopment of molecules specif ically designed for 19F-NMR, (2) e xploitation of the 19F-NMR chemical shift sensitivity inherent in cytotoxic drugs, (3) fluorinated analo gs of e xisting fluorescent indicators. Deutsch and colleagues 125,126 championed the use of 19 F-NMR to measure the intracellular pH based on the series of agents 3-monofluoro-, 3,3-difluoro-, and 3,3,3trifluoro-2-amino-2-methyl propanoic acid. A significant problem is loading indicators into cells, but esters are

relatively per meable, stab le in w ater, and under go nonspecific enzymatic h ydrolysis intracellularl y, liberating the pH-sensitive molecules. Aromatic reporter molecules tend to ha ve a much lar ger chemical shift pH response. Analogs of vitamin B6, for e xample, 6-fluorop yridoxol (6-FPOL) are highl y sensiti ve to pH. 88,127–130 6-FPOL readily enters cells and pro vides w ell-resolved resonances repor ting both intracellular and e xtracellular pH (pHi and pHe), simultaneously, in whole blood130 and the perfused rat heart.128 Fluoronucleotides deri ved in vi vo from 5-FU exhibit sensitivity to changes in pH and could be used to measure intracellular pH. 131–133 Since the pharmacokinetics of 5-FU are repor ted to be pH-sensiti ve, measurements of tumor pH ma y have prognostic value for drug efficacy, and it would be par ticularly attractive to use the dr ug and its o wn metabolites for these measurements. 131,132,134 Tsien 135 developed fluorescent metal-ion chelators and estab lished an approach for loading them into cells using acetoxymethyl esters, for example, 1,2-bis(o-aminopheno xy)ethane-N,N,Nʹ′,Nʹ′tetraacetic acid (B APTA) detects intracellular calcium 136 ions. Metcalfe and colleagues added parafluoro atoms to the aromatic ring yielding a 19F-NMR responsive agent (5,5-difluoro-1,2-bis(o-aminophenoxy)ethane-N,N,Nʹ′,Nʹ′-tetraacetic acid (5F-B APTA)). Upon binding calcium, there is a change in chemical shift. Ideall y, such a repor ter molecule w ould ha ve high specificity for the metal ion of interest. In fact, the F-BAPTA agents are found to bind se veral di valent metal ions, including Ca 2+, Zn 2+, Pb 2+, F e 2+, and Mn2+,137,138 but importantly, each metal-ion chelate has an indi vidual chemical shift, so that the y can be detected simultaneously.139 A third class of repor ter molecule is based on substrates for specific enzyme reactions. FDG is phosphorylated and trapped re vealing metabolicall y acti ve cells. Fluoromisonidazole under goes reduction, w hich is ir reversible in the absence of o xygen re vealing hypoxic tissues. 140,141 Studies have repor ted the fluorinated nitroimidazoles CCI-103F,142 Ro 07-0741, 143 and SR-4554,141,144,145 which contain six, one, and three fluorine atoms per molecule, respecti vely. Subsequent to administration, a washout period sufficient for elimination of unbound marker is required since there is apparently no dif ference detectab le in vi vo in the chemical shifts of the parent molecule and the metabolites. 141 In each case, these methods essentially duplicate standard PET approaches. Transgene acti vity ma y be detected using highly sensitive substrates, as reported for c ytosine deaminase (CD) 146 and β-galactosidase.147

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

19

F-Nanoparticle Imaging and Spectroscopy

Alternatively, the signal generated b y the fluorine atoms within the perfluorocarbon core of perfluorocarbonbased nanoparticles has been introduced as a unique signature for MR molecular imaging. 79,148–150 Because biologic tissues contain little endo genous fluorine, measurement of the fluorine component of tar geted particles has the potential for def initive conf irmation of nanoparticle deposition at the site. While not all nuclei exhibit the magnetic resonance ef fect and therefore cannot be measured directly, 19F, like 1H, has one unpaired proton and no unpaired neutrons and thus with a net spin of 1/2 exhibits the NMR phenomenon. Serendipitousl y, the gyromagnetic ratio ( γ) for 19F is also close to that of 1H (ie, 40.1 MHz/T vs. 42.6 MHz/T , respecti vely), w hich means not only are their resonance frequencies relatively similar but so are their relati ve signal strengths, with 19F being approximately 83% that of 1H. The liquid perfluorocarbon core represents 98% of the total nanopar ticle volume, leading to a substantial 19F-concentration of 100 M within perfluorooctylbromide particles. Furthermore, while the 19F-isotope of fluorine has a natural abundance of near 100%, the biologic presence is virtually zero. Therefore, the 19F-nuclei that are highl y concentrated within the perfluorocarbon core of the nanoparticle emulsions are a prime candidate for direct MR spectroscopy and imaging without sur rounding signal from endo genous fluorine (F igure 14). This recent approach has been demonstrated for molecular imaging

and spectroscopy of fibrin at 4.7 T and for quantifying the concentration of nanopar ticle binding to a selected site based on localized fluorine spectroscop y79 and subsequently for stem cell imaging (see belo w) in vivo.151 Another use for fluorine imaging and spectroscop y of perfluorocarbon par ticles is in reco gnizing dif ferent targeted moieties on the same sample. Due to differences in their local nuclear en vironments (eg, electron shielding, J-coupling), dif ferent fluorine atoms resonate at slightly different frequencies so that they are often readily separable on an NMR spectr um.152 This means that by using perfluorocarbon nanoparticles formulated with different perfluorocarbon species, it is possible to target them to the same sample and quantify their presence separatel y with one spectroscopic scan as shown b y Mora wski and colleagues 79 in F igure 15 for PFOB and cro wn ether compounds for mulated as nanoemulsions. While spectroscop y will ultimatel y be most useful for quantif ication, other techniques allo w imaging of the different perfluorocarbon particles as well. These may include frequency-selective excitation so that only the perfluorocarbon species of interest produces a signal or other for ms of chemical shift imaging that have been developed for differentiating fat from water in clinical imaging.153,154

Perfluoro-15-crown-5 PFOB PFOB

50

70

90

% Total Signal

B

110

130

150

PPM

A A

555

PFOB Crown Ether

100.0 75.0 50.0 25.0 0.0 1:0

B C Figure 14. A, 19F-image (4.7 T) of a single slice through a clot in vitro treated with fibrin-targeted crown ether emulsion. High signal is observed at the clot surface due to bound fluorinated nanoparticles. B, 1H-image (4.7 T) of the same slice showing the anatomy of the clot with significant background 1H-signal. C, False color overlay of 19 F-signal onto 1H-image clearly localizing 19F-signal to clot surface.(Reprinted with permission from Morawski AM et al.79)

3:1 1:1 1:3 0:1 Ratio of Nanoparticles

Figure 15. A, 19F-spectrum acquired at 4.7 T of a clot treated with a mixture of fibrin-targeted crown ether and PFOB emulsions. The crown ether peak and five discernible PFOB peaks are easily detected and individually resolved. B, Percentage of total 19F-signal attributed to crown ether or PFOB for the clots treated with different nanoparticle mixtures, which are listed as the ratio of PFOB to crown ether. Spectral discrimination of crown ether and PFOB allows simultaneous quantification of the two nanoparticle species within a single sample. (Reprinted with permission from: Morawski AM et al.79)

556

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

ParaCEST Nanoparticles

Chemical exchange saturation transfer (CEST) agents have exchangeable protons ( −NH, −OH, etc) that resonate at a chemical shift that is distinguishab le from the bulk w ater signal. RF prepulses applied at the appropriate frequenc y and po wer le vel can saturate the e xchangeable protons, which transfer into the bulk water pool and lead to reduced equilibrium magnetization. 155 Therefore, with the use of CEST agents one can s witch the image contrast “on” and “off ” by simply changing the pulse sequence parameters, an ability that is unique in the realm of MRI. This can minimize the time delays and motion-induced artifacts inherent in normal precontrast and postcontrast imaging protocols. Although se veral agents contain e xchangeable protons and can produce contrast due to saturation transfer,156 the chemical shifts are often very close to the bulk water signal, which makes it difficult to distinguish contrast due to saturation contrast v ersus direct saturation of the bulk water. Paramagnetic ions can be used to shift the bound water frequency further away from the bulk water, allowing distinct saturation of the e xchangeable protons.157 These paramagnetic CEST (P ARACEST) agents consist of paramagnetic chelates that are specif ically designed to e xhibit exchangeable proton or bound w ater peaks and are emerging candidates for molecular imaging applications. The supramolecular P ARACEST agents reported to date have been symmetrical chelates coupled to cationic polymers through ion-pair attractions 158,159 or entrapped within a liposomal vesicle.160 We have recently synthesized and evaluated an asymmetric P ARACEST chelate bifunctional modif ied for chemical coupling to a lipid-encapsulated PFC nanoparticle.80 Very high pa yloads of the lipid-conjugated PARACEST chelate can be incor porated into the nanoparticle surf actant to enhance the ef fectiveness and sensitivity of this PARACEST agent in neat solution. The PARACEST chelate is based on the 1,4,7,10-tetraaza macrocycle using N-substituted gl ycine ethyl ester ligating moieties and a functionalized aromatic group for lipid conjugation (Dow Chemical Co.) (F igure 16). Europium is added at equimolar concentrations to for m the PARACEST constr uct, Eu 3-methoxy-benzyl-DOTA, and is coupled by a lipophilic tail, phosphatidylethanolamine (Avanti Polar Lipids, Inc.), via a thiourea linkage. Proton NMR spectroscop y perfor med on the P ARACEST nanoparticles at 4.7 T re vealed a distinct bound w ater peak at 52 ppm (F igure 17), consistent with a pre viously reported symmetric PARACEST chelate.161 Furthermore, targeting is possible to substrates, such as f ibrin thrombi, for use as a molecular imaging agent (F igure 18).

O

O N H O C2H5O

N H O O C2H5O

O

N H H N

Eu3

OC2H4 H N

O

N

OC2H5

H

H N

H N S

O H3CO

O HO

H3C(H2C)14

O

P

O

O

O O H3C(H2C)14 O

Figure 16. Chemical structure of the lipid-conjugated PARACEST contrast agent. Eu3 is chelated to methoxy-benzylDOTA, which is functionalized with a phospholipid moiety for incorporation into the lipid membrane of perfluorocarbon nanoparticles. (Reprinted with permission from Winter PM et al.80)

Figure 17. A saturation profile of PARACEST nanoparticles shows a clear saturation contrast effect at 52 ppm (arrow), confirming saturation of the bound water peak and effective transfer of magnetization. Control nanoparticles do not show any saturation contrast effects at this chemical shift. (Reprinted with permission from Winter PM et al.80)

NMR and Molecular MRI Applications Atherosclerosis

One motivation for MR molecular imaging is the recognition and discrete localization of precursor events identifying vulnerable plaques that might provide a window of opportunity extending from days to weeks or months to inter vene before serious clinical sequelae ensue. 162

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

557

Fibrin-Targeted Nanoparticle Contrast PARACEST Nanoparticles

Clot

Contrast-To-Noise Ratio

PBS

Control Nanoparticles

Clot Suture

PBS

Suture

12

8

4

0

Figure 18. Top: Images of fibrin-targeted PARACEST (left) or control (right) nanoparticles bound to plasma clots. Images obtained with saturation at −52 ppm show no differences between clots treated with PARACEST or control nanoparticles. Middle: Subtraction images reveal signal enhancement on the surface of the clot treated with PARACEST nanoparticles and no enhancement of the clot treated with control nanoparticles. Bottom: The CNR calculated at the clot surface was significantly higher with PARACEST nanoparticles compared with control nanoparticles. (Reprinted with permission from Winter PM et al.80)

Unstable or disrupted atherosclerotic plaques frequently appear in various stages in arteries with only modest (40 to 60%) stenosis, 163,164 and the y remain diagnosticall y elusive with routine clinical imaging techniques. A sine qua non of the disrupted plaque is fibrin deposition. Not only is f ibrin deposition one of the earliest signs of plaque r upture or erosion, but along with intraplaque hemorrhage, it also for ms a considerab le par t of the core of g rowing lesions. 165 Upon acute r upture, TF is exposed and initiates local thrombosis that ma y lead to total occlusion and inf arction. The diagnosis of disrupted plaque b y detecting small deposits of f ibrin or TF in erosions or microfractures could allo w characterization of a potential “culprit” lesion before a highgrade stenosis has been for med that is detectab le b y cardiac catheterization.

Figure 19. Thrombus in canine external jugular vein targeted with fibrin-specific paramagnetic nanoparticles demonstrating dramatic T1-weighted contrast enhancement in gradient-echo image (arrow) on left and flow deficit (arrow) produced by thrombus in corresponding phase-contrast image on right (three-dimensional phase-contrast angiogram). (Reprinted with permission from Flacke S et al.81)

Paramagnetic PFC-based nanopar ticles ha ve been shown to yield a substantial amplif ication of signal from fibrin clots at 1.5 T both in vitro and in vi vo.81,166,167,78 Figure 19 from Flack e and colleagues 81 shows an e xample of localizing a f ibrin clot in an e xperimental canine model within 1 hour of par ticle injection, with contrastto-noise ratios betw een clot and b lood e xceeding 100. Furthermore, despite the lar ge concentration of par ticles and bound Gd, due to the excessive number of fibrin epitopes available for binding, no diminution of signal due to susceptibility effects was observed. Further confirmation of the ability of this f ibrin-targeted construct to bind to human thrombi in human endar tectomy specimens from disr upted carotid ar teries illustrates the potential clinical signif icance of this approach. 81 The additional ability to image TF-targeted paramagnetic nanopar ticles bound to smooth muscle cell monolayers in cell culture at 1.5 T attests to the potenc y of nanopar ticles agents that carry 50,000 or more Gd chelates. 78 Plaque Angiogenesis and Integrin Imaging

While f ibrin and TF can be used to delineate unstab le cardiovascular diseases, the αvβ3-integrin is a general marker of angiogenesis and plays an impor tant role in a wide variety of disease states including atherosclerosis and cancer.26,168–170 The αvβ3-integrin is a well-characterized heterodimeric adhesion molecule that is widel y expressed b y endothelial cells, monoc ytes, f ibroblasts, and v ascular smooth muscle cells. In par ticular, αvβ3integrin pla ys a critical par t in smooth muscle cell migration and cellular adhesion, 171,172 both of which are required for the for mation of ne w b lood v essels. The αvβ3-integrin is expressed on the luminal surface of activated endothelial cells but not on mature quiescent cells.173 Angiogenesis also plays a critical role in plaque growth and r upture.174,175 In re gions of atherosclerotic

558

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

lesions, angio genic v essels proliferate from the v asa vasorum to meet the high metabolic demands of plaque growth.176,177 Inflammatory cells within the lesion stimulate angio genesis through local molecular signaling, which in turn promotes neovascular growth, thereby providing an a venue for more inflammator y cells to enter the plaque.174 Molecular imaging of expanded vasa vasorum in atherosclerotic lesions in cholesterol-fed rabbits w as f irst demonstrated for MRI b y Winter and colleagues 23 with the use of paramagnetic nanopar ticles tar geted to αvβ3integrin expressing endothelial cells (F igure 20). Signals for specific binding could be detected within 30 minutes of contrast agent injection. Animals on a control diet exhibited no increased signal and backg round was minimal. Expression of αvβ3-integrins in the adventitial layer and be yond w as conf irmed b y colocalized histolo gic staining of αvβ3-integrin and PECAM, a general endothelial marker. This work demonstrated the initial e xamples of the potential for noninvasive detection and quantif ication of angio genesis in atherosclerotic plaque with the use of MRI. Cyrus and colleagues 178 recently used αvβ3integrin-targeted and collagen-III-tar geted paramagnetic nanopar ticles to image ar teries subjected to balloon stretch injury (angioplasty). They observed a high degree of binding associated with the e xposure of native smooth muscle cell integrins after injury and the upregulation of integrins as a component of the inflammatory response. The e xtent of molecular tar geting extended f ar be yond the length of the actual balloon itself indicating the potential for e xtensive y et occult injury beyond the confines of angioplasty and potential

re na l Chol1 a vb 31

mid

stent deplo yment, w hich ma y ha ve consequences for restenosis. Both collagen-III and αvβ3-integrins w ere detectable specif ically, although the MRI inte grin signal exceeded that of the collagen signal.

Cancer

High-resolution imaging of αvβ3-integrin e xpression on the neo vasculature of minute human melanoma tumors implanted in nude mice was demonstrated using targeted paramagnetic nanopar ticles and a clinical 1.5 T MRI scanner for T1-weighted imaging. 179 The a verage xenograft v olume 12 da ys post tumor implantation w as around 30 mm 3 or 2 to 3 mm diameter. Prior to nanoparticle treatment, these minute tumors appeared isointense with respect to skeletal muscle, making it difficult to distinguish them from sur rounding tissues (F igure 21). Injection of paramagnetic αvβ3-integrin-targeted nanoparticles produced abundant MRI signal enhancement in a hetero geneous pattern along the tumor periphery, allowing clear identif ication of the tumor location. Moreover, the enhancing re gion of angio genesis w as four-fold g reater in animals treated with αvβ3-integrintargeted nanopar ticles than in the nontar geted control. By histology, the neo vascular spatial distribution w as consistent with the contrast enhancement patter ns of angiogenesis detected with molecular imaging via αvβ3-integrin-targeted paramagnetic particles. In the VX2 tumor rabbit model, a xenograft implantation in an immunocompetant host, b y 2 hours after injection αvβ3-integrin-targeted paramagnetic nanopar ticles increased T 1-weighted MRI (1.5 T) signal b y 126% in

dia ph

100

Chol1 a vb 32

80 60

Chol2 a vb 31

40 20

Figure 20. Detection of plaque neovascularization in cholesterol-fed rabbits. Left: Aortic cross sections imaged at 1.5 T with αvβ3integrin targeted nanoparticles. Note: heterogeneous distribution of nanoparticle enhancement in aortic cross sections (false-colored contrast enhancement) but little enhancement in nontargeted rabbits (αvβ3-) or rabbits on a standard diet (Chol-). Right: Immunohistochemical staining for αvβ3-integrin at the media-adventitia border of aorta segments. Note abundant red-brown vascular segments. (Reprinted with permission from Winter PM et al.23)

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

A

B

Figure 21. A, T1-weighted MR image (axial view) of an athymic nude mouse before injection of paramagnetic αvβ3-integrin-targeted nanoparticles. Arrow indicates a C32 tumor that is difficult to detect (Ref = Gd in 10 cc syringe). B, Enlarged section of an MR image showing T1-weighted signal enhancement of angiogenic vasculature of early tumors over 2 hr as detected by αvβ3-targeted paramagnetic nanoparticles. (BL = baseline image). (Reprinted with permission from Schmieder A et al.179)

asymmetrically distributed re gions primaril y in the periphery of the tumor .180 Despite their relati vely lar ge size, nanopar ticles could be trapped in leak y tumor neovasculature but did not appreciably migrate into the interstitium, leading to a 56% increase in MR signal magnitude at 2 hours. Pre-tar geting of the αvβ3-integrin with nonparamagnetic nanopar ticles competiti vely blocked the specif ic binding of αvβ3-integrin-targeted paramagnetic nanopar ticles, decreasing the MR signal enhancement (50%) to a level attributable to local extravasation. Interestingly, positive T2-weighted images of tumor masses that w ere ne gative for αvβ3-integrin signal b y T1-weighted imaging of PFC nanopar ticles re vealed no viable tumor mass, and only inflammatory cells, indicative of tumor rejection and death. 19

F-Nanoparticle Imaging

Given their high fluorine content, f ibrin-targeted nanoparticles may provide quantification of exposed fibrin, indicating the size and/or number of r uptures in the fibrous cap. Imaging human carotid endar terectomy samples at 4.7 T, multislice 1H MRI showed high levels of signal enhancement along the luminal surf ace due to binding of targeted paramagnetic nanoparticles to f ibrin deposits.79 A 19F-projection image of the artery, acquired in less than 5 minutes, shows the asymmetric distribution

559

of f ibrin-targeted nanopar ticles around the v essel w all corroborating the 1H-signal enhancement (F igure 22). Spectroscopic quantif ication of nanopar ticle binding allowed calibration of the 19F-MRI signal intensity (see below). Core gistration of the quantitati ve nanopar ticle 1 map with the H-image per mits visualization of anatomic and patholo gic information in a single image. Combining infor mation from 1H- and 19F-MRI could allow prediction of subsequent occlusion or distal embolization and aid clinical decision making for acute invasive intervention versus pharmaceutical therapies. The 19F-signal from site-tar geted nanopar ticles has been measured at 1.5 T with spectroscopy as well as with imaging using a steady-state free precession pulse sequence.149 Both imaging and spectroscopy could detect and distinguish nanopar ticles containing either PFOB or perfluoro-15-crown-5-ether (CE) as the core material. The SNR for PFOB w as lo wer than CE (10 vs. 25, respectively), presumably due to the single CE peak (20 equi valent fluorine atoms) compared with the multiple PFOB peaks (17 fluorine atoms distributed o ver f ive peaks). A clear linear relationship betw een the 19F-signal intensity and perfluorocarbon concentration w as demonstrated for both PFOB and CE using both imaging and spectroscopy. The possibility for performing MR angiography with untargeted 19F-PFC nanopar ticles has been e xplored b y Neubauer and colleagues. 150 Because these agents are sized so as to remain in the vascular space, they might be used to def ine vessel geometry if sufficient signal sensitivity can be achie ved, w hich depends in par t on b lood levels of the 19F. Using a rapid scanning technique and custom-designed coils, high le vel v ascular signals w ere obtained at 1.5 T for in vitro and ex vivo samples. Also, in vivo imaging of rabbit carotid ar teries was successful both during direct intraar terial injection of agents and after noninvasive intravenous dosing and imaging under steady-state conditions. This study represents the f irst demonstration of small-v essel imaging at clinical f ield strengths with sufficient temporal resolution to view both first-pass contrast enhancement and steady-state b lood signals for nanoparticles delivered noninvasively at moderate doses, corresponding to 1.5 to 2.5 mL emulsion per kg body weight (or equi valently, 0.5–0.9 g perfluorocarbon per kg). This dosage is w ell within the “absolute no effect dose” of 2.7 to 9 g PFC/kg deter mined using other PFC-blood substitute emulsions. 16 In addition to molecular imaging of f ibrin, 19F-MRI has been used to track dendritic cells and endothelial precursor cells labeled with PFC nanopar ticles.151,181 Stem cell tracking b y MRI has typicall y relied on iron oxide182–185 or Gd agents,186,187 but definitive identification

560

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

of labeled cells can be difficult due to innate susceptibility artifacts and v ariations in 1H-signal intensity within the body. 19F-MRI pro vides unambiguous identif ication of labeled cells because there is no confounding backg round signal. Ahrens and colleagues 151 recently adapted the fluorine nanoparticle molecular imaging approaches demonstrated originall y b y Lanza and colleagues 148 and Morawski and colleagues 188 by using a crown ether PFC preparation of nanopar ticles to load dendritic immune cells with no loss of viability , and the e xpression of cell surface mark ers, including CD-80 and major histocompatibility comple x, w as not ef fected. Accordingly, the y demonstrated an extension of the use of fluorine imaging at research f ield strengths (11.7 T) to track cells after local and systemic injections. 151 The adv antage to this approach is that no backg round signal e xists because there is no appreciable amount of fluorine in the body to confound the signal from the tar geted cells. Partlow and colleagues 181 recently repor ted methods for labeling of proangio genic endothelial precursor cells with multiple types of perfluorocarbon nanopar ticles for rapid imaging and spectroscop y at clinical (1.5 T) and research (11.7 T) f ield strengths (F igure 23). These cells avidly ingest the PFC nanopar ticles without the need for conjunctive transfection agents, such as lipofectamine, with no unto ward ef fect on cell viability and preser ved ability to contribute to cancer angiogenesis. Moreover, not only can these cells be uniquel y identif ied and track ed with MRI, but their local concentration can be quantif ied based on the 19F-spectral signature.

19

F-NMR Reporter Molecules

Many new industrial pharmaceuticals and agrochemicals incorporate a fluorine g roup providing a tool for NMR investigations.4 To date, most studies ha ve e xamined

A

B

pharmacokinetics and metabolism of fluorop yrimidines, particularly 5-FU. 5-FU is a primary drug in treatment of many cancers, but it has a nar row range of efficacy/toxicity.189–193 Presumably, both response and to xicity are related to pharmacokinetics, and there is interest in assessing the dynamics of uptak e, biodistribution, and metabolism. P atients with enhanced tumor retention of 5-FU (“trappers”) ma y be e xpected to e xhibit better response.194 Such trapping is apparentl y a requisite, though not in itself sufficient for efficacy.2 Following cancer chemotherapeutics, most in vi vo 19 F-NMR has e xamined psychiatric dr ugs.195,196 These can be particularly favorable when they incorporate a CF3 moiety. Se veral repor ts in vestigated fluo xetine (Prozac) with studies ranging from biopsy tissue e xtracts to pre-clinical animal models and human v olunteers.197–199 The primar y goal has been cor relation of concentration with ef ficacy.198 Many gaseous anesthetics are fluorinated , for e xample, halothane, enflurane isoflurane, se voflurane, and desflurane, and NMR studies of fluorinated anesthetics w ere some of the earliest in vivo applications of 19F-NMR.200–202 A recent innovation is the use of 19F-NMR to assess gene activity, in particular transgenes. For cancer, thymidine kinase has the adv antage that the gene ser ves not only as a repor ter, but gene products can themselv es have therapeutic value.203 Cytosin deaminase (CD) activates the minimally toxic 5-fluorocytosine (5-FC) to the highly toxic 5-FU.203,204 The conversion of 5-FC to 5-FU causes a 19F-NMR chemical shift ~1.5 ppm, hence revealing gene activity, which has been demonstrated in a number of systems in vi vo.146,204 A 19F-atom can be substituted for a hydroxyl group in sugars with little o verall structural perturbation. As such, fluorosugars were widely used to e xplore mechanisms of enzyme acti vity.205–207 Yu and colleagues h ypothesized that introducing 19F into the agl ycone moiety of a substrate could re veal β-galactosidase ( β-gal) acti vity and hence the lacZ gene, w hich has historicall y been the

C

Figure 22. A, Optical image ex vivo of a 5 mm cross section of a human carotid endarterectomy sample. This section showed moderate luminal narrowing and several atherosclerotic lesions. B, A 19F-projection image acquired at 4.7 T through the entire carotid artery sample shows high signal along the lumen due to nanoparticles bound to fibrin. C, Concentration map of bound nanoparticles in the carotid sample. (Reprinted with permission from Morawski AM et al.79)

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

A

Localized In Situ Injection

PFC Labels Labeled Cells

B

561

C

E

D

F

Figure 23. Localization of labeled cells after in situ injection. A, To determine the utility for cell tracking stem/progenitor cells labeled with either PFOB (green) or CE (red), nanoparticles were locally injected into mouse thigh skeletal muscle. B–D, At 11.7 T, spectral discrimination permits imaging the fluorine signal attributable to ~1 × 106 PFOB-loaded (B) or CE-loaded cells (C) individually, which when overlaid onto a conventional 1H-image of the site (D) reveals PFOB-labeled and CE-labeled cells localized to the left and right leg, respectively (dashed line indicates 3.3 cm2 field of view for 19F-images). E, F, Similarly, at 1.5 T, 19F-image of ~4 × 106 CE-loaded cells (E) locates to the mouse thigh in a 1H-image of the mouse cross section (F). The absence of background signal in 19F-images (B, C, E) enables unambiguous localization of perfluorocarbon-containing cells at both 11.7 T and 1.5 T. (Reprinted with permission from Partlow KC et al.181)

most popular reporter gene in molecular biolo gy.208 The prototype molecule 4-fluoro-2-nitrophen yl β-D-galactopyranoside (PFONPG) pro ved ef fective as a substrate for β-gal.209 It provides a single 19F-NMR signal with a narrow line width and good stability in solution. It is stable in nor mal WT cells and w hole blood, but exposure to the enzyme or cells transfected to e xpress β-gal causes rapid cleavage in line with anticipated le vels of transfection.209 Upon cleavage of the glycosidic bond a substantial chemical shift dif ference is obser ved, which is suf ficient to permit chemical shift selective imaging to reveal distribution of each entity separatel y.210 To date, the agent has been used to re veal stab ly transfected v ersus WT breast and prostate tumors g rowing in mice, using spectroscop y following direct intratumoral administration of substrates.147,211 Since β-gal exhibits wide promiscuity (lack of substrate specif icity) alter nate substrates ha ve been prepared to enhance signal sensitivity (eg, introduction of a CF3 moiety212) or reduce product toxicity (eg, fluoropyridoxol aglycone in place of nitrophenol). 213

Quantification of MR Signals The sensitivity of MRI for detection of paramagnetic or superparamagnetic nanopar ticulate imaging agents depends on the specif ics of f ield strength, pulse sequence, coil sensiti vities, epitope pre valence, contrast agent concentration, and so forth. In general, the MR signal strength is modest as compared with nuclear imaging applications. F or e xample, at clinical imaging f ield strengths, micromolar concentrations of paramagnetic agents, such as Gd or CEST agents, are required. F or superparamagnetic agents that elicit susceptibility ar tifacts, nanomolar concentrations might suffice. “Contrastto-noise” (CNR) ratios of five or better generally produce readily identif iable (diagnostic) qualitati ve signal enhancement. Examples of optimizing pulse sequences and paramagnetic and fluorinated nanopar ticle concentrations to enhance sensiti vity have been w orked out b y Morawski and colleagues 78,79 with MR signal modeling approaches.

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Comparison of signal intensity changes ef fected by Gd on T1-weighted MRI is a relative measure of signal quantification. More precise measures are deri ved from traditional “quantitative” imaging methods, such as T1-mapping.214 The local effects of MRI contrast agents deter mine both T1 and T 2 relaxation phenomena of nearb y hydrogen nuclei. For the ideal situation and known contrast agent relaxivities, these effects can be modeled mathematically with the use of the Bloch equation 215,216 to predict signal intensity as a function of agent concentration for a given pulse sequence. On the basis of initial w ork by Ahrens and colleagues, 217 Morawski and colleagues 78 modeled the beha vior of paramagnetic perfluorocarbon nanopar ticles and def ined the minimal concentration requirements for Gd-nanopar ticles within an imaging voxel to produce conspicuous MR signal enhancement (for CNR > 5), w hich appears to f all in the high picomolar range for these PFC nanopar ticles for clinical imaging f ield strengths and typical pulse sequence parameters (F igure 24). This par ticle concentration corresponds to the micromolar range for [Gd] and reflects the high Gd pa yload per par ticle (10 5 per particle) and the enhanced ionic relaxi vity due to association with a lar ge, slowly tumbling particle (10× enhancement over Gd-DTPA, for example). An interesting consequence of these modeling approaches is the prediction that impro vements in r1 relaxivities of MR contrast agents produce diminishing increments for tar get detectability abo ve values of around

40 (s*mM) −1 at clinicall y applicab le f ield strengths (see Figure 24). The practical value of this prediction is that the T1-shortening ef fect of a tar geted contrast agent is more dependent on the total concentration of Gd in a v oxel than the relaxation beha vior of single Gd atoms, w hich is a strategy that per mits f ar more simultaneous interactions between Gd and unrelax ed protons than altering the r1 value to increase the proton tur nover rate for single Gd. Thus, for nanopar ticles, the high pa yloads of Gd (10 5 per particle) dominate relaxation efficacy (only 10× greater) as long as the par ticles are relati vely w ell distributed and accessible to free water throughout the voxel. In e xperimental tests of this modeling approach for Gd-nanoparticles bound to TF expressing cells in vitro, the imaged-based signal models ag reed w ell with actual measurements of the concentration of Gd , and hence, the concentration of par ticles bound to cells (F igure 25). Accordingly, the signal modeling gi ves a quantitati ve estimate of the number of molecular binding sites that interact with the tar geted nanopar ticles. This algorithm, while more precise than simple T1-weighted signal

2

20

*

1.5 1

5

0

Targ

A T

NT NT

UT

Targ

0

B

UT

Diagnose cell monolayer region of interest

Figure 24. Calculation of the effect of altering the ionic r1 relaxivity of the nanoparticles on the minimum concentration of nanoparticles [NP] needed for diagnostic contrast enhancement (ie, a CNR = 5) at 1.5 T and 4.7 T. Transverse relaxivity (r2) was assumed to be 1.5 times r1 at 1.5 T, and 3 times r1 at 4.7 T based on data obtained with the current formulation. The asterisks (*) indicate the ionic-based r1 relaxivities of a representative nanoparticle formulation at both field. (Reprinted with permission from Morawski AM et al.78)

10

0.5

C

Model (NT) Data (NT) Model (T) Data (T)

15

* CNR

Paramagnetic PFC Contrast Agents

R1 (1/8)

562

0

1000 2000 TR (ms) T

NT

3000 UT

D

Non-targeted

Figure 25. A, Measured relaxation rates in cellular monolayers for the three treatment groups. Targeted (T) cells are shown on the left, nontargeted (NT) in the center, and untreated (UT) cells on the right. The dotted line indicates the lowest relaxation rate that will produce a contrast-to-noise ratio (CNR) (CNR = 5). B, CNR for targeted and nontargeted cell layers over a range of TRs, as observed in the SE experiment and predicted by the signal model. C, Spin echo images of smooth muscle cell monolayers acquired with an optimized TR of 900 ms. The magnified inset of the targeted and the nontargeted layers shows the region of interest (ROI; outlined in red) used to obtain the signal intensity measurements. ROIs were selected from T1 maps of the cell monolayers and applied to all subsequent images. D, A maximal intensity projection through a three-dimensional stack of T1-weighted images acquired parallel to the cell monolayers shown for one replicate illustrates the sensitivity of this targeting method for displaying changes from small numbers of cells at clinical field strengths. (Reprinted with permission from Morawski AM et al.78)

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

comparisons, is still limited b y assumptions about relaxivities, water exchange rates, local proton concentrations, and other potentially confounding effects. Fluorine Signal Quantification

As an alter native to modeling or measuring indirect effects of contrast agents on w ater molecules (eg, T1 or T2 for 1H), one could measure the signal of the contrast agent directl y as is the case for nuclear medicine radioisotopes. The 19F-nucleus within the perfluorocarbon core of the nanopar ticle emulsions is a prime candidate for direct MR spectroscop y and imaging that is not obscured b y an y sur rounding signal from endo genous fluorine since it is not present normally in the body in detectab le quantities. Using the 19F-signal at 4.7 T from f ibrin-targeted, paramagnetic perfluorocarbon nanoparticles, Morawski and colleagues 79 spectroscopically quantified the varying amounts of PFC nanopar ticles bound to thrombi in vitro and sho wed that these estimates matched the actual measured concentrations of par ticles according to neutron acti vation analysis of the Gd on the nanopar ticles.218 This technique also w as applied to human carotid endar terectomy samples to provide a quantitative measure of microscopic amounts of f ibrin associated with r uptured atherosclerotic plaques in patients with carotid ar tery disease. Extending this to a 1.5 T MR system using rapid steady-state imaging techniques, in vitro f ibrin-bound nanoparticles of two different species of perfluorocarbon (PFOB and CE) in v arying volumes were not only quantified via clinicall y-relevant imaging techniques, but were also identif ied and independently imaged based on their unique MR spectral signatures.149 From this demonstration on f ibrin clots, one might e xtrapolate the potential of using multiple perfluorocarbon nanopar ticle agents, each tar geted to a dif ferent epitope, to perfor m noninvasive, imaging-based immunohistochemistr y, for example, quantifying simultaneousl y the amount of angiogenesis and e xposed f ibrin associated with an atherosclerotic plaque as an indicator of its pathophysiologic significance.

563

PFC nanopar ticles for the detection of angio genesis has been demonstrated in the rabbit VX2 tumor model.29 Reproducible labeling of the nanopar ticles at pa yloads w as achieved at appro ximately 10 nuclides per par ticle. 111Inlabeled αvβ3-integrin-targeted PFC nanopar ticles yielded high tumor-to-muscle ratio signals, which was enhanced for the for mulations with 10 nuclides per par ticle v ersus 1 nuclide per par ticle. The 111In-labeled nanopar ticles provided a high sensiti vity, lo w-resolution signal from the tumor neo vasculature that w as easil y reco gnized within 15 to 30 minutes after injection and w hich persisted for hours.

Optical

On the basis of the ability to conjugate a variety of agents to the PFC nanoparticles through stable lipid anchors, we have developed approaches for comple xation of fluorescent agents that can generate signals for both in vi vo detection and for immunoc ytochemistry. F or e xample, vascular cell adhesion molecule (VCAM) tar geting and imaging in vi vo in mice is demonstrated b y intra vital microscopy (Figure 26) with the use of a mAb ligand to VCAM that was prepared by recombinant genetic methods. The par ticles ha ve been rendered fluorescent with the use of Alexafluor reagents that can be inser ted into the lipid membrane with lipid anchors. These par ticles are injected intra venously and localize to areas of acute laser injury in mouse cremaster arterioles (red arrow) but not in other uninjured se gments, concordant with the upregulation of VCAM in response to the injur y. Immunocytochemical depiction of inflammation is

Other Imaging Applications for PFC Nanoparticles Nuclear Imaging

The capacity of targeted PFC nanoparticles to serve as a universal carrier of imaging agents can be e xtended to nuclear imaging. The utility of 111In-labeled αvβ3-integrin targeted

Figure 26. Vascular cell adhesion molecule targeting (monoclonal antibody (mAb) ligand) with fluorescent nanoparticles (red arrows) in a segment of inflamed arteriole of mouse cremaster muscle after laser injury observed under intravital fluorescent microscopy.

564

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

shown in F igure 27 for VCAM-targeted PFC nanopar ticles in mdx (muscular dystroph y) mice, a model of Duchenne’s muscular dystrophy.

benefit for estimating local dr ug concentrations and developing ne w phar macokinetic and phar macodynamic paradigms to describe this ne w class of agents.

DRUG DELIVERY

Mechanism of Interaction of PFC Nanoparticles with Cells

The potential dual use of nanopar ticles for both imaging and site-tar geted deli very of therapeutic agents offers g reat promise for indi vidualizing therapeutics. Image-based therapeutics with site-selecti ve agents should enab le conclusi ve assurance that the dr ug is reaching the intended tar get and a molecular ef fect is occurring. In the case of par ticulate agents, ho wever, the mechanisms of drug delivery become more complicated such that ser um concentrations are not necessarily indicati ve of the amount of dr ug accessib le to the desired site. Furthermore, by targeting the carrier to the tissue of interest, the dr ug release is also localized to that area, resulting in a much higher effective drug concentration at the site than is indicated b y ser um levels alone. The ability to quantify the local concentrations these par ticles based on image data could be of g reat

As an e xample of this paradigm for dr ug deli very, Lanza and colleagues 148 reported the delivery of paclitaxel to porcine aor tic smooth muscle cells in culture with nanopar ticles that w ere tar geted to tissue f actor. When paclitax el-loaded nanopar ticles w ere applied to the cells, specif ic binding elicited a substantial reduction in smooth muscle cell proliferation. Nontar geted paclitaxel-loaded par ticles applied to the cells (ie, no binding of nanopar ticles to cells occur red) resulted in normal cell proliferation, indicating that selecti ve targeting may be a requirement for effective drug delivery for these emulsions. Similar beha vior w as demonstrated for doxorubicin-containing particles. The unique mechanism of dr ug delivery for highl y lipophilic agents, such as paclitax el, contained within

Figure 27. Inflammation in diaphragm muscle targeted to vascular cell adhesion molecule with rhodamine-labeled nanoparticles in mdx (muscular dystrophy) mouse.

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

emulsions depends on the close apposition betw een the nanoparticle carrier and the targeted cell membrane and has been described as “contact-f acilitated dr ug deli very.”148 In contrast to liposomal dr ug delivery that typically requires endoc ytosis of entire par ticles follo wed by an endosomal release step to liberate the active compounds, the mechanism of dr ug transpor t for the PFC particles in volves par ticle binding follo wed b y lipid mixing of the phospholipid surf ace components between the emulsion vesicle and the targeted cell membrane (Figure 28).35 This mechanism depends in part on the e xtent and the frequenc y of contact betw een tw o lipidic surf aces and the ability to for m a hemifusion complex between the vesicle and the cell lipid la yers to achieve the lipid mixing. Drugs, especially hydrophobic moieties, which are contained in the lipid monola yer of the nanopar ticle then dif fuse out into the cell lipid membrane along with other components (e g, dr ugs, genes) of the nanopar ticle lipid membrane. Ultimatel y, they are transported to the cytoplasm directly by energy requiring processes that in volve lipid raft-dependent internalization.

565

The rate of lipid e xchange and dr ug delivery can be greatly increased by the application of clinically safe levels of ultrasound ener gy that increase the propensity for hemifusion or enhance contact betw een the nanopar ticles and the tar geted cell membrane b y stimulating these interactions betw een nanopar ticles and cell membranes (Figure 29). 35 Previously, disr uptive ultrasonic methods have been applied to micrometer-sized bubbles to perforate cell membranes b y concussive or streaming forces in the hope of enhancing local deli very of dr ugs, genes, and other therapeutic agents through “temporar y” membrane pores.48,219–221 In contrast, we first reported that noncavitational ultrasound energy can markedly augment delivery of lipophilic substances to selected cell types after molecular targeting, which neither produces nor requires disruption of cell membrane integrity. Ultrasound (mechanical index ~1.9) applied with a con ventional ultrasound imaging system to nanoparticles tar geted αvβ3-integrins on C32 melanoma cancer cells in vitro produced no untoward effects. In only 5 minutes, lipid deli very from nanopar ticles into cell cytoplasm w as dramaticall y augmented. This approach should allow focused delivery of sufficient energy to deep tissue to acti vate dr ug deposition from molecularl y targeted nanopar ticle car riers, w hile avoiding an y potentially har mful bioef fects of the ultrasound per se to the targeted or other sur rounding cells. We fur ther sho wed that the responsible physical mechanisms entail the action of quantif iable “radiation forces” on par ticles that are induced b y the tra veling compressional (ultrasound) waves,222,223 which can increase biophysical interactions of nanoparticles with the targeted cell surface. These methods offer additional mechanisms for f acilitating targeted drug delivery with the application of exogenous, safe ultrasound energy in conjunction with therapeutic-targeted nanoparticles. The success of such “doub le-targeting” strate gies ultimately may allow the use of even lower systemic doses of highly effective agents.

Therapeutic Applications

Figure 28. “Lipid streaming” into the plasma membrane. A high-power image of a cell surface (see inset) shows a bound rhodamine-labeled nanoparticle (red) with adjacent lipid mixing into the plasma membrane of a cell transiently expressing a green cytoplasmic marker. The cellular features that can be observed are the nucleus (dark circular region), cell cytoplasm (green), and plasma membrane (directly adjacent to cell cytoplasm, only small portion is labeled with red lipid from nanoparticle). (Reprinted with permission from Crowder KC et al.35)

The contact facilitated mechanism of targeted drug delivery with the perfluorocarbon nanopar ticles w as f irst demonstrated with PFC par ticles targeted to smooth muscle cells (via TF) delivering antiproliferative therapy (doxorubicin or paclitaxel).148 These studies indicated a potent antiproliferative effect for paclitax el regardless of dr ug loading dose (0.2 or 2.0 mole-%). Yet, when untargeted, the paclitax elladen nanoparticles exerted no ef fect on cell proliferation, indicating the specif icity for cell deli very based on the requirement for a nanopar ticle binding e vent. Similar observations were made for do xorubicin, although due to

566

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 29. Ultrasound energy plus targeting augment lipophilic delivery to C32 melanoma cells. Fluorescent lipid transferred from nanoparticles is green. A, Confocal micrographs under normal conditions for nontargeted (NT) cells show minimal nonspecific internalization (dashed line indicates cell boundary). B, αvβ3-integrin targeted (T) cells demonstrate specific targeting of nanoparticles to cell surface (green dots) and delivery of lipids in cell cytoplasm (diffuse green). C, D, After ultrasound (US) insonification for 5 min, marked enhancement of cytoplasmic lipid delivery (diffuse green) was observed for both nontargeted (C) and targeted (D) cells, with the most dramatic effect for the targeted nanoparticles with insonification (D). The cell components observed are the nucleus (dark circular region), cell cytoplasm (bright interior), and cell membrane (bright borders). Scale bar: 2 µm. (Reprinted with permission from Crowder KC et al.35)

its g reater w ater solubility as compared with paclitax el, some antiproliferative effect also w as noted e ven without selective cell targeting. Several examples of the use of this strategy are offered below. In Vivo Targeted Antiangiogenic Therapy for Atherosclerosis

Angiogenesis of the v asa v asorum is required for the progression of atherosclerosis and perhaps for the production of unstable plaque.174,224–227 Plaque neovasculature serves as both a biomark er of early disease, a harbinger of vulnerable plaque, and a useful entr y point for therapeutic deli very. Moulton and colleagues 174 showed that antiangio genic therap y reduces the rate of plaque growth in the apolipoprotein E-insufficient mouse model of atherosclerosis. Administration of the antiangio genic drug TNP-470, a w ater-soluble analog of fumagillin, at 30 mg/kg e very other da y for 16 w eeks inhibited the plaque growth by 70%. Along these lines, Winter and colleagues 228 used the cholesterol-fed rabbit model of earl y atherosclerosis to determine the ef ficacy of antiangio genic therapy. Rabbits on a high cholesterol diet w ere treated with one of three emulsions: αvβ3-targeted fumagillin-loaded nanopar ticles, αvβ3-targeted nanopar ticles without fumagillin, and nontargeted fumagillin-loaded nanopar ticles. All three nanoparticle types included Gd so that MRI could be performed at treatment to assess the le vel of plaque

angiogenesis as indicated b y molecular imaging of αvβ3-integrin e xpression, w hile simultaneousl y conf irming and quantifying the specif ic binding of dr ug-laden nanoparticles, if tar geted. 1 w eek later , all rabbits w ere reimaged using the diagnostic αvβ3-targeted paramagnetic nanoparticles to evaluate the response of the αvβ3-integrin, and hence the plaque neo vasculature, to fumagillan therapy. The g roup that recei ved the tar geted antiangio genic nanoparticles exhibited a signif icant reduction in both the spatial distribution and the le vel of αvβ3-related signal enhancement, whereas the groups that received no drug or nontargeted drug-carrying nanoparticles exhibited no therapeutic effect (Figure 30). Histolo gy conf irmed that neovasculature w as concentrated in areas of the adv entitia overlying regions of intimal thickening, and that neovasculature was significantly suppressed in those rabbits treated with αvβ3-targeted antiangiogenic nanopar ticles. Fur thermore, the T1-weighted MR image signal enhancement on the da y of initial treatment with the αvβ3-targeted antiangiogenic paramagnetic nanopar ticles predicted the therapeutic responsi veness (or reduced αvβ3-integrin signal) on 1 week follow-up MR imaging. In Vivo Targeted Antiangiogenic Therapy for Cancer

Antiangiogenic agents for cancer have assumed a critical role in the therapeutic ar mamentarium for solid tumors subsequent to the demonstration of the efficacy

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

of Avastin™ for v arious tumor types. 64,229 However, although a number of agents have been tested clinically, either lack of ef ficacy or systemic to xicity has limited their clinical adoption. To de velop an approach that might deliver far more agent locally to intended sites of tumor angiogenesis by cell-selective contact-facilitated mechanisms, w hile simultaneousl y requiring f ar less systemic dr ug le vels, w e ha ve used the αvβ3-targeted fumagillan-loaded nanopar ticles to treat se veral tumor types. Current studies underway indicate marked therapeutic ef ficacy for these tar geted compounds at systemic doses that are se veral 100-fold lo wer than ha ve been tested clinicall y for related antiangio genic compounds (TNP-470) (unpublished observations). Targeted Thrombolytic Therapy

Thrombolysis is a clinical mainsta y for rapid inter vention in patients with acute m yocardial inf arction and strok e. However, serious systemic side ef fects can arise in the form of b leeding, hemor rhagic strok e, aller gic reactions, excitotoxicity, etc. Although a number of modif ications to

αvβ3-Targeted With Drug

567

active thrombol ytic agents ha ve been considered and tested to in an attempt to reduce side ef fects of dr ugs, such as streptokinase or tissue plasminogen activator (tPA), tPA remains the dominant agent in clinical use. We recently developed f ibrin-targeted nanoparticles whose surfaces were modif ied to incor porate streptokinase, a traditional thrombolytic enzyme that acts to convert plasmino gen into its acti ve f ibrinolytic for m, plasmin.230 The streptokinase w as co valently coupled to the PFC nanopar ticles with the use of 1.0 mole% 1,2-dipalmitoyl-sn gl ycero-3-phosphoethanolamine-N4-(p-maleimidophenyl)butyramide (MPB-PE; Avanti Polar Lipids) for coupling N-succinimidyl-S-acetylthioacetate (SA TA)-derivatized streptokinase. F ibrin thrombi in vitro were targeted with the nanoparticles and imaged with high-frequenc y ultrasound to illustrate binding based on enhanced acoustic backscatter from the clot surf aces, w hich per mitted v olumetric estimates. Profile backscatter plots of the detected clot surf ace demonstrated that streptokinase-loaded , f ibrin-targeted nanoparticles in the presence of plasmino gen induced rapid f ibrinolysis in less than 60 minutes with nearl y total loss of fibrin volume (Figure 31). In contrast, streptokinase-loaded or control f ibrin-targeted nanopar ticles

αvβ3-Targeted Without Drug Baseline

60 minutes

Treated (Streptokinase Plasminogen)

Segmentation

50

Treatment

Control (Streptokinase)

10

1 Wk Post Figure 30. Cross-sectional MR images of rabbit abdominal aorta showing segmented region of interest (top), false-colored overlay of percent signal enhancement at time of treatment (middle) and 1 week post treatment (bottom). Similar enhancement was noted at the time of treatment via αvβ3-integrin targeted paramagnetic nanoparticles either with (left) or without fumagillin (right), indicating successful delivery of nanoparticles to the vasa vasorum. 1 week after treatment, molecular imaging following reinjection of αvβ3-integrin targeted paramagnetic nanoparticles (no drug) revealed markedly reduced angiogenesis in animals treated with αvβ3-integrin targeted fumagillin nanoparticles vs. their no drug counterparts. (Reprinted with permission from Winter PM et al.228)

Control (Plasminogen)

Figure 31. Fibrin-targeted nanoparticle thrombolytic therapy. Experimental clot profiles detected with ultrasound after nanoparticle binding for treatment groups at baseline (left column) and after 60 minutes (right column). Middle and bottom rows are control groups; only the group in the top row (treated with streptokinasemodified nanoparticles in plasminogen-enriched buffer) exhibits volume change with time since streptokinase must activate plasminogen to achieve clot lytic activity. (Reprinted with permission from Marsh JN et al.230)

568

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

incubated with buffer alone (ie, no plasminogen available for conversion to plasmin) had no ef fect on clot v olume over the course of the study . Mor phologic changes exhibited in the treated g roup w ere accompanied b y temporal and spatial changes in backscatter from the targeted surf aces indicati ve of the local action of the thrombolytic agent. This ne w acoustic nanopar ticlebased thrombol ytic agent ma y pro ve suitab le for rapid and safe administration and could decrease the time from the ischemic event to reperfusion and lessen the morbidity and mortality of stroke.

CONCLUSION Targeted fluorocarbon ( 19F) nanopar ticles and other small-molecule agents can of fer unique and potentiall y quantitative signatures for molecular MRI and MRS with no competing backg round signal. Multimodal imaging can be perfor med simultaneousl y, and multispectral detection with the use of dif ferent perfluorocarbons is possible. Sensiti ve image and spectral signal detection will benef it from impro vements in both hardw are and software as this nascent application mo ves into the clinical evaluation phase.

REFERENCES 1. Martino R, Malet-Martino M, Gilard V. Fluorine nuclear magnetic resonance, a privileged tool for metabolic studies of fluoropyrimidine drugs. Curr Drug Metab 2000;1:271–303. 2. Wolf W, Presant CA, Waluch V. 19F-MRS studies of fluorinated dr ugs in humans. Adv Drug Deliv Rev 2000;41:55–74. 3. Yu JX, Kodibagkar V, Cui W, Mason RP. 19F: a versatile reporter for non-invasive physiology and phar macology using magnetic resonance. Curr Med Chem 2005;12:818–48. 4. Yu J-X, Cui W, Zhao D , Mason RP . Non-in vasive ph ysiology and pharmacology using 19F magnetic resonance. In: Tressaud A, Haufe G, editors. Fluorine and health: Else vier B .V.; 2008. p. 198–276. 5. Zhao D , Jiang L, Mason RP . Measuring changes in tumor oxygenation. Methods Enzymol 2004;386:378–418. 6. Bohm HJ, Banner D , Bendels S, et al. Fluorine in medicinal chemistry. Chembiochem 2004;5:637–43. 7. Krafft MP . Fluorocarbons and fluorinated amphiphiles in dr ug delivery and biomedical research. Adv Dr ug Deli v Re v 2001;47:209–28. 8. Muller K, Faeh C, Diederich F. Fluorine in phar maceuticals: looking beyond intuition. Science 2007;317:1881–6. 9. Croce MA, F abian TC, P atton JH Jr , et al. P artial liquid v entilation decreases the inflammator y response in the alv eolar environment of trauma patients. J Trauma 1998;45:273–80; Discussion 280–72. 10. Hirschl RB, Croce M, Gore D , et al. Prospecti ve, randomized, controlled pilot study of partial liquid ventilation in adult acute respiratory distress syndrome. Am J Respir Crit Care Med 2002;165:781–7. 11. Lane TA. Perfluorochemical-based artificial oxygen carrying red cell substitutes. Transfus Sci 1995;16:19–31.

12. Brown JJ , Duncan JR, Heik en JP, et al. P erfluoroctylbromide as a gastrointestinal contrast agent for MR imaging: use with and without glucagon. Radiology 1991;181:455–60. 13. Wootton SL, Coley BD, Hilton SV, et al. Value of brominated fluorocarbons for the radio graphic diagnosis of small-bo wel obstr uction: comparison with other contrast agents in rats. AJR Am J Roentgenol 1993;161:409–16. 14. Clark LC Jr , Clark EW, Moore RE, et al. Room temperature-stab le biocompatible fluorocarbon emulsions. Prog Clin Biol Res 1983; 122:169–80. 15. Police AM, Waxman K, Tominaga G. Pulmonary complications after Fluosol administration to patients with life-threatening blood loss. Crit Care Med 1985;13:96–8. 16. Flaim SF. Phar macokinetics and side ef fects of perfluorocarbon-based blood substitutes. Artif Cells Blood Substit Immobil Biotechnol 1994;22:1043–54. 17. Keipert PE, Otto S, Flaim SF, et al. Influence of perflubron emulsion particle size on b lood half-life and febrile response in rats. Artif Cells Blood Substit Immobil Biotechnol 1994;22:1169–74. 18. McGoron AJ, Pratt R, Zhang J , et al. P erfluorocarbon distribution to liver, lung and spleen of emulsions of perfluorotributylamine (FTBA) in pigs and rats and perfluorooctyl bromide (PFOB) in rats and dogs by 19F NMR spectroscop y. Artif Cells Blood Substit Immobil Biotechnol 1994;22:1243–50. 19. Wickline SA, Lanza GM. Molecular imaging, targeted therapeutics, and nanoscience. J Cell Biochem 2002;39:90–7. 20. Wickline SA, Lanza GM. Nanotechnolo gy for molecular imaging and targeted therapy. Circulation 2003;107:1092–5. 21. Lanza GM, Wallace KD, Scott MJ, et al. A novel site-targeted ultrasonic contrast agent with broad biomedical application. Circulation 1996;94:3334–40. 22. Harris TD, Kalogeropoulos S, Nguyen T, et al. Design, synthesis, and evaluation of radiolabeled inte grin alpha v beta 3 receptor antagonists for tumor imaging and radiotherapy. Cancer Biother Radiopharm 2003;18:627–41. 23. Winter PM, Morawski AM, Caruthers SD, et al. Molecular imaging of angiogenesis in earl y-stage atherosclerosis with alpha(v)beta3integrin-targeted nanoparticles. Circulation 2003;108:2270–4. 24. Sadeghi MM, Krassilnik ova S, Zhang J , et al. Detection of injur yinduced v ascular remodeling b y tar geting acti vated alpha vbeta3 integrin in vivo. Circulation 2004;110:84–90. 25. Anderson SA, Rader RK, Westlin WF, et al. Magnetic resonance contrast enhancement of neo vasculature with alpha(v)beta(3)-targeted nanoparticles. Magn Reson Med 2000;44:433–9. 26. Sipkins D A, Cheresh D A, Kazemi MR, et al. Detection of tumor angiogenesis in vivo by alphaVbeta3-targeted magnetic resonance imaging. Nat Med 1998;4:623–6. 27. Lanza GM, Wickline SA. Targeted ultrasonic contrast agents for molecular imaging and therapy. Prog Cardiovasc Dis 2001;44:13–31. 28. Lanza GM, Wickline SA. Targeted ultrasonic contrast agents for molecular imaging and therapy. Curr Probl Cardiol 2003;28:625–53. 29. Hu G, Lijo wski M, Zhang H, et al. Imaging of Vx-2 rabbit tumors with alpha(nu)beta(3)-integrin targeted 111In nanoparticles. Int J Cancer 2007;120:1951–7. 30. Lindner JR. Molecular imaging with contrast ultrasound and targeted microbubbles. J Nucl Cardiol 2004;11:215–21. 31. Demos S, Alkan-Onyuksel H, Kane B , et al. In vi vo tar geting of acoustically reflective liposomes for intra vascular and transv ascular ultrasonic enhancement. J Am Coll Cardiol 1999; 33:867–75. 32. Hamilton AJ, Huang SL, Warnick D, et al. Intra vascular ultrasound molecular imaging of atheroma components in vi vo. J Am Coll Cardiol 2004;43:453–60. 33. Morawski AM, Lanza GM, Wickline S. Targeted contrast agents for magnetic resonance imaging and ultrasound. Cur r Opin Biotechnol 2005;16:89–92.

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

34. Crowder KC, Hughes MS, Marsh JN , et al. Augmented and selective delivery of liquid perfluorocarbon nanopar ticles to melanoma cells with noncavitational ultrasound. Proceedings of IEEE Ultrasonics Symposium; 2003;03CH37476C:532–5. 35. Crowder KC, Hughes MS, Marsh JN, et al. Sonic activation of molecularly-targeted nanopar ticles accelerates transmembrane lipid deli very to cancer cells through contact-mediated mechanisms: implications for enhanced local drug delivery. Ultrasound Med Biol 2005;31:1693–700. 36. Zhou S, Schmelz A, Seuf ferlein T, et al. Molecular mechanisms of low intensity pulsed ultrasound in human skin f ibroblasts. J Biol Chem 2004;279:54463–9. 37. Lanza GM, Trousil RL, Wallace KD, et al. In vitro characterization of a novel, tissue-tar geted ultrasonic contrast system with acoustic microscopy. J Acoust Soc Am 1998;104:3665–72. 38. Hall CS, Marsh JN, Scott MJ, et al. Time evolution of enhanced ultrasonic reflection using a f ibrin-targeted nanopar ticulate contrast agent. J Acoust Soc Am 2000;108:3049–57. 39. Hughes MS, Marsh JN , Zhang H, et al. Characterization of digital waveforms using ther modynamic analogs: detection of contrasttargeted tissue in vivo. IEEE Trans Ultrason Ferroelectr Freq Control 2006;53:1609–16. 40. Hughes MS, Marsh JN , Wallace KD , et al. Sensiti ve ultrasonic detection of dystrophic sk eletal muscle in patients with duchenne muscular dystroph y using an entrop y-based signal receiver. Ultrasound Med Biol 2007;33:1236–43. 41. Hughes MS, McCarthy JE, Marsh JN, et al. Properties of an entropybased signal recei ver with an application to ultrasonic molecular imaging. J Acoust Soc Am 2007;121:3542–57. 42. Mattrey RF. The potential role of perfluorochemicals (PFCs) in diagnostic imaging. Artif Cells Blood Substit Immobil Biotechnol 1994;22:295–313. 43. Lanza GM, Wallace KD, Miller JG, Wickline SA. Development of a novel site targeted ultrasonic contrast agent. In: Nanda N, Schlief R, Goldberg B, editors. Advances in echo imaging using contrast enhancement. Norw ell (MA): Kluw er Academic Pub lishers; 1997. p. 655–67. 44. Lanza GM, Abendschein DR, Hall CS, et al. Molecular imaging of stretch-induced tissue f actor e xpression in carotid ar teries with intravascular ultrasound. Invest Radiol 2000;35:227–34. 45. Lanza GM, Abendschein DR, Hall CS, et al. In vivo molecular imaging of stretch-induced tissue factor in carotid arteries with ligandtargeted nanoparticles. J Am Soc Echocardiogr 2000;13:608–14. 46. Hughes MS, Marsh JN , Hall CS, et al. In vi vo ultrasonic detection of angiogenesis with site-tar geted nanoparticle contrast agents using measure-theoretic signal receivers. Proceedings of the 2004 IEEE Ultrasonic Symposium. 2004. 47. Unger E, Shen D, Fritz T, et al. Gas-filled lipid bilayers as ultrasound contrast. Invest Radiol 1994;29 Suppl 2:S134–6. 48. Unger EC, McCreer y TP, Sw eitzer RH, et al. Acoustically acti ve lipospheres containing paclitax el-a ne w therapeutic ultrasound contrast agent. Invest Radiol 1998;33:886–92. 49. Takeuchi M, Ogunyankin K, Pandian N, et al. Enhanced visualiation of intravascular and left atrial appendage thrombus with the use of a thromb us-targeting ultrasono graphic contrast agent (MRX480A1): in vi vo e xperimental echocardio graphic studies. J Am Soc Echocardiogr 1999;12:1015–21. 50. Tiukinhoy SD , Maho wald ME, Shi vely VP, et al. De velopment of echogenic, plasmid-incor porated, tissue-tar geted cationic liposomes that can be used for directed gene deli very. Invest Radiol 2000;35:732–8. 51. Ferrara KW, Merritt CRB, Burns PN, et al. Evaluation of tumor angiogenesis with US: imaging, doppler , and contrast agents. Acad Radiol 2000;7:824–39. 52. Klibanov AL, Hughes MS, Wojdyla JK, et al. Targeting of ultrasound contrast material: selecti ve imaging of microbubb les in vitro. Acad Radiol 1998;5:S243–6.

569

53. Lindner JR, Song J , Christiansen J , et al. Ultrasound assessment of inflammation and renal tissue injur y with microbubbles targeted to P-selectin. Circulation 2001;104:2107–12. 54. Hughes MS, Marsh JN , Hall CS, et al. Acoustic characterization in whole blood and plasma of site-tar geted nanopar ticle ultrasound contrast agent for molecular imaging. J Acoust Soc Am 2005; 117:964–72. 55. Marsh JN, Partlow KC, Abendschein DR, et al. Molecular imaging with targeted perfluorocarbon nanoparticles: quantification of the concentration dependence of contrast enhancement for binding to sparse cellular epitopes. Ultrasound Med Biol 2007;33:950–8. 56. Hall CS, Marsh JN, Scott MJ, et al. Temperature dependence of ultrasonic enhancement with a site-tar geted contrast agent. J Acoust Soc Am 2001;110:1677–84. 57. Klibanov AL, Rasche PT, Hughes MS, et al. Detection of indi vidual microbubbles of ultrasound contrast agents —imaging of freefloating and targeted bubbles. Invest Radiol 2004;39:187–95. 58. Soman NR, Marsh JN , Hughes MS, et al. Acoustic activation of targeted liquid perfluorocarbon nanopar ticles does not compromise endothelial integrity. IEEE Trans Nanobioscience 2006;5:69–75. 59. Lanza GM, Wallace KD, Fischer SE, et al. High frequency ultrasonic detection of thrombi with a tar geted contrast system. Ultrasound Med Biol 1997;23:863–70. 60. Broze GJJ . Tissue f actor pathw ay inhibitor . Thromb Haemost 1995;74:90–3. 61. Oltrona L, Speidel CM, Recchia D, et al. Prolonged inhibition of tissue f actor-mediated coagulation mark edly attenuates stenosis after balloon-induced ar terial injur y in minipigs. Circulation 1997;96:646–52. 62. Lanza G, Abendschein D, Hall C, et al. In vivo molecular imaging of stretch-induced tissue f actor in carotid ar teries with ligand targeted nanoparticles. J Am Soc Echocardiogr 2000;44:433–9. 63. Folkman J. Tumor angiogenesis: therapeutic implications. N Engl J Med 1971;285:1182–6. 64. Folkman J . Angiogenesis and apoptosis. Semin Cancer Biol 2003;13:159–67. 65. Hughes MS. Analysis of ultrasonic w aveforms using shannon entropy. Proceedings of the 1992 IEEE. Ultrasonics Symposium 1992;92CH3118-7:1205–9. 66. Hughes MS, Marsh JN , Hall CS, et al. Characterization of digital waveforms using ther modynamic analogs: applications to detection of materials defects. IEEE Trans Ultrason F erroelectr F req Control 2005;52:1555–64. 67. Hughes MS. A comparison of shannon entropy versus signal energy for acoustic detection of artificially induced defects in plexiglass. J Acoust Soc Am 1992;91:2272–5. 68. Hughes MS. Analysis of digitized waveforms using shannon entropy. J Acoust Soc Am 1993;91:892–906. 69. Hughes MS. NDE imaging of fla ws using rapid computation of shannon entrop y. Proceedings of the 1993 IEEE. Ultrasonics Symposium 1993;93CH3301-9:697–700. 70. Hughes MS. Analysis of digitized w aveforms using Shannon entropy. II. High-speed algorithms based on Green’ s functions. J Acoust Soc Am 1994;95:2582–8. 71. Arbeit J, Munger K, Ho wley P, Hanahan D . Progressive squamous epithelial neoplasia in K14-human papilloma virus type 16 transgenic mice. J Virol 1994;68:4358–68. 72. Coussens L, Hanahan D, Arbeit J. Genetic pre-disposition and parameters of malignant progression in K14-HPV16 transgenic mice. Am J Pathol 1996;149:1899–917. 73. Smith-McCune K, Zhu YH, Hanahan D, Arbeit J. Crossspecies comparison of angio genesis during the premalignant stages of squamous carcino genesis in the human cer vix and K14-HPV16 transgenic mice. Cancer Res 1997;57:1294–300. 74. Arbeit J, Olson D, Hanahan D. Upregulation of fibroblast growth factors and their receptors during multi-stage epider mal carcinogenesis in K14-HPV16 transgenic mice. Oncogene 1996;13:1847–57.

570

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

75. Marsh JN, Hall CS, Scott MJ , et al. Impro vements in the ultrasonic contrast of tar geted perfluorocarbon nanopar ticles using an acoustic transmission line model. IEEETrans Ultrason Ferroelectr Freq Control 2002;49:29–38. 76. Marsh JN, Crowder KC, Scott MJ, et al. Quantitative acoustic signatures of “tissue f actor” expression by single smooth muscle cells targeted with liquid perfluorocarbon nanopar ticles. Circulation 2003;108:644. 77. Marsh JN, Hail CS, Scott MJ, et al. Kinetic modeling of ultrasonic contrast enhancement b y tar geted agents using acoustic microscopy. J Am Coll Cardiol 2000;35:478A. 78. Morawski AM, Winter PM, Crowder KC, et al. Targeted nanoparticles for quantitati ve imaging of sparse molecular epitopes with MRI. Magn Reson Med 2004;51:480–6. 79. Morawski AM, Winter PM, Yu X, et al. Quantitati ve “magnetic resonance immunohistochemistr y” with ligand-tar geted (19)F nanoparticles. Magn Reson Med 2004;52:1255–62. 80. Winter PM, Cai K, Chen J , et al. Targeted PARACEST nanoparticle contrast agent for the detection of fibrin. Magn Reson Med 2006; 56:1384–8. 81. Flacke S, F ischer S, Scott MJ , et al. No vel MRI contrast agent for molecular imaging of f ibrin: implications for detecting vulnerable plaques. Circulation 2001;104:1280–5. 82. Burnell EE, Cullis PR, de Kr uijff B. Effects of tumbling and lateral diffusion on phosphatidylcholine model membrane 31P-NMR lineshapes. Biochim Biophys Acta 1980;603:63–9. 83. Tilcock C, Ahkong QF, Koenig SH, et al. The design of liposomal paramagnetic MR agents: ef fect of v esicle size upon the relaxi vity of surf ace-incorporated lipophilic chelates. Magn Reson Med 1992;27:44–51. 84. Winter PM, Caruthers SD, Yu X, et al. Improved molecular imaging contrast agent for detection of human thrombus. Magn Reson Med 2003;50:411–6. 85. Winter P, Athey P, Kiefer G, et al. Improved paramagnetic chelate for molecular imaging with MRI. J Magn Magn Mater 2005; 293:540–5. 86. Sherry AD, Cacheris WP, K uan KT . Stability constants for Gd3+ binding to model DTP A-conjugates and DTP A-proteins: implications for their use as magnetic resonance contrast agents. Magn Reson Med 1988;8:180–90. 87. Laurent S, Elst LV, Copoix F, Muller RN. Stability of MRI paramagnetic contrast media: a proton relaxometric protocol for transmetallation assessment. Invest Radiol 2001;36:115–22. 88. Mason RP. Transmembrane pH g radients in vi vo: measurements using fluorinated vitamin B6 deri vatives. Curr Med Chem 1999; 6:481–99. 89. Parhami P, Fung BN. Fluorine-19 relaxation study of perfluorochemicals as oxygen carriers. J Phys Chem 1983;87:1928–31. 90. Thomas SR. The biomedical applications of Fluorine-19 NMR. In: Partain CL, Price RR, P atton JA, et al., editors. Magnetic resonance imaging. Vol 2. London, UK: W.B. Saunders Co.; 1988. p. 1536–52. 91. Kodibagkar VD, Wang X, Mason RP. Physical principles of quantitative nuclear magnetic resonance o ximetry. F ront Biosci 2008;13:1371–84. 92. Shukla HP, Mason RP, Woessner DE, Antich PP. A comparison of three commercial perfluorocarbon emulsions as high f ield NMR probes of oxygen tension and temperature. J Magn Reson B 1995; 106:131–141. 93. Hamza MA, Serratice G, Stebe M-J, Delpuech J-J. Fluorocarbons as oxygen car riers. II. An NMR study of par tially or totall y fluorinated alkanes and alkenes. J Magn Reson 1981;42:227–41. 94. Hamza MA, Ser ratice G, Stebe M-J , Delpuech J-J . Solutesolvent interactions in perfluorocarbon solutions of o xygen. An NMR study. J Am Chem Soc 1981;103:3733–8. 95. Eidelberg D, Johnson G, Barnes D, et al. 19F NMR imaging of blood oxygenation in the brain. Magn Reson Med 1988;6:344–52.

96. Lai C-S, Stair S, Miziorko H, Hyde JS. Effect of oxygen and the spin label TEMPO-Laurate on 19F and proton relaxation rates of the perfluorochemical b lood substitute FC-43 emulsion. J Magn Reson 1984;57:447–52. 97. Thomas SR, Pratt RG, Millard RW, et al. Evaluation of the influence of the aqueous phase bioconstituent en vironment on the F-19 T1 of perfluorocarbon b lood Substitute emulsions. J Magn Reson Imaging 1994;4:631–5. 98. Mason RP, Shukla HP, Antich PP. In vi vo oxygen tension and temperature: simultaneous deter mination using 19F spectroscop y of perfluorocarbon. Magn Reson Med 1993;29:296–302. 99. Babcock EE, Mason RP, Antich PP. Effect of homonuclear J modulation on 19F spin-echo images. Magn Reson Med 1991; 17:178–88. 100. Mason RP, Bansal N , Babcock EE, et al. A novel editing technique for 19F MRI: molecule-specif ic imaging. Magn Reson Imaging 1990;8:729–36. 101. Sotak CH, Hees PS, Huang HN, et al. A new perfluorocarbon for use in fluorine-19 MRI and MRS. Magn Reson Med 1993;29:188–95. 102. Mason RP, Rodbumrung W, Antich PP. Hexafluorobenzene: a sensitive 19F NMR indicator of tumor o xygenation. NMR Biomed 1996;9:125–34. 103. Dardzinski BJ, Sotak CH. Rapid tissue o xygen tension mapping using 19F inversion-recovery echo-planar imaging of Perfluoro15-crown-5-ether. Magn Reson Med 1994;32:88–97. 104. Mason RP, Antich PP, Babcock EE, et al. Non-invasive determination of tumor o xygen tension and local v ariation with g rowth. Int J Radiat Oncol Biol Phys 1994;29:95–103. 105. McIntyre DJO, McCoy CL, Griffiths JR. Tumour oxygenation measurements b y 19F MRI of perfluorocarbons. Cur r Sci 1999; 76:753–62. 106. Berkowitz BA, Wilson CA, Hatchell DL, London RE. Quantitati ve determination of the par tial oxygen pressure in the vitrectomized rabbit e ye In vi vo using 19F NMR. Magn Reson Med 1991; 21:233–41. 107. Duong TQ, Iadecola C, Kim SG. Ef fect of h yperoxia, hypercapnia, and hypoxia on cerebral interstitial o xygen tension and cerebral blood flow. Magn Reson Med 2001;45:61–70. 108. Barker BR, Mason RP, Bansal N, Peshock RM. Oxygen tension mapping by 19F echo planar NMR imaging of sequestered perfluorocarbon. J Magn Reson Imaging 1994;4:595–602. 109. Mason RP, Ran S, Thorpe PE. Quantitative assessment of tumor oxygen dynamics: molecular imaging for prognostic radiology. J Cell Biochem 2002;39 Suppl 87:45–53. 110. Zhao D, Jiang L, Hahn EW, Mason RP. Tumor physiological response to combretastatin A4 phosphate assessed b y MRI. Int J Radiat Oncol Biol Phys 2005;62:872–80. 111. Zhao D , Constantinescu A, Jiang L, et al. Pro gnostic radiolo gy: quantitative assessment of tumor oxygen dynamics by MRI. Am J Clin Oncol 2001;24:462–6. 112. Zhao D, Constantinescu A, Chang C-H, et al. Cor relation of tumor oxygen dynamics with radiation response of the dunning prostate R3327-HI tumor. Radiat Res 2003;159:621–31. 113. Zhao D , Constantinescu A, Hahn EW , Mason RP . Tumor o xygen dynamics with respect to growth and respiratory challenge: investigation of the dunning prostate R3327-HI tumor . Radiat Res 2001;156:510–20. 114. Zhao D, Constantinescu C, Hahn EW, Mason RP. Differential oxygen dynamics in tw o diverse dunning prostate R3327 rat tumor sublines (MAT-Lu and HI) with respect to g rowth and respirator y challenge. Int J Radiat Oncol Biol Ph ys 2002;53:744–56. 115. Zhao D, Ran S, Constantinescu A, et al. Tumor oxygen dynamics: correlation of in vi vo MRI with histolo gical f indings. Neoplasia 2003;5:308–18. 116. Hunjan S, Zhao D , Canstandtinescu A, et al. Tumor o ximetry: demonstration of an enhanced dynamic mapping procedure using fluorine-19 echo planar magnetic resonance imaging the dunning

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

117.

118.

119.

120.

121.

122.

123.

124.

125.

126. 127.

128.

129. 130.

131.

132.

133.

134.

135. 136.

137.

prostate R3327-At1 rat tumor . Int J Radiat Oncol Biol Ph ys 2001;49:1097–108. Kim JG, Zhao D, Constantinescu A, et al. Inter play of tumor vascular oxygenation and tumor pO2 observed using NIRS, oxygen needle electrode, and 19F MR pO 2 mapping. J Biomed Opt 2003; 8:53–62. Le D, Mason RP, Hunjan S, et al. Regional tumor oxygen dynamics: 19F PBSR EPI of he xafluorobenzene. Magn Reson Imaging 1997;15:971–81. Song Y, Constantinescu A, Mason RP. Dynamic breast tumor oximetry: the development of prognostic radiology. Technol Cancer Res Treat 2002;1:471–8. Xia M, Kodibagkar V, Liu H, Mason RP . Tumour oxygen dynamics measured simultaneousl y b y near infrared spectroscop y and 19 F magnetic resonance imaging in rats. Ph ys Med Biol 2006; 51:45–60. Mason RP, Constantinescu A, Hunjan S, et al. Re gional tumor oxygenation and measurement of dynamic changes. Radiat Res 1999; 152:239–49. Mason RP, Hunjan S, Constantinescu A, et al. Tumor oximetry: comparison of 19F MR EPI and electrodes. In: Dunn JF , Swartz HM, editors. Oxygen transpor t to tissue XXIV . Vol 530. Ne w York: Kluwer; 2003. p. 19–28. Gu Y, Bourke V, Kim JG, et al. Dynamic response of breast tumor oxygenation to h yperoxic respirator y challenge monitored with three oxygen-sensitive parameters. Appl Opt 2003;42:1–8. Bourke VA, Zhao D, Gilio J, et al. Cor relation of radiation response with tumor o xygenation in the dunning prostate R3327-A T1 tumor. Int J Radiat Oncol Biol Ph ys 2007;67:1179–86. Deutsch CJ, Taylor JS. 19F NMR measurements of intracellular pH. In: Gupta RK, editor . NMR spectroscop y of cells and or ganisms 1987. Boca Raton (FL): CRC Press; 1987. p. 55–73. Deutsch CJ, Taylor JS. Intracellular pH measured b y 19F NMR. Ann N Y Acad Sci 1987;508:33–47. He S, Mason RP, Hunjan S, et al. Development of novel 19F NMR pH indicators: synthesis and e valuation of a series of fluorinated vitamin B6 analogs. Bioorg Med Chem 1998;6:1631–9. Hunjan S, Mason RP, Mehta VD, et al. Simultaneous intra- and extracellular pH measurement using 19F NMR of 6-Fluorop yridoxol. Magn Reson Med 1998;39:551–6. Korytnyk W, Singh RP. Proton magnetic resonance spectra of compounds in the vitamin B6 group. J Am Chem Soc 1963;85:2813–7. Mehta VD, Kulkarni PV, Mason RP, et al. 6-Fluoropyridoxol: a novel probe of cellular pH using 19F NMR spectroscop y. FEBS Lett 1994;349:234–8. Lutz NW , Hull WE. Assignment and pH dependence of the 19F-NMR resonances from the fluorouracil anabolites in volved in fluoropyrimidine chemotherap y. NMR Biomed 1999; 12:237–48. McSheehy PMJ, Lemaire LP, Griffiths JR. Fluorine-19 MRS: applications in oncology. In: Grant DM, Har ris RK, editors. Enclyclopedia of NMR. Chichester, UK: Wiley; 1996. p. 2048–51. Sijens PE, Baldwin NJ , Ng TC. Multinuclear MR in vestigation of the metabolic response of murine RIF-1 tumor to 5-fluorouracil chemotherapy. Magn Reson Med 1991;19:337–85. Guerquin-Kern J-L, Leteur tre F, Croisy A, Lhoste J-M. pH dependence of 5-fluorouracil uptak e obser ved b y in vi vo 31P and 19F NMR spectroscopy. Cancer Res 1991;51:5770–3. Tsien RY. A non-disr uptive technique for loading calcium buf fers and indicators into cells. Nature 1981;290:527–8. Metcalfe JC, Hesk eth TR, Smith GA. F ree cytosolic Ca 2+ measurements with fluorine labelled indicators using 19F NMR. Cell Calcium 1985;6:183–95. Benters J , Flo gel U, Schafer T, et al. Study of the interactions of cadmium and zinc ions with cellular calcium homoeostasis using F-19-NMR spectroscopy. Biochem J 1997;322:793–9.

571

138. Smith GA, Hesk eth R T, Metcalfe JC, et al. Intracellular calcium measurements b y F-19 Nmr of fluorine-labeled chelators. Proc Natl Acad Sci U S A-Biological Sciences 1983;80:7178–82. 139. Gupta RK, Gillies RJ . 19F NMR measurement of intracellular free calcium ions in intact cells and tissues. In: Gupta RK, editor . NMR spectroscop y of cells and or ganisms. Vol 2. Boca Raton (FL): CRC; 1987. p. 45–53. 140. Procissi D, Claus F, Burgman P, et al. In vi vo F-19 magnetic resonance spectroscop y and chemical shift imaging of tri-fluoro-nitroimidazole as a potential h ypoxia repor ter in solid tumors. Clin Cancer Res 2007;13:3738–47. 141. Robinson SP, Griffiths JR. Current issues in the utility of 19F nuclear magnetic resonance methodologies for the asssessment of tumour hypoxia. Philos Trans R Soc Lond B Biol Sci 2004;359:987–96. 142. Cline JM, Rosner GL, Raleigh JA, Thrall DE. Quantification of CCI103F labeling hetero geneity in canine solid tumors. Int J Radiat Oncol Biol Phys 1997;37:655–62. 143. Workman P, Maxw ell RJ , Grif fiths JR. Nonin vasive MRS in ne w anticancer drug development. NMR Biomed 1992;5:270–2. 144. Seddon BM, Maxwell RJ, Honess DJ, et al. Validation of the fluorinated 2-nitroimidazole SR-4554 as a noninvasive hypoxia marker detected b y magnetic resonance spectroscop y. Clin Cancer Res 2002;8:2323–35. 145. Seddon BM, P ayne GS, Simmons L, et al. A phase I study of SR-4554 via intra venous administration for nonin vasive investigation of tumor h ypoxia by magnetic resonance spectroscop y in patients with malignancy. Clin Cancer Res 2003;9:5101–12. 146. Corban-Wilhelm H, Hull WE, Becker G, et al. Cytosine deaminase and th ymidine kinase gene therap y in a dunning rat prostate tumour model: absence of b ystander effects and characterisation of 5-fluoroc ytosine metabolism with 19F-NMR spectroscop y. Gene Ther 2002;9:1564–75. 147. Liu L, K odibagkar VD, Yu J-X, Mason RP . 19F-NMR detection of lacZ gene e xpression via the enzymic h ydrolysis of 2-fluoro-4nitrophenyl β-D-galactopyranoside in vivo in PC3 prostate tumor xenografts in the mouse. FASEB J 2007;21:2014–9. 148. Lanza GM, Yu X, Winter PM, et al. Targeted antiproliferative drug delivery to v ascular smooth muscle cells with a magnetic resonance imaging nanoparticle contrast agent: implications for rational therapy of restenosis. Circulation 2002;106:2842–7. 149. Caruthers SD, Neubauer AM, Hockett FD, et al. In vitro demonstration using 19F magnetic resonance to augment molecular imaging with paramagnetic perfluorocarbon nanopar ticles at 1.5 Tesla. Invest Radiol 2006;41:305–12. 150. Neubauer AM, Car uthers SD , Hock ett FD , et al. Fluorine cardiovascular magnetic resonance angio graphy in vi vo at 1.5 T with perfluorocarbon nanopar ticle contrast agents. J Cardiovasc Magn Reson 2007;9:565–73. 151. Ahrens ET, Flores R, Xu HY, Morel PA. In vivo imaging platform for tracking immunotherapeutic cells. Nat Biotechnol 2005;23:983–7. 152. Bovey F, Jelinski L, Mirau P . Nuclear magnetic resonance spectroscopy. San Diego (CA): Academic Press; 1988. 153. Doyle M, Mansf ield P. Chemical-shift imaging: a h ybrid approach. Magn Reson Med 1987;5:255–61. 154. Noth U, Jager LJ, Lutz J, Haase A. Fast 19F-NMR imaging in vi vo using FLASH-MRI. Magn Reson Imaging 1994;12:149–53. 155. Ward KM, Aletras AH, Balaban RS. A new class of contrast agents for MRI based on proton chemical exchange dependent saturation transfer (CEST). J Magn Reson 2000;143:79–87. 156. Zhou J, Wilson DA, Sun PZ, et al. Quantitative description of proton exchange processes betw een w ater and endo genous and e xogenous agents for WEX, CEST, and APT experiments. Magn Reson Med 2004;51:945–52. 157. Terreno E, Castelli DD , Cravotto G, et al. Ln(III)-DO TAMGly complexes: a v ersatile series to assess the deter minants of the ef ficacy of paramagnetic chemical e xchange saturation transfer agents for

572

158.

159.

160.

161.

162.

163. 164. 165. 166.

167.

168.

169. 170.

171.

172.

173.

174.

175.

176.

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

magnetic resonance imaging applications. In vest Radiol 2004; 39:235–43. Aime S, Delli Castelli D, Terreno E. Supramolecular adducts between poly-L-arginine and [TmIIIdotp]: a route to sensiti vity-enhanced magnetic resonance imaging-chemical exchange saturation transfer agents. Angew Chem Int Ed Engl 2003;42:4527–9. McMahon MT, Gilad AA, Zhou J, et al. Quantifying e xchange rates in chemical exchange saturation transfer agents using the saturation time and saturation power dependencies of the magnetization transfer effect on the magnetic resonance imaging signal (QUEST and QUESP): Ph calibration for poly-L-lysine and a starburst dendrimer. Magn Reson Med 2006;55:836–47. Aime S, Delli Castelli D, Terreno E. Highly sensitive MRI chemical exchange saturation transfer agents using liposomes. Angew Chem Int Ed Engl 2005;44:5513–5. Adair CR, Zhao P, Pasha A, et al. Synthesis and spectral proper ties of a P ARACEST europium chelate: a bifunctional inter mediate for targeted imaging applications. Inorg Chem [in press]. Ojio S, Takatsu H, Tanaka T, et al. Considerable time from the onset of plaque rupture and/or thrombi until the onset of acute m yocardial infarction in humans: coronar y angiographic findings within 1 w eek before the onset of inf arction. Circulation 2000; 102:2063–9. Goldstein JA. Multifocal coronar y plaque instability . Pro g Cardiovasc Dis 2002;44:449–54. Goldstein JA. Angiographic plaque complexity: the tip of the unstab le plaque iceberg. J Am Coll Cardiol 2002;39:1464–7. Constantinides P. Plaque f issuring in human coronar y thrombosis. J Atheroscler Res 1966;6:1–17. Lanza GM, Lorenz CH, F ischer SE, et al. Enhanced detection of thrombi with a novel fibrin-targeted magnetic resonance imaging agent. Acad Radiol 1998;5:S173–6. Yu X, Song SK, Chen J, et al. High-resolution MRI characterization of human thrombus using a no vel f ibrin-targeted paramagnetic nanoparticle contrast agent. Magn Reson Med 2000;44:867–72. Friedlander M, Brooks PC, Shaf fer RW, et al. Def inition of tw o angiogenic pathw ays b y distinct a v integrins. Science 1995; 270:1500–2. Kerr JS, Mousa SA, Slee AM. Alpha(v)beta(3) integrin in angiogenesis and restenosis. Drug News Perspect 2001;14:143–50. Brooks PC, Montgomery AMP, Rosenfeld M, et al. Inte gin αvβ3 antagonists promote tumor regression by inducing apoptosis of angiogenic blood vessels. Cell 1994;79:1157–64. Bishop GG, McPherson J A, Sanders JM, et al. Selecti ve alpha(v)beta(3)-receptor blockade reduces macrophage infiltration and restenosis after balloon angioplasty in the atherosclerotic rabbit. Circulation 2001;103:1906–11. Corjay MH, Diamond SM, Schlingmann KL, et al. alpha vbeta3, alphavbeta5, and osteopontin are coordinately upregulated at early time points in a rabbit model of neointima for mation. J Cell Biochem 1999;75:492–504. Brooks PC, Stromblad S, Klemke R, et al. Antiintegrin alpha v beta 3 blocks human breast cancer growth and angiogenesis in human skin. J Clin Invest 1995;96:1815–22. Moulton KS, Heller E, Konerding MA, et al. Angiogenesis inhibitors endostatin or TNP-470 reduce intimal neo vascularization and plaque g rowth in apolipoprotien E-def ficient mice. Circulation 1999;99:1653–5. Tenaglia AN, Peters KG, Sketch MH Jr, Annex BH. Neovascularization in atherectomy specimens from patients with unstable angina: implications for patho genesis of unstab le angina. Am Hear t J 1998;135:10–4. Wilson SH, Her mann J, Ler mann LO, et al. Simv astatin preser ves the structure of coronar y adventitial vasa vasorum in experimental hypercholesterolemia independent of lipid lo wering. Circulation 2002;105:415–8.

177. Zhang Y, Clif f WJ, Schoefl GI, Higgins G. Immunohistochemical study of intimal micro vessels in coronar y atherosclerosis. Am J Pathol 1993;143:164–72. 178. Cyrus T , Abendschein DR, Car uthers SD , et al. MR threedimensional molecular imaging of intramural biomark ers with targeted nanoparticles. J Cardiovasc Magn Reson 2006;8:1–7. 179. Schmieder A, Winter PM, Caruthers SD, et al. Molecular MR imaging of melanoma angio genesis with αvβ3-targeted paramagnetic nanoparticles. Magn Reson Med 2005;53:621–7. 180. Winter PM, Car uthers SD, Kassner A, et al. Molecular imaging of angiogenesis in nascent Vx-2 rabbit tumors using a novel αvβ3-targeted nanopar ticle and 1.5 Tesla MRI. Cancer Res 2003; 63:5838–43. 181. Partlow KC, Chen J, Brant JA, et al. 19F magnetic resonance imaging for stem/progenitor cell tracking with multiple unique perfluorocarbon nanobeacons. FASEB J 2007;21:1647–54. 182. Yeh TC, Zhang W, Ildstad ST, Ho C. Intracellular labeling of T-cells with super paramagnetic contrast agents. Magn Reson Med 1993;30:617–25. 183. Ahrens ET, Feili-Hariri M, Xu H, et al. Receptor -mediated endocytosis of iron-o xide par ticles pro vides ef ficient labeling of dendritic cells for in vi vo MR imaging. Magn Reson Med 2003; 49:1006–13. 184. Bulte JW , Kraitchman DL. Monitoring cell therap y using iron oxide MR contrast agents. Cur r Phar m Biotechnol 2004; 5:567–84. 185. Kircher MF, Allport JR, Gra ves EE, et al. In vi vo high resolution three-dimensional imaging of antigen-specif ic c ytotoxic T-lymphocyte trafficking to tumors. Cancer Res 2003;63:6838–46. 186. Jacobs RE, Fraser SE. Magnetic resonance microscopy of embryonic cell lineages and movements. Science 1994;263:681–4. 187. Modo M, Mellodew K, Cash D, et al. Mapping transplanted stem cell migration after a stroke: a serial, in vivo magnetic resonance imaging study. Neuroimage 2004;21:311–7. 188. Morawski AM, Winter PM, Car uthers SD , et al. Magnetic resonance immunoc ytochemistry: characterization of “T issue factor” e xpression b y smooth muscle cells with tar geted paramagnetic nanoparticles. Circulation 2003;108:139. 189. Gu Y, Mason RP, Liu H. Estimated fraction of tumor v ascular blood contents sampled by near infrared spectroscopy and 19F magnetic resonance spectroscopy. Opt Express 2005;13:1724–33. 190. Heidelberger C. Fluorinated p yrimidines, a ne w class of tumour inhibitory compounds. Nat Chem Biol 1957;179:663. 191. Isanbor C, O’Hagan D. Fluorine in medicinal chemistry: a review of anti-cancer agents. J Fluor Chem 2006;127:303–19. 192. Sogabe T, Imaizumi T, Mori T, et al. Ef fects of v asodilators on the signal intensity of perfluorocarbon monitored b y in vi vo F-19-NMR spectroscopy. Magn Reson Imaging 1997;15:341–5. 193. van Laarhoven HWM, Punt CJ A, Kamm YJL, Heerschap A. Monitoring fluoropyrimidine metabolism in solid tumors with in vi vo 19F magnetic resonance spectroscop y. Crit Re v Oncol Hematol 2005;56:321–43. 194. Presant CA, Wolf W, Waluch V, et al. Association of intratumoral pharmacokinetics of fluorouracil with clinical response. Lancet 1994;343:1184–7. 195. Bartels M, Albert K. Detection of psychoactive drugs using 19F MR spectroscopy. J Neural Transm Gen Sect 1995;99:1–6. 196. Passe TJ, Charles HC, Rajagopalan P, Krishnan KR. Nuclear magnetic resonance spectroscop y: a re view of neuropsychiatric applications. Nuclear magnetic resonance spectroscop y: a review of neuropsychiatric applications. Prog Neuropsychopharmacol Biol Psychiatry 1995;19:541–63. 197. Bolo N R, H ode Y, Macher JP . Long-ter m sequestration of fluorinated compounds in tissues after fluvoxamine or fluoxetine treatment: a fluorine magnetic resonance spectroscop y study in vivo. MAGMA 2004;16:268–76.

Fluorocarbon Agents for Quantitative Multimodal Molecular Ima ging and Targeted Therapeutics

198. Henry ME, Schmidt ME, Hennen J, et al. A comparison of brain and serum pharmacokinetics of R-fluoxetine and racemic fluoxetine: a 19-F MRS study. Neuropsychopharmacol 2005;30:1576–83. 199. Strauss WL, Unis AS, Co wan C, et al. Fluorine magnetic resonance spectroscopy measurement of brain fluvoxamine and fluoxetine in pediatric patients treated for pervasive developmental disorders. Am J Psychiatry 2002;159:755–60. 200. Takeda T, Makita K, Ishika wa S, et al. Uptak e and elimination of sevoflurane in rabbit tissues—an in vivo magnetic resonance spectroscopy study. Can J Anaesth 2000;47:579–84. 201. Wyrwicz AM, Conbo y CB . Multiecho 19F NMR imaging of halothane in rabbit brain. Proceedings 7th Society for Magnetic Resonance in Medicine; San Francisco: 1988. 202. Xu Y, Tang P, Zhang WG, et al. F-19 Nuclear -magnetic-resonance imaging and spectroscopy of sevoflurane uptake, distribution, and elimination in rat-brain. Anesthesiology 1995;83:766–74. 203. Freytag SO, Khil M, Strick er H, et al. Phase I study of replicationcompetent adenovirus-mediated double suicide gene therap y for the treatment of locall y recur rent prostate cancer . Cancer Res 2002;62:4968–76. 204. Stegman LD, Rehemtulla A, Beattie B, et al. Nonin vasive quantitation of c ytosine deaminase transgene e xpression in human tumor x enografts with in vi vo magnetic resonance spectroscopy. Proc Natl Acad Sci U S A 1999;96:9821–6. 205. Fairweather JK, Faijes M, Driguez H, Planas A. Specificity studies of bacillus 1,3-1,4-beta-glucanases and application to glycosynthasecatalyzed transglycosylation. Chembiochem 2002;3:866–73. 206. Millett F , Rafter y MA. Fluorine-19 nuclear magnetic resonance study of the binding of trifluoroacetylglucosamine oligomers to lysozyme. Biochemistry 1972;11:1639–43. 207. Withers SG, Street IP , P ercival MD . Fluorinated carboh ydrates as probes of enzyme specif icity and mechanism. ACS Symp Ser 1988;374:59–77. 208. Yu JX, Otten P , Ma Z, et al. A novel NMR platfor m for detecting gene transfection: synthesis and e valuation of fluorinated Phen yl β-D-Galactosides with potential application for assessing LacZ Gene expression. Bioconjug Chem 2004;15:1334–41. 209. Cui W, Otten P, Li Y, et al. A novel NMR approach to assessing gene transfection: 4-fluoro-2-nitrophen yl-β-D-galactopyranoside as a prototype repor ter molecule for b-galactosidase. Magn Reson Med 2004;51:616–20. 210. Kodibagkar VD, Yu J, Liu L, et al. Imaging b-galactosidase acti vity using 19F chemical shift imaging of LacZ gene-repor ter molecule 2-fluoro-4-nitrophenol-β-D-galactopyranoside. Magn Reson Imaging 2006;24:959–62. 211. Yu JX, Kodibagkar VD, Liu L, Mason RP. A 19F NMR approach using repor ter molecule pairs to assess β-Galactosidase in human x enograft tumors in vi vo. NMR Biomed 2008. [in press] 212. Yu JX, Liu L, K odibagkar VD, et al. Synthesis and e valuation of novel enhanced gene repor ter molecules: detection of β-Galactosidase acti vity using 19F NMR of Trifluoromethylated Aryl β-D-Galactopyranosides. Bioorg Med Chem 2006;14:326–33.

573

213. Yu JX, Ma Z, Li Y, et al. Synthesis and e valuation of a no vel gene reporter molecule: detection of β-galactosidase activity using 19F NMR of a fluorinated vitamin B6 conjugate. Med Chem 2005;1:255–62. 214. Look DC, Locker DR. Time saving in measurement of NMR and EPR relaxation times. Rev Sci Instrum 1970;41:250–1. 215. Hinshaw WS, Lent AH. An introduction to NMR imaging: from the bloch equation to the imaging equation. Proceedings of the 1983 IEEE 1983;71:338–54. 216. Haake EM, Brown RW, Thompson MR, Venkatesan R. Magnetic resonance imaging: ph ysical principles and sequence design. Ne w York: John Wiley & Sons; 1999. 217. Ahrens ET, Rothbacher U, Jacobs RE, F raser SE. A model for MRI contrast enhancement using T1 agents. Proc Natl Acad Sci U S A 1998;5:8443–8. 218. Robertson JD, Crane SB, Wickline SA, Lanza GM. Characterization and biodistribution of a no vel MRI molecular imaging agent b y neutron acti vation anal ysis. J Radioanal Nucl Chem 2005; 263:511–4. 219. Guzman HR, Nguyen DX, Khan S, Prausnitz MR. Ultrasound-mediated disruption of cell membranes. I. Quantif ication of molecular uptake and cell viability. J Acoust Soc Am 2001;110:588–96. 220. Taniyama Y, Tachibana K, Hiraoka K, et al. Local deli very of plasmid DN A into rat carotid ar tery using ultrasound. Circulation 2002;105:1233–9. 221. Deng CX, Sieling F, Pan H, Cui J. Ultrasound-induced cell membrane porosity. Ultrasound Med Biol 2004;30:519–26. 222. Dayton P, Klibanov A, Brandenburger G, Ferrara K. Acoustic radiation force in vi vo: a mechanism to assist tar geting of microbubbles. Ultrasound Med Biol 1999;25:1195–201. 223. Ter Haar G, Wyard SJ. Blood cell banding in ultrasonic standing wave fields: a physical analysis. Ultrasound Med Biol 1978;4:111–23. 224. McCarthy MJ, Loftus IM, Thompson MM, et al. Angiogenesis and the atherosclerotic carotid plaque: an association betw een symptomatology and plaque mor phology. J Vasc Surg 1999;30:261–8. 225. O’Brian ER, Garvin MR, Dev R, et al. Angiogenesis on human coronary atherosclerotic plaque. Am J Pathol 1994;145:883–94. 226. Moreno PR, Pur ushothaman KR, Sirol M, et al. Neo vascularization in human atherosclerosis. Circulation 2006;113:2245–52. 227. Jain RK, F inn AV, K olodgie FD, et al. Antiangiogenic therap y for normalization of atherosclerotic plaque v asculature: a potential strategy for plaque stabilization. Nat Clin Pract Cardio vasc Med 2007;4:491–502. 228. Winter PM, Neubauer AM, Car uthers SD , et al. Endothelial alpha(v)beta3 inte grin-targeted fumagillin nanopar ticles inhibit angiogenesis in atherosclerosis. Arterioscler Thromb Vasc Biol 2006;26:2103–9. 229. Ferrara N, Hillan KJ, Gerber HP, Novotny W. Discovery and development of be vacizumab, an anti-VEGF antibody for treating cancer. Nat Rev Drug Discov 2004;3:391–400. 230. Marsh JN, Senpan A, Hu G, et al. F ibrin-targeted perfluorocarbon nanoparticles for tar geted thrombol ysis. Nanomedicine 2007; 2:533–43.

36 APTAMERS

FOR

MOLECULAR IMAGING

BERTRAND TAVITIAN, MD, PHD

Molecular imaging is lar gely based on in vi vo administration of molecular probes that selectively recognize molecular targets of interest. Increasing the power of molecular imaging for medical diagnosis therefore requires development of molecular probes with e xquisite targeting proper ties. Most molecular probes are deri vatives of enzyme substrates or receptor ligands of a small molecular size (< 1,000 Da) that bind to protein pock ets that naturally fit to accommodate their biolo gical ligand. In contrast, the portfolio of molecular probes that reco gnize proteins other than enzymes or receptors, or other parts of tar get proteins than their natural ligand binding pockets, or nonprotein tar gets, is still limited. Cur rent efforts are being made to develop molecular probes capable of binding to sites for w hich there are no kno wn or available natural binders of small molecular w eight. These ne w imaging probes are often macromolecular polymers (peptides, proteins, nucleic acids) that can be tailored to reco gnize an y molecular tar get of interest. Whether they are based on amino acid ornucleotide polymers, these macromolecular “aperiodic cr ystals”1 adopt spatial confor mations that constitute the basis for tar get recognition. Macromolecular imaging probes are f abricated by artificial adaptations of physiological processes for the synthesis of recognition macromolecules. Molecular probes based on antibodies mak e use of the capacity of immune competent cells to produce pol ymers of amino acids capab le of reco gnizing an antigen epitope (the antigenic determinant, ie, the part of the antigen that is recognized by the antibody). Nucleic acids can for m a m yriad of threedimensional str uctures, some of w hich possess a catalytic activity or interact with proteins or other par tners2 (Figure 1). This property is the basis for the implication of nucleic acids in the regulation of cellular pathways and stands in contrast to their role in the storage and the

574

transfer of genetic infor mation, w hich is based on hybridization of complementar y sequences. In 1990, combinatorial approaches w ere developed to create ar tificial nucleic acids binding to specific targets on a structural basis. 3–10 The nucleic acid-based ligands found b y this technique w ere ter med aptamers, from the Latin “aptus,” meaning “to f it,”9 and the method w as named Systematic Evolution of Ligands by Exponential Enrichment (SELEX).10 The manner in which an aptamer binds depends on the nature of its tar get. As an e xample, the crystal structure of an aptamer directed against malachite green (4-[(4-dimethylaminophenyl)-phenyl-methyl]-N, N-dimethyl-aniline; MG), a to xic chemical primaril y used to dye materials such as silk, leather, and paper,11 in the presence of MG or high-affinity analogs of MG w as studied by 2.8 Å X-ray12 and nuclear magnetic resonance (NMR) spectroscopy.13,14 Interestingly, not only does the aptamer str ucture its ligand binding pock et in the presence of the ligand (“adapti ve folding”) but the ligand itself also under goes confor mational changes (“induced fit”) that can e ven confer intrinsic catal ytic properties to the aptamer.15 Due to their unique recognition capacities, aptamers rival antibodies in ter ms of af finity and specif icity for their targets and have entered the fields of medical diagnosis and therap y (re viewed in the studies b y Cerchia and colleagues 16 and Brody and Gold 17). Their application to molecular imaging is an exciting emerging field.

PRINCIPLES OF APTAMER SELECTION Although various names ha ve been gi ven to the method of aptamer selection, that is, in vitro genetics, 4 SELEX,10 or directed molecular e volution,18 the underlying principles illustrated in F igure 2 are the same. They have been described on many occasions in the literature. 19,20

Aptamers for Molecular Imaging

A

B

D

C

E

G

575

F

H

Figure 1. Examples of elementary 3D structures that short nucleic acid sequences can fold into. Intrachain interactions can form GCAA (A) and GAGA (B) tetraloops, hairpins (C), and pseudoknots (D); intra and interchain interactions can form CA loops (E), GU wobbles (F), a variety of loop-loop interactions (G), and G-quartets (H).

Figure 2. General principle for the selection of aptamers by SELEX. N designates any one of the four nucleobases of DNA (A, T, C, G, ie, Adenine, Thymine, Guanine, Cytosine) or RNA (A, U, G, C, ie, Adenine, Uracil, Guanine, Cytosine) incorporated randomly during the synthesis of the oligonucleotide library. See text for description of SELEX.

The initial step consists in the production, by chemical synthesis, of a population of oligonucleotides including a random sequence obtained b y the introduction, with the same probability, of an adenine, a th ymine (or uracil for RNA), a guanine, or a c ytosine in each position of the sequence. Since there are 4 n = 10 nlog(4) ≈ 100.6n different sequences in a population containing a random sequence of n nucleotides, the theoretical diversity in a usual SELEX library with a random stretch of 40nucleotides is 440 ≈ 1024

individual sequences. This represents a huge mass of material, and in practice, the comple xity of a w orkable initial pool of sequences engaged in SELEX is in the order of 10 13 to 10 15 different sequences. The random sequence is flanked on both sides by two constant sequences that are used as primers during the enzymatic amplif ication steps making up the selection process. This random population is then subjected to a selection criterion such as the af finity for a gi ven target. Oligonucleotide sequences that pass successfull y through the selection process are separated , from those that do not, by various separation methods, for example, membrane filtration, immunoprecipitation, a finity chromatography, gel or capillary electrophoresis, gel retardation, magnetic beads, centrifug ation, surf ace plasmon resonance, flo w c ytometry, etc (re viewed in a study b y Gopinath21). These “winning” sequences are then replicated enzymaticall y to for m a daughter population that will undergo the next selection cycle. The separation and amplif ication steps are repeated several times (usuall y 4 to 20 rounds), and a Darwinian evolution of the library leads to preferential amplification of the sequences that best f it the selection pressure. Finally, aptamers are cloned , sequenced , and tested for binding or catal ytic acti vity. Usuall y, conser ved predictive structures or sequences are found in dif ferent clones that ma y be g rouped into se veral classes, in w hich the covariation and conser vation of cer tain bases are indicative of functional minimum domains, and can be

576

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

exploited to deduce the critical str uctural motifs. However, this type of approach assumes a unique relation between the sequence and the structure of an aptamer and neglects the str uctural pleiomor phism of oligonucleotides, that is, their ability to adopt both intramolecular and inter molecular confor mations in the presence of their target. SELEX can use either DN A or RN A. In that latter case, a re verse-transcription step must be included before the complementar y DN A thereb y obtained can be amplified y pol ymerase chain reaction (PCR), and the amplified D A is subsequently transcribed into RNA to be used for a new round of selection. Most bacterial polymerases used in the amplificatio step naturally generate a high rate of mutation, approximately one mutation for every 2 × 104 nucleotides incorporated. Although this is detrimental to applications for which the f idelity of sequence replication is essential, such as the detection of patholo gies by PCR, mutations generated by the polymerases play a key role in the Darwinian e volution of the sequence population during SELEX, b y generating di versity among the selected aptamers, thereby allowing the evolution of aptamers that were not present in the original pool. In some cases, in vitro evolution is sought by using conditions in which the fidelity of the po ymerase is intentionally reduced. The repetitive nature and the comple xity of the selection procedure make it difficult for a single person to carry out several parallel selections manuall y and ha ve encouraged v arious teams to automate the selection process. Since 1998, automated in vitro selection processes for high-rate selection of aptamers have been described,22 and several biotechnolo gy companies using aptamers ha ve been set up. 17 Although all SELEX selection methods are based on the same principle, it is impor tant to stress that the conditions during selection affect the final result: • Selection pressure, for e xample, increasing the stringency of washing conditions during selection, changing ion concentration, reducing incubation time, lowering the concentration of the oligonucleotide library or of the tar get, e ven changing the selection system, or adding non-amplifia le competitors. • Elimination of ar tifacts, such as sequences selected for their ability to bind on the filter or chromato graphic surf ace, b y including suitab le counter selection steps. Hence, candidates are often sie ved on a pre-chromato graph without tar get, and onl y those with no af finity for the chromato graph surf ace are

then used for selection. Other authors also use a specific elution y vir tue of the free tar get. Another type of specific elution consists in unlocking th aptamers from their target by adding the natural ligand of the rele vant target to the reaction medium. In the end, the best solution for a voiding artifacts may be to alternate different selection methods. Accordingly, the SELEX technolo gy can be readil y modified to orient selection of aptamers designed for a specific application. In “b lended SELEX,” high-affinity aptamer binders are obtained b y attaching covalently the oligonucleotide library to a weak inhibitor for an enzyme. The aptamer moiety stabilizes contacts of the inhibitor with the enzyme and impro ves its inhibition ef ficacy.23 “Toggle SELEX” alternates selection against two closely related target molecules to create aptamer ligands able to bind to both tar gets, an approach recentl y repor ted for selection of RNA aptamers that bound to both human and porcine thrombin.24 It has been sho wn that RN A molecules can for m surfaces that functionall y mimic those of proteins. A strategy based on the anti-idiotype approach uses antibodies directed against interfaces of protein–protein interactions to isolate aptamer mimics of one of the interaction domains. This strate gy w as used to isolate aptamers binding to a neutralizing antibody directed against the insulin receptor.25 An antibody raised against the HIV-Rev-nuclear export signal (NES) has been used to select “e xport aptamers” mimicking NES. These export aptamers bound to the NES-receptor in vitro and inhibited Rev-dependent e xport in vi vo.26,27 The selection has been impro ved fur ther by combining the antiidiotype approach with a strate gy related to the “blended SELEX” to isolate aptamers mimicking the substrate of the mito gen- and stress-acti vated protein kinase MSK1. The “bifunctional” aptamer obtained was a substrate mimic that bound to the kinase and inhibited specifically kinase activity in vitro. Aptamers bind to a small epitope of their tar get and binding is exquisitely dependent on the confor mation of that epitope. Therefore, it is essential to ensure during selection that the tar get is in a confor mation as close as possible to the one that it will adopt in the conditions intended for the use of the aptamer. Because many pharmaceutical tar gets are e xtracellular domains of membrane proteins, se veral g roups applied SELEX to isolated cell membranes or li ve or ganisms including viruses, parasites, and mammalian cells. 28,29 Pestourie and colleagues 30 compared dif ferent strate gies to select

Aptamers for Molecular Imaging

aptamers for the transmembrane protein tyrosine kinase receptor RET. It was shown that the aptamers selected in vitro against the recombinant e xtracellular domain of RET did not bind the functionally active protein inserted into the cell membrane. To achie ve reco gnition of the active RET protein, selection had to be conducted against a confor mationally rele vant for m of RET correctly inser ted in the cell membrane. 33 Particularly interesting are SELEX protocols aiming at the selection of aptamers capab le of reco gnizing cell surf ace molecules that differ between two states of cell ph ysiology. Blank and colleagues 31 thus obtained aptamers that targeted a protein e xpressed during the proliferation of endothelial cells of rat brain tumor micro vessels.32 Cerchia and colleagues e xpressed a mutated for m of the oncogenic for m of RET in rat pheochromoc ytoma cells and selected aptamers that specif ically recognized (and, additionall y, neutralized the acti vity of ) the mutant RET.32 Interestingly, when performed on complex systems such as live cells, the subtracti ve procedure of SELEX yields a series of aptamers in w hich each indi vidual aptamer recognizes selectively any one of the molecular differences between the counter-selection and selection conditions. In principle, this can be e xploited to recognize epitopes differing between the cells used for the selection step, for instance, cells e xpressing a particular onco gene and the cells used for counter selection that do not e xpress that onco gene, thereb y yielding aptamers that def ine the differential pattern of expression of cell surf ace proteins induced b y the oncogene.30 Somalogic (Boulder, Colorado) has de veloped a methodology called “photochemical SELEX” in which photoacti vatable nucleotides are incor porated into the aptamer librar y. Upon UV ir radiation, aptamers are co valently photocrosslinked to the tar get protein. This method has yielded aptamers capab le of recognizing recombinant basic fibroblast growth factor, a potent inducer of neo vascularization, and to distinguish it from the related v ascular endothelial growth factor (VEGF) and platelet-derived growth factor (PDGF). Its development is promoted by Somalogic as a generic tool to identify protein signatures of disease states in the f ield of biomark er disco very and assessment.33 Finally, it should be noted that SELEX allows the fabrication of aptamers against molecular tar gets that are too toxic for antibody production or that are not immunogenic. Originally, man y e xperiments using the SELEX methodology w ere perfor med to study natural interactions between RNA structures and their partners

577

(for review see Hermann and Patel34). SELEX has also been used to e xplore the catal ytic mechanisms of ribozymes, that is, natural RN As endo wed with an enzymatic activity, and to design ribozymes with novel catalytic proper ties (for re view see the studies b y Breaker35 and K umar and Ellington 36). Hundreds of aptamers have been selected against a wide v ariety of targets, from small molecules to proteins and e ven whole organisms, and man y have been introduced into biotechnological applications such as biosensors, regulators of gene e xpression, as w ell as for tar get validation and small-molecule dr ug identif ication in high-throughput systems (for review see the studies b y Rimmele37 and Burgstaller and colleagues 38). Aptamers exhibit high affinity for their targets with Kd values (the dissociation constant that characterizes the strength of the interaction betw een a ligand and its target) in the lo w nanomolar to picomolar range. Besides antibodies, they are the only other class of molecules from which specific binding molecules against a variety of targets can be isolated (reviewed in the study by Brody and Gold17). They usually retain their binding capacity after immobilization 39 or after labeling with various functional g roups.40 An interesting proper ty of aptamers that stands in contrast with proteins is their high ther mal stability, which allows the use of drastic chemical conditions for labeling or other modifications of aptamers. They can be delivered into cells by generic methods for nucleic acid transfection, or , when in natural DNA or RNA chemistry, expressed within cells, 41 and delivered into animals. 42 Finally, no immune reactions after aptamer administration to animals or humans have been observed so far, suggesting that they have no or v ery little immunogenicity.43 A comparison of some key properties of aptamers and antibodies that are relevant to their potential use as imaging probes is given in Table 1. Many aptamers were found to possess a neutralizing activity against their tar get, and se veral ha ve been

Table 1. GENERAL PROPERTIES OF APTAMERS RELEVANT TO THEIR USE AS IMAGING PROBES COMPARED TO THAT OF ANTIBODIES Aptamers

Antibodies

Size

9–15 kDa

150 kDa

Discrimination

+++

+/++

Affinity

0.5–100 nM

0.05–100 nM

Source

synthetic

animal

Toxicity

no evidence

+/++

578

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

developed as dr ugs against major phar macological targets.42 Aptamer NX 1838 (now referenced as MacugenTM or pegaptanib sodium) inhibits the VEGF, a key factor of angiogenesis,43 and w as the f irst to obtain an FD A approval for the treatment of the w et form of age-related macular degeneration (AMD), a major cause of blindness. NX 1838 also demonstrated promising anti-angio genesis effects in glomer ular disease in rats, 43 in mice models of Wilms tumor growth,45 and in a mouse xenograft model of neuroblastoma.46 As an e xample of the manner in w hich aptamers have been translated from the molecular biology laboratory into the clinics, NX 1838 was originally developed by NeXstar Phar maceuticals group and Gilead Sciences47 and after ef forts from EYEtech in collaboration with Pf izer reached the US mark et in Januar y 2005. The chemistry of NX 1839 w as e xtensively modif ied for its clinical use: (i) its size w as reduced to 29 nucleotides; (ii) 12 purines of 14 w ere changed into 2'O-Me purines to impro ve nuclease resistance; and (iii) its e xtremity w as conjugated to a 40 kDa pol yethylene glycol (PEG) to enhance its plasmatic residence time. 44 Other aptamers that are under clinical testing include ARC1779, an antithrombotic tar geting the v onWillebrand factor,48 and RB006, an anticoagulant tar geting factor IXa.49 RB006 uses an original antidote concept to reverse the anticoagulation acti vity of the aptamer: the antisense sequence binds to the aptamer sequence b y nucleobase hybridization and induces loss of the 3D structure of the aptamer , which becomes incapable of binding its target protein.49 It is note worthy that most aptamers de veloped for therapy are either administered locall y (Macugen™; the anti-TGFb2 ARC81) 50 or tar get intra vascular proteins (ARC1779 and RB006). This reflects the deli very limitations that aptamers encounter in vi vo, a consequence of their nucleic acid nature that will be discussed belo w.

CHEMICAL MODIFICATIONS One of the limitations in the use of aptamers in animal models of disease and in humans is the reduced stability of natural nucleic acids in biolo gical media. Major efforts have therefore been directed to ward improving the stability of aptamers b y a v ariety of approaches. Endogenous nucleases are present in the b lood and tissues of mammals, and the plasmatic half-life of single stranded DNA and RNA injected intravenously (IV) are on the order of minutes and seconds, respecti vely. For aptamers to be of practical use in vi vo, the resistance of aptamers to nucleasic de gradation must therefore be signif icantly augmented. Chemical

modifications of the ʹ′ position of ribose b y replacement of the hydroxyl g roup with a fluorine, an amine or an O-methyl, increase the stability of oligoribonucleotides. These modif ications can be done b y “post-SELEX,” substitution of nucleotides with the corresponding 2'-fluoro, 2'-amino or 2'-O-alk yl v ariants.17,51,52 However, modified nucleotides induce steric variations of the natural nucleic acid motif and change the folding of oligonucleotides, w hich may lead to the loss of the binding properties of the aptamer. Because it is impossible to predict the consequences of chemical modifications introduced after selection, the e fect of any nucleotide change on the aptamer proper ties must be carefully controlled, a time-consuming and tedious task. To circumv ent this limitation, SELEX has been performed directl y in the presence of 2'-modif ied nucleotides that are accepted for enzymatic polymerization by T7 RNA polymerase (T7pol). Much research has been conducted to study w hich modifications of the 2' position are compatib le with T7pol activity, and mutations of T7pol have been obtained that allo w the incor poration of 2ʹ′-fluoro and ʹ′-amino ribonucleotides, 53,54 and more recently, 2ʹ′-OMe.55,56 Using mutant T7pol, modified 2ʹ′aminopyrimidine or 2 ʹ′-fluoro yrimidine nucleotides can be incor porated directl y in the star ting pools and during the subsequent rounds of SELEX. An alternative approach to replace the h ydroxyl at 2 ʹ′ on the ribose is to modify the backbone of the nucleic acids.57 However, the chemical g roups used are rarel y inte grated b y the enzymes required for SELEX, with the notab le exception of the phosphorothioate nucleotides and boranophosphate nucleotides, w here the nonbridging oxygen of the phosphodiester g roup is replaced with a sulfur and a borane (BH3) g roup. Modifications a fecting the bases are less common although a successful selection using 5-(1-pentyn yl)-2ʹ′-deoxyuridine has been repor ted.58 Another strate gy for obtaining nuclease-resistant aptamers e xploits the chirality of biomolecules: when an aptamer is selected against the non-natural enantiomer of a target, such as a peptide or protein motif built from amino acids in the L conf iguration, the enantiomer of this aptamer (a L-DN A or LRNA) will reco gnize the natural tar get. The advantage of these enantiomeric aptamers, called Spiegelmers® in their for mat e xploited b y No xxon Phar ma AG, is that they are full y resistant to natural nucleases that are specific for pentoses of the D enantiomeric series 59,60 This has been conf irmed b y an imaging study with [18F]-labeled L-DNA or L-RN A Spie gelmers in mice, rats, and primates 60 (Figure 3).

Aptamers for Molecular Imaging

579

Figure 3. Dynamic positron emission tomography scan of a baboon after injection of a tracer dose of [18F]-L-DNA (top) or [18F]-L-RNA (bottom) Spiegelmers. Numbers indicate the middle of the time frame intervals for each image. Images were normalized for the injected dose and decay corrected. Color scale is from black (no radioactivity) to red (maximal radioactivity concentration). Reproduced with permission from Boisgard R et al.60

APTAMERS FOR CELLULAR IMAGING Basic studies support aptamers as ligands with excellent binding proper ties, in some cases superior to that of antibodies. Accordingly, aptamers have been substituted for monoclonal antibodies (mAbs) in diagnostic applications, for instance ELISA-lik e assa ys61,62 and flo w cytometry.63,64,32 In microscop y, Blank and colleagues used a fluorescein isothiocyanate-conjugated aptamer to visualize the neoangiogenetic microvasculature in tissue sections of rat gliob lastoma.32 Other imaging applications appear unique to the nucleic acid str ucture of aptamers, for instance, the capacity of aptamers to fold into a stable spatial structure in the presence of their ligand (“structuration on the ligand ” 65), which is now routinely used to tur n aptamers into beacons producing a fluorescent signal upon str uctural rear rangement in the presence of the tar get.66,67 In the absence of the tar get, the fluorophore attached to the aptamer is quenched b y a complementary sequence carrying a quencher moiety, whereas in the presence of the tar get, the aptamer folds around it, releasing the quencher and generating a fluorescent signal (for instance see F igure 4). Babendure and colleagues 68 showed that the MG aptamer, w hose structure has been repor ted above, can enhance tremendously (over 2,000 fold) the lo w intrinsic fluorescence of MG. By linking the MG aptamer to DN A aptamers recognizing small molecules, “modular aptamer sensors” that repor t on the presence of endo genous (ATP, flavine mononucleotide) or e xogenous (theoph ylline) compounds in comple x media w ere created. 69 Another elegant use of aptamers is to e xpress inside the cells a

natural RN A aptamer sequence for a fluorophore that can repor t on the re gulation of the e xpression system. Studies by Tsien and others ha ve shown that the insertion of an aptamer sequence in the noncoding re gion of a mRNA can provide a means to control the e xpression of the mRNA upon addition of the relevant target of the aptamer.70,71 It would therefore be attractive to report on mRNA transcription directl y inside the cells b y inser ting an aptamer reco gnizing MG (or other fluorescent probes activated upon binding to their cognate aptamer) in the 5’UTR noncoding sequence of a mRN A.68 Due to their small size (10 to 20 kDa), aptamers experience limited steric hindrance for binding to tar gets in complex environments. Combined with the high selectivity of tar get recognition, this confers aptamers with a high capacity to discriminate selecti vely their tar get inside a mixture, with an ef ficiency that has been reported superior to that of antibodies. This property has been e xploited for the isolation of stem cells based on recognition of specif ic cell surf ace biomark ers b y aptamers72 and for cellular imaging. 29,73 Lorger and colleagues73 selected aptamers against the v ariant surf ace glycoprotein (VSG) of Trypanosoma brucei, the parasite that causes sleeping sickness in humans. VSG was visualized b y incubating tr ypanosomes with biotin ylated aptamers, follo wed b y staining with fluorophore-conjugated strepta vidin, w hile VSG remained inaccessib le to immunostaining with antibodies. Atomic force microscopy (AFM) imaging with a DN A aptamer recognizing Immuno globulins E (IgE) sho wed that binding was ef ficient, specif ic, and not interfered with b y the presence of other proteins even in large amounts.74

580

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 4. Quantum dot-aptamer conjugates for synchronous cancer imaging, therapy, and sensing of drug delivery based on bifluorescence resonance energy transfer. A, Schematic illustration of the Bi-FRET system. In the first step, the CdSe/ZnS core-shell quantum dots (QDs) are surface functionalized with the A10 prostate-specific membrane antigen (PSMA) aptamer (Apt). The intercalation of the anticancer drug Doxorubicin (Dox) within the A10 PSMA aptamer on the surface of QDs results in the formation of the QD-Apt(Dox) and quenching of both QD and Dox fluorescence through a Bi-FRET mechanism: the fluorescence of the QD is quenched by Dox, whereas simultaneously the fluorescence of Dox is quenched by intercalation within the A10 PSMA aptamer resulting in the “OFF” state. B, Schematic illustration of specific uptake of QD-Apt(Dox) conjugates into target cancer cell through PSMA mediate endocytosis. The release of Dox from the QD-Apt(Dox) conjugates induces the recovery of fluorescence from both QD and Dox (“ON” state) thereby sensing the intracellular delivery of Dox and enabling the synchronous fluorescent localization and killing of cancer cells. Reproduced with permission from Bagalkot V et al.81

However, the AFM study of Lin and colleagues74 also showed that the strength of binding, measured by the average force necessar y to break the aptamer -protein bond , was lower than that of antibodies. The general observation that the af finity of aptamers for their tar get is lower than that of antibodies questions their capacity to produce a high contrast-to-noise ratio for molecular imaging, especially in conditions of low abundance of the molecular target where lower affinity may lead to insufficient contrast. A favored approach to circumv ent this problem has been to create multi valent binders b y conjugating se veral aptamers together75,76 or with molecular scaffolds such as polymers, nanopar ticles. Huang and colleagues 77 at the University of Florida conjugated fluorescein-labeled aptamers to Au-Ag nanorods (NR). Up to 80 copies of an aptamer selected for cell reco gnition of a human acute lymphoblastic leuk emia cell line (CCRF-CE cells) w ere conjugated to 12 × 56 nm Au-Ag NR and tested for binding to the tar get cells and other cells. The authors reported an increase of ca. 30 fold in the af finity of the aptamer-conjugated NR with respect to free aptamer binding, which resulted in an increase in fluorescence signal bound to the cells of over 300 fold. Yigit and colleagues 78 at the Uni versity of Illinois produced an ele gant demonstration of the magnetic

resonance imaging (MRI) “smart probe” concept based on aptamer. Revisiting the principle of magnetic nanopar ticles as sensors for the detection of molecular interactions,79 they cross-linked a super paramagnetic iron oxide nanoparticle (SPIO) with tw o complementar y DN A sequences. One of the sequences w as ter minated b y an aptamer sequence reco gnizing adenosine. In the absence of adenosine, the complementary DNA strands hybridized together and complexed dextran-coated SPIOs into a cluster that dephased the spins of neighboring w ater protons, therefore decreasing the T2 relaxation time. Addition of adenosine that bound the tar get aptamer changed its conformation thereb y destabilizing the DN A duple x and inducing dispersion of the SPIOs from the cluster . The resulting increase in the T2 relaxation time could be detected in a 4.7 T magnet. The detection limit of adenosine concentrations in the mixture w as 10 µM, and the T2 relaxation time increased linearly with adenosine concentration over a 0.1 to 1 mM range. Interestingl y, the specificity of the aptamer for adenosine allo wed detecting adenosine selectively with respect to the other nucleosides cytidine, uridine, or guanosine, and detection was retained in the presence of 10% human ser um. A similar concept w as applied to MRI detection of thrombin, using SPIOs conjugated to tw o dif ferent

Aptamers for Molecular Imaging

aptamers binding to tw o different sites of thrombin: one is that bound to the f ibrinogen-recognition e xosite and another one is that bound to the heparin-binding e xosite. Upon addition of thrombin consisting of both f ibrinogen and heparin, the aptamers induced assembly of thrombinSPIO complexes by binding to the tw o exosites thereby increasing the size of the SPIO clusters and decreasing T2 relaxation times. Concentrations as low as 25 nM thrombin could be detected in human serum, and detection was selective for thrombin (Figure 5).80 Omid Farokhzad and Rober t Langer at Har vard had explored another type of nanopar ticle-aptamer conjugate.81 The nanoparticle construction was built as a dr ug-delivery system comprising three distinct elements: (i) the nanoparticle scaffold, a fluorescent quantum dot (QD), (ii) the anticancer fluorescent dr ug do xorubicin, and (iii) the cancer targeting agent, a 2'fluorop yrimidine RNA aptamer tar geting the prostate-specif ic membrane antigen (PSMA) protein. Fluorescence from the QD w as quenched b y doxorubicin, w hereas fluorescence from do xorubicin w as quenched by intercalation into the aptamer , resulting in a double-quench system in the “off ” state when all three elements were associated together. Upon delivery into LNCaP target cells, doxorubicin was released from the aptamer, dequenching the fluorescence signals from both the dr ug and the QD (see Figure 4). Because delivery was dependent on

A

581

the inter nalization of the comple x through binding of the aptamer to PSMA, the ter nary constr uction by Farokhzad and Langer represents an interesting proof-of-concept on the use of an aptamer as an enhanced drug-delivery system. Accordingly, staining of LNCaP cells, but not of PSMAnegative PC3 cells, with the QD-aptamer conjugate w as observed. Although this conjugate w as not tested in vi vo, the same g roup reported efficient delivery to prostate cancer cell line x enografts in nude mice of nanopar ticles loaded with docetax el and conjugated to the same aptamer.82 Imaging was not performed in that study, but the spectacular effect on tumor growth of the aptamer-targeted, drug-containing nanopar ticles highlights the potential of aptamers as tumor targeting agents for in vivo applications. Chu and colleagues at the Uni versity of Texas at Austin labeled tumor cells with fluorescent CdSe and CdTe nanocrystals (QD) conjugated to a 2' fluorop yrimidine RNA aptamer targeting the PSMA protein. 83 They showed specif ic tar geting of the LNCaP tumor cells overexpressing PSMA, whereas PC3 cells, which do not overexpress PSMA, did not tak e up the aptamer . The number of PSMA binding sites w as estimated at approximately 10 6 sites per cell, and the af finity of the aptamer for whole cells (estimated by measuring its Kd) was found to be f ive times lower than its affinity for the isolated PSMA protein, estimated from its Ki (the dissociation constant for the inhibitory activity). Interestingly, the selectivity of aptamer reco gnition was demonstrated by its capacity to reco gnize tumor cells in a collagen matrix simulating a biolo gical tissue. The authors concluded that aptamer-QD conjugates offer a powerful and general tool for cellular imaging and may help to visualize tumorigenesis at the cellular and molecular le vel, while in vivo imaging applications may be dependent on biodistribution and toxicity issues.

IN VIVO MOLECULAR IMAGING WITH APTAMERS

B Figure 5. Magnetic resonance imaging detection of thrombin using an aptamer. Cross-linked dextran-coated superparamagnetic iron oxide (CLIO) nanoparticles (shown as red spheres) have been modified with either Thrm-A, a DNA aptamer (shown as blue lines) that binds to fibrinogen-recognition exosite of thrombin, or Thrm-B, a DNA aptamer (shown as green lines) that binds to the heparinbinding exosite of thrombin. Addition of thrombin consisting of both fibrinogen (as blue donuts) and heparin (as green donuts) exosites results in aggregation of CLIO nanoparticles, reducing the T2 relaxation time. Reproduced with permission from Yigit MV et al.80

Regarding in vi vo molecular imaging of aptamers, the requirement for a method with a high detection sensitivity has called in f avor of nuclear imaging techniques such as single photon emission computerized tomo graphy (SPECT) and positron emission tomography (PET).

Radiolabeling of Oligonucleotides Several methods to radiolabel oligonucleotides with isotopes for nuclear imaging have been described (reviewed in the studies by Tavitian,84 Younes and colleagues,85 and

582

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Tavitian86). All radiolabeling methods repor ted so f ar modify the oligonucleotide b y addition of a chemical group, which may significantly affect its biodistribution. Indeed, it has been repor ted that, depending on the radiometal chelating group, 4 hours following the injection of [ 99mTc]-labeled DNA, the percentage of injected dose per g ram recovered in mouse or gans v aried from 2.5 to 30.1 in kidne y and 3.3 to 24 in li ver, for MA G3 and HYNIC, respectively.87 In contrast, using a competitive h ybridization assa y,88 we showed that conjugation with a [ 18F]-fluorobenzyl at the 3' end did not modify significantly the blood kinetics of oligonucleotides. 89 In any case, the influence of the size and nature of the group added to the aptamer on its biodistribution should be carefully examined for in vivo imaging applications. A few studies have explored the general pharmacological profile of oligonucleotides in the absence of a def ined target. Hnatowich and colleagues90 imaged [99mTc]-labeled phosphodiester and phosphorothioate 22-mer oligonucleotides in nor mal mice. Both compounds sho wed lo w molecular w eight metabolites, demonstrating that phosphorothioate is not totally stable in vivo and presents high levels of protein binding. Whole body clearance was much slower for the phosphorothioate than for the phosphodiester due to a high hepatic uptake of the phosphorothioate. The authors concluded the phosphodiester DN A may be the prefer red [ 99mTc]-labeled oligonucleotide to a void the high and persistent li ver uptak e obser ved with the phosphorothioate DN A. Our laborator y perfor med comparative imaging studies in monk eys with 3 ʹ′- end [ 18F]labeled phosphodiester , phosphorothioate, and 2ʹ′-O-methyl RNA oligonucleotides.89 We observed a rapid degradation of the phosphodiester (half-life in the plasma 3 to 5 min), w hereas labeled phosphorothioate and 2 ʹ′-Omethyl RN A remained intact in plasma during at least 2 hours follo wing injection. Phar macodistribution of the radioactivity showed both renal and digesti ve elimination of the phosphodiester , whereas the phosphorothioate and 2ʹ′-O-methyl RN A sho wed onl y renal e xcretion, and the phosphorothioate accumulated in the li ver (0.1 % of injected dose per mL of tissue between 20 and 80 min after injection). With respect to in vi vo imaging, the most promising candidate appeared to be the 2ʹ′-O-methyl RNA, because it was both nuclease resistant and de void of nonspecific liver uptake.

Reports on Nuclear Imaging of Aptamers To be of practical use as in vi vo imaging agents , aptamers must exhibit adequate biodistribution, appropriate systemic clearance, efficient delivery, and biostability. Much of the

efforts to ward in vi vo use of aptamers for molecular imaging ha ve dealt with the delicate tailoring of their pharmacokinetic properties. A g roup at NeXstar Phar maceuticals pub lished the first report on the use of an aptamer for in vivo imaging in 1997.23 This group used an aptamer obtained by “blended” SELEX, in which valyl diphenyl ester phosphonate, a weak inhibitor of the enzyme elastase, was incorporated into the DNA oligonucleotide library during SELEX. Elastase is a soluble protease secreted by neutrophils at sites of inflammation that concentrates at those sites through interaction with the α1 proteinase inhibitor and binds to the neutrophil surface. When administered IV in a rat re verse passi ve Arthus inflammation model, the aptamer labeled with 99m Tc concentrated in the inflamed limb in respect to the control, noninflamed limb and the le vel of binding cor related with the enzymatic activity. The target-to-background ratio (ratio of uptake between the two limbs) reached 4.3 at 2 hours postinjection, to be compared with the 3.1 ratio at 3 hours obtained with the antibody clinically used to image inflammation. However, a control aptamer also sho wed a significant level of binding. Impor tantly, binding w as not saturable even at concentrations as high as 1 µM, and the addition of nonspecific DNA only partially competed with the binding of the aptamer. In contrast, aptamer binding to fixed cells w as lower and could be saturated , suggesting that the aptamer may have been internalized in vivo. Taken together, these results led the authors to propose that their aptamer was submitted to tw o types of nonspecif ic binding, saturab le and nonsaturab le, to gether with specif ic binding. Regrettably, there w as no repor t on the b lood pharmacokinetics and metabolism of the aptamer , and therefore it can be questioned w hether the imaging data reflects binding of the aptamer itself or of some degradation product. Ne vertheless, this repor t had the merit to trigger the interest for in vivo imaging with aptamers and to raise a number of issues that are of general v alue: • The labeling method w as critical with respect to the activity of the aptamer. In the case of the anti-elastase 99m aptamer, the N3S1–type cage for Tc labeling reduced the affinity significantly.20 • In comparison with a reference antibody , the aptamer showed a faster clearance leading to a higher and earlier tar get-to-background uptak e. Ho wever, the le vel of uptake (hence, sensitivity) was lower than with the antibody. • In vivo, the le vel of nonspecif ic binding w as signif icant and could not be predicted from in vitro data.The question of nonspecif ic binding w as reco gnized as

Aptamers for Molecular Imaging

crucial, e ven though in the NeXstar study it ma y have been in par t due to the tar get, because neutrophils are known to produce the Mac-1 protein that internalizes DNA. The conclusion of that repor t was that aptamer ligands could become useful as diagnostic imaging agents and may offer signif icant advantages over mAbs. Ho wever, it was not until 9 y ears later that a second e xample of targeted in vivo aptamer imaging appeared in the literature.91 In this repor t from the Schering compan y, working together with a g roup from NeXstar , a modif ied 13 kDa RNA aptamer raised against human tenascin-C w as carefully tailored to impro ve its stability in vi vo. Tenascin-C is a he xameric e xtracellular matrix protein that is highl y overexpressed (1 µmol/L) in se veral types of tumors. Tenascin-C has six binding sites for the aptamer, coded TTA-1, which was built from ribose modified in the 2' position of the p yrimidine bases b y the replacement of a h ydroxyl g roup with a fluorine atom. The 2' hydroxyl of the ribose of the purine bases was substituted with 2'-OMe in se veral positions and a 3'-3' inverted phosphodiester linkage w as added at the 3' extremity. TTA-1 was labeled b y 99mTc, and the authors demonstrated that conjugation labeling did not alter affinity (Kd = 5 nM) for the protein tar get. [99mTc]-TTA-1 was IV injected into nude mice bearing tumor xenografts of human U251 glioma, and the animals were imaged on a gamma camera. The initial tumor uptake was maximal at 10 minutes postinjection, and not different from that of a control aptamer . Ho wever, the control aptamer w ashed out from the tumor w hile TTA-1 w as retained in the tumor and cleared from the b lood, yielding a tumor-to-blood ratio of 50 at 3 hours and 180 at 16 hours postinjection. In contrast, a tenascin-C antibody sho wed a tumor-to-blood ratio of onl y 5 at 40 hours postinjection. Quantification of radioacti vity in mouse tissues sho wed that 2.7% of the injected dose per g ram of tumor w as recovered 1 hour after injection. The routes of excretion of [ 99mTc]-TTA-1 were through both the kidne ys (50%) and the liver (50%), and 95% of the radioacti vity was cleared from the body at 24 hours. PEGylation of the aptamer could limit its hepatobiliar y clearance. F inally, at 3 hours postinjection, radioactivity represented essentially metabolites in the plasma, w hereas 60% of radioacti vity w as unchanged TTA-1 in the tumor. The conclusions of this carefull y conducted study confirm and complement those of the initial report by the NeXstar g roup on the possibility of using aptamers as imaging agents to recognize protein targets.92 Several key points can be emphasized:

583

• The importance of the labeling group in respect to the biodistribution of the aptamer. Hicke and colleagues90 reported that changing the radiometal chelator had dramatic consequences on tissue uptake and clearance patterns. Similar obser vations on the influence of chelating agents on oligonucleotide biodistribution had been repor ted pre viously b y Zhang and colleagues.87 • The faster uptake and clearance and the lower level of uptake of aptamers, in respect to that of antibodies. In that respect, it is note worthy that w hile both studies showed a lower level of uptake for the aptamers than for the respective antibodies, they also evidenced better tumor-to-background ratios for the aptamers. The tradeoff betw een uptak e le vel and w ashout, w hich constitutes the basis for image contrast, therefore appears in f avor of the aptamers as protein tar geting agents. • The limited kno wledge on the actual mechanism b y which aptamers bind to their target in vivo. In the case of the DNA elastase aptamer, binding in the inflamed tissue appeared to reflect a signif icant propor tion of “nonspecific” uptak e. In contrast, in the case of the tenascin aptamer, binding w as both aptamer specif ic (ie, there w as no binding of the control aptamer) and target specific (ie, there was no binding to a tumor not expressing tenascin). However, the observation that the tumor uptake was dose dependent is hard to reconcile with a classical protein-ligand interaction. Increasing the dose by 100-fold increased tumor uptak e by a factor of 3 and decreased clearance rate, but this dose effect w as identical w hen a control, nonbinding aptamer was used. The authors suggested the possibility of saturab le clearance mechanisms independent of the aptamer activity, a hypothesis that deserves further exploration. A g roup in the UK led b y Alan Perkins and Sotiros Missailidis de veloped DN A aptamers against MUC1, a mucin that is overexpressed and aberrantly glycosylated on the surface of cancer cells. MUC1 is the antigen in commercial tumor mark er assays and considered a promising target for imaging and therap y. An aptamer reco gnizing MUC1 w as obtained b y af finity chromato graphy-based SELEX, labeled with rhodamine red and sho wn to bind 93 MCF-7 breast cancer cells in culture. The authors mentioned that the aptamer association with its tar get was rapid, w hereas dissociation w as slo w, and the aptamer could displace binding of the antibody to the tar get. In another repor t, radiolabeling of the aptamer with 99mTc, using either the MA G3 chelator or a no vel c yclen-based

584

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

chelator, was reported.75 No in vivo images were provided in that repor t although it w as mentioned that the tumor visualization w as “poor ,” and the authors suggested that this was the consequence of a rapid renal excretion. Biodistribution at 6 hours after injection indicated a high liver and kidney uptak e with the MA G3 conjugate, and a high uptake in the stomach with the c yclen conjugate, suggesting signif icant release of technetium from the chelator. In an attempt to increase circulation times of the aptamer, tetrameric comple xes of the aptamer w ere attached to either DO TA or carbo xyporphyrin and radiolabeled. Again, biodistribution w as suggestive of the release of free technetium from the comple xes. F inally, ex vivo autoradio graphy sho wed e xtensive dif fusion of radioactivity inside the tumor mass, described b y the authors as cor responding to intact aptamer , although no evidence w as pro vided for that. It w as concluded that MUC1 aptamers could potentiall y become interesting agents for tumor imaging and possib ly radiotherapy once the issues of label stability and pharmacokinetics would be solved. The group at TRIUMF in Vancouver evaluated DNA aptamers directed against thrombin as thrombus imaging agents.94 In an extensive series of in vitro binding assays, it w as sho wn b y this g roup that the site of binding to thrombin w as critical for imaging applications. One of their aptamers binding e xosite 1 competed with f ibrin for binding to thrombin and displaced thrombin from the thrombus. This aptamer w as therefore considered unfavorable for imaging. A second aptamer binding exosite 2 (the heparin binding site) sho wed no interference with the thrombin binding to the f ibrin clot, and its binding to thrombi w as fur ther e xplored in vi vo in a rabbit model of jugular v ein thrombus. Aptamers w ere labeled with iodine-125 or iodine-123, administered to the animals and thrombi imaged b y scintig raphy or counted after e xcision. Modif ications of the 3' or 5' ends of the aptamers did not change their af finity for thrombin. Ho wever, the uptak e of the aptamer in the thrombus was not higher than that of labeled o valbumin, indicating the absence of a specif ic uptak e. Two explanations can be proposed for that ne gative f inding: (i) thrombin concentration in the clot (37 nM) could have been too lo w to achie ve signif icant uptake; (ii) the time necessary for the aptamer to bind to the thrombus, on the order of se veral hours, ma y ha ve been too long with respect to the circulation time of the aptamer , which was on the order of minutes. The authors concluded that increasing circulation time of the aptamers might improve imaging.

IN VIVO DELIVERY IS THE CRITICAL POINT FOR MOLECULAR IMAGING WITH APTAMERS Concerning resistance to nucleases, modif ications at the 2'-position of the ribose and in verted ter minal cap str uctures dramaticall y impro ve aptamer stability in vi vo. In addition to de gradation b y nucleases, another issue of concern is the systemic clearance of aptamers, which are subject to renal elimination. A nuclease-resistant oligonucleotide administered IV exhibits an in vi vo half-life of less than 10 minutes.89,60 Fortunately, a wide range of possibilities is a vailable to introduce changes in aptamers to modify their phar macokinetics proper ties. F or instance, the clearance rates of aptamers can be altered to lengthen circulation times by anchoring them to liposome bila yers or by coupling them to inert large molecules such as PEG or h ydrophobic g roups.94 Dougan and colleagues 95 showed that biotin conjugation of the 3' end of an anticoagulant aptamer against thrombin protected its de gradation b y nucleases but did not increase its circulation time, whereas conjugation of the biotin ylated aptamer to streptavidin increased b lood lifetime b y 10 to 20 fold without hampering its affinity for thrombin. Working at Archemix Cor p, Heal y and colleagues 96 examined the ef fects of conjugation of small molecules, peptides, and polymer terminal groups on the pharmacokinetics and biodistribution of an anti-TGF β2 aptamer in vivo. They found that conjugation to PEG could dramatically increase aptamer residence in b lood circulation and modify the distribution to tissues. Depending on the chemical composition, aptamer mean residence times in the blood could be extended from 0.6 to 16 hours. A mixed composition aptamer containing both 2ʹ′-F and 2ʹ′-OMe stabilizing modifications persisted signif icantly longer in the blood stream than did a full y 2 ʹ′-O-methylated composition. Furthermore, complexation with a 20 kDa PEG polymer proved nearly as effective as a 40 kDa PEG polymer in preventing aptamer elimination. 96 Globally, it appears relati vely straightforw ard to improve the phar macokinetics of aptamers directed against intra vascular tar gets. Ho wever, the capacity of either unconjugated or PEGylated aptamers to escape the vasculature and distribute to or gans and tissues in vi vo, especially in the case of intracellular deli very, is still largely unsolv ed. In the Archemix study , though the primary effect of PEGylation w as on aptamer clearance, the prolonged blood circulation of the 20 kDa PEG-conjugated aptamer appeared to f acilitate its distribution to tissues, par ticularly those of highl y perfused or gans.96

Aptamers for Molecular Imaging

Conversely, the advantages of decreasing blood clearance by increasing the apparent molecular w eight may be balanced by the f act that the relati vely small size of unconjugated aptamers (8 to 15 kDa) is lik ely to f avor tissue penetration. Previous studies in volving antisense oligonucleotides have e xplored the ef fects of v arious conjugation chemistries on phar macokinetics and biodistribution. 97 Conjugation with v arious lipids such as cholesterol has been repor ted to increase the circulation half-life of antisense oligonucleotides. 98 We ha ve tested the ef fect of oligonucleotide incor poration into ar tificial v ectors on their biodistribution b y PET imaging. 99 The phar macokinetics and bioa vailability of an anti-HIV 1 PO oligonucleotide delivered either in a free form or using cationic or anionic synthetic car rier systems was compared by whole body dynamic quantitati ve imaging. The anionic v ector dramatically enhanced the uptake in several organs, including the lungs, spleen, and brain, with a prolonged accumulation of radioactivity in the brain. With this vector, intact oligonucleotide was detected in plasma for up to 2 hours after injection and the elimination half-life (T 1/2β) and distribution volume increased by 4 and 7 fold, respectively. Another approach has been the use of cellpenetrating peptides able to transport antisense oligonucleotides across cellular membranes. Examples of these conjugates include a fragment of the HIV Tat protein,100 a sequence derived from the third helix of the Drosophila antennapedia (Ant) homeotic protein, 101 and positi vely charged peptides composed of polyarginine.102 However, with respect to antisense oligonucleotides, aptamers are longer, possess dif ferent types of chemical modif ications, and fold into more complex tertiary structures, and the in vivo biodistribution of aptamers is not readily predictable from studies with antisense oligonucleotides. Like for many macromolecules, the major limitation to in vi vo applications of aptamers is deli very, par ticularly when their biolo gical targets are located inside the intracellular compar tment of specif ic or gans. Aptamers targeting extracellular markers represent a less stringent pharmacological challenge.

CONCLUSION Because the y ri val antibodies in ter ms of af finity and specificity for their tar gets, aptamers are considered promising for anal ytical applications, tar get v alidation, drug discovery, in vitro and in vivo diagnostics, and therapeutic agents. A typical aptamer is 10 to 15 kDa in size, binds its target with nanomolar affinity, and discriminates among closely related targets. Aptamers are generated by

585

a synthetic selection process in w hich the e xperimental setting may be fully controlled for the f abrication of ligands endo wed with specif ic reco gnition characteristics. Many possibilities are a vailable to introduce changes in aptamer str ucture through def ined chemical modif ications, and aptamers can be easil y conjugated to a wide array of labels or other molecules. This allows testing of a lar ge number of modif ications designed to impro ve their phar macokinetics or confer aptamers with unique properties dedicated for specif ic applications. Aptamers can address a very wide range of targets and have shown little or no toxicity or immunogenicity. The way aptamers engage their tar get molecules is dif ferent from that of antibodies and renders them unique as reco gnizing agents. Aptamers can be selected against tar gets that are too toxic to be used as antigens for raising antibodies in a living or ganism, as w ell as against tar gets that do not elicit antibody response. As drugs, several aptamers have reached the clinical development stage, and “escor t” aptamers are a budding concept in w hich the aptamer ma y be used to deli ver an active drug, radionuclide, toxin, or cytotoxic agent to the desired site for diagnostic tests and therap y. Their discrimination and tar geting capacities stem from their unique binding characteristics and mak e them attracti ve as imaging agents for noninvasive diagnostic procedures. However, the f ield of molecular imaging with aptamers is presently in its infancy and further work is necessary to better define their potential as molecular imaging probes. Trends in the fur ther development of aptamers as imaging agents concer n impro ved in vi vo stability , improvement in their binding affinity, and increasing circulation times. Kno wledge concer ning e xtravasation and the capacity of aptamers to access diseased tissues or the interior of tumors is required to better def ine the spectrum of tar gets accessib le for imaging. Similarl y, the minimal target concentration that can be recognized by aptamers is still a matter of question. There is lar gely room for the improvement of aptamers, for instance, through multimerization or conjugation to nano-scaf folds to improve delivery to mak e these “nucleic acid biotools” e ven more attractive for in vivo molecular imaging. Finally, numerous ways in which aptamers can be used are still e volving and this young domain of research has the signif icant advantage of a great technological diversity.

ACKNOWLEDGMENTS I thank the members of m y laborator y, F rédéric Ducongé, Carine P estourie, Elina Zue va, Fabien Chauveau, Daniel Miott-Dupont, Bertrand Kuhnast, Frédéric

586

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Dollé, and Raphaël Boisgard for their w ork on aptamer imaging, Vittorio de Franciscis and Laura Cerchia (University of Naples, Italy), Sven Klussman and Jens-Peter Fürste (Noxxon Pharma, Germany) and Domenico Libri (CGM Gif-sur -Yvette, F rance), for fr uitful collaborations, and Jean-Jacques Toulmé (University of Bordeaux, F rance) and Andrew Stephens (for merly at Schering, Germany) for inspiring discussions. Research in the author’s laboratory is suppor ted by the European networks of excellence EMIL and DIMI, by the Agence nationale de la Recherche and by the Institut national du Cancer.

REFERENCES 1. Schrödinger E. What is life? The physical aspect of the li ving. New York: The Macmillan Company; 1946. 2. Cech TR. RNA finds a simpler way. Nature 2004;428:263–4. 3. Robertson DL, Jo yce GF. Selection in vitro of an RN A enzyme that specifically clea ves single-stranded DN A. Nature 1990; 344:467–8. 4. Green R, Ellington AD, Szostak JW. In vitro genetic analysis of the Tetrahymena self-splicing intron. Nature 1990;347:406–8. 5. Thiesen HJ, Bach C. Target detection assay (TDA): a versatile procedure to determine DNA binding sites as demonstrated on SP1 protein. Nucleic Acids Res 1990;18:3203–9. 6. Blackwell TK, Kretzner L, Blackw ood EM, et al. Sequence-specif ic DNA binding by the c-Myc protein. Science 1990;250:1149–51. 7. Blackwell TK, Weintraub H. Dif ferences and similarities in DN Abinding preferences of My oD and E2A protein comple xes revealed by binding site selection. Science 1990;250:1104–10. 8. Pollock R, Treisman R. A sensitive method for the deter mination of protein-DNA binding specif icities. Nucleic Acids Res 1990; 18:6197–204. 9. Ellington AD, Szostak JW. In vitro selection of RN A molecules that bind specific ligands. Nature 1990;346:818–22. 10. Tuerk C, Gold L. Systematic e volution of ligands b y e xponential enrichment: RNA ligands to bacteriophage T4 DNA polymerase. Science 1990;249:505–10. 11. Grate D , Wilson C. Laser -mediated, site-specif ic inacti vation of RNA transcripts. Proc Natl Acad Sci USA 1999;96:6131–6. 12. Baugh C, Grate D, Wilson C. 2.8 A crystal structure of the malachite green aptamer. J Mol Biol 2000;301:117–28. 13. Nguyen DH, DeFina SC, Fink WH, Dieckmann T. Binding to an RNA aptamer changes the char ge distribution and confor mation of malachite green. J Am Chem Soc 2002;124:15081–4. 14. Flinders J, DeFina SC, Brackett DM, et al. Recognition of planar and nonplanar ligands in the malachite g reen-RNA aptamer complex. Chembiochem 2004;5:62–72. 15. Brackett DM, Dieckmann T. Aptamer to ribozyme: the intrinsic catalytic potential of a small RNA. Chembiochem 2006;7:839–43. 16. Cerchia L, Hamm J , Libri D , et al. Nucleic acid aptamers in cancer medicine. FEBS Lett 2002;528:12–16. 17. Brody EN, Gold L. Aptamers as therapeutic and diagnostic agents. J Biotechnol 2000;74:5–13. 18. Joyce GF . Directed molecular e volution. Sci Am 1992; 267:90–7. 19. Osborne SE, Matsumura I, EllingtonAD. Aptamers as therapeutic and diagnostic reagents: prob lems and prospects. Cur r Opin Chem Biol 1997;1:5–9. 20. Fitzwater T, P olisky B . A SELEX primer . Methods Enzymol 1996;267:275–301.

21. Gopinath SC. Methods de veloped for SELEX. Anal Bioanal Chem 2007;387:171–82. 22. Cox JC, Rudolph P , Ellington AD. Automated RN A selection. Biotechnol Prog 1998;14:845–50. 23. Charlton J , Sennello J , Smith D . In vi vo imaging of inflammation using an aptamer inhibitor of human neutrophil elastase. Chem Biol 1997;4:809–16. 24. White R, Rusconi C, Scardino E, et al. Generation of species crossreactive aptamers using “to ggle” SELEX. Mol Ther 2001; 4:567–73. 25. Doudna JA, Cech TR, Sullenger BA. Selection of an RN A molecule that mimics a major autoantigenic epitope of human insulin receptor. Proc Natl Acad Sci USA 1995;92:2355–9. 26. Hamm J, Huber J, Luhrmann R. Anti-idiotype RNA selected with an anti-nuclear e xport signal antibody is acti vely transpor ted in oocytes and inhibits Rev- and cap-dependent RNA export. Proc Natl Acad Sci USA 1997;94:12839–44. 27. Hamm J, Fornerod M. Anti-idiotype RNAs that mimic the leucinerich nuclear export signal and specifically bind to CRM1/exportin 1. Chem Biol 2000;7:345–54. 28. Morris KN, Jensen KB, Julin CM, et al. High affinity ligands from in vitro selection: comple x tar gets. Proc Natl Acad Sci USA 1998;95:2902–7. 29. Ohuchi SP, Ohtsu T, Nakamura Y. Selection of RNA aptamers against recombinant transfor ming g rowth f actor-beta type III receptor displayed on cell surface. Biochimie 2006;88:897–904. 30. Pestourie C, Cerchia L, Gomber t K, et al. Comparison of dif ferent strategies to select aptamers against a transmembrane protein target. Oligonucleotides 2006;16:323–35. 31. Cerchia L, Ducongé F, Pestourie C, Boulay J, Aissouni Y, Gombert K, Tavitian B, de F ranciscis V, Libri D . Neutralizing aptamers from w hole-cell SELEX inhibit the RET receptor tyrosine kinase. PLoS Biol. 2005 Apr;3(4):e123. 32. Blank M, Weinschenk T, Priemer M, Schluesener H. Systematic evolution of a DN A aptamer binding to rat brain tumor microvessels. selective targeting of endothelial re gulatory protein pigpen. J Biol Chem 2001;276:16464–8. 33. Service RF. Proteomics. Searching for recipes for protein chips. Science 2001;294:2080–2. 34. Hermann T, Patel DJ. Adaptive recognition by nucleic acid aptamers. Science 2000;287:820–25. 35. Breaker RR. DN A aptamers and DN A enzymes. Cur r Opin Chem Biol 1997;1:26–31. 36. Kumar PK, Ellington AD. Artificial evolution and natural ribozymes. FASEB J 1995;9:1183–95. 37. Rimmele M. Nucleic acid aptamers as tools and dr ugs: recent developments. Chembiochem 2003;4:963–71. 38. Burgstaller P, Girod A, Blind M. Aptamers as tools for target prioritization and lead identification. Drug Discov Today 2002;7:1221–8. 39. Farokhzad OC, Jon S, Khademhosseini A, et al. Nanoparticle-aptamer bioconjugates: a new approach for targeting prostate cancer cells. Cancer Res 2004;64:7668–72. 40. Blind M, Grättinger M, Mayer G. Nucleic acid biotools: accelerating the discovery of lead compounds. Screening 2002;6:35–7. 41. Blind M, K olanus W, Famulok M. Cytoplasmic RN A modulators of an inside-out signal-transduction cascade. Proc Natl Acad Sci USA 1999;96:3606–10. 42. Nimjee SM, Rusconi CP , Sullenger B A. Aptamers: an emer ging class of therapeutics. Annu Rev Med 2005;56:555–83. 43. Ostendorf T, Kunter U, Eitner F, et al. VEGF(165) mediates glomerular endothelial repair. J Clin Invest 1999;104:913–23. 44. Ruckman J , Green LS, Beeson J , et al. 2'-Fluorop yrimidine RNA-based aptamers to the 165-amino acid for m of v ascular endothelial growth factor (VEGF165). Inhibition of receptor binding and VEGF-induced vascular permeability through interactions requiring the e xon 7-encoded domain. J Biol Chem 1998; 273:20556–67.

Aptamers for Molecular Imaging

45. Huang J, Moore J , Soffer S, et al. Highl y specif ic antiangiogenic therapy is ef fective in suppressing g rowth of e xperimental Wilms tumors. J Pediatr Surg 2001;36:357–61. 46. Kim ES, Serur A, Huang J, et al. Potent VEGF blockade causes regression of coopted v essels in a model of neurob lastoma. Proc Natl Acad Sci USA 2002;99:11399–404. 47. Drolet DW, Nelson J, Tucker CE, et al. Phar macokinetics and safety of an anti-v ascular endothelial g rowth f actor aptamer (NX1838) following injection into the vitreous humor of rhesus monk eys. Pharm Res 2000;17:1503–10. 48. Gilbert JC, DeF eo-Fraulini T, Hutabarat RM, et al. F irst-in-human evaluation of anti v on Willebrand f actor therapeutic aptamer ARC1779 in healthy volunteers. Circulation 2007;116:2678–86. 49. Dyke CK, Steinhub l SR, Kleiman NS, et al. F irst-in-human experience of an antidote-controlled anticoagulant using RN A aptamer technology: a phase 1a phar macodynamic evaluation of a dr ugantidote pair for the controlled regulation of f actor IXa acti vity. Circulation 2006;114:2490–7. 50. McCauley TG, Kurz JC, Merlino PG, et al. Phar macologic and pharmacokinetic assessment of anti-TGFbeta2 aptamers in rabbit plasma and aqueous humor. Pharm Res 2006;23:303–11. 51. Hicke BJ, Stephens AW. Escort aptamers: a delivery service for diagnosis and therapy. J Clin Invest 2000;106:923–28. 52. Usman N, Blatt LM. Nuclease-resistant synthetic ribozymes: developing a new class of therapeutics. J Clin In vest 2000;106:1197–202. 53. Green LS, Jellinek D, Bell C, et al. Nuclease-resistant nucleic acid ligands to v ascular per meability f actor/vascular endothelial growth factor. Chem Biol 1995;2:683–95. 54. Pagratis NC, Bell C, Chang YF, et al. P otent 2'-amino-, and 2'-fluoro-2'-deoxyribonucleotide RN A inhibitors of k eratinocyte growth factor. Nat Biotechnol 1997;15:68–73. 55. Burmeister PE, Lewis SD, Silva RF, et al. Direct in vitro selection of a 2'-O-methyl aptamer to VEGF. Chem Biol 2005;12:25–33. 56. Chelliserrykattil J, Ellington AD. Evolution of a T7 RNA polymerase variant that transcribes 2'-O-meth yl RN A. Nat Biotechnol 2004;22:1155–60. 57. Micklefield J . Backbone modif ication of nucleic acids: synthesis, structure and therapeutic applications. Cur r Med Chem 2001;8:1157–79. 58. Latham J A, Johnson R, Toole JJ . The application of a modif ied nucleotide in aptamer selection: no vel thrombin aptamers containing 5-(1-pentyn yl)-2'-deoxyuridine. Nucleic Acids Res 1994;22:2817–22. 59. Vater A, Klussmann S. Toward third-generation aptamers: Spiegelmers and their therapeutic prospects. Curr Opin Drug Discov Devel 2003;6:253–61. 60. Boisgard R, K uhnast B, Vonhoff S, et al. In vi vo biodistribution and pharmacokinetics of 18F-labelled Spie gelmers: a ne w class of oligonucleotidic radiophar maceuticals. Eur J Nucl Med Mol Imaging 2005;32:470–77. 61. Drolet D W, Moon-McDer mott L, Romig TS. An enzyme-link ed oligonucleotide assay. Nat Biotechnol 1996;14:1021–25. 62. Baldrich E, Restrepo A, O’Sulli van CK. Aptasensor de velopment: elucidation of critical parameters for optimal aptamer performance. Anal Chem 2004;76:7053–63. 63. Davis KA, Lin Y, Abrams B, Jayasena SD. Staining of cell surf ace human CD4 with 2'-F-p yrimidine-containing RNA aptamers for flow cytometry. Nucleic Acids Res 1998;26:3915–24. 64. Ringquist S, P arma D. Anti-L-selectin oligonucleotide ligands recognize CD62L-positi ve leuk ocytes: binding af finity and specificity of uni valent and bi valent ligands. Cytometr y 1998; 33:394–405. 65. Carothers JM, Szostak JW . In vitro selection of functional oligonucleotides and the origins of biochemical activity. In: Klussmann S, editor. The aptamer handbook. Weinheim, Germany: Wiley-VCH Verlag GmbH & Co. KgaA; 2006. p. 3–28.

587

66. Hamaguchi N , Ellington A, Stanton M. Aptamer beacons for the direct detection of proteins. Anal Biochem 2001;294:126–31. 67. Yamamoto R, Katahira M, Nishika wa S, et al. A novel RNA motif that binds efficiently and specifically to the Ttat protein of HIV and inhibits the trans-acti vation by Tat of transcription in vitro and in vivo. Genes Cells 2000;5:371–88. 68. Babendure JR, Adams SR, Tsien R Y. Aptamers s witch on fluorescence of triphen ylmethane dy es. J Am Chem Soc 2003; 125:14716–7. 69. Stojanovic MN , K olpashchikov DM. Modular aptameric sensors. J Am Chem Soc 2004;126:9266–70. 70. Werstuck G, Green MR. Controlling gene e xpression in li ving cells through small molecule-RN A interactions. Science 1998; 282:296–98. 71. Babendure JR, Babendure JL, Ding JH, Tsien RY. Control of mammalian translation b y mRN A str ucture near caps. RN A 2006; 12:851–61. 72. Schafer R, Wiskirchen J, Guo K, et al. Aptamer-based isolation and subsequent imaging of mesench ymal stem cells in ischemic myocard by magnetic resonance imaging. Rofo 2007;179:1009–15. 73. Lorger M, Engstler M, Homann M, Goringer HU . Targeting the variable surface of African trypanosomes with v ariant surface glycoprotein-specific, serum-stable RNA aptamers. Eukaryot Cell 2003; 2:84–94. 74. Lin L, Wang H, Liu Y, et al. Reco gnition imaging with a DN A aptamer. Biophys J 2006;90:4236–8. 75. Borbas KE, F erreira CS, P erkins A, et al. Design and synthesis of mono- and multimeric tar geted radiophar maceuticals based on novel c yclen ligands coupled to anti-MUC1 aptamers for the diagnostic imaging and tar geted radiotherapy of cancer. Bioconjug Chem 2007;18:1205–12. 76. Perkins AC, Missailidis S. Radiolabelled aptamers for tumour imaging and therapy. Q J Nucl Med Mol Imaging 2007;51:292–96. 77. Huang YF, Chang HT , Tan W. Cancer cell tar geting using multiple aptamers conjugated on nanorods. Anal Chem 2008;80:567–72. 78. Yigit MV, Mazumdar D, Kim HK, et al. Smar t “turn-on” magnetic resonance contrast agents based on aptamer -functionalized superparamagnetic iron o xide nanopar ticles. Chembiochem 2007;8:1675–8. 79. Perez JM, Josephson L, Weissleder R. Use of magnetic nanopar ticles as nanosensors to probe for molecular interactions. Chembiochem 2004;5:261–4. 80. Yigit MV, Mazumdar D , Lu Y. MRI detection of thrombin with aptamer functionalized super paramagnetic iron o xide nanopar ticles. Bioconjug Chem 2008;19:412–17. 81. Bagalkot V, Zhang L, Levy-Nissenbaum E, et al. Quantum dot-aptamer conjugates for synchronous cancer imaging, therap y, and sensing of drug delivery based on bi-fluorescence resonance ener gy transfer. Nano Lett 2007;7:3065–70. 82. Farokhzad OC, Cheng J, Teply BA, et al. Targeted nanoparticle-aptamer bioconjugates for cancer chemotherapy in vivo. Proc Natl Acad Sci USA 2006;103:6315–20. 83. Chu TC, Shieh F, Lavery LA, et al. Labeling tumor cells with fluorescent nanocr ystal-aptamer bioconjugates. Biosens Bioelectron 2006;21:1859–66. 84. Tavitian B. In vivo imaging with oligonucleotides for diagnosis and drug development. Gut 2003;52 Suppl 4:40–7. 85. Younes CK, Boisgard R, Tavitian B . Labelled oligonucleotides as radiopharmaceuticals: pitfalls, problems and perspecti ves. Cur r Pharm Des 2002;8:1451–66. 86. Tavitian B . In vi vo antisense imaging. Q J Nucl Med 2000; 44:236–55. 87. Zhang YM, Liu N , Zhu ZH, et al. Influence of dif ferent chelators (HYNIC, MAG3 and DTPA) on tumor cell accumulation and mouse biodistribution of technetium-99m labeled to antisense DN A. Eur J Nucl Med 2000;27:1700–07.

588

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

88. Boutet V, Delaunay V, De Oli veira MC, et al. Real-time monitoring of the h ybridization reaction: application to the quantif ication of oligonucleotides in biolo gical samples. Biochem Bioph ys Res Commun 2000;268:92–8. 89. Tavitian B , Terrazzino S, K uhnast B , et al. In vi vo imaging of oligonucleotides with positron emission tomo graphy. Nat Med 1998;4:467–71. 90. Hnatowich DJ, Mardirossian G, Fogarasi M, et al. Comparati ve properties of a technetium-99m-labeled single-stranded natural DN A and a phosphorothioate derivative in vitro and in mice. J Pharmacol Exp Ther 1996;276:326–34. 91. Hicke BJ , Stephens AW, Gould T, et al. Tumor tar geting b y an aptamer. J Nucl Med 2006;47:668–78. 92. Gambhir SS. Using radiolabeled DNA as an imaging agent to recognize protein targets. J Nucl Med 2006;47:557–8. 93. Ferreira CS, Matthews CS, Missailidis S. DNA aptamers that bind to MUC1 tumour mark er: design and characterization of MUC1binding single-stranded DN A aptamers. Tumour Biol 2006;27:289–301. 94. Willis MC, Collins BD, Zhang T, et al. Liposome-anchored vascular endothelial g rowth f actor aptamers. Bioconjug Chem 1998; 9:573–82. 95. Dougan H, Lyster DM, Vo CV, et al. Extending the lifetime of anticoagulant oligodeo xynucleotide aptamers in b lood. Nucl Med Biol 2000;27:289–97.

96. Healy JM, Lewis SD, Kurz M, et al. Pharmacokinetics and biodistribution of novel aptamer compositions. Pharm Res 2004;21:2234–46. 97. Manoharan M. Oligonucleotide conjugates as potential antisense drugs with impro ved uptak e, biodistribution, tar geted deli very, and mechanism of action. Antisense Nucleic Acid Dr ug De v 2002;12:103–28. 98. de Smidt PC, Le Doan T, de F alco S, v an Berkel TJ. Association of antisense oligonucleotides with lipoproteins prolongs the plasma half-life and modif ies the tissue distribution. Nucleic Acids Res 1991;19:4695–700. 99. Tavitian B, Marzabal S, Boutet V, et al. Characterization of a synthetic anionic v ector for oligonucleotide deli very using in vi vo whole body dynamic imaging. Phar m Res 2002;19:367–76. 100. Vives E, Brodin P , Leb leu B . A tr uncated HIV-1 Tat protein basic domain rapidl y translocates through the plasma membrane and accumulates in the cell nucleus. J Biol Chem 1997;272:16010–17. 101. Pietersz GA, Li W, Apostolopoulos V. A 16-mer peptide (RQIKIWFQNRRMKWKK) from antennapedia preferentiall y tar gets the Class I pathway. Vaccine 2001;19:1397–405. 102. Rothbard JB, Kreider E, VanDeusen CL, et al. Arginine-rich molecular transpor ters for dr ug delivery: role of backbone spacing in cellular uptake. J Med Chem 2002;45:3612–8.

37 NONCLINICAL PRODUCT DEVELOPMENTAL STRATEGIES, SAFETY CONSIDERATIONS, AND TOXICITY PROFILES OF MEDICAL IMAGING AND RADIOPHARMACEUTICALS PRODUCTS SUNDAY AWE, PHD, MBA, SIHAM BIADE, PHARMD, PHD, SALLY J. HARGUS, PHD, TUSHAR KOKATE, PHD, ADEBAYO LANIYONU, PHD, AND YANLI OUYANG MD, PHD, DABT*

This chapter provides an overview of the nonclinical product de velopmental strate gies, safety considerations, and toxicity prof iles of medical imaging including radiopharmaceutical products. These consist of agents used with medical imaging modalities, such as ultrasonography (US), magnetic resonance imaging (MRI), computed tomography (CT), radiography, and radionuclide imaging (single photon emission computed tomo graphy [SPECT] and positron emission tomography [PET]). In general, medical imaging agents can be classif ied into at least tw o cate gories: (1) contrast agents and (2) diagnostic radiopharmaceuticals. Contrast agents impro ve the visualization of tissues, or gans, and ph ysiologic processes b y increasing the relative difference of imaging signal intensities in adjacent regions of the body. These agents include: (1) iodinated compounds used in radio graphy and CT ; (2) paramagnetic metallic ions, such as ions of gadolinium (Gd), iron, and manganese, link ed to a v ariety of molecules used in MRI; and (3) microbubb les, microaerosomes, and related micropar ticles used in ultrasonography. The U.S. Food and Dr ug Administration (FDA) defines a diagnostic radiopharmaceutical as (1) an article that is intended for use in the diagnosis or monitoring of a disease or a manifestation of a disease in humans, and that exhibits spontaneous disintegration of unstab le nuclei with the emission of nuclear par ticles or photons; or (2) any nonradioactive reagent kit or

nuclide generator that is intended to be used in the preparation of such an ar ticle.1

NONCLINICAL PRODUCT DEVELOPMENTAL STRATEGIES In general, medical imaging agents are go verned b y the same regulations as other dr ug and biolo gical products. 2,3 However, because medical imaging agents are used to diagnose and monitor diseases or conditions as opposed to treating them, their development programs can be tailored to reflect these par ticular uses. 4 For product approval purposes, the nonclinical phar macology and toxicology studies are car ried out to elucidate the phar macology and toxicology of a dr ug for mulation at v arious doses and applicable routes of administration in appropriate animal and in vitro models. 5 Information from the nonclinical studies is incor porated into the o verall assessment of human safety and effectiveness. In the early stages of drug development, nonclinical data are needed to pro vide a rationale for the proposed clinical use and to determine the target organ toxicity as well as a safe starting dose for firstin-human use. 6 During later stages of dr ug development, additional nonclinical data are needed to justify the safe administration of the dr ug to lar ger numbers and broader populations of subjects in clinical trials ultimatel y leading to the approval of the drug.

Disclaimer: The views expressed in this chapter are those of the authors and do not necessaril y reflect the views or policies of the U.S. Food and Drug Administration. In addition, no official support or endorsement by this agency is intended or should be infer red. * Authors’ names arranged alphabetically; all authors contributed equally to the chapter. 589

590

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

The nonclinical de velopmental strate gy for an investigational dr ug should be customized based upon sound scientif ic principles, intended clinical use, and the drug’s unique chemistry including those of its components, metabolites, and impurities. The special characteristics of an imaging agent that can lead to a more focused nonclinical safety e valuation include its dose or cold mass, its route of administration, its frequenc y of use, and for a diagnostic radiophar maceutical, its biolo gical, ph ysical, and ef fective half-li ves. In par t, the types of nonclinical studies necessar y to suppor t mark eting appro val of an investigational dr ug depend upon the phase of the dr ug’s development, the infor mation available about the dr ug, its pharmacologic class, its proposed use, and the intended patient population.4 Nonclinical studies consist of phar macology and safety studies. Phar macologic studies are car ried out to determine the pharmacological properties of a dr ug associated with its intended clinical use. F or medical imaging drugs, this usuall y means providing data that (1) suppor t the proposed mechanism of selecti ve localization to the anatomic or functional target sites and (2) justify administration to humans b y sho wing that the agent selecti vely distributes to the target site in an in vi vo model. Nonclinical safety studies are car ried out to predict the potential adverse effects in humans, including effects that may arise from an extension of the drug’s pharmacodynamic effects or unpredicted to xicologic features. 7,8 In general, neither pharmacodynamic nor toxicologic effects should occur at diagnostic doses for medical imaging dr ugs. Therefore, safety pharmacology studies generally determine if clinically relevant effects are possible at doses within the diagnostic range. The objecti ves are (1) to identify the undesirable phar macodynamic effects relevant to human safety, (2) to e valuate the adv erse phar macodynamic and/or pathoph ysiologic ef fects obser ved in to xicology and/or clinical studies, and (3) to study the mechanism of the adv erse phar macodynamic ef fects obser ved or suspected. The design of the nonclinical safety studies is in large part driven by the characteristics of the drug. For example, the design may need to focus upon specific organ toxicity if the dr ug is lik ely to e xhibit selective biodistribution to an organ, especially if that or gan is compromised in the target patient population. Nonclinical safety studies should be carried out using multiple dose levels extending beyond (e g, 100 times) the clinical dose to deter mine a dose-response relationship, identify or gan systems sensitive to the dr ug, and estab lish a no obser vable adv erse effect level (NOAEL) for clinically relevant effects. Safety pharmacology studies ma y assess ef fects on the

cardiovascular, central ner vous, respiratory, renal/urinary, or other systems. F or example, the cardio vascular safety studies are especially relevant for medical imaging agents used in imaging the hear t or intended for patients with potentially compromised cardio vascular function. In this situation, the safety studies ma y par ticularly assess the effects of the imaging agent on cardiac function as well as detect histopathologic changes or other signs of to xicity. Similarly, the central nervous system (CNS) safety studies are rele vant for dr ugs (e g, radiophar maceuticals) that localize in the brain. These types of dr ugs may localize through a variety of mechanisms, for example, the binding of the drug to specific receptors or through disr uptions in the blood-brain barrier. Adverse effects on the CNS can be assessed using biochemical, histologic, neurophysiologic, or behavioral tests. Most medical imaging drugs are usually administered infrequently (e g, for initial diagnosis or to monitor response to a therapeutic agent/disease pro gression). For such products, data from e xpanded acute to xicity studies are impor tant for estab lishing a meaningful mar gin of safety.4 Accordingly, adv erse e vents related to long-ter m use or due to dr ug accumulation are less lik ely to occur with these agents than with agents administered repeatedly to the patient. Therefore, the nonclinical development programs for such single-use products usually may omit longterm repeat-dose safety studies of three-month duration or longer. F or carcino genicity and reproducti ve to xicology studies, a sponsor may request a waiver. For medical imaging agents, repeat-dose to xicity studies of 14- to 28-day duration are usually carried out to evaluate the ef fects of e xaggerated dose re gimens or to mimic chronic clinical administration. Such studies ma y sho w effects not detected in clinical studies of small sample size, identify clinical effects to monitor in clinical trials, or reveal effects that might occur in sensiti ve indi viduals. Special toxicity studies may be needed if the str ucture/property of the drug or its conditions of use raises a concern, for example, in renal-impaired subjects, or if previous nonclinical or clinical findings of the product or of related products ha ve indicated special to xicologic concer ns. Such studies ma y include the use of animal models of renal impair ment and immunotoxicology studies. Most medical imaging dr ugs are administered intravenously, thus are a vailable systemicall y. Therefore, the similarities or dif ferences betw een human and animal pharmacokinetics would be based on parameters such as volume of distribution, maximum b lood concentration (Cmax), area under the concentration cur ve (AUC), organ distribution, routes and rates of elimination, degree of protein binding, and extent and array of metabolism.

Nonclinical Product Developmental Strategies, Safety Considerations. . .

The formulation used to establish safety margins in nonclinical studies should be similar to the for mulation that will be used in clinical trials and that is intended for clinical mark eting. Any dif ferences betw een the formulations used in the clinical trials and the nonclinical studies should be specif ied so that any effect on the adequacy of the nonclinical studies can be deter mined. Bridging studies ma y be helpful w hen changes in the formulation have the potential to change the phar macokinetics, the phar macodynamics, or the safety characteristics of the dr ug, including manuf acturing changes for biologic purposes. When it is infeasible or impractical to administer the intended clinical for mulation to animals in multiples of the clinical dose (eg, the volume to be injected into animals ma y be e xcessive), alter native strate gies can be used , such as di viding the dail y mass dose within the 24-hour period or by using a more concentrated formulation.

TIMING OF STUDIES Nonclinical studies should be timed so that the y help facilitate the timel y conduct of clinical trials including appropriate clinical safety monitoring based on f indings in nonclinical studies and to reduce the unnecessar y use of animals and other resources.5 In general, safety pharmacology information on major organs and organ systems that the drug is intended to visualize, to xicokinetics/pharmacokinetics, expanded acute single dose (with reco very), and in vitro genoto xicity are e valuated before Phase I. When warranted, special to xicology e valuations before Phase I investigation can include route ir ritancy, misadministration/extravasation, and b lood compatibility . Repeat-dose toxicity studies and the complete standard batter y of genotoxicity studies (test for gene mutation in bacteria, in vitro assessment of chromosomal damage using mammalian cells or an in vitro mouse l ymphoma tk assa y, and an in vi vo test for chromosomal damage using rodent hematopoietic cells) are usuall y required before Phase II. When there is cause for concern, based on molecular structure, biodistribution pattern, drug class, or clinical/nonclinical signal, an immunoto xicologic e valuation ma y be needed. The recommended timing of the nonclinical studies for medical imaging drugs is summarized in Table 1. Drug developers are encouraged to tak e advantage of the e xisting FD A re gulations that allo w fle xibility in the amount and the type of data required for earl y-phase clinical development. For example, the exploratory investigational ne w dr ug (IND) mechanism allo ws sponsors to initiate clinical trials of limited scale with reduced nonclinical requirements. 9 The e xploratory IND mechanism is

591

particularly suited for imaging trials that e valuate human radiation dosimetry and biodistribution parameters. Among its man y adv antages is pro viding the potential to test multiple drug candidates at the same time and at a fraction of the cost of the traditional IND path.

SAFETY CONSIDERATIONS AND TOXICITY PROFILES Toxicity prof iles dif fer considerab ly between classes of medical imaging agents. In this section, w e will discuss the to xicity prof iles of se veral cate gories of medical imaging agents. A class-related to xicity effect has been identified for some agents, for e xample, Gd-based contrast agents (GBCAs). For others, for example, PET diagnostic radiopharmaceuticals, there has been no clear class toxicity prof ile reported, perhaps due to the v ery heterogeneous nature of these dr ugs. These dr ugs commonl y differ in tar geted receptors, biochemical and ph ysiological pathways, or elimination mechanisms.

GBCAS GBCAs are widel y used as contrast agents for MRI because of the magnetic proper ty of Gd. Gadopentetate dime glumine (Magne vist®),10 gadodiamide (Omniscan®),11 gadoversetamide (OptiMARK ®),12 gadoteridol (ProHance ®),13,14 gadobenate dime glumine (MultiHance®),14 gadoxetate disodium (Eo vist®)15 and gadofosveset trisodium (Vasovist®) are approved by U.S. FDA while gadobutrol (Gado vist®) and gadoterate me glumine (Dotarem ®), are mark eted in other countries. Clinical safety considerations for GBCAs relate to their dose, osmolality , potential for transmetalation of complexes of Gd, and the potential effect of tissue or cellular accumulation on or gan function, par ticularly if intended to image a diseased human or gan system. The clinical and nonclinical safety prof iles for GBCAs are generall y similar, although each member of this class has unique features that ma y some what dif fer depending upon the dose of the agent and cer tain molecular features. Summarized below are safety f indings generall y considered as typical for the class of GBCAs.

RENAL TOXICITY Proximal tubular v acuolation, the c ytoplasmic v acuole formation in renal pro ximal tubular epithelial cells observed in histopatholo gic e xamination, has been reported in animals administered single or multiple doses

592

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 1. TIMING OF NONCLINICAL STUDIES FOR NONBIOLOGICAL PRODUCTS SUBMITTED TO AN INVESTIGATIONAL NEW DRUG APPLICATION Study Type Safety pharmacology Toxicokinetic pharmacokinetic

Before Phase 1

Before Phase 2

Major organs and organ systems the drug is intended to visualize See International Conference on Harmonization (ICH) guidelines Expanded acute single dose

Expanded singledose toxicity Short-term (2 to 4 weeks) repeat-dose toxicity Special toxicology Conduct as necessary based on route-irritancy, blood compatibility, protein flocculation, misadministration, extravasation Radiation If applicable dosimetry Genotoxicity In vitro

Before Phase 3

Before NDA

—

—

—

—

—

—

—

—

—

—

—

—

—

—

—

—

—

Repeat-dose toxicity

Immunotoxicity

—

Complete standard battery —

Reproductive and developmental toxicity Drug interaction Other based on data results

—

—

—

—

of GBCAs. In these studies, the vacuolation was dose- and time-dependent and partially or completely reversible after a recovery period follo wing treatment cessation. 16,17 Vacuolation appeared to be independent of the osmolality and the viscosity of the contrast agents.17 Transient increases in urine v olume, albumin, br ush border enzymes, such as γ-glutamyltransferase and alkaline phosphatase, and c ytoplasmic enzymes, such as lactate deh ydrogenase and alanine aminopeptidase, w ere obser ved in a rat study w hen gadopentetate dime glumine or gadodiamide w as intravenously administered at 4.59 mmol/kg as a single bolus dose. The highest increase w as noted during the f irst 2 hours after dosing, and the levels were similar to those of saline-treated rats 2 da ys after dosing. The increases in urine volume and albumin le vels appeared to be osmolality-dependent. No change in the l ysosomal tubule enzymes, such as N-acetyl- β-D-glucosaminidase and

May be needed based on molecular structure, biodistribution pattern, class concern, or clinical or nonclinical signal Needed or waiver obtained —

—

—

As needed As needed

glucose, was noted. 17 In a study comparing GBCAs and iodinated contrast agents in an ischemic porcine model, renal histomorphologic changes, including proximal tubular and glomer ular necrosis, hemor rhage/congestion, and protein-filled tubules in the cor tex and medulla, w ere observed.18 The de gree of necrosis and hemor rhage/congestion was related to the de gree of impair ment of renal function but in versely related to v acuolation and tubular protein filling.

NEPHROGENIC SYSTEMIC FIBROSIS Nephrogenic systemic f ibrosis (NSF) is a relati vely recently identif ied clinical disorder that manifests with fibrosis of the skin and other organs. NSF has been associated with GBCAs administration to patients with se verely impaired renal function although other co-morbidities,

Nonclinical Product Developmental Strategies, Safety Considerations. . .

such as thrombosis, endothelial injur y, and metabolic acidosis, have also been repor ted in many patients with NSF. NSF was f irst described in the medical literature in 2000, the f irst case is generall y thought to ha ve occur red in 1997.19 The clinical presentation of NSF generally consists of thickening and discoloring of the skin with pr uritus and sharp pain in affected areas, primarily on the limbs and the trunk. The systemic f ibrosis often resulted in contractures and may cause organ functional impair ment, such as lung or heart abnormalities. Extremity s welling is often noted. Common histologic findings of the af fected skin included the presence of elastic f ibers and collagen bands, numerous CD34/procollagen-I dual-positive spindle cells (f ibrocytes), numerous f actor XIIIa-positi ve cells and occasionally CD68-positi ve cells (macrophages), and mucin. Hemosiderin and calcium w ere sometimes noted while ossif ication occur red in rare cases. The etiology of NSF is still under study .20 However, the administration of GBCAs to patients with se vere renal impair ment has been temporally associated with NSF , and GBCAs administration has been identif ied as one of the most common features among these patients. In the repor ted cases, NSF generally appeared within a fe w days to a fe w months of administration of GBCAs and w as commonl y associated with the use of higher doses of the GBCAs (e g, when the agent w as used for magnetic resonance ar teriography) or when multiple doses were administered in a relatively short period. Severe renal impair ment, or clinical setting associated with severe renal impairment, has also been a hallmark underlying this condition. In some patients with NSF , Gd, associated with calcium and phosphorous, w as detected in skin biopsy samples and bones and was noted to remain for as long as 3 years after the last administration of a GBCA.21 The most widel y accepted h ypothesis relates to the de-chelation of less stable GBCAs, leading to the release of free Gd3+.22 The free Gd3+ may attract CD34+, CD45+, and procollagen+ circulating fibrocytes via the release of chemokines. This inflammator y reaction is thought to culminate in f ibrosis. When GBCAs remain inside the body for a long period , as may occur in the setting of severe renal f ailure, the risk of de-chelation and subsequently NSF ma y uniquely increase. A bo xed w arning has been added to the labeling for GBCAs to identify the increased risk for NSF in patients with acute or chronic severe renal insuf ficiency (glomer ular f iltration rate < 30 mL/min/1.73 m2) or acute renal insufficiency of any severity due to hepato-renal syndrome or in the perioperative liver transplantation period. 23,24 A few animal studies ha ve been car ried out to study the role and the mechanism of GBCAs in NSF. A recently published study 25 suggested that Gd pla yed a role in

593

GBCA-induced skin lesions in rats. Following the administration of gadodiamide (2.5 mmol/kg, 5 times/week for 4 weeks, standard formulation with 5% caldiamide, a drug intended to chelate free Gd) four of six rats developed skin lesions, which consisted of epidermal ulceration and acanthosis, der mo-epidermal clefts, minimal-to-slight der mal fibrosis, and increased der mal inf iltration of dif ferent cells including CD34-positive cells. These lesions shared similar characteristics with those seen in patients with NSF. No animal de veloped skin lesions follo wing the administration of caldiamide (0.5 mmol/kg); chelation appeared to alleviate GBCA-induced skin lesions. All animals (n = 6) developed skin lesions, from treatment day 8 onwards, follo wing the administration of 2.5 mmol/kg gadodiamide without caldiamide. The treatment w as discontinued after the 10th injection because of the se verity of the skin lesions. In addition, the study suggested that the ability of GBCAs to induce skin lesions may be agentspecific. Four of six animals de veloped the skin lesions following the administration of 2.5 mmol/kg gadodiamide, w hile none of the animals de veloped the skin lesions after the administration of gadopentetate dimeglumine at the same dose. Gd w as detected in skin, femur , and liver in rats treated with GBCAs. Gd concentrations in skin were much higher in animals given gadodiamide than in animals gi ven gadopentetate dime glumine w hile the differences were less robust in li ver and femur; in these studies, the anal ytical methods used could not dif ferentiate between the chelated and the free Gd that is central to the putati ve role of free Gd in NSF . The dif ferential effect w as assumed to result from the dif ferent physiochemical proper ties with gadopentetate dimeglumine possessing higher ther modynamic and conditional stability than gadodiamide. In contrast to these nonclinical data, clinical data ha ve pro ven much more challenging when attempts have been made to distinguish differential risks for NSF based upon the unique characteristics of each member of the GBCA drug class. The rarity of NSF has par ticularly limited the ability to reach conclusions regarding each GBCA’s unique risks.

REPRODUCTIVE AND DEVELOPMENTAL TOXICITIES Reproductive and developmental toxicities of GBCAs have been shown in animal studies. 23,24 A relatively consistent finding was reproductive toxicities in male animals. Dosedependent reduction and de generation of sper matocytes, reduced weight of testes and epididymides, and v acuolation in testes w ere frequentl y obser ved in repeat-dose toxicity studies in rats. F or example, vacuolation in testes

594

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

and abnor mal sper matogenic cells w ere obser ved w hen gadobenate dime glumine was intravenously administered to male rats at 3 mmol/kg/d (5 times the human dose based on body surface area [BSA]) for 28 days. The effects were not re versible follo wing a 28-da y reco very period. The effects were not reported in dog (28 days dosing) and monkey (14 days dosing) studies at doses up to about 11 and 10 times the human dose based on BSA, respecti vely. To illustrate with another e xample, gado versetamide, when administered to rats at 2.0 mmol/kg/d (4 times the human dose based on BSA) for 7 weeks, was shown to produce ir reversible reduction and de generation of sper matocytes and impaired male fer tility. In a separate 28-da y repeat-dose study in rats, gado versetamide administered at intra venous doses of 3.0 mmol/kg/d (6 times the human dose based on BSA) produced ir reversible reduction in male reproducti ve organ weights, degeneration of the germinal epithelium of the testes, presence of ger m cells in the epididymides, and reduced sperm count. These effects were not observed at 0.5 or 0.6 mmol/kg/d (equi valent to the human dose based on BSA) in the fer tility or repeat-dose studies or in a single dose study at 0.5 to 15 mmol/kg (1 to 25 times the human dose based on BSA). These effects were not observed in similar studies car ried out in do gs. The potential risks in humans are unkno wn. Moreo ver, for animals that recei ved repeated doses of GBCAs, their overall e xposure w as signif icantly higher than that achieved with the standard single dose administered to humans. GBCAs cross the placenta in animals and are thought capable of crossing the placenta in humans. De velopmental toxicities have been sho wn by animal studies. In general, GBCAs induced embr yo-fetal to xicities resulting in preimplantation and/or postimplantation losses, intrauterine deaths, and absorption when pregnant rats or rabbits were given GBCAs during the gestation phase at doses equal to or g reater than 3.2 times the human doses based on BSA comparison. In addition, some GBCAs induced terato genic effects w hen gi ven during the or ganogenesis phase in rats and rabbits. For example, gadobenate dimeglumine administered at 2 mmol/kg/d (6 times the human dose based on BSA) during or ganogenesis (gestation da ys 6 to 18) was teratogenic in rabbits with microphthalmia/ small eye and/or focal retinal fold obser ved in fetuses. Gadoversetamide caused forelimb fle xures and cardiovascular changes in fetuses from female rabbits gi ven 0.4 and 1.6 mmol/kg/d (1 and 4 times the human dose based on BSA, respecti vely) during organogenesis. The cardiovascular changes consisted of malformed thoracic

arteries, septal defect, and abnor mal v entricle. These effects were not obser ved at 0.1 mmol/kg/d (0.3 times the human dose based on BSA). Mater nal toxicity was not observed at any doses. On the basis of the risks identified in animals, the market labels for GBCAs emphasize that these agents are to be used during pre gnancy only if the potential benef it to the mother justifies the potential risk to the fetus.

INJECTION SITE REACTIONS A comparative local toxicity study was carried out in mice using gadopentetate dime glumine, gadoteridol, gadodiamide, and gado versetamide.26 Each mouse recei ved a subcutaneous injection in the hindlimb of 0.3 mL of the contrast agent (0.5 mol/L) or saline. The severity of local necrosis w as as follo ws: gadopentetate dime glumine > gadoversetamide > gadodiamide > gadoteridol > saline. The severity appeared to cor relate with osmolality: 1960, 1110, 789, and 630 mOsmol/kg, respecti vely. Similarl y, perivenous injection of gadobenate dime glumine to rabbits provoked more se vere local reactions than the intravenous injection. 23,24 Local reactions, including eschar and necrosis, persisted through da y 8 post peri venous injection of gadobenate dime glumine. Local intolerance reactions, including moderate interstitial hemor rhage, edema, and focal muscle f iber necrosis, w ere obser ved after intramuscular administration of gado xetate disodium.15 The general patter n of injection site reaction to GBCAs in animals contributed to the clinical recommendations to use special precautions w hen intra venously administering GBCAs to a void inadv ertent local extravasation.

ULTRASOUND CONTRAST AGENTS Ultrasound contrast agents (microbubb les) are gasfilled microspheres administered intra venously to improve images obtained during echocardio graphic studies. Microbubbles have a high degree of echogenicity, the ability to reflect the ultrasound wave, which generally differs from that of human tissue. This difference in reflectivity improves the ultrasound signal b y creating a contrast betw een the gas core and the cardiac anatomic str uctures. P erflutren Lipid Microsphere (Definity®) and Perflutren Protein-Type A Microspheres (Optison™) are the tw o microb ubble agents cur rently approved in the United States. Both products contain fluoro-propane gas. They dif fer in the microbubb le shell; albumin for Optison™ and lipid for Def inity®.

Nonclinical Product Developmental Strategies, Safety Considerations. . .

A third microbubb le product, SonoV ue®, contains sulfurhexafluoride gas in a lipid shell and is appro ved in Europe.27 These ultrasound contrast agents are used in echocardiography and assist in the diagnosis of cardiac conditions. Ho wever, ultrasound contrast agents are undergoing clinical trials for use in other conditions and settings, such as the detection of stenoses within the peripheral vasculature and the detection of li ver abnormalities. Targeted microbubb les for use as molecular imaging agents in cancer applications are also under active study.28,29 The uniqueness of the ph ysical and chemical properties of the ultrasound contrast agents, the technicalities of the ultrasound f ield, and the potential for modulation of physicochemical properties of the ultrasound contrast agents by the ultrasound f ield present special challenges and safety considerations in the nonclinical development of these agents. Some of the unique safety considerations include e valuation of the susceptibility of microbubbles to the range of ultrasound mechanical index (MI), a measure of the intensity of the acoustic pressure output of the ultrasound system commonly used in clinical settings, and ho w the MI af fects the fragility of the microbubb les. Because of their small sizes (1 to 10 µm) and particulate features, animal studies are v ery important to help understand the beha vior of the microbubbles in the micro vasculature. Animal and in vitro studies are used to evaluate microbubbles’ size and deformability, percentage of capillaries obstr ucted, and potential for agg regation in appropriate animal models after intra venous and intra-ar terial administration. The intra-arterial route of administration is important to help in risk identif ication because the intra-ar terial route models the clinical risks for patients with right-to-left, bidirectional cardiac shunt. In these patients, microbubbles can bypass the pulmonary-filtering mechanisms and directly enter the arterial circulation.30,31 To deter mine the risk of to xicity to patients with compromised pulmonar y v asculature, the cardiopulmonary effects of the microbubb les should generally be studied in an animal model of chronicall y compromised pulmonary circulatory disease. Pulmonary artery hemodynamic changes potentially induced by ultrasound contrast agents ma y be handled reasonab ly well in health y humans whereas these changes may present greater risks for patients with compromised lung function or other causes for pulmonary hypertension. Additionally, animal models with compromised cardiovascular function, such as atherosclerotic and hypertensive models, may be relevant to assess properl y the cardio vascular ef fects of microbubbles.

595

HEMODYNAMIC EFFECTS OF MICROBUBBLES The effects of microbubbles with or without ultrasound exposure on cardiopulmonary function have been extensively e valuated in v arious species, including rat, do g, monkey, and pig. Depending on the species, microb ubbles at clinically equivalent or higher doses induce transient but severe hemodynamic changes. The patterns of these reactions ha ve been proposed as models for rare but serious hypotensive reactions seen in humans. 27 In a cardiopulmonar y safety study in anesthetized pigs, acute intra venous administration of SonoV ue® and another microbubb le product induced dose-dependent marked hemodynamic alterations. 27 The alterations w ere characterized b y increase in pulmonar y ar tery pressure (~three-fold), decrease in systemic blood pressure (~50%), increase in heart rate, decrease in lung compliance and partial pressure of o xygen, increase in airw ay resistance and electrocardiogram abnor malities. The hemodynamic changes occurred within 1 to 2 minutes of administration of microbubbles at clinically relevant doses and returned to baseline v alues b y 10 minutes postdosing. The intensity and the sequence of hemodynamic alterations were similar for both products and w ere not potentiated b y an increase in the intensity of ultrasound e xposure (MI: 0.25 and 1.6). The hemodynamic alterations obser ved in pigs w ere proposed to result from thrombo xane release induced b y microbubble administration since there w ere transient increases in the level of a thromboxane metabolite, TXB2, and the ef fects were completely blocked by pretreatment with aspirin or indomethacin kno wn to inhibit the cyclooxygenase pathway involved in thromboxane synthesis. Similar adv erse hemodynamic ef fects in pigs w ere reported for Definity®.32 SonoVue® and another ultrasound contrast product produced transient but signif icant systemic hypotension in rats.27 However, in rats, the ef fects were less mark ed and occurred at doses 10-fold higher than the human equivalent dose. In contrast to pigs, the adv erse hemodynamic effects in rats were not blocked by aspirin but by platelet activating factor (PAF) antagonist ABT-491, suggesting that PAF, not thromboxane, pla ys a major role in the mechanism of microbubble-induced hypotension in rats. In dogs, transient systemic h ypotension w as induced b y administration of SonoVue® at high doses (8 times maximum human dose [MHD]), whereas no cardiovascular effects were observed in monkeys at doses up to 17 times the human dose. In a cardiovascular safety study ,30 Definity®, administered to anesthetized dogs at 27-times MHD, caused a transient but significant increase in respirator y rate (300%) and

596

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

pulmonary ar tery pressure (188%), w hereas the ar terial pressure and m yocardial contractility decreased b y 44% and 55%, respecti vely.30 The NOAEL for this ef fect w as 14-times MHD. One of the four do gs died sho wing signs consistent with the cardiopulmonary collapse. In an acute to xicity study in rats, Def inity ® (8-times MHD) caused lethality and clinical signs, such as abnor mal respiration and increased hear t rate, observed before death, which were consistent with cardiopulmonary effects.30 In monkeys, acute administration of Def inity® (48-times MHD) produced abnor mal electrocardiographic changes, including ST-T segment depression within 1 minute of administration. This was followed b y v entricular e xtrasystoles, v entricular tachycardia, f irst degree and complete atrio ventricular block, and transient de velopment of right bundle branch b lock. The ST -T depression be gan to retur n toward baseline betw een 8 and 10 minutes after the beginning of drug infusion. 30 Among the tested species, the pig w as highly sensitive to the adv erse hemodynamic ef fects induced b y microbubbles. The species rank order of sensiti vity for adverse hemodynamic ef fects appeared to be pigs > dogs > rats > monk eys. It has been postulated that the sensitivity of pigs ma y be because of high concentration of pulmonary intravascular macrophages relative to other species including humans. These macrophages react to particulates, such as liposomes, and micellar lipids in the bloodstream by activation of phagocytosis and thromboxane release, resulting in transient but se vere pulmonar y hypertension and secondar y systemic h ypotension. Pigs are kno wn to produce se vere reactions to injections of particulates.33 Therefore, the pig ma y not be a good model to assess the quantitative risk but may be an appropriate species to assess the qualitati ve risk for microbubbles to induce adverse hemodynamic changes.

EFFECTS OF MICROBUBBLES ON MICROVASCULATURE CIRCULATION Microbubbles ma y ha ve the potential to obstr uct the microcirculation. If microbubbles pass the pulmonary circulation, the y are systemicall y a vailable and ma y be trapped in other microvessels. Therefore, it is important to evaluate the behavior of microbubbles in ex vivo preparations, such as the rat spinotrapezius muscle microcirculation model and the hamster cheek pouch model. In these models, the flo w of fluorescent microbubb les and their entrapment/retention within the circulation is obser ved using intravital microscopy.

Intra-arterial administration of Def inity®, SonoVue®, and Optison™ into the rat spinotrapezius muscle microcirculation sho wed that a por tion of microbubbles w as entrapped within small ar terioles especially at branch points and in capillaries.27,30 Retention of bubb les of size smaller than 5 microns w as mainly because of their sticking to the endothelium. Some of the lar ge-sized bubb les (>5 microns) w ere entrapped in the capillaries with subsequent defor mation (elongation). The bubb les e ventually reduced in size and disappeared possibly because of the dissolution of the gas within the microbubb les. No coalescence or aggregation of bubb les w as obser ved. The transient occlusion ma y adv ersely af fect the cardiopulmonar y function in patients with compromised pulmonar y microvasculature, such as patients with chronic lung disease. In addition, systemic embolization may present special risks for patients with right-to-left or bidirectional cardiac shunt where microbubbles can bypass the pulmonary-filtering mechanisms and directly enter into the arterial circulation.

EFFECTS OF MICROBUBBLES IN ANIMAL MODELS OF PULMONARY HYPERTENSION Definity®, Optison™, and SonoV ue® have been e valuated in a do g model of acute pulmonar y h ypertension.27,30 In this model, moderate (+10 to 15 mm Hg) to severe (+30 mm Hg) pulmonar y h ypertension w as induced b y administration of microspheres (Sephade x glass beads). Microbubb les at clinicall y rele vant and higher doses (5 to 10 times MHD) did not affect any of the hemodynamic parameters including pulmonar y arterial pressure. The relevancy of this acute do g pulmonary h ypertension model to humans with chronic pulmonary dysfunction is not kno wn. Ef fects of microbubbles in a chronically compromised pulmonary circulation disease model have not been studied.

EFFECTS ON ENDOTHELIAL INTEGRITY It is impor tant to e valuate the pathoph ysiologic consequences of the ultrasound-microbubb le combinations, in par ticular, the ef fects on v ascular endothelial integrity. Such studies ha ve been car ried out for Optison™,27 and no endothelial damage in jugular vein wall, aor tic w all, m yocardium, and kidne y w as noted when Optison™ was administered to dogs with concurrent exposure to ultrasound at high acoustic pressures (MI 0.8 and 1.8).

Nonclinical Product Developmental Strategies, Safety Considerations. . .

IN VITRO COMPLEMENT AND BASOPHIL ACTIVATION STUDIES SonoVue® and another ultrasound contrast agent w ere evaluated for their potential to acti vate the immune system.27 Both products dose-dependently activated the complement system as assessed b y the production of anaphylotoxins C3a and C5a after in vitro incubation with pig and human ser um. A low level of complement acti vation w as obser ved for both contrast agents at concentrations comparab le with the human imaging dose. There were no ef fects on basophil acti vation. Whether complement activation constituted an early triggering event in the hemodynamic adv erse reactions seen in some patients remains to be estab lished. Future clinical studies e valuating the effects of microbubbles on complement acti vation and other potential biomarkers of toxicity may be of value.

DIAGNOSTIC PHARMACEUTICALS Diagnostic radiophar maceuticals are radioacti ve dr ug or biological products that contain a radionuclide usuall y linked to a ligand or car rier. Typical ligands or car riers include lipids, carboh ydrates, nucleic acids, peptides, small proteins, or antibodies. Diagnostic radiopharmaceuticals are used in nuclear medicine modalities, such as planar imaging, SPECT, PET, or in combination with other radiation detection probes. The radiation dose, the type of radiation, and the phar macology and to xicology of the carrier or ligand are important determinants of the adverse reaction profile of a diagnostic radiophar maceutical. Dose-related adverse events are less lik ely to occur when the ligand or car rier is administered at the lo w end of the phar macological dose-response cur ve. It is also important that the radiation dose be as lo w as reasonably achievable thereby reducing the risk of long-ter m radiation-induced carcino genesis w hile allo wing for optimal imaging. The nonclinical requirements for diagnostic radiopharmaceuticals differ from those for conventional therapeutic dr ugs.4 Because diagnostic radiophar maceuticals are not intended for chronic use and the radiation risk is low, their nonclinical evaluation can be uniquely tailored to address the limited clinical use proposed for these products. To e xpedite the de velopment of diagnostic radiopharmaceuticals in the United States, the FD A has developed a few unique regulatory mechanisms, such as the e xploratory IND for in vestigational dr ug de velopment or the Radioacti ve Dr ug Research Committee (RDRC) pathw ay for research studies of radiophar maceuticals. Regardless of the re gulatory considerations or

597

clinical de velopment pathw ay, w hen dealing with diagnostic radiopharmaceutical safety concerns, it is pertinent to consider both the mass dose and the radioacti ve dose.

MASS DOSE Although the mass dose of the car rier/ligand is generally small and is often refer red to as a tracer dose or “microdose,” some diagnostic radiophar maceuticals contain ligands that may have pharmacologic or toxicologic effects.1 In addition, many of the car rier moieties have agonist or antagonist acti vity on receptors, transporters, or metabolic processes, for w hich the y sho w high affinity and/or selectivity at nanomolar concentrations.34–37 Therefore, safety concer ns re garding the administered dose of both the radionuclide and its carrier must be considered. Important factors to consider when using a biological product include its ph ysical parameters (size and str ucture), route of administration, tissue distribution, immunogenicity, and its biolo gical activity/pharmacodynamic effects.38,39 For example, biological products ma y have limited e xtravascular tissue distrib ution, w hich in addition to their size and surface charge, may prevent targeting of the diagnostic radiophar maceutical to or gans and areas of tissues with limited perfusion. Nonclinical studies should include phar macokinetic and phar macodynamic characterization of the biolo gical product, as well as the tar get organ and tissue distribution, to understand potential dose-related to xicities and to sho w the presence of the product-specif ic tar get in the par ticular organ or tissue of interest. Many protein products sho w species and/or tissue specificity, therefore, safety e valuation programs should include the use of rele vant species. A relevant species is one in which the test material is pharmacologically active because of the expression of the receptor or an epitope (in the case of monoclonal antibodies). A v ariety of techniques (eg, immunochemical or functional tests) can be used to identify a relevant species. Immunogenicity is another unique issue in safety evaluation of protein products. Many protein products intended for humans are immuno genic in animals. Therefore, measurement of antibodies associated with administration of these types of products should be performed when carrying out repeated dose toxicity studies to aid in the interpretation of these studies. 40 Antibody responses should be characterized (eg, titer, number of responding animals, neutralizing or nonneutralizing), and their appearance should be cor related with an y phar macologic and/or to xicologic changes. Specif ically, the effects of antibody for mation on

598

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

pharmacokinetic/pharmacodynamic parameters, incidence and/or severity of adv erse effects, complement acti vation, or the emergence of new toxic effects should be considered when interpreting the data. Attention should also be paid to the evaluation of possible pathologic changes related to the immune complex formation and deposition. Other biolo gical products (e g, antibodies, g rowth factors, cytokines) used as car riers for radiophar maceuticals ma y ha ve safety concer ns in addition to those related to their anticipated phar macologic ef fects. For example, biolo gical products ma y cause infusion/hypersensitivity reactions follo wing dosing or result in an immuno genic response that ma y cross-react and neutralize an endo genous, homolo gous protein. Other undesired safety concer ns may include antibodydependent, cell-mediated cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC) mediated b y the Fc re gion of a full y human or humanized monoclonal antibodies of IgG subclasses. Fur thermore, potential agonistic or antagonistic function of the biological product on the endo genous tar get ma y produce unintended toxicity (e g, c ytokine release syndrome follo wing agonistic stimulation of T-cell receptors). These issues should be addressed completely in the nonclinical safety studies before use of the biologic radiopharmaceutical in humans.

ABSORBED RADIATION DOSE Because the radiation doses are lo w compared with the doses administered with therapeutic radiopharmaceuticals, acute radiation toxicity is not expected to occur with diagnostic agents. However, estimation of the potential for radiation toxicity is essential during the preclinical and clinical development of all radiophar maceuticals. Radiation dosimetry studies are usually carried out in rodents or primates before f irst use in humans to estimate the radiation doses potentially absorbed by the organs of patients and to determine the target organs of radiation. 40 Dosimetry data may re veal une xpected distribution of the radiophar maceutical. For example, uptake by the eye or testes would be particularly useful to kno w about since this signal in animals ma y w arrant specif ic e ye or testicular precautions when the radiopharmaceutical is studied in humans.

PHARMACOLOGIC/ TOXICOLOGIC PROFILE OF RADIOPHARMACEUTICALS Identifying the pharmacologic and toxicologic profile of a diagnostic radiophar maceutical ma y require specif ic evaluation using unique in vi vo and/or in vitro models,

depending on the chemistry of the drug and the intended use. It is also impor tant to address concer ns of nonspecific dr ug localization due to uptak e at multiple sites. For e xample, meta-iodobenzylguanidine radiolabeled with 131I (I-131 mIBG) 41 is used in the detection and localization of neurob lastomas and pheochromoc ytomas. The to xicologic e valuation of cold (nonradiolabeled) mIBG in mice, rats, and do gs identif ied the cardiovascular system as one of the tar gets for drug toxicity in e xposed animals. Clinical signs detected in animals treated with mIBG suggested transient, postsynaptic adrenergic-like effects. In some animals, a single injection of mIBG increased diastolic and systolic blood pressure and decreased hear t rate. For some peptides, the ligand’s agonist or antagonist activity at a receptor ma y not translate into phar macologic effects. For example, indium In-111 pentetreotide (OctreoScan™)42 binds competiti vely to somatostatin receptors and is used for localizing tissues rich in somatostatin receptors, such as cer tain types of tumors. However, the somatostatin-lik e effects (eg, inhibition of growth hor mone release) ha ve not been obser ved at diagnostic doses. On the other hand , some peptides may display phar macological ef fects at clinicall y rele vant doses but raise little to no to xicological concerns stemming from the pharmacological ef fects. An e xample is technetium Tc-99m Apcitide (AcuTect™)43 developed for the detection and the localization of acute v enous thrombosis. Apcitide is a synthetic peptide that binds to membrane gl ycoprotein GP-IIb/IIIa receptors on activated platelets that ma y lead to inhibition of platelet aggregation. The inhibition of platelet agg regation was shown in human platelets in vitro and in do gs treated ex vivo. In dogs administered 30 and 100 times the MHD of Apcitide, agg regatory responses to adenosine diphosphate (ADP) declined by 43% and 98%, respectively. No antiaggregatory ef fect w as noted in do gs administered the human equi valent diagnostic dose of AcuTect™ injection. Because no inter mediate dose w as e valuated between 1 and 30 times the MHD , it is unclear w hether the agg regatory responses could ha ve been obser ved at dose multiples much lower than 30 times the MHD. The U .S. FD A, in its literature re view for some PET products, notab ly 18F-fluorodeoxyglucose ( 18FDG), N-13 ammonia, 44 and 18FluoroDOPA ( 18FDOPA),45 examined whether there were reports of pharmacology or toxicology studies typicall y conducted to suppor t the safety of a New Drug Application (NDA). The literature sur vey did not identify an y study that examined the toxicity profile of FDOPA or that quantified impurities or excipients potentially present in clinical formulations. However, L-DOPA (usually administered with

Nonclinical Product Developmental Strategies, Safety Considerations. . .

a decarbo xylase inhibitor) is an appro ved dr ug product with more than 30 years of clinical use in the treatment of parkinsonism and related disorders. The side ef fects of L-DOPA are lar gely attributed to dopamine for med from its decarboxylation by aromatic L-amino acid decarbo xylase. The side ef fects include nausea, confusion, and hallucination especiall y in the elderl y and patients with preexisting co gnitive dysfunction. Actions of dopamine on the cardiovascular system include hypotension and cardiac ar rhythmia especiall y in patients with pree xisting conductance disturbance. These side ef fects of L-DOP A occur at much higher dose range than those anticipated for 18 FDOPA use in PET imaging studies. F or the 18FDOPA18 FDOPA usuall y PET procedure, a single dose of contains approximately 7 to 10 mg of the dr ug substance. The a vailability of e xtensive clinical e xperience with FDOPA with no repor ts of safety concer n, a safety database for L-DOPA, and the f act that 18FDOPA would be administered in a tracer quantity , ameliorated the lack of nonclinical data. Similar to the FDA’s review findings for 18FDOPA, no nonclinical study addressing the to xicology of N-13 ammonia was identified. However, the pathophysiology of excess ammonia is w ell described. The nor mal human blood ammonia concentration is generall y less than 35 µmol/L. F or the N-13 ammonia-PET procedure, the total amount of ammonia usually administered is less than 0.02 µmol/L and w as not e xpected to cause a signif icant alteration in the circulating level of ammonia. The clinical evidence suggests that N-13 ammonia as used in PET procedure is essentially nontoxic to humans. Although not all data typicall y required for phar macology/toxicology in support of an NDA was identified in the literature survey, the FDA concluded that the need for additional nonclinical data could be waived. With regard to [18F]FDG,46 no studies addressing its toxicology w ere identif ied in the FD A’s re view of the pub lished literature. The LD50 for FDG w as 600 mg/kg in mice and rats injected intraperitoneall y.47 In mice injected intraperitoneally with up to 1000 times the human dose b y w eight and do gs injected intravenously with up to 50 times the human dose, there was no abnor mality in brain, hear t, spleen, li ver, kidneys, or lungs. 48 FDG w as c ytotoxic in mouse lymphoma (L5178Y) cells in vitro at a concentration of 1 mg/mL. There were human experiences with 2-deoxyD-glucose (2DG). F asting patients w ho recei ved 50 mg/kg DG sho wed h ypoglycemic-like responses, such as s weating, hunger , and thirst, that w ere less severe and of shorter duration compared with the hypoglycemic response to insulin in the same patients. One

599

of these patients became comatose.49,50 In all the human studies, the doses producing adv erse effects are f ar in excess of that needed for PET studies. Moreover, extensive clinical experience with [ 18F]FDG, with no repor ts of safety-related concerns, ameliorated the lack of more comprehensive animal data. Therefore, the need for typical phar macology and to xicology studies w as w aived for this PET agent. These PET e xamples illustrate the range of considerations in the nonclinical and clinical development of PET and other radiopharmaceutical products. It also emphasizes the need for the developers of PET products to engage the FDA early on the nonclinical requirements for their products.

IODINATED CONTRAST MEDIA Iodinated contrast media (ICM) are used widel y in diagnostic radiography and CT.51,52 The iodine-car rier molecules in use as ICM possess v arying numbers of iodine atoms, are of varying hydrophilicity and lipophilicity, and confer v ariable le vels of osmolality w hen administered intravascularly. In general, ICM are most commonl y categorized as high-osmolar ionic monomers, lo w-osmolar ionic monomers, lo w-osmolar nonionic monomers, and iso-osmolar nonionic dimers. The physico-chemical characteristics influence the radio-contrast quality, safety, and tolerability of various ICM.53,54 The osmolality of high-osmolar ionic monomeric contrast media (HOCM) is generall y f ive to eight times higher than nor mal blood osmolality and has generall y resulted in a higher incidence of serious adv erse events compared with lo w-osmolar nonionic monomeric contrast media (LOCM). LOCM, such as metrizamide (Amipaque®), iohe xol (Omnipaque ®), and io versol (Optiray®), and iso-osmolar nonionic contrast media, such as iodixanol (V isipaque®), are sometimes cited as having improved safety prof iles compared with HOCM. One published review concluded that the clinical risk of severe adv erse reactions is about six times lo wer with LOCM than with HOCM; the re view suppor ted the results of animal studies, w hich had sho wn a tw o- to three-fold g reater safety mar gin for LOCM compounds than for HOCM compounds.

ADVERSE EFFECTS OF ICM Two main cate gories of acute adv erse reactions are generally associated with intravascular ICM administration. The first category, often referred to as “chemotoxicity,” is dependent on the ph ysicochemical properties of

600

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

the molecules: lipophilicity , osmolality , viscosity of infusion solution, and phar macologic proper ties, and these adverse events are mostl y predictable in patients. The second cate gory consists of a constellation of “idiosyncratic” adverse reactions based on poorly understood mechanisms that do not follo w a known paradigm but are usuall y ter med “anaphylactoid” or “h ypersensitivity,” although these reactions also ma y include unpredictable serious or fatal cardiopulmonary reactions.55,56

CARDIOVASCULAR AND PULMONARY EFFECTS ICM have been associated with rare but se vere adverse cardiac and pulmonary effects, such as ventricular fibrillation (VF), cardiac ar rest, se vere bronchospasm, hypotension, bradycardia, and depressed sinoatrial and atrioventricular nodal acti vity; underl ying cardiac and pulmonary disease are thought to e xacerbate the frequency and severity of these reactions. 57 Animal studies have shown that the frequenc y and the se verity of cardiopulmonary adverse events were lower with intra vascular administration of the nonionic medium, metrizamide, compared with ionic media. 58 In animal studies of ICM-induced VF, results showed that the supplementation of ionic contrast media with calcium resulted in moderation of HOCM-induced VF. Studies of the nonionic medium iodixanol, in animals and in isolated perfused hear ts, showed that the supplementation of the infusion solution with sodium and calcium reduced the incidences of VF and m yocardial depression.59,60 Studies of cardiac and hemodynamic tolerability of ICM in anesthetized rats sho wed that the use of low osmolar , especiall y isotonic, dimeric CM, sho w a clear benef it o ver HOCM re garding cardio vascular adverse effects.61 Despite these nonclinical f indings, the differential risks for the v arious CM are much more difficult to characterize in humans, perhaps because of the rarity of the reactions.

ICM-INDUCED NEPHROTOXICITY ICM-induced nephropathy is a well-recognized risk of ICM administration in high-risk patients. When the nephropathy occurs, it may progress to acute renal failure.51,53,54,57,62 The mechanisms of pathoph ysiology involved in ICM-induced nephropathy are thought to include altered rheolo gic properties, per turbation of renal hemodynamics, re gional hypoxia, auto- and paracrine factors (adenosine, endothelin, and reacti ve o xygen species [R OS]), and direct c ytotoxic effects.51,62,63 The dimeric ICM, such as ioxaglate (ioxaglate

meglumine sodium; He xabrix™) and iodixanol (V isipaque®), have higher viscosity than b lood and ma y e xert additional stress on renal circulatory systems and glomerular f iltration capacity.62 The effects on renal function and morphology of the nonionic dimeric ICM, iodixanol, w ere studied in rats, rabbits, and monk eys and compared with other ICM. Iodixanol af fected renal function to the same degree as did the nonionic monomeric and dimeric comparative media, but to a lesser de gree than the ionic monomers. In addition, iodixanol administration in Wistar rats resulted in decreased renal blood flow and lower blood pressure compared with iobitridol, a monomeric nonionic agent. This effect was long-lasting and was not alleviated by increasing the hydration rate.64 Studies using proximal tubular cells showed impaired cell proliferation, vacuolization, perturbation of mitochondrial enzyme activity, and mitochondrial membrane potential.62 Dimeric iso-osmolar contrast media molecules, such as iodixanol, had a g reater potential for direct c ytotoxic effects in a renal pro ximal tubular cell line (LLC-PK1) than did monomeric contrast media molecules. 65 In addition, the degree of renal pro ximal tubular cell v acuolation induced by iodixanol seems to be species-dependent, being less pronounced and more quickl y re versed in monk eys than in rats. 66 As pre viously noted, clinical cor relates of differential risks for each CM are dif ficult to conf irm despite the extent of the animal data.

NEUROTOXIC EFFECTS OF ICM Entry of intravascularly administered contrast agents into the CNS is nor mally limited but ma y be increased b y osmotic opening of the b lood-brain bar rier with cerebral arteriography or arch aortography. Most neurotoxic effects are thought to represent direct effects of the contrast agent on brain or spinal cord. Adverse effects with arteriography include seizures, transient cortical blindness, brain edema, and spinal cord injur y. Seizures ma y occur with intravenous administration, especiall y in patients with brain tumors or other processes disr upting the blood-brain barrier. The most common adv erse ef fects obser ved with myelographic agents include seizures and transient encephalopathy with metrizamide. 67

IDIOSYNCRATIC TOXICITIES AND ANAPHYLACTOID ADVERSE EVENTS A primar y cate gory of reco gnized, but poorl y understood, safety concerns for ICM consists of idiosyncratic serious adv erse e vents. While the clinical e xperience with ICM is e xtensive, the mechanisms underl ying

Nonclinical Product Developmental Strategies, Safety Considerations. . .

idiosyncratic adv erse e vents, such as anaph ylactoid reactions, are not well understood. True anaphylaxis has not been established in most cases. A central hypothesis in the mechanism of anaph ylactoid reactions is acti vation of basophils and mast cells by ICM, resulting in the release of histamine and production of a cascade of inflammatory mediators including leuk otrienes, prostaglandins, enzymes, and a v ariety of c ytokines.68 In both clinical and nonclinical studies, results ha ve been inconsistent between investigators.68 Currently, no well-characterized animal models sho w reproducib le anaphylactoid responses to ICM administration.

NANOPARTICLES The FD A Guidance for Industr y; De veloping Medical Imaging Drugs and Biological Products noted that as technology advances, new products may emerge that do not fit into traditional cate gories (eg, agents for optical imaging, magnetic resonance spectroscop y, combined contrasts and functional imaging). It w as hoped that the general principles of the guidance could appl y to these ne w diagnostic products. Cer tain aspects of nanopar ticles research and some of its emerging imaging modalities aptly fit this anticipation. There has been a gradual development of nanoparticles for use in medical imaging as a ne w innovation in diagnosis and treatment. 69 Several nanosystems including quantum dots, liposomes, gold nanopar ticles, immunotargeted nanoshells, silica nanopar ticles, and magnetic nanocrystals have been studied for imaging. 69–71 However, the identification of factors that can predict toxicity, permit target screening, and help in emer gence of safer nanoparticles with adequate consideration accorded to structure-toxicity infor mation remains a challenge. 72 Therefore, the issues of obtaining adequate information on the mechanism of toxicity of nanoparticles, reaction with the internal body environment, and the contribution of the par ticle size to safety actively engage the attention of toxicologists. Nanotechnology has pro vided some ne w options for medical imaging especiall y in cancer diagnosis; ho wever, the potential adv erse human health ef fects resulting from exposure to nanopar ticles are of special safety concer n in light of the ine xperience with the products. The potential toxicity of nanopar ticles is af fected b y biodistribution including mo vement through tissues, phagoc ytosis, opsonization, and endocytosis of nanoparticles.73 Oxidative stress inflammatory reactions have been reported for many nanoparticles.74 There are repor ts of some to xicity in the major organs like the li ver, bone mar row, kidneys, lymph nodes, heart, and lungs to these tar geted nanoparticles.69,71 Therefore, it becomes imperati ve to identify the safety

601

issues of potentiall y useful nanopar ticles for imaging especially in ter ms of the mechanism of to xicity, reaction with the inter nal body en vironment, and par ticle size and characteristics.

CONCLUSION This chapter pro vides an o verview of the nonclinical product de velopmental strate gies, safety considerations, and to xicity prof iles of medical imaging and radiopharmaceutical products. Toxicity prof iles dif fer considerably betw een classes of medical imaging agents. For some agents, for e xample, GBCAs, cer tain class-related toxicity effect has been identified whereas for other products, for e xample, PET diagnostic radiopharmaceuticals, no clear class to xicity prof ile has emerged presumably because of the very heterogeneous nature of the g roup. The regulatory pathways for the nonclinical development of these agents are relati vely broad. In general, the number as w ell as the nature of the nonclinical studies required for development of imaging products depend on the phar macologic class of the dr ug and the anticipated safety profile. Multiple precedents have been established to assist in the development of these products.

ACKNOWLEDGMENT The authors wish to thank Anne M. Pilaro, M. Stace y Ricci, and Haw-Jyh Chiu for their v aluable input in the Diagnostic Radiopharmaceuticals section.

REFERENCES 1. US Code of F ederal Re gulations Title 21, F ood, Dr ug, and Cosmetic Act, P ART 315.2-Diagnostic radiophar maceuticals (2008) a vailable at http://www .accessdata.fda.gov/scripts/ cdrh /cfdocs/ cfCFR/CFRSearch.cfm?fr=315.2 (accessed Feb 12, 2009). 2. US Code of F ederal Regulations Title 21 P ART 312-Investigational New Dr ug Application (2008) http://www .accessdata.fda.gov/ scripts/cdrh/cfdocs/cfCFR/CFRSearch.cfm?CFRPart=312 (2008) (accessed Feb 17, 2009). 3. US Code of F ederal Regulations Title 21 P ART 312-Investigational New Dr ug Application (2008) http://www .accessdata.fda.gov/ scripts/cdrh/cfdocs/cfCFR/CFRSearch.cfm?CFRPart=601 (accessed Feb 17, 2009). 4. US FD A. Guidance for industr y; de veloping medical imaging drug and biological products, P arts 1, 2, and 3. 1994. Available at: http://www.fda.gov/cder/guidance/5742prt1.htm (accessed Feb 12, 2009). 5. International Conference on Har monization. Guidance on nonclinical safety studies for the conduct of human clinical trials and marketing authorization for phar maceuticals M3(R2) 2008. Published in the Federal Register, 1997;62:62922. Available at: http://www.ich.org/LOB/ media/MEDIA4744.pdf.

602

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

6. US FDA. Guidance for industry; estimating the maximum safe starting dose in initial clinical trials for therapeutics in adult healthy volunteers. 2005. Available at: http://www .fda.gov/cder/guidance/ 5541fnl.htm. 7. International Conference on Harmonization. Safety pharmacology studies for human phar maceuticals S7A 2000. Pub lished in the F ederal Register, 2001;66(135):36791–92. Available at: http://www.ich.org/ LOB/media/MEDIA504.pdf. 8. International Conference on Harmonization. The non-clinical evaluation of the potential for dela yed v entricular repolarization (Qt Interval Prolongation) b y human phar maceuticals S7B 2005. Published in the F ederal Register, 2005;70(202):61133–4. Available at: http://www.ich.org/ LOB/media/MEDIA2192.pdf. 9. US FD A. Guidance for industr y, in vestigators, and re viewers. Exploratory IND studies 2006. Available at: http://www.fda.gov/ cder/guidance/7086fnl.pdf. 10. FDA Approval P ackage; NDA 19-596, Gadopentetate dime glumine (Magnevist) June 1988. 11. FDA Approval P ackage; ND A 20-123, Gadodiamide (Omniscan) January 1993. 12. FDA Approval P ackage; ND A 20-937, 20-975; Gado versetamide (OptiMARK) December 1999. 13. FDA Approval P ackage; ND A 20-131; Gadoteridol (ProHance) November 1992. 14. FDA Approval P ackage; ND A 21-357; Gadobenate dime glumine (MultiHance) November 2004. 15. FDA Approval P ackage; ND A 22-090; Gado xetate disodium (Eo vist) July 2008. 16. Dohr O, Hofmeister R, Treher M, Schweinfurth H. Preclinical safety evaluation of Gd-EOB-DTP A (Primo vist). In vest Radiol 2007; 42:830–41. 17. Thomsen HS, Dorph S, Larsen S, et al. Urine profiles and kidney histology after intravenous injection of ionic and nonionic radiologic and magnetic resonance contrast media in nor mal rats. Acad Radiol 1994;1:128–35. 18. Elmstahl B, Leander P, Grant D , et al. Histomor phological changes after Renal X-Ra y ar teriography using iodine and gadolinium contrast media in an ischemic porcine model. Acta Radiol 2007;48:1109-1119. 19. Cowper SE. Nephrogenic systemic fibrosis: a review and exploration of the role of gadolinium. Adv Dermatol 2007;23:131–54. 20. Agarwal R, Br unelli SM, Williams K, et al. Gadolinium-based contrast agents and nephro genic systemic f ibrosis: a systematic review and meta-anal ysis. Nephrol Dial Transplant 2008.NDT Advance Access published online on October 24, 2008, Nephrology Dialysis Transplantation, doi:10.1093/ndt/gfn593 21. Abraham JL, Thakral C. Tissue distribution and kinetics of gadolinium and nephro genic systemic f ibrosis. Eur J Radiol 2008; 66:200–7. 22. Idee JM, Port M, Medina C, et al. Possible involvement of gadolinium chelates in the pathophysiology of nephrogenic systemic fibrosis: a critical review. Toxicology 2008;248:77–88. 23. Optimark Product Label. F ebruary 17, 2009. Available at: http:// dailymed.nlm.nih.gov/dailymed/dr ugInfo.cfm?id=4667 (accessed Feb 17, 2009). 24. Multihance Product Label. F ebruary 17, 2009. Available at: http:// dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?id=8934 (accessed Feb 17, 2009). 25. Sieber MA, Pietsch H, Walter J, et al. A preclinical study to investigate the de velopment of nephro genic systemic f ibrosis: a possib le role for gadolinium-based contrast media. In vest Radiol 2008; 43:65–75. 26. Runge VM, Dickey KM, Williams NM, Peng X. Local tissue toxicity in response to e xtravascular extravasation of magnetic resonance contrast media. Invest Radiol 2002;37:393–98.

27. US FD A. Advisory Committee Brief ing document; Cardio vascular and Renal Dr ugs Advisory Committee: safety considerations in the development of ultrasound contrast agents (6-24-2008). Available at: http://www .fda.gov/ohrms/dockets/AC/08/briefing/ 2008-4369b1-01.pdf (accessed Feb 17, 2009). 28. Lindner JR. Microbubb les in medical imaging: cur rent applications and future directions. Nat Rev Drug Discov 2004;3:527–32. 29. Willmann JK, van Bruggen N, Dinkelborg LM, Gambhir SS. Molecular imaging in dr ug development. Nat Re v Dr ug Discov 2008; 7:591–607. 30. FDA Approval Package; NDA 21-064; Perflutren (Definity) July 2001. 31. Perflutren Protein-Type A Microspheres Label (Optison) 2008. Available at: http://www.fda.gov/cder/foi/label/2008/ 020899s011lbl.pdf (accessed Feb 17, 2009). 32. Grauer SE, P antely GA, Xu J , et al. My ocardial imaging with a ne w transpulmonary lipid-fluorocarbon echo contrast agent: e xperimental studies in pigs. Am Heart J 1996;132:938–45. 33. Staub NC. Pulmonar y intravascular macrophages. Annu Rev Physiol 1994;56:47–67. 34. Graff-Guerrero A, Willeit M, Ginovart N, et al. Brain region binding of the D-2/3 agonist [C-11]-(+)-PHNO and the D-2/3 antagonist [C-11]raclopride in health y humans. Hum Brain Mapp 2008; 29:400–10. 35. Willeit M, Gino vart N , Graf f A, et al. F irst human e vidence of d-amphetamine induced displacement of a D-2/3 agonist radioligand: A [C-11]-(+)-PHNO positron emission tomography study. Neuropsychopharmacology 2008;33:279–89. 36. Mu LJ, Drandarov K, Bisson WH, et al. Synthesis and binding studies of epibatidine analo gues as ligands for the nicotinic acetylcholine receptors. Eur J Med Chem 2006;41:640–50. 37. Horti AG, Chefer SI, Mukhin AG, et al. 6-[F-18]fluoroA-85380, a no vel radioligand for in vi vo imaging of central nicotinic acetylcholine receptors. Life Sci 2000;67:L463–9. 38. US FD A. P oints to consider in the manuf acturing and testing of monoclonal antibody products for human use 1997. Available at: http://www .fda.gov/cber/gdlns/ptc_mab.pdf (accessed Sep 10, 2008). 39. Internal Conference on Har monisation: “Guidance for Industr y S6, Preclinical Safety Evaluation of Biotechnology-Derived Pharmaceuticals” (1997) Available at: http:// www .fda.gov/cder/guidance/1859fnl.pdf (1997) (accessed Sep 10, 2008). 40. International Commission on Radiolo gical Protection. Radiation Dose to P atients from Radiophar maceuticals. ICRP Pub lication 53, Annals of the ICRP Volume 18/1-4 (1988) Elsevier. 41. FDA Approval Package; NDA 20-084; Meta-iodobenzylguanidine sulfate, I-131. August 1993. Available online at www . Pharmapendium.com (accessed Sep 10, 2008). 42. FDA Approval Package; NDA 20-314; Indium In-111 Pentetreotide Kit. march 1993. Available at www .pharmapendium.com (accessed Sep 10, 2008). 43. FDA Approval Package; NDA 20-887; Technicium Tc-99m Apcitide September 1998. Available at www .phar mapendium.com (accessed Sep 10, 2008). 44. FDA Approval Package; NDA 22-119; Ammonia, N-13. August 2007. Available at www.pharmapendium .com (accessed Sep 10, 2008). 45. Informal discussion of literature re views of safety and ef ficacy data for PET : 15O-W ater and 18F-Fluorodopa. 2000. Available at: http://www.fda.gov/cder/regulatory/PET/july_28_2000_informal_discussion.htm (accessed Sep 10, 2008). 46. FDA Approval P ackage; ND A 21-768; Fludeo xyglucose, F-18. july 2006. Available at www.pharmapendium.com (accessed Sep 10, 2008). 47. Bessell EM, Courtenay VD, Foster AB, et al. Some in vivo and in vitro antitumour ef fects of the deo xyfluoro-D-glucopyranoses. Eur J Cancer 1973;9:463–70.

Nonclinical Product Developmental Strategies, Safety Considerations. . .

48. Reivich M, K uhl D , Wolf A, et al. The [18F]fluorodeo xyglucose method for the measurement of local cerebral glucose utilization in man. Circ Res 1979;44:127–37. 49. Thomas DG, Duthie HL. Use of 2 deoxy-D-glucose to test for the completeness of surgical vagotomy. Gut 1968;9:125–28. 50. Gough AL, K eddie NC. An assessment of the reproducibility and safety of 2-deoxy-D-glucose as a gastric acid stimulant in duodenal ulcer patients. Gut 1975;16:171–76. 51. Gerst SR, Hricak H. Chapter 6. Radiolo gy of the urinar y tract. In: Tanagho EA, McAninch J , editors. Smith’s general urolo gy, Har rison’s principles of inter nal medicine. 17th ed. Ne w York: Lange Medical Books/McGraw-Hill, Medical Pub lishing Division; 2008. Accessed through StatRef Website http://online.statref.com/document .aspx?fxid=12& docid=36 (accessed Aug 12, 2008). 52. Dillon WP. 2008. Chapter 362. Neuroimaging in neurolo gic disorders. In: Tan agho EA, McAninch J, editors. Diagnosis of neurologic disorders (Section 1) in neurolo gic disorders (P art 16) in Harrison’s principles of inter nal medicine. 17th ed. Ne w York: Lange Medical Books/McGra w-Hill, Medical Pub lishing Di vision; 2008. Accessed through StatRef Website http://online. statref.com/Document/Document.aspx?DocId=3280&FxId=55& Scroll=5&Index=1&SessionId=D767DDHOOSPYVTEI (accessed Aug 12, 2008). 53. Multiple Authors. Chapter 46, Radiological contrast agents, iodinated contrast media. In: Aronson JK, editor . Me yler’s side ef fects of drugs; The International Encyclopedia of Adverse Drug Reactions and Interactions. 15 ed. Else vier; 2006. Accessed through Pharmapendium W ebsite https://www.pharmapendium.com/ index.jsp (accessed Aug 4, 2008). 54. Grainger RG. 1997. Chapter 2, Intra vascular contrast media. In: Grainger RG, Allyson DJ, editors. Diagnostic radiolo gy, a te xtbook of medical imaging. 3rd ed. Ne w York: Churchill Li vingstone; p. 35–48. 55. Visipaque(TM) Iodixanol Injection Product Insert. 2003. Available at: http://www.fda.gov/medwatch/SAFETY/ 2003/03DEC_PI/V isipaque_PI.pdf (accessed Aug 18, 2008). 56. Omnipaque(TM) Iohexol Injection Product Label 2007. Available at: http://dailymed.nlm.nih.gov/dailymed/ fda/fdaDr ugXsl.cfm?id= 6848&type=display (accessed Aug 18, 2008). 57. American Colle ge of Radiolo gy. Manual on Contrast Media Version 6. 2008. Available at: http://www.acr.org/SecondaryMainMe nu Categ orie s/q ua lity_ sa fe ty/c on trast_ma nu al. aspx (accessed Aug 12, 2008). 58. Almen T. Angiography with metrizamide. Animal e xperiments and preliminary clinical e xperiences. Acta Radiol Suppl 1977; 355:419–30.

603

59. Morris TW. A review of coronar y ar teriography- and contrast mediainduced ventricular fibrillation. Acta Radiol Suppl 1995;399:100–4. 60. Baath L, Almen T. Reduction of the risk of v entricular f ibrillation in the isolated rabbit heart by small additions of electrolytes to nonionic monomeric contrast media. Acta Radiol 1989;30:327–33. 61. Muschick P, Wehrmann D, Schuhmann-Giampieri G, Krause W. Cardiac and hemodynamic tolerability of iodinated contrast media in the anesthetized rat. Invest Radiol 1995;30:745–53. 62. Persson PB, Hansell P, Liss P. Pathophysiology of contrast mediuminduced nephropathy. Kidney Int 2005;68:14–22. 63. Rudnick MR, Goldf arb S. P athogenesis of contrast-induced nephropathy: e xperimental and clinical obser vations with an emphasis on the role of osmolality . Rev Cardiovasc Med 2003;4 Suppl 5:S28–33. 64. Idee JM, Lancelot E, Ber thommier C, et al. Ef fects of non-ionic monomeric and dimeric iodinated contrast media on renal and systemic haemodynamics in rats. Fundam Clin Phar macol 2000;14:11–18. 65. Heinrich MC, K uhlmann MK, Gr gic A, et al. Cytoto xic effects of ionic high-osmolar , nonionic monomeric, and nonionic isoosmolar dimeric iodinated contrast media on renal tubular cells in vitro. Radiology 2005;235:843–9. 66. Walday P, Heglund IF, Golman K, et al. Renal ef fects of iodixanol in experimental animals. Acta Radiol Suppl 1995;399:204–12. 67. Junck L, Marshall WH. Neurotoxicity of radiological contrast agents. Ann Neurol 1983;13:469–84. 68. Morcos SK. Review article: acute serious and f atal reactions to contrast media: our cur rent understanding. Br J Radiol 2005; 78:686–93. 69. Liu Y, Miyoshi H, Nakamura M. Nanomedicine for drug delivery and imaging: a promising a venue for cancer therap y and diagnosis using tar geted functional nanopar ticles. Int J Cancer 2007; 120:2527–37. 70. Panyam J, Labhasetwar V. Biodegradable nanoparticles for dr ug and gene deli very to cells and tissues. Adv Dr ug Deli v Re v 2003;55:329–47. 71. Liu YY, Chen XQ, Xin JH. Silica nanopar ticles-walled microcapsules. J Mater Sci 2006;41:5399–401. 72. Drobne D. Nanotoxicology for safe and sustainab le nanotechnology. Arh Hig Rada Toksikol 2007;58:471–78. 73. Garnett MC, Kallinteri P. Nanomedicines and nanoto xicology: some physiological principles. Occup Med (Lond) 2006;56:307–11. 74. Borm P, Robbins D, Haubold S, et al. The potential risks of nanomaterials: a re view car ried out for ECET OC. Available at: http://www.particleandf ibretoxicology.com/content/3/1/11 (accessed July 8, 2008).

38 OVERVIEW

OF

MOLECULAR

AND

CELL BIOLOGY

HARVEY R. HERSCHMAN, PHD AND HIDEVALDO B. MACHADO, PHD

This chapter will describe, for the nonmolecular biologist/ cell biologist, (1) many of the fundamental molecular and cellular processes that occur in man and in model or ganisms and (2) alterations in these processes that occur in disease initiation, pro gression, and response to therap y. By necessity, this chapter will be terse—perhaps to the extent of being superf icial—in its attempt to be comprehensive and y et sta y within the conf ines of allo wable length. We f irst need to def ine the topic “cell biolo gy.” The Journal of Cell Biology and Nature Cell Biology include topics such as nuclear organization and structure, nuclear transport, deo xyribonucleic acid (DN A) replication and repair, transcription and chromatin str ucture, cell c ycle and cell division, cell growth, protein and membrane trafficking, signal transduction, cellular adhesion and motility, proteolysis, and cell survival and death. Both journals emphasize the overlap of “cell biolo gy” with disciplines such as immunolo gy, developmental biology, neurobiology, and disease—topics discussed in the molecular imaging applications considered in man y other chapters of this book. To begin, we will def ine the structure(s) of a cell; the critical cellular components and compar tments in which molecular and subcellular processes occur . Subsequently, we will emphasize topics in cell biolo gy w hose consequences ha ve or gan-wide and or ganism-wide consequences since the imaging technologies discussed in Part I and the imaging applications described in Parts III-VI deal primarily with imaging at the w hole cell, multicellular , organ and/or or ganismal levels. Following (1) the section on cell str ucture, we will consider (2) the “life c ycle” of individual gene products, (3) the cell di vision c ycle, (4) transport of small molecules across membranes, (5) outside to inside the cell; signal transduction, (6) cell cycle, signal transduction and cancer , (7) the life c ycles of cells—embryology, dif ferentiation, de velopment, and 604

death (8) embryonic stem cells (ES cells), adult stem cells, induced pluripotent stem cells and cancer stem cells, and (9) transgenic animals. F or readers w ho would like more extensive descriptions of these fundamentals of cell biology, we recommend two excellent upper division/graduate level text books; Molecular Biology of the Cell 1 and Molecular Cell Biology.2 In all candor, the Wikipedia Web site is also an e xcellent primer on man y of these topics . To tie these fundamental concepts in cell biolo gy to the common theme of this book, e xamples of molecular imaging applications or insights will be presented. As will become apparent, animal models are important for development of molecular imaging technologies and imaging probes, and in using molecular imaging applications to increase our understanding of basic biology. Animal models are also essential both to increase our understanding of disease pro gression and for development and testing of therapeutic approaches and alter natives. Because de velopment of transgenic mice in w hich cells and tissues can be mark ed with “reporter genes” w hose e xpression can be nonin vasively, repeatedl y, and quantitati vely imaged has become such an important part of molecular imaging at the research le vel, and of fers such promise for future studies, a short “primer” on the characteristics and uses of transgenic mice will conclude the chapter .

CELL STRUCTURE We will be gin with an in ventory of components of the cell, considering it as a static, multicompar tment entity (Figure 1). In reality , of course, the cell is an e xtraordinarily acti ve, ener gy producing and consuming microcosm, car rying out hundreds of thousands of molecular and biochemical processes that require both interactions and material transfer among these compartments, as well

Overview of Molecular and Cell Biolo gy

Figure 1. diagram.

605

The organelles, compartments, and components of cells. The components of a typical eukaryotic cell are shown in the

as modif ication of the compar tments themselves as the cell goes about both its autonomous functions and its integrated, or ganismal functions. Ho wever, before discussing the roles and the interactions of the pla yers, we have to identify and describe them. The cell is bounded b y a plasma membrane; a lipid bilayer that contains loosel y attached ( peripheral) and strongly attached ( integral) proteins. The plasma membrane protein components car ry out a v ariety of functions; for e xample, acting as channels or pumps to mediate transpor t of small molecules (including probes for man y imaging procedures), acting as receptors for proteins such as hor mones, g rowth f actors, and cytokines that modulate cell behavior, and providing the mechanism(s) for “pinching of f ” por tions of the lipid bilayer membrane to bring materials from the outside of the cell to the inside. Coated pits bud from the plasma membrane and for m endocytic vesicles that car ry cargo from the exterior of the cell into the cytoplasm. The cell membrane can also include areas of specialization that mediate passage of small molecules from one cell to another when cells are in direct contact ( gap junctions), specialized regions of the membranes that f acilitate the adherence of one cell to another (adherens junctions and desmosomes) to provide structural integrity to multicellular tissues, and tight junctions in w hich re gions of membranes of adjacent epithelial cells for m an

impenetrable barrier that separates the apical (or upper) regions of a polarized epithelial cell layer from the basolateral (or lower) regions of the epithelium. Eukaryotes, to w hich w e and the other or ganisms used in imaging studies discussed in this book belong, are distinguished b y the f act that our cells contain a nucleus. The nuclear envelope, tw o lipid bila yer rings separated by the perinuclear space, segregates the cytoplasmic compartment of the cell from the nucleus. The double-layered nuclear membrane contains nuclear pores, through w hich macromolecules, lipids, small molecules, and other cellular components can exchange between the c ytoplasm and the nucleus. Within the nucleus are the chromosomes, where the g reat majority of the genes (other than mitochondrial genes) that control cell function reside. The DN A sequences that encode our genes are coated with specif ic proteins; these nucleoprotein complexes are known as chromatin. The major proteins associated with the acidic DNA molecules, to compact the DN A into chromosomes, are the basic histones. Chromatin interactions f acilitate the packing of the roughly seven linear feet of DNA present in each cell into the chromosomes. Dynamic alterations in chromatin str ucture, primaril y as a result of posttranslational, re versible histone modif ications, par ticipate in mediating replication of DN A as cells pass through the cell c ycle, stable differential expression of

606

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

genes in distinct cell types, and transient re gulation of gene expression in cells in response to extracellular signaling cues. Thus, chromatin-mediated differential gene expression results both in the differentiation of our cells into alter native tissues and in the ability of cells to respond to e xtracellular signals and change their biochemical proper ties and functions. Within the nucleus we f ind the nucleolus, the site both of synthesis of the ribonucleic acid (RN A) and protein components of the ribosomal subunits as w ell as the site of ribosome assembly (ribosomes are in volved in protein manuf acture, as described subsequently). Within the c ytoplasm, but e xcluded from the nucleus, are the mitochondria—the organelles that are the source of ener gy generation via adenosine triphosphate (A TP) production. Mitochondria are composed of se veral compar tments; the mitochondrial outer membrane, the inter membrane space , the inner membrane, the cristae (formed by foldings that occur in the inner mitochondrial membrane), and the mitochondrial matrix. Although most well publicized for their roles in energy generation for the cell, the mitochondria also participate in other biolo gical processes—cell death, regulating some aspects of cell c ycle control, modulating cellular dif ferentiation, and par ticipating in the intracellular responses to signaling pathw ays activated by extracellular modulators. Mitochondria are the onl y organelle other than the nucleus kno wn to contain DNA. The mitochondrial g enome encodes a relati vely small number of RNAs and proteins that help for m the mitochondrial ribosomes. Because mitochondria are inherited b y f ission, the mitochondrial genome is maternally inherited. Ribosomes are the RN A-protein machines on which new proteins are synthesized. The small ribosomal subunit and the large ribosomal subunit form the active protein synthesizing w orkbench. Mitochondrial ribosomes and cytoplasmic ribosomes are distinct, both in their molecular (RN A and protein) composition and in the subsets of proteins that the y synthesize. The genes that encode the RN A and protein components of cytoplasmic ribosomes are all found in nuclear DN A. In contrast, nuclear genes and mitochondrial genes encode proteins that are present in mitochondrial ribosomes. Cytoplasmic ribosomes are found in tw o compartments. Ribosomes that are synthesizing proteins destined to become either secreted proteins or proteins that will become inte gral membrane proteins are associated with the rough endoplasmic reticulum, or rough ER. In contrast, ribosomes that are engaged in the translation of messages for other proteins—those

proteins not destined either to be inte gral membrane components or secreted proteins—are found free in the cytoplasm. “F ree” ribosomes and ribosomes bound to the rough ER are identical; the distinction that determines whether they will be “free” or bound to the ER is a function of the protein that the y are synthesizing. The name “rough endoplasmic reticulum” suggests an alter native—the smooth endoplasmic r eticulum, or smooth ER. And, indeed, the ER, containing both rough ER and smooth ER, is a continuous or ganelle that resembles a netw ork of tubules and sacs, stretching from the nucleus to the cell membrane. The ER is, in fact, continuous with the outer nuclear en velope. The smooth ER does not contain ribosomes and , consequently, it is not in volved in protein synthesis. Ho wever, the smooth ER contains enzymes in volved in many biosynthetic pathways in cells, for example, lipid production. The ER is the site of synthesis for essentially all the cellular membranes, including the proteins and lipids of both the plasma membrane and the membranes for other cellular or ganelles. The membrane components destined for other or ganelles, but synthesized in the ER, are shuttled to their alternative sites by transport vesicles that bud from the ER membrane. The ER is one of the most flexible of cellular organelles; for example, cells that specialize in synthesizing secreted proteins (e g, antibody producing cells) ha ve v astly expanded rough ER to facilitate synthesis of these proteins w hile cells in volved in deto xification of dr ugs expand their smooth ER to accommodate the increased cellular production of the enzymes in volved in these biochemical processes. The Golgi appar atus (also kno wn as the Golgi body or Golgi complex), like the ER, is a membranous network that looks like a collection of bags and tubules. Alberts and colleagues 1 describe the Golgi as “resembling a stack of pancak es” and others describe the Golgi apparatus as resemb ling a set of stack ed dinner plates. The Golgi apparatus acts as a site for modif ication, packaging, sor ting and tar geting of secreted proteins and proteins destined to become components of the various cellular or ganelles. Proteins synthesized in the ER and destined for a v ariety of tar gets are transferred from the ER to the Golgi apparatus b y transport vesicles, entering from the cis side of these stacks (on the nuclear f ace). The proteins pass through the Golgi, where they are modif ied by glycosylation, phosphor ylation, and sulf ation reactions (and other posttranslational modifications) that target the individual proteins for e xport to the proper compar tment (e g, the plasma membrane, the lysosome, secretion). Transport vesicles

Overview of Molecular and Cell Biolo gy

carrying the matured, targeted proteins depart the Golgi from the trans-Golgi apparatus. The lysosomes are membrane-bound or ganelles that specialize in de gradation of macromolecules and “outdated” or ganelles. To car ry out their de gradative function, l ysosomes contain a v ariety of h ydrolytic enzymes (lipases, proteases, nucleases, gl ycosidases, phospholipases) that are synthesized in the ER, modified for tar geted l ysosomal deli very in the Golgi apparatus and transported from the trans-Golgi apparatus to the l ysosome by transport vesicles. Mutations in a subset of these enzymes are responsib le for a class of human diseases kno wn as the lysosomal stor age diseases.3 Molecules and membranes from inside the cell that are to be digested ma y reach the l ysosome by autophagy of damaged intracellular or ganelles. Alternatively, molecules from outside the cell ma y be targeted to the lysosome for de gradation follo wing receptor-mediated endocytosis of extracellular ligands, while bacteria and other e xtracellular entities ma y reach the l ysosome by the cellular engulfment process known as phagocytosis. Peroxisomes are another group of membrane bound organelles that deto xify cellular components; in this case, primaril y o xidized f atty acids. The pero xisomes are, like the lysosomes, generated from membrane components synthesized in the ER and modif ied in the Golgi before being transported into the maturing peroxisome. P eroxisomes contain enzymes that perfor m b-oxidation reactions to degrade, in a step-wise fashion, long chain f atty acids. Peroxisomes also contain o xidative enzymes that produce h ydrogen pero xide in the process of oxidizing substrates. The hydrogen peroxide is then used by catalase, another enzyme resident in the peroxisome, to oxidize other substrates. As an example, ethanol is o xidized to acetaldeh yde b y a pero xisomemediated reaction. On occasion, these o xidative enzymes are at such high le vels in the pero xisome that they for m a distincti ve cr ystalline core that distinguishes this or ganelle in electron micro graphs. Peroxisomes are also di verse or ganelles; in alter native cell types the y can contain considerab ly dif ferent complements of enzymes. Within the cytoplasm are a variety of f ibrous proteins or filaments (actin, microtubules, microfilaments) that are collectively referred to as the cellular cytoskeleton. Collectively, the c ytoskeleton deter mines the str ucture of the cytoplasm and controls the shape of the cell. Each of these f ilaments is a dynamic str ucture, composed of monomeric components that polymerize and depolymerize in response to changes in cellular en vironment, cell-cell

607

contact, e xternal ligands, cell c ycle pro gression, etc. Cytoskeleton str uctures and their reor ganization pla y major roles in processes such as microvilli projection from the cell surface, pseudopodia formation at the cell surface, muscle contraction, cell motility , axonal flo w, retro grade transport in neurons, mitosis, and intracellular v esicle transport. The centrosome (also kno wn as the microtubule organizing center) is a cellular or ganelle involved in the coordination of mitosis and in the str ucture of the cellular microtubule skeleton. Two centrioles are found within centrosomes, surrounded by the centrosome matrix, also known as the pericentriolar material. During interphase of the cell c ycle (see section “The Cell Di vision Cycle”) the centrosome, w hich lies near the nuclear membrane, serves as a nucleating center for a microtubule netw ork that provides a cytoskeleton for most cell types. At mitosis, the centrosome duplicates and “sheds” it microtubules. The two centrosomes of the mitotic cell mig rate to new positions in the cell and ser ve as nucleation centers for the for mation of the spindle apparatus that separates the duplicated chromosomes to the sib ling cells during mitosis. The obser vation of cellular str uctures in li ving animals is still technicall y very challenging, considering the high resolution required to identify these str uctures inside the cell. Therefore, most of the studies on cell structure, intracellular organelles, and dynamics of organelle str ucture and function are still dependent on microscopic technolo gies. The adv ent of fluorescent labeling techniques (fluorescent protein fusions, quantum dots, etc) and related optical technolo gies no w facilitate the study, repeatedly and with high resolution, of both the str ucture of living cells as well as activities and alterations of organelles within the cell in response to stimuli. Such technolo gies include but are not limited to, fluorescence resonance energy transfer (FRET), fluorescence microscop y, laser scanning confocal, spinning disk confocal, and multiphoton microscop y. Many researchers are no w studying cells and their internal components in li ving animals, using intra vital microscopy techniques. 4–6 In par ticular, the use of multiphoton and confocal microscopy can be used noninvasively in small animals such as Xenopus embryos 7–10 or Zebra f ish, due to their transparent nature. Although the remainder of this chapter and the bulk of the studies discussed in this book emphasize nonin vasive imaging at the or ganismal le vel, cellular and subcellular intravital imaging at the other end of the spectrum in the conte xt of tissues and or gans provide an important lower boundary to this topic.

608

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

THE “LIFE CYCLE” OF INDIVIDUAL GENE PRODUCTS With the singular e xceptions of the immuno globulin genes in antibody-producing B cells and the genes encoding antigen-specif ic receptors in T cells, all cells in our body have identical DN A sequences. Thus, the complement of genes, which w e will def ine here as DN A sequences that are transcribed into RNA molecules, are identical in all our cells. What distinguishes one differentiated cell type from another , then, must be the dif ferences in the alter native subsets of genes transcribed into RNA in dif ferent cell types. In addition, alter native cell responses to changes in external stimuli—growth factors, hormones, c ytokines, and other signaling molecules— frequently depends on stimulus-induced increases and/or decreases in e xpression of subsets of genes in “tar get” cells. The signaling pathways that modulate gene expression in response to stimuli are discussed (see section “Outside to Inside the Cell; Signal Transduction”). Historically, characterization of gene e xpression has centered primaril y on genes w hose RN A transcription products are subsequentl y translated into proteins. The focus of this section will be on the production of messenger RN A (mRN A) from protein-encoding genes, the mechanism and regulation of translation of mRN As into proteins, the posttranslational modif ications of proteins, the interactions of proteins in cells, and protein de gradation. Ho wever, increasing interest and research has recently been focused on the e xpression and function of RNA transcripts that are not translated into proteins, RNA molecules that directl y car ry out biolo gical functions. The functions of se veral classes of untranslated RNAs ( transfer RNAs, microRNAs) will be discussed in the context of their roles in the mechanism and regulation of protein expression. To begin, we need to def ine more full y the “gene. ” Initially concei ved as the unit of inheritance, most cell and molecular biolo gists now consider the gene to be a linear DN A re gion that includes the sequences that encode the mRN A transcription product of RNA pol ymerase and the adjacent DN A regulatory r egions that modulate transcription (Figure 2). The promoter is a regulatory re gion immediatel y adjacent to the star t site of mRNA transcription (although promoters can extend over several kilobases); transcription factors bind to the promoter, for ming a DN A-protein comple x that modulates RNA pol ymerase transcription of the RN A-encoding DNA sequence (the transcription unit ) into a complementary RN A molecule. Transcription f actors are bipartite; the y contain a DNA-binding domain that

recognizes specific DNA sequences and a transcriptional activation domain (or transcriptional silencing domain ) that modulates the RN A polymerase activity of the transcriptional machinery. The “upstream” 5ʹ′ noncoding portions of genes also include enhancer sequences that bind protein transcriptional activator s and transcriptional repressors. The interactions of the proteins bound at the enhancer sequences and proteins bound at the promoter regulate transcription of the RN A coding re gion of the gene (Figure 2). The RN A pol ymerase products of gene coding regions are the primary RN A tr anscripts. The primar y transcripts of most genes are subsequentl y reduced in length by RNA splicing, to create the f inal RNA product, the mRNA. The DNA sequences encoding primar y transcript regions that remain in the f inal RNA product are the exons; the sequences encoding primar y transcript regions eliminated b y splicing are the introns. Unique sequences at the ends of e xons and introns identify the sites b y w hich the spliceosome, the RN A-protein complex that car ries out the splicing reaction, reco gnizes the intronic RNA sequences to be removed from the primary transcript. For genes w hose f inal products are proteins, the spliced RN A product, the mRNA, has a protein coding region that is translated into the protein amino acid sequence. The sequential three nucleotide codons of the mRNA protein coding re gion are the templates for the sequential binding of the charged tr ansfer RN As that carry the indi vidual amino acids to be pol ymerized into the protein sequence during the translation process. The coding region of the mRNA has a start codon at its beginning, and a stop codon at its end. The protein coding region of the mRN A is preceded b y a 5ʹ′ untranslated region, or 5ʹ′UTR, and is follo wed by the 3ʹ′ untranslated region, or 3ʹ′UTR. F ollowing splicing, a set of proteins bind to facilitate transport of the mature mRNA from the nucleus through the nuclear pore and into the c ytoplasm of the cell. Translation is initiated by binding of translation initiation factors to the mRNA 5ʹ′ cap sequence first and subsequent binding of the small (40S) ribosome subunit at this site. After the addition of se veral other proteins, this preinitiation complex moves down the mRNA to the star t codon. As a result of reco gnition between the star t triplet codon sequence in the mRN A and the complementar y anticodon sequence of the appropriate transfer RN A, the transfer RNA carrying the f irst amino acid in the protein sequence joins the initiation complex, binding to the small ribosomal sub unit as directed b y the start codon of the mRNA. The large (60S) ribosomal sub unit is then added

Overview of Molecular and Cell Biolo gy

609

Figure 2. The “life cycle” of individual gene products. (upper panel) From gene to message; a brief outline of gene expression. For our purposes, the gene includes enhancer and promoter regulatory regions, to which transcription factors and enhancer-binding proteins bind, in order to modulate ribonucleic acid (RNA) polymerase-mediated transcription, as well as the exons and introns that are the template for the primary RNA transcript. The primary transcript can, as a result of alternative splicing by the spliceosome machinery, be converted to alternatively spliced messenger RNAs (mRNAs) that may have distinct 3ʹ′ untranslated regions (UTRs), distinct protein coding sequences, and distinct 5ʹ′ UTRs. (lower panel) mRNA to protein and the fate of cellular proteins. Mature mRNAs are recognized by a set of molecules that aid in transporting the message from the nucleus through the nuclear pore into the cytoplasm. The 5ʹ′UTR cap region is recognized by a group of cytoplasmic translation initiation factors and the 40S ribosomal subunit. Once this preinitiation complex finds the translation start codon, the 60S ribosomal subunit is added, and the elongation complex begins its cyclic addition of amino acids from the charged transfer RNAs to the growing peptide chain. As the initial elongation complex moves down the mRNA, new initiation complexes and elongation complexes can assemble and become active on the message, generating polysomes. Those polysomes synthesizing membrane or secreted proteins become attached to the endoplasmic reticulum (ER), forming rough ER. After their synthesis, proteins made in the rough ER are moved by transport vesicles to the Golgi body, where they are subjected to posttranslational modifications such as glycosylation and sulfation. After posttranslational modification, the proteins are moved to their targets by transport vesicles. Newly synthesized cytoplasmic proteins are also subjected to posttranslational modifications that modulate their location, function, stability, etc. Damaged proteins are often recognized by specific mechanisms in the cell and shuttled to the lysosome for degradation. Regulated degradation of proteins, as occurs for example in the cell cycle, is carried out by specific polyubiquitination of proteins; proteins with polyubiquitin chains are subsequently degraded by the proteasome.

610

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

to for m the elongation complex. Elongation is a sequential, ATP-dependent cyclic reaction in which (1) the appropriate char ged transfer RN A is positioned on the ribosome-mRNA comple x as directed b y the mRN A codon sequence, (2) the peptide bond to the g rowing polypeptide chain is for med, (3) the transfer RN A is released, and (4) the mRNA is moved one codon along the coding region. The process is repeated until the ter mination codon is reached and the completed pol ypeptide chain is released from the elongation comple x. Because mRNAs are often quite long in mammalian cells, translation initiation can reoccur on an mRN A that is still in the process of being translated, once the mRNA 5ʹ′ cap is again available to recognize the initiation factors and another small ribosomal subunit following movement of the elongation comple x do wn the mRN A. Consequently, multiple ribosomes can be engaged in translation of the protein coding region of a single mRNA. The complex of multiple ribosomes engaged in sequentiall y synthesizing proteins from a single message is kno wn as a polyribosome or polysome. P olysomes synthesizing membrane proteins or secreted proteins are targeted, during the translation process, to the rough ER by a sequence present in the N-ter minal region of their translated products; polysomes synthesizing nonmembrane/nonsecreted proteins are found free in the c ytoplasm. The initiation of protein translation, the rate of protein translation, and the stability of mRNA molecules are regulated by specif ic proteins and shor t RNAs that bind to specif ic sequences in the 5 ʹ′ UTRs and 3 ʹ′ UTRs of mRNAs. For e xample, mRNAs w hose stability is re gulated b y g rowth f actors, c ytokines, and other signaling molecules often contain in their 3 ʹ′ UTRs multiple AU-rich elements that bind proteins that stabilize and/or destabilize the message. MicroRNAs (miRNAs) discovered only relatively recently, are short, nonprotein coding RNAs that bind to complementar y sequences in tar get mRNAs and can also modulate either mRN A stability and/or the ef ficacy of the translational machiner y. The practical application of this obser vation, RNA interf erence, recei ved the 2006 Nobel Prize. 11,12 The posttranscriptional modulation of mRNA stability and translation by mRN A binding f actors is an area of g reat cur rent interest and study. Imaging technologies to noninvasively study these regulatory events in vivo are rapidly developing and are awaited with great anticipation.13,14 Although we have only approximately 30,000 genes, the number of proteins encoded by these genes is actually much greater. As indicated in Figure 2, the primar y transcript encoded from most genes contains sequences both for a number of introns and for a number of e xons. For

many proteins, dif ferent exons from a common primar y transcript can be retained in alter native mRN As, as a consequence of “ alternative splicing .” If alter native exons that contain protein coding re gions are chosen b y alternative splicing, the proteins e xpressed from the alternative mRNAs will ha ve dif ferent b locks of amino acid sequence. If, as a result of alternative splicing within coding exons, the triplet nucleotide code is frame shifted, the proteins encoded b y the alter natively spliced messages will have different amino acid sequences at the site of translation distal to the alternative splice site. The ability of cells to make distinct, but related, proteins by “mixing and matching e xons” pro vides the cell with a mechanism to mak e e xtensive f amilies of proteins that perform related functions with modif ied proper ties. Estimates range from one to tw o orders of magnitude more proteins expressed in cells than genes present, as a result of alter native splicing. Dif ferentiated tissues often express some what dif ferent proteins from a common gene, as a consequence of tissue-specif ic alter native splicing; for e xample, the w ell-known src gene has a unique neuronal for m that includes an e xtra six amino acids, a consequence of including an additional e xonencoded sequence. 15 More dynamicall y, in response to alternative signals, cells can alter the splicing patter ns of targeted messages, to produce proteins with altered properties in response to changing en vironmental cues. If alternative messages are created with alternative 3ʹ′ UTRs or 5ʹ′ UTRs, these alternative messages may have distinct translation capabilities and/or distinct stabilities. The ability to nonin vasively monitor the appearance of ne w protein products of alter native splicing and/or the changes in message stability that occur as a result of dynamic modifications of splicing choices are among the opportunities and challenges for molecular imaging. As an example of pro gress in this area, repor ter genes that encode fluorescent protein products ha ve been used to analyze alternative splicing activity in distinct cells or tissues in transgenic mice. 16–18 Once a protein is synthesized—in fact, even during the process of pol ypeptide elongation—amino acids in proteins can be chemicall y modif ied b y a v ariety of enzymatic mechanisms. Among the man y enzymatic posttranslational protein modif ications are hydroxylation, phosphorylation, sulfation, meth ylation, acetylation, m yristylation, farnesylation, palmitolation, and glycosylation. Most of these reactions are re versible; the enzymes that modulate these modif ications and their re verse reactions (deacetylation, dephosphor ylation, demethylation, etc) are often re gulated b y e xtracellular signaling e vents. In response to

Overview of Molecular and Cell Biolo gy

signal-mediated posttranslational modif ications, the biological acti vities of proteins are often modif ied to effect rapid, transient cellular responses to these changing environmental signals. P osttranslational modif ications can modulate protein localization, interactions with other macromolecules (e g, with other proteins, DNA, RN A, cellular matrix), enzymatic acti vity, and protein stability. The ability to image transient, functionally signif icant protein modif ications, and their immediate consequences (e g, altered protein-protein interactions, binding to target DN A sequences, enzymatic acti vity) in li ving indi viduals in e xperimental models is a major target of current noninvasive imaging research programs.19–22 A number of these questions are dealt with in substantially more detail in other chapters in this book. Alterations in the mechanisms of re versible, transient posttranslational protein modif ications and the consequent alterations in the biolo gical functioning of these proteins pla y key roles in the man y diseases. Our ability to nonin vasively monitor , in response to therapeutic interventions, alterations of posttranslational protein modifications is another major goal for nonin vasive imaging, both in the preclinical, translational research arena and in the clinical realm. In addition to regulating protein expression by modulating their translation and re gulating protein function and/or location b y posttranslational modif ication, cells alter the quantity of indi vidual proteins by altering their rates of de gradation. The intrinsic half-li ves of indi vidual proteins vary from minutes to da ys. However, modifications of some proteins in response to appropriate extracellular signals and consequent protein posttranslational modif ications can alter their de gradation rates by orders of magnitude. Although proteins can be degraded by several mechanisms, the most well-studied mechanism is the process of ubiquitin-mediated protein degradation. Ubiquitin, a 76 amino acid protein, is found in all eukar yotic cells. Proteins tar geted for de gradation b y the ubiquitinmediated pathway are covalently modified by the attachment of a string of ubiquitin molecules to specific lysine residues, to produce polyubiquitinated pr oteins. The polyubiquitination reaction is catal yzed b y indi vidual members of a f amily of hundreds of ubiquitin lig ases, enzymes that ha ve remarkab le tar get specif icity for identifying distinct proteins for degradation. Polyubiquitinated proteins are recognized by a multiprotein cellular protease “machine, ” the proteasome that de grades the ubiquitinated tar get protein and retur ns the ubiquitin molecules to the cellular monomeric ubiquitin pool.

611

Ubiquitination of a tar get protein and subsequent proteasome degradation in response to a cellular signal can be modulated either (1) b y re gulated posttranslational modif ication (e g, phosphorylation) of the tar get protein that now makes it recognizable by a constitutive ubiquitin ligase or, alternatively, (2) by a regulated posttranslational modif ication of the appropriate ubiquitin ligase, increasing its ability to ubiquitinate an unmodified protein substrate. Chimeric fusions betw een proteins of interest (e g, the I κB re gulator of NF- κB transcription f actor acti vation,23 hypoxia-inducible factor,24 the p27 c yclin kinase inhibitor [ CKI]25) and luciferase have become useful tools for the nonin vasive imaging of alterations in proteasome-mediated protein degradation in response to cellular signals. These studies demonstrate the bur geoning power and potential of this technology. The ubiquitination process for tar geted protein de gradation has been most completely examined in the context of regulation of the cell di vision cycle (vida infra). However, altered functioning of the ubiquitin-proteasome pathway for regulated protein degradation is postulated to play a role in a v ariety of diseases, including Huntington’ s disease, Alzheimer’s disease, P aget’s disease, P arkinson’s disease, and v on Hippel Lindau disease. De velopment of tools to monitor targeted protein degradation in the clinical context is an important, but as yet essentially unexplored, target. Another k ey re gulated protein de gradation e vent in cells is the initiation of apoptotic cell death through the activation of the caspase proteolytic pathw ay. Targeted protein de gradation during apoptosis (see section “The Life Cycles of Cells—Embryology, Differentiation, Development, and Death”) is also a major topic of in vestigation in cell biology; caspase-mediated apoptosis pla ys a major role both in development of normal tissues and in response to cellular injur y and insult. In contrast, inhibition of caspase-mediated apoptosis is a major contributor to the progression of many cancers. Our ability to nonin vasively monitor the degradation of individual proteins via both the ubiquitin-proteasome mediated pathw ay and the caspaseapoptosis pathway during the de velopment of disease and in response to presumpti ve therapies should contribute greatly to our understanding of both disease processes and to their therapeutic responses. 25–30

THE CELL DIVISION CYCLE Cells in our bodies exist in three states with regard to cell division. (1) Cells that are al ways in the cell c ycle—cells that are al ways di viding (e g, intestinal cr ypt cells), (2) postmitotic cells that will never divide again (eg, platelets,

612

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

red b lood cells, man y neurons), and (3) cells that are “resting,” but can be induced b y appropriate, often celltype specif ic, signals to di vide (e g, hepatoc ytes in response to hepatectom y, antigen-responsi ve B and T cells, mammary cells during pre gnancy). To go from one to two cells, a cell must do two things: (1) double its content of DNA and (2) di vide its replicated DN A equall y between two siblings. Mitosis, the process of ph ysical division of one cell into two, is the most easily observable phenomenon in tissues when we use a microscope; mitotic cells are round , shiny, and refractile because the microtubule network that forms the cytoskeleton has reformed to create the spindle apparatus necessary for the chromosomes to separate. The DNA content of the cell is divided to the sibling cells during mitosis. Although mitosis is di vided into a series of complicated and w ell-investigated biochemical and morphological stages, the y are not rele vant in the conte xt of noninvasive molecular imaging at the or ganismal le vel and will not be discussed in this chapter . Mitosis, the process of chromosomal se gregation and cell di vision, usually takes from 30 to 60 minutes, star t to f inish. The period between one cellular mitosis and the next mitosis is the interphase period; interphase may last anywhere from 4 hours to da ys for somatic cells. The questions w e will address in the rest of this section are “Ho w does the cell prepare to duplicate its DNA, when and how does it duplicate the DNA, and how does it prepare for mitosis?” “Pulse labeling” of exponentially growing populations of cells with a radioactive DNA precursor such as tritiated thymidine for a brief (e g, 30 minutes) time, follo wed by washing, f ixation, and autoradio graphy, re veals that no mitotic cells (recognized by their distinctive round, refractile mor phology) incor porate radioacti ve label, demonstrating that mitotic cells do not synthesize DN A. In contrast, some—but not all—inter phase cells (ie, flat, adherent cells) become labeled during this brief e xposure to radioactive thymidine. This result suggests that there is a discrete period during inter phase when DNA synthesis occurs. The period of inter phase during w hich DN A is replicated is called the S phase. By labeling cells for a brief period of time with a radioacti ve DNA precursor, removing the label, and then monitoring b y autoradiography the times at w hich cells with incor porated label (ie, cells that were in the S phase) enter into mitosis (distinguished morphologically as new mitotic cells that are now radioactively labeled), we can determine that there is a gap betw een the time at w hich the S phase ends and mitosis be gins and a second gap betw een the time at w hich mitosis is f inished and S phase be gins. The gap betw een the conclusion of mitosis and the initiation of the S phase is called g ap 1 or

G1 phase; the gap betw een the end of the S phase and the beginning of mitosis (the time during w hich the cell prepares to di vide its chromosomes to daughter cells) is called gap 2 or G2 phase. The four cell cycle phases, G1, S, G2, and M are shown in Figure 3. As a result of traversing the cell cycle, the amount of DNA in the G2 phase nucleus—prior to distribution to the daughter cells in mitosis—is twice that of the G1 nucleus. To compact the increased DNA into the chromatin of the segregating chromosomes, the cell must also doub le its content of histones. F or most cell types, histone mRN A and protein synthesis are tightly regulated, occurring only during the S phase—that is, histone and DN A synthesis occur concomitantly during the cell cycle. “Resting” cells, that is, cells that are not in the process of traversing the cell division cycle but can reenter the c ycle in response to general or cell-specif ic growth stimulating agents, are generall y in a quiescent stage ter med “G 0.” G 0 cells ha ve a G 1 complement of DNA; as cells traverse the early portion of G 1, they must make a decision w hether to commit to another round of DNA replication and mitosis or to enter the G0, quiescent state and wait for a command to reenter the proliferation cycle. The mechanisms and signal transduction pathways by w hich growth factor s and other mitogenic stimuli induce resting, G 0 cells to reenter the cell c ycle will be discussed later in the chapter . (see section “Outside to Inside the Cell; Signal Transduction.”) The questions with re gard to cell c ycle that are of greatest concern to cell biologists, and whose answers we would like to be able to monitor in living individuals are: “How are the transitions from the v arious cell c ycle phases controlled?” “What are the w ays in which proper progression through the cell c ycle is monitored and , if necessary, by which errors are corrected?” “What are the consequences for the cell, for the organ, and for the individual if mistakes occur during cell cycle progression?” Research in the 1980s identif ied the k ey molecules that regulate the progress of cells through the cell c ycle. A protein kinase (an enzyme that transfers phosphate from ATP to a target protein substrate) required for the G2 to M transition was identified by a combination of genetics and biochemistr y. Subsequently, it became clear that this protein kinase was a member of a class of structurally related protein kinases w hose distinct but related functions are required at critical cell c ycle transition points; for the G1 to S transition, for continued DNA synthesis in S phase, for the S to G 2 transition, for the G2 to M transition, and for man y of the sequential steps in mitosis involved in chromosome/spindle or ganization and chromosome separation. Although their enzymatic

Overview of Molecular and Cell Biolo gy

activities are onl y required at the times of the cell c ycle transitions they regulate, these protein kinase molecules are present at essentiall y constant le vels throughout the cell c ycle. Their constituti ve presence and temporall y restricted enzymatic activities lead, of course, to the question of how their kinase activities are turned on and off at appropriate times during the cell c ycle. About the same time that the protein kinases that regulate cell cycle progression were being uncovered, a class of molecules named cyclins, because of their cyclic accumulation and disappearance during pro gression through the cell c ycle, were identif ied. Genetic studies predicted and biochemical studies conf irmed that c yclins and the kinases described abo ve, w hich are no w called cyclindependent kinases (or CDKs), for m heterodimeric complexes in which the catalytic activity of the kinase subunit is active; in the absence of its c yclin partner, the CDK is enzymatically inacti ve. Thus, the c yclic e xpression and interaction of the cyclin regulatory subunits are responsible for the transient catal ytic acti vities of the co gnate CDKs at the proper stage in the cell c ycle. Distinct cyclins (eg, G 1 cyclins, S cyclins, G 2 cyclins) have peak expression values at different phases of the cell c ycle, to activate the appropriate CDKs required for the appropriate cell cycle transition. A lar ge—and e ver g rowing— catalog of CDK/cyclin complexes whose transient protein kinase acti vities re gulate cell-c ycle transitions has been described.31 The kinase reactions that modulate cell c ycle transitions must be rapid , accurate, highl y ef fective, and reversible. Cyclin accumulations occur o ver a relati vely extended time, within the conte xt of cell c ycle parameters. Consequentl y, c yclin-CDK comple xes be gin to assemble before the kinase acti vity required for a shar p cell cycle transition is necessary. To provide a more temporally precise acti vation mechanism, the kinase acti vities of the CDK enzymes in the CDK/c yclin complexes are re gulated b y tw o additional mechanisms. The catalytic kinase subunits of the CDK comple xes are subject to regulatory phosphorylation and dephosphor ylation by specific CDK kinases and phosphatases; the addition and removal of these phosphate g roups modulate the kinase activity of the CDK/cyclin complex. In the early 1990s another group of proteins, the CKIs were identified. CKI proteins bind to what should be enzymatically active CDK-cyclin heterodimers and inhibit their kinase activity. Before the CDK-c yclin can phosphor ylate its substrate, the CKI must be remo ved/degraded. Thus, a temporal balance betw een CDK-c yclin heterodimerization, phosphor ylation of the catal ytic kinase domain, and removal of inhibitor y CKI molecules re gulates the CDKdependent control of cell cycle progression (Figure 3).

613

While there are man y CDK-c yclin comple xes that operate to control steps in cell c ycle progression, several of them are k ey, for our interests. The G 1 cyclins—in man, primaril y c yclin D—acti vate CDK4, leading to phosphorylation e vents that acti vate the synthesis of a number of proteins that prepare the cell to enter S phase. One of these proteins, c yclin E, activates CDK2; CDK2cyclinE drives the cell into S phase.Another cyclin whose synthesis is induced in the CDK4/c yclinD response, Cyclin A, forms a complex with CDK2 to initiate the G 2 to M transition. The mitotic c yclins (e g, Cyclin B), in concert with CDK1 (the first CDK to be identified), promote the transition through the v arious phases of mitosis (eg, chromosome condensation, spindle for mation). Cyclins come and go during the cell c ycle; so do the CKIs. And so do other proteins that regulate cell cycle transitions. Ho w are the c yclins, CKIs, and other cellcycle mediators eliminated at the appropriate times? Many cell cycle regulatory molecules that modulate CDK activity and pro gression through the cell c ycle are degraded by cell-cycle regulated ubiquitination and subsequent proteasomal proteol ytic de gradation. There are two ubiquitin ligases that mark cell c ycle regulatory proteins for timed proteolytic degradation; they are both dedicated protein machines. Each of these “ubiquitination machines” has s witchable modular components that recognize appropriate targets as they appear and are modified to identify them for de gradation (for instance, b y phosphorylation) during the c ycle. One of these multiprotein component ubiquitin ligases, the Skp1-Cullin-F-box protein (SCF) complex, is responsib le for ubiquitination of many of the cyclins, for many of the CKIs, and other cell cycle re gulatory molecules. F ollowing ubiquitination b y the SCF complex, these cell cycle regulatory proteins are targeted to and de graded in the proteasome. The F-box proteins recognize their target proteins once the targets are marked for de gradation and then bring them to the SCF for ubiquitination. We now know there are dozens of Fbox proteins, each of w hich promotes the de gradation of its target protein(s) as the cell pro gresses through the cell cycle, through SCF complex-mediated ubiquitination and subsequent delivery to the proteasome. The anaphase-promoting complex, or APC, the second cell c ycle multiprotein-component ubiquitin ligase, ubiquitinates the mitotic c yclins, man y of the proteins that are involved in chromosome alignment prior to chromatid separation, and other substrates. Lik e the SCF complex, the APC has modular protein elements that target cognate substrates to the APC E3 ligase for ubiquitination and proteasomal de gradation, once these proteins are biochemicall y mark ed for de gradation. The APC is also known as the cyclosome.

614

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 3. The cell division cycle. (upper panel) Cells resting in the G0 state, when stimulated by mitogens/growth factors, enter the cell cycle. After preparing for DNA synthesis in G1, they replicate their DNA in S phase before preparing for cell division in G2. After proceeding through mitosis (M) and entering into G1, the cells reach a “restriction” point where they make a decision to commit to another cell cycle or, alternatively, to exit to the G0 compartment. The transitions from one cell cycle phase to the next are controlled by cyclin-dependent protein kinases (CDKs) whose catalytic activity is regulated by the cyclins. In mammalian cells, distinct CDK proteins activated by cell cycle phasespecific cyclins regulate the various cell cycle transitions. Checkpoints, which examine the integrity of cell cycle progression, exist at a variety of points as cells traverse the division cycle. (middle panel) The catalytic subunits of the CDKs are present at all times. Their activity is regulated in four ways; by association with the appropriate cyclin at the appropriate time in the cell cycle and timed degradation of the G1-, S-, and G2/M-phase cyclins at the appropriate cell cycle times, by phosphorylation at some sites on the CDK to activate catalytic activity and dephosphorylation to inactivate the kinase, by dephosphorylation at other sites on the CDK to activate catalytic activity and phosphorylation to inactivate the kinase, and as a result of inhibition by cyclin kinase inhibitors (CKIs). (lower panel) Ubiquitination and subsequent proteasomal degradation removes cyclins, CKIs, and many other cell cycle regulatory proteins at appropriate times in the cell cycle. Proteins to be destroyed during cell cycle progression are targeted for ubiquitination and degradation by two “machines” in the cytoplasm, the anaphasepromoting complex (APC; also known as the cyclosome) and the Skp1-Cullin-F-box protein (SCF) complex. Both the APC and the SCF complexes are multiprotein machines that have components that recognize target molecules at the correct time in the cycle and present these substrates to the ubiquitin ligase machinery of the corresponding complex. Following ubiquitination, the substrates are degraded by the proteasome.

Overview of Molecular and Cell Biolo gy

The importance of proteolysis of the molecules that drive the cell through the cell cycle cannot be overstated; degradation of these components of the cell c ycle regulatory machinery guarantees that the cell c ycle will be ir reversible. The combination of phosphor ylation reactions to re gulate cell c ycle transitions and proteol ysis to ter minate, in an unequi vocal f ashion, inhibitor y and catalytic steps regulating cell cycle progression is a continuing area of investigation. The c yclin/CDK mechanism of cell c ycle pro gression has been conser ved from unicellular eukar yotes (eg, yeast) to man. In addition to the CDKs and c yclins that are involved in the cycles of all eukaryotes, multicellular or ganisms ha ve e volved both specialized CDKcyclin systems that modulate cell c ycle transitions in differentiated cell types and CDK-c yclin systems that participate in the control mechanisms that mediate the signaling systems that direct resting, G 0 populations of cells to reenter the cell c ycle. Clearly, errors and aberrations in the cell cycle could be catastrophic for the cell and for the or ganism. Consequently, a number of cell cycle checkpoints have evolved to provide mechanisms by which the cell can either correct errors or eliminate cells prior to completion of a cell cycle that locks an error into the genetic apparatus of subsequent progeny. One such mechanism is the DNA damage c heckpoint. Several molecular mechanisms e xist to determine whether DNA damage has occur red, either as a result of er rors during DNA replication in other phases of the cell c ycle (or in the DN A of resting cells). DN A replication during the cell division cycle is delayed and/or stopped until the damage can be repaired and replication is then per mitted to continue. If the damage is too g reat, the cell is initiated down a pathway leading to apoptosis. Postreplication DNA integrity is monitored b y the chk1 gene; mutation of this gene will result in aber rant DNA repair. At the G1 or restriction checkpoint, proliferating cells assess their en vironment—including nutrients, signals from their neighbors, and signals from other re gulatory sources—to deter mine w hether the y should commit to another cycle or should shunt off to the resting, G 0 stage. Clearly, if this checkpoint f ails and cells that should not divide do not exit the cell cycle, the possibility for uncontrolled g rowth is g reatly ele vated. One of the CKI inhibitors, p16, pla ys a major role in this “decision, ” blocking the acti vity of CDK4/c yclinD mediated transition to prepare for S phase. The p16 gene is an e xample of a tumor suppressor g ene; loss of both copies of this gene will result in a f ailure to properl y inhibit cellular proliferation.32,33 Clearly, a nonin vasive w ay to monitor

615

the presence and acti vity of the p16 CKI w ould be of great value in the study of tumor biolo gy.34 The G2 damage checkpoint occurs at the end of the G2 phase. If cells are not properl y prepared for mitosis, they will not be allo wed to pass through this bar rier and enter mitosis. In order for the mitotic CDK/c yclin complex to become active, the CDK catalytic subunit must be dephosphorylated. The phosphatase that acti vates this kinase ser ves as a checkpoint molecule; it will not dephosphorylate the kinase until a series of biochemical criteria indicating G 2 completion has been met. Once cells are in mitosis, another checkpoint, the mitotic spindle c heckpoint, or metaphase/anaphase c heckpoint, determines whether all the chromosomes are properly aligned along the mitotic plate. If all is w ell, degradation of cyclin B b y ubiquitination and subsequent proteasomal degradation occurs, per mitting chromosomal se gregation, and the completion of mitosis and entr y into G 1.

TRANSPORT ACROSS MEMBRANES Although an impor tant topic in cell biolo gy, membrane transport often does not have the emphasis in cell biology textbooks that w e f ind for man y of the other topics discussed in this book; for example, cell cycle, regulation of gene expression, and signal transduction. However, in the context of this compendium on molecular imaging, transport takes on a more prominent role; man y of the probes used in molecular imaging must tra verse the cell membrane to gain access to their tar gets. Although often not appreciated, or at least accounted for in pub lications, limitations of probe access rather than tar get concentration can certainly modulate and even be the determining factor in the strength of the imaging signal. This chapter is intended primarily for non-biologists. For chemists synthesizing probes and for ph ysicists and engineers designing new imaging instrumentation, an appreciation for the biology of intracellular probe access to target at the molecular level is an impor tant corollary to pharmacokinetic and pharmacodynamic considerations such as probe metabolism, probe b lood half-life, and probe elimination pathways. For those cell biolo gists reading this chapter— and, unfortunately, perhaps even more importantly for those cell biologists who do not read this chapter—it is important to reco gnize the limitations of intracellular probe access in designing animal models and e xperimental protocols for preclinical imaging in vestigations. It is critical that chemists, biolo gists, and instr ument designers, as w ell as modelers, be a ware of the mechanisms and limitations underlying probe access to target.

616

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

The cell membrane is essentiall y a lipid bila yer containing a number of embedded proteins. Although gases (eg, O 2, CO 2) and some small unchar ged molecules (e g, urea, ethanol) can pass through the plasma membrane b y simple dif fusion, man y (indeed , most) small h ydrophilic molecules, which constitute a large proportion of the probes used in molecular imaging, cannot pass through the membrane by this mechanism. Moreover, while some small molecules move from a higher e xtracellular concentration to a lower intracellular concentration w hen the y pass through the cell membrane, other molecules must mo ve “uphill”; they must be transpor ted from an en vironment of lo wer concentration to an environment of higher concentration (a primary example of this is the ability of some cells to pump out chemotherapeutic agents against a g radient). How do the h ydrophilic ions, nutrients, substrates, and other molecules required for cell function tra verse the plasma membrane and gain access to the c ytoplasm, if not b y diffusion? And as a corollary, how do many of the probes used in molecular imaging make this journey? Transport of small molecules across the plasma membrane and into the interior of the cell is usually medi-

ated b y specialized membrane proteins. Ev en for those molecules that can diffuse through the plasma membrane, the rate at which they can enter cells is often not sufficient to meet the biolo gical demand; transmembrane transpor t proteins are present that help increase the transpor t of these molecules across the plasma membrane b y facilitated dif fusion (Figure 4). In addition to facilitated diffusion (also known as facilitated transport), we will discuss active transport and cotransport (also known as secondary active transport). Until relati vely recentl y, man y in vestigators w ere of the opinion that most dr ugs enter the cell b y passi ve membrane diffusion; indeed the “Lipinski r ule of 5” 35 for predicting drug transport into cells is based on the properties of dif fusion of compounds across a lipid membrane. Since many molecular imaging probes are based on dr ug structures or their deri vatives, it has been assumed that at least some probes enter and e xit the cell membrane b y simple diffusion across the membrane. However, since the advent of “e xpression cloning, ” hundreds of cellular transporters ha ve been identif ied, and man y dr ugs ha ve been sho wn to enter cells b y f acilitated transpor t, acti ve

Figure 4. Membrane transport proteins. “Exterior” refers to the luminal side of the plasma membrane, “interior” refers to the cytoplasmic side of the plasma membrane. Concentration gradients are indicated by the triangles; the tips show the side of the membrane with the lower concentration. Pumps use coupled adenosine triphosphate (ATP) hydrolysis to drive movements of small molecules or ions against an electrochemical gradient. Channels provide a mechanism for moving ions against a gradient. In the case shown, a “gated” channel can exist in the “open” or “closed” state; cells can regulate the open or closed state of the channel. Uniporter transporters facilitate movement of substrates down a gradient. Symporters facilitate the movement of one substrate against a gradient by coupling this reaction with movement of a substrate in the same direction, taking advantage of a gradient in the opposite direction. Antiporters move their substrates in opposite directions across the membrane; the driving substrate moves down the gradient, while the driven substrate moves up its gradient.

Overview of Molecular and Cell Biolo gy

transport and/or cotranspor t. (Expression cloning is a technique in w hich complementar y DN As (cDN As) are used to e xpress proteins in cells and the cells e xpressing desired clones are identified by their production of the corresponding protein.) By analo gy, we have every expectation that these modes of entr y into cells will be tr ue for molecular imaging probes as well. Glucose and its positron emission tomo graphy (PET) probe analo g fluorodeo xyglucose (FDG) are transported into cells b y f acilitated diffusion (or f acilitated transpor t). The GLUT1 glucose transpor ter is a transmembrane protein that for ms a channel in the plasma membrane. By vir tue of its size, shape, and charge, this channel allows glucose to pass do wn a concentration gradient through the membrane, without glucose having to penetrate the lipid bilayer of the plasma membrane, a much slo wer process. Thus, the rate of facilitated dif fusion is much g reater than is the rate of diffusion through the lipid bilayer. The rate at which substrate is transpor ted into cells b y facilitated transport is a function of the number of transpor ter molecules in the membrane, the properties of the transporter, and the concentration gradient between the outside and the inside of the cell. Each transpor ter of this type is relati vely specific, transpor ting onl y a single molecule or class of molecules with g reat str uctural similarity; for e xample, glucose and FDG. Most, but not all, mammalian cells use glucose in the b lood as their source of ener gy and transport glucose into the cell via GLUT1. The GLUT1 transpor ter f aces outw ard and has an unoccupied glucose binding site. When glucose binds, the transporter undergoes a conformational shift that now positions the bound glucose in a site that f aces inward, toward the c ytosol. Once the bound glucose molecule is released into the c ytosol, the confor mation of the transporter shifts again to position the binding site to the luminal side of the plasma membrane. The direction of facilitated transpor t is entirel y concentration dependent; if the intracellular glucose concentration is g reater inside the cell than outside the cell, glucose will be moved in the opposite direction. The kinetics of glucose transpor t can be described in the same way that enzymatic reactions are described; this anal ysis is used in the compar tmental models that describe PET metabolic rates. All uniporters for facilitated diffusion share these characteristics. Many small molecules must be transpor ted into or out of a cell in the face of a concentration gradient opposite to the desired direction of transpor t. In this case, the cell uses two processes, active transport and cotransport (or secondary active tr ansport). Active transport of ions and small molecules against a concentration g radient is

617

carried out by transmembrane proteins that use ATP-generated energy to “pump” their substrates across the membrane in the face of the gradient. The best known of these active transpor t pumps is the sodium-potassium pump, which maintains the lo w sodium/high potassium characteristics of the cell in the f ace of opposing g radients between the cytosol and the exterior. The ABC active tr ansporters (ATP-binding cassette transporters) are a superfamily of transporters for sugars, amino acids, phospholipids, cholesterol, peptides, pol ysaccharides, and proteins. The name of the superfamily is derived from the sequence/structure of their ATP-binding domains. The multidrug r esistance pr otein (MDR), or P-glycoprotein, is a member of the ABC transpor ter family. The apparent role of this transpor ter is to remove toxic substances from cells; no “natural” substrate has been identif ied to date. The MDR transporter is w ell known in cancer research; it—unfor tunately—pumps many chemotherapeutic compounds out of cells in the face of a higher concentration outside the cell, thus “detoxifying” the tumor cell b y removing the chemicals designed to kill them. Amplification (the development of multiple copies) of the MDR gene in cancers and consequent overexpression of this protein leads to dr ug resistance during cancer chemotherap y. A number of radionuclide-labeled PET and single photon emission computed tomo graphy (SPECT) probes and galliumlabeled magnetic resonance imaging agents ha ve been described for nonin vasive functional imaging of MDRdependent dr ug resistance in tumor therap y.36–38 The related ABC transporter ABCG2/BCRP appears to have a role in D-luciferin transport,39 suggesting that the users of the f irefly luciferase repor ter gene need to be co gnizant of this potential limitation in inter pretation of optical imaging results. To our knowledge, no other ABC transporter is yet to be functionally imaged in living animals. In contrast to acti ve transpor ters, w hich use ATPdriven reactions to mo ve substrates across membranes against a concentration g radient, symporters and antiporters use transport of one molecule down a gradient to drive transport of a second molecule against a gradient. Symporters drive both molecules in the same direction; in this case, the g radient for the cotranspor ted molecule and the substrate of interest are in the opposite directions (see Figure 4). For antiporters, the substrate of interest and the cotransported molecule are dri ven in opposite directions since the g radients for the tw o molecules are in the same direction. This type of transpor t is refer red to as cotransport or secondary active tr ansport. The sodium/iodide symporter is among the best characterized PET repor ter genes used for nonin vasive tracking of v ector and

618

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

cell-mediated therapies.40–42 Hundreds of articles describing PET and SPECT studies of the norepinephrine, serotonin, dopamine, and 5-h ydroxytryptamine sympor ters have also been published.43–48 Similarly, radiolabeled probes for PET and SPECT imaging for a number of monoamine antipor ters and the acetylcholine antipor ters have been described. 49 Finally, it should be pointed out that, in man y cases, individual drugs have been repor ted to be transpor ted by more than one transpor ter.50 So, w e should not be surprised, as we discover more about the comple xity of drug transport, to lear n that there are similar comple xities in understanding the transport of molecular imaging probes. Thus, for e xample, the literature repor ts that the nucleoside analo gs ac yclovir and ganciclo vir are both transported into cells by two transporters.51,52 We should not be surprised, therefore, to lear n that the radionuclide-labeled probes used for PET and SPECT anal ysis of the HSV1thymidine kinase repor ter gene, based on these ac ycloguanosine structures, will also be taken up by multiple membrane transporters. Indeed, the increasing elucidation of transporter genes and proteins, and the small molecules whose transpor t the y mediate, should help suggest ne w imaging probes both for imaging of endogenous transport processes and for imaging repor ter genes to be used both for transgenic mouse models as w ell as for tracking of cell-based and vector-based gene therapies.

OUTSIDE TO INSIDE THE CELL: SIGNAL TRANSDUCTION All cells respond to chemical and ph ysical signals “from the outside”; even unicellular organisms respond to extracellular signals such as pH, ion concentrations, nutrient levels, temperature, pressure, and radiation. In multicellular organisms, cells talk to one another both via chemical messengers produced by one cell and reco gnized by another and via cell-to-cell contact-mediated signaling. Two distinct types of cell responses can be elicited either in response to chemical messengers or in response to cell-to-cell signaling. Some cellular responses to e xtracellular signaling do not require gene e xpression. They elicit alterations in proper ties of pree xisting targets; that is, posttranslational protein modifications of existing proteins, the stability of e xisting mRN As, exocytosis and endocytosis. For these cellular responses to e xtracellular chemical messengers and to cell-cell signaling, no ne w gene expression occurs. Other signaling responses, which occur over a longer time frame, require changes in gene expression; that is, either increases or decreases in tar get gene transcription rates. In a gene-e xpression mediated

response to an e xtracellular signal, infor mation must reach the nucleus to cause an alteration in the biolo gy of the cell. Molecular imaging is used both to nonin vasively monitor signal-induced changes in nor mal biochemical and physiological processes and to noninvasively monitor cellular pathologies that result from genetically or xenobiotically altered signaling pathways. Hormones, neurotransmitters, cytokines, and growth factors are among the many classes of “chemical messengers” that cells use to con verse with one another . Hormones f all into tw o classes; small molecules (e g, glucocorticoids, estro gen, pro gesterone, prostaglandins) and protein hormones (eg, insulin, parathyroid hormone). Growth f actors (e g, epider mal g rowth f actor [EGF], fibroblast g rowth f actor [FGF], platelet-deri ved g rowth factor [PDGF], transforming growth factor β [TGFβ]) are proteins that regulate cell proliferation and/or dif ferentiation. Cytokines (e g, the v arious interleukins) w ere originally described as proteins that modulate de velopment and function both of antibody-producing cells (B cells) and of cells that mediate cellular immunity (T cells). As we learn more about the functional di versity of indi vidual g rowth f actors and c ytokines and as w e identify more of these protein chemical messengers, the definitions/classifications of growth factors and cytokines increasingly blur and overlap. Chemical messengers such as hor mones, g rowth f actors, and c ytokines ma y act o ver long distances, via the blood stream. Consequently, chemical messenger signaling systems must ha ve g reat sensiti vity; tar get cells must be able to respond to very low circulating chemical messenger concentrations. Steroid hormones and other small-molecule chemical messengers are effective at concentrations of l0−10 to 10−9 M. Protein hormones, growth factors, and cytokines typically operate at even lower concentrations, 10 −11 to 10− 10 M. Although many hormonal and growth factor-mediated events take place over long distances, chemical messenger signaling actions may also be targeted to adjacent cells. For hormones, signaling at a distance and signaling to adjacent cells are, respecti vely, refer red to as endocrine and paracrine signaling. Cells ha ve tw o types of receptors for ligands (an alternative ter m often used for chemical messengers). Some small-molecule chemical messengers (e g, steroid hor mones, retinoic acid deri vatives, vitamin D) are h ydrophobic. Hydrophobic chemical messengers are carried in the blood by carrier proteins. Although h ydrophobic ligands can dif fuse into the cell, more and more transpor ters that f acilitate the entry of these molecules into cells are being identif ied (vida supra). The receptors for h ydrophobic chemical

Overview of Molecular and Cell Biolo gy

messengers are usuall y intracellular (F igure 5). Hydrophobic signaling molecules bind to their appropriate intracellular receptors, either in the cytoplasm or the nucleus, depending on the specif ic ligand-receptor system. For systems in w hich the unoccupied receptor is in the cytoplasm, ligand binding activates a series of reactions that results in the translocation of the ligand-receptor comple x into the nucleus (e g, binding of cor tisol and its analo gs to the glucocor ticoid receptor). These comple xes of small-molecule ligand and co gnate receptor are ligand-dependent tr anscription factors; they increase or decrease gene expression. Monitoring the le vel and location of such receptors

619

with imaging probes that are analogs of the endogenous ligands is a major molecular imaging application. 53,54 Other chemical messengers (e g, protein hor mones, growth f actors, c ytokines, neurotransmitters) are hydrophilic ligands; the y are unab le to pass through the cell’s lipid bila yer plasma membrane. F or these ligands, cells ha ve transmembrane r eceptors—integral transmembrane recognition proteins for the ligands—imbedded in the cell’s plasma membrane (F igure 5). Two questions should immediately be apparent: (1) Ho w does the infor mation imparted b y ligand interaction with the receptor at the membrane get inside the cell? (2) For responses that require gene expression, ho w does the infor mation get from the

Figure 5. Outside to inside the cell: signal transduction. Examples of signal transduction pathways that lead to changes in gene expression in response to hydrophobic ligands that signal through intracellular receptors, hydrophilic ligands that signal through G-protein coupled receptors (GPCRs) and hydrophilic ligands that signal through receptor tyrosine kinases. Cell-cell activated signaling by membrane-bound ligands and by gap junctions are also illustrated. Selected examples of guanosine triphosphate (GTP)/guanosine diphosphate (GDP)-mediated signaling (G-protein, Ras), second messenger signaling (eg, cyclic adenosine monophosphate (cAMP) activation of protein kinase A) and protein kinase cascades (Raf-MEK-MAPK-ERK) are also shown. Because so much of interest in signal transduction is related to stimulation of resting cells to enter the cell proliferation cycle as a result of new gene expression, particularly with respect to molecular imaging applications, this point is emphasized in the figure. Coated vesicle/endosomal uptake, delivery to lysosomes and degradation of growth factor receptor and ligand are included to remind readers that pathways other than ligand-mediated kinase activation are postulated to play a role in receptor tyrosine kinase signaling in response to cognate ligands.

620

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

receptor-ligand complex at the membrane to the nucleus to alter gene e xpression? Man y of the same signaling pathways acti vated in response to endo genous chemical messengers are also modulated b y e xogenous xenobiotic agents, often with consequences that are deleterious to the cell, organ, and individual. We find frequently that the same ligand can elicit different responses in dif ferent cell types. F or e xample, acetylcholine can cause muscle cells to contract, but pancreatic acinar cells secrete am ylase when stimulated b y acetylcholine. Why the dif ference? Because of their distinct de velopmental histories, dif ferent cell types ha ve distinct reper toires of pree xisting proteins and distinct repertoires of genes that respond to a common ligandreceptor induced intracellular signal. To restate, the response of a cell to a ligand-receptor acti vated intracellular signaling pathw ay depends both on the signal generated and on w hat protein tar gets of that signaling are present inside the cell. Chemical messengers must do tw o things; (1) the y must bind to their receptors and (2) the y must cause the ligand-receptor comple x to generate the intracellular signal. Consider a molecule that can bind to a receptor but cannot elicit the intracellular signal. Such a molecule is an antagonist; it will compete for and block the activity of the normal ef fector ligand (hor mone, g rowth f actor, neurotransmitter, c ytokine). Molecules that bind to receptors and acti vate the intracellular signal transduction mechanism are agonists. The actions of agonists and antagonists represent the crosso ver of phar macology, biochemistr y, endocrinology, ph ysiology, neurolo gy, oncolo gy, infectious diseases, etc. Imaging probes for receptors, lik e endogenous ligands, agonists, or antagonists, must bind to the target receptor. Ideally, imaging agents, while interacting with cellular receptors, should not per turb the biology of the cell. Cells also signal to one another via cell-cell contact. Recall that small molecules can be passed from one cell to another via gap junctions in adjoining membranes. Cells also contain membrane bound ligands in addition to membrane-spanning receptors; interactions of membrane bound ligands (e g, delta) on one cell with co gnate membrane bound receptors (eg, notch) on an adjacent cell can activate intracellular signal transduction pathw ays. These types of cell-cell interactions are usuall y de velopmental cues that regulate cell fates and are usually not involved in transient, reversible responses to environmental changes. We have now reached the point w here we will consider specif ic types of intracellular receptor -mediated signaling pathways. Figure 5 is a very simplified diagram of alter native signal transduction pathw ays that signal

from cell surface receptors, following either ligand binding or as a result of cell-cell interactions, to the nucleus. Two e xamples of solub le ligands and one of cell-cell mediated signaling will ha ve to suf fice to indicate the general nature of signal transduction pathw ays following ligand-receptor interaction. Interested readers can consult a wide variety of texts and Web sites on the topics shown in the illustration and refer red to below. The G-protein coupled r eceptors (GPCRs) are “seven membrane pass” proteins (ie, the single pol ypeptide chain of GPCRs thread back and for th through the plasma membrane se ven times). In response to ligand binding, GPCRs engage an intracellular membranebound heterotrimeric G-protein complex composed of Gα, Gβ, and Gγ protein subunits. Binding to the ligandreceptor comple x initiates a confor mational shift that allows the G α protein component of the G-protein complex to discard its bound guanosine diphosphate (GDP), bind guanosine triphosphate (GTP), dissociate from the ligand/receptor/G-protein complex and bind to and activate a “downstream” enzyme. Depending on the par ticular ligand , GPCR, and G-protein comple x, the downstream ef fector ma y be an adenylate cyc lase, a phospholipase or an ion c hannel. These downstream effectors, in response to activation by the interaction with the GTP-bound G α subunit, either catal yze the production of small molecules (e g, cyclic AMP, diacyl glycerol, inositol triphosphate), or modulate intracellular ion concentrations ( calcium ion influx ). These small molecules then activate signal transduction cascades of sequential enzyme reactions that eventually lead to the activation of preexisting, latent transcription f actors to stimulate ne w gene transcription and protein production. Historicall y, the small molecules (cAMP , D AG, IP3) produced b y GPCR acti vation ha ve been called second messeng ers, because they propagate the activation of the signal transduction pathways initiated by the primary chemical messengers. Signal transduction pathw ays should be re versible; cells need to be ab le to “turn off ” the response when the signal is removed/reduced. The Gα GTP-binding subunit is also an enzyme; a GTP ase. When the G α GTPase activity clea ves GTP, the G α subunit lea ves its do wnstream effector (eg, adenylate cyclase, phospholipase C) and again joins the other tw o subunits (G β and G γ) to reform the heterotrimeric G-protein, and—in the absence of ligand-bound GPCR—no longer dri ves the signal transduction pathway. Many g rowth f actor receptors (e g, EGFR, PDGFR ) and hor mone receptors (e g, insulin r eceptor) are transmembrane ligand-stimulated protein tyrosine kinases

Overview of Molecular and Cell Biolo gy

(RTKs). When ligand engages the RTK extracellular binding site, the intracellular protein tyrosine kinase domain is activated to transfer phosphate from ATP to target tyrosine residues present in specif ic intracellular tar get proteins (including the RTK itself). These phosphorylated proteins then trigger a signal transduction cascade, usuall y involving a series of sequential targeted phosphorylations, again leading to latent transcription factor activation and modulation of gene e xpression. F requently, proteins that bind GTP and ha ve GTP ase acti vity (e g, Ras, Rho, Ral proteins) also pla y an inter mediate role in R TK-mediated pathways; the ligand-induced R TK acti vation cascade stimulates GTP uptake by these small G proteins, driving their consequent acti vation of do wnstream kinases (eg, Raf, MEK in Figure 5). GTP cleavage by the GTPase activity of the small G protein, phosphatase remo val of phosphate from the R TK, and phosphatase remo val of phosphate from the acti vated kinase cascade proteins assure re versibility of these pathw ays w hen the ligand level is reduced. In man y cases, R TK-ligand comple xes are tak en into the cell b y endocytic vesic les (or endosomes) and transpor ted to the l ysosome, where both the receptor and the ligand (g rowth factor, hormone) may be degraded. Although some of the studies are controversial, the cumulati ve data suppor t roles for endoc ytosis and degradation as w ays to do wnregulate the signaling pathway, to pro vide ligand fragments necessar y in the cell to elicit biological activity, and/or to pro vide receptor fragments necessary in the cell to elicit biolo gical activity. The delta-notch signaling pathw ay is one of man y cell-cell ligand-receptor interactions that lead to intracellular signaling, ultimatel y altering gene e xpression and biological proper ties. When the transmembrane delta ligand engages the notch receptor on an adjacent cell, the interaction acti vates a tw o-step clea vage reaction that results in the release of an intracellular portion of the notch protein. The intracellular notch clea vage product then migrates to the cell nucleus, where it acts as a transcription factor to acti vate ne w gene e xpression. This remarkab le “one-off ” signaling mechanism, in which the ligand-receptor interaction results in creation of a single acti ve transcription factor molecule derived from the receptor, plays a key role in determining a large number of cell fate decisions during differentiation and development. Clearly, in this introduction to signal transduction w e ha ve pro vided onl y a fe w e xamples of signaling mechanisms to illustrate fundamental concepts and ha ve not discussed the multiple interactions that occur betw een and among alter native signaling pathw ays. The inter play betw een multiple pathways activated in response to a single ligand, and

621

the interactions occur ring as the result of multiple simultaneous ligand stimulations, culminate in enormously complicated and intricate intracellular v ariations in cell signaling. A more complete discussion of this complexity is presented in the chapters in this book on systems biology.

CELL CYCLE, SIGNAL TRANSDUCTION AND CANCER At its most fundamental le vel, cancer is a loss of g rowth control b y indi vidual cells; cells continue to proliferate when and where they should not. Many of the oncogenic activations leading to inappropriate cell proliferation occur in genes in volved in cell c ycle control or in genes that mediate entry into or exit from the cell cycle; hence, the emphasis on these two topics in this chapter. Recall that man y cells e xist in the quiescent, G 0 state and a wait appropriate cell-type specif ic cues to reenter the cell c ycle. When g rowth f actors (or mitogens) bind to their appropriate R TKs, the y stimulate signal transduction pathw ays that lead to new gene expression necessar y for the G 1 to S transition. Suppose, however, that an RTK is mutated so that its kinase activity is constitutively activated; that is, that the RTK no longer requires ligand occupanc y for its protein kinase activity. This is the case both for some lung cancers and for some brain cancers; mutated EGF receptors ha ve strong constituti ve, ligand-independent kinase activity that appears to be causal in the development of the cancer. Or suppose that a small GTP-binding protein in a g rowth-regulating pathw ay that normally has GTP ase activity is mutated so that it can no longer clea ve GTP and , instead , al ways has GTP bound. As a result, the mutated GTP-binding protein can activate its do wnstream kinase pathw ay that leads to cell c ycle entry, even in the absence of ligand binding to the upstream receptor (F igure 5). In nearl y all pancreatic tumors one member of the Ras gene f amily is mutated in a way that blocks activation of its GTPase activity. Genes in w hich there e xists a mutation that will “push” cells into the cell c ycle are refer red to as oncogenes. The mutated EGFR and Ras genes described here are kno wn as dominant onco genes because they are ab le to dri ve proliferation w hen present in cells that also car ry copies of the wild-type, unmutated gene. Recall that there also e xist genes that re gulate progression through the cell c ycle (F igure 3). Consider what w ould happen if genes that control the G 1 checkpoint decision (the p53 gene, the retinoblastoma

622

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

gene) are mutated to an inacti ve form. Missing one copy might be OK; the heterozygous cell might sur vive unscathed. However, if both copies are rendered inactive, the checkpoint will not be effective and cells will proceed around the cycle when they should not. If both copies of DNA damage or G 2 damage checkpoint genes are nonfunctional, cells will replicate damaged DN A, f ix it in their genomes, and pass it on to sib ling cells. Genes whose homozygous loss results in dysregulated cell cycle control are tumor suppressor genes. They are also called recessive oncogenes because loss of both gene copies is required for tumorigenesis. Much of the research, de velopment and application in translational clinical molecular imaging research involves de velopment and use of molecular imaging probes to noninvasively monitor altered expression, function or consequences of oncogenic transformation, and to monitor responses to therapeutic measures addressing the consequences of these mutations. Transgenic animal models in w hich reconstitution of the genetic onco gene causes of cancers is coupled with nonin vasive repor ter genes for identifying tumor initiation, for repeated monitoring of tumor pro gression, and for e valuation of response to therap y, are a bur geoning area of research application of molecular imaging tools. 55–57

THE LIFE CYCLES OF CELLS— EMBRYOLOGY, DIFFERENTIATION, DEVELOPMENT, AND DEATH The somatic cells of our body are diploid; they contain an equal complement of chromosomes from each of our parents. As a result of a meiotic division, the sperm and egg cells (or gametes) are haploid; they contain one copy of the parental genome. When the e gg is fertilized, the resulting zygote (Figure 6) is once again diploid and contains equal contributions of the pater nal and mater nal genomes (with the exception, as discussed previously, of the mitochondrial genome that is inherited mater nally). The zygote is a totipotent cell; it can go through the processes of cell di vision and directed cell specialization. Its progeny can differentiate into all the specialized cell types, each represented by its unique pattern of gene expression, in the body. Following fertilization, the zygote undergoes a series of rapid cell clea vages, to for m the morula and then the blastula, a ball of appro ximately 100 cells (with no increase in o verall size). The blastula then matures to a blastocyst; a structure that contains an inner cell mass or embryoblast which will subsequentl y gi ve rise to the embryo and an outer cell mass or trophoblast that will

subsequently for m the placenta. At the b lastula stage, implantation into the uterine w all occurs. During the ensuing gastrulation, cells that ha ve mig rated into the center of the blastula form the three germ layers: the ectoderm (which gives rise to the central nervous system, pigment cells and skin), the mesoderm (which gives rise to bone, cartilage, muscle, spleen, and blood vessels), and the endoderm (which gives rise to the li ver, colon, stomach, intestines, por tions of the lungs, and th yroid). Cells in the various germ layers are pluripotent; their ability to give rise to differentiated cells is now restricted to cells, tissues, and or gans deri ved from these ger minal la yers. Following the de velopment of the ger m la yers, the process of organogenesis is initiated. Neurulation is one of the earliest of these processes in all species, along with formation of the heart. During the process of organogenesis the capacity of pluripotent cells to differentiate is further restricted; for example, astrocytes, oligodendrocytes, and neurons are descended from one multipotent progenitor, w hile er ythroblasts, T cells, B cells, macrophages, mast cells, and platelets are descended from a distinct multipotent progenitor. Mature cells ha ve a f inite lifetime; the y are eliminated and the organs are repopulated from the appropriate pro genitor cell population. Most damaged cells are removed b y apoptosis, one of the alter native mechanisms of programmed cell death. Apoptotic cells are characterized b y plasma membrane blebbing, nuclear fragmentation, chromatin condensation, and endonuclease-mediated DNA cleavage. Apoptosis is also a par t of many nor mal de velopmental processes; e xtra cells required during inter mediate stages of or gan and tissue development are eliminated b y apoptosis during fur ther differentiation. A classic e xample is the elimination, b y programmed apoptosis, of the cells forming the webbing between the fingers and toes in the fetus as development of the hands and feet proceeds. In addition, damage from radiation, toxic chemicals, DNA damage from a v ariety of sources, viral infections, and other assaults can induce apoptosis in mature cells. One major role of damageinduced apoptosis is to commit cells to death, rather than fixing mutations into the genome that remo ve g rowth control, pre venting damaged cells from gi ving rise to transformed, pretumorigenic cells. If cells with DN A damage are unable to undergo the apoptotic option, they may then become tumor cells as a result of “f ixing” oncogenic mutations during DNA replication and subsequent mitosis. The biochemical process that ultimatel y leads to apoptotic cell death includes the induced expression/activation of caspases (a group of c ysteine proteases) and SMA C ( secondary mitoc hondria-derived activators of caspase) proteins. The process of apoptosis

Overview of Molecular and Cell Biolo gy

causes the cell to be broken into apoptotic bodies that are then phagocytized by macrophages. Apoptosis involves both the e xpression of genes that are not expressed in their nondying precursor cells and the rearrangement of cell membranes that e xposes previously inaccessible molecules. Extensi ve research to de velop imaging reagents that will allow us to monitor the presence of cells under going alternative forms of pro grammed cell death is currently ongoing. There exist imaging probes that can monitor the appearance of apoptotic indicators on the cell surface29,58 and reporter genes that identify the acti vation of caspases in tumor xenografts.59,60 If we can develop tools to monitor these processes in li ving individuals, we will be able to noninvasively examine both the presence or absence of these processes and the therapeutic modif ication of their progress in pathologies such as neurodegenerative diseases and cancers.

ES CELLS, ADULT STEM CELLS, INDUCED PLURIPOTENT STEM CELLS, AND CANCER STEM CELLS At the morula or blastocyst stage (Figure 6), cells from the inner cell mass can give rise, following transplantation, into all the cells of the adult body—these ES cells are totipotent. Both murine and human ES cells can be e xpanded in cul-

623

ture and, if expanded in the proper media, maintained with both self-rene wal and totipotent dif ferentiation potential. More dif ferentiated pluripotent stem cells (e g, mesenchymal stem cells, neural stem cells, hematopoietic stem cells) can similarl y be cultured in en vironments that maintain both the self-rene wal and pluripotenc y proper ties. Stem cells with restricted cell lineages, present in mature tissues, are often referred to as adult stem cells or multipotent progenitor cells . Stem cells are operationall y characterized b y their ability to undergo asymmetric division, with one progeny cell retaining the self-renewing property and one progeny cell e xercising the totipotential (for ES cells) or pluripotential (for adult stem cells) proper ty to move down an appropriate differentiation pathway. The use of both ES cells and adult stem cells for therapy of disease has, of course, been a topic of g reat enthusiasm, research and controversy. Stem cell therapies using adult (tissue-specif ic) stem cells ha ve been proposed for neurodegenerative disorders, stroke, traumatic brain injury, immunodeficiency diseases, diabetes, cancer , hear t disease, hematopoiesis, deafness, blindness and even baldness and tooth replacement; successful preclinical results ha ve been repor ted for man y e xamples, and some (albeit frequently contro versial) clinical benef its ha ve been described. Although many researchers and clinicians feel strongly that ES cells, with their g reater pluripotent capacity and potential, will be better “starting points” with

Figure 6. Embryology. The haploid male sperm fertilizes the haploid female egg to form the zygote. Following fertilization, the zygote undergoes a number of rapid cell divisions, without enlarging in volume, and develops into the morula, then the blastula and then the blastocyst. The blastocyst is surrounded by a trophoblast cell layer that promotes implantation into the uterine endometrium and develops into the placenta. In the center of the blastocyst is the inner cell mass, or embryoblast. It is from the inner cell mass that totipotent embryonic stem cells can be derived. After implantation, the embryo undergoes gastrulation, a major reorganization during which the mesoderm, ectoderm, and endoderm emerge. During subsequent development and differentiation, these germinal layers give rise to restricted sets of more differentiated cells as a consequence of continuing restriction of their pluripotent potentials.

624

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

which to embark on stem cell therap y research and applications for all diseases, the ethical and moral controversies sur rounding the use of ES cells—w hich must be derived from embr yos—have restricted research acti vities and clinical applications with these cells. There is no question that the ability to nonin vasively, repeatedly, and quantitati vely track the location, expansion, and dif ferentiation of stem cells follo wing in vivo administration is essential to the development of stem cell therapies in preclinical research conte xts, in the use of stem cells in clinical trials and , eventually, in standard therapies. 61–65 If stem cell therapeutic trials result in some patients sho wing benef it and others having either no response or adv erse ef fects, w hich will undoubtedly be the case in trial after trial, it will be imperative to kno w where these cells w ent, how many survived, for ho w long, and w hat biolo gical pathw ays they follo wed in patients with dif fering clinical responses. Onl y nonin vasive molecular imaging techniques can provide the answers. Although ES cell therapies ha ve great potential benefit, there are at least two major obstacles to their success. One is the hotl y debated moral objection, discussed above, to destro ying a human embr yo to generate the stem cell population. The second is the problem of transplantation barriers; unless the population of stem cells to be transplanted can be deri ved from the patient to be treated, rejection of the therapeutic cells ma y present an overwhelming complication. This second obstacle applies to both ES and adult stem cell therapies. Great excitement in the field of stem cell therapy has been generated by the development of methods to create induced pluripotent stem cells (iPS cells or IPSCs) from fibroblasts,66,67 liver cells, 68 and other dif ferentiated cells. Transferring stem cell-associated genes ( Oct4, Sox2, c-Myc, Klf4) into these differentiated cells, both in mice and in humans, can reprogram these cells to iPS cells that are in man y respects identical to ES cells. If iPS cells do, indeed , have the pluripotential capabilities of ES cells, the y o vercome both the moral objection (since no embr yo needs to be destro yed) and the transplantation bar rier (if the iPS cell is deri ved from the patient to be treated). F or patients with somaticall y altered disease-causing genes, no further genetic manipulation of the f ibroblast-derived iPS cells is needed. F or patients car rying germ line mutations, the genetic er ror must be cor rected before the iPS cell can be used for therapy. F or mice, the iPS cell-based therap y for a genetically inherited disease has, in principle, been performed. iPS cells were created from skin fibroblasts of a mouse carrying the sickle cell anemia gene. The iPS cell

hemoglobin gene mutation w as “cured” b y homologous recombination with the nor mal DN A sequence before iPS cell therapy was initiated. 69 Clearly, as with ES cell and adult stem cell therapies, the ability to noninvasively monitor the location, number and function of iPS cells following administration will be essential to the de velopment and application of this ne w therapy. At least one of the genes, c-Myc, used in the initial development of iPS cells from dif ferentiated tissues is causal in the de velopment of some cancers. Ho wever, subsequent studies have shown that c-Myc can be replaced with other stem cell-associated genes, and that ne w cocktails of stem cell-associated genes can impro ve reprogramming of differentiated cells into iPSCs. 67,70 Great adv ances are occurring in this very, very active field of cell biology. One of the most vexing problems in cancer therapy is frequent tumor reoccurrence following response to radiation or chemotherap y. It appeared to man y cancer therapists and cancer researchers that there e xists, for man y cancers, a resistant tumor cell subpopulation that expands to refor m a tumor follo wing therap y. Man y oncolo gy researchers suggest that included in the w ell-known cellular diversity of tumors is a population of cancer stem cells that retain the fundamental proper ties of stem cells—the ability to under go asymmetric di visions that give rise both to another self-rene wing cell and to a cell that proceeds do wn the pathw ay of limited di vision followed by “differentiation.” It is the cancer stem cells, and not the non self-rene wing progeny, that are postulated to be tumorigenic. Cancer stem cells are thought to be resistant to man y con ventional therapies; the therapies are effective for the bulk of the nontumorigenic cells, but not on the minority population of self-renewing, tumorigenic cancer stem cells. F ollowing therap y, the tumorigenic cells emerging from asymmetric di vision of the resistant stem cell population re generate the tumor . Solid, albeit hotly debated, data now exist for stem cell populations in breast, brain, colon, pancreatic, prostate, lymphoid, ovarian, and other cancers. There is substantial debate on the origins of cancer stem cells. One suggestion is that cancer stem cells originate by mutations in the genomes of nor mal stem cells. The second suggestion is that cancer stem cells originate from tumor cells that, in addition to ha ving lost the cell cycle and proliferation controls characteristic of other tumor cells, have also suffered additional alterations that confer on them the self-rene wing proper ties of stem cells. Regardless of their origin, if the cancer stem cell theory is cor rect, then identifying the cancer stem cells, understanding their biolo gy, identifying the alterations that make them tumorigenic, and devising therapies that

Overview of Molecular and Cell Biolo gy

attack them become paramount in ir revocably eradicating tumors. By using repor ter gene imaging techniques in geneticall y modif ied mice, along with imaging procedures that monitor the unique endogenous properties of cancer stem cells, w e will gain substantial insight both into cancer stem cell origins and into w ays in which they can be eradicated. At the clinical le vel, if we can def ine cancer stem cell markers that can be noninvasively imaged and if w e can nonin vasively distinguish stem cells from the bulk of the tumor , then monitoring disease pro gression, reoccurrence, and therapeutic responses based on stem cell eradication becomes an obtainable objective.

TRANSGENIC MICE The use of geneticall y engineered mice to study nor mal development, disease initiation and progression, diagnosis and therapeutic alter natives is an e xtraordinarily valuable and rapidl y g rowing preclinical research area. As w e increase our understanding of the genetic basis of biolo gical processes and as w e ref ine the ability to manipulate the mouse genome, applications of these models g row exponentially. In the last few years, addition of new instrumentation and the development of probes both for endogenous targets and for imaging reporter gene-reporter probe combinations have opened up extraordinary opportunities to e xpand our understanding of the biolo gy of de velopment and disease. The last section of this chapter will describe the cur rent state of transgenic mouse de velopment and the potential for utilization of combined transgenic models and molecular imaging technologies. We now have the ability to create targeted “knock-in” mutations in the mouse genome b y using homologous recombination; we can target and alter a single nucleotide in the genome of an ES cell (eg, changing a single amino acid in hemo globin, to create a sickle cell mouse) and convert that ES cell into an estab lished mouse strain that we can breed. We can also “knock-in” repor ter imaging genes; for e xample, our laborator y created a mouse in which the luciferase protein is e xpressed from the cyclooxygenase 2 gene promoter in a mouse strain, allowing us to noninvasively image when transcription of this gene is acti vated in response to inflammator y stimuli,71 or during tumor for mation (unpub lished studies). We can also “knock-out” targeted genes, eliminating their functions, b y similar processes. We can also add ne w DNA to the mouse genome, rather than replacing or eliminating e xisting DN A. In this w ay, w e can ectopically express genes in tissues that w ould not nor mally express them, b y dri ving the coding re gion of the gene to be

625

expressed with the re gulatory region of a dif ferent gene. Applying this technique to nonin vasive imaging questions, we can create mice that express reporter genes in a tissue-specif ic manner . Alternatively, b y creating mice in w hich a repor ter gene is e xpressed in all tissues from an ubiquitously expressed promoter,72 one can create mice in which the location and f ate of cells from an y tissue used for transplantation into a wild-type host can be monitored. With the adv ent of cyclization r ecombination and locus of X-o ver P1 (Cr e-Lox) recombination, and other technologies that work in similar w ays, we can now target to indi vidual cell types deletions of specif ic genetic regions. Cell-type specif ic targeting is accomplished b y directing, with appropriate promoters, the tar geted Cre nuclease excision of DN A sequences in the mouse genome that w e ha ve flank ed with 34 base pair loxP sequences. Targeted Cre e xcision of the lo xP flank ed sequence (the floxed sequence ) results in a tissuespecific deletion. With modif ications of the Cre recombinase enzyme to render it active only in the presence of the drug tamoxifen,73,74 researchers can now activate tissue-specific Cre recombinase e xpression at alter native times and initiate targeted gene deletion or ectopic gene expression as a result of lo xP cleavage. In this w ay, we can exert cell-type specif ic, temporal control over either gene deletion or ectopic gene e xpression. By dri ving small interfering RN A (siRN A) e xpression with inducible promoters in tissue-specif ic conte xts, we can now “turn off ” or “turn on” endogenous genes or exogenous genes at the time and place of our choice, for the length of time of our choice. 75 Combining these techniques with repor ter gene-repor ter probe imaging systems and with imaging probes for endo genous functions,76 we have within our g rasp the opportunity to manipulate both nor mal de velopment and disease initiation and progression in any fashion imaginable and monitor the consequences nonin vasively, quantitatively, and repeatedly.

REFERENCES 1. Alberts B, Johnson A, Lewis J, et al. Molecular biolo gy of the cell. New York: Garland Science; 2007. 2. Lodish HF, Berk A, Kaiser CA, et al. Molecular cell biolo gy. Ne w York: W.H. Freeman; 2008. 3. Walkley SU. P athogenic mechanisms in l ysosomal disease: a reappraisal of the role of the l ysosome. Acta P aediatr Suppl 2007;96:26–32. 4. Dunn KW, Sutton TA. Functional studies in living animals using multiphoton microscopy. ILAR J 2008;49:66–77. 5. Croix CS, Zipfel WR, Watkins SC. P otential solutions for confocal imaging of living animals. Biotechniques 2007;43:14–9.

626

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

6. Makale M. Intra vital imaging and cell in vasion. Methods Enzymol 2007;426:375–401. 7. Carradice D, Lieschke GJ. Zebrafish in hematology: sushi or science? Blood 2008;111:3331–42. 8. Kirby BB, Takada N, Latimer AJ, et al. In vi vo time-lapse imaging shows dynamic oligodendroc yte pro genitor beha vior during zebrafish development. Nat Neurosci 2006;9:1506–11. 9. Laughlin ST, Baskin JM, Amacher SL, Bertozzi CR. In vivo imaging of membrane-associated glycans in developing zebrafish. Science 2008;320:664–7. 10. McLean DL, Fetcho JR. Using imaging and genetics in zebraf ish to study de veloping spinal circuits in vi vo. De v Neurobiol 2008;68:817–34. 11. Mello CC. Retur n to the RN Ai w orld: rethinking gene e xpression and evolution. Cell Death Differ 2007;14:2013–20. 12. Fire AZ. Gene silencing b y double-stranded RNA. Cell Death Dif fer 2007;14:1998–2012. 13. Bogdanov AA Jr. Merging molecular imaging and RNA interference: early e xperience in li ve animals. J Cell Biochem 2008;104: 1113–23. 14. Medarova Z, Pham W, F arrar C, et al. In vi vo imaging of siRN A delivery and silencing in tumors. Nat Med 2007;13:372–7. 15. Martinez R, Mathe y-Prevot B , Ber nards A, Baltimore D . Neuronal pp60c-src contains a six-amino acid inser tion relati ve to its nonneuronal counterpart. Science 1987;237:411–5. 16. Bonano VI, Oltean S, Brazas RM, Garcia-Blanco MA. Imaging the alternative silencing of FGFR2 e xon IIIb in vi vo. RN A 2006;12:2073–9. 17. Bonano VI, Oltean S, Garcia-Blanco MA. A protocol for imaging alternative splicing re gulation in vi vo using fluorescence reporters in transgenic mice. Nat Protoc 2007;2:2166–81. 18. Wagner EJ, Baines A, Albrecht T, et al. Imaging alter native splicing in living cells. Methods Mol Biol 2004;257:29–46. 19. Paulick MG, Bo gyo M. Application of acti vity-based probes to the study of enzymes in volved in cancer pro gression. Cur r Opin Genet Dev 2008;18:97–106. 20. Villalobos V, Naik S, Piwnica-Worms D. Detection of protein-protein interactions in li ve cells and animals with split f irefly luciferase protein fragment complementation. Methods Mol Biol 2008;439:339–52. 21. Villalobos V, Naik S, Piwnica-Worms D. Current state of imaging protein-protein interactions in vi vo with geneticall y encoded reporters. Annu Rev Biomed Eng 2007;9:321–49. 22. Massoud TF, Paulmurugan R, De A, et al. Repor ter gene imaging of protein-protein interactions in living subjects. Curr Opin Biotechnol 2007;18:31–7. 23. Gross S, Piwnica-W orms D . Real-time imaging of ligand-induced IKK acti vation in intact cells and in li ving mice. Nat Methods 2005;2:607–14. 24. Safran M, Kim WY, O’Connell F, et al. Mouse model for noninvasive imaging of HIF prolyl hydroxylase activity: assessment of an oral agent that stimulates er ythropoietin production. Proc Natl Acad Sci U S A 2006;103:105–10. 25. Zhang GJ, Kaelin WG Jr. Bioluminescent imaging of ubiquitin ligase activity: measuring Cdk2 activity in vivo through changes in p27 turnover. Methods Enzymol 2005;399:530–49. 26. Gross S, Piwnica-Worms D. Monitoring proteasome activity in cellulo and in li ving animals b y bioluminescent imaging: technical considerations for design and use of geneticall y encoded repor ters. Methods Enzymol 2005;399:512–30. 27. Luker GD, Pica CM, Song J , et al. Imaging 26S proteasome acti vity and inhibition in living mice. Nat Med 2003;9:969–73. 28. Yoo B , Raam MS, Rosenb lum RM, P agel MD. Enzyme-responsi ve PARACEST MRI contrast agents: a ne w biomedical imaging approach for studies of the proteasome. Contrast Media Mol Imaging 2007;2:189–98.

29. Schoenberger J, Bauer J, Moosbauer J, et al. Inno vative strategies in in vivo apoptosis imaging. Cur r Med Chem 2008;15:187–94. 30. Corsten MF, Reutelingsper ger CP, Hofstra L. Imaging apoptosis for detecting plaque instability: rendering death a brighter facade. Curr Opin Biotechnol 2007;18:83–9. 31. Bloom J, Cross FR. Multiple levels of cyclin specificity in cell-cycle control. Nat Rev Mol Cell Biol 2007;8:149–60. 32. Gil J, Peters G. Re gulation of the INK4b-ARF-INK4a tumour suppressor locus: all for one or one for all. Nat Re v Mol Cell Biol 2006;7:667–77. 33. Kim WY, Sharpless NE. The regulation of INK4/ARF in cancer and aging. Cell 2006;127:265–75. 34. Sasaki M, Sugio K, K uwabara Y, et al. Alterations of tumor suppressor genes (Rb, p16, p27 and p53) and an increased FDG uptake in lung cancer. Ann Nucl Med 2003;17:189–96. 35. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Experimental and computational approaches to estimate solubility and per meability in drug discovery and development settings. Adv Drug Deliv Rev 2001;46:3–26. 36. Bae KT, Piwnica-Worms D. Pharmacokinetic modeling of multidrug resistance P-gl ycoprotein transpor t of gamma-emitting substrates. Q J Nucl Med 1997;41:101–10. 37. Kurdziel KA, Kalen JD, Hirsch JI, et al. Imaging multidrug resistance with 4-[18F]fluoropaclitaxel. Nucl Med Biol 2007;34:823–31. 38. Sharma V. Radiophar maceuticals for assessment of multidr ug resistance P-glycoprotein-mediated drug transport activity. Bioconjug Chem 2004;15:1464–74. 39. Zhang Y, Bressler JP, Neal J, et al. ABCG2/BCRP expression modulates D-Luciferin based bioluminescence imaging. Cancer Res 2007;67:9389–97. 40. Serganova I, Ponomarev V, Blasberg R. Human repor ter genes: potential use in clinical studies. Nucl Med Biol 2007; 34:791–807. 41. Dingli D, Kemp BJ, O’Connor MK, et al. Combined I-124 positron emission tomo graphy/computed tomo graphy imaging of NIS gene e xpression in animal models of stab ly transfected and intravenously transfected tumor . Mol Imaging Biol 2006; 8:16–23. 42. Dingli D, Russell SJ, Morris JC III. In vi vo imaging and tumor therapy with the sodium iodide sympor ter. J Cell Biochem 2003;90:1079–86. 43. Brooks DJ , Piccini P . Imaging in P arkinson’s disease: the role of monoamines in behavior. Biol Psychiatry 2006;59:908–18. 44. Ding YS, Fowler J. New-generation radiotracers for nAChR and NET. Nucl Med Biol 2005;32:707–18. 45. McConathy J, Kilts CD , Goodman MM. Radioligands for PET and SPECT Imaging of the central noradrenergic system. CNS Spectr 2001;6:704–9. 46. Elsinga PH, Hatano K, Ishiw ata K. PET tracers for imaging of the dopaminergic system. Curr Med Chem 2006;13:2139–53. 47. Schou M, Pik e VW, Halldin C. De velopment of radioligands for imaging of brain norepinephrine transporters in vivo with positron emission tomography. Curr Top Med Chem 2007;7:1806–16. 48. Dhaenen H. Imaging the serotoner gic system in depression. Eur Arch Psychiatry Clin Neurosci 2001;251 Suppl 2:II76–II80. 49. Bohnen NI, F rey KA. Imaging of choliner gic and monoaminer gic neurochemical changes in neurode generative disorders. Mol Imaging Biol 2007;9:243–57. 50. Dobson PD, K ell DB . Car rier-mediated cellular uptak e of phar maceutical dr ugs: an e xception or the r ule? Nat Re v Dr ug Discov 2008;7:205–20. 51. Mahony WB, Domin BA, McConnell RT, Zimmerman TP. Acyclovir transport into human er ythrocytes. J Biol Chem 1988;263: 9285–91. 52. Mahony WB, Domin BA, Zimmerman TP. Ganciclovir permeation of the human er ythrocyte membrane. Biochem Phar macol 1991;41:263–71.

Overview of Molecular and Cell Biolo gy

53. de Vries EF, Rots MG, Hospers GA. Nuclear imaging of hor monal receptor status in breast cancer: a tool for guiding endocrine treatment and dr ug de velopment. Cur r Cancer Dr ug Targets 2007;7:510–9. 54. Van Den Bossche B , Van de Wiele C. Receptor imaging in oncolo gy by means of nuclear medicine: cur rent status. J Clin Oncol 2004;22:3593–607. 55. Raman V, Pathak AP, Glunde K, et al. Magnetic resonance imaging and spectroscopy of transgenic models of cancer . NMR Biomed 2007;20:186–99. 56. Vignjevic D, Fre S, Louvard D, Robine S. Conditional mouse models of cancer. Handb Exp Pharmacol 2007;263–87. 57. Gross S, Piwnica-Worms D. Spying on cancer: molecular imaging in vivo with genetically encoded reporters. Cancer Cell 2005;7:5–15. 58. Kenis H, Hofstra L, Reutelingsper ger CP. Annexin A5: shifting from a diagnostic towards a therapeutic realm. Cell Mol Life Sci 2007; 64:2859–62. 59. Bullok KE, Maxwell D, Kesarwala AH, et al. Biochemical and in vivo characterization of a small, membrane-permeant, caspase-activatable far-red fluorescent peptide for imaging apoptosis. Biochemistry 2007;46:4055–65. 60. Coppola JM, Ross BD, Rehemtulla A. Noninvasive imaging of apoptosis and its application in cancer therapeutics. Clin Cancer Res 2008;14:2492–501. 61. Li Z, Suzuki Y, Huang M, et al. Comparison of reporter gene and iron particle labeling for tracking f ate of human embr yonic stem cells and differentiated endothelial cells in li ving subjects. Stem Cells 2008;26:864–73. 62. Sutton EJ, Henning TD, Pichler BJ , et al. Cell tracking with optical imaging. Eur Radiol 2008. (in press) 63. Lee Z, Dennis JE, Gerson SL. Imaging stem cell implant for cellularbased therapies. Exp Biol Med (Maywood) 2008;233:930–40. 64. Kraitchman DL, Bulte JW . Imaging of stem cells using MRI. Basic Res Cardiol 2008;103:105–13. 65. Schroeder T. Imaging stem-cell-dri ven re generation in mammals. Nature 2008;453:345–51.

627

66. Takahashi K, Tanabe K, Ohnuki M, et al. Induction of pluripotent stem cells from adult human f ibroblasts by def ined factors. Cell 2007;131:861–72. 67. Yu J, Vodyanik MA, Smuga-Otto K, et al. Induced pluripotent stem cell lines deri ved from human somatic cells. Science 2007;318: 1917–20. 68. Aoi T, Yae K, Nakagawa M, et al. Generation of pluripotent stem cells from adult mouse li ver and stomach cells. Science 2008;321: 699–702. 69. Hanna J, Wernig M, Markoulaki S, et al. Treatment of sickle cell anemia mouse model with iPS cells generated from autolo gous skin. Science 2007;318:1920–3. 70. Nakagawa M, K oyanagi M, Tanabe K, et al. Generation of induced pluripotent stem cells without Myc from mouse and human fibroblasts. Nat Biotechnol 2008;26:101–6. 71. Ishikawa TO, Jain NK, Taketo MM, Herschman HR. Imaging cyclooxygenase-2 (Cox-2) gene expression in living animals with a luciferase knock-in repor ter gene. Mol Imaging Biol 2006;8:171–87. 72. Okabe M, Ika wa M, K ominami K, et al. ‘Green mice’ as a source of ubiquitous green cells. FEBS Lett 1997;407:313–9. 73. Feil R, Brocard J , Mascrez B , et al. Ligand-acti vated site-specif ic recombination in mice. Proc Natl Acad Sci U S A 1996;93: 10887–90. 74. Zhang Y, Riesterer C, Ayrall AM, et al. Inducible site-directed recombination in mouse embryonic stem cells. Nucleic Acids Res 1996; 24:543–8. 75. Dickins RA, Hemann MT, Zilfou JT, et al. Probing tumor phenotypes using stab le and re gulated synthetic microRN A precursors. Nat Genet 2005;37:1289–95. 76. Sundaresan G, Paulmurugan R, Berger F, et al. MicroPET imaging of Cre-loxP-mediated conditional acti vation of a her pes simple x virus type 1 th ymidine kinase repor ter gene. Gene Ther 2004; 11:609–18.

39 SYSTEMS BIOLOGY GREGORY FOLTZ, MD AND LEROY HOOD, PHD

The challenge for scientific and engineering disciplines in the 21st centur y is dealing with the prob lem of complexity. One reason why biology will be a dominant science in the 21st century, just as chemistry was in the 19th century and physics was in the 20th centur y, is that powerful new approaches to deal with comple xity are emerging. A systems approach to biology, holistic in nature rather than strictly reductionistic––inte grating both detailed molecular data and higher le vel phenotypic obser vations to create predictive models explaining biology or disease, is the k ey to deal with biolo gical complexity. These systems approaches are driving the development of novel high-throughput measurement and visualization technologies as well as the creation of powerful new computational and mathematical tools. Finally, the systems view of biology is g rounded in the realization that biology is fundamentally an infor mation science. Some ha ve argued that energy or or ganization constitute the core of biolo gy, yet the increasing complexity of life, driven by evolution’s ability to solv e increasingl y comple x biolo gical challenges, is basicall y a prob lem of infor mation. A systems view of disease will also allo w powerful new approaches to understand the comple xity of patholo gy and it will as well suggest striking new strategies for diagnosis, therapy, and eventually prevention.

integrated to gether to mediate the f ive fundamental processes of life: evolution, development, physiological responses, aging, and disease. Second , biolo gic information is captured, transmitted, integrated, and modulated b y biolo gic netw orks that often transmit this processed infor mation to simple and complex molecular machines for execution. Third, there are hierarchical levels of biologic information that are multiscalar in dimensionality––DNA, RNA, protein, protein interactions, biologic networks (functional interactions of DN A, RN A, proteins, and other biolo gical molecules), cells, tissues or organs, individuals, populations, and ecolo gies. The en vironmental infor mation ma y impinge upon and modify the digital output to this informational hierarchy at any of these le vels. Increasingly, complex phenotypic measurements may be made at each of these le vels to be gin to assess the ef fects of the accumulating environmental modif ications at these various information levels. Indeed, to really understand how systems function, one must accumulate infor mation at as many levels as possible from the highest level phenotypic measurements back to the digital genome and integrate these so that one can capture the cumulative ef fects of the en vironmental modif ications—how to do this is another of the grand challenges of systems biology.

BIOLOGY AS AN INFORMATION SCIENCE The informational view of biology is predicated on several simple concepts. F irst, there are tw o fundamental types of biolo gic infor mation: the digital infor mation encoded in the genomes of li ving or ganisms and the environmental infor mation that impinges upon and modifies the outputs of digital infor mation. Thus, one of the grand challenges of systems biology is to understand ho w digital and en vironmental infor mation are

628

FEATURES OF CONTEMPORARY SYSTEMS BIOLOGY In a sense, systems biology has been around for more than a century; at the turn of the 19th century physiologists were interested in homeostasis and later in the 20th centur y developmental biologists, immunologists, and neurobiologists all became interested in understanding ho w their respective systems functioned. Several features distinguish

Systems Biology

the systems biolo gy of the 21st centur y from its earlier counterparts. First, it follows the vir tuous cycle of experimentation and modeling: a model of the system is for mulated based upon e xisting data and kno wledge about the system (the model may be initially descriptive, then graphical [biolo gical netw orks] and f inally, hopefull y, mathematical); g enetic o r e nvironmental p erturbations a re carried out to test a h ypothesis, molecular and phenotypic data are gathered and integrated and then compared against the old model, w hich can then be updated in vie w of new insights. This cycle can be continued until the e xperimental data are in accord with the predictions of the model. In this process, as the model becomes more complex, computational approaches may be used to optimize the choices of future per turbations. Indeed , as the model becomes more complete, simulations may be required to explore the appropriate dimensions of hypothesis (and data) space, that is, exploring new hypotheses and the meaning of new data types. Second , the data gathered should be as global or comprehensive as possible. This means that the beha viors of all genes, all mRN As, all microRNAs, all proteins, etc, must be measured. Because of the system’ s comple xity, hypotheses about which genes or proteins are important for the system are always incomplete, so it is important to look at all possib le features. The genomics measurements can generally be global (e g, comprehensi ve or complete) in nature, the proteomics techniques are not y et comprehensive. Third, measurements should be dynamical––as biologic networks change their architectures (eg, the nature of the connections of the elements (genes, mRN As, microRNAs, proteins, metabolites)) and features of the nodal (ie, the chemical nature of these elements, e g, the chemical modification of proteins) elements during the execution of biologic functions such as de velopment, ph ysiological responses, aging, and disease pro gression. The dynamically changing networks must be captured for a complete systems analysis of the system. F ourth, data must be captured from as man y le vels of the infor mation hierarch y (from the highest le vel phenotype back to the molecular measurements of DN A sequence and modif ications, mRNA identity, modifications, editing and quantif ication, etc) and integrated as described above. Fifth, the data measurements should be as quantitative as possible for the accurate formulation of models. This presents an enormous challenge as most lar ge-scale data sets ha ve significant noise––ho w to deal with this noise is, once again, a signif icant challenge for systems biolo gy. Sixth, system biolo gy often initiates its study of a system with discovery science, that is, defining all of the elements of an object without being dri ven b y h ypotheses. The core of systems biolo gy must star t with the complete genome

629

sequence of the or ganism being studied because this permits the inference of a complete par t list of the genes, mRNAs, microRNAs, proteins, (and other informational features) in the organism. The sequence and annotation of a genome represent pure discovery science. However, once an object such as the genome has had its elements enumerated, then one can begin the virtuous cycle of hypothesis-driven systems biolo gy described abo ve. One of the current challenges of systems biolo gy is the f act that w e only know about the architecture of a por tion of the networks relevant to individual biological phenomena. Moreover, the netw orks change dynamicall y during the progression of de velopment, physiological responses, and disease, and it is e xtremely dif ficult to capture these dynamics. These are areas of major technolo gy development, and some of the most e xciting w ork is emer ging from the applications of microfluidics to deter mine the structure and dynamics of netw orks. Ev en though our understanding of networks is incomplete, enormously useful insights can be gleaned from these incomplete studies on the nature of biology and disease.

SYSTEMS MEDICINE A systems view of medicine is driven by the concept that disease arises as a consequence of one or more diseaseperturbed netw orks and that these netw orks change dynamically during the course of the disease. Diseaseperturbed networks alter the nor mal patterns of information e xpressed (e g, mRNAs, microRNAs, and proteins) and these dynamicall y altered patter ns of infor mation expression encode the pathoph ysiology and hence the basic mechanisms of disease. Advances in high throughput DNA sequencing will enable each person’s genome to be sequenced rapidly, and at a reasonab le cost providing digital information for each person and making possib le the prediction of increasingly accurate individual probabilistic health futures. Also, the advance of high-throughput measurement technologies will enable the assessment of dynamic environmental information emerging from the integration of genome and environmental information for each indi vidual, as reflected , for e xample, b y the changing levels of proteins in the b lood, thus pro viding a real-time (current) health assessment of the indi vidual. These technolo gies will generate tremendousl y lar ge, dynamical data sets about health states and disease states and hence about the states of relevant biological networks of the indi vidual patients. Hence, an understanding of how the networks are perturbed, and how these perturbed networks change dynamicall y, can be gin to pro vide fundamental insights into the disease process. As we will see

630

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

below, they also provide new strategies for the diagnosis, treatment, and e ventually the pre vention of disease. Hence, we suggest that the informational view of biology, the systems approach to disease, the new next generation DNA sequencing technolo gies, high-throughput measurement and visualization technologies, new approaches for the measurement of increasingl y complex aspects of phenotype, powerful new computational and mathematical tools for data capture, assessment, storage, analysis, integration, and modeling will e ventually replace our current largely reactive medicine with a medicine that is predictive, personalized , pre ventive, and par ticipatory (P4 medicine) o ver the ne xt 10 to 20 years. Ho wever, there will always be a role for the integrative capacities of the physician to see the patient as a w hole and inte grate into the data-dri ven picture described abo ve the subtle phenotypic signals, easil y missed, that ma y be the real key to the diagnosis of disease in comple x cases. The central premise of systems medicine, diseaseperturbed networks alter dynamical patterns of expression of the genes and proteins that encode, in turn, the dynamic pathophysiology of the disease, requires the de velopment of techniques to detect and delineate disease-per turbed networks in the relevant organs and to assess their altered disease-perturbed molecular f ingerprints in the clinical setting. With recent advances in measurement and visualization technolo gies and computational tools for tissue and blood analyses, we anticipate the capacity to interpret these disease-perturbed molecular fingerprints in the context of a multitude of diagnoses to distinguish health from disease and to deter mine the nature of an y underl ying pathology. The ability to perform multiparameter analyses of biological information is the fundamental requirement for tracking these altered patter ns in disease-per turbed networks. Traditional diagnostic methods ha ve relied on just one or a fe w clinical parameters to def ine a disease state (e g, prostate specif ic antigen for the diagnosis of prostate cancer), w hich has signif icantly restricted our ability to track diseases that are highl y heterogeneous in the genetically polymorphic general population. Cur rent and emerging technologies are creating a transition into a new era w here molecular f ingerprints can be routinel y detected from serum or tissue samples allowing for multiparameter anal ysis of disease per turbation in indi vidual patients, across populations, and dynamicall y through time. A causal disease perturbation could be the result of specific, disease-causing DN A mutations, patho genic organisms, or other patholo gical en vironmental f actors such as to xins. Molecular f ingerprints of patholo gic processes can tak e on man y molecular for ms, including the analyses of proteins, 1 DNA,2 RNA,3 and metabolites,4

as well as informative, post-translational modifications to these molecules such as protein phosphor ylation. Once collected, these molecular datasets represent a dynamic and information-rich profile of disease that can be used to direct a personalized characterization of the disease course in individual patients. This will lead to signif icant advances in our ability to predict and therefore potentially prevent the onset of disease, or at least, the se verity of its course, in individual patients. Signals related to health and disease can be found in multiple sites. F or instance, man y bodily fluids such as the blood, urine, saliva, and cerebrospinal fluid can be sampled to identify evidence of perturbed molecular f ingerprints reflecting the altered e xpression patter ns of genes and proteins in disease-per turbed biolo gical networks. Of these, the blood is likely the most informationrich organ (or fluid) in that it bathes all of the organs and tissues in the body , all or gans secrete proteins into the blood, and it is easil y accessib le for diagnostic procedures. In addition to the biomolecules secreted into the blood from cells and tissues throughout the body , the transcriptomes and proteomes of cells circulating in the blood (eg, white blood cells) are also potentially an abundant source of biomedicall y impor tant infor mation.5 Thus, the amount of infor mation available in the b lood about health and disease is potentiall y enor mous if w e learn to read and inter pret the molecular signals.

AN EXAMPLE OF A SYSTEMS APPROACH TO DISEASE: PRION DISEASE IN MICE As an example of an en vironmental disease per turbation, we have studied the dynamic onset of the infectious prion disease in several strains of mice at the level of mRNA in the brain (the affected organ) and showed that a series of interlocking protein networks that surround the prion protein, as w ell as se veral other prion-disease-rele vant networks, are significantly perturbed across the 150-day span from disease initiation to death (F igure 1). Prion disease was chosen as one of the f irst models because (1) the onset and pro gression of the disease can be precisel y defined, hence its dynamics can be anal yzed effectively, (2) a g reat deal is kno wn about the histopatholo gy of the disease and thus one can cor relate molecular phenotype with histological phenotype, and (3) because of the a vailability of different prion strains affecting different inbred strains of mice in different ways, the biology of the disease can be used to deal with the enor mous signal to noise issues arising from most high-throughput data sets. Figure 1 shows a por tion of the dif ferential networks that w ere derived from comparisons of mRN A expression patter ns

Systems Biology

18 wks No Clinical Signs

2 wks

631

22 wks Clinical Signs

12 wks

20 wks

Figure 1. A central sub-network involved in neuronal pathology in mouse prion disease, shown at three times after inoculation. Red circles indicate increased levels of gene expression in brains of diseased animals relative to controls.

in normal and diseased brains at each of three time points after infection. They involve 67 proteins in the prion replication and accumulation netw ork. The initial netw ork changes occur well before the clinical signs of the disease can be detected and predict later widespread histopathologic events. These dynamicall y changing, disease-per turbed networks lead to two important conclusions. First, some significant netw ork nodal points change before the related clinical or histologic changes are evident (a nodal point is an element in the netw ork, in this case a mRN A, and the change is a concentration change). Therefore, labeled biomarkers that are specif ic for the changing nodes or the biologic processes they regulate could be used for in vivo imaging diagnostics e ven before symptoms arise, as has already been shown in patients.6–9 Alternatively, if some of these altered nodes encode secreted proteins, the y could provide readily accessible in vitro diagnostic blood markers for early disease detection. It is important to stress that the networks are distinct in dif ferent brain cell types and that these studies on the w hole brain provide information for networks specif ic for neurons, glial cells, and m yelinating cells. The cell-specif ic netw orks w ere check ed against the localization of 20,000 mRN As in dif ferent mouse brain cell types (from the Allen Institute Brain Atlas) and 128/128 of our assignments were cor rect. So these networks do pro vide f ascinating cell type specif ic information. Second, man y of the sub-netw orks of proteins that change during the onset of disease af fect changes in phenotypic traits that are consistent with the pathology of the disease. Approximately 300 per turbed mRNAs appear to encode the core prion-disease response,

and during the pro gression of the prion-disease process, perhaps a hundred or so under go significant gene expression changes well before the clinical signs of prion disease appear. Many of these potential earl y disease “sentinels” are predicted to be secreted into the b lood and therefore represent potential protein biomark ers for earl y disease diagnosis through b lood protein analysis. The impor tant point is that the disease-perturbed networks will be unique and distinguishing for each different brain disease; hence if multiple mark ers are assessed , the y will pro vide a unique fingerprint (ie, changes in the levels at which they are expressed) for each disease (see the discussion belo w on organ-specific blood proteins.)

MULTIPARAMETER BLOOD-BORNE BIOMARKERS In principle, it is clear ho w b lood samples containing secreted molecules from all of the tissues in the body can be used as a windo w into health or disease states. 10–14 In practice, however, the task of identifying mark ers of disease states in the vast array of secreted proteins can seem daunting. There are millions of different proteins present in the blood; they are expressed at levels that probably differ by 12 orders of magnitude (10 12); and perhaps 21 different proteins constitute 99% of the blood proteins by weight (albumin is 51%). Thus, there are indeed very significant challenges associated with defining relevant biomarkers in the plasma proteome, as is e videnced by the paucity of b lood protein mark ers found thus far despite significant ef forts.15 This f act highlights the need to develop well-founded systems approaches to diagnostics

632

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

to hasten the identif ication of such mark ers and to harness the tremendous infor mation potential of the blood. While the potential infor mation available to diagnose health and disease is enor mous, it is also tr ue that the challenges of separating signal from backg round noise (biolo gical or measurement noise) are also v ery significant. Sources of noise include er ror in measurements, polymorphisms in the population, en vironmental variations, s tochastic v ariations, a nd s ignal d ilution through mixing and other processes of molecule transport. Thus, enhancing signal as w ell as reducing noise will be a primar y theme of research in predicti ve medicine for the foreseeable future. An impor tant strate gy for dealing with noise arising from genetic pol ymorphisms and health histories in the population is through the use of dynamic, subtracti ve analyses car ried out in the indi vidual patient. In other words, measurements tak en from indi vidual patients at different time points (longitudinal data) can be used to perform subtractive analyses where only the dif ferences observed are considered. Thus, each patient becomes their own control, w hich eliminates many of the sources of noise. Eventually, it will be important for each patient to ha ve bi-annual b lood anal yses tak en so that changes can be obser ved relati ve to the backg round of w hat is normal in each individual, rather than relative to the population at large. Such databases of measurements will be essential to enable personalized and predictive medicine. Using each patient as their o wn control is one of the essential features of the emerging personalized medicine. It is true that at this point one does not know the extent to which variation of protein mark ers occurs in indi viduals across the 24 hour cycle of eating, working, and sleeping, with re gard to de velopment or ph ysiological responses, or with re gard to the aging process. Ne vertheless, these studies are now underway, and over the next 5 years we will gain tremendous insights into the variation of individuals that will be essential as a backdrop to vie w the individual as their own control.

ORGAN-SPECIFIC BLOOD PROTEIN FINGERPRINTS We discuss the organ-specific blood protein f ingerprint approach to diagnostics because it lies at the v ery heart of the predictive medicine that will emerge over the next 10 years. Our feeling is that this will be one of the central data-gathering strategies for predictive and personalized medicine and , accordingl y, w e will discuss it in some detail. The idea is that disease arises from dynamically changing disease-perturbed networks, the diseased

organs will secrete proteins into the b lood, and if these proteins are encoded b y disease-per turbed netw orks, their levels of expression in the blood will be altered in a manner that reflects the specif ic nature of the disease. Indeed, this idea is the basis for the v ery broad range of blood biomark er studies that are being car ried out b y many scientif ic centers. 10–14 The major dif ficulty with this simple vie w is that if y ou identify multiple b lood proteins whose changes are specific for a particular disease state (as compared against nor mal controls) and then examine the same blood markers for that disease in bloods dra wn from indi viduals, for e xample with 10 other pathologies, these mark er proteins can change in unpredictab le w ays because multiple or gans control the expression of most b lood proteins and these or gans respond differently to various environmental signals. The important point is that if a marker protein that is synthesized in f ive organs changes in the b lood, we cannot be certain w hich or gan(s) is the origin of the change. Although that mark er may sample a biolo gical network relevant to disease diagnosis, since its origin is not clear, its use as a disease biomark er may raise more questions than it answers. (Imaging, of course can be used to localize the site of change for one or a few markers, but imaging is dif ficult with the thousands of mark ers that w e propose to use to evaluate health and disease; see below.) The solution to this dilemma is clear—use organ-specific blood protein biomarkers whose changes must therefore reflect changes onl y in the or gan itself. We can determine candidate organ-specific mRNA transcripts by analyzing the transcriptome (quantitati ve deter mination of the population of mRN As present in the or gan) of the relevant or gan (e g, prostate) against the transcriptomes of all other organs and tissues. We have determined that from 50 to se veral hundred or gan-specific transcripts may be present in each of the 50 or so organs tested in humans and mice. The key question is what fraction of these organ-specific transcripts encode proteins that are secreted into the blood, and it turns out that for the initial tw o or gans tested in a preliminary manner for humans and mice, the numbers exceed 20 and will probably eventually be a hundred or more. If enough of these organ-specific blood proteins are sampled, they will represent a sur vey of man y dif ferent biolo gic networks in the organ of interest and will provide the ability to detect any disease in the organ as each disease will perturb distinct sets of netw orks and, accordingly, will change the levels of expression of differing combinations of the organ-specific proteins. For example, recent studies have demonstrated that three brain-specif ic b lood proteins in the mouse are altered in prion disease, and indeed that

Systems Biology

these markers can be used to car ry out presymptomatic detection of disease, identifying diseased animals 8 weeks before the onset of symptoms. Thus the organspecific blood protein fingerprint will not be a biomarker for indi vidual diseases, but rather will diagnose all diseases of the relevant organ. Eventually, correlation of these organ-specific biomarkers with more general biomarkers may prove to be the best long-term strategy for developing diagnostic blood fingerprints. We ha ve se veral lines of preliminar y e vidence that suggest this or gan-specific b lood protein approach will be ef fective. In prostate cancer , for e xample, there are disease-mediated altered patter ns of mRN A and protein expression in the diseased tissue. Some of these genes are expressed primarily in the prostate (organ-specific products) and some of these or gan-specific proteins are secreted in the blood, where they collectively constitute a protein molecular f ingerprint comprised of sa y 100 or more proteins whose relative concentration levels probably repor t the status of the biolo gic netw orks in the prostate gland. We have demonstrated that changes in the blood concentrations of several of these prostate-specif ic blood proteins reflect the v arious stages of prostate cancer and, as discussed above, various brain-specific blood proteins also reflect the progression of prion disease in mice. By comparison of the brain transcriptomes of mice and humans against the transcriptomes of more than 50 other organs, it is possib le to demonstrate more than 200 brain-specific or relatively brain-specific transcripts. At least 100 of these are secreted in the b lood. We propose that the distinct expression levels of the indi vidual proteins in each fingerprint represent a multiparameter (and therefore potentially information rich) diagnostic indicator reflecting the dynamic beha vior of, for e xample, the disease-perturbed networks from which they arise. The anal ysis of 50 or so or gan-specific proteins should allow us to stratify both diseases in the or gan and determine their stage of pro gression. We have identif ied ten to hundreds of or gan-specific transcripts in each of the 50 or so organs that we have examined in mouse and human. Most of these mark ers are perfect or thologs to one a nother an d h ence a ssay p recisely o rthologous biologic networks. We can en vision a time o ver the ne xt 5 to 10 y ears when 50 or so organ-specific blood proteins will be identified for each of the 50 or so major or gans and tissues in humans so that computational analyses of the relative concentrations of the protein components in these organspecific f ingerprints will enab le b lood to become the primary window into health and disease. When we analyze data from these b lood indicators, w e also ma y be in a

633

position to identify the dynamicall y changing diseaseperturbed network(s). The analysis of these dynamic networks will allow us to study in detail the pathoph ysiology of the disease response and hence be in a position to think of new approaches to therap y and pre vention. To generate the ability to assess simultaneously all 50 organ blood protein fingerprints in patients, we ultimately need to de velop the microfluidic or nanotechnolo gy tools for making perhaps 2500 rapid protein measurements (e g, 50 proteins from each of 50 human or gans) from a droplet of b lood. Nanotechnology is necessar y because onl y this striking miniaturization can allow the necessary thousands of measurements from a single drop of b lood. To reach this stage, we will also ha ve to create the appropriate computational tools to capture, store, anal yze, integrate, model, and visualize the information arising from this approach.

HIGH-THROUGHPUT PLATFORMS FOR SYSTEMS ANALYSIS Medicine in the future will focus on the dynamical measurements of the individual and will need to integrate the detailed molecular data re vealing the operation of disease-perturbed (or nor mal) netw orks with higher le vel phenotypic data (e g, hear t rate, b lood pressure, skin color, presence of pain, palpable abnormalities, history––recent and long ter m). Ho wever, increasing the detailed molecular data (DNA, mRNA, microRNA, protein, int eractions, ne tworks, etc) wi ll pro vide dee p insights into the onset and progression of disease in individuals. Hence, w e review here some of the contemporary measurement platforms for obtaining this data at the DNA, RN A, protein, and interaction le vels. All ph ysicians will need to have at least a general understanding of this molecular data and how it is obtained.

GENOMICS Complete genome sequencing provided the foundation for systems biolo gy b y enab ling in vestigations that could examine specific molecular hypotheses in the context of a completely def ined catalo gue of genes and proteins for humans. The f irst c omplete genome o f a f ree-living species, Haemophilus influenzae, was completed in 199516 and was quickly followed by other important human bacterial pathogens,17–19 yeast,20 human,21 mouse,22 and chimpanzee.23 By 2005, onl y a decade later , more than 1000 complete genomes had been completed. This massi ve expansion of sequence data was enabled by technological improvements in the Sanger sequencing method ,24 which

634

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

allowed for increased automatization and throughput. 25,26 Despite these advances, significant infrastructure and cost are required for sequencing e ven a relati vely small genome. As a result, the majority of genome sequencing is still perfor med at lar ge, dedicated genome centers and the majority of completed genomes represent onl y one or a fe w sampled or ganisms.27 Recent de velopments in microfluidics, image processing, and enzymolo gy promise to massively increase the speed and capacity of DNA sequencing at a signif icantly reduced cost. Currently, this “ne xt generation” high-throughput DN A sequencing (from companies such as 454, Solexa, Applied Biosystems, and Helicos) is being used to characterize cancer-associated mutations across e xtended patient populations.28 Once indi vidual genome sequencing is economically achie ved, comprehensi ve identif ication of individualized markers of disease susceptibility and treatment response will enab le predicti ve and pre ventative strategies to treat and prevent disease. One powerful application that has emerged from largescale genomic sequencing is the identif ication and characterization of pol ymorphisms, either single nucleotide polymorphisms (SNPs) or simple sequence repeats, in genes that identify and predict variations in biolo gic response, behavior, and predisposition to disease states (by DNA ar rays and hybridization or b y DNA sequencing of various types). Most common diseases are belie ved to be resulted from a mixture of genetic and en vironmental factors. Many demonstrate a comple x genetic predisposition believed to result from the contribution of small variations in several genes. More than 10 million putative SNPs have been identif ied in the human genome. Because the y are often highl y abundant, occur ring on a verage e very fe w hundred base pairs in the genome, and relati vely stab le, they provide useful mark ers for linkage anal ysis of genes involved in the patho genesis of comple x disease. The development of high-throughput genotyping methods makes genome-wide linkage disequilibrium mapping of SNPs a viab le approach to the study of complex disease susceptibility, as has been recentl y demonstrated in colorectal and renal epithelial cancer .29–31 Another important application of SNP analysis is the identification of genetic variants that influence a patient’s response to a dr ug, usually used to predict phar macological efficacy or the likelihood of har mful side ef fects. In the case of w arfarin, the most commonly used oral anticoagulant, genetic polymorphisms strongly predict inter-individual variability in anticoagulant ef fect, an impor tant cause of morbidity in the general patient population. 32–34 An important extension of this application is the identif ication of SNPs that alter cellular r esponses t o b iologic s ignaling d uring n ormal

development and function. This lar gely untapped area offers vast potential for defining the genetic basis of normal biological variation in the development of disease.

TRANSCRIPTOMICS Advance in genome sequencing technolo gy enab led the systematic measurement and comparati ve anal ysis of complete transcriptional pro grams of cells and tissues at various points in time and at any given developmental, pathological, or functional stage. The most widel y used methodologies for transcriptome anal ysis include DN A microarrays, serial anal ysis of gene expression (SAGE), microbead-based massively parallel signature sequencing (MPSS), and the massively parallel sequencing by synthesis (SBS). DNA microarrays are a powerful tool for highthroughput identif ication and quantif ication of nucleic acids in biological systems. DNA arrays typically consist of thousands of shor t gene-specif ic DNA molecules spatially ar ranged on a solid surf ace. Nucleic acid specif ic hybridization allo ws for the precise identif ication and quantification of transcript le vels in a cellular system at two or more different states. DNA microarrays either can be spotted arrays of oligonucleotides (25–60 bp in length) or cop y or cDN A molecules or can be oligonucleotide arrays produced by piezoelectric deposition or in situ synthesis. Recentl y, custom-designed ar rays ha ve become available containing up to 400,000 oligonucleotides. This flexibility in ar ray design allo ws for economical approaches to study specif ic disease states in systems biology. In principle, oligonucleotide arrays are more specific than the cDN A ar rays and ha ve the capability to distinguish between single-nucleotide differences. This method has the advantage of distinguishing between transcripts derived from indi vidual members of multigene f amilies and alternatively spliced variants. The widespread acceptance of DN A ar ray technology has led to the de velopment of applications be yond transcriptome anal ysis that have impact on the study of disease. DN A array technology has been used in genotyping studies to identify SNPs and conf irm the sequence identity of kno wn re gions of DNA. Cur rently, DN A ar ray applications include promoter anal ysis, ChIP (chromatin immunoprecipitation)on-chip studies of protein-promoter binding site occupancy, mutation anal ysis, comparati ve genomic hybridization, and genome resequencing. DN A sequences can also be tagged or labeled in such a way that the y can be identif ied in solution. A powerful new application of sequencing technology, MPSS or SBS, combines adv ances in microfluidics, enzymolo gy, and

Systems Biology

image proce ssing t echnologies to allo w millions of different sequences of up to 16 to 20 residues to be determined simultaneously per sample. 35

PROTEOMICS Proteomics can be def ined as the global characterization of proteins in comple x mixtures, including protein identity, abundance, processing, chemical modif ications, interactions in protein comple xes, and subcellular localization within a cell or tissue. At present, no proteomics technology a pproaches t he t hroughput a nd l evel o f automation of genomic technolo gies. Strategies for protein identif ication and quantif ication can be divided into mass spectrometr y (MS)-based techniques, designed to provide unbiased global measurements of protein abundance, and antibody-based techniques designed to identify kno wn proteins in a biolo gical sample. These and other strate gies to reduce sample comple xity, dif ferentially label protein samples, and impro ve relati ve quantification of proteins by MS analysis and antibody arrays promise to enhance a systems approach to disease.

Proteomics: MS Techniques The standard approach to proteome analysis is based on the separation of comple x protein samples b y tw odimensional gel electrophoresis (2DE) and the subsequent identif ication of selected separated protein species by one of a v ariety of mass spectrometric techniques.36 This approach is fundamentall y limited because specific classes of proteins are either not represented in the gels (e g, membrane proteins, small proteins, v ery basic proteins) or undetectab le because of their li mited abundance. Fur thermore, this method remains labor intensi ve despite automation of 2DE gel computerized patter n matching, protein e xtraction and digestion, and MS-based analysis. Furthermore, the enormous dynamic range of protein abundances found in biological systems, ranging from 1 to 10 6 copies or greater in cells and up to 1 to 1012 in serum, is a major impediment for d etecting lo w abundance p roteins. Improved throughput is pro vided b y direct anal ysis using tandem mass spectrometr y (MS/MS) of peptides generated by the digestion of complex, unseparated protein mixtures. 37 The k ey feature of this method is the ability of a tandem mass spectrometer to collect sequence infor mation from a specif ic peptide, e ven if numerous other peptides are concurrently present in the sample. This is accomplished in the instr ument by the

635

isolation of the peptide ion of interest from other peptides, fragmentation of the peptide ion in a collision cell (collision-induced dissociation [CID]), and the acquisition of the fragment ion masses in a computer. It is these fragment ion masses that represent unique identifiers for a peptide and the sequence of the peptide and therefore the identity of a protein is deter mined b y correlating the CID spectr um with the contents of sequence databases.38 Recently, protein separation techniques have been enhanced b y the use of multidimensional liquid chromatography (LC) followed by specific protein/peptide capture strategies.39 The development of technologies for global comparative measurements of proteomes from cells or tissues of different states, for e xample, healthy versus disease, is a fundamental requirement for a systems approach to disease. Stable isotope labeling of proteins/peptides enab les high-throughput relative quantif ication of proteins using MS on a scale approaching se veral thousand per sample. The general strate gy involves differentially labeling proteins or proteol ytic peptides with stab le isotopes, mixing of labeled samples at a 1:1 ratio, followed by combined sample processing and subsequent MS anal ysis. Because the labeling reagents possess almost identical chemical properties, the labeled peptides appear closel y paired in the LC and MS processes. Relati ve quantif ication is achieved by comparing ion signal intensities or peak areas of isotope-encoded peptide pairs observed in the corresponding mass spectra. Current methods for the introduction of mass tags to proteins and peptides include isotope-coded affinity tags (ICATs), stable isotope labeling by amino acids in cell culture (SILA C), and isobaric tag for relative and absolute quantitation (iTRAQ). The ICAT technique involves differential labeling of two dif ferent protein populations on the side chain of reduced c ysteinyl residues using one of tw o chemically identical but isotopicall y dif ferent ICAT reagents. 40 By incorporating a biotin affinity tag into the ICAT reagents, selective isolation and purif ication of labeled peptides substantially reduces sample comple xity. The ICA T approach has been successfully applied to the systematic identification and quantif ication of proteins contained in the microsomal fraction of cells. 41 It has also been applied, in conjunction with DNA microarray analysis, to identify differential expression profiles of hematopoietic progenitor cells. 42 A major dra wback of the ICA T technique is that it onl y labels the fraction of proteins containing c ysteine residues. An alter native approach, SILAC, involves growing two populations of cells, under identical conditions except that the culture medium for one population contains all 20 essential amino acids in

636

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

their naturall y occur ring isotopic for ms (“light” population), w hereas the other population is g rown in medium w here one or more amino acids are replaced with stable, heavy isotope labeled analo gs.43 The incorporation of a heavy amino acid into a peptide, referred to as metabolic labeling, results in a kno wn mass shift relative to the peptide that contains the light v ersion of the amino acid. The advantage of using metabolic labeling is that it allows mixing of labeled and unlabeled cells before the fractionation and purif ication steps and therefore avoids introduction of an y er rors in relati ve quantif ication in subsequent sample preparation. Fur thermore, all peptides within the sample can be analyzed, not just those containing c ysteine residues, increasing conf idence in both identification and quantif ication. Isobaric tag for relati ve and absolute quantitation (iTRAQ) is a multiple xed strate gy that allo ws up to four samples to be anal yzed simultaneously by MS in the same experiment.44 Peptides are labeled on the free amine g roups at the amino ter minus and on l ysine residues. Unlik e other isotopic labeling strate gies, the iTRAQ label reagents are designed to provide quantitative information upon peptide fragmentation. This technique modif ies peptides b y linking a mass balance group (carbonyl group) and a repor ter group (based on N-methylpiperazine). Designed to be isobaric (ha ving the same mass), the iTRA Q reagents are chromatographically indistinguishab le in the LC step, causing the ion peak for each of the identically labeled peptides to be detected simultaneousl y by the mass spectrometer. When MS/MS is used for anal ysis, the mass balancing carbon yl moiety is released as a neutral fragment thereb y liberating isotope-encoded repor ter ions that pro vide relati ve quantitati ve infor mation on protein abundance. Because four dif ferent iTRA Q reagents are cur rently a vailable, comparati ve anal ysis of a set of tw o to four samples is feasib le within a single MS run. As is the case with SILAC, all peptides are labeled in iTRA Q e xperiments. An adv antage of this method is that the multiplexed nature of iTRAQ greatly reduces the amount of MS time required to characterize indi vidual samples, increasing instr umentation throughput.

Proteomics: Reducing Complexity Protein concentrations in biolo gical systems span 10 12 orders of magnitude, w hereas the most commonl y available MS-based methods onl y allow for the identif ication of proteins spanning appro ximately three orders of magnitude in concentration from a given sample. Several

methods ha ve been adv anced that select for specif ied fractions of the proteome to reduce the complexity of the sample suf ficiently to identify biolo gically interesting proteins. Protein glycosylation, one of the most common post-translational modif ications, is characteristic of secreted proteins and cell-surf ace markers but not found on the predominant ser um proteins such as albumin. Recent approaches selecting for N-linked or O-linked glycosylated peptides using affinity capture techniques or solid-phase extraction followed by stable isotope labeling enrich for these biologically-active proteins.45 Zhang and colleagues examined glycosylated proteins from a number of tissues, cells, and plasma and compared the glycoproteins identif ied in the tissues and cells with those identified in the plasma. 46,47 A signif icant o verlapping was obser ved. This study demonstrates that tissuederived proteins are indeed present and detectab le in the plasma via direct MS analysis of captured glycopeptides, proving the feasibility of an MS-based approach for plasma protein discovery and analysis. A significant improvement on this technique has been the capture of glycopeptides, rather than glycoproteins.48 Multiple reaction monitoring (MRM) is a highl y selective, highly sensitive MS approach for detecting the presence of par ticular peptide species in a comple x mixture such as plasma. 49 A specif ic tr yptic peptide is selected as a stoichiometric representati ve of the protein from which it is cleaved. This peptide is quantified by MS against a spik ed inter nal standard, a synthetic stab le isotope-labeled version of the peptide, to yield a measure of protein concentration. In principle, such an assay requires only knowledge of the masses of the selected peptide and its fragment ions and an ability to make the stable isotopelabeled version. This method can reliably quantify protein concentrations over a dynamic range of 4.5 orders of magnitude in human plasma using a multiple xed approach. MRM assa ys coupled with enrichment of proteins b y immunodepletion and size exclusion chromatography,50 or enrichment of peptides b y antibody capture, ha ve also been repor ted.51 Stable isotope standards and capture b y anti-peptide antibodies (SISCAP A) has been sho wn to extend the sensitivity of a peptide assay by at least two orders of magnitude and with fur ther de velopment appears capable of e xtending the MRM method to cover the full known dynamic range of plasma (ie, to the pg/mL level).51 In systems approaches to disease, man y important bioactive proteins are presumed to be secreted in the blood as key regulators of systems processes. These and other strategies to reduce or overcome the complexity of serum hold g reat promise for identifying k ey proteins active in disease states.

Systems Biology

protein abundance can be monitored in real time allowing for assessment of binding dynamics and (2) slides can be regenerated allowing for cost-effective screens of multiple samples. Another recentl y de veloped technique, DN A-encoded antibody libraries (DEAL), is a highly sensitive measurement technique, which can detect protein and single-stranded DNA simultaneously on a single chip. 59 DNA-encoded antibodies are labeled with single-stranded DN A oligomers. DN A-encoded antibodies and secondar y (fluorescentl y) labeled antibodies are added to the biological sample containing the protein of interest. The entire complex is then captured by nucleic-acid h ybridization onto a spot that w as prepatterned with the complementary single-stranded DNA oligomer. This approach has been used for the rapid detection of multiple proteins within a single microfluidic channel with a lo wer detection limit of 10 femtomolar, 150 times more sensitive than the analog ELISA (Figure 2) A critical component of systems medicine is the capacity to measure and inte grate multiple biolo gical interactions, for instance protein-protein or protein-DNA interactions, w hich c an t hen b e u sed t o c onstruct dynamic molecular netw orks associated with the progression of disease. The complete set of interactions between all indi vidual biomolecules constituting the complex re gulatory, signal transduction, and feedback networks underlying health and disease provides the theoretical framework for higher order anal ysis refer red to as the interactome. These dynamic associations include protein-protein interactions, protein-DN A interactions, transcriptional regulation, and post-transcriptional gene silencing b y shor t-interfering RN A and microRN A.60 Multiple high-throughput methods ha ve been developed

Proteomics: Antibody-Based Arrays An alter native strate gy to global proteome anal ysis using MS is the use of antibody-based ar ray techniques.52–54 A variety of methods have been developed based on antibody binding, all of which are limited by (1) dependence on the affinity and specificity of the antibodies used for detection, (2) relatively high cost of generating monoclonal antibodies, and (3) potential cross reactivities in comple x protein mixtures. Despite these limitations, antibody arrays have the advantage of providing a quantitati ve and comparati ve platfor m for rapid screening of proteomes from dif ferent disease states such as lung,55 pancreatic,56 and prostate cancer.57 One emer ging approach with tremendous promise is surface plasmon resonance (SPR), w hich enables realtime, label-free measurement of protein ab undance.58 SPR is a ph ysical phenomenon that occurs w hen electromagnetic waves, such as light, are reflected off a thin metal f ilm at specif ic incident angles and w avelengths. A fraction of the light ener gy (either pol ychromatic, many colors, or monochromatic, one color) interacts with and transfers energy to the surface plasmons, electromagnetic waves that oscillate along the surface interface, thus reducing the reflected light intensity at a sharply def ined angle or at a specif ic wavelength. Any modifications on the metal surf ace, such as that occurs with the interaction betw een antibody and antigen, will affect the SPR condition and can be used to detect and monitor specif ic molecular interactions. Sensiti vity of SPR for low abundance proteins is estimated to be in the picogram/cm2 range. Current SPR-based chips have 800 unique antibodies arrayed at approximately 4 µm spatial resolution. Significant advantages of this method are (1)

Capture antibody-DNA conjugation O N

1

O

O

O

N

N H

N

SANH (Succinimidyl 6-hydrazinonicotinate acetone hydrazone) O

1

N

O

O

O

O

NH2

H

SFB (Succinimidyl 4-formylbenzoate)

Figure 2A.

637

DEAL coupling chemistry. Heterobifunctional molecules SANH and SFB are used to couple ssDNA to antibodies.

638

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

complementary DNA array

Antibody W/ DNA

Protein mixture

Protein array

Protein array with bound analyte

Figure 2B.

300 nanoliters of plasma

cells out

Assay region

Figure 2C (top). DEAL for in vitro molecular diagnostics: integrated nanotech/microfluidics platform. Separate plasma & rapidly quantitate protein biomarker panels, profile health status of individual organs, select appropriate therapies or combination therapies, profile positive & adverse responses to therapies.

Organ 1

Organ 2

Tox response

Inflammation

20 mm 20

Figure 2C (bottom). Panel of protein biomarkers measured in a single microfluidics channel (15 min assay time). Figures are courtesy of the laboratory of James R. Heath at the California Institute of Technology.

to characterize the interaction betw een proteins, including the yeast two-hybrid system,61 SPR,62 and MS-based analysis of immunoprecipitated protein comple xes.63,64 High-throughput methods for characterizing proteinDNA interactions include approaches where an antibody specific for a DN A binding protein is used to immunoprecipitate the associated protein-DN A comple x after which the DN A can be isolated and anal yzed b y massively parallel single-molecule sequencing 65 or ar raybased techniques. 66 These methods include ChIP-sequencing, ChIP-promoter array analysis (ChIPchip), and DNA sequencing of transcription f actor associated-DNA regulatory elements.67–69 These studies have provided a rich resource to constr uct interactome networks for protein-protein and protein-DN A interactions. In general, identifying multi-protein and multigene interactions is a comple x computational and statistical task. Cur rent studies often e xamine pair-wise interaction data betw een biomolecules, such as w hether two proteins are positi vely cor related or ne gatively

correlated in vivo. These techniques ha ve been used to profile por tions of the y east, fl y, w orm, and human interactomes, aided by automated data mining programs such as Reactome 70 and curated databases including BioGRID.71 New advances in microfluidic measurement techniques will allo w for the rapid and accurate collection of larger numbers of measurements from bodily fluids, signif icantly enab ling our ability to constr uct interactomes for key re gulatory pathw ays underl ying disease states. 72 The computational anal ysis of these biological networks will re veal multivariate mark ers of pathology w here traditional, single-parameter studies have pro ven insuf ficient. This represents a promising new domain for the characterization, diagnosis, and potential treatment of human disease.

Case Study: In Vivo Molecular Diagnostics As it relates to personalized and predictive medicine, in vivo molecular diagnostics will also require the

Systems Biology

development of a di verse librar y of molecular imaging probes. These modular tools can be used to (1) identify the specif ic location of disease-per turbed netw orks in patients, (2) link in vi vo molecular measurements in diseased tissue in patients to in vitro measurements, (3) rapidly assess the ef ficacy of personalized therapeutics, and (4) validate that a dr ug is hitting its tar get and inducing the desired pharmacological outcome. Although there are man y in vi vo imaging modalities, perhaps the best current method from the point of vie w of personalized and predicti ve medicine in patients is positron emission tomo graphy (PET) molecular imaging. 73 For PET, trace quantities of radiolabeled molecular probes are injected into the patient. As the probes circulate through the body and its v arious or gan systems, the y interact with tar get proteins to pro vide imaging assa ys, for example, the rate of metabolic processes, the concentration of receptors in signal transduction, enzyme activity, DNA-replication rates, hormone status, pharmacokinetics and pharmacodynamics.

COMPUTATIONAL APPROACHES IN SYSTEMS MEDICINE A systems approach to disease requires the integration of vast amounts of quantitative biological data generated by global genomic and proteomic anal yses to (1) comprehensively identify key molecular components defining disease and health y states and (2) deter mine how these components interact in biological netw orks in a predictable way. Initial efforts primarily integrated and analyzed large databases of gene e xpression data to identify subsets of genes with predictive value for disease stratification and pro gnosis. A v ariety of methods ha ve been applied to disease diagnostics including approaches based on support vector machines and relative expression reversals, among many others. Application of these methods has led to the disco very of molecular classif iers of varying degrees of accuracy to identify prognostic signatures for breast cancer , o varian cancer , colon cancer , prostate cancer, and brain cancer.74,75 One interesting new diagnostic method evaluates relative expression reversals in protein concentrations or gene e xpression le vels for pairs of informational features. This procedure eliminates the need for data nor malization or the estab lishment of population-wide thresholds. This approach has been successfully used to identify robust and accurate classifiers for prostate cancer ,76 sarcomas,77 and a v ariety of other cancers,78 as w ell as to predict treatment outcomes in breast cancer,79 demonstrating the ef ficacy of e ven relatively low-order systems analysis in medicine.

639

With the development of protein-protein and proteinDNA interaction databases, gene e xpression data can be mapped onto interaction networks to identify relevant biologic pathways active in specif ic disease states or experimental perturbations. This type of approach is useful for assigning disease-specif ic rele vance to dif ferentially expressed genes or molecular pathw ays and has been applied in a number of human diseases, most notably cancer. Although these interaction netw orks are v ery useful tools for visualizing lar ge data sets, the y are not computable, predictive network models, w hich are those that hold the most promise for predicti ve medicine and dr ug development. One ne w approach uses an inte grated framework, P ointillist,80 for combining di verse datasets and inferential algorithms to generate model netw orks, which can then be incor porated into Cytoscape 81 or Biotapestry82 for visualization and simulation of protein and gene re gulatory netw orks, respecti vely. The inte gration methodology of Pointillist is able to handle data of different types and sizes (e g, interactions, protein e xpression, gene expression) to create a higher confidence interaction network than that resulting from a single dataset alone. A novel aspect of this methodology is that it does not require a “gold standard” set of data to be used for training nor does it make assumptions about the underlying statistical distributions in the for m of parametric models. This involves designing an efficient deterministic optimization algorithm to minimize the numbers of misses and f alse positives via an iterati ve element b y element procedure. The methodolo gy is general pur pose so that it can be applied to integrate data from any existing and future technologies.83 Other approaches for inferring genetic regulatory networks include parsimonious linear re gression models, probabilistic Boolean netw orks, or Ba yesian netw orks from expression data (both steady state and time course).84 Probabilistic Boolean netw orks are robust in the f ace of biologic and measurement uncer tainty and offer the ability to characterize and simulate global netw ork dynamics using the inferred model structure.85 It also provides a natural way to deter mine the influences of par ticular genes on the global network behavior.86 Thus, the model can be used to predict the ef fects of per turbations on netw ork dynamics, an impor tant goal for understanding disease development and treatment response. These and other predictive models stemming from mathematical descriptions of biochemical reaction networks and statistical influence models are critical for identifying disease-per turbed networks in disease states. Dynamic and predicti ve network models have been developed for impor tant signaling networks in disease such as cancer. Such approaches are now

640

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

used to predict response to netw ork per turbations in mammalian systems using an algorithm for the Reconstruction of Accurate Cellular Networks (ARACNe).87 These and other computational modeling approaches will play a k ey role in the identif ication of disease-per turbed networks, identifying potential therapeutic tar gets. With the development of comprehensive databases, hypothesisdriven global anal ysis methods, and predicti ve netw ork models, systems biolo gy has matured as a predicti ve science, capable of generating testab le hypotheses based on network models with predictable behaviors. Applying this systems approach to disease holds g reat promise for the development of new therapies.

CONCLUSIONS AND PERSPECTIVES Technologies are just no w emerging that will transfor m systems medicine, and the next 10 years will see striking developments in the speed, cost, and accuracy of genomic sequencing so that indi vidual genome sequencing will become a reality. In the ne xt 10 y ears, one will de velop powerful ne w types of chemical protein capture agents that are required by the protein chips to detect thousands of proteins (to replace the dif ficult to produce antibodies). New techniques will emerge for the large-scale production of protein chips with the capacity for making thousands of measurements, rapidl y and ine xpensively from a fraction of a droplet of b lood (and similar chips will be de veloped for mRN As, microRNAs, etc). Lik ewise, nanotechnolo gy and microfluidic sensing chips with integrated microfluidic delivery devices will be developed over the next 5 to 10 y ears so that we reach a point at which thousands of protein or RN A interactions can be anal yzed quantitatively from onl y a fraction of a droplet of blood. There will emerge powerful new molecular imaging techniques, both in vitro and in vi vo, for interrogating the molecular operations of the disease processes. And all of these data will be v alidated, mined, integrated, and modeled to gi ve deep insights into health and disease for indi vidual humans. These technolo gies will lead to tw o of the most fundamental aspects of predictive and personalized medicine, namel y the ability to predict future health histories from analyses of individual genomic sequences (coupled with relevant environmental data) and the ability to assess the cur rent health/disease status of vir tually e very or gan in the body at 6 month intervals through the anal ysis of or gan-specific b lood protein f ingerprints. Similarl y, in vi vo imaging agents will per mit the functional visualization of infor mational molecules and drugs in humans, which will provide powerful and infor mative in vi vo diagnostics approaches.

Discussions of the emerging visualization techniques that will permit the in vi vo imaging in humans of indi vidual molecules are pro vided else where in this book. But an attractive possibility will be to use b lood organ-specific protein measurements to identify disease in an organ, say the brain, and then to use in vi vo molecular imaging techniques to localize the position and distribution of disease-perturbed networks, say in a tumor such as glioblastoma. These techniques to gether could pro vide earl y detection of cancer with the possibility of managing the disease ef fectively with con ventional therapies. These measurement technologies, together with both ne w computational approaches to extract information from these data and the systems view of medicine, will lead to three important revolutions in medicine. First, over the next 10 to 20 years, a medicine that is predictive, personalized, preventive, and participatory will emer ge. Obtaining indi vidual human genome sequences will lead to individual predictive health histories, whereas biannual blood measurements of thousands of proteins from the molecular fingerprints of human organs will give us a real-time assessment of indi vidual health. Predictive medicine will result in a personalized medicine that focuses on the illnesses of individuals and eventually their wellness, preventing disease rather than treating it. The systems approach to disease will in time permit the identif ication of the k ey disease-per turbed sub-networks and the identif ication of impor tant nodal points, which, if perturbed by drugs, could make the network behave in a more nor mal fashion or at least delete the more deleterious effects of the disease-perturbed network. Drugs of the future will be multi-component as biological netw orks will probab ly require tw o or more perturbations to alter their disease-per turbed behaviors. This capability will provide a new approach for drug-target discovery and a powerful and rapid new approach for developing drugs. Indeed, eventually we may be able to design drugs that will prevent disease. For example, suppose that one’s genome leads to the prediction that y ou have an 80% chance for lung cancer by the time you are 50 years old. The hope would be that one could de velop a pre ventive dr ug, w hich if tak en star ting at age 35 would reduce this probability to 3%. Predicti ve, personalized, and pre ventive medicine, if appropriatel y orchestrated with patient-oriented inter pretations, will also enab le patients to understand more deepl y and actively participate in personal choices about illness and wellness. P articipatory medicine will necessitate the development of po werful ne w approaches to handle enormous amounts of personal infor mation in a secure manner, and to a ne w for m of medical education of

Systems Biology

individual patients as w ell as their ph ysicians. Over the next 5 to 20 years, medicine will become predictive, personalized, preventive, and participatory (P4 medicine). Another very exciting idea is that, in the long-term, systems analysis of blood offers the potential to open up a ne w a venue for studying human biolo gy—when w e learn to read and interpret the information inherent in secreted organ-specific protein patterns. The k ey patterns in these secreted proteins represent dif ferent network per turbations, and hence, dif ferent diseases. When secreted proteins enter the blood, they can provide a novel means to study biolo gy in higher organisms and to identify dr ug targets through linking b lood measurements to per turbations in underl ying biolo gical networks. Impor tantly, lear ning to study per turbations in underlying netw orks through secreted protein patter ns (to whatever resolution is possible) will provide access to study biological networks in vivo that are not amenable to direct e xperimentation. De veloping the capacity to identify in vi vo network per turbations through secreted protein measurements in the blood will open up a new avenue to dr ug tar get identif ication and will pro vide a novel means to disco ver the per turbed sub-netw orks. Because taking b lood is relati vely non-in vasive, it has less potential than biopsies or other invasive techniques for greatly distorting the system being studied. Thus, this approach has the potential to strongly complement existing approaches to study human biolo gy. The ef fect of drugs, toxicity, human development, and even aging may all be amenab le to study through the b lood if w e can learn to read and inter pret the signals in the proteins, which has the potential to be v ery signif icant in human biology. In f airness to our future vie w of medicine and ho w rapidly it will be realized, one should add real cautionary notes. Some scientists or ph ysicians see these technologies as reaching fr uition in 20 or 30 years rather than 10 years. We suspect that their assessments will be incorrect and ours will be much closer to the mark. There are serious debates about ho w much one can project from individual genome sequences in the w ay of future health consequences. These are reasonab le concer ns, and our view is that we will increasingly be able to integrate into the digital genome infor mation indi vidual responses to environment stimuli, and this inte gration will pro vide much deeper insights into future health histories. Some have ar gued that the or gan-specific b lood f ingerprint strategy is at a v ery earl y stage, and w e reall y do not know the real potential of this technique. F or e xample, will w e be ab le to see the or gan-specific f ingerprints

641

from small organs such as the beta cells of the pancreas? We believe there will be suf ficient sensitivity in cur rent techniques, which are cer tainly going to increase in sensitivity by orders of magnitude over the next 10 years, but this remains to be proven. And many additional questions may be raised. For the student––one should be appropriately sk eptical––but open to the v ery real possibilities that the impact of these transfor mational emerging technologies on the future practice of medicine will be lar ge and short term. Second, as the sensitivity of measurements increases (both in vitro and in vivo), we will achieve a digitalization of biology and medicine, that is, the ability to extract relevant information content from single individuals, single cells, and ultimately single molecules, with its attendant e conomies o f s cale. J ust a s M oore’s l aw ( the prediction that the number of computing elements that can be placed on a computer chip will double every 18 months––as it has for the last 35 years) led in time to the widespread digitalization of infor mation technolo gies and communications, the exponentially increasing ability to extract quantized biologic information from individual cells and molecules will transfor m biology and medicine in ways that we can only begin to imagine. Finally, all of these changes, a systems approach to medicine with its focus on disease pre vention and more efficient drug discovery, the introduction of increasingly inexpensive nanotechnolo gy based diagnostics and in vivo measurement technologies, the highly accurate and specific molecular characterization of the systems biology of disease, and the digitalization of medicine, will start to reduce the ine xorably increasing costs of health care. This can, in principle, enable us to afford to provide health care for the more than 45 million people in the United States w ho cur rently lack health insurance, to reduce the crushing costs of health care to society, and to export our digital predictive and preventive medical approach to the developing world. Just as the mobile phone has become a fundamental communication mode in de veloping countries and has changed the li ves of much of the world’s population, so digital medicine of the 21st century can bring to the world’s citizens a global and strongly improved state of human health and health care. In our vie w, P4 medicine will become the foundational frame work for global medicine. We can no w understand how systems biology (and medicine) in its ability to deal with comple xity will allow us to be gin to attack some of the most fundamental challenges of humankind, in this case global health care for all individual humans.

642

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

REFERENCES 1. Wang X, Yu J, Sreekumar A, et al. Autoantibody signatures in prostate cancer. N Engl J Med 2005;353:1224–35. 2. Papadopoulou E, Da vilas E, Sotiriou V, et al. Cell-free DN A and RNA in plasma as a new molecular marker for prostate and breast cancer. Ann N Y Acad Sci 2006;1075:235–43. 3. Scherzer CR, Eklund AC, Morse LJ, et al. Molecular markers of early Parkinson’s disease based on gene e xpression in blood. Proc Natl Acad Sci U S A 2007;104:955–60. 4. Solanky KS, Baile y NJ , Beckwith-Hall BM, et al. Application of biofluid 1H nuclear magnetic resonance-based metabonomic techniques for the anal ysis of the biochemical ef fects of dietar y isoflavones on human plasma prof ile. Anal Biochem 2003;323: 197–204. 5. Buttner P, Mosig S, Lechter mann A, et al. Ex ercise af fects the gene expression profiles of human w hite b lood cells. J Appl Ph ysiol 2007;102:26–36. 6. Golub TR, Slonim DK, Tamayo P, et al. Molecular classif ication of cancer: class disco very and class prediction b y gene e xpression monitoring. Science 1999;286:531–7. 7. Quackenbush J. Microarray analysis and tumor classification. N Engl J Med 2006;354:2463–72. 8. Ramaswamy S, Tamayo P, Rifkin R, et al. Multiclass cancer diagnosis using tumor gene e xpression signatures. Proc Natl Acad Sci U S A 2001;98:15149–54. 9. Wang Y, Klijn JG, Zhang Y, et al. Gene-expression profiles to predict distant metastasis of l ymph-node-negative primary breast cancer. Lancet 2005;365:671–9. 10. Anderson NL, Anderson NG. The human plasma proteome: history, character, and diagnostic prospects. Mol Cell Proteomics 2002;1:845–67. 11. Fujii K, Nakano T, Kawamura T, et al. Multidimensional protein profiling technolo gy and its application to human plasma proteome. J Proteome Res 2004;3:712–8. 12. Hood L, Heath JR, Phelps ME, Lin B . Systems biology and new technologies enab le predicti ve and pre ventative medicine. Science 2004;306:640–3. 13. Lathrop JT, Anderson NL, Anderson NG, Hammond DJ. Therapeutic potential of the plasma proteome. Cur r Opin Mol Ther 2003;5:250–7. 14. Lee HJ, Lee EY, Kwon MS, P aik YK. Biomark er discovery from the plasma proteome using multidimensional fractionation proteomics. Curr Opin Chem Biol 2006;10:42–9. 15. Wilson JF. The rock y road to useful cancer biomark ers. Ann Inter n Med 2006;144:945–8. 16. Fleischmann RD, Adams MD, White O, et al. Whole-genome random sequencing and assembly of Haemophilus influenzae Rd. Science 1995;269:496–512. 17. Blattner FR, Plunk ett G III, Bloch CA, et al. The complete genome sequence of Escherichia coli K-12. Science 1997;277:1453–74. 18. Gardner MJ, Shallom SJ, Carlton JM, et al. Sequence of Plasmodium falciparum chromosomes 2, 10, 11 and 14. Nature 2002;419: 531–4. 19. Hall N, Pain A, Ber riman M, et al. Sequence of Plasmodium f alciparum chromosomes 1, 3-9 and 13. Nature 2002;419:527–31. 20. Goffeau A, Park J, Paulsen IT, et al. Multidrug-resistant transport proteins in y east: complete in ventory and ph ylogenetic characterization of y east open reading frames with the major f acilitator superfamily. Yeast 1997;13:43–54. 21. Lander ES, Linton LM, Birren B, et al. Initial sequencing and analysis of the human genome. Nature 2001;409:860–921. 22. Waterston RH, Lindblad-Toh K, Birney E, et al. Initial sequencing and comparative anal ysis of the mouse genome. Nature 2002;420:520–62.

23. Mikkelsen T. Initial sequence of the chimpanzee genome and comparison with the human genome. Nature 2005;437:69–87. 24. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A 1977;74:5463–7. 25. Madabhushi RS. Separation of 4-color DN A sequencing e xtension products in nonco valently coated capillaries using lo w viscosity polymer solutions. Electrophoresis 1998;19:224–30. 26. Prober JM, Trainor GL, Dam RJ, et al. A system for rapid DNA sequencing with fluorescent chain-terminating dideoxynucleotides. Science 1987;238:336–41. 27. Hall N. Advanced sequencing technologies and their wider impact in microbiology. J Exp Biol 2007;210:1518–25. 28. Margulies M, Egholm M, Altman WE, et al. Genome sequencing in microfabricated high-density picolitre reactors. Nature 2005;437:376–80. 29. Fan JB, Chee MS, Gunderson KL. Highl y parallel genomic assa ys. Nat Rev Genet 2006;7:632–44. 30. Mullighan CG, Goorha S, Radtk e I, et al. Genome-wide anal ysis of genetic alterations in acute l ymphoblastic leukaemia. Nature 2007;446:758–64. 31. Tomlinson I, Webb E, Car vajal-Carmona L, et al. A genome-wide association scan of tag SNPs identifies a susceptibility variant for colorectal cancer at 8q24.21. Nat Genet 2007;39:984–8. 32. Krynetskiy E, McDonnell P . Building indi vidualized medicine: prevention of adverse reactions to warfarin therapy. J Pharmacol Exp Ther 2007;322:427–34. 33. Thomas FJ, McLeod HL, Watters JW. Pharmacogenomics: the influence of genomic variation on drug response. Curr Top Med Chem 2004;4:1399–409. 34. Wu AH, Wang P, Smith A, et al. Dosing algorithm for w arfarin using CYP2C9 and VKORC1 genotyping from a multi-ethnic population: comparison with other equations. Phar macogenomics 2008;9:169–78. 35. Brenner S, Johnson M, Bridgham J , et al. Gene e xpression analysis by massively parallel signature sequencing (MPSS) on microbead arrays. Nat Biotechnol 2000;18:630–4. 36. Patterson SD , Aebersold RH. Proteomics: the f irst decade and beyond. Nat Genet 2003;33 Suppl:311–23. 37. Link AJ, Eng J, Schieltz DM, et al. Direct anal ysis of protein complexes using mass spectrometry. Nat Biotechnol 1999;17:676–82. 38. Sadygov RG, Eng J, Durr E, et al. Code developments to improve the efficiency of automated MS/MS spectra inter pretation. J Proteome Res 2002;1:211–5. 39. Wolters DA, Washburn MP, Yates JR III. An automated multidimensional protein identif ication technology for shotgun proteomics. Anal Chem 2001;73:5683–90. 40. Gygi SP, Rist B , Gerber SA, et al. Quantitati ve analysis of comple x protein mixtures using isotope-coded af finity tags. Nat Biotechnol 1999;17:994–9. 41. Han DK, Eng J , Zhou H, Aebersold R. Quantitati ve prof iling of differentiation-induced microsomal proteins using isotope-coded affinity tags and mass spectrometr y. Nat Biotechnol 2001;19:946–51. 42. Tian Q, Stepaniants SB , Mao M, et al. Inte grated genomic and proteomic analyses of gene expression in Mammalian cells. Mol Cell Proteomics 2004;3:960–9. 43. Ong SE, Blagoev B, Kratchmarova I, et al. Stable isotope labeling by amino acids in cell culture, SILA C, as a simple and accurate approach to e xpression proteomics. Mol Cell Proteomics 2002;1:376–86. 44. Ross PL, Huang YN, Marchese JN, et al. Multiplexed protein quantitation in Saccharomyces cerevisiae using amine-reactive isobaric tagging reagents. Mol Cell Proteomics 2004;3:1154–69. 45. Zhang H, Li XJ, Martin DB, Aebersold R. Identification and quantification of N-link ed gl ycoproteins using h ydrazide chemistr y,

Systems Biology

46. 47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60. 61.

62.

63.

64.

stable isotope labeling and mass spectrometr y. Nat Biotechnol 2003;21:660–6. Zhang H, Liu AY, Loriaux P, et al. Mass spectrometric detection of tissue proteins in plasma. Mol Cell Proteomics 2007;6:64–71. Zhang H, Yi EC, Li XJ, et al. High throughput quantitati ve analysis of serum proteins using glycopeptide capture and liquid chromatography mass spectrometry. Mol Cell Proteomics 2005;4:144–55. Sun B , Ranish J A, Utle g AG, et al. Shotgun gl ycopeptide capture approach coupled with mass spectrometr y for comprehensive glycoproteomics. Mol Cell Proteomics 2007;6:141–9. Anderson L, Hunter CL. Quantitative mass spectrometric multiple reaction monitoring assa ys for major plasma proteins. Mol Cell Proteomics 2006;5:573–88. Liao H, Wu J, Kuhn E, et al. Use of mass spectrometry to identify protein biomarkers of disease severity in the synovial fluid and serum of patients with rheumatoid ar thritis. Arthritis Rheum 2004;50:3792–803. Anderson NL, Anderson NG, Haines LR, et al. Mass spectrometric quantitation of peptides and proteins using Stab le Isotope Standards and Capture by Anti-Peptide Antibodies (SISCAPA). J Proteome Res 2004;3:235–44. Haab BB, Dunham MJ, Brown PO. Protein microarrays for highly parallel detection and quantitation of specif ic proteins and antibodies in comple x solutions. Genome Biol 2001;2: RESEARCH0004. Huang RP, Huang R, F an Y, Lin Y. Simultaneous detection of multiple cytokines from conditioned media and patient’s sera by an antibodybased protein array system. Anal Biochem 2001;294:55–62. Olle EW, Sreekumar A, Warner RL, et al. De velopment of an internally controlled antibody microar ray. Mol Cell Proteomics 2005;4:1664–72. Gao WM, Kuick R, Orchekowski RP, et al. Distinctive serum protein profiles in volving ab undant proteins in lung cancer patients based upon antibody microarray analysis. BMC Cancer 2005; 5:110. Orchekowski R, Hamelinck D, Li L, et al. Antibody microarray profiling reveals individual and combined serum proteins associated with pancreatic cancer. Cancer Res 2005;65:11193–202. Miller JC, Zhou H, Kwekel J, et al. Antibody microarray profiling of human prostate cancer sera: antibody screening and identif ication of potential biomarkers. Proteomics 2003;3:56–63. Lyon LA, Musick MD, Natan MJ. Colloidal Au-enhanced surface plasmon resonance immunosensing. Anal Chem 1998;70: 5177–83. Bailey RC, Kwong GA, Radu CG, et al. DNA-encoded antibody libraries: a unif ied platfor m for multiple xed cell sor ting and detection of genes and proteins. J Am Chem Soc 2007;129: 1959–67. Ruvkun G. Molecular biolo gy. Glimpses of a tin y RNA world. Science 2001;294:797–9. Ito T, Chiba T, Ozawa R, et al. A comprehensive two-hybrid analysis to explore the yeast protein interactome. Proc Natl Acad Sci U S A 2001;98:4569–74. Usui-Aoki K, Shimada K, Nagano M, et al. A novel approach to protein e xpression prof iling using antibody microar rays combined with surf ace plasmon resonance technolo gy. Proteomics 2005;5:2396–401. Tian Q, F eetham MC, Tao WA, et al. Proteomic anal ysis identif ies that 14-3-3zeta interacts with beta-catenin and facilitates its activation by Akt. Proc Natl Acad Sci U S A 2004;101:15370–5. Tu LC, Yan X, Hood L, Lin B. Proteomics analysis of the interactome of N-myc downstream regulated gene 1 and its interactions with the androgen response program in prostate cancer cells. Mol Cell Proteomics 2007;6:575–88.

643

65. Robertson G, Hirst M, Bainbridge M, et al. Genome-wide prof iles of STAT1 DN A association using chromatin immunoprecipitation and massively parallel sequencing. Nat Methods 2007;4:651–7. 66. Acevedo LG, Iniguez AL, Holster HL, et al. Genome-scale ChIPchip analysis using 10,000 human cells. Biotechniques 2007;43: 791–7. 67. Bulyk ML. DN A microar ray technolo gies for measuring proteinDNA interactions. Curr Opin Biotechnol 2006;17:422–30. 68. Euskirchen GM, Rozowsky JS, Wei CL, et al. Mapping of transcription factor binding regions in mammalian cells by ChIP: comparison of ar ray- and sequencing-based technolo gies. Genome Res 2007;17:898–909. 69. Wu J, Smith L T, Plass C, Huang TH. ChIP-chip comes of age for genome-wide functional anal ysis. Cancer Res 2006; 66:6899–902. 70. Vastrik I, D’Eustachio P, Schmidt E, et al. Reactome: a knowledge base of biolo gic pathw ays and processes. Genome Biol 2007;8:R39. 71. Stark C, Breitkreutz BJ , Reguly T, et al. BioGRID: a general repository for interaction datasets. Nucleic Acids Res 2006;34:D535–9. 72. Heath JR, Da vis ME. Nanotechnolo gy and cancer . Annu Re v Med 2008;59:251–65. 73. Czernin J, Phelps ME. P ositron emission tomography scanning: current and future applications. Annu Rev Med 2002;53:89–112. 74. Mischel PS, Cloughesy TF, Nelson SF. DNA-microarray analysis of brain cancer: molecular classif ication for therap y. Nat Re v Neurosci 2004;5:782–92. 75. Sotiriou C, Piccart MJ. Taking gene-expression profiling to the clinic: when will molecular signatures become rele vant to patient care? Nat Rev Cancer 2007;7:545–53. 76. Xu L, Tan AC, Naiman DQ, et al. Robust prostate cancer mark er genes emer ge from direct inte gration of inter -study microar ray data. Bioinformatics 2005;21:3905–11. 77. Price ND , Trent J , El-Naggar AK, et al. Highl y accurate two-gene classif ier for dif ferentiating gastrointestinal stromal tumors and leiom yosarcomas. Proc Natl Acad Sci U S A 2007;104:3414–9. 78. Tan AC, Naiman DQ, Xu L, et al. Simple decision r ules for classifying human cancers from gene expression profiles. Bioinformatics 2005;21:3896–904. 79. Ma XJ, Wang Z, Ryan PD, et al. A two-gene expression ratio predicts clinical outcome in breast cancer patients treated with tamo xifen. Cancer Cell 2004;5:607–16. 80. Hwang D, Rust AG, Ramsey S, et al. A data integration methodology for systems biolo gy. Proc N atl Acad Sci U S A 2005;102: 17296–301. 81. Shannon P, Markiel A, Ozier O, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 2003;13:2498–504. 82. Longabaugh WJ, Da vidson EH, Bolouri H. Computational representation of developmental genetic regulatory networks. Dev Biol 2005;283:1–16. 83. Hwang D, Smith JJ, Leslie DM, et al. A data integration methodology for systems biolo gy: e xperimental v erification. Proc Natl Acad Sci U S A 2005;102:17302–7. 84. Price ND, Shmulevich I. Biochemical and statistical netw ork models for systems biology. Curr Opin Biotechnol 2007;18:365–70. 85. Shmulevich I, Dougherty ER, Kim S, Zhang W. Probabilistic Boolean Networks: a rule-based uncertainty model for gene regulatory networks. Bioinformatics 2002;18:261–74. 86. Shmulevich I, Dougherty ER, Zhang W. Gene perturbation and intervention in probabilistic Boolean netw orks. Bioinfor matics 2002;18:1319–31. 87. Basso K, Margolin AA, Stolovitzky G, et al. Reverse engineering of regulatory networks in human B cells. Nat Genet 2005;37:382–90.

40 PROTEIN ENGINEERING MOLECULAR IMAGING

FOR

ANNA M. WU, PHD

Protein engineering—the ability to design, modify , and produce proteins at will—has revolutionized the fields of biochemistry, molecular biology, and cell biology. As the emphasis shifts to molecular imaging—detection and measurement of specif ic molecules, molecular e vents, and molecular processes in li ving or ganisms—protein engineering is also pla ying an increasingl y impor tant role. The vast majority of molecular imaging biomark ers and tar gets are proteins, so it is not sur prising that the ability to isolate, engineer , and manipulate proteins is playing an increasingly important role in the development of new molecular imaging approaches. This chapter will provide a survey of current protein engineering technolo gies and then pro vide e xamples of the contributions of protein engineering to molecular imaging. A star ting point is the imaging tar get itself— many of the original molecular imaging approaches are based on the ability to detect specif ic proteins, w hich includes enzymes (e g, he xokinase, th ymidine kinase [tk]), receptors (for neurotransmitters, g rowth f actors, hormones, etc), tissue or dif ferentiation mark ers, and other specific proteins. Progress in the ability to identify new molecular imaging tar gets has been dri ven b y a deeper understanding of biolo gy—as adv ances in genomics and proteomics allo w investigators to identify molecular signatures that are characteristic of nor mal tissue development and function, or representative of disease states. Furthermore, validation of targets as potential intervention points opens the w ay for de velopment of molecularly targeted therapies. Once a key molecular target has been identified, molecular imaging needs to pro vide tw o functions: specificity, to single out the tar get of interest; and signal generation (sensitivity), such that acti vity can be monitored noninvasively in living subjects. Again, proteins can 644

provide k ey star ting points. F or e xample, antibodies represent a class of proteins w hose natural pur pose is to recognize tar get molecules with e xquisite specif icity. Adaptation for use in molecular imaging requires engineering to optimize their af finity if needed , as w ell as modification of phar macokinetics and reduction of immunogenicity. Finally, a means for detection needs to be incor porated—by conjugation or fusion to signalgenerating moieties. The need for antibodies or antibodylike proteins to pro vide specif ic tar get binding and recognition (sometimes refer red to as “affinity reagents” or “binders”) has spa wned technologies, such as librar y display, where large reper toires of proteins with di verse potential binding acti vities are e xpressed on microbes, such as phage (bacterial vir uses) or y east, or on mammalian cells. Application of rigorous and w ell-designed selection protocols to large diverse libraries has proven to be a robust pathw ay for generation of reagents with an y desired recognition properties1,2 (see Chapter 41, “Phage Display for Imaging Agent Development”). Indirect imaging through the use of repor ter genes is a complementary approach for noninvasive monitoring of molecular and cellular e vents. Repor ter gene imaging relies on cellular expression of a gene encoding a protein that produces a detectab le signal. Widely used e xamples of reporters include fluorescent proteins and luciferases, which produce visib le light that can be detected using sensitive cooled char ge-coupled de vices (CCDs). Introduction of enzymes that will enhance trapping of radioactive probes (such as her pes simplex vir us tk (HSV1-tk) paired with radiolabeled nucleoside analo gs, such as fluorinated p yrimidine analo gs or ac ycloguanosines) provide a means for detection of reporter gene activity by positron emission tomo graphy (PET) imaging. Applications of repor ter gene imaging continue to e xpand and

Protein Engineering for Molecular Imaging

currently include cell tracking, monitoring gene e xpression, assessing protein-protein interactions, and other novel approaches. The ability to engineer repor ter proteins and enzymes has e xtended their utility . See Chapter 42, “Molecular Imaging of Gene Therapy,” Chapter 43, “De veloping Diagnostic and Therapeutic Viral Vectors,” Chapter 47, “Molecular Imaging of Protein-Protein Interactions,” and Chapter 48, “Fluorescence Readouts of Biochemistry in Live Cells and Organisms,” for further details.

BASIC CONCEPTS IN PROTEIN ENGINEERING Proteins represent a broad and di verse class of biolo gic macromolecules. This diversity of structure and function is rooted in the 20 building b locks—the amino acids— that are assemb led into pol ypeptide chains. Amino acid side chains include h ydrophilic, h ydrophobic, acidic, basic, aromatic, chemicall y reacti ve, and neutral moieties, and span a v ariety of sizes from minimal (e g, glycine) to bulk y (e g, isoleucine). P olypeptide chains fold into comple x, highl y specif ic three-dimensional structures w hich can contain common motifs (alphahelices, beta-pleated sheets) or unique loops and folds; these in turn can assemble into multimeric structures that may become co valently linked as w ell; and posttranslational modifications further serve to f ine-tune the structures and functions of proteins. Specif ic alterations to proteins can be achie ved through chemical and biochemical means, for e xample, by chemical conjugation. Nonnatural amino acids can also be incor porated into proteins and peptides, through protein engineering or synthetic approaches, to impro ve stability or introduce new functions.3 One of the main challenges to working with proteins is their diversity—minor differences in sequence or structure can lead to lar ge dif ferences in their biochemical and biophysical proper ties. As a result, introduction of specific modif ication into proteins can be challenging. An e xtensive range of chemical and biochemical approaches to protein modif ication has been de veloped; however, application to indi vidual protein projects often requires signif icant experimentation and optimization to achieve the desired results. Instead of direct modif ication of mature pol ypeptides and proteins, protein engineering provides a shortcut around some of the challenges of manipulating proteins. One simpl y creates or modif ies a deo xyribonucleic acid (DN A) sequence that encodes the

645

desired protein, to change its primar y sequence and ultimately the proper ties of the f inal expressed protein. Current recombinant DNA technologies enable full control over a DNA sequence, and thus, the encoded protein sequence. Using recombinant DNA approaches, one can assemble and reassemble protein domains in an y order, create repeats or multimers, fuse protein se gments to combine functions into no vel combinations, and introduce highl y specif ic indi vidual modif ications for a variety of purposes.

Sources of Target Genes Completion of the sequencing of the human genome opens access to all of the appro ximately 25,000 genes. Gene expression prof iling, proteomics approaches, systems biolo gy anal yses, and e xtensive v alidation w ork have led to the identif ication of genes that encode proteins of interest as diagnostic or therapeutic tar gets. Many cloned complementar y DN A (cDN A) and genomic sequences are a vailable through repositories, such as the American Type Culture Collection (A TCC). Alternatively, if the sequence of a gene of interest is known, it is a straightforward process to construct a synthetic gene. This is typically accomplished by synthesizing a set of long, overlapping DNA oligonucleotides that encode the sequence of the protein of interest, then assembling and amplifying the gene using the pol ymerase chain reaction (PCR). Although this can be readily achieved in standard molecular biolo gy laboratories, it is also possib le to simply order genes of interest from commercial entities, and the cost of these ser vices is steadily dropping.

Sources of Probe Genes (Antibodies and Beyond) Antibodies—proteins produced b y the mammalian immune systems for specif ic responses to patho gens— represent the prototype for protein-based molecular probes. The de velopment of h ybridoma technolo gy b y Milstein in 1975 opened the door to renewable sources of monoclonal antibodies with high af finity and e xquisite specificity for their cognate antigens. Recombinant DNA technology enab led retrie val of the genes encoding the variable regions of individual monoclonal antibodies with desired specificities; this in turn has enabled engineering of derivatives that retain the original binding specif icity encoded in those v ariable regions, while allowing broad freedom to manipulate an y adjoining protein sequences.

646

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Specific applications (such as chimerization, humanization, enhancement of ef fector functions, production of engineered fragments and fusion proteins) will be described in a following section. Recently the broad reach of antibody technologies has been extended even further through display technologies.4 Instead of rel ying on the mammalian immune system, display technolo gies rel y on creation of lar ge, di verse libraries of binding proteins (such as F ab fragments or single-chain antibody fragments [scFv]). These libraries can be encoded in bacteriophage or y east in a w ay that “displays” the binding protein on the surf ace of the microbe or particle. Powerful selection methods can result in retrie val of indi vidual clones from the librar y with very high specificity and affinity for the target of interest2 (see Chapter 41, “Phage Displa y for Imaging Agent Development”). A k ey feature of displa y technolo gies is that the y provide a means to obtain proteins with desired binding properties in a w ay that is linked to the DN A that encodes the protein. This physical linkage—between the phenotype (ability of a protein to bind a v ery specif ic target) and genotype (the DN A encoding this binding specificity), allows retrieval and further manipulation of the DN A sequence to produce no vel binding proteins that retain the desired specif icity. Librar y displa y approaches essentiall y mimic the mammalian immune system (in which B-lymphocytes displaying or secreting a monoclonal antibody pro vide a source of the immunoglobulin variable re gion genes that encode the desired specif icity). Cur rent phage displa y, y east display, ribosome display, or in vitro compartmentalization strategies use microbial or cell-free systems to accelerate the selection, amplif ication, and affinity-maturation procedures, with man y steps perfor med in vitro. In the end, the goal is the same—rapid isolation of a DN A sequence that encodes a protein moiety with highly specific binding to the tar get of interest, w hich can then serve as a building b lock for fur ther engineering. Display technologies have also been extended to antibodies from sources be yond the murine immune system, including naïve (non-immunized) antibody reper toires, human antibody libraries, and e ven fully synthetic antibody repertoires.4 The advent of display technology has also spawned a bevy of alternative protein scaffolds that can serve the same general function as antibody domains—a conserved, stab le protein core with e xposed strands or loops that can be randomized to generate a broad ar ray of binding specif icities.5 These scaf folds can of fer

advantages o ver antibody v ariable domains, such as alternative form factors that might yield higher affinity ligands for proteins of dif ferent shape classes, ease of production in microbial systems, smaller size, and higher stability. (Antibodies as a class are quite stab le, particularly under ph ysiological conditions, but some of the ne wer scaffolds can tolerate more e xtreme conditions of high temperature, non-physiological pH, etc). Examples include camellid antibodies (“nanobodies”),6 single domain human antibodies (dAbs),7 and scaffolds based on a protein A domain (“af fibodies”),8 a small human receptor domain (“a vimers”),9 ankyrin repeats (“DARpins”), 10 cysteine-knot miniproteins, 11 and others.

PROTEIN ENGINEERING BASICS Designer Genes Classic recombinant DNA technology—restriction digestion and ligation to reassemble preexisting DNA segments into new configurations, remains the backbone of protein engineering. PCR has added an e ven g reater le vel of control over the DNA sequences that can be produced, and thus, the sequence of the encoded proteins. PCR methods allow amplif ication of the star ting sequence and simultaneous incorporation of desired modifications by including the ne w sequence infor mation in the oligonucleotide primers used in the amplif ication. PCR-based methods can be used to inser t or change restriction endonuclease sites (to enab le further gene constr uction or assemb ly by cleavage and ligation); for site-directed mutagenesis (b y using oligonucleotides encoding the desired change); insertion of shor t peptide tags (using long oligonucleotides that result in addition of e xtra sequences to the amplified DNA sequence); and fusion of protein domains (by incor porating long oligonucleotides that generate an overlap, with or without a link er peptide betw een the elements being fused). In addition to engineering the desired acti vity or function, additional modif ications can be incor porated into a DN A sequence to enhance the utility of the encoded protein. Additional options include selection of optimal promoters (and an y required enhancers and introns for mammalian e xpression), incor poration of appropriate ribosome reco gnition sequences for ef ficient initiation of translation, codon optimization to match the cell type chosen for e xpression, and termination and polyadenylation sites (Figure 1). Cellular localization can be dictated b y incor poration of signal

Protein Engineering for Molecular Imaging

647

Cell secreting recombinant protein

Gene of interest P, rbs

ss Coding sequence

Fusion partner Linker

Cell expressing reporter protein (intracellular) Tags

Figure 1. Construction and expression of engineered proteins. The many components of a recombinant protein can readily be assembled by polymerase chain reaction (PCR). Typically, one requires a promotor and ribosome binding site (P, rbs; red) for efficient transcription and translation in the cell of choice. If cell surface expression or secretion is desired, a signal sequence (ss; orange) is required. The example shown indicates a fusion protein comprised of the gene of interest (green) and its fusion partner (blue) connected by a linker peptide (yellow). Codon optimization can be used to optimize expression in the selected microbial or mammalian system. A variety of short peptide tags (purple) can be added to aid in detection and purification of the recombinant protein. On the right are two general pathways for expressing engineered proteins. Recombinant proteins can be secreted and purified for use as detection and signal-generating agents. Alternatively, they can be expressed intracellularly (shown) or on the cell surface to serve as reporters.

sequences. For example, leader peptides can direct proteins to the cell surf ace for membrane e xpression or secretion. Classic e xamples of manipulation of cellular localization signals in the repor ter gene f ield include deletion of the nuclear localization signal (NLS) from the native HSV tk gene, and similar deletion of the natural pero xisomal localization signal from f irefly luciferase, to allow cytoplasmic expression of these proteins and enhance activity. Another important modification is the incorporation of additional short peptides that can be used to enhance detection of the engineered protein, including commonl y used myc, FLAG™ tags, and hexahistidine tags. Corresponding myc-, FLAG-, or

6His-specific antibodies can be used in enzyme-link ed immunosorbent assa ys (ELISAs), Western b lots, or in flow c ytometry to conf irm e xpression of the recombinant protein. These tags can also be used for af finity capture w hen the intent is to secrete and purify the encoded protein (Figure 1).

Mutagenesis—Site-Directed or Random Manipulation of DN A sequences allo ws in vestigators to introduce specif ic changes in the amino acid residues that comprise a protein. Generall y, there are two broad approaches to mutagenesis of proteins,

648

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Rational mutagenesis based on: crystallographic structure sequence homology molecular modeling known mutations

RLuc8 BDC

N53Q

RLuc8

F181Y

F262W

A123S/ D162E/I163L

Irrational mutagenesis: error-prone PCR mutator strains

A123S/ D162L/I163V

RLuc8.62 535 RLuc8.62 547

RLuc8 CV

RLuc8.62 535 CV

RLuc8.62 545 RLuc8.62 547 CV CV

Figure 2. Rational versus irrational mutagenesis. The primary structure and function of a protein can be modified by a rational process, in which data from X-ray crystallographic or other structural analyses, sequence homology, mutation analysis, or molecular modeling can be used to identify key amino acid residues and anticipate functional improvements when these residues are modified. Alternatively, in an “irrational” approach, the investigator generates a large, random, diverse set of variants from the original protein and simply screens for mutants with the desired properties. An example combining both approaches can be found in Loening et al.46

which can be refer red to as “rational” v ersus “irrational.” (F igure 2) In rational mutagenesis, one typically has a good understanding of the critical regions and amino acid residues required for protein activity (eg, binding ability or enzymatic acti vity), and specific, predeter mined alterations are designed. The underlying str uctural and acti vity infor mation can be gleaned a number of w ays, including analysis of X-ray crystallographic, nuclear magnetic resonance (NMR), or other str uctures, computer modeling of protein structures, comparison to homolo gous proteins with similar activities, and through kno wledge of canonical sequences in classes of proteins, such as the hypervariable loops in antibodies. Specif ic individual amino acid residue changes can be introduced by site-directed mutagenesis, using synthetic oligonucleotides that match the coding region of the target gene, with the e xception that the oligo encodes the desired altered codon. PCR is used to amplify the tar get gene, incorporating the modif ied

codon. A v ariety of commercial kits are a vailable, making deliberate incorporation of specific amino acid changes readily achievable.

Directed Evolution of Proteins Mutagenesis is also a powerful means for diversifying an existing protein, so that one can select impro ved variants (the “irrational” approach). For example, an investigator might desire to shift the proper ties of an enzyme (to a different substrate or to impro ve catalysis rates) or ma y want to improve the affinity of an antibody to its antigen. For saturation mutagenesis of specific residues or regions of a protein, the site-directed mutagenesis approaches described abo ve can be used , but instead of using a defined oligonucleotide containing a specif ic codon change, degenerate oligonucleotides encoding all 20 possible amino acids are incor porated. If desired, potentially problematic amino acid residues can be omitted from the

Protein Engineering for Molecular Imaging

Therapeutic gene

Reporter 1

649

Reporter 2

A Individual cistrons

B IRES

C Fusion protein

D Self-cleaving fusion protein Figure 3. Linked expression of therapeutic/reporter genes. Strategies for coordinated expression of therapeutic genes and/or multimodal reporter genes include the following: A, coexpressing each gene as an individual transcription/translation unit; B, using internal ribosome entry site (IRES) sequences to reinitiate translation on a polycistronic messenger ribonucleic acid (mRNA); C, directly fusing the individual component genes to encode a single fusion protein, with or without additional peptide linkers; and D, assembling and producing a fusion protein incorporating self-cleaving peptides in the linkers, resulting in release of each individual protein in equimolar amounts.

mix, such as prolines (w hich force a bend in the peptide backbone) and cysteines (which can induce cross-linking and aggregation if unpaired). Broad sequence di versification across the entire coding region of a target protein is also easy to achieve. In this approach, one creates a “librar y” of sequences derived from the original DN A/protein sequence. Each member has one or a small number of random amino acid substitutions. The library can be screened for variants with the desired impro ved proper ties. Two of the most popular methods for generating these libraries from a kno wn cloned sequence are er ror-prone PCR and the use of mutator strains of bacteria for plasmid propagation. In er ror-prone PCR, the gene of interest is amplified under intentionall y suboptimal conditions. F or example, increasing the Mg ++ concentration or adding Mn++ will weaken the fidelity of the polymerase used in the reaction, resulting in random incor poration of incorrect bases. The frequency of misincorporation can be roughly titrated to produce random mutations at the desired frequency (eg, one per 400 bases). The resultant mixture of PCR products needs to be efficiently ligated into an appropriate plasmid vector to generate a library. Proteins are e xpressed from the librar y, and selected and screened for the desired impro vements. Alternatively, a plasmid car rying the gene of interest can be propagated using a mutator strain of Escherichia coli (E. coli) as the host. These strains, commercially available, are def icient in DN A repair pathw ays, w hich results in the steady, low-level incorporation of random base substitutions, deletions, and inser tions into DN A as it is replicated.

Fusion Proteins Protein engineering techniques ha ve added a major element of v ersatility to the molecular imaging f ield. In particular, it is often desirab le to link the e xpression of two or more proteins for repor ting purposes—for example, a repor ter gene and a therapeutic gene, or repor ter genes representing different detection modalities (F igure 3). Expression has been link ed transcriptionally through the use of tandem promoters or bidirectional promoters. Translational linkage, through the use of inter nal ribosome entr y site (IRES) sequences allo ws e xpression of two proteins from the same messenger ribonucleic acid (mRNA) transcript, although signif icantly lower expression of the do wnstream protein is typicall y obser ved, since the IRES site is typically not as efficient as a translation initiation site compared with the primary ribosome binding site.12 A more direct approach is to produce fusion proteins, by joining the genes of interest in-frame, with or without connecting peptide link ers. For example, fluorescent, bioluminescent, and PET reporter proteins have been combined into a triple fusion protein, representing an example of a uni versal repor ter gene. Although this approach ensures that all three of the component proteins are present in equimolar amounts, activity levels of the indi vidual components in the fusion ma y be lo wer than the unfused proteins. Incomplete folding or steric hindrance (e g, occlusion of acti ve site) might be encountered, w hich ma y cur tail the acti vity of one or more of the components of the fusion protein. Unfor tunately, there are few clear guidelines for optimization of fusion proteins. In general, introduction of longer linker

650

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

peptides should allow each component g reater freedom to fold properly, but long, exposed linkers also increase the possibility of inadv ertent proteol ytic clea vage, which w ould destro y the linkage of the component activities. More recentl y, self-clea ving peptide link ers have been used in fusion protein strate gies. In this approach, the component protein units are synthesized as a contiguous protein precursor , with a self-clea ving peptide sequence, such as the 18-amino acid 2A peptide from foot-and-mouth disease vir us (FMD V), incor porated as a linker between the components.13 The 2A peptide causes “skipping” of its C-ter minal peptide bond , resulting in its release along with upstream pol ypeptide sequences. The ribosome and mRNA remain associated and continue ef ficient translation of do wnstream polypeptides. The net result is equimolar e xpression of each of the indi vidual component proteins in highl y active for m for therap y and/or detection using the appropriate repor ter probe(s). Protein ligation and protein trans-splicing also pro vide alter native approaches for fusing functional domains. 14,15 Another major contribution of fusion protein approaches to molecular imaging has been through the linkage of tar get reco gnition domains with signalgenerating domains—to ph ysically combine reco gnition with detection. An e xample (described in g reater detail belo w) is the production of antibody-luciferase fusion proteins. These serve, in effect, as “reporter proteins” and produce an optical signal when localized (in contrast to repor ter gene approaches, w here a foreign gene/protein must be introduced into the system for detection). A novel approach has been the use of signaling proteins, such as g reen fluorescent protein (GFP), as a scaffold for display of binding peptides. 16 A v ariation on the use of fusion proteins has been their use in “pro ximity” assa ys for molecular imaging. This approach allows the investigator to ask whether two proteins of interest are close to or in contact with each other. Each target protein can be expressed as a fusion to a signaling moiety that is silent unless the tw o components are in close pro ximity. Classic examples are based on fluorescence resonance energy transfer (FRET) where emission from one fluorophore can excite an adjacent fluorophore to emit a distincti ve signal. Fusions of proteins of interest to the appropriate fluorescent proteins ha ve similarly been used. A more recent variation is the use of fusion proteins in bioluminescent resonance energy transfer (BRET).17 In this case, target protein “A” is fused to a luciferase and tar get protein “B” to an appropriate fluorescent protein; onl y if A and B are in close association

will addition of the appropriate luciferin result in energy transfer and emission by the fluorescent protein component. Engineering Fusion Proteins

Construction of genes encoding fusion proteins has been considerably simplif ied by the adv ent of PCR methods. In par ticular, “o verlap e xtension” PCR is a uni versal method for fusing gene se gments. In this approach, each of the two component genes is amplif ied by PCR, using long oligonucleotides that incor porate an e xtension that overlaps with the DN A segment to be joined. The result of this f irst set of PCR reactions is that each amplif ied product has additional DN A appended, w hich creates a long o verlapping sequence. The tw o fragments can be mixed and e xtended, and the o verlapped molecules are further amplif ied b y PCR to generate quantities of the now fused gene for fur ther cloning and e xpression. As noted above, a link er peptide can be easil y incor porated between the fused domains, b y including the necessar y additional sequences in the long oligonucleotides used to create the extensions. Fusion Protein versus Conjugation

An o verarching question is w hy one should utilize a fusion protein approach at all, w hen it is possib le to join two proteins simpl y b y chemical cross-linking or conjugation, using methods estab lished decades ago. Synthesizing the appropriate gene constr uct and coaxing cells into stab le, high-le vel production of the encoded protein is laborious and time-consuming; furthermore, the process must be repeated for each individual protein combination. Chemical conjugation, in contrast, of fers the potential for freel y mixing and matching protein components, using of f-the-shelf reagents to link them. Nonetheless, for man y applications, production of fusion proteins remains the best and sometimes onl y option. Chemical approaches to linking proteins are typically nonspecific; cross-linking reagents will react with all surf ace accessib le amino acid residues (e g, lysines, cysteines, or other residues, depending on the selected chemistr y). Random chemical modif ication can inadv ertently destro y binding or enzymatic activity. Control of the extent of conjugation can be challenging, requiring careful titration of the two protein components and the linking reagent. In short, the resultant protein conjugates are often quite

Protein Engineering for Molecular Imaging

heterogeneous biochemicall y, and it ma y be dif ficult to reproducibly obtain preparations with the desired stoichiometry and activity. In contrast, a fusion protein approach allows extensive control over the final bifunctional/multifunctional protein and ensures a more uniform product.

Split Proteins Protein engineering has also enabled “splitting” of signalgenerating enzymes, such as luciferases, for use in complementation assays. This has been a par ticularly fruitful approach for imaging protein-protein interactions. Typically, one protein under study is fused to the N-ter minal portion of a split enzyme, and the putative interacting protein is fused to the C-terminal portion. If and when the two proteins interact, the enzymatic activity is regenerated and a signal (e g, bioluminescence) is produced. GFP , f irefly luciferase (FLuc), and Renilla luciferase (RLuc) ha ve been successfully split for use in complementation assa ys for molecular imaging of protein-protein interactions. 18 See Chapter 47, “Molecular Imaging of Protein–Protein Interactions,” for further details.

APPLICATIONS OF PROTEIN ENGINEERING IN MOLECULAR IMAGING Engineered Antibodies for Molecular Imaging Classic monoclonal antibodies, deri ved from mice, ha ve grown to occup y a central position in laborator y applications, forming the basis of assays including ELISAs, Western b lots, flo w c ytometry, immunofluorescence, and immunohistochemical staining. Shor tly after the de velopment of murine monoclonal antibodies, in vestigators also began to explore the possibility of radiolabeling antibodies for detection of cancer in patients. 19 This ma y represent one of the earliest examples of clinical molecular imaging of a disease-specif ic target in patients. Since then, fur ther implementation of antibodies in the clinic has relied hea vily on protein engineering to modify and optimize their properties. Antibodies can be used as the basis of highl y specific molecular imaging agents. The greatest strength of this approach is the near -universal ability to obtain antibodies with any binding specificity. A broad overall limitation of antibody imaging in vivo is that this strategy is limited to targets that are accessib le on the cell surf ace.

651

Nonetheless, the plasma membrane represents the interface between the cell and its environment, with a host of highly specif ic proteins as potential molecular imaging targets. Cell surf ace biomarkers include se veral classes of proteins including hor mone and g rowth factor receptors, transpor ters, pumps and ion channels, adhesion molecules, proteases and de gradative enzymes, tissuespecific proteins, and dif ferentiation and acti vation markers. A handful of antibody-based imaging agents have gained U.S. Food and Dr ug Administration (FDA) approval o ver the y ears, but adoption b y the medical community has been limited. 20 In retrospect, these f irstgeneration antibody imaging agents suf fered shor tcomings including immuno genicity due to their murine origins, and suboptimal tar geting and phar macokinetic properties. An initial challenge for all clinical applications of antibodies was to reduce the immuno genicity of murine monoclonal antibodies, w hich are reco gnized as foreign proteins by the human immune system. Sequential development of chimeric (ie, mouse/human) antibodies, humanization methods, and f inally, fully human antibodies has led to proteins that e xhibit the high af finity and specificity of traditional murine monoclonal antibodies, but with the constant domains increasingl y replaced b y the cor responding domains from human antibodies. Humanized and/or human antibodies (from phage display or transgenic mice) ha ve become the de facto standard for clinical use. 4,21 Further engineering of antibody constant re gions has led to impro vements of the functional properties of antibodies, including enhanced biolo gic activity, optimized phar macokinetics, and addition of novel functions b y creation of antibody fusion proteins. For e xample, antibodies ha ve been fused to enzymes, cytokines, to xins, and other moieties for therapeutic purposes. Pharmacokinetics and Engineered Antibody Fragments

Intact, nati ve antibodies, w hich inherentl y possess long circulating half-li ves, are optimal for man y therapeutic applications, but fall short as imaging agents. An imaging agent needs to retain high specif icity for its target in vivo but also needs to clear quickly from the blood and nontarget tissues to allo w the imaging signal to be detected above background. This limitation was initially addressed by generating antibody fragments by enzymatic digestion. However, con ventional Fab’ and F(ab’) 2 fragments ha ve been superseded b y the adv ent of engineered antibody

652

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Intact

Minibody

Diabody

scFv

AA

FF

Blood activity

50

60

40

50 %ID/g

%ID/g

Tumor uptake

30 20

scFv diabody minibody scFv-Fc intact

40 30 20

10 0

D D

C C

B B

EE

ScFv-Fc

10 0

10 20 30 40 50 60 70 80 Time (h)

0

0

10

20

30 40 50 Time (h)

60

70

80

Figure 4. Engineered antibody fragments for in vivo positron emission tomography (PET) imaging of cell surface markers. The top panel shows a native, intact antibody with the variable regions that form the binding site shown in green and the constant domains in blue. Engineered fragments depicted include single-chain variable fragments (scFv), the diabody (a dimer of scFv fragments), minibody (fusion of scFv and CH3 domain), and scFv-Fc fragments (fusions of the scFv and the full Fc regions—CH2 and CH3). The middle panel shows coronal slices of co-registered microPET/microCT (micro computed tomography) images, all acquired at 20 h following intravenous administration of a radiolabeled engineered antibody fragment in athymic mice carrying different subcutaneous human tumor xenografts (orange arrows). A, Imaging of an LS174T colon cancer tumor using a carcinoembryonic antigen (CEA)-specific diabody labeled with 124I. B, Imaging of a B-cell lymphoma using CD20-specific minibody labeled with 124I. C, Detection of an LAPC-9 prostate cancer xenograft using a minibody that recognizes prostate stem cell antigen (PSCA), labeled with 124I. D, Detection of an MCF-7 breast cancer tumor overexpressing HER2 using a 64Cu-DOTA-conjugated HER2-specific scFv-Fc double mutant antibody fragment. E, F, Biodistribution data summarized for radioiodinated T84.66 anti-CEA engineered antibody fragments in the LS174T colon cancer xenograft model. Uptakes in tumor (E) or blood activity (F) are expressed as percent injected dose per gram of tissue (% ID/g).

fragments (F igure 4, top). Two engineering approaches have been used to modify antibody phar macokinetics: control of antibody size and modif ication of the antibody constant regions that control serum persistence. An important first step in engineering antibody fragments w as the de velopment of single-chain antibodies. Antibody specif icity resides in the protein sequences of the variable regions of the antibody light chain and heavy chain (VL and VH). Fusion of VL and VH, with a connecting peptide (typicall y 15–20 amino acid residues), gi ves rise to the single-chain Fv fragment (scFv) with the following k ey proper ty—antigen binding encoded in a single DN A molecule/single pol ypeptide.22 Due to the conservation of the immuno globulin (Ig) fold (Ig domains are based on sandwiches of disulf ide-linked β-pleated sheet str uctures), refor matting of the v ariable regions of a nati ve antibody into an scFv can readil y be accomplished, resulting in cor rectly folded, soluble pro-

tein that retains specific binding to the target. The process is not perfect, and often there is a tw o- to ten-fold loss in binding affinity. In instances where the loss in af finity is severe, af finity maturation can be conducted to restore the needed acti vity. Specif ic mutations designed to restore the native conformation of the binding site can be modeled and introduced. Alternatively, a small, di versified librar y can be produced (e g, b y er ror-prone PCR), and candidates with higher af finity can be selected for further development. ScFv fragments ha ve been used in numerous assa ys and applications and also ser ve as a con venient building block for larger engineered antibody fragments. A versatile and con venient fragment is the diabody , a nonco valent dimer of scFv fragments, w hich can readil y be produced by shortening the interdomain peptide linker in an scFv. Linkers of 10 amino acid residues and below are too short to allow correct association of tethered VL and

Protein Engineering for Molecular Imaging

VH domains. Instead , ener getics f avor tw o molecules coming to gether associating in a cross-paired f ashion. The net result is ef ficient production of a 55 kDa engineered antibody fragment (commonl y called a “diabody”) with two sites for antigen binding. 23 Larger fragments can be built from scFv fragments, including minibodies and related small immunoproteins (SIPs), w hich consist of an scFv fused to an immuno globulin constant domain that induces dimerization, such as the human IgG1 C H3 domain. 24,25 A number of laboratories ha ve also explored fusion of scFv fragments to the entire IgG Fc region (C H2 and C H3), which opens the door to more precise tailoring of effector functions and pharmacokinetics (see section “Engineering Fc:FcRn Interactions to Control Antibody Phar macokinetics”). Higher valencies can be achie ved through the production of triabodies and tetrabodies, or tandem diabodies and other variants. Protein engineering also opens the possibility of generating bispecif ic antibodies based on any of these for mats, since variable region genes from antibodies reco gnizing tw o dif ferent tar gets can be readily combined. 26 The molecular w eight of engineered antibody fragments has a major impact on phar macokinetics, since the threshold for f irst-pass renal clearance of proteins is around 60 kDa. 27 As a result, smaller engineered fragments, such as scFv and diabodies, are subject to renal filtration and have relatively short persistence in the circulation. Lar ger fragments are cleared and metabolized through the liver. A number of strategies have been developed to fur ther modify the phar macokinetics of antibody fragments. Conjugation to pol yethylene glycol (PEG) of varying molecular weights is a general method for half-life e xtension that is highl y applicable to small engineered proteins and can be used to enhance targeting and imaging properties. Engineering Fc:FcRn Interactions to Control Antibody Pharmacokinetics

The constant regions of an antibody are typically a focus of interest due to their role in therapeutic interactions— the Fc region is the site of interaction with complement components, Fc receptors on immune ef fector cells, and even key bacterial protein such as protein A. Of interest to imagers is the key role that the Fc por tion of antibodies plays in phar macokinetics, through interaction with the FcRn receptor (neonatal Fc receptor) in endothelium and nor mal tissues. Binding of immuno globulins to Fc receptors results in recycling of antibodies back into the

653

circulation and prolongs ser um half-life. A number of laboratories ha ve studied this interaction and demonstrated that introduction of specif ic mutations at sites that enhance FcRn binding prolong half-life for therapeutic applications, or interfere with binding to accelerate clearance. Using this engineering approach, one can precisely tailor the phar macokinetics of an antibody or fragment to suit an application in molecular imaging. 28 In summar y, using combinations of fragment engineering and modulation of FcRn binding, recombinant fragments suitable for rapid targeting and imaging of an y of a v ariety of cell surf ace tar gets in cancer ha ve been generated (Figure 4, bottom). Fusion to Albumin or Albumin-Binding Domains

An alter native strate gy for increasing the circulating half-life of molecules, including engineered antibody fragments, is to someho w link them to proteins that exhibit ser um persistence, such as ser um albumin. F or example, b y fusing small proteins to albumin-binding domains, the albumin essentially acts as a carrier for the small proteins, e xtending their half-li ves. Two recent examples include production of AB.Fabs and AlbudAbs.29,30 As an alter nate approach, scFv fragments ha ve been fused directl y to albumin to e xtend circulation time.31

Linkage of Target Recognition with Signal Generation Regardless of whether the investigator selects an antibody Fv re gion as the source of a protein-based reco gnition domain, or instead opts for an alter native platfor m (scaffolds based on single antibody domains, f ibronectin domains, ankyrin repeats, Kunitz domains, or any of a variety of alternatives under evaluation), a means of appending a signal-generating moiety must be identif ied. Attachment of a radiolabel (a gamma- or positron-emitting radionuclide) enab les detection using γ cameras, single photon emission computed tomo graphy (SPECT) scanners, or PET scanners. Alternatively, the recognition protein can be linked to fluorescent dyes/proteins or luciferases for detection using optical imaging systems. Engineering for Conjugation

The chemical reacti vity of specif ic amino acid side chains has long been e xploited for protein modif ication.

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A Anti-CEA diabody

B

C 80,000 60,000

RLuc8

40,000

photons/sec/cm2/sr

654

20,000 Figure 5. Optical bioluminescence imaging using an antibody-luciferase reporter fusion protein. A, Model structure of anticarcinoembryonic antigen (CEA) diabody-Renilla luciferase (RLuc8) fusion protein based on the X-ray crystallographic structures of the component proteins.65,66 Antibody variable domains are green, with the complementarity determining regions (CDRs) that define the binding site in white. RLuc8 monomers are shown in blue and the linker peptides are fuschia. B, C, show optical imaging of mice carrying CEA-expressing xenografts in the left shoulder area and target-negative tumors on the right. Mice were injected intravenously with the CEA-specific diabodyRLuc8 fusion protein (B) or RLuc8 alone (C). 4 hours later, coelenterazine (a substrate for Rluc8) was injected via the tail vein, and images acquired with a cooled charge-coupled device (CCD) camera. Antibody-mediated delivery of luciferase protein to the target-positive tumor is evident.37

Amine-specific, carbo xyl-specific, and thiol-specif ic conjugation reagents and protocols are w ell established. Further control o ver the location and stoichiometr y of chemical modif ication can be obtained b y using sitespecific mutagenesis to add chemicall y reacti ve amino acid residues. F or e xample, introduction of c ysteine residues is frequentl y used as a strate gy to foster sitespecific conjugation. C-ter minal c ysteine residues appended to scFv antibody fragments have been used for homodimer formation32 and for attachment of functional groups including enzymes, radioisotopes, fluorescent dyes, and nanopar ticles.33 Similarly, C-ter minal modif ication of diabodies pro vides a small, bi valent antibody fragment for conjugation to functional g roups, with the added advantage that the thiol g roups are protected in a disulfide bond that can be reduced at the time of conjugation.34 In both cases, use of a C-ter minal cysteine for coupling ensures that the chemical linkage occurs on the Fv fragment at a location well removed from the antigencombining site, ensuring that high binding acti vity is retained. This approach has also been used to engineer a cysteine residue at the N-terminus of annexin V, for thiolspecific conjugation to 18F- FB ABM is N-[4-[(4-[ 18F] f luorobenzylidene)aminooxy]butyl]maleimide. Membrane binding af finity w as retained , pro viding a potential new agent for in vivo imaging of apoptosis. 35 Fusion to Optical Signaling Domains

As alluded to abo ve, if the signal-generating moiety is itself a protein, it can be link ed to binding proteins using two general approaches. Chemical conjugation (which can be nonspecif ic or site-specif ic, if the protein par tners are

appropriately engineered) pro vides a uni versal approach to linking tw o protein acti vities, but as noted abo ve, requires optimization to achie ve the desired conjugation efficiency and also risks inacti vation of one or both parties. Alternatively, one can construct and produce a fusion protein. For e xample, diabodies specif ic for carcinoembryonic antigen (CEA) w ere fused to RLuc and Gaussia luciferase (GLuc), in essence producing a “repor ter protein” which can localize to a cell surface target in vivo and produce an optical signal w hen the substrate coelenterazine is administered systemically36,37 (Figure 5).

Engineering to Improve Reporter Genes Reporter genes, and the proteins they encode, must serve a v ariety of needs. An o verarching principle is that a reporter system for in vi vo imaging must be ab le to generate a signif icant detectab le signal, in the absence of background activity. Reporter genes should be otherwise biologically inert since the goal is to mak e a noninvasive measurement without per turbing the host cell, tissue, or organism. Needless to sa y, protein engineering has allowed investigators to adapt enzymes and other proteins to suit reporter gene applications, by modifying substrate specificity, signal w avelength of optical repor ters, and protein stability. Ultimatel y, to transition repor ter gene technology into clinical applications, the de velopment of nonimmunogenic reporter genes will be required. Substrate Specificity

One of the best kno wn examples of engineered substrate specificity in reporter gene applications is the development

Protein Engineering for Molecular Imaging

of the HSV-tk SR39 mutant. SR39tk (bearing f ive amino acid substitutions in the nucleoside binding site) and related mutants w ere originally developed using directed evolution, b y e valuating a focused librar y of tk v ariants containing mutations.38,39 The original goal was to obtain a derivative in w hich the substrate preference w as shifted toward ganc yclovir and ac yclovir, instead of the natural substrate thymidine, improving the utility of tk for suicide gene therapy applications. In parallel, a spectr um of 18Fand 124I-labeled nucleoside analogs have been developed as PET reporter probes for use in conjunction with parental or mutant tks. Extensive evaluation of the in vivo distribution, pharmacokinetics, and phosphor ylation and trapping of radiolabeled analo gs including 1-(2 ʹ′-fluoro-2ʹ′-deoxy-Darabinofuranosyl)-5-methyluracil (FMA U), 2 ʹ′-fluoro-2ʹ′deoxyarabinofuranosyl-5-ethyluracil (FEA U), 2 ʹ′-fluoro2ʹ′ -deoxy-β-D-arabinofuranosyl-5-iodouracil (FIAU), penciclovir (PCV), and others ha ve been conducted in search of optimal combinations for PET repor ter gene imaging.40,41 In addition, identification of a supermutant tk that preferentially phosphor ylates ac ycloguanosine analo gs, with limited appetite for pyrimidine analogs, was produced based on molecular modeling and site-specif ic mutagenesis.42 Use of this variant as a reporter gene allows imaging even in the presence of therapeutic le vels of gancyclovir.43 Color Variants

One of the strong points of optical imaging is the potential for multicolor detection, w hich enab les imaging of multiple tar gets at the same time. The a vailability of multiple small-molecule dy es with different fluorescence emission maxima has accelerated in vitro applications, such as multicolor flow cytometry, chromosome painting, and countless applications in fluorescence microscop y. In an analo gous f ashion, w avelength-shifted v ariants of monomeric red fluorescent protein (mRFP , emission maximum, 607 nm) w ere produced b y randomization of residues near the chromophore, to yield a f amily of fruit (colored) variants with emission maxima from 537 to 610 nm. 44 This molecular e volution ef fort also yielded mutant fluorescent proteins with increased quantum yield, stability, or per missiveness to ward N- or C-ter minal fusions (although not all in the same molecule). In parallel, a set of b lue fluorescent protein v ariants ha ve been produced.45 In an analo gous f ashion, color -shifted v ersions of RLuc have been produced, using a combination of rational (site-directed) mutagenesis, as well as by screening a small library of v ariants containing randoml y introduced mutations. RLuc exhibits high sequence homolo gy to bacterial

655

haloalkane hydrogenases; and alignment to kno wn crystal structures and identif ication of a catal ytic triad motif allowed identif ication of the putati ve acti ve site of the luciferase. Site-directed saturation mutagenesis resulted in the generation of RLuc v ariants with emission maxima at wavelengths up to 50 nm longer than the parental RLuc (482 nm). Fur ther screening of a librar y generated b y error-prone PCR e xtended the w avelength shift e ven further (e g, RLuc 8.6-547). Combinations of the w avelength-shifted RLuc v ariants with coelenterazine analo gs resulted in light emissions that are shifted e ven fur ther compared with the original b lue light of RLuc. Importantly, detailed spectral anal ysis shows that many of these mutant luciferases have significantly increased light output at 600 nm or higher w avelengths, a re gion critical for in vivo applications.46 Stability Variants

Availability of repor ter proteins with a range of stabilities is impor tant to meet the requirements of v arious applications. F or e xample, if the imaging application involves cell tracking in vivo, use of a stable variant will allow buildup and maintenance of the signal o ver time. Stabilized RLuc, or GLuc, w hich is quite robust in its native for m, w ould be prefer red for bioluminescence detection. In contrast, repor ter gene approaches are often used to monitor transient e vents, such as acti vation of gene expression within cells or tissues. To follow the kinetics of acti vation, it is often desirab le to incorporate a reporter protein with high light output but relatively f ast protein tur nover, such that upw ard and downward shifts in transcription/translation of the gene of interest can be follo wed. For example, a destabilized HSV1-tk gene w as created with rapid tur nover for dynamic monitoring of transcription events.47 Use of an overly stab le repor ter protein will obscure an y subtle shifts in expression over time. A special case is the use of luciferases as repor ter proteins. In the e xample of an antibody-luciferase fusion protein described abo ve, a highly stab le luciferase protein w as needed for the fusion protein to maintain activity during the time interval required for in vivo localization and clearance from nontarget tissues (up to 24 h). Native RLuc has an activity half-life of onl y 30 min at 37°C, necessitating the use of the RLuc8 variant (activity half-life > 200 h)48 or the naturally stable GLuc protein 49 for this application. Finally, in addition to biochemical stability , rational design and directed evolution were used to derive orange and RFPs with impro ved bioph ysical proper ties— enhanced photostability.50

656

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Multimodal Reporter Genes

Fusion protein approaches ha ve been used to generate reporter genes that function in a v ariety of imaging modes; for example, fluorescence, bioluminescence, and PET imaging. 51–53 Major adv antages of multimodality imaging repor ter genes include (1) the ability to mo ve quickly from cell culture to small-animal imaging, (2) reduced cost of using optical imaging, w hile retaining the ability to s witch quickly to PET imaging in the same models, and (3) earl y assessment of the potential for clinical translation of a PET imaging system or probe. Nonimmunogenic Reporter Genes

The use of repor ter genes in conjunction with gene therapy or cell-based therapies will pro vide an impor tant window for assessment of these therapies in living organisms, including patients. P otential clinical applications have been focused on the use of PET repor ter genes, which foster trapping of a cor responding PET repor ter probe in cells and tissues e xpressing the gene. The radioactive emissions can then be detected using a PET scanner in preclinical or clinical settings. Among the most widely used PET reporter gene systems under evaluation are the HSV1-tk and related HSV1-SR39tk derivative. F ollowing initial human studies of 9-[4[18F]fluoro-3-(hydroxymethyl)butyl] guanine (18F-FHBG) in normal volunteers,54 Penuelas and colleagues55 demonstrated PET imaging of HSV1-tk gene e xpression using 18 F-FHBG in follo wing intratumoral injection of an adenoviral vector in patients with li ver cancer. However, the use of repor ter genes can be ne gatively impacted b y immunogenicity (if the reporter is recognized as a foreign protein), endo genous e xpression, or unw anted biolo gic activity. In 23 patients treated with HSV1-tk-transduced donor l ymphocytes, se ven de veloped a T-cell response 56 specifically directed against the foreign tk protein. Immune responses against cells e xpressing repor ter genes w ould be par ticularly detrimental if repeated administration and imaging are planned or if long-ter m monitoring of transduced therapeutic cells is necessary. It is also important to ensure that the reporter genes are biologically inert; approaches to minimizing biolo gic activity including introducing site-specif ic point mutations to ablate acti vity, and deleting or substituting functional domains, are readily accomplished using standard protein engineering technology. These concer ns ha ve prompted de velopment of additional repor ter genes of human origin, to reduce the

possibility of an immune response in patients. 57 Numerous laboratories have studied the use of the human sodium iodide sympor ter (hNIS; w hich transpor ts and concentrates iodide, pertechnetate, and other similar anions), 58,59 norepinephrine transporter,60 and human mitochondrial tk type 2 (w hich can phosphor ylate and trap FEA U and FIAU).61 Clinical proof-of-principle was demonstrated in a study treating patients with prostate cancer with an adenoviral vector car rying two suicide genes and hNIS as a reporter gene. 58 Noninvasive SPECT imaging using 99m Tc-pertechnetate was used to demonstrate gene expression in the prostate up to 7 da ys following administration of the adeno virus v ector. Enzyme and transpor ter-based reporter genes ha ve the adv antage that the y can act catalytically and concentrate tracers intracellularl y, against a concentration gradient. An alter native approach is the con version of proteins of human origin into no vel cell surf ace receptors. It w ould be preferab le to select genes that encode proteins that are human in origin, b ut have little or no biologic acti vity, in addition to minimal e xpression in normal tissues. It is also necessar y to achie ve stab le, high-level expression in tar get cells without impacting on cellular function. An example of this strate gy is the reformatting of CEA into a transmembrane-anchored reporter protein, used in conjunction with imaging-optimized anti-CEA antibody fragments as the PET reporter 124 probes 62 (Figure 6A). Imaging using I-labeled chimeric anti-CEA scFv-Fc fragments demonstrated that tumors e xpressing a transfected CEA “mini-gene” were detected by microPET imaging, but only at higher levels of cell surface expression. A reverse approach has been de veloped in w hich cell surf ace e xpression of a high-affinity antihapten antibody has been used as a reporter gene. 63,64 In this e xample, an antibody that binds comple xes of yttrium and the chelating agent 1,4,7,10-tetraazacyclododecane-N, N', N'', N'''tetraacetic acid (DOTA) was reformatted into a chimeric single-chain, membrane-anchored for m, and e xpressed stably on the surf ace tumor cells (F igure 6B). Xenografts were readily detected by microPET imaging using 86Y-DOTA comple xes.64 Use of full y humanized or human antibodies in these strate gies will fur ther reduce potential immuno genicity, and careful selection of spacer re gions (immunoglobulin constant re gions or other human protein domains) and mutation, if needed , can ensure that the reporter protein is biologically inert. A limitation of receptor -based repor ter gene strate gies is that the maximum signal that can be reached will be limited by saturation of binding to the receptors. Ef fective receptor -based repor ter gene strate gies need to

Protein Engineering for Molecular Imaging

A A

B B

657

established biotechnolo gy industr y that has estab lished methodologies and infrastr ucture and has v alidated the utility of producing engineered proteins for in vitro and in vivo applications. As a result, as research on molecular imaging including protein engineering matures, there are clear pathways to broader de velopment and implementation in biomedical research.

ACKNOWLEDGMENTS

Figure 6. Nonimmunogenic reporter gene strategies. In (A) the human protein carcinoembryonic antigen (CEA) (which is expressed during fetal development and in limited tissues in adults) was engineered into a stable membrane protein, which can be detected using a radiolabeled CEA-specific antibody as a probe.62 In (B) the antibody itself (with specificity for yttrium-DOTA) was engineered into a stable membrane protein; detection is accomplished using 86Y-DOTA as the tracer.64

balance the requirement of high le vels of receptor expression with the need to avoid perturbing normal cell function.

SUMMARY Protein engineering offers a powerful, straightforward, and efficient approach for modifying the structure and function of proteins, with broad applications in the field of molecular imaging. Engineering can be used to generate di versity and isolate imaging agents with the desired specif icity, optimize detection, and enhance suitability for clinical use. Engineering proteins for use as repor ter genes is w ell established, the major remaining concer ns being gene delivery, potential immunogenicity, and strength of the signal that can be produced. Production of recombinant proteins is not without its challenges. Predicting a protein’ s folding and function from its primary sequence remains an incomplete science. Fur thermore, synthesis of recombinant proteins still relies largely on cells (microbial or mammalian) for production. In vitro transcription/translation systems are broadl y a vailable but f all shor t in ter ms of scalability. Countering these potential hurdles is a highl y

The author is grateful to past and present members of her laboratory, as w ell as her man y collaborators at UCLA, the City of Hope, and other institutions. The author also thanks James Strommer and Drs. Tove Olafsen, Vania Kenanova, and Andy Loening for their assistance with the figures. Research suppor t has been pro vided b y NIH grants CA086306 (UCLA ICMIC), CA092131 (UCLA SPORE in Prostate Cancer), CA098010 (UCLA Scholars in Oncologic Molecular Imaging), and CA119367 (Stanford Center for Cancer Nanotechnology Excellence).

REFERENCES 1. Marks JD, Hoo genboom HR, Bonner t TP, et al. By-passing immunization: Human antibodies from V-gene libraries displa yed on phage. J Mol Biol 1991;222:581–97. 2. Vaughan TJ, Williams AJ, Pritchard K, et al. Human antibodies with sub-nanomolar af finities isloated from a lar ge non-immunized phage display library. Nat Biotechnol 1996;14:309–14. 3. Link AJ, Mock ML, Tirrell DA. Non-canonical amino acids in protein engineering. Curr Opin Biotechnol 2003;14:603–9. 4. Hoogenboom HR. Selecting and screening recombinant antibody libraries. Nat Biotechnol 2005;23:1105–16. 5. Nuttall SD , Walsh RB . Displa y scaf folds: protein engineering for novel therapeutics. Curr Opin Pharmacol 2008;8:609–15. 6. Tijink BM, Laeremans T, Budde M, et al. Improved tumor targeting of anti-epidermal growth factor receptor Nanobodies through albumin binding: taking adv antage of modular Nanobody technolo gy. Mol Cancer Ther 2008;7:2288–97. 7. Holt LJ, Herring C, Jespers LS, et al. Domain antibodies: proteins for therapy. Trends Biotechnol 2003;21:484–90. 8. Orlova A, Magnusson M, Eriksson TL, et al. Tumor imaging using a picomolar affinity HER2 binding affibody molecule. Cancer Res 2006;66:4339–48. 9. Silverman J , Liu Q, Bakk er A, et al. Multi valent a vimer proteins evolved by exon shuffling of a family of human receptor domains. Nat Biotechnol 2005;23:1556–61. 10. Binz HK, Amstutz P, K ohl A, et al. High-af finity binders selected from designed ank yrin repeat protein libraries. Nat Biotechnol 2004;22:575–82. 11. Kolmar H. Alternative binding proteins: biological activity and therapeutic potential of c ystine-knot miniproteins. FEBS J 2008;275: 2684–90. 12. Ngoi SM, Chien AC, Lee CG. Exploiting internal ribosome entry sites in gene therapy vector design. Curr Gene Ther 2004;4:15–31. 13. de Felipe P, Luke GA, Hughes LE, et al. Eunum pluribus: multiple proteins from a self-processing pol yprotein. Trends Biotechnol 2006;24:68–75.

658

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

14. Bhaumik S, Walls Z, Puttaraju M, et al. Molecular imaging of gene expression in li ving subjects b y spliceosome-mediated RN A trans-splicing. Proc Natl Acad Sci U S A 2004;101:8693–8. 15. Muralidharan V, Muir TW. Protein ligation: an enabling technology for the bioph ysical anal ysis of proteins. Nat Methods 2006;3:429–38. 16. Dai M, Temirov J, Pesavento E, et al. Using T7 phage display to select GFP-based binders. Protein Eng Des Sel 2008;21:413–24. 17. De A, Gambhir SS. Nonin vasive imaging of protein-protein interactions from li ve cells and li ving subjects using bioluminescence resonance energy transfer. FASEB J 2005;19:2017–9. 18. Villalobos V, Naik S, Piwnica-Worms D. Current state of imaging protein-protein interactions in vi vo with geneticall y encoded reporters. Annu Rev Biomed Eng 2007;9:321–49. 19. Goldenberg DM, DeLand F , Kim E, et al. Use of radiolabeled antibodies to carcinoembr yonic antigen for the detection and localization of di verse cancers b y e xternal photoscanning. N Engl J Med 1978;298:1384–8. 20. Boswell CA, Brechbiel MW. Development of radioimmunotherapeutic and diagnostic antibodies: an inside-out vie w. Nucl Med Biol 2007;34:757–78. 21. Lonberg N. Human antibodies from transgenic animals. Nat Biotechnol 2005;23:1117–25. 22. Huston JS, Levinson D, Mudgett-Hunter M, et al. Protein engineering of antibody binding sites: Recovery of specific activity in an antidigoxin single-chain Fv analo gue produced in Escherichia coli. Proc Natl Acad Sci U S A 1988;85:5879–83. 23. Holliger P, Winter G. Diabodies: small bispecif ic antibody fragments. Cancer Immunol Immunother 1997;45:128–30. 24. Hu S, Shi vely L, Raubitschek AA, et al. Minibody: A no vel engineered anti-CEA antibody fragment (single-chain Fv-CH3) which exhibits rapid , high-le vel tar geting of x enografts. Cancer Res 1996;56:3055–61. 25. Borsi L, Balza E, Bestagno M, et al. Selecti ve targeting of tumoral vasculature: comparison of different formats of an antibody (L19) to the ED-B domain of fibronectin. Int J Cancer 2002;102:75–85. 26. Muller D , K ontermann RE. Recombinant bispecif ic antibodies for cellular cancer immunotherap y. Cur r Opin Mol Ther 2007;9: 319–26. 27. Wu AM, Senter PD. Arming antibodies: prospects and challenges for immunoconjugates. Nat Biotechnol 2005;23:1137–46. 28. Kenanova V, Wu AM. Tailoring antibodies for radionuclide deli very. Expert Opin Drug Deliv 2006;3:53–70. 29. Dennis MS, Jin H, Dugger D, et al. Imaging tumors with an albuminbinding Fab, a novel tumor-targeting agent. Cancer Res 2007;67: 254–61. 30. Holt LJ, Basran A, Jones K, et al. Anti-serum albumin domain antibodies for e xtending the half-li ves of shor t lived dr ugs. Protein Eng Des Sel 2008;21:283–8. 31. Yazaki PJ , Kassa T, Cheung CW , et al. Biodistribution and tumor imaging of an anti-CEA single-chain antibody-albumin fusion protein. Nucl Med Biol 2008;35:151–8. 32. Adams GP, McCar tney JE, Tai M-S, et al. Highl y specif ic in vi vo tumor targeting by monovalent and divalent forms of 741F8 antic-erbB-2 single-chain Fv. Cancer Res 1993;53:4026–34. 33. Albrecht H, Burke PA, Natarajan A, et al. Production of soluble ScFvs with C-ter minal-free thiol for site-specif ic conjugation or stab le dimeric ScFvs on demand. Bioconjug Chem 2004;15:16–26. 34. Olafsen T, Cheung CW, Yazaki PJ, et al. Covalent disulfide-linked antiCEA diabody allows site-specific conjugation and radiolabeling for tumor targeting applications. Protein Eng Des Sel 2004;17:21–7. 35. Li X, Link JM, Stekho va S, et al. Site-specif ic labeling of anne xin V with F-18 for apoptosis imaging. Bioconjug Chem 2008;19:1684–8. 36. Venisnik KM, Olafsen T, Gambhir SS, Wu AM. Fusion of Gaussia luciferase to an engineered anti-carcinoembryonic antigen (CEA)

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51. 52.

53.

54.

antibody for in vi vo optical imaging. Mol Imaging Biol 2007;9:267–77. Venisnik KM, Olafsen T, Loening AM, et al. Bifunctional antibodyRenilla luciferase fusion protein for in vi vo optical detection of tumors. Protein Eng Des Sel 2006;19:453–60. Gambhir SS, Bauer E, Black ME, et al.A mutant herpes simplex virus type 1 thymidine kinase reporter gene shows improved sensitivity for imaging repor ter gene e xpression with positron emission tomography. PG—2785-90. Proc Natl Acad Sci U S A 2000;97: 2785–90. Kokoris MS, Black ME. Characterization of her pes simple x vir us type 1 thymidine kinase mutants engineered for impro ved ganciclovir or acyclovir activity. Protein Sci 2002;11:2267–72. Kang KW, Min JJ, Chen X, Gambhir SS. Comparison of [14C]FMA U, [3H]FEAU, [14C]FIAU, and [3H]PCV for monitoring reporter gene expression of wild type and mutant her pes simple x vir us type 1 thymidine kinase in cell culture. Mol Imaging Biol 2005;7:296–303. Miyagawa T, Gogiberidze G, Ser ganova I, et al. Imaging of HSV -tk Reporter gene e xpression: comparison betw een [18F]FEA U, [18F]FFEAU, and other imaging probes. J Nucl Med 2008; 49:637–48. Degreve B, Esnouf R, De Clercq E, Balzarini J. Selective abolishment of pyrimidine nucleoside kinase acti vity of her pes simplex vir us type 1 th ymidine kinase b y mutation of alanine-167 to tyrosine. Mol Pharmacol 2000;58:1326–32. Likar Y, Dobrenkov K, Olszewska M, et al. A new acycloguanosinespecific super mutant of her pes simple x vir us type 1 th ymidine kinase suitab le for PET imaging and suicide gene therap y for potential use in patients treated with p yrimidine-based cytotoxic drugs. J Nucl Med 2008;49:713–20. Shaner NC, Campbell RE, Steinbach P A, et al. Impro ved monomeric red, orange and y ellow fluorescent proteins deri ved from Discosoma sp. red fluorescent protein. Nat Biotechnol 2004;22:1567–72. Ai HW, Shaner NC, Cheng Z, et al. Exploration of new chromophore structures leads to the identif ication of improved blue fluorescent proteins. Biochemistry 2007;46:5904–10. Loening AM, Wu AM, Gambhir SS. Red-shifted Renilla renifor mis luciferase v ariants for imaging in li ving subjects. Nat Methods 2007;4:641–3. Hsieh CH, Chen FD, Wang HE, et al. Generation of destabilized herpes simple x vir us type 1 th ymidine kinase as transcription reporter for PET reporter systems in molecular genetic imaging. J Nucl Med 2008;49:142–50. Loening AM, F enn TD, Wu AM, Gambhir SS. Consensus guided mutagenesis of Renilla luciferase yields enhanced stability and light output. Protein Eng Des Sel 2006;19:391–400. Tannous BA, Kim DE, Fernandez JL, et al. Codon-optimized Gaussia luciferase cDNA for mammalian gene expression in culture and in vivo. Mol Ther 2005;11:435–43. Shaner NC, Lin MZ, McK eown MR, et al. Impro ving the photostability of bright monomeric orange and red fluorescent proteins. Nat Methods 2008;5:545–51. Ray P, De A, Min JJ, et al. Imaging tri-fusion multimodality repor ter gene expression in living subjects. Cancer Res 2004;64:1323–30. Ponomarev V, Doubrovin M, Serganova I, et al. A novel triple-modality reporter gene for w hole-body fluorescent, bioluminescent, and nuclear noninvasive imaging. Eur J Nucl Med Mol Imaging 2004; 31:740–51. Ray P , Tsien R, Gambhir SS. Constr uction and v alidation of improved triple fusion repor ter gene v ectors for molecular imaging of living subjects. Cancer Res 2007;67:3085–93. Yaghoubi S, Bar rio JR, Dahlbom M, et al. Human phar macokinetic and dosimetry studies of [(18)F]FHBG: a reporter probe for imaging her pes simple x vir us type-1 th ymidine kinase repor ter gene expression.PG - 1225-34. J Nucl Med 2001;42:1225–34.

Protein Engineering for Molecular Imaging

55. Penuelas I, Mazzolini G, Boan JF , et al. P ositron emission tomo graphy imaging of adenoviral-mediated transgene expression in liver cancer patients. Gastroenterology 2005;128:1787–95. 56. Traversari C, Marktel S, Magnani Z, et al. The potential immunogenicity of the TK suicide gene does not prevent full clinical benefit associated with the use of TK-transduced donor lymphocytes in HSCT for hematologic malignancies. Blood 2007;109:4708–15. 57. Serganova I, Ponomarev V, Blasberg R. Human reporter genes: potential use in clinical studies. Nucl Med Biol 2007;34:791–807. 58. Barton KN, Stricker H, Brown SL, et al. Phase I study of noninvasive imaging of adeno virus-mediated gene e xpression in the human prostate. Mol Ther 2008;16:1761–9. 59. Park SY, Kwak W, Thapa N, et al. Combination therapy and noninvasive imaging with a dual therapeutic v ector e xpressing MDR1 short hair pin RNA and a sodium iodide sympor ter. J Nucl Med 2008;49:1480–8. 60. Doubrovin MM, Doubrovina ES, Zanzonico P, et al. In vi vo imaging and quantitation of adoptively transferred human antigen-specific T cells transduced to express a human norepinephrine transporter gene. Cancer Res 2007;67:11959–69.

659

61. Ponomarev V, Doubro vin M, Sha vrin A, et al. A human-deri ved reporter gene for nonin vasive imaging in humans: mitochondrial thymidine kinase type 2. J Nucl Med 2007;48:819–26. 62. Kenanova V, Barat B, Olafsen T, et al. Recombinant carcinoembryonic antigen as a reporter gene for molecular imaging. Eur J Nucl Med Mol Imaging 2008;36:104–14. 63. Roffler SR, Wang HE, Yu HM, et al. A membrane antibody receptor for nonin vasive imaging of gene e xpression. Gene Ther 2006;13:412–20. 64. Wei LH, Olafsen T, Radu C, et al. Engineered antibody fragments with inf inite affinity as repor ter genes for PET imaging. J Nucl Med 2008;49:1828–35. 65. Carmichael JA, Power BE, Gar rett TP, et al. The crystal structure of an anti-CEA scFv diabody assembled from T84.66 scFvs in V(L)to-V(H) orientation: implications for diabody fle xibility. J Mol Biol 2003;326:341–51. 66. Loening AM, F enn TD, Gambhir SS. Cr ystal str uctures of the luciferase and g reen fluorescent protein from Renilla renifor mis. J Mol Biol 2007;374:1017–28.

41 PHAGE DISPLAY FOR IMAGING AGENT DEVELOPMENT KIMBERLY A. KELLY, PHD

Early work to develop targeted diagnostic and therapeutic molecules has focused on the use of antibodies for tumor recognition and dr ug deli very.1,2 However, antibodies produced in traditional methods require tar get antigens to be immunogenic and can cause adverse effects in patients. They can also be costly to produce, and there is difficulty associated with producing an acti ve conjugated for m.3 Small molecule screening efforts, which work well against concave surfaces, can be difficult to design against e xtracellular protein–protein interactions that often in volve large flat contact surf aces.4 Small peptides, w hich ha ve also been used for tar geting, are typicall y nonimmunogenic, can adopt confor mations that compliment such extended surfaces and combine high affinity and selectivity with g reater tolerance for modif ication (see Chapters 25, “Hyperpolarized 13C Magnetic Resonance Imaging—Principles and Applications” and 34, “Magnetic Nanopar ticles”). Traditional methods in volving the rational design of peptides are predicated on the knowledge of the str ucture of the tar get with one peptide taking years of iterations to perfect. 5 To circumvent these issues and to aid in the de velopment of tar geted imaging agents, combinatorial chemistry–based approaches have been used. Phage display uses a population of bacteriophages genetically modified to display a librar y on v arious phage coat proteins. Phage display of fers a number of impor tant adv antages such as rapid and economical biolo gical e xpansion (rather than time consuming chemical re-synthesis), co vering a lar ge area of di versity space, a rapid screening process, and the availability of man y types of phage clones and libraries such as peptide, cDN A, and antibody (for re view6–11). Libraries with diversities as high as 10 10 are routinely constructed and are also commerciall y available. Indeed, billions of clones can be screened within a w eek using phage 660

display. The screening process is highly flexible with selection limited onl y b y the imagination of the in vestigator. Finally, the startup costs for a phage display experiment are relatively low because it requires no special equipment and the libraries are self-replicative. Another important advantage is that bacteriophages, unlike higher or ganisms, ha ve onl y one cop y of each gene, so protein expression is not dependent on the interaction of multiple genes. Each gene leads to one protein and each protein has one gene, that is, genotype equals phenotype. A clone isolated based on phenotypic properties is easy to identify by sequencing the appropriate portion of the phage genome. The genotype equals phenotype phenomenon enab les screening in a single well, thereb y reducing the amount of star ting material (proteins, cells, tissue, etc) needed for the screen and also allowing the competition of displa yed entities against each other. Phage display technology has had a major impact on immunology,12 cell biology,13 drug discovery,14 and pharmacology and is increasingly gaining importance in molecular imaging. 3,11,15 It has been used to select for specific antibodies and to humanize and af finity mature conventional murine antibodies, and has also been used for the identif ication of disease specif ic peptides. The aim of this chapter is to pro vide a practical overview on the principles and applications of phage displa y in the context of identifying lead compounds for imaging agents.

PHAGE LIBRARIES Since it was f irst described by George Smith in 1985, 9 there ha ve been o ver 3,700 pub lications in PubMed using phage displa y. The f ilamentous phage Ff, of

Phage Display for Imaging Agent Development

661

Figure 1. Schematic representation of M13 bacteriophage. pVIII is the major coat protein with ~2,700 copies. pIII and pIV are present in four to five copies on one end of the phage, whereas pVII and pV are present on the opposite end. Right—TEM of M13 bacteriophage courtesy of Debadyuti Ghosh and Angela Belcher, Departments of Material Science and Energy and Biological Engineering, Massachusetts Institute of Technology, Cambridge MA 02421.

which M13 is a f amily member, is the most commonl y used platform for display. Filamentous phages are flexible rods about 6 nm in diameter with the length v arying based on the size of the DN A plasmid inser ted (Figure 1). 16 Roughly, the length is around 1 µm with the shor test being a “microphage” v ariant of 50 nm long.17 Cloning into the nonessential re gions produces longer v ariants; ho wever, the longer the phage, the more fragile and subject to breakage the y become. Phage particles are composed of five coat proteins with the major component being pVIII (2,700 copies; 50 amino acids) and minor components being located on the ends (F igure 2). 18 The b lunt end (b y electron microscopy) has pVII and pIX (3 to 5 copies; 33 and 32 amino acids). The pointed end is composed of pIII and pVI (3 to 5 copies; 406 and 112 amino acids). All f ive coat proteins have been used in phage displa y systems although pIII and to a lesser e xtent pVIII enjo y preferred status. The most commonl y used libraries are random peptide sequences, which can be found in great diversity containing linear , disulf ide-constrained, or v arious peptide lengths from 7 amino acids up to 15 amino acids. In these libraries, the DN A inser ts are deri ved from de generate oligonucleotides, w hich are synthesized b y adding mixtures of nucleotides to a g rowing nucleotide chain. 10 In the absence of detailed str uctural infor mation about the protein being tar geted for imaging, it is dif ficult to

predict w hich phage displa y librar y will be the most useful and productive. The ease and rapidness of screening facilitates the use of multiple libraries in the case that the screen does not yield a satisfactory outcome. The random seven amino acid peptide librar y contains o ver 10 9 random sequences, and the resultant peptide sequences are easy and cost ef fective to synthesize and also easy to chemically modify into tar geted imaging agents. Combined with the librar y being commercially available (Table 1), it has been one of our prefer red choices of library for imaging agent development. If, however, after completing the selection process, there are no binders with appropriate characteristics, other libraries have been used with some degree of success. Disulf ide-constrained peptides bind more tightl y than the same sequence in a linear librar y because of impro ved entrop y of binding; however, there is a g reater chance of f inding no binders because of the imposed str uctural constraint.19,20 In addition, conjugation chemistries in volve the use of thiols, making the disulfide-constrained peptides difficult to use and expensive to produce. The peptide libraries containing larger than seven amino acids, that is, 12 and 15 mer, are useful because the y allow the adoption of str uctural conformations that might be needed for target interaction. However, because of the length of the peptides and the limitations of librar y synthesis, the lar ger phage displa y libraries ha ve generall y less sequence di versity than the 7-mer libraries and are more costl y to synthesize.

662

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

promoter peptide

pVIII

pIll or pVIII

pIII

A

promoter

peptide

pIll or pVIII

phagemid

promoter

pVIII

pIll or pVIII

pIII

helper phage

B Figure 2. Types of phage display. A, Multivalent phage display on pIII or pVIII phage coat proteins. B, Monovalent phage display through the use of a phagemid vector.

Table 1. SOME COMMERCIALLY AVAILABLE SOURCES OF PHAGE DISPLAY VECTORS AND LIBRARIES Manufacturer

Product Name

Vector

Reagents

Website

New England Biolabs

PhD™

M13KE pIII phage vector

Premade peptide libraries with control target and eluant (streptavidin/biotin); M13KE phage vector for library synthesis

www.neb.com

Novagen

T7 Select

T7 select vectors for cDNA display libraries

Premade cDNA libraries, kits for synthesis of cDNA libraries

Spring Bioscience

Phage display cDNA screening kit

pHD9

Premade cDNA libraries

www.springbio.com

In addition to the length and constraint, an important consideration when choosing a library is the number of peptides being displa yed, as peptides can be displa yed monovalently on pIII or multi valently on pIII or pVIII coat proteins. Whether to use mono valent or multi valent libraries is dependent on the end use of the peptide. Typically, multi valent uses generate the highest apparent

affinity agents because of their a vidity ef fects, and as such, are the main choice for the development of imaging agents.21 However, if monovalent peptides are going to be the endpoint agent, then a monovalent library would produce optimum results. Peptide libraries in different forms (ie, linear, disulfide, 7 and 12 amino acid lengths) are commerciall y

Phage Display for Imaging Agent Development

available and represent the easiest out-of-the-bo x method for high-throughput screening (see Table 1). A close second in usage are phage libraries w here the phage has been engineered to displa y antibody fragments from the ser um of diseased patients. 22–24 Less widely used, especially as lead compounds for imaging agents, are phages that displa y w hole proteins. These libraries are comple x to constr uct, and the synthesis of the subsequent agent apar t from the phage is costl y, time-consuming, and proteins are generall y not amenable as imaging agents.

pVIII The pVIII protein is 50 amino acids in length and is the major protein comprising the phage coat. There are ~2,700 pVIII proteins in the coat, oriented at a 20° angle for ming a right-handed helix. Displa yed sequences are fused into the amino ter minus between the signal sequence and the start of the mature protein-coding sequence (see Figure 2). Because larger size of molecules restricts the proper packaging of the phage, onl y shor t peptide sequences, betw een six and eight amino acid residues, can be displa yed on all 2,700 pVIII proteins. Therefore, displa y of lar ger peptides on pVIII requires supplementation with wild-type pVIII particles to produce mosaic phage. 25–27 This system has been used to de velop monovalent display and will be discussed in the mono valent library section.

pIII Similar to pVIII, the foreign DN A to encode the foreign peptide must be spliced betw een the signal sequence and the amino terminus of the mature protein required for phage viability (see F igure 2). pIII, ho wever, is present in only three to five copies on the blunt end of the phage and is responsib le for the proper infection of Escherichia coli (E. coli) by attaching to the F pilus on the bacterium. 16 Historically, pIII is the protein of choice for phage displa y e xperiments because it can accommodate lar ger inser tions and there are commerciall y available libraries and v ectors that reduce the star tup time for a phage displa y experiment. Also, like pVIII, it is compatib le with monovalent displa y systems. Although it can tolerate lar ger insertions (up to 15 amino acids), it is dif ficult to express proteins on all three to f ive copies because infectivity of the phage is substantiall y reduced. This limitation can be o vercome b y producing mosaic phage similar to what was described for pVIII.

663

Monovalent Display Typically, phage vectors are constructed so that there is one copy of the gene with the spliced inser t, so in the phage, all the copies of the protein produced ha ve the foreign peptide displayed. This can produce limitations to the size of the displa yed peptide. To overcome these limitations, se veral v ectors ha ve been produced to allow the production of mosaic phage. In addition, these systems have been exploited to produce a monovalent peptide displa y system, w hich is used to select for peptides with high af finity for the tar get. In a type 88 v ector system (created b y Geor ge Smith, detailed extensively on his w ebsite: , and reviewed in 10), the phage genome has two copies of gene VIII, which encodes the wild-type protein and the fusion protein. The result is a longer phage but one that retains viability because it is comprised of both wild-type and recombinant pVIII coat proteins. The 33 v ector system is identical to the 88 system e xcept that it is the pIII protein that is present in two copies. Dr. Smith has also produced a type 8 + 8 vector system and library in which the VIII genes are on two separate genomes. The wild-type version is in the genome of a helper phage that has a compromised origin leading to inef ficient packaging of the virus (Figure 2B). The recombinant v ersion of VIII is cloned under a w eak promoter into a phagemid , which is a plasmid containing only a plasmid origin of replication but also a phage origin to allo w production of a single-stranded vector and to allo w efficient packaging of virus once the bacteria is infected with helper phage. The E. coli carrying the phagemid is infected with the helper phage resulting in the production and secretion of phage that contains both wild-type pVIII and recombinant pVIII. The genotype equals phenotype link is preserved because the helper phage genome is poorl y packaged, and therefore nearly all the phages contain the phagemid vector. Again, 3 + 3 systems (F igure 2B) are constructed the same way as the 8 + 8 with the difference being the 3 + 3 systems are displa yed on pIII instead of pVIII. In these systems, not all the phages will have a peptide displayed, and in f act, most will be completel y wild-type with around 10% being mono valent, and a small percentage having more than one copy.6

Other Libraries M13 displa yed cDN A libraries can be used to identify natural binding par tners for receptors or peptides or

664

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

domains used as imaging agents. Although the libraries have been displa yed with dif ferent for mats, no single format is uni versally applicab le probab ly because of the fact that onl y a subset of eukar yotic proteins can be expressed in E. coli .28 The most successful system for expressing cDN A has been a dif ferent vir us alto gether, the T7 virus, and the a vailability of libraries and v ectors for this system mak es this vir us the choice of in vestigators pursuing a natural protein project (see Table 1). Because of their high affinity and specificity for their targets, antibodies ha ve been the subject of intensi ve research with several being clinically approved therapeutics.29,30 Fab (fragments that contain the variable and first constant region of the hea vy (H) chain and the light (L) chain) and Fv (fragments that contain the variable regions of the H and L chain), can be cloned into phage displa y vector cassettes and are functionall y e xpressed and assembled in E. coli . As a result, phage libraries displaying antibody fragments ha ve become practical tools for drug discovery. Compared to peptide screens, phage antibody screens are more difficult because libraries are not commercially a vailable and must be synthesized. Most libraries have been constructed from the natural immune repertoire either b y immunizing an animal against a known antigen or using sera from patients with disease.31 As a result, antibodies ha ve been generated that bind proteins,32,33 nucleic acids, 34,35 and carboh ydrates36,37 important in cancer and autoimmune diseases. This approach has the limitation that a ne w librar y must be constructed before each screening e xperiment. To circumvent this limitation, libraries ha ve been constructed that are more di verse, uni versal, and unbiased using unimmunized animals w hose antibody reper toire is dominated by IgMs with a specif icity for a v ariety of antigens.38

SCREENING Once a phage displa y librar y has been acquired or constructed, the task is to screen the librar y to produce clones ha ving the highest af finity for the target (Figure 3). Phage displa y clones ha ving the highest affinity for the tar get are enriched via stringent washing conditions after multiple rounds of selection. All screening procedures share the same characteristics and involve (1) selection or “biopanning,” (2) washing and elution, and (3) amplif ication. These steps constitute a round of selection and are repeated betw een three and six times (see F igure 3). Additional Web-based resources are listed in Table 2.

Figure 3.

Flow diagram of a typical phage display experiment.

Selection Wild-type phages have no known tropism for mammalian cells and cannot breach the plasma membrane; therefore, cell surface receptors are the primar y targets when contemplating a screen for in vi vo imaging agents. Traditional in vitro screens commonl y use purif ied tar get proteins immobilized on plates, 39 columns, and beads. 21 Target proteins, carboh ydrates, or inor ganic molecules can be adsorbed directly (ie, to specially coated immunoplates or nitrocellulose membranes) or co valently attached to solid suppor ts. Although this process is the most straightforw ard, the orientation of the tar get molecule on the surf ace is random and not all molecules will be in the cor rect orientation for phage binding. An alternative is to indirectly attach the target to the suppor t by coupling it to biotin or producing a recombinant

Phage Display for Imaging Agent Development

protein with a FLAG, poly-histidine, or Fc tag. Either the labeled tar get is captured on the suppor t with the compatible binding partner such as streptavidin for biotin, and then the phage incubated with the captured tar get or the phage is incubated with tar get, and then the comple x incubated with the modif ied support. A number of modif ications to traditional phage screens ha ve been de veloped o ver the last decade. Screens ha ve been perfor med on li ve cells with tar gets expressed in their nati ve en vironments, w hich has allowed the de velopment of modif ied screens that bias towards cell-inter nalized phage, 40,41 binding under flo w conditions40,41 or other biolo gical processes. This approach ensures that the selected peptide sequence binds in the context of other biomolecules. It also allo ws selection against membrane proteins that are dif ficult to express and purify because of their ph ysical characteristics. Additionally, screening on cells allo ws the selection of peptides that bind to inter nalizing cell surf ace receptors. Receptor-mediated internalization has been a mechanism to increase signal to noise.40,41 The imaging moiety conjugated to the inter nalizing sequence is sequestered and concentrated into the endosomes. This mechanism also increases the apparent af finity of the agent with the koff being on the order of days with degradation being the route of elimination. Ho wever, the selection process can be more difficult because of nonspecif ic binding. One of the most e xciting recent de velopments has been the use of in vi vo phage displa y to yield disease- or organ-specific phage clones. 42 For example, a number of atherosclerosis-targeted phages43 have recently been developed and endothelial bed specif ic clones have been found in both mice and humans. 42,44 Likewise, in vivo screening for tumor vascular targets has also produced peptides specific for tumor-associated endothelial cells.45 In addition to vascular targets, researchers have extended the technology to include in vi vo screening for tumor epithelial cells and have found specif ic peptides for cancers such as prostate. 7 Irrespective of the selection method used, stringency versus yield of binders is an impor tant consideration (a thorough treatise on the topic can be found at , , and re viewed in 10). The stringency of an af finity selection is counter to the yield of binders. In order to recover all interesting clones, yield is extremely impor tant in the f irst round of selection. In a typical phage display experiment, the input phage library has ~100 copies of each clone. If the yield is low, there is a chance that interesting clones will be lost amidst the ever present background nonspecific binders.

665

Stringency is controlled by the choice of binding and elution conditions (detailed in the elution section). F or example, the presence of deter gent in the binding buf fer has been sho wn to reduce nonspecif ic interactions with lower detergent amounts resulting in higher eluate titers but potentiall y more nonspecif ic binders, w hereas high detergent concentrations could result in fe w binders. Likewise, temperature can be v aried depending on whether the interaction is enthalpicall y or entropicall y driven. Stringenc y can be increased b y shor tening the binding time because as KD = Koff/Kon peptides with rapid on rates will be f avored. In later rounds of selection, the target concentration can be lo wered to select for higher affinity peptides although this strate gy works best with purified target and to some e xtent cells b y reducing the overall number of cells that are plated. F inally, the number of rounds can be v aried with selections consisting of an ywhere from three to six rounds. With each round of selection then subsequent amplif ication, the pool of phage clones becomes enriched for peptides that bind to the tar get. Ho wever, care must be tak en w hen increasing the number of rounds because the phenomenon of o verselection ma y occur .6 In this case, peptides with the desired proper ties, that is, high af finity and specificity, have been lost in f avor of clones that have an amplification bias or better display properties. Another technique to increase specif icity and selectivity for the target is to use subtraction or negative selection. In this procedure, phage clones either from the initial phage librar y or after the f irst round of selection are incubated with a moiety closel y related to the tar get. For example, if cancer cell–binding peptides are desired , the phage pool would be subtracted of ubiquitous binders by incubating with normal cells or tissue. One of the benefits of in vi vo selection is that subtraction naturall y occurs with the phage pool being e xposed to man y millions of different types of cells in their native environment. When and ho w man y rounds of subtraction to perform is another area to influence tar get hits. A round of subtraction is incubating with one w ell of tar get. Amplification is not car ried out between rounds of subtraction because amplif ication increases the number of phage by 106 to 107 making it more difficult to subtract clones. Subtracting before the first round of selection has been used successfull y; ho wever, the chances of an incomplete subtraction are high because billions of phage clones are being screened and not all may be able to bind because of steric effects. In addition, nonspecific binding of phage clones to the tar get ma y remo ve clones with optimal properties. To increase the efficiency of subtraction, subtraction steps have been carried out after the first

666

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

round of selection but before amplif ication. Ho wever, optimal clones may be lost because of inefficient binding in the selection step then nonspecif ic binding in the subtraction step. In our hands, three rounds of subtraction have typically been sufficient to remove nonspecific and nonselective clones with additional rounds increasing the chances of losing clones because of nonspecif ic binding of the phage coat proteins. 46 In addition, we typically do not perfor m a second subtraction step betw een rounds two and three for reasons detailed abo ve, namel y that amplification of the phage pool makes it more difficult to effectively subtract.

Washing and Elution As in selection, there are man y variations that have been used in the w ashing and elution step to produce phage clones with desired af finities and specificities. The purpose of the washing step is to remo ve nonspecif ic phage clones, while also increasing the stringenc y of the selection. Typically, selection wells are washed with the same buffer as the selection process but with the addition of a low percentage of deter gent (ie, 0.1% Tween-20). Wash steps, between 3 and 10 for each phage selection round , can be quick (seconds) or long (5 to 30 minutes) and are dependent on the tar get. If w ashing is too stringent, the overall number of phage clones is reduced, and the population ma y be dominated b y high-af finity binders with low specif icity. On the contrar y, if the stringenc y is too low, then the population ma y ha ve a g reater range of binding specificities and affinities in addition to nonspecific binders that must all be tested to f ind the appropriate clone. Elution is another area where investigators have been extraordinarily creative. A wide variety of elution conditions ha ve been used e xploiting M13’ s stability to extremes of pH, ionic strength, denaturants, and e ven most proteases. 6 Nonspecific elution conditions are intended to weaken receptor–peptide interactions without regard to their specif icity.10 Protocols commonl y used include pH e xtremes (either acidic or Alkaline), high molar salt concentrations, denaturing conditions (ie, comurea), proteases, and reducing agents (disulf ideconstrained libraries.) In choosing an elution condition, it is impor tant to remember that KD = Koff/Kon and higher affinity clones can be selected by lengthening the elution time. Specif ic elution assumes prior kno wledge of the receptor that the peptide is binding. Antibodies, purif ied proteins, and inorganic materials, such as EGTA for calcium-specific receptor binding, can g reatly increase the specificity of the elution step. Ho wever, selecting

peptides with high af finity may be dif ficult because the ligand must be able to elute the phage clone from the target and the natural ligand–receptor interaction ma y have a lo w af finity. It is not al ways necessar y to elute the phage from the tar get. E. coli can be applied directl y to the protein, cell, or even excised tissue to recover the captured phage and permit amplification.

Amplification The ne xt step of the process is to amplify the eluted phage clones before the ne xt round of selection to increase the number of phages bearing the same clone. The initial input in the first round of selection consists of all clones in the librar y, and each clone is onl y represented b y a fe w par ticles. If onl y 1 to 10% of a given clone is reco vered, then that clone has a good chance of being lost. To pre vent this, the entire eluate from the f irst round is amplif ied and used as the input for the second round. Amplification results in each clone being represented b y millions of phage, and no clones will be lost in the subsequent rounds. There are some caveats to amplif ication, namel y that some peptide sequences ha ve an amplif ication, or displa y bias and may be underrepresented. F or e xample, ar ginine interferes with pIII secretion; hence, peptides containing arginine are selected against. 47 Cysteines, except where expressed for disulf ide constraints, are rarel y obser ved in peptides presumab ly because of inappropriate disulfide bond formation interfering with proper display. Reports ha ve been pub lished using eluted phage without amplif ication for the input of the ne xt round of selection.48 This strate gy may reduce backg round problems and help reduce the number of nonspecif ic phages that are car ried through the selection process. Ho wever, because the o verall yield of clone binding e ven for the highest affinity binders rarely approaches 100%, the risk of losing v aluable clones is g reater using this strate gy but may be overcome by using more input phages in the initial round.

VALIDATION AND TARGET IDENTIFICATION Irrespective of the method used, it is common for a given screen to yield tens to hundreds of potential phage clones that subsequentl y require anal ysis and v alidation. After the last round of selection, the enriched pool is titered , colonies pick ed, amplif ied, and DN A sequenced to produce homo genous, characterized indi vidual phage samples that are ready for analysis.

Phage Display for Imaging Agent Development

Data Analysis and Display The easiest and most straightforw ard method to assess the af finity and specif icity of indi vidual clones is via enzyme link ed immunosorbent assa y (ELISA). By far, the most common anal ysis and presentation of specificity and af finity data obtained from ELISA is the bar g raph (F igure 4A). When anal yzing up to 30 clones, the bar g raph method of visualization is adequate, ho wever, selecting the optimum clone from larger lists is f acilitated b y a more intuiti vely visual presentation such as a heat map (Figure 4B), which was made popular in cDN A microar ray data displa y. An added layer of statistical stringency can be incorporated by processing the ELISA data using multidimensional analysis.49 In this computation method , the af finities and specif icities of phage clones are processed via background subtraction using the median v alue of mock-treatment (wild-type phage) w ells from each

667

assay plate. Backg round-subtracted v alues for mocktreatment wells were accumulated across multiple assay plates to af ford tw o mock-treatment distrib utions reflecting assay noise, one corresponding to target cells and the other cor responding to subtracted cells. These mock-treatment distributions are used to nor malize independently each v alue cor responding to a phagetreated well, affording Z-nor malized (Znorm) values for each w ell in volving both tar get cells and subtracted cells. Znorm is a dimensionless quantity deri ved b y subtracting the population mean from an individual raw score and then dividing the difference by the population standard de viation. To view the data, the Znorm values from each of these e xperiments are assemb led to produce an outcome “signature” for each phage clone. Unbiased hierarchical clustering of these data sho ws a collection of outcomes with a range of af finities and specificities, w hich can be visualized as a heat map (Figure 4B).

A

B Figure 4. Presentation of ELISA data. A, Typical presentation of ELISA data via bar graph. B, Heat maps generated through multidimensional analysis. Horizontal rows represent individual phage clones. Blocks are colored to represent distance from the mean score with green being above the mean and red being below the mean. To the right of the arrow are composite scores for the affinity and specificity of the phage clones.

668

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Once the pool is nar rowed do wn to a fe w (5 to 10) interesting clones, the homo genous phage clones can then be further validated by labeling the phage coat proteins with fluorochromes such as FITC 50 then used in flow cytometry and also fluorescence microscop y. For in vi vo validation, phages ha ve been used directl y as imaging agents. By conjugating reporter molecules to the phage coat proteins, the phage becomes a targeted imaging agent that is replenishable and cheap, and the peptides (or other displayed moieties) are in the context in which they were screened.50,51 After the clones with desired proper ties ha ve been isolated, peptides can be synthesized using standard FMOC chemistry. The amino terminus is a good star ting place to attach fluorochromes or other imaging moieties because the chemistry is easy and relatively cheap. However, on the phage, the amino ter minus of the peptide is free and able to participate in binding; therefore, modif ication of the free amine should be done with caution and controls for affinity and specificity performed. An alternative to modify ing the amino ter minus has been to add a peptide link er Gl y-Gly-Ser, w hich is a common link er sequence between the peptide and the phage coat protein. After the Ser , a c ysteine, lysine, or an y amino acid ma y be added to facilitate conjugation to nanoparticles, biotin, fluorochromes, etc. Again, affinity and specif icity measurements should be performed. Once the peptides ha ve been synthesized , detailed kinetic measurements can be made using surf ace plasmon resonance. Using this technology, the kon and koff rates and af finity can be deter mined. Scatchard anal ysis for af finity deter mination can also be done. Specif icity measurements are deter mined by incubating the peptide with the target and with a negative control then analyzing using FACs for cells or tar get captured on microbeads, ELISA, or microscop y. For in vi vo specif icity, the animals are injected, and then histology is performed. If after analysis and validation is performed no clones with appropriate characteristics are identif ied, then selection can be performed using a different library, modifying the selection parameters, or making a new library with the consensus sequence from the initial phage selection constant but varying the flanking amino acids. The latter is a process known as sequence optimization or combinatorial affinity maturation. 52 Affinity maturation has been successfully used to improve binding by an ErbB-2 peptide.53

Target Identification Target identif ication of the phage displa y isolated molecules generates a mechanism w hereby potential no vel biomarkers of disease can be ascertained. There are many

methods for tar get identif ication ranging from in silico, such as searching peptide databases, to biochemical (ie, affinity chromato graphy) and genetic, such as cDN A cloning or yeast two hybrid. In silico searches of peptide specific databases may direct investigators in a direction by providing lists of “hits” or candidate tar gets that can then be v alidated with binding assa ys. Computational predictions are relatively fast and inexpensive but require links to peptide sequence databases with algorithms such as basic local alignment tool (BLAST) or Smith–W aterman search tools. P eptide databases such as Receptor Ligand Contacts (RELIC) 54 and Ar tificially Selected Proteins/Peptides Database (ASPD)55 are well curated but have onl y 1,717 and 3,632 peptide sequences, respectively. A ne wly pub lished database, P epBank (),56 has a Web-based user interface, Google-like search function, advanced text search, BLAST, and Smith–Waterman search tools, and a larger content of 21,126 peptides (F igure 5). PepBank is also a useful tool to remove sequences with known undesirable properties, that is, toxicity or nonspecific binding. If the in silico methods of tar get identif ication produce too man y or too fe w possible outcomes, e xperimental methods using affinity chromatography or genetic methods may be used. In affinity chromatography, biotin peptides are used as “bait” and either incubated with the cell or tissues positi ve for binding or immobilized on a solid suppor t then incubated with the cell or tissues of interest.57 In the f irst case, the peptide/tar get mixture is then incubated with strepta vidin-coated beads, w ashed, then eluted with e xcess biotin. In the second e xample, the beads/l ysate mixture is w ashed then eluted with either free peptide or lo w pH buffers. The resulting eluate is then r un on SDS P AGE and unique bands cut, digested with tr ypsin, then anal yzed via mass spectroscopy to identify the target protein. A disadvantage to this method is that tar gets of the de veloped imaging agents are typicall y membrane bound or associated proteins and therefore difficult to purify from the plasma membrane. Conditions needed to remo ve these proteins from the membrane are usuall y not conduci ve to protein–protein comple x stability . Re versible chemical cross-linkers with shor t link er ar ms to thw art nonspecific cross-linking are useful to circumv ent this limitation. Another potential dra wback of this method is that large amount of cells or tissues usuall y are needed to generate enough tar get to be anal yzed via mass spectroscopy. Several genetic-based methods have also been used for target identif ication. Yeast 2 hybrid screening using the peptide as the “bait” has been used with success.58,59

Phage Display for Imaging Agent Development

669

Figure 5. PepBank as a tool to identify peptide targets. Typical user workflow: user enters peptide data into either a quick search or advanced search field, the results are returned in a table that links to a more detailed page.

Because the proteins are produced in y east, however, if the peptide–target interaction relies on posttranslational modifications, then the tar get will be dif ficult if not impossible to identify . An interesting and ne wly used approach has been the use of re verse phage selection. Reverse phage selection proceeds b y immobilizing the peptide on beads or immunoplates then using cDN A phage libraries (see Table 1) to select for proteins that bind to the peptide.60 As in yeast 2 hybrid screening, the production of the protein in a bacterium is counter to interactions that rel y on eukar yotic posttranslational modifications. Creating an e xpression librar y from the cells or tissue of interest then e xpressing them in eukaryotic cells, typically COS-7 or 293T, is a potential

way to incor porate posttranslation modif ications into the protein being e xpressed. In this system, pools of cDNA (10 wells of ~100,000 each) are transfected into high producer cells, then the cells are anal yzed for peptide binding. The pool of cDN A that confers positi ve peptide binding to the w ell is then split into 10 more pools (10 w ells of 10,000 each) and the process is repeated until the pool is sufficiently narrowed to allow identification of the peptide target via DNA sequencing. In any genetic method, the screen is only as good as the cDNA librar y being produced. This method also assumes that one peptide binds to one protein, w hich may not be the case if the tar get requires cooperati ve binding between hetero subunits.

670

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

PHAGE AS IMAGING AGENTS The development of validated and optimized peptides into imaging agents has been discussed in other chapters (ie, Chapters 25, “Hyper polarized 13C Magnetic Resonance Imaging—Principles and Applications,” 27, “Optical Imaging Agents,” 29, “Multimodality Agents,” 33, “Theranostics: Agents for Diagnosis and Therapy,” 34, “Magnetic Nanopar ticles”). Interestingl y and most of all promising is the utilization of phage themselv es as nanoscaffolds for tar geted imaging agents. Phages ha ve been labeled with fluorochromes using standard isothiocyante or N-hydroxysuccinimide ester (NHS) chemistry.50,51 The resulting agents ha ve been used to identify VCAM-1 expression in inflamed vessels found in atherosclerotic plaques 50 and ha ve also been used to detect expression of molecular mark ers of in vasion and metastasis in prostate cancer (F igure 6). 50,51 Phage screening on inor ganic molecules has led to the identif ication of peptides that can bind quantum dots, 61 near infrared fluorochromes,62 and superparamagnetic iron o xide nanoparticles (A. Belcher personal communication). Coupled with new generations of phage v ectors that allow expression of multiple peptides on pVIII, 63 phages ha ve been created that ha ve biolo gical tar geting capability on pIII and are nanoscaf folds with the ability to bind multiple

detection moieties. Using the phage as nanoscaf folds carries se veral adv antages namel y: (1) the phages are replenishable and cost-ef fective and (2) the peptides are presented in the for mat in w hich the y w ere originall y screened for bioacti vity ob viating the potential to lose the desired ph ysiological proper ties through chemical synthesis of the peptides. Recentl y, selections ha ve been performed in patients with melanoma, pancreatic, and breast cancers with no adverse affects generating hope for the not-too-distant use of tar geted phage as clinicall y relevant imaging agents. 64

CONCLUSIONS Phage displa y technolo gy provides a po werful approach for the rapid and cost-ef fective disco very of lead compounds for the development into targeted imaging agents. With this technique, targeting molecules with optimum in vivo pharmacokinetics, specif icity, and af finity can be isolated. The technique is maturing and be ginning to produce agents that are being tested in clinical trials and also FDA-approved drugs. In 2003, the f irst antibody isolated from phage display, Humira® (Abbott laboratories),13 was FDA-approved to treat rheumatoid arthritis. In addition, a phage displa y–derived peptide, DX-88 (Dyax 65 and Genzyme), which inhibits kallikrein, a protein involved in vascular permeability, underwent phase I, II, and III clinical trials. 65 These successes presage a w ave of tar geted agents that have been identif ied through phage display.

REFERENCES

Figure 6. Phage as imaging agents. Phage can be fluorochrome labeled and used in whole body imaging to detect tumors (top right) or used to image cellular processes such as VCAM-1 expression in endothelium (bottom right).

1. Folli S, Westermann P, Braichotte D, et al. Antibody-indocyanin conjugates for immunophotodetection of human squamous cell carcinoma in nude mice. Cancer Res 1994;54:2643–9. 2. Neri D, Carnemolla B, Nissim A, et al. Targeting by affinity-matured recombinant antibody fragments of an angio genesis associated fibronectin isoform. Nat Biotechnol 1997;15:1271–5. 3. Goldsmith SJ. Receptor imaging: competiti ve or complementar y to antibody imaging? Semin Nucl Med 1997;27:85–93. 4. Cochran AG. Antagonists of protein-protein interactions. Chem Biol 2000;7:R85–94. 5. Moreira IS, F ernandes PA, Ramos MJ . Hot spots—a re view of the protein-protein interf ace deter minant amino-acid residues. Proteins 2007;68:803–12. 6. Russel M, Lo wman HB, Clackson T. Introduction to phage biolo gy and phage displa y. In: Clackson T, Lowman HB, editors. Phage display: a practical approach. New York: Oxford University Press; 2004. p. 1–26. 7. Landon LA, Deutscher SL. Combinatorial discovery of tumor targeting peptides using phage displa y. J Cell Biochem 2003;90:509–17. 8. Sidhu SS, Lowman HB, Cunningham BC, Wells JA. Phage display for selection of no vel binding peptides. Methods Enzymol 2000;328:333–63.

Phage Display for Imaging Agent Development

9. Smith GP. F ilamentous fusion phage: no vel e xpression v ectors that display cloned antigens on the virion surf ace. Science 1985;228:1315–7. 10. Smith GP, Petrenko VA. Phage display. Chem Rev 1997;97:391–410. 11. Willats WG. Phage displa y: practicalities and prospects. Plant Mol Biol 2002;50:837–54. 12. Ashfield R, Jak obsen BK. Making high-af finity T-cell receptors: a new class of targeted therapeutics. IDrugs 2006;9:554–9. 13. Marx J . Cancer research. Ne w insights into metastasis. Science 2001;294:281–2. 14. Hassan R, Ler ner MR, Benbrook D , et al. Antitumor acti vity of SS(dsFv)PE38 and SS1(dsFv)PE38, recombinant antimesothelin immunotoxins against human gynecolo gic cancers g rown in organotypic culture in vitro. Clin Cancer Res 2002;8:3520–6. 15. Tolmachev V, Orlova A, Nilsson FY, et al. Affibody molecules: potential for in vi vo imaging of molecular tar gets for cancer therap y. Expert Opin Biol Ther 2007;7:555–68. 16. Makowski L, Russel M. Str ucture and assembly of f ilamentous bacteriophages. In: Chiu W, Burnett RM, Garcea RL, editors. Str uctural biolo gy of vir uses. Ne w York: Oxford Uni versity Press; 1997. p. 352–380. 17. Specthrie L, Bullitt E, Horiuchi K, et al. Constr uction of a microphage v ariant of f ilamentous bacteriophage. J Mol Biol 1992;228:720–4. 18. Marvin DA, Hale RD, Nave C, Helmer-Citterich M. Molecular models and str uctural comparisons of nati ve and mutant class I f ilamentous bacteriophages Ff (fd, f1, M13), If1 and IKe. J Mol Biol 1994;235:260–86. 19. Prezzi C, Nuzzo M, Meola A, et al. Selection of antigenic and immunogenic mimics of hepatitis C vir us using sera from patients. J Immunol 1996;156:4504–13. 20. Cheadle C, Ivashchenko Y, South V, et al. Identif ication of a Src SH3 domain binding motif b y screening a random phage displa y library. J Biol Chem 1994;269:24034–9. 21. Bass S, Greene R, Wells JA. Hormone phage: an enrichment method for v ariant proteins with altered binding proper ties. Proteins 1990;8:309–14. 22. Rhyner C, Weichel M, Hübner P, et al. Phage displa y of human antibodies from a patient suffering from coeliac disease and selection of isotype-specif ic scFv against gliadin. Immunolo gy 2003;110:269–74. 23. Ditzel HJ . Human antibodies in cancer and autoimmune disease. Immunol Res 2000;21:185–93. 24. Clark MA, Hawkins NJ, Papaioannou A, et al. Isolation of human antic-erbB-2 Fabs from a l ymph node-deri ved phage displa y librar y. Clin Exp Immunol 1997;109:166–74. 25. Felici F, Castagnoli L, Musacchio A, et al. Selection of antibody ligands from a large library of oligopeptides expressed on a multivalent exposition vector. J Mol Biol 1991;222:301–10. 26. Greenwood J, Willis AE, Perham RN. Multiple display of foreign peptides on a f ilamentous bacteriophage. Peptides from Plasmodium falciparum circumsporozoite protein as antigens. J Mol Biol 1991;220:821–7. 27. Iannolo G, Minenkova O, Petruzzelli R, Cesareni G. Modifying f ilamentous phage capsid: limits in the size of the major capsid protein. J Mol Biol 1995;248:835–44. 28. Sidhu SS. Phage display in pharmaceutical biotechnology. Curr Opin Biotechnol 2000;11:610–6. 29. von Mehren M, Adams GP, Weiner LM. Monoclonal antibody therapy for cancer. Annu Rev Med 2003;54:343–69. 30. Plosker GL, F iggitt DP . Rituximab: a re view of its use in nonHodgkin’s lymphoma and chronic lymphocytic leukaemia. Drugs 2003;63:803–43. 31. Hoogenboom HR, de Bruïne AP, Hufton SE, et al. Antibody phage display technolo gy and its applications. Immunotechnolo gy 1998;4:1–20.

671

32. Rozemuller H, Chowdhury PS, Pastan I, Kreitman RJ. Isolation of new anti-CD30 scFvs from DN A-immunized mice b y phage displa y and biolo gic acti vity of recombinant immunoto xins produced b y fusion with tr uncated pseudomonas e xotoxin. Int J Cancer 2001;92:861–70. 33. Ridgway JB, Ng E, K ern JA, et al. Identif ication of a human antiCD55 single-chain Fv b y subtractive panning of a phage librar y using tumor and nontumor cell lines. Cancer Res 1999;59:2718–23. 34. Calcutt MJ, Kremer MT, Giblin MF, et al. Isolation and characterization of nucleic acid-binding antibody fragments from autoimmune mice-deri ved bacteriophage displa y libraries. Gene 1993;137:77–83. 35. Calcutt MJ, Komissarov AA, Marchbank MT, Deutscher SL. Analysis of a nucleic-acid-binding antibody fragment: constr uction and characterization of hea vy-chain complementarity-deter mining region switch variants. Gene 1996;168:9–14. 36. Mao S, Gao C, Lo CH, et al. Phage-displa y library selection of highaffinity human single-chain antibodies to tumor-associated carbohydrate antigens sial yl Le wisx and Le wisx. Proc Natl Acad Sci USA 1999;96:6953–8. 37. Dinh Q, Weng NP, Kiso M, et al. High affinity antibodies against Lex and sial yl Le x from a phage displa y librar y. J Immunol 1996;157:732–8. 38. Marks JD, Hoo genboom HR, Bonner t TP, et al. By-passing immunization. Human antibodies from V-gene libraries displa yed on phage. J Mol Biol 1991;222:581–97. 39. Koivunen E, Ga y DA, Ruoslahti E. Selection of peptides binding to the alpha 5 beta 1 inte grin from phage displa y librar y. J Biol Chem 1993;268:20205–10. 40. Brown KC. New approaches for cell-specific targeting: identification of cell-selective peptides from combinatorial libraries. Cur r Opin Chem Biol 2000;4:16–21. 41. Kelly KA, Allport JR, Tsourkas A, et al. Detection of v ascular adhesion molecule-1 e xpression using a no vel multimodal nanopar ticle. Circ Res 2005;96:327–36. 42. Arap W, K olonin MG, Trepel M, et al. Steps to ward mapping the human vasculature by phage display. Nat Med 2002;8:121–7. 43. Liu C, Bhattacharjee G, Boisvert W, et al. In vivo interrogation of the molecular display of atherosclerotic lesion surf aces. Am J Pathol 2003;163:1859–71. 44. Rajotte D, Arap W, Hagedor n M, et al. Molecular hetero geneity of the v ascular endothelium re vealed b y in vi vo phage displa y. J Clin Invest 1998;102:430–7. 45. Peters EA, Schatz PJ , Johnson SS, Do wer WJ. Isolation of a colon tumor specif ic binding peptide using phage displa y selection. Neoplasia 2003;5:437–44. 46. Kelly KA, Jones DA. Isolation of a colon tumor specific binding peptide using phage display selection. Neoplasia 2003;5:437–44. 47. Peters EA, Schatz PJ , Johnson SS, Do wer WJ. Membrane inser tion defects caused by positive charges in the early mature region of protein pIII of f ilamentous phage fd can be cor rected by prlA suppressors. J Bacteriol 1994;176:4296–305. 48. Markland W, Rober ts BL, Ladner RC. Selection for protease inhibitors using bacteriophage displa y. Methods Enzymol 1996;267:28–51. 49. Kelly KA, Clemons P A, Yu AM, Weissleder R. High-throughput identification of phage-deri ved imaging agents. Mol Imaging 2006;5:24–30. 50. Kelly KA, Waterman P, Weissleder R. In vivo imaging of molecularly targeted phage. Neoplasia 2006;8:1011–8. 51. Newton JR, Kelly KA, Mahmood U, et al. In vi vo selection of phage for the optical imaging of PC-3 human prostate carcinoma in mice. Neoplasia 2006;8:772–80. 52. Yu J, Smith GP. Affinity maturation of phage-displa yed peptide ligands. Methods Enzymol 1996;267:3–27.

672

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

53. Urbanelli L, Ronchini C, Fontana L, et al. Targeted gene transduction of mammalian cells expressing the HER2/neu receptor by filamentous phage. J Mol Biol 2001;313:965–76. 54. Mandava S, Makowski L, Devarapalli S, et al. RELIC—a bioinformatics server for combinatorial peptide anal ysis and identif ication of protein-ligand interaction sites. Proteomics 2004;4:1439–60. 55. Valuev VP, Afonnikov D A, P onomarenko MP, et al. ASPD (Ar tificially Selected Proteins/Peptides Database): a database of proteins and peptides evolved in vitro. Nucleic Acids Res 2002;30:200–2. 56. Shtatland T, Guettler D, Kossodo M, et al. PepBank—a database of peptides based on sequence text mining and public peptide data sources. BMC Bioinformatics 2007;8:280. 57. Rajotte D, Ruoslahti E. Membrane dipeptidase is the receptor for a lung-targeting peptide identif ied by in vivo phage display. J Biol Chem 1999;274:11593–8. 58. Zhang L, Hof fman J A, Ruoslahti E. Molecular prof iling of hear t endothelial cells. Circulation 2005;112:1601–11. 59. Smothers JF, Henikoff S. Predicting in vi vo protein peptide interactions with random phage displa y. Comb Chem High Throughput Screen 2001;4:585–91.

60. Essler M, Ruoslahti E. Molecular specialization of breast vasculature: a breast-homing phage-displayed peptide binds to aminopeptidase P in breast vasculature. Proc Natl Acad Sci USA 2002;99:2252–7. 61. Lee SW, Mao C, Fl ynn CE, Belcher AM. Ordering of uantum dots using genetically engineered viruses. Science 2002;296:892–5. 62. Kelly KA, Carson J , McCar thy JR, Weissleder R. No vel peptide sequence (“IQ-tag”) with high af finity for NIR fluorochromes allows protein and cell specific labeling for in vivo imaging. PLoS ONE 2007;2:e665. 63. Whaley SR, English DS, Hu EL, et al. Selection of peptides with semiconductor binding specificity for directed nanocrystal assembly. Nature 2000;405:665–8. 64. Krag DN, Shukla GS, Shen GP, et al. Selection of tumor -binding ligands in cancer patients with phage displa y libraries. Cancer Res 2006;66:7724–33. 65. Dyax. Product Pipeline: DX-88. 2007.

42 MOLECULAR IMAGING

OF

GENE THERAPY

MARÍA VERÓNICA LÓPEZ, PHD, QIANA L. MATTHEWS, PHD, DAVID T. CURIEL, MD, PHD, AND ANTON V. BOROVJAGIN, PHD

Molecular imaging techniques that emer ged in biology as a means to monitor intracellular trafficking and interaction of macromolecules have lately become a powerful tool also for detection of transgene deli very b y viral v ectors. The necessity of monitoring and quantif ication of v ectordelivered transgene e xpression prompted de velopment of new generation viral vectors. Those encoded a special class of reporter genes, whose expression in vector-targeted cells and tissues allo wed quantitati ve assessment of v ectormediated gene transfer and thereb y a noninvasive real-time tracking of the viral vectors at the tissue and the whole-body levels. Those imaging reporters are classif ied in this review according to their biochemical nature, type of the generated imaging signal, substrate requirements, and type of tracer/substrate used for signal generation. This chapter reviews imaging strate gies, used for adeno viral vectors, as a major class of gene therap y v ectors, as w ell as other (nonadenoviral) common vir us-based v ector systems. Special attention is gi ven to imaging strate gies, used to monitor replicati ve v ectors for oncol ytic therap y applications. One of those is a ne w approach of genetic capsid labeling that potentiall y allows for direct monitoring of the viral pro geny par ticles, as opposed to repor ter transgene expression, typicall y used in v ector imaging. The major advancements in imaging v ector development are re viewed in the context of their applications for research and clinical purposes.

Since the emer gence of the gene therap y f ield the necessity of monitoring and quantif ication of v ectordelivered transgene e xpression, in volving localization of the transgene, timing, duration, and magnitude of its expression, prompted development of a new generation of viral v ectors. Those vectors encoded a special class of reporter genes, whose expression in target cells and tissues allo wed tracing of the v ectors both in vitr o and in vivo by the contemporar y optic or radiolo gic imaging methods. A large variety of viral v ectors have been used so f ar for transgene imaging applications with adenovirus (Ad)-based vectors taking the lead.

The f irst-generation adeno viral v ectors used for imaging pur poses w ere represented b y replicationdeficient (or nonreplicati ve) Ads, capab le of producti ve replication only in transgenic helper cell lines. The main goals of the f irst v ector imaging application were as follows: (i) to allo w detection of vir us-infected cells, thereby estab lishing a “proof-of-principal” for the concept that some cells can be labeled b y vir uses upon infection, and (ii) to demonstrate that a specif ic cell population in a specif ic tissue can be targeted by viral infection. Subsequentl y, it w as realized that if the vir us expressing an imaging transgene w as replication competent, it would allow visualization of not onl y the primary viral infection but also spread of the viral progeny within the infected tissues. Fur thermore, a tissue and/or tumor specific replication could potentiall y be obser ved if the virus had the ability to replicate onl y in cer tain types of target cells as opposed to all cell types, adjacent to the site of v ector administration. Indeed , de velopment of conditionally replicative viral vectors allowed monitoring of viral lateralization and spread, which has an important utility for the cancer gene therap y f ield, par ticularly for oncolytic cancer therap y or so-called “virotherap y.” In this re gard, tw o distinct genetic approaches ha ve been used to engineer imaging modalities into the viral v ectors. One of them in volves a replication-dependent expression of an imaging repor ter dri ven b y viral or exogenous regulatory elements (promoters), w hereas the other is based on genetic labeling (tagging) of cer tain viral str uctural proteins with an imaging modality that allows not onl y its coordinated e xpression with other structural proteins of the vir us but also incor poration of the imaging tag into the viral capsid , thereb y directl y labeling the viral progeny particles. A noninvasive imaging of in vivo gene expression requires a repor ter gene and a technolo gy capab le of 673

674

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

noninvasively detecting/monitoring its acti vity in a particular tissue or in the whole subject (laboratory animal or a patient) in a real-time f ashion. In this chapter we re view the major advancements in imaging v ector development along with their applications for research and clinical purposes.

IMAGING OF ADENOVIRAL VECTORS Nonreplicative Adenoviral Vectors Expressing Imaging Reporter Transgenes A broad variety of reporter transgenes have been used in the conte xt of adeno viral vectors for imaging pur poses. Most of them are briefl y described and classif ied in this section, according to their biological function and type of substrates/tracers, used for detection of their activity. The available substrates are discussed in conjunction with the existing imaging techniques capable of detecting them in a noninvasive fashion. With regard to most imaging repor ters an enzymatic activity is required for generation of a detectable imaging signal. Generation of this signal is typically dependent on utilization of radioactive or nonradioactive substrates. With re gard to their acti vity inside the tar get cell, reporter genes can be classified into enzymes, receptors, transporters, and reporters (see Figure 1). The enzymatic type reporters chloramphenicol acetyl transferase (CAT) and β-galactosidase (β-gal) were one of the first imaging modalities, used for gene therap y v ector imaging. 1,2 The major disadv antage of the abo ve repor ters is that detection of their e xpression cannot be perfor med in a noninvasive f ashion and requires either a biopsy or an euthanasia of the subjects. Cer tain secretor y proteins, such as carcinoembr yonic antigen and human chorionic gonadotropin β subunit, whose intrinsic biologic properties allo w to follo w their ser um le vels b y standard immunologic bioassa ys, can also be used for imaging purposes.3,4 Over time other repor ter systems w ere de veloped that use enzymes, w hose intracellular acti vities chemically modify systemicall y-delivered nonto xic radiolabeled substrates (tracers), thereb y converting them into toxic products, ir reversibly trapped inside the tar get cells. This allo ws accumulation of a radioacti ve signal and visualization (imaging) of the vir us-infected tar get cells, e xpressing the repor ter enzyme, b y radiolo gic methods following clearance of the radioactive substrate from noninfected tissues. Thus the essential requirement for a compound to function as a molecular probe in those enzyme-based repor ter systems is an inducib le preven-

tion of its ability to cross the plasma membrane of the target cell upon modif ication (typicall y phosphor ylation) b y a repor ter enzyme (see F igure 1A). The most common reporting enzymes of that category are cytosine deaminases (CDs) from bacteria and y east and her pes simplex virus-1 thymidine kinase (HSV-tk).5–7 An alternate approach to the enzyme-based repor ter systems uses receptor molecule-based repor ters. Expression of a transgene repor ter in the for m of receptor molecule, localized to the surf ace of the v ector-transduced tar get cells, can be imaged by virtue of a specif ic, high-affinity binding of the receptor to its cognate radiolabeled ligand, agonists, or antagonists (see F igure 1B), deli vered systemically as radioactive tracers following vector administration. This system relies on the intrinsic ability of the receptor molecule for internalization upon its binding to the ligand. In this case, a radioacti ve tracer can be detected inside the receptor -expressing tar get cell as a result of its receptor -mediated inter nalization8 (see Figure 1B). An example of this approach is utilization of the human type 2 somatostatin receptor (hSSTr2) as a receptor-type reporter for imaging of a v ector-mediated gene transfer. In the proof-of-principal study the authors used an adenoviral vector (Ad5-CMV-hSSTr2), carrying the hSSTr2 as a reporter gene. Expression of this reporter was imaged noninvasively with a gamma camera, using 99mTc and 188Re-labeled somatostatin analog (P829), a synthetic peptide with high affinity for hSST r2.9 The imaging readout data in that study correlated with the Ad vector biodistribution data, demonstrating utility and high sensiti vity of the receptor reporter-based approach. The abo ve approach has been extended by its combination with HSV-tk imaging. In this regard, Zinn and colleagues demonstrated that a bicistronic Ad vector, encoding both hSSTr2 and HSV-tk, can be imaged in vivo by a simultaneous detection of both reporters, with SST r2 detection system being more sensitive.10,11 The dopamine D2 receptor (D2R) is another reporter, commonly used for v ector imaging. It is a 415 amino acid protein expressed in the brain, primarily in the striatum and pituitar y gland. 12 Several radionuclide-labeled probes for D 2R ha ve been de veloped,13–16 and the D2R/FESP (3-(2 ʹ′-[18F] fluoroeth yl) spiperone) system has been used to develop an adenoviral vector for a noninvasive imaging of the repor ter gene e xpression in living mice.12 Gene therap y has also used v ectors, encoding genes for cell surface-expressed transporter proteins. Such transporter proteins allo w a labeled probe to be transferred across the membrane and accumulate inside the target cell

Molecular Imaging of Gene Therapy

675

Figure 1. Various reporter systems used in gene therapy. Adenoviral vector-mediated delivery and expression of various reporter transgenes in the form of enzymes, receptors, transporters, or fluorescent reporters. A, An externally added substrate with imaging properties is taken up by cells, where it undergoes modification by a reporter enzyme (for instance, by phosphorylation (P)). This results in trapping and accumulation of the modified substrate (tracer) inside the enzyme-producing cells and thereby allows their detection by radiologic or lightbased methods. B, A specific ligand, used as a tracer, can bind to its cognate receptor, resulting in the accumulation of the ligand in the receptor transgene-expressing cells. C, An imaging substrate is selectively transported into the target cell, expressing a transporter molecule-encoding transgene, where it accumulates and generates an imaging signal. D, An adenoviral vector expressing a fluorescent protein as an imaging reporter. After stimulation with light of a certain wave length (excitation), light photons will be produced (emission) and captured by a photosensitive device.

(see F igure 1C). The probe can then be detected noninvasively by using an appropriate imaging technique. For example, the human sodium-iodide symporter (hNIS), normally responsible for the accumulation of iodide in the thyroid gland ,17 can ser ve as an imaging repor ter, w hen used with free 131I radioiodide as a tracer . Unlik e other imaging or therapeutic transgenes, such as HSV -tk, sodium-iodide sympor ter (NIS) is an endo genous mammalian gene, w hich mak es it less immuno genic in full y immunocompetent mammalian systems, thereby minimizing the problem of transgene-mediated immune responses. The expression and activity of hNIS, delivered by Ad vector into human glioma cells, has been demonstrated both in vitro and in vivo. In a study by Cho and colleagues subcutaneous human gliomas, g rafted in nude mice, w ere intratumorally injected with adeno viral v ector encoding hNIS (rAd-CMV-hNIS). An approximately 25-fold higher

accumulation of 125I w as obser ved in gliomas compared with the spleen or saline-injected control tumors. 18 Subcutaneous SiHa tumor xenografts, injected with Ad, encoding NIS gene from rat (rAd-CMV -rNIS), w ere successfull y imaged by a gamma camera, follo wing an intraperitoneal injection of 123I as an alternate tracer.19 A number of radiolabeled analo gs of the compounds, ser ving as substrates for v arious repor ting enzymes, have been de veloped.20–23 Some examples of those are mentioned in this and other sections belo w. The in vivo distribution pattern of a radiolabeled tracer, following its interaction with an appropriate repor ting enzyme, can be reconstructed from γ ray signals by contemporary imaging techniques such as single photon emission computed tomo graphy (SPECT) or positron emission tomo graphy (PET). Each of these imaging techniques requires utilization of a special class

676

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

radionuclide (isotope), which can be incorporated in the probe, modif ied upon interaction with the cor responding reporter molecule. SPECT involves scanning of the subject with a mo ving gamma camera to capture a radioactive tracer-emitted γ radiation and reconstitute a tomographic image of spatial distribution of the radioactive signals from their different projections. This occurs by conversion of γ rays into light signals (photons) b y a cr ystal detector inside the imaging de vice, followed b y signal collection, inte gration, and computer-based image reconstruction. PET is an alter nate three-dimensional imaging technique, based on indirect detection of positrons, emitted b y radioactive probes, as a result of nuclear deca y of their radionuclide components. Upon collision of the short-lived positrons with electrons (annihilation), the electromagnetic energy of positrons is transformed into two 511 keV γ rays, emitted in opposite directions (180°). These γ rays are simultaneously detected b y a circular ar ray of detectors inside the PET scanner. This allows determining the spatial location of each signal and reconstr ucting a three-dimensional image of radioactivity distribution in the subject. In contrast to the aforementioned imaging approaches, based on radiolo gic methods of signal detection, another distinct category of commonl y used nonin vasive imaging techniques uses light-based signal detection. This category comprises bioluminescent and fluorescent imaging approaches. The adv antages of the light-based imaging over radiologic methods include a higher spatial resolution of the image and con venience of using nonradioacti ve substrates or even no substrates at all. Bioluminescent imaging (BLI), 24 using f irefly (FLuc) or renilla (RLuc) luciferases 25,26 as bioluminescence reporter enzymes, has been widely used for monitoring adenoviral gene transfer both in vitro and in vivo. The chemical reaction, catalyzed by the above enzymes in the presence of substrate “luciferin,” in volves con version of the substrate into oxyluciferin in an ATP-dependent manner. This generates light photons that can be captured nonin vasively by a highly sensiti ve char ge-coupled de vice (CCD) camera. 27 The in vivo application of BLI typicall y in volves an intraperitoneal injection of luciferin in the e xperimental animals following administration of an adeno viral v ector, encoding a luciferase (Luc) reporter. The advantages of BLI over fluorescent imaging, described belo w, include high sensitivity, independence from e xcitation light, and deeper penetration of luminescence in animal tissues with no adverse biological effect on the cells. Besides, a shor t halflife of nati ve luciferases mak es them ideal imaging reporters for time-dependent changes in gene e xpression such as temporal changes of promoter acti vity.

Fluorescent proteins represent a separate class of imaging repor ters. Upon e xpression of a fluorescent reporter protein and its specific stimulation with a distinct spectrum light source (e xcitation), a light of a dif ferent spectrum is generated b y the e xcited fluorophore (emission) and detected by a sensitive optic device (fluorescent microscope and/or CCD camera) (see F igure 1D). The classical g reen fluorescent protein (GFP) from jell y f ish Aequorea victoria and its advanced version, the enhanced GFP (EGFP) ha ve been so f ar the most popular imaging reporter genes. They have been widely used for repetitive, noninvasive imaging in li ving cultured cells, single-cell organisms, and multicellular transparent organisms28,29 by using a broad v ariety of v ector systems, including adenoviral expression vectors.30,31 GFP is a 27 kDa protein that produces green fluorescence upon illumination with ultraviolet light. 28 It has been repor ted that genetic fusion of GFP to other proteins does not signif icantly alter its fluorescent characteristics on the one hand and does not interfere with intracellular localization of the proteins, fused to it, on the other . This allows utilization of GFP as a fluorescent tag to generate v arious labeled proteins and visualization of those in specific cell compartments as well as monitoring dynamic processes, w hich involve those proteins in living cells.32 By using a site directed mutagenesis approach to alter sequence in the chromophore re gion of the GFP gene sequence, se veral spectral v ariants of this fluorescent reporter, named blue- (BFP), cyan- (CFP), and yellow (YFP) fluorescent proteins, ha ve been generated. These v ariants ha ve dif ferent emission spectra, thereb y allowing simultaneous monitoring of multiple proteins within the same cell or organism.33,34 The major disadvantage of GFP as an imaging modality is lo w tissue penetrating ability of its fluorescence, w hich limits the ability to detect GFP in animal tissues beyond 1 to 2 mm from the source of the signal. 35 To impro ve sensiti vity of GFP detection in vivo the f iber optics technology has recentl y been coupled with confocal microscopy.36 Another g roup of fluorescent tags used for imaging of Ad vectors is represented by a family of proteins with red fluorescence spectra. The red fluorescent protein (dsRed or drFP583), originall y cloned from Discosoma coral, is a spectrall y distinct companion or substitute for GFP. By extending the spectrum of available colors to red wavelength, the dsRed pro vided a useful tool for multicolor tracking of gene expression and protein localization studies in the past fe w years. Moreover, together with GFP dsRed pro vides a donor/acceptor pair for fluorescence resonance ener gy transfer (FRET), a no vel molecular imaging technique for monitoring of protein interactions that is superior to the GFP-deri ved

Molecular Imaging of Gene Therapy

cyan/yellow FRET pair. Despite its attractiveness as a fluorescent modality , the originall y cloned dsRed protein possesses slow maturation kinetics and requires tetramerization for its acti vity. A directed evolution of dsRed led to a gradual improvement of its molecular characteristics, such as shor tening of the protein maturation time and decreasing of its oligomeric state. This resulted in generation of dsRed2 (Clontech), T1, HcRed1 (Clontech), dimer2, tdimer2(12), and f inally monomeric red fluorescent protein 1 (mRFP1), de veloped in Ro ger Tsien’s laboratory.37 Recently, a new family dsReds, collectively named “mFruits” for their monomeric nature and v ariety of colors, reflected in the protein names: mHone ydew, mBanana, mOrange, dT omato, tdT omato, mT angerine, mStrawberry, and mCher ry were generated by combination of random mutagenesis and directed e volution of mRFP1.38 While being v ery closel y related b y their amino acid sequences, most of the ne w family members differ from mRFP1 b y spectra, fluorescence quantum yield, brightness, photostability, and maturation time. The maxima for fluorescence emission spectra of mF ruits range from 537/562 (mHone ydew) to 610 (mCher ry). Most mFruits are monomeric, except for the tomato variants, and contain N- and C-ter minal peptides, deri ved from GFP protein, w hich pro vide tolerance for N-and C-terminal fusions with other proteins, similar to that known for GFP. The next advancement in the evolution of dsReds w as generation of f ar-red fluorescent proteins, such as HcRed14, mPlum15, and AQ143, w hose emission maxima reach the 650 nm bar rier (the f avorable “optical windo w” for visualization in li ving tissues is approximately 650 to 1,100 nm). 39 Finally, a new far-red fluorescent protein named “Katushka” has been reported. This protein is claimed to be 7- to 10-fold brighter compared with the spectrall y close HcRed or mPlum and is characterized b y f ast maturation as w ell as a high pHstability, brightness, and photostability. The superiority of Katushka or its monomeric for m “mKate” for w holebody imaging, demonstrated by a direct comparison with other red and f ar-red fluorescent proteins, mak es these new dsReds potentiall y the ones of choice for visualization in living tissues.40 The vast majority of imaging studies, using red fluorescent reporters, have been conducted with nonadenoviral v ectors, predominantl y lenti viral and retro viral vectors. In contrast to GFP, so far there have been only a very few reports, describing expression of dsReds in the conte xt of adeno viral v ectors. Le and colleagues constructed capsid-labeled E1/E3-deleted and wildtype adenoviruses (Ad-wt) b y fusing the minor capsid protein IX (pIX) with dsReds mRFP1 and tdimer2 (12), resulting in Ad-IX-mRFP1, Ad-IX-tdimer2 (12), and

677

Ad-wt-IX-mRFP1. Athymic mice ( n = 4), car rying xenograft tumors that w ere deri ved from A549 lung adenocarcinoma cells, w ere intratumorall y inoculated with Ad-wt-IX-mRFP1, and Ad replication was dynamically monitored with a fluorescence noninvasive imaging system. 41 In another study , adeno viral v ector expressing dsRed2 was used to transduce l ymphoma B model cells expressing artificial Ad receptor.42 As illustrated above, the contemporary vector imaging approaches use cutting-edge technologies and fully depend on the state of the ar t high precision instr umentation and computing. The astonishing advancements in vector imaging, observed for the past se veral years, have been g reatly facilitated by a concomitant de velopment of the con ventional imaging technolo gies and their applications for clinical monitoring of disease progression. Despite the broad variety of imaging modalities developed to date, only few of them have been used in the context of nonreplicati ve imaging Ad v ectors. As sho wn in Figure 2, the majority of the imaging repor ters have been engineered and e xpressed from the immediate-earl y ( E1) region of the adenoviral genome, although some have also been incorporated also in other genomic locales such asE3 and E4 regions (see F igure 2). Niu and colleagues 43 have constructed dual-expression adenoviral vectors with hNIS, cloned into the E3 region, and therapeutic genes, either manganese supero xide dismutase (MnSOD) and tumor protein 53 (TP53 or p53), incor porated into the E1 region of the Ad genome. Lowe and colleagues44 sought to restore the balance betw een the Bcl-2 f amily members to induce apoptosis in prostate cancer cells. To accomplish this, they expressed Bax gene driven by a prostate-specific promoter (probasin promoter, ARR2PB) using a second-generation adenoviral vector, expressing HA-tagged Bax gene and a GFP reporter in the E4 region.44

Transcriptional Control of Reporter Gene Expression in Nonreplicative Adenoviral Vectors Most of the reporter genes used in the context of Ad vectors are dri ven b y nonre gulatable uni versal promoters, such as c ytomegalovirus (CMV) promoter , to pro vide a constitutive (always “on”) mode of transgene expression. However, promoters that are active in a wide range of cell types have their limitations. 45 Therefore, achieving high levels of transcription in a def ined cell population through a specific cellular promoter or other re gulatory element is of a considerab le importance.45 It is noteworthy that activity of any exogenous specif ic promoter can be significantly altered, when it is placed in the context of

678

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

E1A region

ITR

E3 region

ITR

Promoter Reporter gene E4 region

Figure 2. Schematic representation of a recombinant adenoviral genome. The arrows indicate three possible regions within the adenoviral genome, suitable for cloning of imaging transgenes. ITR-Inverted Terminal Repeats; Ψ-Ad DNA packaging signal.

Ad genome, and therefore, promoters, to be used for regulation of viral replication, should be tested in the context of a nonreplicative Ad vector first. For this purpose, specific promoters are incorporated in the desired position of the genome to drive expression of a reporter gene such as luciferase.46,47 A large number of studies have been carried out to isolate and use tissue- or cancer -specific promoters to dri ve reporter gene e xpression in the conte xt of adeno viral vectors. Some of those will be briefly mentioned below. Dwyer and colleagues48 described the use of the mucin-1 (MUC1) promoter to drive expression of the NIS in breast and ovarian cancer x enografts in nude mice. MUC1-dri ven NIS gene transfer resulted in effective imaging and treatment of breast tumor x enografts in an animal model, resulting in 80% reduction in the tumor v olume.48 The same strate gy was recently used for pancreatic tumors. 49 Recently, Zeng and colleagues 50 developed an Ad-prostate-specific membrane antigen (PSMA)(E-P)EGFP vector, where EGFP was driven by enhancer/promoter of the PSMA as a re gulatory element, specif ic for prostate adenocarcinoma. Ne xt, they demonstrated inhibition of prostate tumor g rowth in the BALB/c mice upon treatment with the Ad-PSMA (E-P)CD/5-FC system, 50 expressing a suicide gene CD under control of the tumor-specific PSMA promoter. In another study an early growth response 1 promoter was placed upstream from the tumor necrosis f actor (TNF)-α gene to de velop a “TNF erade” Ad. This promoter is activated by ionizing radiation, thus allowing for temporal and spatial control of TNF-α release.51 Although adenoviral vectors have been predominantly used for cancer therap y applications, tar geting of other diseases was endeavored as well. For example, Chang and colleagues52 demonstrated that micro–positron emission tomography (microPET) imaging can be used to assess the expression of HSV-tk reporter gene in the m yocardium. To this end , an Ad encoding a silenced HSV -tk gene w as injected into the left v entricular wall of the hear t of male

transgenic mice, whose genomes carried a Cre recombinase gene, dri ven b y a cardiac-specif ic α-myosin hea vy chain promoter.52 A tissue-specif ic e xpression of the cardiac-delivered HSV-tk gene was achieved by means of cardiac-specific acti vation of Cre and the recombinasedependent removal of the silencer sequence. HSV-tk in vivo imaging w as carried out follo wing injection of the 9-[4-[18F]-fluoro-3-(hydroxymethyl)butyl]guanine ([ 18F]FHBG), as a tracer , resulting in the cardiac-specif ic tracer uptake.52 In another study, Hou and colleagues 53 combined a neuron-restrictive silencer elements, h ypoxia responsive elements, and CMV minimal promoter in the context of a nonreplicative adenoviral vector to selectively target gene e xpression to neurons in a h ypoxia-regulated manner .53 Recently Takeuchi and colleagues 54 used an Ad encoding a sterol re gulatory element-binding protein (SREBP)-1c-promoter -driven luciferase gene in transgenic mice to demonstrate that re gulation of SREBP-1c, w hich is a “master re gulator” of lipogenic gene e xpression in li ver, occurs at the transcriptional level.

Applications of Nonreplicative Adenoviral Vector Imaging for Cancer and Other Diseases The central questions in the e valuation of safety and efficacy of a gene therap y agent include the le vel, persistence, and location of a therapeutic transgene expression. For most cancers, the collection of repeated biopsy specimens is not a feasib le option. However, the use of noninvasive imaging allo ws for a consecuti ve assessment of v ector localization to be made o ver a period of time. F or this pur pose, a treatment v ector usuall y includes a second transgene, encoding a receptor that can be detected with a radiolabeled ligand, administered systemically.55

Molecular Imaging of Gene Therapy

Undoubtedly, one of the most used applications of a nonreplicative Ad in cancer treatment is suicide gene therapy. This cancer therapy approach involves a vector-mediated transfer of a suicide gene into tumor cells to render them sensitive to prodr ugs that are relati vely nontoxic to normal tissues. In this re gard, it is impor tant to note that most of the suicide gene systems can be adapted to function for both therapeutic and imaging purposes. One of the most popular and best studied suicide gene therap y systems uses ganciclo vir (GCV) in conjunction with the HSV-tk.56,57 HSV-tk phosphorylates GCV (a nontoxic prodrug), thereby converting it into a toxic, intracellularly trapped metabolite capable of killing the target tumor cells (see F igure 1A). The f irst successful demonstration of in vivo imaging of gene transfer in an animal tumor model was perfor med b y using HSV -tk gene. 58 The HSV -tkexpressing tumor cells w ere detected b y measuring the local accumulation of radioacti vity following injection of a radiolabeled derivative of the prodrug 5-[131I]-2ʹ′-deoxy2ʹ′-fluoro-β-D-arabinofuranosyl-5-iodouracil ( 131I-FIAU), by a gamma camera and a SPECT imaging. This same approach for imaging of gene transfer w as e xtended to PET using a specific, positron generating 18F-labeled prodrug derivatives.6 Besides inhibition of tumor g rowth, the other important goal of cancer gene therapy is to correlate a therapeutic gene expression with the tumor response to the v ector delivered transgene. Bar ton and colleagues 59 described a quantitative method for measuring the magnitude and distribution of a transgene e xpression in canine prostate. They developed an adenoviral vector, encoding the hNIS as a repor ter gene. F ollowing systemic administration of Na99mTcO4 tracer, autoradio graphs of prostate sections, depicting hNIS-dependent 99mTcO4 uptake, were digitized and stacked to produce a three-dimensional reconstruction of gene expression magnitude in vivo.59 In the study mentioned earlier, Niu and colleagues 43 used radiologically detectable dual-e xpressing adenoviral v ectors with hNIS as a repor ter gene that allo wed them to nonin vasively monitor transfer of TP53 and MnSOD genes in vivo and directly cor relate expression of both therapeutic genes with the hNIS imaging signal. Peñuelas and colleagues 60 have used PET with a 18Flabeled penciclovir analog as a tracer to monitor th ymidine kinase gene e xpression after intratumoral injection of a f irst-generation recombinant Ad in patients with hepatocellular carcinoma. 2 da ys after Ad injection, the transgene expression in the tumor w as detectable in all patients who received no less than 10 12 viral particles.60 Hemminki and colleagues55 showed that a retargeted Ad RGDTKSSTR, which expresses HSV-1-tk for mole-

679

cular chemotherap y and the hSST r2 for nonin vasive in vivo imaging, had an in vivo antitumor efficacy and provided ability to image the v ector-mediated SST r expression for 15 days in a murine ovarian cancer model. Another application of noninvasive imaging is monitoring the therapeutic ef fects of somaticall y transfer red genes on the processes of tumor growth, progression, vascularization, and metabolism. This is typicall y accomplished by using the con ventional imaging technolo gies: gamma camera, magnetic resonance imaging (MRI), ultrasound and PET.5 Chaudhuri and colleagues 61 used an Ad vector, encoding GFP, to infect human breast cancer cells. Then GFP-positi ve breast cancer cells w ere implanted in nude mice, treated with do xorubicin 24 hours after implantation and imaged at dif ferent da ys. Doxorubicin therapy killed GFP-positive cancer cells and gradually eliminated in vivo GFP fluorescence. This treatment could be follo wed b y nonin vasive imaging, using light microscopy.61 In another study Wu and colleagues 62 developed an Ad (Ad-CMV -VEGF121-CMV-HSV1-sr39tk) with CMV promoter -driven e xpression of angio genesisinducing v ascular endothelial g rowth f actor (VEGF) linked to a common PET repor ter gene, a modif ied sr39tk. The g roup then used a microPET imaging to assess the uptake of a PET reporter probe [18F]-FHBG by cells, e xpressing the HSV -sr39tk repor ter gene at the peri-infarct region of rat hear t.62 As discussed abo ve, the functional coe xpression of an engineered TK-GFP fusion protein w as assessed in gliosarcoma, and carcinoma cells. 32 This fusion protein served both as a screening mark er for fluorescence microscopy and fluorescence-acti vated cell sor ting and also as a therapeutic repor ter, rendering GFP-e xpressing cells, sensitive to GCV.

Oncolytic Ads Expressing Imaging Transgenes While man y molecular therapies use viral v ectors for delivery of therapeutic transgenes to target cells, virotherapy, using oncol ytic viruses, does not require therapeutic gene delivery since targeted oncolytic viruses can provide a direct therapeutic ef fect due to their intrinsic ability to kill tar get cells. Since oncol ytic agents are potentiall y toxic to the host, it is critical to monitor accurac y of their delivery, extent of dissemination, and ultimately therapeutic efficacy. In attempts to o vercome limitations, associated with the virotherapy approach, imaging paradigms ha ve also been applied to replication-competent oncol ytic vectors.

680

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

These paradigms follo w similar schemas that w ere demonstrated abo ve for nonreplicati ve vir uses. With respect to oncolytic viruses, two distinct approaches have been used to engineer imaging modalities into these v ectors. The f irst strate gy in volves e xpression of imaging modalities from the replication competent viral genomes. The second strategy involves tagging of a viral str uctural protein with an imaging modality that allo ws its coordinated e xpression with the vir us-specific component in the form of a fusion protein, on the one hand , and incorporation of the imaging tag into the viral particles, resulting in direct labeling of the viral progeny, on the other. Conditionally replicative adenoviruses (CRAds) are attractive candidate virotherapy agents due to their clinical safety, high infectivity and ability to propagate to high titers. However, despite the demonstrated safety prof ile, the overall clinical ef ficacy of CRAds has been limited. The progress in clinical improvement of CRAds has been hampered by insufficient understanding of CRAd biodistribution, replication and spread in human patients, which could be attributed in par t to the lack of adequate noninvasive CRAd monitoring systems. To compensate for this deficiency, contemporar y molecular imaging techniques have been implemented to allo w in vivo tracking of CRAd replication. Ono and colleagues h ypothesized that Ad replication could be monitored b y means of detection of EGFP , expressed from the genomic E3 region, transcription from which at late stages of Ad infection is dri ven by the viral major late promoter . Fluorescence imaging in vivo confirmed the ability to noninvasively detect fluorescent signal, reflecting the underlying level of viral DNA replication.63 In another study , Lamfers and colleagues 64 investigated transgene e xpression of an oncol ytic Ad in an intracranial murine model for glioma. The goal of this study was to increase le vels of transgene e xpression following oncolytic Ad administration to a void necessity in readministration of the v ector. This g roup e xamined the effect of immunosuppressi ve dr ug cyclophosphamide on the CRAd-encoded transgene e xpression by using CMVLuc, encoded in the Ad E3 region,64 as a reporter gene. By using a nonin vasive BLI technique, the y demonstrated a substantial decrease in viral replication and the transgene expression after intratumoral injection of oncol ytic Ad in mouse brain. Ho wever, treatment with c yclosphamide, attenuating antiadenoviral immune response, w as able to prolong CRAd-mediated gene expression.64 An alter nate approach, using luciferase imaging, was recently taken to assess the efficacy of oncolytic Ad replication in vivo . In this inno vative study, Guse and colleagues65 performed intratumoral co-injections of a

luc-expressing replication-deficient virus with eight different CRAds in two different murine xenograft models. They found cor relation betw een photon emission and infectious vir us production, suggesting that this system is suitab le for a nonin vasive imaging, quantitation of amplitude, persistence and dynamics of viral replication in vivo.65 Recently, strate gies ha ve been attempted to impro ve the oncol ytic potenc y of replication-competent Ads b y inserting therapeutic transgenes into the viral genomes. This strate gy, known as “ar ming,” provided a signif icant therapeutic benef it when applied to a number of CRAds. However, studies on CRAds, “ar med” with therapeutic transgenes, have been affected by limited knowledge about the timing and efficiency of “arming” transgene expression in vivo. To address this issue, Hakkarainen and colleagues incorporated CMV promoter dri ven HSV-tk-GFP fusion gene into the E3 region of the Ad5Δ24 CRAd, embodying an Rb-binding site mutation in the E1A gene, which confers cancer selecti vity of Ad replication. Flow c ytometry and fluorescent microscop y analyses revealed an increase in number of GFP-positi ve cells o ver time, w hich cor related with the CRAd replication. The production of HSVtk-GFP in infected cells, correlated with their sensitivity to GCV in vitro, suggesting that production of HSV-tk can be adequately monitored b y the GFP fluorescence. Ho wever, it was found that e xpression of the therapeutic gene in the presence of prodr ug GCV does not impro ve oncol ytic potency of the Ad5Δ24TK-GFP CRAd in a subcutaneous mouse model of o varian cancer and e ven appears to be counteractive with the CRAd-mediated oncolysis.66

Viral Particle Labeling Strategies for Adenoviral Vector Imaging Applications As it w as illustrated abo ve, the principle of the e xisting strategies for in vivo imaging of oncolytic viral vectors is based on the readout of an imaging transgene e xpression in the vir us-infected cancer cells. This requires that the target cells remain viab le throughout the life c ycle of the virus to express the virus-encoded imaging reporter. However, the infected cells suppor ting viral replication are ultimately killed by a replicating oncolytic virus as a result of a natural l ytic mechanism. Thus, the e xisting imaging strategies appear to be at cross purposes with the oncolytic function of CRAd agents and, therefore, fail to adequately monitor the dynamics of their replication. This warrants development of a ne w CRAd-imaging strate gy, allowing for a direct monitoring of the viral pro geny particles and providing a potentiall y more accurate assessment of the viral replication in vivo.

Molecular Imaging of Gene Therapy

One such recentl y de veloped strate gy in volves a direct labeling of the Ad capsid with an imaging tag through a genetic fusion of the tag-coding sequence to the gene of the adeno viral minor str uctural pIX. The resulting fusion protein can efficiently incorporate into the viral par ticle (capsid) without af fecting biolo gic characteristics of the vir us and allows visualization of the virions both in the solution and in the vir us infected cells (Figure 3). As a proof-of-principle, Le and colleagues67 showed that a large protein tag EGFP could be fused to pIX without compromising pIX incor poration in the viral capsid, capsid assemb ly, replication, and pro geny production of the wild-type Ad v ector. A similar f inding using a nonreplicative Ad vector was repor ted by Meulenbroek and col leagues.68 The presence of fluorescent tags on the viral capsids pro vided an adv antage of obser ving Ad-pIX-TK-Luc genome

Flag

E1 LITR

IX

E3 TK-Luc

RITR

Ad-pIX-TK-Luc virion

pIX TK-Luc

pIX-TK-Luc

pIX-TK-Luc Figure 3. Strategy for the construction of an adenovirus (Ad), carrying a herpes simplex virus-1 thymidine kinase (TK)-firefly luciferase (Luc) fusion reporter (TK-Luc), incorporated at the C-terminus of the minor capsid protein IX (pIX). Human Ad can be genetically engineered to express a modified capsid pIX and incorporate it in the viral capsid upon virus assembly. An E1/E3-deleted Ad, containing a pIX-TK-Luc carboxy-terminal fusion gene in place of the wild-type pIX (Ad-pIXTK-Luc), has been generated. An 18-amino acid linker SADYKDDDDKLAGSGSG, containing an octapeptide FLAG-tag sequence (underlined), was incorporated between the pIX and the TK-Luc coding regions (Adapted from Matthews QL et al 71).

681

Ad par ticles in infected cells b y flo w c ytometry and fluorescent microscopy in tissue sections. 67 To fur ther expand to this concept, Le and colleagues 69 fluorescently labeled canine Ad type 2 (CA V2) by fusing its pIX protein with EGFP . The purif ied labeled CA V2 virions could be visualized b y fluorescent microscop y and allowed tracking of CAV2 particles in canine cells as w ell as their tissue localization shor tly after infection. These results indicated that capsid labeling is also applicable and functional in the conte xt of a canine Ad xenotype. In that repor t it w as suggested that the capsid-labeled CA V2 could ha ve utility for v ectordevelopment studies and for monitoring of replication and spread of the CA V2-derived oncol ytic Ad in syngeneic canine cancer models. Le and colleagues fur ther e xtended the capsid labeling approach to v alidate its utility for nonin vasive monitoring of viral replication in vivo in a mouse model of adenocarcinoma. F or this pur pose they labeled adenoviral capsid with a mRFP1 or tdimer2 (12) b y fusing them to the carbo xy-terminus of pIX in the conte xt of E1/E3-deleted as w ell as replication competent Ad with the wild-type E1 genes. The advantage of using mRFP1 over EGFP is that the red fluorescent signal has better penetration ability through the tissues compared with green fluorescence, which can only be detected within 1 to 2 mm radius from the signal source. The replicating virus could be clearl y detected b y fluorescent imaging technique using a CCD camera in a flank tumor of a li ve mouse 1 to 2 da ys after intratumoral injection. The fluorescent signal increased over time and reached maximum at 3 to 4 da ys postinjection, reflecting viral replication, then showed g radual decrease, completel y disappearing between days 9 and 15. 41 A similar capsid-labeling strategy has been used to generate a replication competent Ad (carrying wild-type E1) with a clinicall y compatib le HSV-tk on the capsid. This work, perfor med by Li and colleagues,70 was a v ery timel y preclinical study . This vector demonstrated c ytotoxicity in the presence of prodrug GCV due to the acti vation of capsid-incor porated therapeutic gene (HSV -tk) and simultaneousl y a ne w imaging utility owing to the capsid incorporation of HSVtk as an imaging reporter. This study made also an important contribution to vector development as, at that time, it defined the lar gest size pIX-fused ligand that could be incorporated in adenoviral capsid without compromising its assemb ly. Subsequentl y, this size limit w as fur ther increased by the study of Matthews and colleagues71 who labeled Ad capsid with a doub le imaging repor ter HSVtk-Luc b y fusing it to pIX, using the same genetic approach. However, in contrast to the abo ve mentioned

682

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

studies, this pIX modif ication w as perfor med in the context of an E1-deleted , nonreplicati ve vir us. They demonstrated that both components of the Luc fusion retain their respecti ve enzymatic functions at the pIX locale on the capsid and function in an in vivo context, thus making imaging anal ysis feasible using a reper toire of techniques (Figure 4).71 Prior to the development of the genetic capsid labeling technique, strate gies for adeno viral capsid labeling were based on e x vivo nonspecific chemical conjugation of purif ied viral par ticles with synthetic fluorophores such as fluorescein isothioc yanate (FITC), Texas Red, or carbocyanine (DiR) dy es. Ho wever, chemical conjugation to fluorophores w as used exclusively for addressing basic virologic questions such as molecular mechanisms of adenoviral entry,72,73 the effect of fiber modification on intracellular traf ficking of Ad,74,75 or interactions of Ad with cellular f ilaments.76 Besides labeling of adeno viral capsids, ef forts were also made to de velop strate gies for labeling adeno viral genome as a unique tool for following adenoviral genome delivery, an imperative step in both natural infection and gene transfer. In earlier studies, an ele gant DNA-protein interaction-based approach was used to label Ad genomic DNA. A high-affinity specif ic binding betw een tetR protein and tetO DN A sequence repeats w as used to tether terR-EGFP fusion protein to Ad DNA via incorporated tetO tandem repeats. 77 Although successful, this DNA labeling system had limitations including its dependence of on the tetR-EGFP producer cell line and a recombination-related instability of the tetO sequence tags within the Ad genome. The rationale for labeling of adenoviral core proteins originates from the idea of an alternate labeling of the Ad genome via an endo genous viral DN A-binding protein expressed from the vector itself, as opposed to exogenous protein label such as tetR-EGFP . Le and colleagues 78 hypothesized that the adenoviral core proteins V, VII, and X (Mu), organizing structure of viral DNA in the particle through a nonco valent association, ma y ser ve as tar gets for labeling of Ads with fluorescent proteins. They constructed various Ad vectors including replication competent (wt-E1) variants with each of the fusion core protein genes (Mu-EGFP , V-EGFP, preVII-EGFP , or matVIIEGFP) placed in the deleted E3 region while retaining the core protein genes in their nati ve genomic locales. Expression from the E3 region allowed a robust production of the fusion genes, dri ven b y the viral major late promoter. The level of the fusion core protein e xpression was equi valent to that of the nati ve core proteins. An efficient labeling, allowing visualization of purified virus

A

B

Figure 4. Micro–positron emission tomography (microPET) and bioluminescent imaging analysis of mice infected with an adenovirus (Ad) containing a capsid-incorporated pIX-TK-Luc fusion protein. Athymic nude mice were implanted with subcutaneous xenografts of the human head and neck squamous cell carcinoma cell line (FaDu). Tumor nodules were injected with Ad and then subjected to imaging analyses. A, MicroPET imaging, animals were injected with a control Ad, encoding HSV-tk (TK), Ad-CMV-TK (1 × 1010 viral particles [VPs] in 0.1 mL phosphate-buffered saline [PBS]) on the left flank tumor and with the dual-modality vector AdpIX-TK-Luc (1 × 1010 VPs in 0.1 mL PBS) on the opposite side tumor. Mice were subjected to microPET analysis 48 hours after injection. B, Bioluminescence imaging, animals were injected intratumorally with a control Ad encoding luciferase, Ad-CMV-luc (1 × 1010 VPs in 0.1 mL PBS) on the left flank and with the dual-modality vector on the right flank (Adapted from Matthews QL et al 71).

Molecular Imaging of Gene Therapy

and tracking of the viral core during earl y infection, was obtained for Ad-wt-E3-V-EGFP and Ad-wt-E3-preVIIEGFP vectors. These vectors maintained their biolo gical function, including ther mostability, viral DN A replication, viral DNA encapsidation, and the ability to induce a cytopathic effect. This study was the f irst to demonstrate that genetic labeling of core proteins offers a unique way to follo w the Ad core with potential for studying Ad infection and biolo gy as w ell as tracking adeno viral vectors in gene therapy applications. The major limitation of this approach w as the mosaic nature of the corelabeled particles, composed of both GFP-fusion chimeras and native forms of the corresponding core proteins. As a result, efficiency of par ticle labeling was reduced due to competitive incorporation of the wild-type core proteins, expressed from their native locales.

IMAGING OF NONADENOVIRAL VECTORS Nonadenoviral Vectors, Expressing Imaging Transgenes Many g roups have used a nonin vasive bioluminescence imaging (BLI) for detection of vir us-encoded luciferase transgene e xpression in li ve animals. This allo ws realtime monitoring of viral replication and localization in intact subjects. Mouse models of her pes simple x vir us type 1 (HSV-1) infection pro vide essential infor mation about viral and host genes that re gulate the disease patho genesis. However, the conventional assays, commonly used to evaluate viral replication and spread , require that the infected animals be sacrificed. To develop a noninvasive means of HSV -1 detection in li ving mice, Luk er and colleagues used KOS HSV-1 recombinant strain, w hich expresses both f irefly ( Photinus p yralis) and Renilla (Renilla r eniformis) luciferase repor ter proteins, and monitored viral infection in mice using a CCD camera. Viral infection in mouse footpads, brain, peritoneal cavity, eyes, and brain could be detected by BLI of firefly luciferase. 79 This w ork w as fur ther e xtended to establish a transgenic repor ter mouse model for BLI imaging of HSV-1 infection in living mice. 80 To investigate the effects of interferons on spatial and temporal pro gression of v accinia virus infection, scientists generated recombinant vaccinia viruses that express firefly luciferase or a monomeric orange fluorescent protein. These vir uses af forded imaging of v accinia vir us infection with BLI and fluorescence imaging. 81 Many g roups ha ve used BLI to image lenti viral vector-mediated gene transfer in the brain. Gene transfer

683

into the central ner vous system is a rapidl y adv ancing therapeutic strategy for a range of neurolo gic conditions, including neurode generation. This approach benef its from ne wly designed imaging technolo gies that could pinpoint the magnitude, extent, and duration of transgene 82 constructed expression. Deroose and colleagues lentiviral vectors based on human immunodeficiency virus type 1 (HIV-1), encoding f irefly luciferase. Those v ectors were injected into the brains of mice and rats and could be traced by BLI, which allowed characterization of vector localization in the brain as w ell as kinetics and reproducibility of infection. 82 The most recent studies on adeno-associated vir uses (AAVs) demonstrate that recombinant AAV is a promising vector for gene therap y of photoreceptor -based diseases. Previous studies showed that AAV-2 and -5 transduce both rod and cone photoreceptor cells in rodents and dogs, and these serotypes can tar get rods but not cones in primates. Mancuso and colleagues 83 have repor ted a successful targeting of GFP reporter gene to primate cones, using an AAV-5 vector, car rying a human cone-specif ic enhancer and promoter to regulate GFP expression. The time course of GFP expression was monitored in a living animal using the RetCam digital imaging system. 83 Greenberg and colleagues constr ucted both transcriptionally and transductionall y targeted HIV-1-based lentiviral v ectors, capab le of transducing Muller glia cells of health y and diseased retinas. The conclusion from this study , based on a nonin vasive imaging of vector-mediated EGFP e xpression, w as that pseudotyped lenti viruses, containing glia-specif ic promoters, efficiently transduce retinal Muller glia, allo wing a sustained transgene e xpression in the retina. Such v ectors will be useful for e xperimental treatment of retinal disease, as w ell as for ph ysiologic and de velopmental studies of the retina. 84 Multimodality repor ter fusion proteins ha ve been recently constructed by Gambhir’s group to enable simultaneous noninvasive live imaging of laboratory animals by using BLI, fluorescence light imaging approaches, and PET.53,85,86 A number of retro viral vectors, which express various dual and triple reporter fusion proteins, have been designed.86 To use a multimodality imaging capacity in vivo, Kim and colleagues constr ucted a lenti viral vector, encoding a tri-fusion repor ter gene, comprising sr39tk, synthetic Renilla luciferase, and EGFP. Using this new multimodality system, the y could detect tumor specific mig ration of v ector-transduced l ymphocytes b y both microPET scan and BLI. 86 Imaging strategies have also been applied to nonadenoviral oncolytic agents. A replication-competent HSV-1

684

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

oncolytic virus with its native HSV-tk gene under control of various promoters has been studied by PET imaging to assess the impact of promoter characteristics on viral infection and replication. These in vivo studies demonstrated that as little as 1 × 107 particles of the replicati ve HSV-1 v ector, injected into a 0.5 cm human colorectal tumor, can be detected b y a labeled FIA U-mediated PET scanning. Bennett and colleagues demonstrated the ability of PET imaging to detect subtle changes in tk expression due to local replication of the vir us. Noninvasive imaging studies, such as these, are of significance to demonstrate biodistribution of HSV-1-based therapies.87 Adusumilli and colleagues 88 studied a replicationcompetent HSV-1 as a means to minimize the use of invasive oncologic surgery. These researchers sought to establish whether a strain of HSV -1 NV1066 could distinguish a tumor tissue from a normal tissue. Their HSV1 v ariant e xpressed GFP and specif ically replicated in cancer cells, w hen administered either locall y (intratumoral or intraca vitary) or systemicall y and detected loco-regional and distant disease throughout the body . Such cancer selectivity was conf irmed in 110 types of cancer cells from 16 different primar y organs. Fluorescence-aided minimall y in vasive endoscopy re vealed microscopic tumor deposits, unreco gnized b y con ventional laparoscopy/thoracoscopy. Fur thermore, the ability of NV1066 to transit and infect tumor and metastases was proven in syngenic and transplanted tumors in different animal models, both immunocompetent and immunodeficient. These studies for m the basis for realtime, intraoperati ve diagnostic imaging of tumor and metastases b y minimall y in vasive endoscopic technology. The signif icance of this study w as that it e xpanded the use of HSV-guided fluorescence from solid tumor to distinct body cavities. This study also demonstrated that administered vir us can tra vel along v arious paths to identify cancer metastases. 87 Along those same lines, Adusumilli and colleagues used the replication-competent HSV -1 (NV1066) to investigate the potential of identifying l ymph node metastases intraoperati vely b y using her pes vir usguided cancer-specific expression of GFP. After infection of human mesothelioma cancer cell lines, NV1066 replicated (5-fold–17,000-fold) and e xpressed GFP in all cancer cells, w hile lea ving nor mal l ymphocytes uninfected. When injected into primar y tumors of murine models of l ymphatic metastasis, the vir us was able to locate and infect l ymph node metastases producing GFP that w as visualized b y means of fluorescent imaging. Histolo gy conf irmed l ymphatic metastases, and immunohistochemistry confirmed viral

presence in re gions that e xpressed GFP. The authors concluded that herpes virus-guided cancer cell-specific production of GFP can f acilitate accurate diagnosis of small metastatic deposits of pleural and peritoneal cancers, invisible to the nak ed eye. The sensitivity of this system could f acilitate oncologic surgery and impro ve cancer diagnostics accuracy.89 Since regulatory elements of oncol ytic viruses play a critical role in the viral life c ycle, Yamamoto and colleagues90 incorporated luciferase cassettes into HSV-1 vectors to monitor in vivo activity of the immediate-early and strict-late HSV-1 promoters in preclinical tumor treatment studies for e valuation of their role in HSV -1-mediated tumor oncolysis. MRI represents another imaging technique that can be used for the visualization of viral v ector deli very in vivo.91,92 Räty and colleagues 91 were able to produce vector-related MRI contrast in the choroid ple xus cells of rat brain in vivo over a period of 14 da ys by conjugating avidin-coated baculo viral v ectors (Baa vi) with biotin ylated ultrasmall super paramagnetic iron oxide particles.91 In another study , Wang and colleagues 92 analyzed pharmaceutical ef ficacy of a “suicide” gene CD , con verting prodrug 5-fluorocytosine to 5-fluorouracil, by using fluorine-19 nuclear magnetic resonance spectroscop y in a rat intracranial tumor model. 92

Capsid Labeling of Nonadenoviral Vectors Nonadenoviral v ectors ha ve been broadl y used for gene therapy and v accine schemas. In this re gard, imaging of non-Ads, as it relates to vir us traf ficking and transgene expression, has been paramount for v ector de velopment. As it has been demonstrated above for Ad vectors, the successful imaging of viral par ticles requires the ability to incorporate an imaging tag in the viral par ticle with high efficiency and detect the incor porated imaging modality with high level of sensitivity, using an appropriate imaging technology. The ef forts in labeling viral par ticles can be classified into genomic and capsid/envelope labeling. Genomic Labeling

Chemical conjugation with fluorescent dyes, such as DAPI (4ʹ′,6ʹ′,-diamidino-2-phenylindole), YoPro-1 {4-[3-methyl2,3-dihydro-(benzo-1,3-oxazole)-2-methylmethyledene]1(3ʹ′-trimethylammoniumpropyl)-quinoliniumdiiodide} and SYBR g reen has been used to enumerate vir us-like particles in marine samples. One such early attempt was to stain marine viruses with a nucleic acid stain, SYBR gold.

Molecular Imaging of Gene Therapy

Chen and colleagues 91 used SYBR gold for digital image analysis and flo w cytometry to enumerate marine vir uses in costal samples. Marine viruses stained with SYBR gold yielded bright and stab le fluorescent signal that could be detected with CCD camera or b y flow cytometry. Digital images of the SYRB-stained viruses were created to determine concentration of viral particles, using digital analysis software. The estimates based on digitized images were 1.3 times higher than those based on direct counting using an epifluorescence microscope. These strate gies allo w identification of viral subpopulations within anal yzed samples and ha ve a g reat impact on marine ecosystem studies.91 Capsid/Envelope Labeling

Chemical conjugation labeling of viral particles was used for elucidating the molecular mechanism of a membraneenveloped Sendai virus fusion with cell membranes. This study by Rocheleau and P eterson94 was aimed at distinguishing betw een the tw o e xisting fusion mechanisms: the pore-forming mechanism and the hemifusion mechanism. Labeling viral lipids with fluorescent dy es such as 4-(4-(dihexadecylamino)styryl)-N-methyl-quinolinium iodine (DiQ) and viral proteins with FITC with subsequent analysis of vir us-cell fusions by image cor relation spectroscopy supported the mechanism, involving formation of a hemifusion intermediate. The hemifusion model, also known as the proximity model, requires involvement of viral proteins to bring the virus and the cell into a close contact. Studies in volving labeling of viral par ticles w ere conducted also on v accinia vir us, the prototype of poxviruses. Recombinant v accinia vir uses with GFP , genetically fused to the most abundant protein of the extracellular enveloped virus P37 (F13L), were generated to study their intracellular transpor t mechanisms. These studies unco vered a second , microtubule-dependent mechanism for intracellular transpor t of en veloped vaccinia virions.95 A similar study was performed by Ward and Moss, 96 who produced infectious vaccinia virus that expressed the BR5 envelope glycoprotein, fused to GFP.94 This labeling allowed visualization of intracellular mo vement of the virus in real time. One of the f indings of this study w as that direction, speed, and saltatory motion of the intracellular enveloped virus (IEV) were consistent with the role of microtubules in intracellular transpor t of IEV.96 Additional studies b y Rodger’ s g roup demonstrated that replacement of short consensus repeats of the e xtracellular domain of the vaccinia virus envelope protein B5R by

685

EGFP produced smaller plaques but did not abolish transport of the viral par ticles to the cell surf ace.97 A similar imaging approach was used in studies with new subtypes of influenza A vir us responsib le for pandemics, such as the Spanish flu pandemic in 1918 to 1919.98 Since the new virus subtypes are distinct from the existing viral populations due to the dif ferences in viral membrane gl ycoprotein hemagglutinin (HA), it w as important to deter mine if HA of the ne w viral subtypes can potentially cause pandemics b y testing the ne w isolates against a variety of human cell types. The approach involved labeling of the influenza vir uses with DiD , a fluorescent lipophilic dy e that spontaneousl y par titions into the viral en velope membrane. This labeling scheme allowed detection of a single vir us-mediated membrane fusion e vent and the viral ph ysical trajector y in the infected cell. The infection mechanism of influenza virus has also been in vestigated by tracking viral par ticles in living cells using a fluorescence microscopy.99,100 To impro ve on pre viously used fluorescence dequenching assays, Sakai and colleagues 101 developed a dual-wavelength imaging technique that detects indi vidual virus-endosome fusion events as a change in the color of fluorescent signal, associated with viral par ticles, which can be detected b y fluorescence microscopy. This approach allowed visualizing fusion events between virus and very small endosomes, w hich could not be detected previously, indicating high sensiti vity of the approach. The dual-wavelength detection imaging method is beneficial for virologic research, involving studies on mechanisms of viral infection. The dual-w avelength imaging technique also elucidated the role of the HA fusion events in a specif ic host cell response. 101 To in vestigate the mechanism of en veloped HIV virus entr y and deli very of its genome into susceptib le cells, Lampe and colleagues 102 constructed a double-colored HIV deri vative, car rying a g reen fluorescent label attached to the viral matrix protein and a red label fused to the viral VPr protein. This doub le-labeling approach allowed distinguishing between the complete virions and the sub viral par ticles, lacking matrix after membrane fusion. This study demonstrated that the abo ve doub lelabeling system is a novel tool that can be combined with live cell imaging and is suitab le for real-time analyses of HIV-cell interaction.102 Viral entr y of nonen veloped vir uses has also been investigated b y means of imaging anal ysis. This w as accomplished b y Brandenbur g and colleagues, 103 who labeled capsid and RNA of a poliovirus (PV). Their findings define early events in PV infection. Also, these findings define a pathway in which PV binds to receptors on

686

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

the cell surface and enters the cell by tyrosine kinase and actin-dependent endocytic mechanism.103

CONCLUSION It is obvious that gene therapy enters a new era, in which noninvasive imaging strate gies will pla y an impor tant role in the de velopment, biologic validation, and clinical applications of the viral v ectors. Noninvasive imaging in its various forms will not only become an integral part of most gene therap y applications in the near future, but is lik ely to ha ve a positi ve impact on the whole gene therapy field development. As illustrated in this review chapter, γ ray imaging has been already extensively used for imaging gene transfer in animal models and is lik ely to be translated into similar studies in humans. Utilization of red fluorescent repor ters with far red emission spectra that allow for a deeper tissue penetration of the light signal in animal models of human diseases is lik ely to become the ne xt step in e volution of noninvasive imaging and could also translate in the human clinical settings upon improvement of the signal detection sensitivity of instr umentation. A rapid de velopment of optical and digital imaging techniques, accompanied b y that of high-technolo gy instr umentation, opens wide perspectives for noninvasive imaging of targeted gene therapy vectors, using dsReds of the “mFruits” family as imaging repor ters. Specif ically, the emer ging ne w spectral imaging technology104 is capable of discriminating a fluorescent tag-associated signal from autofluorescence background of cells and tissues, based on differences in spectral prof iles of the tw o types of fluorescence. The associated image processing softw are can “unmix” digital images and “extract” a specific fluorescence signal. The importance of noninvasive imaging is underscored by its inte gration in the protocols of human clinical trials. Two such trials are cur rently planned at our institution.

ACKNOWLEDGMENTS This w ork w as suppor ted b y g rants from the National Institute of Health (5R01CA111569 and 5T32CA075930) to David T. Curiel.

REFERENCES 1. Leite JP, Niel C, D’Halluin JC. Expression of the chloramphenicol acetyl transferase gene in human cells under the control of earl y adenovirus subg roup C promoters: ef fect of E1A gene products from other subgroups on gene expression. Gene 1986;41:207–15. 2. Naciff JM, Behbehani MM, Misa wa H, Dedman JR. Identif ication and transgenic analysis of a murine promoter, that targets cholinergic neuron expression. J Neurochem 1999;72:17–28.

3. Kanerva A, Zinn KR, P eng KW, et al. Nonin vasive dual modality in vivo monitoring of the persistence and potenc y of a tumor targeted conditionall y replicating adeno virus. Gene Ther 2005;12:87–94. 4. Peng KW, Facteau S, Wegman T, et al. Non-invasive in vivo monitoring of trackab le vir uses expressing soluble marker peptides. Nat Med 2002;8:527–31. 5. Brix G, Bellemann ME, Haberkorn U, et al. Assessment of the biodistribution and metabolism of 5-fluorouracil as monitored b y 18F PET and 19F MRI: a comparati ve animal study . Nucl Med Biol 1996;23:897–906. 6. Gambhir SS, Barrio JR, Herschman HR, Phelps ME. Assays for noninvasive imaging of repor ter gene e xpression. Nucl Med Biol 1999;26:481–90. 7. Blasberg RG, Tjuvajev JG. Herpes simplex virus thymidine kinase as a marker/reporter gene for PET imaging of gene therapy. Q J Nucl Med 1999;43:163–9. 8. Briat A, Vassaux G. Preclinical applications of imaging for cancer gene therapy. Expert Rev Mol Med 2006;8:1–19. 9. Zinn KR, Buchsbaum DJ, Chaudhuri TR, et al. Noninvasive monitoring of gene transfer using a reporter receptor imaged with a highaffinity peptide radiolabeled with 99mTc or 188Re. J Nucl Med 2000;41:887–95. 10. Rogers BE, Zinn KR, Buchsbaum DJ . Gene transfer strate gies for improving radiolabeled peptide imaging and therap y. Q J Nucl Med 2000;44:208–23. 11. Zinn KR, Chaudhuri TR, Buchsbaum DJ, et al. Simultaneous evaluation of dual gene transfer to adherent cells by γ-ray imaging. Nucl Med Biol 2001;28:135–44. 12. MacLaren DC, Gambhir SS, Satyamurthy N, et al. Repetitive, noninvasive imaging of the dopamine D2 receptor as a repor ter gene in living animals. Gene Ther 1999;6:785–91. 13. Ehrin E, Farde L, De Paulis T, et al. Preparation of 11C-labelled Raclopride, a new potent dopamine receptor antagonist: preliminary PET studies of cerebral dopamine receptors in the monk ey. Int J Appl Radiat Isot 1985;36:269–73. 14. Hume SP, Myers R, Bloomfield PM, et al. Quantitation of carbon-11labeled raclopride in rat striatum using positron emission tomo graphy. Synapse 1992;12:47–54. 15. Hall H, K ohler C, Ga well L, et al. A ne w selecti ve ligand for the dopamine-D2 receptors. Pro g Neuropsychophar macol Biol Psychiatry 1988;12:559–68. 16. Barrio JR, Satyamurthy N, Huang SC, et al. 3-(2 ʹ′-[18F] fluoroethyl) spiperone: in vivo biochemical and kinetic characterization in rodents, nonhuman primates, and humans. J Cereb Blood Flo w Metab 1989;9:830–9. 17. Cho JY. A transporter gene (sodium iodide sympor ter) for dual purposes in gene therap y: imaging and therap y. Cur r Gene Ther 2002;2:393–402. 18. Cho JY, Xing S, Liu X, et al. Expression and activity of human Na+/Isymporter in human glioma cells b y adeno virus-mediated gene delivery. Gene Ther 2000;7:740–9. 19. Boland A, Ricard M, Opolon P, et al. Adenovirus-mediated transfer of the thyroid sodium/iodide sympor ter gene into tumors for a targeted radiotherapy. Cancer Res 2000;60:3484–92. 20. Iyer M, Bar rio JR, Namavari M, et al. 8-[18F] Fluoropenciclo vir: an improved repor ter probe for imaging HSV1-tk repor ter gene expression in vivo using PET. J Nucl Med 2001;42:96–105. 21. Gambhir SS, Bauer E, Black ME, et al. A mutant her pes simple x virus type 1 th ymidine kinase repor ter gene sho ws impro ved sensitivity for imaging repor ter gene e xpression with positron emission tomo graphy. Proc Natl Acad Sci U S A 2000;97: 2785–90. 22. Namavari M, Bar rio JR, Toyokuni T, et al. Synthesis of 8-[(18)F] fluoro guanine deri vatives: in vivo probes for imaging gene e xpression with positron emission tomo graphy. Nucl Med Biol 2000;27:157–62.

Molecular Imaging of Gene Therapy

23. Liang Q, Satyamur thy N, Bar rio JR, et al. Nonin vasive, quantitative imaging in li ving animals of a mutant dopamine D2 receptor reporter gene in w hich ligand binding is uncoupled from signal transduction. Gene Ther 2001;8:1490–8. 24. Zhang Y, Bressler JP, Neal J, et al. ABCG2/BCRP expression modulates D-Luciferin based bioluminescence imaging. Cancer Res 2007;67:9389–97. 25. Wilson T, Hastings JW. Bioluminescence. Annu Rev Cell De v Biol 1998;14:197–230. 26. Yu YA, Timiryasova T, Zhang Q, et al. Optical imaging: bacteria, viruses, and mammalian cells encoding lightemitting proteins re veal the locations of primar y tumors and metastases in animals. Anal Bioanal Chem 2003;377:964–72. 27. Klerk CP, Overmeer RM, Niers TM, et al. Validity of bioluminescence measurements for nonin vasive in vivo imaging of tumor load in small animals. Biotechniques 2007;43:7–13. 28. Chalfie M. Green fluorescent protein. Photochem Photobiol 1995;62:651–6. 29. Chalfie M, Au M. Genetic control of dif ferentiation of the Caenorhabditis ele gans touch receptor neurons. Science 1989;243:1027–33. 30. de Mar tin R, Raidl M, Hofer E, Binder BR. Adenovirus-mediated expression of g reen fluorescent protein. Gene Ther 1997;4:493–5. 31. Smith GM, Berry RL, Yang J, Tanelian D. Electrophysiological analysis of dorsal root g anglion neurons pre- and post- coexpression of green fluorescent protein and functional 5-HT3 receptor. J Neurophysiol 1997;77:3115–21. 32. Jacobs A, Dubrovin M, He wett J, et al. Functional coe xpression of HSV-1 thymidine kinase and g reen fluorescent protein: implications for noninvasive imaging of transgene e xpression. Neoplasia 1999;1:154–61. 33. Ellenberg J, Lippincott-Schwartz J, Presley JF. Dual-colour imaging with GFP variants. Trends Cell Biol 1999;9:52–6. 34. Livet J, Weissman TA, Kang H, et al. Transgenic strategies for combinatorial expression of fluorescent proteins in the ner vous system. Nature 2007;450:56–62. 35. Shah K, Jacobs A, Breakefield XO, Weissleder R. Molecular imaging of gene therapy for cancer. Gene Ther 2004;11:1175–87. 36. Ilyin SE, Flynn MC, Plata-Salaman CR. F iber-optic monitoring coupled with confocal microscop y for imaging gene e xpression in vitro and in vivo. J Neurosci Methods 2001;108:91–6. 37. Campbell RE, Tour O, Palmer AE, et al. A monomeric red fluorescent protein. Proc Natl Acad Sci U S A 2002;99:7877–82. 38. Shaner NC, Campbell RE, Steinbach PA, et al. Improved monomeric red, orange and y ellow fluorescent proteins deri ved from Discosoma sp. red fluorescent protein. Nat Biotechnol 2004;22: 1567–72. 39. Shkrob MA, Yanushevich YG, Chudakov DM, et al. F ar-red fluorescent proteins e volved from a b lue chromoprotein from Actinia equina. Biochem J 2005;392:649–54. 40. Shcherbo D , Merzl yak EM, Chepur nykh TV, et al. Bright f ar-red fluorescent protein for w hole-body imaging. Nat Methods 2007;4:741–6. 41. Le LP, Le HN , Dmitriev IP, et al. Dynamic monitoring of oncol ytic adenovirus in vivo by genetic capsid labeling. J Natl Cancer Inst 2006;98:203–14. 42. Mättö, M, Nuutinen UM, Hakkarainen T, et al. hCAR-EGFP fusion receptor in human follicular l ymphoma B cells—a model for adenoviral gene therapy for B cell malignancies. Int J Mol Med 2006;17:1057–62. 43. Niu G, Anderson RD, Madsen MT, et al. Dual-e xpressing adenoviral vectors encoding the sodium iodide symporter for use in noninvasive radiological imaging of therapeutic gene transfer . Nucl Med Biol 2006;33:391–8.

687

44. Lowe SL, Rubinchik S, Honda T, et al. Prostate-specific expression of Bax deli vered b y an adeno viral v ector induces apoptosis in LNCaP prostate cancer cells. Gene Ther 2001;8:1363–71. 45. Nettelbeck DM, Jerome V, Muller R. Gene therap y: designer promoters for tumour targeting. Trends Genet 2000;16:174–81. 46. Zhu ZB, Makhija SK, Lu B , et al. Transcriptional targeting of adenoviral vector through the CXCR4 tumor -specific promoter. Gene Ther 2004;11:645–8. 47. Rein DT , Breidenbach M, Curiel DT . Cur rent de velopments in adenovirus-based cancer gene therap y. Future Oncol 2006;2:137–43. 48. Dwyer RM, Ber gert ER, O’Connor MK, et al. in vivo radioiodide imaging and treatment of breast cancer x enografts after MUC1driven e xpression of the sodium iodide sympor ter. Clin Cancer Res 2005;11:1483–9. 49. Dwyer RM, Ber gert ER, O’Connor MK, et al. Adenovirus-mediated and tar geted e xpression of the sodium-iodide sympor ter per mits in vivo radioiodide imaging and therapy of pancreatic tumors. Hum Gene Ther 2006;17:661–8. 50. Zeng H, Wei Q, Huang R, et al. Recombinant adeno virus mediated prostate-specific enzyme pro-dr ug gene therap y re gulated b y prostate-specific membrane antigen (PSMA) enhancer/promoter . J Androl 2007:28:827–35. 51. McLoughlin JM, McCarty TM, Cunningham C, et al. TNFerade, an adenovector car rying the transgene for human tumor necrosis factor α, for patients with adv anced solid tumors: sur gical experience and long-term follow-up. Ann Surg Oncol 2005;12: 825–30. 52. Chang GY, Cao F, Krishnan M, et al. P ositron emission tomography imaging of conditional gene acti vation in the hear t. J Mol Cell Cardiol 2007;43:18–26. 53. Hou, S. T., Jiang, S. X., Huang, D ., et al. A no vel adeno viral vector-mediated neuronal selecti ve gene e xpression in neonatal mouse brain in response to h ypoxia. Neurosci Lett. 2007; 419(1):23–7. 54. Takeuchi Y, Yahagi N, Nakagawa Y, et al. in vivo promoter analysis on refeeding response of hepatic sterol re gulatory elementbinding protein-1c e xpression. Biochem Bioph ys Res Commun 2007;363:329–35. 55. Hemminki A, Zinn KR, Liu B, et al. in vivo molecular chemotherapy and nonin vasive imaging with an infecti vity-enhanced adenovirus. J Natl Cancer Inst 2002;94:741–9. 56. Moolten FL, Wells JM, Heyman RA, Evans RM. Lymphoma regression induced b y ganciclovir in mice bearing a her pes thymidine kinase transgene. Hum Gene Ther 1990;1:125–34. 57. Adachi Y, Tamiya T, Ichikawa T, et al. Experimental gene therap y for brain tumors using adeno virus-mediated transfer of c ytosine deaminase gene and uracil phosphoribosyltransferase gene with 5-fluorocytosine. Hum Gene Ther 2000;11:77–89. 58. Tjuvajev JG, Finn R, Watanabe K, et al. Noninvasive imaging of herpes virus thymidine kinase gene transfer and expression: a potential method for monitoring clinical gene therap y. Cancer Res 1996;56:4087–95. 59. Barton KN, Xia X, Yan H, et al. A quantitative method for measuring gene expression magnitude and volume delivered by gene therapy vectors. Mol Ther 2004;9:625–31. 60. Peñuelas, I, Mazzolini G, Boán, JF, et al. Positron emission tomography imaging of adenoviral-mediated transgene expression in liver cancer patients. Gastroenterology 2005;128:1787–95. 61. Chaudhuri TR, Cao Z, Rodriguez-Burford C, et al. A non-in vasive approach for monitoring breast tumor cells during therapeutic intervention. Cancer Biother Radiopharm 2002;17:205–12. 62. Wu JC, Chen IY , Wang Y, et al. Molecular imaging of the kinetics of vascular endothelial g rowth f actor gene e xpression in ischemic myocardium. Circulation 2004;110:685–91.

688

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

63. Ono HA, Le LP, Davydova JG, et al. Noninvasive visualization of adenovirus replication with a fluorescent repor ter in the E3 re gion. Cancer Res 2005;65:10154–8. 64. Lamfers ML, Fulci G, Gianni D , et al. Cyclophosphamide increases transgene e xpression mediated b y an oncol ytic adeno virus in glioma-bearing mice monitored b y bioluminescence imaging. Mol Ther 2006;14:779–88. 65. Guse K, Dias JD, Bauerschmitz GJ, et al. Luciferase imaging for evaluation of oncol ytic adeno virus replication in vivo . Gene Ther 2007;14:902–11. 66. Hakkarainen T, Hemminki A, Curiel DT, Wahlfors J. A conditionally replicative adeno virus that codes for a TK-GFP fusion protein (Ad5Δ24TK-GFP) for e valuation of the potenc y of oncol ytic virotherapy combined with molecular chemotherap y. Int J Mol Med 2006;18:751–9. 67. Le LP, Everts M, Dmitriev IP, et al. Fluorescently labeled adenovirus with pIX-EGFP for v ector detection. Mol Imaging 2004;3:105–16. 68. Meulenbroek RA, Sargent KL, Lunde J, et al. Use of adenovirus protein IX (pIX) to displa y large polypeptides on the virion–generation of fluorescent vir us through the incor poration of pIX-GFP. Mol Ther 2004;9:617–24. 69. Le LP, Ri vera AA, Glasgo w JN , et al. Infecti vity enhancement for adenoviral transduction of canine osteosarcoma cells. Gene Ther 2006;13:389–99. 70. Li J, Le L, Sib ley DA, et al. Genetic incor poration of HSV-1 thymidine kinase into the adeno virus protein IX for functional displa y on the virion. Virology 2005;338:247–58. 71. Matthews QL, Sib ley DA, Wu H, et al. Genetic incor poration of a herpes simple x vir us type 1 th ymidine kinase and f irefly luciferase fusion into the adeno virus protein IX for functional display on the virion. Mol Imaging 2006;5:510–9. 72. Leopold PL, Ferris B, Grinberg I, et al. Fluorescent virions: dynamic tracking of the pathway of adenoviral gene transfer vectors in living cells. Hum Gene Ther 1998;9:367–78. 73. Nakano MY, Greber UF. Quantitative microscopy of fluorescent adenovirus entry. J Struct Biol 2000;129:57–68. 74. Bernt KM, Ni S, Gaggar A, et al. The effect of sequestration by nontarget tissues on anti-tumor efficacy of systemically applied, conditionally replicating adeno virus v ectors. Mol Ther 2003;8: 746–55. 75. Miyazawa N, Leopold PL, Hackett NR, et al. Fiber swap between adenovirus subgroups B and C alters intracellular traf ficking of adenovirus gene transfer vectors. J Virol 1999;73:6056–65. 76. Kelkar SA, Pf ister KK, Cr ystal RG, Leopold PL. Cytoplasmic dynein mediates adeno virus binding to microtubules. J Virol 2004;78:10122–32. 77. Glotzer JB , Michou AI, Bak er A, et al. Microtubule-independent motility and nuclear tar geting of adenoviruses with fluorescently labeled genomes. J Virol 2001;75:2421–34. 78. Le LP, Le HN , Nelson AR, et al. Core labeling of adeno virus with EGFP. Virology 2006;351:291–302. 79. Luker GD, Bardill JP, Prior JL, et al. Nonin vasive bioluminescence imaging of her pes simplex vir us type 1 infection and therap y in living mice. J Virol 2002;76:12149–61. 80. Luker KE, Schultz T, Romine J , et al. Transgenic repor ter mouse for bioluminescence imaging of herpes simplex virus 1 infection in living mice. Virology 2006;347:286–95. 81. Luker KE, Hutchens M, Schultz T, et al. Bioluminescence imaging of vaccinia vir us: ef fects of interferon on viral replication and spread. Virology 2005;341:284–300. 82. Deroose CM, Reumers V, Gijsbers R, et al. Noninvasive monitoring of long-term lenti viral v ector-mediated gene e xpression in rodent brain with bioluminescence imaging. Mol Ther 2006;14:423–31. 83. Mancuso K, Hendrickson AE, Connor TB Jr, et al. Recombinant adenoassociated virus targets passenger gene e xpression to cones in primate retina. J Opt Soc Am A Opt Image Sci Vis 2007;24:1411–6.

84. Greenberg KP , Geller SF , Schaf fer D V, Flanner y JG. Targeted transgene expression in muller glia of normal and diseased retinas using lenti viral v ectors. In vest Ophthalmol Vis Sci 2007;48: 1844–52. 85. Chin FT, Namavari M, Le vi J, et al. Semiautomated radiosynthesis and biological evaluation of [(18)F]FEAU: a novel PET imaging agent for HSV1-tk/sr39tk repor ter gene expression. Mol Imaging Biol 2008;10:82–91. 86. Kim YJ, Dubey P, Ray P, et al. Multimodality imaging of lymphocytic migration using lenti viral-based transduction of a tri-fusion reporter gene. Mol Imaging Biol 2004;6:331–40. 87. Bennett JJ, Tjuvajev J, Johnson P, et al. P ositron emission tomography imaging for herpes virus infection: implications for oncolytic viral treatments of cancer. Nat Med 2001;7:859–63. 88. Adusumilli PS, Stiles BM, Chan MK, et al. Real-time diagnostic imaging of tumors and metastases b y use of a replication-competent her pes v ector to f acilitate minimall y in vasive oncolo gical surgery. FASEB J 2006;20:726–8. 89. Adusumilli PS, Eisenberg DP, Stiles BM, et al. Intraoperative localization of lymph node metastases with a replication-competent her pes simplex virus. J Thorac Cardiovasc Surg 2006;132: 1179–88. 90. Yamamoto S, Deckter LA, Kasai K, et al. Imaging immediate-earl y and strict-late promoter acti vity during oncol ytic her pes simplex virus type 1 infection and replication in tumors. Gene Ther 2006;13:1731–6. 91. Räty JK, Liimatainen T, Wirth T, et al. Magnetic resonance imaging of viral par ticle biodistribution in vivo . Gene Ther. 2006;13(20):1440-6. 92. Wang W, Tai CK, K ershaw AD, et al. Use of replication-competent retroviral v ectors in an immunocompetent intracranial glioma model. Neurosurg Focus. 2006;20(4):E25. 93. Chen F, Lu JR, Binder BJ , et al. Application of digital image anal ysis and flow cytometry to enumerate marine vir uses stained with SYBR gold. Appl Environ Microbiol 2001;67:539–45. 94. Rocheleau JV, P etersen NO . The Sendai vir us membrane fusion mechanism studied using image cor relation spectroscopy. Eur J Biochem 2001;268:2924–30. 95. Geada MM, Galindo I, Lorenzo MM, et al. Mo vements of v accinia virus intracellular enveloped virions with GFP tagged to the F13L envelope protein. J Gen Virol 2001;82:2747–60. 96. Ward BM, Moss B . Visualization of intracellular mo vement of vaccinia vir us virions containing a g reen fluorescent protein-B5R membrane protein chimera. J Virol 2001; 75:4802–13. 97. Rodger G, Smith GL. Replacing the SCR domains of v accinia virus protein B5R with EGFP causes a reduction in plaque size and actin tail formation but enveloped virions are still transpor ted to the cell surface. J Gen Virol 2002;83:323–32. 98. Taubenberger JK, Reid AH, Krafft AE, et al. Initial genetic characterization of the 1918 “Spanish” influenza vir us. Science 1997;275:1793–6. 99. Lakadamyali M, Rust MJ, Babcock HP, Zhuang X. Visualizing infection of indi vidual influenza vir uses. Proc Natl Acad Sci U S A 2003;100:9280–5. 100. Rust MJ, Lakadamyali M, Zhang F , Zhuang X. Assembly of endocytic machinery around indi vidual influenza vir uses during viral entry. Nat Struct Mol Biol 2004;11:567–73. 101. Sakai T, Ohuchi M, Imai M, et al. Dual w avelength imaging allows analysis of membrane fusion of influenza vir us inside cells. J Virol 2006;80:2013–8. 102. Lampe M, Briggs J A, Endress T, et al. Doub le-labelled HIV-1 particles for study of vir us-cell interaction. Virology 2007;360:92–104. 103. Brandenburg B, Lee LY, Lakadamyali M, et al. Imaging polio virus entry in live cells. PLoS Biol 2007;5:1643–1555 104. Levenson RM. Spectral imaging perspecti ve on c ytomics. Cytometry A 2006;69:592–600.

43 DEVELOPING DIAGNOSTIC AND THERAPEUTIC VIRAL VECTORS KHALID SHAH, PHD

Viral v ectors are essential to de velop no vel and ef ficient gene and cell-based therapies. In recent y ears, man y advances have been made in designing no vel viral v ectors, primarily including adenovirus, adeno-associated virus, herpes simple x vir us, and retro virus. These v ectors dif fer in their suitability for different applications, depending on factors such as size of transgene, route of deli very, tropism, duration/regulation of gene e xpression, and possib le side effects. One of the most critical issues for ensuring success of viral vector therapy is the development of technology for noninvasive monitoring of the location, magnitude and duration of vector-mediated gene expression and the distribution and targeting of vector particles in vivo. A variety of factors play a role in deter mining the choice of specif ic v ector system in molecular imaging, some of w hich include the imaging requirements (single or repeated), intended use (animal or human), and spatial requirements (or gans vs cellular resolution and depth). This chapter pro vides descriptions of the de velopment of different vector systems and their applications to imaging dif ferent parameters of vector-mediated gene expression in vitro and in vivo.

genome of the packaging cell line or incor porated in a plasmid. Basic studies in vir us and v ector biology have shown the essential cis- and trans-acting elements that provide the viral genome with the ability to persist in the nucleus of the host cell. Although some of these elements allow episomal persistence of vir us genomes, others induce specif ic or random inte grations into host cell chromosomes, such as those from adeno-associated virus (AAV) and retro virus (R V). Although the deli very of transgenes via recombinant vir uses is highl y ef ficient, remaining obstacles include tar geting, control of transgene fate after delivery, stable and regulated gene expression, and possib le adverse effects on the cell. Though a number of viruses have been developed, interest has centered on four types: R V (including lenti viruses [LV]), adenovirus (AV), AAV, and her pes simplex virus (HSV)1. The characteristics of dif ferent viral v ectors are discussed in detail, and they are also summarized in Table 1.

AV VIRAL VECTORS Viruses are natural vehicles for transfer of foreign nucleic acid into cells and are, therefore, optimal tools for gene therapy.1 Although some vir uses e xhibit an e xtremely restricted host range, others are capab le of infecting a wide v ariety of cell types. They ha ve the ability to influence cellular gene e xpression, modulate apoptosis pathways, and escape the immune response of the organism in a multitude of w ays. Recombinant viral vectors can be engineered b y replacing genes that are needed for the replication phase of their life c ycle (the nonessential genes) with foreign genes of interest. To produce such recombinant viral v ectors, the nonessential genes are pro vided in trans, either inte grated into the

The AV is a nonen veloped par ticle, containing linear double-stranded DN A (dsDN A). The wild type AV genome is approximately 35 kb of which up to 30 kb can be replaced with foreign DN A.2,3 There are four earl y transcriptional units (E1, E2, E3 and E4), which have regulatory functions and a late transcript, w hich codes for structural proteins. After the onset of DN A replication, the major late promoter dri ves much of the viral transcription. Viral-encoded functions can be separated into cis and trans elements. Whereas the cis genes, such as those responsible for the origin of replication or the packaging signal that condenses the DN A (protein IX, ψ), must generall y be car ried b y the vir us itself, the trans genes can be complemented or replaced with inser ted 689

Recombinant and “gutless” (dsDNA)

Parvovirus (ssDNA)

Recombinant Herpes simplex type I II (dsDNA)

Retrovirus (RNA)

Alpha virus (RNA)

Picarnovirus (RNA)

Adeno-associated Virus

HSV-1

Lentivirus/MoMLV

Sindbis

Poliovirus replicon

Virion/Vector Type

Adenovirus

Vector No

No

No Yes

No

No

20–30 nm, 1010–1013

120–300 nm, 1011 100 nm, 106–109

60–65 nm, 107

30 nm, 109

Chromosomal Integration

100 nm, 1010–1012

Particle Size and Titers (tu/mL)

6; Moderate

6; Moderate

9; Moderate (1010)

~30–50; Moderate (109)

4.9; High (1012)

8–10; High (1012)

Cloning Capacity (kb) and Vector Yield

Table 1.

Infects motor neurons very efficiently

express their gene products in the cytoplasm without the need for nuclear entry. Robust gene expression within 8 hrs of infection

Can integrate into the genome of both dividing and non-dividing cells

Transduces both dividing and non-dividing cells,

Stably retained in the and nucleus of both dividing non-dividing cells, low immunogenicity

Transduces both dividing and non-dividing cells;

Advantages

Small transgene capacity

Toxic, small transgene capacity, kills host cells

Risk of activating a proto-oncogene or inactivating a critical gene; limited cloning capacity

Immunogenic, some toxicity

Limited transgene capacity

Immunogenic, instability of transgene expression, can be toxic

Disadvantages

Developing Diagnostic and Therapeutic Viral Vectors

“foreign” DNA. Progenitor vectors have either the E1 or the E3 gene inactivated, with the missing gene being supplied in trans by either a helper vir us or plasmid , integrated into a helper cell genome (human fetal kidne y cells, line 293 4). The majority of AV v ectors are de veloped based on serotypes 2 and 5 b y replacing either the E1 or the E3 gene with a transgene, resulting in the loss of viral replicability. These recombinant viruses replicate only in cells that e xpress the E1 or E3 gene products at very high concentrations, creating a highly efficient, controllable viral vector system for therapeutic applications. The AV has se veral features that mak e them w ell suited for gene deli very. They are ubiquitous; AVs ha ve been isolated from a lar ge number of dif ferent species, and more than 100 dif ferent serotypes ha ve been repor ted. Most adults have been exposed to the AV serotypes most commonly used in gene therap y (serotypes 2 and 5). In addition, the y have the ability to rapidl y infect a broad range of dividing and nondividing human cells, have low pathogenicity in humans, can accommodate relati vely large se gments of DN A (up to 7.5 kb), transduce these transgenes in nonproliferating cells and are relati vely easy to manipulate using recombinant DNA techniques.

AAV AAVs are 4.7 kb single-stranded DN A vir uses that depend on helper vir uses such as AV for replication. They are capab le of infecting both di viding and nondividing cells; and in the absence of a helper vir us, they inte grate into a specif ic point of the host genome (19q 13-qter) at a high frequenc y.5 The genome of AAV encodes two proteins, namely Rep, which is a nonstr uctural protein in volved in rescue and replication of the virus, and Cap, w hich for ms icosahedral capsid within which the replicated genome is packaged. In addition to the Rep and Cap genes, the AAV genome also contains two inverted terminal repeats (ITRs) on either end of the genome that are ~145 bases each in length. The ITRs are the sole elements required for rescue, replication, packaging, and inte gration of AAV.6,7 Five primate AAV serotypes ha ve been characterized and are designated AAV types 1 to 5 (AAV1 to AAV5).8,9 Serological studies suggest that AAV1 to AAV3 and AAV5 frequently infect human populations, w hereas AAV4 infects monkeys.10 The advantages of AAV vectors include par ticle stability, the ability to inte grate into host chromosomes, and the potential for transducing nor mal cells in vitro and in vivo. The initial cloning of the AAV genome into a plasmid v ector f acilitated a wide range of molecular manipulations that led to the understanding of se veral

691

key e vents in AAV biolo gy (Samulski 1982). Because the AAV v ector genome lacks viral coding sequences, the vector itself has not been associated with to xicity or inflammatory response (e xcept for the generation of neutralizing antibodies that may limit re-administration).

RV RVs are a class of enveloped viruses containing a singlestranded RNA molecule as the genome. The viral genome is approximately 10 kb, containing at least three genes: gag (coding for core proteins), pol (coding for re verse transcriptase), and env (coding for the viral envelope protein). At each end of the genome are long-ter minal repeats (LTRs), which include promoter/enhancer regions and sequences in volved with inte gration. In addition, there are sequences required for packaging the viral DNA (psi) and RN A splice sites in the env gene. Lentiviruses (LVs) are a subclass of R Vs that are considerab ly more complicated than simple R Vs, containing an additional six proteins, tat, re v, vpr , vpu, nef, and vif. The L V genome derived from immunodeficiency viruses, such as human immunodeficiency virus-1 (HIV-1), has been split into multiple fragments to minimize the potential formation of replication-competent vir uses. These components are brok en into the follo wing cate gories: (1) transfer or inte grating vector, (2) str uctural and packaging plasmids, and (3) envelope plasmids. During LV vector production, there are tw o main re gions within the transfer plasmid that e xpress the viral RN A genome following transfection: (1) a multiple cloning site for the insertion of various expression cassettes and (2) flanking LTRs that have several distinct functions. The RNA vector genome is re verse transcribed and inte grated into the genomic DNA as a provirus using viral proteins obtained from the str uctural and packaging plasmids in trans during v ector production. Two impor tant cis-acting DN A elements, the central polypurine tract sequence (cppt) and the woodchuck postregulatory element (WPRE), are also included in the transfer (inte grating) v ector to enhance the transduction ef ficiency and transcript stability , respectively.11,12 The cppt is a small DNA fragment found in the pol gene of HIV that is usuall y cloned 5 ʹ′ to the internal promoter region, whereas the WPRE is cloned 3ʹ′ to the inserted transgene so that it is in close proximity to the poly(A) signal in the 3 ʹ′-LTR. Another important feature in the lentiviral transfer plasmid is a 400-bp deletion in the U3 re gion of the 3 ʹ′-LTR, which debilitates the 5 ʹ′LTR RN A pol II promoter acti vity follo wing inte gration.12,13 LVs have the unique ability among RVs of being able to infect nonc ycling cells. As the par ticles are often

692

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

pseudotyped with the envelope of the vesicular stomatitis virus (VSV), the v ector can ser ve to introduce genes into a broad range of tissues and can be used in vivo. Furthermore, it has been sho wn that in vi vo expression is sustained for se veral months without detectab le pathology. Most studies using lentiviral vectors have used powerful constitutive promoters such as PGK1, CMV, CAG, and EF1α.14 Lentiviral v ectors are par ticularly amenab le to developing regulated systems, as their copy number in target cells is reasonab ly predictab le and stab le without silencing. This is in contrast with AAV vectors, which have limited car go capacity and ma y produce such high cop y numbers that regulators are potentially diluted out. The LV vector-based regulated systems use re gulator proteins, for example, the ones deri ved from bacterial operons such as the Tet repressor or engineered eukar yotic transcription factors.15–17 These systems are now being refined to ensure that the basal e xpression is lo w and once induced the expression levels are high.16 As all extraneous viral coding regions are assiduously removed from lentiviral vectors, it is counter productive to introduce foreign re gulators that may be immuno genic or ma y influence the cell biology. Current work is therefore aimed at de-immunizing re gulator proteins or at developing systems that can be regulated by small molecule interactions with the pro viral DNA or transcripts.18

Herpes Simplex Virus (HSV)-1 HSV-1 is currently the most extensively engineered herpes virus for gene transfer. HSV has a large genome composed of 152 kb of linear dsDN A containing at least 84 almost entirely contiguous (unspliced) genes, appro ximately half of which are nonessential for vir us replication in cell culture. HSV-1 has many features that make it a suitable gene delivery vehicle: (1) the sequence of the HSV-1 genome is known, (2) more than 20% of the viral genes are nonessential for replication in cell culture and can, therefore, be replaced with foreign DN A, (3) HSV -1 has a wide host range that includes both di viding and nondi viding cells, and (4) during latenc y in neurons, the HSV -1 genome remains in a relati vely stab le state that suppor ts at least some transcriptional acti vity. Two fundamentally different HSV-1-based v ector systems ha ve been de veloped: (1) recombinant and (2) amplicon HSV. Recombinant HSV-1 vectors are created by replacing one or several virus genes with transgene sequences. Depending on the vir us genes that are replaced, recombinant HSV-1 vectors can be replication competent, conditional, or defecti ve. The choice of the replicative state of a v ector depends on the pur pose of

gene delivery and the target tissue, for example, replication defective for gene replacement therap y19,20 or replication competent/conditional for cancer gene therap y and v accination.21–24 HSV-1 amplicons have been derived from naturally occur ring defecti ve vir uses obser ved in HSV -1 stocks that have been serially passaged at high multiplicities of infection. 25 Amplicons contain less than 1% of the HSV-1 genome, in par ticular ori and pac, and, therefore, require HSV-1 helper functions for replication and packaging into virions. The basic structure of the HSV-1 amplicon has not changed o ver the past 20 y ears. However, safety issues have initiated the de velopment of innovative means to provide helper functions. These include the use of replication-conditional mutants of HSV -1, as opposed to wild type HSV-1, as helper vir uses26–29 and, more recentl y, the use of helper vir us-free packaging systems. 30 Amplicon vector stocks produced by using helper vir us-free packaging systems have titers of up to 10 7 transducing units (tu) per mL of cell culture medium (and up to 10 9 tu/mL after purification and concentration) and can ef ficiently transduce many cell types.

TARGETING VIRAL VECTORS The development of technologies that allow targeting of specific cells with viral vectors has progressed substantially. In recent years, several types of viral vectors have been used to date in 70% of gene therapy trials.31,32 Currently, novel w ays of impro ving viral v ector tar geting, focused mainl y on systemic tar geting, are being explored. One of the major obstacles, the f inal step of infecting the target cell, is among a number of obstacles that need to be o vercome in order for a systematicall y applied v ector to reach its tar get cells. To accomplish this, the viral v ector must displa y a suitab le ligand for binding to the tar get cell receptor . The natural tropism of some vir uses matches their v ector utility, as is the case for HSV , w hich can be used for neuronal gene delivery,33 but in man y cases, the v ector must be engineered to ha ve a ne w tropism. Most pro gress in v ector development has been achieved using AV, AAV, and vectors that are derived from RV, particularly LVs. The main approaches are discussed , also summarized in Table 2, and illustrated in Figure 1.

Vector Targeting Using Adaptors Adaptors are molecules with dual specif icities: one end of the adaptor binds the receptor on the tar get cell and the other binds the viral attachment protein. The advantages of using adaptors for viral tar geting are its g reat

Developing Diagnostic and Therapeutic Viral Vectors

693

Table 2. Approach

Principle

Advantages

Disadvantages

Adapter System Receptor ligand

A native viral receptor is fused to the targeting ligand

Easy preclinical testing

Correct folding of each new receptor–ligand pair must be determined

Bispecific antibody

Two antibodies are coupled, with the resulting molecule having specificity for the vector and target

Using existing reagents, antibody is easy to make; screening for different targets is readily possible

Binding affinity of the targeting complex to the vector can vary

Chemical linkage

Targeting moiety is bound to the vector by chemical means

A covalent bond is formed with the targeting complex, thus no adaptor dissociation from the vector

Technically more demanding than other adaptor systems (but scaleable for clinical applications)

Avidin–biotin

Biotin is coupled to the vector and then bound to the avidin–ligand complex

High-affinity binding of the targeting complex to the vector; allows easy vector purification

Some risk for toxicity in clinical applications (biotin from the circulation could be complexed)

Small targeting motifs

Small peptides are inserted into the capsid or viral attachment protein

Minimal disturbance of vector structure

Broadens tropism; limited number of available motifs; not applicable for all cell types

Serotype switching

Use of a different serotype from within the same virus family

Biological compatibility makes it feasible

Limited availability of serotypes; the precise cellular receptor is frequently unknown

Mosaic viral attachment proteins

Two viral attachment proteins with different properties are combinesd, allowing targeting, production or imaging in parallel

True multifunctionality in a virion can be achieved

Desired stoichiometry can be difficult to achieve

Ablation of native tropism

Mutation of the amino acids responsible for native tropism

Can be combined with other techniques

Can confound production in packaging cell line

Pseudotyping

Use of a viral attachment protein from a different virus strain or family

Technically easy when the biology is supportive or compatible

Limited availability of pseudotypes that fit the desired target cell; possi ble reduction of transfec tion efficiency (retrovirus)

Genetic systems

Adapted from Waehler R et al 32 with permission.

flexibility as different adaptors can readily be coupled to the same v ector and the f act that it does not require changes in v ector structure that could be detrimental to vector production or gene transfer .32 Most adaptors can achieve the two main goals of targeted delivery: ablating native tropism and confer ring a no vel tropism to ward the desired target. Adaptor systems have proved particularly useful for proof-of-principle preclinical studies,

allowing easy testing of se veral tar get receptors. 34 A number of different adaptor systems have been used: (1) receptor-ligand system, (2) chemically conjugated adaptors, (3) avidin–biotin system, and (4) monoclonal antibodies as adaptors (see Figure 1). Recent studies showed that a fusion of the extracellular domain of the avian sarcoma and leuk emia vir us (ASL V) retro viral receptor (tumor virus subgroup A receptor; TVA) to heregulin-β1

694

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

C

B

D

A

E

F

G

Figure 1. Different targeting strategies using viral vectors. Adaptors (A–D) An adenoviral vector is coupled with a receptor–ligand fusion (A); a biotin-acceptor peptide is integrated into the fiber knob, biotinylated and coupled to an avidin-containing ligand (B). An antibody-binding domain is genetically incorporated into the adeno-associated virus (AAV) capsid to couple a monoclonal antibody to the vector (C). A bispecific antibody is attached to the AAV capsid (D) Genetic incorporation of a targeting ligand (E–F) A single-chain antibody (single-chain variable fragment (scFv) against a particular antigen and a matrix metalloprotease (MMP) cleavage site are coupled to the viral envelope protein (Env). This allows binding to tumor cells that express the antigen, followed by cleavage of the matrix metalloprotease (MMP) cleavage site by tumor-secreted MMP (E). The incorporation of a small targeting ligand (eg, an RGD peptide) can be used to target a vector to integrin receptors (F). Pseudotyping (G). A retroviral (lentiviral) vector is pseudotyped with an envelope protein (Env) from a different virus. CAR, Coxsackie and adenovirus receptor. (Adapted from Waehler R et al 32 with permission from Mcmillian Publishers.)

successfully tar geted the v ector to cells e xpressing heregulin receptors, 35 thus, providing a potential therapeutic strategy for the treatment of various malignancies. Adaptors that incor porate the ASLV receptor as the virus-binding moiety have been explored in particular as they can trigger confor mational changes in the ASLV envelope gl ycoprotein that are required for membrane fusion and vir us entr y. P olyethylene gl ycol (PEG) and PEG-derived polymers have been used to couple AV vectors to ligands such as f ibroblast growth factor-2, to target ovarian cancer cells, 36 and arginine–glycine–aspartic (RGD) peptide and E-selectin antibody to tar get endothelial cells.37 PEGylation has the potential to shield the vector from the innate immune system in vivo,38 and it allo ws infection in the presence of AV antibodies, 39 which might enab le repeated v ector application. The high-affinity binding between avidin and biotin provides an adaptor strate gy that has been e xploited for se veral viral v ectors and for ms a v aluable basis for tar geting studies. The feasibility of this approach for an enveloped virus was shown when an ecotropic RV was coated with a biotinylated antienvelope antibody and strepta vidin to redirect its attachment to human major histocompatibility complex class I (MHCI). AAV vectors can be biotinylated in a similar w ay to AV vectors, and this system is now being used as a platform for purification and targeting of this vector type.40 The avidin–biotin system seems

to be suitab le for an y v ector type that allo ws incorporation of a biotin acceptor peptide (B AP) or that can be chemically biotinylated and is expected to be used in man y more applications in the future. Viral v ectors have been geneticall y modif ied to allo w coupling of monoclonal antibodies; for example, a region of a bacterial immunoglobulin (Ig)-binding protein (usuall y the Z domain of the staphylococcus protein A) can be inserted into the viral-attachment proteins of v arious v ector systems.41–44 In this w ay, the unmodif ied antibody can work lik e an adaptor , bridging the Ig-binding domain that is incor porated in the v ector to the tar get receptor through its antibody specif icity. This approach has been successfully used both in vitro and in vi vo.45 A limited use of v arious adaptors in vi vo has limited our understanding of their potential utility . The systems based on the B AP ha ve g reat fle xibility and are e xpected to be widely applied, including clinics. The other systems are likely to have their greatest potential in preclinical proofof-principle studies.

Genetically Engineered Targeted Ligands Genetic incorporation of targeting ligands into the capsid or the en velope protein yields a single virion molecule that reco gnizes the tar get cell. The use of such single component systems pro vide homo genous retar geted

Developing Diagnostic and Therapeutic Viral Vectors

vector particles and also f acilitates high-titer production by eliminating the need to create a separate adaptor molecule. The genetic incor poration of pol ypeptide ligands into viral surf ace proteins gi ves vectors new and highl y specific tropisms for cells e xpressing the tar get antigen. The display of a single-chain antihapten antibody fused near the N terminus of the murine leukemia virus (MLV) surface (SU) component of the envelope glycoprotein on the surf ace of an en veloped vir us was shown to bind to hapten via the displa yed antibody .46 The single-chain antibody approach has been applied to se veral v ectors AAV,47 AV,48 RV,49 and HSV,50 emphasizing the versatility of this approach. It has been sho wn that tar geted vir us attachment does not necessarily lead to targeted entry. For example, in the case of R V envelope glycoproteins, displayed tar geting ligands usuall y impede infecti vity and lead to direction of the vir us into a nonfunctional entr y pathway, steric b locking of its natural receptor interactions, or pre vention of confor mational changes that are required for effective fusion.51 To circumvent these problems, a concept of in verse tar geting, w hereby the viral envelope glycoprotein is modif ied to selecti vely destroy its infecti vity for cells e xpressing a tar geted receptor ,52 has emerged. For example, amphotropic retroviral vectors that display epidermal growth factor (EGF) or stem-cell factor (SCF) are selecti vely noninfectious for cells that express the cognate receptors (EGFR or KIT), but remain fully infectious for other (receptor ne gative) human cells,52,53 thus providing a useful way to detarget the liver (EGFR positi ve) or mar row stem cells (KIT positi ve). Several animal studies ha ve sho wn the feasibility of in vivo transductional targeting, using retroviral and lentiviral vectors with genetically incorporated polypeptide ligands. In one study , the li ver was successfully detargeted by pseudotyping LV with an amphotropic MLV envelope glycoprotein that displa yed EGF as an N-ter minal fusion.54 The reduced hepatic transduction in this case was resulted from in verse targeting, which was possible because EGF receptors w ere e xpressed abundantl y on hepatocytes. The RV Rexin-G, which expresses a cytocidal dominant-negative form of cyclin G1, has been tested in the clinic. This v ector displa ys the collagen-binding portion of v on Willebrand factor (vWF) incor porated in its Env protein, allo wing preferential v ector deli very to the tumor site w here angio genesis and collagen matrix exposure occur.55 Rexin-G is targeted to the extracellular matrix of tumor tissue 56 and has been tested for its antitumor activities in three clinical studies.57 The limitations that are encountered during R V or L V v ector tar geting using engineered pol ypeptide ligands do not necessaril y apply for other enveloped viruses.

695

Small peptide motifs are generall y 3 to 20 amino acids in size and are capab le of changing the tar geting characteristics of a v ector dramatically. Two small peptide tar geting moieties that ha ve been commonl y used are (1) the peptides with the RGD motif and (2) pol ylysine (pK7) peptide. Small peptides containing an RGD motif, which targets the vectors to integrins, have facilitated targeting of the vasculature and of tumor cells (see Figure 1). Such targeting has been achieved for AAV (in vitro),58 AV (e x vi vo in tissue-slice assa ys59 and in vivo60), RV (in vitro)61 and for a phage–AAV hybrid vector (in vivo)62 pK7 peptide that targets vectors to heparan sulphates, w hich are o verexpressed in a number of malignancies63 and other pathologies. AV vectors carrying pK7 in their f iber knob sho wed an increased transduction of v arious Co xsackie and adeno virus receptor (CAR)-deficient tar gets, such as sk eletal muscle in vivo.64 In addition, RGD and pK7 ha ve recentl y been used together in the AV capsid to improve the efficiency of vector delivery in a murine model of cancer.65 Generally, both of these modifications broaden vector tropism, which mak es them especiall y useful for local administrations. Although small peptide motifs are v ersatile and can be used to tar get viral v ectors to se veral cell types, other cell types cannot be tar geted in this w ay and require different targeting approaches. To a void the transduction of nontar get tissue, the native v ector tropism of viral v ectors might need to be ablated for systemic applications. In general, each v ector requires specif ic changes in the viral attachment protein to ablate its native tropism. These modifications can then be used in a wide range of applications and can be done by the addition of a tar geting ligand that reduces the native tropism suf ficiently. For e xample, the incor poration of peptides that tar get human v enous endothelial cells into AAV capsids resulted in signif icantly lo wer hepatocyte transduction, b ut g reatly increased v enouscell transduction.66

Vector Targeting by Pseudotyping Pseudotyping involves transferring viral attachment proteins either between strains within a family of viruses or between virus f amilies. This can be achie ved by co-transfection of plasmids, with one encoding the attachment protein to be pseudotyped and with separate plasmids encoding all other vector components. This approach is used routinel y to pseudotype AAV and R V or LV vectors. Alternatively, the viral attachment protein can be e xpressed in trans from the production cell line or genetically incorporated into the viral genome, an approach that is par ticularly well suited to the

696

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

generation of AV pseudotypes. Pseudotyping has been used most e xtensively to modulate the host-cell tropisms of enveloped RV (including LV) vectors, and the most widel y used retroviral vector pseudotypes are those that incorporate the attachment glycoprotein of the VSV-G.67 The incorporation of VSV-G allows both the production of high-titer v ector stocks and confers a broad host range. The concept of pseudotyping has recentl y been e xtended to the incor poration of host-cell viral receptors (CD4 and CXCR4 or CCR5) into viral en velopes for tar geted entr y into HIV -infected cells. The feasibility of this approach has been sho wn b y using a replication-competent rhabdo virus (VSV) and nonreplicating lentiviral or MLV vectors to mediate the targeted destruction of HIV-infected cells by redirecting them to use the HIV-derived glycoprotein HIVgp120 as a receptor .68–70 Pseudotyping has also been used for nonen veloped vectors, including AAV and AV, by substituting coat proteins with homologous proteins of other related serotypes, gi ving rise to a new tropism without changing the rest of the genome, and thus, enab ling the use of estab lished cloning systems that have been developed for the previous serotype. The disadv antage of using eukar yotic vir uses is that their nati ve tropism must be ab lated to achie ve liganddirected targeting upon systemic administration. Prokar yotic viruses infect mammalian cells with a lo w efficiency but can be adapted to bind mammalian receptors by genetically engineering a eukar yotic ligand into their capsid. In one study, a hybrid vector that comprised an AAV cassette inserted in the phage genome and that targeted αv integrins through an RGD-4C peptide motif that w as displayed on the phage capsid w as constr ucted.62 Following systemic application in nude mice, this v ector showed specif ic targeting to tumors derived from injection of human prostate cancer cells, and tumor shrinkage w as achie ved w hen a therapeutic transgene was introduced into the vector. Monitoring the biodistribution of the v ectors in this study w as facilitated using in vivo vector imaging, which is discussed in the below section. Remarkably, the chimeric vector was also highly effective for antitumor therapy in the context of immunocompetent mice, e ven following prior phage v accination and high antiphage antibody titers. A similar strategy could be used for other v ectors with doub le-stranded genomes, such as AV.

IMAGING VECTOR TARGETING AND GENE EXPRESSION Different classes of mark er genes (encoding proteins, enzymes, or cell surface receptors) that can be detected by different imaging modalities ha ve been described in previous studies. 71,72 Various e xisting imaging technolo gies

differ in a variety of aspects: spatial and temporal resolution, depth of penetration, ener gy e xpended for image generation, a vailability of the injectab le probes, and the detection threshold of probes for a gi ven technology.73 Molecular imaging has been sho wn to g reatly facilitate the e valuation of tar geting approaches. F or e xample, a L V v ector pseudotyped with the en velope (En v) protein of the Sindbis vir us was successfully retargeted to metastatic melanoma in vi vo.45 Specifically, Sindbis Env w as mutated to ab late the li ver or spleen tropism, and the ZZ domain (Ig-binding domain) of proteinA was genetically fused to this protein. ZZ Sindbis-pseudotyped vir us v ectors e xpressing f irefly luciferase (Fluc) were used to quantify the le vel and specif icity of targeting in transduced cells in the or gans of li ve mice. Mice injected systemicall y with a pseudotyped L V sho wed strong liver and spleen tropism owing to domains within the Sindbis Env protein that target the vector to these tissues. Mutating these domains largely ablated this natural tropism. The use of a tumor -cell-specific antibody in mice bearing lung tumor metastases and injected with the LV vectors lead to transduction of tumor cells. Mice injected with a nonspecific control antibody and LV vectors did not sho w an y transduction of lung metastases. This study shows how combining several targeting techniques (pseudotyping, ab lation of nati ve tropism, and adaptor coupling) can enhance tar geted gene transfer after systemic application and ho w imaging is essential for the anal ysis. Recentl y, imaging ligands ha ve been incorporated into the capsid of AV v ectors. Green and red fluorescent proteins 74–76 and an HSVtk–luciferase fusion protein 77 have been fused to the capsid protein pIX, allo wing multimodality imaging. The labeling of the capsid is promising in the conte xt of tracking the intracellular fate of vectors and for observing the spread of tar geted oncol ytic v ectors. Recentl y, magnetic resonance imaging (MRI) has been used for direct v ector imaging b y using an a vidin-coated baculo virus conjugated with biotin ylated super paramagnetic iron o xide particles.78 This technique should be easily applicable to other capsid-coated viruses. Imaging of gene e xpression has become essential for improving the diagnosis of disease and predicting its response to therap y.79–82 Gene e xpression imaging in vivo has been f acilitated b y the a vailability of a wide variety of viral v ectors and animal models, the ease of recombinant DN A technolo gy, and the impor tance to preclinical e valuation and clinical applications. Viral proteases, w hich are not ubiquitousl y e xpressed in mammalian cells, of fer a possibility to de velop such protease/substrate systems. In recent y ears, w e ha ve

Developing Diagnostic and Therapeutic Viral Vectors

developed near-infrared fluorescence (NIRF) probes for HIV-1 and HSV -1 proteases (PR) for imaging of gene delivery to tumors. We ha ve sho wn specif ic fluorescence acti vation of this HIV -1 NIRF probe in human Gli36 gliomas injected with viral v ector e xpressing HIV-1PR,83 showing that viral proteases can be imaged in live animals and can be used as transgene mark ers in tumor therapy in vivo. To develop regulatable therapeutic proteins w hose re gulation can be controlled b y the imageable protease, we have coupled the regulation of a therapeutic protein, tumor necrosis-f actor related apoptosis-inducing ligand (TRAIL) with the acti vation of HSV-1-specific NIRF probe using the HSV -1-specific protease.84 TRAIL is a type II transmembrane protein that induces apoptosis in tumor cells of di verse origins by binding to death-domain-containing receptors, DR4 (TRAIL-R1) and DR5 (TRAIL-R2), 85,86 on the tumor cell surf ace. We ha ve engineered a secretab le for m of TRAIL (S-TRAIL) and shown that this for m of TRAIL is acti vely secreted from the cells and allo ws a “bystander ef fect,” w hereby transduced cells cause death of sur rounding nontransduced tumor cells. 84,87,88 We ha ve de veloped means to control the secretion of TRAIL using a viral protease b y engineering endoplasmic reticulum (ER)-tar geted TRAIL and sho wn that it was inacti vely retained in the ER until selecti vely released by the HSV-1 viral protease. Expression of ERtargeted HSV-1 protease in the (ER)-tar geted TRAIL expressing cells resulted in the release of S-TRAIL from the ER and subsequent induction of apoptosis in glioma cells. The same HSV -1 protease w as used to monitor gene deli very in vi vo by systemic administration of HSV -1 protease specif ic NIRF probe acti vated by the protease (Figure 2).84 The mode of cancer therapy used in this study has tw o impor tant adv antages: (1) control of the conversion of S-TRAIL from a nonapoptotic resident of the ER to a apoptosis-inducing protein by selective protease activation and (2) in vi vo imaging of protease acti vity using a NIRF probe. Release of S-TRAIL in tumors can be controlled b y co-injection of viral vectors encoding ER-S-TRAIL and the viral protease and should be compatib le with clinical trials for accessible tumor foci. The luciferases from Renilla (Rluc) and firefly (Fluc) have different substrates, coelenterazine and D-luciferin, respectively, and can be imaged in tumors in the same living mouse with kinetics of light production being separable in time b y separate injections of these tw o substrates.89,90 Dual bioluminescence imaging has been used to monitor gene delivery via a therapeutic vector and to follow the ef fects of the therapeutic protein TRAIL in

697

gliomas.90 In a w ork co-ordinated in our laborator y, glioma cells stab ly expressing Fluc w ere implanted subcutaneously into nude mice, and the tumor g rowth w as monitored in vi vo o ver time b y luciferin administration and BLI. HSV amplicon v ectors bearing the genes for TRAIL and Rluc w ere injected directl y into these Flucpositive gliomas allowing super imposition of gene delivery to the tumor b y coelenterazine administration and BLI. This dual-imaging approach has direct applications in studying the deli very of gene therap y v ectors and simultaneously monitoring therapeutic effects in vivo. The th ymidine kinase (TK) enzyme acts to con vert this relati vely nonto xic substrate into compounds selectively to xic to di viding cells, resulting in tumor cell death. PET probes, like [18F]FHBG (9-[4-[18F]fluoro3-(hydroxymethyl)butyl]guanine) and [124I]FIAU (5-iodo2ʹ′-fluoro-2ʹ′-deoxy-1-β-D-arabino-furanosyl-uracil), which are specif ically phosphor ylated by the HSV-1 TK enzyme and trapped within the cell, ha ve been successfully used for the nonin vasive localization of R V,72 HSV,91,92 and AV93 vector-mediated HSV-1-TK expression in animal models. Fur ther, nonin vasive PET monitoring has been carried out in clinical gene therapy paradigms. In a preliminar y study , Jacobs and colleagues 91,92 showed that [ 124I]FIAU PET imaging w as useful in monitoring experimental HSV-1 tk suicide gene treatment of glioblastomas. [ 124I]FIAU imaging sho wed the location, magnitude, and e xtent of v ector-mediated HSV -1-tk gene expression delivered by direct convection-enhanced intratumoral infusion of cationic liposomes in a phase I/II clinical trial for recur rent glioblastoma in patients. Although still in v alidation, it appears that the PET reporter-gene imaging will be able to provide supplementary gene delivery information in gene therap y trials. Detection of aberrant gene e xpression in unintended sites will be particularly helpful in e valuating safety features of the vector. Transplantation of geneticall y manipulated cells of human origin to the central ner vous system of fers immense potential for the treatment of se veral neurological disorders. Monitoring the expression levels of transgenes and following cells at a cellular resolution in vivo are critical in assessing the efficacy of such therapies in vivo. We have engineered lentiviral vectors bearing fusions betw een fluorescent and bioluminescent marker proteins, and w e used bioluminescence imaging and intra vital microscop y to study the f ate of human neural stem cells (hNSCs) in relation to gliomas in vivo. We used g reen fluorescent protein (GFP)-Rluc e xpressing malignant human glioma model and implanted hNSC e xpressing Fluc-DsRed2 intracraniall y. Intravital

698

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

C

B Figure 2. Regulation of apoptosis and in vivo imaging of herpes simplex virus (HSV)-1 proteases (PR) activity. A, Schematic view of apoptosis and imaging after transduction of cells with vector encoding ER-S-TRAIL followed by vector encoding ER-HSV-1PR and incubation with HSV-1 protease specific NIRF probe. B, Gli36 glioma cells were left uninfected or infected with S-TRAIL, ER-S-TRAIL or co-infected with ER-S-TRAIL and ER-HSV PR amplicon vectors, and cell viability was assessed 24 h later. C, Mice were implanted with Gli36 glioma cells on both sides in the upper, lateral abdomen, and 7 days later, tumors were injected with 20 µL of 3.5 × 108 transducing units (tu)/mL of ER-HSV-1PR amplicon vector or HGCX control vector; 36 h later mice were administered IV with 2.5 nanomoles of HSV-1PR-NIRF probe and imaged 24 h later using a charge-coupled device camera (Adapted with permission from Shah K et al.84)

microscopy sho wed that transduced hNSC mig rates extensively toward and into glioma tumors and sur vive longer in mice with gliomas than in nor mal brain (Figure 3). Similar studies, using primar y neural stem cell (NSC) transduced with lenti viral v ectors bearing both luciferase GFP , ha ve been perfor med to nonin vasively assess the sur vival and residence time of transplanted NSC in the spinal cord injur y models in li ving animals.94 Recently, a triple-fusion repor ter system has been used for multimodal imaging to monitor human mesenchymal stem cell (hMSC) transplants. hMSCs were transduced with a triple-fusion repor ter, fluc–mrfp–ttk (encoding f irefly luciferase, monomeric red fluorescent protein [mrfp], and truncated HSV 1 sr39 thymidine kinase [ttk]) b y use of a lenti viral vector, and their potential to dif ferentiate into bone, car tilage, and

fat was assessed b y fluorescence, bioluminescence, and small-animal PET imaging. 95 These studies sho w the combined application of different imaging modalities in evaluating the fate of NSC in real time.

CONCLUSIONS AND PERSPECTIVES The viral v ectors discussed abo ve are not inclusi ve but represent those used in cur rent clinical trials or under advanced preclinical de velopment. No single v ector system is likely to be optimal for all the potential gene therapy applications. Ho wever, for a specif ic application, a “perfect” v ector will be administered b y nonin vasive delivery routes, tar geted to the desired number of cells within target tissue, and regulated to express a therapeutic amount of transgene product for a def ined length of time.

Developing Diagnostic and Therapeutic Viral Vectors

699

A

Figure 3. Expression of bi-modal imaging transgenes in neural stem cell (NSC) and tumor cells using lentiviral vectors. A, selfinactivating lentiviral system based on HIV-1 (CS-CGW) was used to construct vectors: fusion between GFP and Fluc, GFP and Rluc, Fluc and DsRed2, and Rluc and DsRed2 under the CMV promoter (B–E) Bioluminescence imaging of mice implanted with human neural stem cell (hNSC) transduced with LV bearing GFP-Fluc in mice with established gliomas expressing Rluc-DsRed2. Fluc images of mice on day 3 (B), day 7 (C), and day 10 (D), and Rluc image on day 10 (E) are shown. F–H, Intravital microscopy of Fluc-DsRed2 hNSC implanted in mice with established GFP-Rluc gliomas were imaged by intravital microscopy on day 3 (F), 7 (G), and 10 (H) after hNSC implantation. ×30 original magnification. Mice brains were sectioned, and confocal microscopy was performed. Fluorescent image showing hNSC (red) infiltrating the tumor (green) (I); 40 × original magnification. Representative images of brain sections immunostained for nestin (J) and Ki67 (K). (Adapted from Shah et al 2007).

700

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Though w e are lik ely to see fur ther successes of viral vector therapy in the near future, the tr ue fruition of these therapies cannot be realized until the cur rent vectors are perfected or ne w v ectors de veloped with proper ties described above. Furthermore, noninvasive monitoring of the cor relation between primar y transduction and therapeutic efficiency of viral v ectors will be highl y informative and can widely influence the development of safe and efficient uses of viral vectors in clinical settings.

REFERENCES 1. Kay MA, Glorioso JC, Naldini L. Viral vectors for gene therap y: the art of tur ning infectious agents into v ehicles of therapeutics. Nat Med 2001;7:33–40. 2. Verma IM, Somia N . Gene therap y—promises, prob lems and prospects. Nature 1997;389:239–42. 3. Smith AE. Viral v ectors in gene therap y. Annu Re v Microbiol 1995;49:807–38. 4. Graham FL, Smile y J , Russell WC, Nair n R. Characteristics of a human cell line transfor med b y DN A from human adeno virus type 5. J Gen Virol 1977;36:59–74. 5. Kotin RM, Siniscalco M, Samulski RJ, et al. Site-specif ic integration by adeno-associated vir us. Proc Natl Acad Sci USA 1990;87: 2211–5. 6. Lusby E, Fife KH, Berns KI. Nucleotide sequence of the inverted terminal repetition in adeno-associated vir us DN A. J Virol 1980; 34:402–9. 7. Srivastava A, Lusby EW, Berns KI. Nucleotide sequence and or ganization of the adeno-associated vir us 2 genome. J Virol 1983;45: 555–64. 8. Atchison R W, Casto BC, Hammon WM. Adenovirus-associated defective virus particles. Science 1965;149:754–6. 9. Parks WP, Melnick JL, Ronge y R, Ma yor HD . Ph ysical assa y and growth cycle studies of a defecti ve adeno-satellite vir us. J Virol 1967;1:171–80. 10. Georg-Fries B, Biederlack S, Wolf J, zur Hausen H. Analysis of proteins, helper dependence, and seroepidemiology of a new human parvovirus. Virology 1984;134:64–71. 11. Follenzi A, Ailles LE, Bakovic S, et al. Gene transfer by lentiviral vectors is limited b y nuclear translocation and rescued b y HIV-1 pol sequences. Nat Genet 2000;25:217–22. 12. Park F, Ohashi K, Ka y MA. Therapeutic levels of human f actor VIII and IX using HIV-1-based lentiviral vectors in mouse liver. Blood 2000;96:1173–6. 13. Zufferey R, Dull T, Mandel RJ, et al. Self-inactivating lentivirus vector for safe and ef ficient in vi vo gene deli very. J Virol 1998; 72:9873–80. 14. Ramezani A, Hawley TS, Hawley RG. Lentiviral vectors for enhanced gene expression in human hematopoietic cells. Mol Ther 2000;2: 458–69. 15. Galimi F, Saez E, Gall J , et al. De velopment of ecdysone-regulated lentiviral vectors. Mol Ther 2005;11:142–8. 16. Pluta K, Luce MJ, Bao L, et al. Tight control of transgene expression by lenti virus v ectors containing second-generation tetrac yclineresponsive promoters. J Gene Med 2005;7:803–17. 17. Sirin O , P ark F. Re gulating gene e xpression using self-inacti vating lentiviral v ectors containing the mifepristone-inducib le system. Gene 2003;323:67–77. 18. Suess B, Hanson S, Berens C, et al. Conditional gene e xpression by controlling translation with tetrac ycline-binding aptamers. Nucleic Acids Res 2003;31:1853–8.

19. Natsume A, Mata M, Wolfe D, et al. Bcl-2 and GDNF deli vered by HSV-mediated gene transfer after spinal root a vulsion provide a synergistic effect. J Neurotrauma 2002;19:61–8. 20. Glorioso JC, Fink DJ. Use of HSV v ectors to modify the ner vous system. Curr Opin Drug Discov Devel 2002;5:289–95. 21. Rampling R, Cruickshank G, Papanastassiou V, et al. Toxicity evaluation of replication-competent herpes simplex virus (ICP 34.5 null mutant 1716) in patients with recur rent malignant glioma. Gene Ther 2000;7:859–66. 22. Markert JM, P arker JN, Gillespie GY, Whitley RJ. Genetically engineered human her pes simple x vir us in the treatment of brain tumours. Herpes 2001;8:17–22. 23. Loudon PT, Blak eley DM, Boursnell ME, et al. Preclinical safety testing of DISC-hGMCSF to suppor t phase I clinical trials in cancer patients. J Gene Med 2001;3:458–67. 24. Herrlinger U, Jacobs A, Quinones A, et al. Helper vir us-free her pes simplex vir us type 1 amplicon v ectors for g ranulocytemacrophage colony-stimulating factor-enhanced vaccination therapy for experimental glioma. Hum Gene Ther 2000;11:1429–38. 25. Spaete RR, F renkel N . The her pes simple x vir us amplicon: a ne w eucaryotic defecti ve-virus cloning-amplifying v ector. Cell 1982;30:295–304. 26. Lim F, Har tley D , Star r P, et al. Generation of high-titer defecti ve HSV-1 v ectors using an IE 2 deletion mutant and quantitati ve study of e xpression in cultured cor tical cells. Biotechniques 1996;20:460–9. 27. Ho DY, Saydam TC, Fink SL, et al. Defective herpes simplex virus vectors e xpressing the rat brain glucose transpor ter protect cultured neurons from necrotic insults. J Neurochem 1995;65:842–50. 28. Geller AI, Break efield XO . A defecti ve HSV -1 v ector e xpresses Escherichia coli beta-galactosidase in cultured peripheral neurons. Science 1988;241:1667–9. 29. During MJ, Naegele JR, O’Malley KL, Geller AI. Long-term behavioral recovery in parkinsonian rats b y an HSV v ector expressing tyrosine hydroxylase. Science 1994;266:1399–403. 30. Saeki Y, Fraefel C, Ichikawa T, et al. Improved helper virus-free packaging system for HSV amplicon v ectors using an ICP27-deleted, oversized HSV-1 DNA in a bacterial ar tificial chromosome. Mol Ther 2001;3:591–601. 31. Edelstein ML, Abedi MR, Wixon J, Edelstein RM. Gene therapy clinical trials w orldwide 1989–2004-an o verview. J Gene Med 2004;6:597–602. 32. Waehler R, Russell SJ, Curiel DT. Engineering targeted viral vectors for gene therapy. Nat Rev Genet 2007;8:573–87. 33. Burton EA, F ink DJ , Glorioso JC. Replication-defecti ve genomic HSV gene therapy vectors: design, production and CNS applications. Curr Opin Mol Ther 2005;7:326–36. 34. Parrott MB, Adams KE, Mercier GT, et al. Metabolically biotinylated adenovirus for cell tar geting, ligand screening, and v ector purif ication. Mol Ther 2003;8:688–700. 35. Snitkovsky S, Young JA. Targeting retroviral vector infection to cells that e xpress here gulin receptors using a TVA-heregulin bridge protein. Virology 2002;292:150–5. 36. Lanciotti J , Song A, Doukas J , et al. Targeting adeno viral v ectors using heterofunctional polyethylene glycol FGF2 conjugates. Mol Ther 2003;8:99–107. 37. Ogawara K, Rots MG, Kok RJ, et al. A novel strategy to modify adenovirus tropism and enhance transgene deli very to activated vascular endothelial cells in vitro and in vi vo. Hum Gene Ther 2004;15:433–43. 38. Mok H, Palmer DJ, Ng P, Barry MA. Evaluation of polyethylene glycol modif ication of f irst-generation and helper -dependent adenoviral v ectors to reduce innate immune responses. Mol Ther 2005;11:66–79. 39. Eto Y, Gao JQ, Sekiguchi F, et al. PEGylated adenovirus vectors containing RGD peptides on the tip of PEG sho w high transduction

Developing Diagnostic and Therapeutic Viral Vectors

40.

41. 42.

43.

44.

45.

46. 47.

48.

49.

50.

51. 52.

53.

54.

55. 56.

57.

58.

59.

60.

61.

efficiency and antibody e vasion ability . J Gene Med 2005;7: 604–12. Arnold GS, Sasser AK, Stachler MD, Bartlett JS. Metabolic biotin ylation provides a unique platfor m for the purif ication and tar geting of multiple AAV vector serotypes. Mol Ther 2006;14:97–106. Gigout L, Rebollo P , Clement N , et al. Altering AAV tropism with mosaic viral capsids. Mol Ther 2005;11:856–65. Korokhov N, Mikheeva G, Krendelshchikov A, et al. Targeting of adenovirus via genetic modif ication of the viral capsid combined with a protein bridge. J Virol 2003;77:12931–40. Ohno K, Sa wai K, Iijima Y, et al. Cell-specif ic targeting of Sindbis virus vectors displaying IgG-binding domains of protein A. Nat Biotechnol 1997;15:763–7. Tai CK, Lo gg CR, P ark JM, et al. Antibody-mediated tar geting of replication-competent retro viral v ectors. Hum Gene Ther 2003;14:789–802. Morizono K, Xie Y, Ringpis GE, et al. Lentiviral vector retargeting to P-glycoprotein on metastatic melanoma through intra venous injection. Nat Med 2005;11:346–52. Russell SJ, Hawkins RE, Winter G. Retroviral vectors displaying functional antibody fragments. Nucleic Acids Res 1993;21:1081–5. Yang Q, Mamounas M, Yu G, et al. Development of novel cell surface CD34-targeted recombinant adenoassociated vir us v ectors for gene therapy. Hum Gene Ther 1998;9:1929–37. Hedley SJ, Auf der Maur A, Hohn S, et al. An adenovirus vector with a chimeric f iber incor porating stabilized single chain antibody achieves targeted gene delivery. Gene Ther 2006;13:88–94. Chowdhury S, Chester KA, Bridge water J, et al. Ef ficient retroviral vector tar geting of carcinoembr yonic antigen-positi ve tumors. Mol Ther 2004;9:85–92. Menotti L, Cerretani A, Campadelli-Fiume G. A herpes simplex virus recombinant that e xhibits a single-chain antibody to HER2/neu enters cells through the mammar y tumor receptor, independently of the gD receptors. J Virol 2006;80:5531–9. Larochelle A, Peng KW, Russell SJ. Lentiviral vector targeting. Curr Top Microbiol Immunol 2002;261:143–63. Fielding AK, Maurice M, Morling FJ , et al. In verse tar geting of retroviral vectors: selective gene transfer in a mix ed population of hematopoietic and nonhematopoietic cells. Blood 1998;91:1802–9. Cosset FL, Takeuchi Y, Battini JL, et al. High-titer packaging cells producing recombinant retro viruses resistant to human ser um. J Virol 1995;69:7430–6. Peng KW, Pham L, Ye H, et al. Or gan distribution of gene e xpression after intra venous infusion of tar geted and untar geted lentiviral vectors. Gene Ther 2001;8:1456–63. Morse M. Technology evaluation: Rexin-G, Epeius Biotechnolo gies. Curr Opin Mol Ther 2005;7:164–9. Gordon EM, Chen ZH, Liu L, et al. Systemic administration of a matrix-targeted retroviral v ector is ef ficacious for cancer gene therapy in mice. Hum Gene Ther 2001;12:193–204. Gordon EM, Lopez FF, Cornelio GH, et al. Pathotropic nanoparticles for cancer gene therap y Re xin-G IV: three-y ear clinical e xperience. Int J Oncol 2006;29:1053–64. Stachler MD , Bar tlett JS. Mosaic v ectors comprised of modif ied AAV1 capsid proteins for efficient vector purification and targeting to vascular endothelial cells. Gene Ther 2006;13:926–31. Stoff-Khalili MA, Stof f A, Rivera AA, et al. Gene transfer to carcinoma of the breast with fiber-modified adenoviral vectors in a tissue slice model system. Cancer Biol Ther 2005;4:1203–10. Wu H, Han T, Lam JT, et al. Preclinical evaluation of a class of infectivity-enhanced adenoviral vectors in ovarian cancer gene therapy. Gene Ther 2004;11:874–8. Gollan TJ, Green MR. Redirecting retro viral tropism b y insertion of short, nondisr uptive peptide ligands into en velope. J Virol 2002;76:3558–63.

701

62. Hajitou A, Trepel M, Lille y CE, et al. A h ybrid v ector for liganddirected tumor tar geting and molecular imaging. Cell 2006;125: 385–98. 63. Fjeldstad K, K olset SO. Decreasing the metastatic potential in cancers—targeting the heparan sulfate proteoglycans. Curr Drug Targets 2005;6:665–82. 64. Bouri K, Feero WG, Myerburg MM, et al. Polylysine modification of adenoviral f iber protein enhances muscle cell transduction. Hum Gene Ther 1999;10:1633–40. 65. Mahasreshti PJ, Kataram M, Wu H, et al. Ovarian cancer targeted adenoviral-mediated mda-7/IL-24 gene therap y. Gynecol Oncol 2006;100:521–32. 66. White SJ, Nicklin SA, Buning H, et al. Targeted gene delivery to vascular tissue in vi vo by tropism-modif ied adeno-associated vir us vectors. Circulation 2004;109:513–9. 67. Zavada J. VSV pseudotype particles with the coat of avian myeloblastosis virus. Nat New Biol 1972;240:122–4. 68. Endres MJ, Jaffer S, Haggar ty B, et al. Targeting of HIV- and SIVinfected cells b y CD4-chemokine receptor pseudotypes. Science 1997;278:1462–4. 69. Schnell MJ, Johnson JE, Buonocore L, Rose JK. Constr uction of a novel vir us that tar gets HIV-1-infected cells and controls HIV -1 infection. Cell 1997;90:849–57. 70. Somia NV, Miyoshi H, Schmitt MJ, Verma IM. Retroviral vector targeting to human immunodef iciency vir us type 1-infected cells b y receptor pseudotyping. J Virol 2000;74:4420–4. 71. Gambhir SS, Bar rio JR, Phelps ME, et al. Imaging adeno viraldirected reporter gene expression in living animals with positron emission tomography. Proc Natl Acad Sci USA 1999;96:2333–8. 72. Tjuvajev JG, Stockhammer G, Desai R, et al. Imaging the e xpression of transfected genes in vivo. Cancer Res 1995;55:6126–32. 73. Bogdanov A Jr, Weissleder R. The development of in vi vo imaging systems to study gene e xpression. Trends Biotechnol 1998; 16:5–10. 74. Le LP, Everts M, Dmitrie v IP, et al. Fluorescentl y labeled adeno virus with pIX-EGFP for vector detection. Mol Imaging 2004;3:105–16. 75. Le LP, Le HN , Dmitriev IP, et al. Dynamic monitoring of oncol ytic adenovirus in vivo by genetic capsid labeling. J Natl Cancer Inst 2006;98:203–14. 76. Le LP, Li J , Ternovoi VV, et al. Fluorescentl y tagged canine adenovirus via modif ication with protein IX-enhanced g reen fluorescent protein. J Gen Virol 2005;86:3201–8. 77. Matthews QL, Sibley DA, Wu H, et al. Genetic incor poration of a herpes simple x vir us type 1 th ymidine kinase and f irefly luciferase fusion into the adeno virus protein IX for functional displa y on the virion. Mol Imaging 2006;5:510–9. 78. Raty JK, Liimatainen T, Wirth T, et al. Magnetic resonance imaging of viral particle biodistribution in vivo. Gene Ther 2006;13:1440–6. 79. Blasberg RG, Tjuvajev JG. Herpes simplex virus thymidine kinase as a marker/reporter gene for PET imaging of gene therapy. Q J Nucl Med 1999;43:163–9. 80. Shah K, Jacobs A, Breakefield XO, Weissleder R. Molecular imaging of gene therapy for cancer. Gene Ther 2004;11:1175–87. 81. Shah K, Weissleder R. Molecular optical imaging: applications leading to the development of present day therapeutics. NeuroRx 2005; 2:215–25. 82. Weissleder R, Ntziachristos V. Shedding light onto li ve molecular targets. Nat Med 2003;9:123–8. 83. Shah K, Tung CH, Chang CH, et al. In vivo imaging of HIV protease activity in amplicon v ector-transduced gliomas. Cancer Res 2004;64:273–8. 84. Shah K, Tung CH, Yang K, et al. Inducib le release of TRAIL fusion proteins from a proapoptotic for m for tumor therapy. Cancer Res 2004;64:3236–42. 85. Pan G, Ni J, Wei YF, et al. An antagonist decoy receptor and a death domain-containing receptor for TRAIL. Science 1997;277:815–8.

702

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

86. Pan G, O’Rourk e K, Chinnaiyan AM, et al. The receptor for the cytotoxic ligand TRAIL. Science 1997;276:111–3. 87. Shah K, Bureau E, Kim DE, et al. Glioma therapy and real-time imaging of neural precursor cell mig ration and tumor re gression. Ann Neurol 2005;57:34–41. 88. Shah K, Tung CH, Breakefield XO, Weissleder R. In vivo imaging of S-TRAIL-mediated tumor re gression and apoptosis. Mol Ther 2005;11:926–31. 89. Bhaumik S, Gambhir SS. Optical imaging of Renilla luciferase reporter gene expression in living mice. Proc Natl Acad Sci USA 2002;99:377–82. 90. Shah K, Tang Y, Breakefield X, Weissleder R. Real-time imaging of TRAIL-induced apoptosis of glioma tumors in vi vo. Onco gene 2003;22:6865–72. 91. Jacobs A, Tjuvajev JG, Dubrovin M, et al. P ositron emission tomography-based imaging of transgene e xpression mediated b y

92.

93.

94.

95.

replication-conditional, oncol ytic her pes simple x vir us type 1 mutant vectors in vivo. Cancer Res 2001;61:2983–95. Jacobs AH, Winkeler A, Hartung M, et al. Impro ved herpes simplex virus type 1 amplicon v ectors for propor tional coe xpression of positron emission tomo graphy mark er and therapeutic genes. Hum Gene Ther 2003;14:277–97. Gambhir SS, Bauer E, Black ME, et al.A mutant herpes simplex virus type 1 thymidine kinase reporter gene shows improved sensitivity for imaging repor ter gene e xpression with positron emission tomography. Proc Natl Acad Sci USA 2000;97:2785–90. Okada S, Ishii K, Yamane J, et al. In vivo imaging of engrafted neural stem cells: its application in e valuating the optimal timing of transplantation for spinal cord injury. FASEB J 2005;19:1839–41. Love Z, Wang F, Dennis J, et al. Imaging of mesench ymal stem cell transplant b y bioluminescence and PET . J Nucl Med 2007;48: 2011–20.

44 CELL VOYEURISM USING MAGNETIC RESONANCE IMAGING NASER MUJA, PHD, CHRISTOPHER M. LONG, BS, AND JEFF W. M. BULTE, PHD

THE NEED FOR CELL VOYEURISM The coordinated mo vement of cells across v ast tissue expanses is critical for the de velopment and maintenance of multicellular organisms. Indeed, we are often ob livious to the incessant tor rent of cells underw ay within our o wn bodies, right before our very eyes. Visual access to the quotidian life of cells serves to enlighten our understanding of how these minute building blocks of tissues and organs are assembled and how disturbances in the brick and mortar of life cause disease. 1–4 During embr yogenesis, appropriate cell position and fate is attained through the active process of migration from the site of bir th to the f inal target location. Failure of cells to mig rate or mig ration to incor rect locations can ha ve deleterious consequences including failure of pregnancy, fetal death, mental retardation, vascular disease, and congenital abnor malities in limb de velopration is central to ment.5–9 In the adult, cell mig homeostatic processes such as the mounting of immune responses and the repair of damaged tissue. 10 Understanding the mechanisms that initiate, maintain, and terminate the migration of cells is pivotal to our knowledge of nor mal de velopment and disease. Moreover, elucidation of the mechanisms underl ying cell migration is important for emerging areas of biotechnology which focus on tissue engineering, 11–13 development of optimal therapeutic strate gies for stem cell transplanrowth and tation,14–16 and the control of tumor cell g metastasis.9,17 Noninvasive imaging of cell movements in live tissue also pro vides v aluable data re garding the impact of local cues on cell mig ration and will adv ance our kno wledge of ho w therapeutic cell transplants respond to the myriad factors present at sites of disease or injury over long treatment periods. Fur thermore, noninvasive imaging can be applied to measure the rate of

accumulation of therapeutic cells at sites of disease or injury so that cell recr uitment at these sites can be directly correlated with therapeutic outcome. In addition to active cell migration, noninvasive imaging is useful for routine conf irmation of stab le cell position follo wing therapeutic delivery or for conf irmation of the inhibition of cell migration as in the case of metastatic tumors. Magnetic resonance imaging (MRI) has a long history of application in clinical diagnosis and is rapidl y developing as method for nonin vasive imaging of cells during animal development and in laborator y models of disease.18 Established means for magnetic resonance (MR) diagnosis of human disease will f acilitate the scrutiny of e xperimental cell therapeutics. One good example of the potential synergy between clinic and laboratory is the mer ger betw een adv ancements in MR detection of central ner vous system (CNS) lesion progression in animal models of multiple sclerosis (MS)19–23 and e xperimental ef forts to nonin vasively monitor neural stem cell recr uitment to these dem yelinated lesions. 24–28 Methods for MR monitoring of the temporal progression of CNS lesions in human MS have recently been established,29,30 and through careful translation of research adv ances to the clinical practice, it may 1 day be possible to use MRI to observe a cessation or reduction in lesion pro gression in response to cell therapy.

NONINVASIVE IMAGING AND INVASION OF CELL PRIVACY Most cell types spend their lifetimes cached deep beneath skin and bone, generall y well hidden from the insatiable curiosity of investigators. To obtain a glimpse of their pri vate li ves, in vestigators often isolate and 703

704

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

enrich cells from tissue for microscopic anal yses in vitro. However, for the development of cell therapies for disease, there is increasing demand for imaging methodology that exposes cells into plain vie w without significantly destroying their anatomic conte xt. Noninvasive anal yses of cell mig ration therefore require means to indelibly label a well-characterized cell population so that they can be routinely distinguished at any tissue depth within a li ving or ganism at re gular time intervals. Several imaging modalities are available for noninvasive imaging of cell mig ration including ultrasound , multiple photon confocal microscop y, computerized tomography (CT), positron emission tomo graphy (PET), single photon emission computed tomo graphy (SPECT) and MRI. While multiphoton confocal microscopy offers exquisite detail of dynamic changes in cell morphology and position, the application of this methodology is restricted to superf icial tissue preparations. In comparison, ultrasound, CT, PET, SPECT, and MRI allow tracking of cell position at an y tissue depth at the e xpense of (sub)cellular detail and real-time visualization. Unlik e CT , PET , and SPECT , w hich detect potentiall y har mful ionizing radiation, MRI measures the response of the proton that for ms the 1H nucleus in water, fat, and other biomolecules to applied radio w aves in a strong homo geneous magnetic f ield. Besides the detection of re gional dif ferences in w ater abundance in various tissues, MRI can be used to detect local inhomogeneities in proton relaxation produced by paramagnetic or super paramagnetic iron o xide (SPIO) nanoparticles to generate specif ic contrast in labeled cells, tissues, or body fluids. One such SPIO contrast agent, Feridex or Endorem, is FD A-approved for clinical application, streamlining the transfer of experimental research f indings to clinical settings. 31–35 Though particulate MR contrast agents are lar ger in size than highly solub le radioisotope conjugated PET contrast agents and require special means for cell incorporation, MRI per mits safe, periodic, high resolution (do wn to ~50 µm in animals) detection of contrast-labeled cells deep within an opticall y opaque, li ving organism with reasonable scan duration. A signif icant adv antage of MRI is that in addition to identifying transplanted cells in their anatomic conte xt, it can pro vide infor mation about the surrounding milieu (ie, edema, inflammation, and lesion size). In this chapter, we detail the current state of the art of cell labeling for MR detection and noninvasive MRI of cell mig ration in li ve animals with emphasis on applications in immunology and stem cell therapeutics.

LABELING CELLS FOR MR DETECTION To detect cells of interest within tissue, a biocompatible MR contrast agent must be incor porated and retained b y the cells throughout the duration of experimental anal ysis. Endo genous contrast in MRI originates from local variations in tissue water concentration and the chemicall y bound state of protons. 36 Two relaxation time constants, T1 and T 2 which are associated with the time deca y of h ydrogen proton magnetization back to equilibrium follo wing a radiofrequency (RF) pulse, are applied to measure intrinsic tissue variations that generate anatomic contrast. T1, or spin-lattice relaxation, characterizes the ener gy transition of the nuclear spin to equilibrium follo wing a RF pulse, whereas T2, or spin-spin relaxation, cor responds to the loss of coherenc y among adjacent nuclear spins. T1 and T 2 time constants are af fected b y the local microenvironment of w ater molecules, w hich ma y b y influenced b y temperature, viscosity , dif fusion, bulk flow, and pro ximity to inhomo geneities induced b y macromolecules or paramagnetic and super paramagnetic ions (also refer red to as T2* relaxation). 37–39 Via shortening of T1 and T 2 relaxation, paramagnetic and superparamagnetic contrast agents can be applied to distinguish transplanted cells from backg round tissue contrast by MRI. The initial challenges to in vivo cellular MRI consisted of the successful incor poration of paramagnetic and superparamagnetic contrast agents into the cells for subsequent MR detection and w hether there e xisted any adv erse cell responses to contrast agent follo wing incor poration. Paramagnetic metals with unpaired electrons such as manganese among the transition metals and gadolinium from the lanthanide f amily (elements 58–71) primaril y decrease T1 relaxation, producing a h yperintense or positive (white) signal in the re gion where they distribute. Gadolinium ion stabilized by a shell of dieth ylenetriamine pentaacetic acid (Gd-DPTA) or tetraazac yclododecanetetraacetic acid and related Gd chelate derivatives are routinely applied as fluid phase (blood, cerebrospinal fluid, etc) contrast agents. One approach to cell labeling with Gd in volves pre-incubation of Gd-DPTA with albumin protein to for m a complex that may reduce the intracellular to xicity of Gd-DPT A w hile maintaining its T1 relaxation proper ties for nonin vasive imaging.40 However, due to relati vely lo w T1 contrast effects and unkno wn intracellular biocompatibility , examples of cell labeling using Gd are limited 41–48 and have not been y et introduced in the clinic. Manganese chloride, another T 1 enhancing contrast agent commonl y used to label cardiac muscle cells or neurons for functional MRI

Cell Voyeurism Using Magnetic Resonance Imaging

following uptak e, has onl y recently been applied to label T lymphocytes and B cells,49 which are notoriously difficult to label with iron o xides. Due to their direct dipole-dipole interaction, the T1 relaxation effects of paramagnetic agents are, on a molar basis, weaker than iron oxide nanoparticles which dephase multiple proton spins via their ef fects on local magnetic f ield g radients. In addition, once inter nalized b y cells, paramagnetic contrast agents ma y e xhibit reduced T 1 relaxivity compared with their unbound counterparts in solution. Advances in chelation and par ticle formation yielding T1 contrast agents that are more amenab le to cell labeling would expand options for cell tracking, and when combined with MRI of cells labeled with SPIO nanoparticles, it ma y be possib le to simultaneousl y follow two independently labeled cell populations. 50 However, in practice, a hyperintense signal and hypointense signal colocalized in space might cancel each other for detection under certain conditions.

CELLS PUMPING IRON Iron o xide nanopar ticles are e xtensively applied in cell imaging and cell tracking studies due to their strong contrast effect, improved biocompatibility, variety in core size and coating surface, and ease of microscopic detection by Prussian Blue staining. SPIO nanopar ticles (50–200 nm diameter) and their ultrasmall SPIO (USPIO) variants51,52 (~35 nm diameter) are composed of stabilized maghemite (Fe2O3) or magnetite (Fe3O4) and act locally to reduce T2 and T2* relaxation via the induction of strong f ield inhomogeneities upon imaging with the appropriate pulse sequences to produce a h ypointense or ne gative (b lack) signal.53 Dextran, carboxydextran, or silo xane coating of the iron oxide core ensures iron oxide nanoparticle stability and solubility in biolo gic media and is also thought to minimize an y deleterious ef fects on cell function upon cellular uptak e. In addition, pol ystyrene m icrospheres containing iron oxides (MPIO) or so-called Bangs particles ha ve recentl y been applied for single-cell detection.54–60 Though the y are quite lar ge (1 µm), these particles are the most sensitive contrast agent available for cell tracking, due to their high iron content (relaxi vity) and reduced par tial v olume ef fects. Unlik e SPIOs and USPIOs, Bangs par ticles are not biode gradable by intracellular dextranases, are not supplied as sterile materials, and the ef fects of long-ter m retention of Bangs par ticles on cell physiology are not known—limiting their application to animal studies. While cells with phagocytic abilities readily ingest iron oxide nanopar ticles,61 many cells lack substantial

705

phagocytic function and require alter nate means of contrast agent incor poration.62 Early approaches for paramagnetic and super paramagnetic contrast agent introduction into cells used either iron oxide coupled to lectin,63 iron oxide incor porated into viral capsids for subsequent eukar yotic cell membrane docking and fusion,64 or binding of a cell-specif ic biotin ylated monoclonal antibody complexed with streptavidin, and biotinylated dextran magnetite. 65 Subsequent developments in cell labeling included the use of cationic twin arginine translocase (T AT) peptides 66 or dendrimers 67 to coat iron oxides and facilitate their electrostatic association with the anionic e xtracellular membrane and subsequent engulfment by cells with macropinoc ytotic activity. Presentl y, a panoply of cell labeling methods are a vailable including bombardment, 68,69 receptor mediated or fluid phase endoc ytosis,70 lipofection,71,72 polycation-based transfection agents (pol y-L-lysine and pol yarginyl rich protamine sulf ate),32,33,73,74 and magnetoelectroporation.75,76 Unlike other methods of contrast agent incorporation, mechanical methods such as electroporation rapidl y introduce contrast agent in the c ytoplasm thereb y ob viating the need to hold cells in vitro for prolonged periods and reducing potential alterations in cell proper ties during this incubation period. In addition, mechanical methods per mit contrast agent introduction into cells that require regulatory certified preparation or are difficult to transfect due to their small size (l ymphocytes), lack of phagocytic activity, differentiation state, or lo w proliferative rate. Though each of the abo ve cell labeling methods effectively transpor t contrast agents into cells, the parameters for each labeling approach must be carefull y adjusted to ensure sufficient MR contrast agent uptake in the absence of intracellular aggregation and precipitation, adverse effects on cell proliferation, dif ferentiation, and migration, or cell to xicity.77–79 In addition, for MRI experiments in w hich cells are monitored for se veral days, it is necessary to verify contrast agent retention and characterize contrast agent dilution follo wing cell di vision.80–82 Also, because it is dif ficult to unifor mly label and track an entire mig ratory cell population, usuall y a subset of the population of interest is sampled o ver time. Alternatively, the intracellular concentration of MR contrast agent ma y be optimized to allo w for impro ved discrimination of independent cells: k eeping in mind the balance betw een cell tolerance of the contrast agent, intensity of the MR signal, percentage of labeled cells, and dilution of contrast agent due to cell proliferation.

706

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Increasing applications of MRI in no vel and unique research settings ha ve e xpanded the limits of nonin vasive cell tracking. F or example, recent de velopments in engineered nanopar ticles and biolo gic repor ters present exciting possibilities. When chemicall y cross-link ed in the presence of epichlorohydrin and ammonia, monocrystalline iron o xide nanopar ticles (MION) and free dextran yield an amine-terminated cross-linked iron oxide (CLIO) nanopar ticle.83,84 The incor poration of functional amine g roups enables the coupling of a wide variety of link ers that increase the functionality and specificity of CLIO nanopar ticles. In par ticular, molecules such as anne xin V,85 oligonucleotides, fluorescent dyes,86,87 quantum dots, 88 and TAT peptides86,89,90 have been conjugated to CLIO nanopar ticles.

FLUORINE As pre viously described, (super)paramagnetic labeling agents cause local f ield inhomo geneities allo wing for the detection of the agent by measuring its effect on the surrounding 1H-rich environment. Thus, these contrast agents are not actuall y directl y detected w hen used in the setting of cellular imaging. In the case of iron o xide contrast agents, the presence of the agent will create a signal decrease on T2/T2* w eighted images and a resulting blackout in the areas containing labeled cells. These “blackouts” are very difficult to distinguish from many other dark spots in an image which may be caused by necrotic tissue, hemor rhage, or inherentl y T2 dark tissues. In many cases, a priori kno wledge is needed to locate the cells of interest. Fluorine ( 19F), however, can

A

be detected directly, and the lack of endogenous fluorine atoms in the body be yond trace amounts assures the absence of an y backg round signal. Due to the f act that there is v ery little 19F tissue signal, fluorine-labeled cells can be imaged in a “hot spot” fashion akin to imaging modalities such as PET, SPECT, and bioluminescent imaging.91 Fluorine was recognized as an e xcellent tracer element for MR spectroscopy and imaging very early in the history of MRI w hen it w as f irst attempted in the 1970s.92 Features of 19F that make it well suited for MR include a gyromagnetic ratio close to that of 1H, a spin 1⁄ nucleus, it has 100% natural ab undance, and the rel2 ative sensitivity is 83% that of 1H. However, it still took 30 y ears before this MRI technolo gy w as f irst introduced for in vi vo cell tracking. 93 As an agent for cell tracking, 19F does not require a priori knowledge of possible cell locations. The method is highl y selective for labeled cells, and w hen con ventional 1H images are acquired simultaneously, the superimposition of the two scans allows for the “hot spot” location of cells to be determined. Figure 1 shows a first example of this technique, w here dendritic cells (DCs) are labeled with polyfluorpolyether (PFPE) emulsions and the two nuclei imaged following various routes of injection. In addition, as 19F is being directl y detected , as opposed to (super)paramagnetic agents, the signal can be quantified based on spectroscopy. A direct consequence of this ability to quantify the fluorine label is the potential for accurate determination of trafficking cell numbers. 94,95 To date, all attempts at 19F imaging ha ve in volved intracellular labeling of cells with tracer agents following in vitro labeling.93–95 Most often, the agent of choice is a

B

C

Figure 1. A, Mouse quadriceps after intramuscular injection of PFPE-labeled dendritic cells (DCs) (asterisk indicates injection site). Shown (from left to right) are the 19F and 1H images and a merged 19F/1H image. B, Merged 19F/1H of DC migration into the popliteal lymph node following hind footpad injection. C, Merged 19F/1H image following intravenous injection of PFPE-labeled DCs. Cells are apparent in the liver (L), spleen (S), and also sparsely in the lungs (Lu). Adapted from Ahrens ET et al.93

Cell Voyeurism Using Magnetic Resonance Imaging

perfluorocarbon (PFC). Of course, an intracellular contrast agent that af fects viability , dif ferentiation, or function of the labeled cell is of little utility; therefore, the effects of labeling cells with fluorine contrast agents must be look ed at in detail. The PFC compounds currently being used in MRI ha ve proven to be safe in cell types such as DCs, T cells, and stem cells.93–95 This result is not entirel y une xpected as PFCs ha ve a high o xygen solubility coef ficient and maintain high o xygen par tial pressures. They have e ven been used as o xygen “reservoirs” for har vested organs in pancreas or gan transplantation,96,97 and related fluorine compounds, such as LiquiVent® and Oxygent ®, have been used in clinical trials as ar tificial b lood substitutes. Unlik e the de xtrancoated iron o xides often used for cell tracking, w hich become biolo gically a vailable, PFCs are biolo gically inert and in many cases eventually leave the body through exhalation in the form of a gas. The use of fluorinated agents may eventually overcome some of the limitations associated with traditional 1H MRI contrast agents. Ho wever, the sensiti vity of this method is still lacking as its strength can also be its w eakness. The “hot spot” nature of this technique means the onl y signal obtained is from the cells themselves; therefore, these cells must be hea vily labeled and in close pro ximity to each other in order to obtain suf ficient signal. Although, (super)paramagnetic agents are not directl y detected, they do dra w upon an incredib ly v ast 1H sur rounding w hich increases their sensitivity greatly. Even with this limitation, the current clinical use of 19F agents may pave the way for possible clinical translation of this promising technolo gy.

A

B

707

REPORTER GENES So far, imaging contrast agents ha ve been discussed which require the cells of interest to be labeled with an exogenous agent. This labeling ma y occur in vi vo or e x vivo and is quite ef ficient, but once the cells of interest ha ve been labeled, they have the potential to dilute this contrast as they divide. For terminally differentiated cells, such as DCs, this may not present a prob lem, but for rapidl y dividing cells, such as stem cells and T cells, one ma y see a precipitous decline in the ability to track the cells after the y have been labeled. Our g roup has e ven shown that asymmetric cell division may occur where the contrast agent is not divided evenly among daughter cells 82 (Figure 2). The ability to track cells o ver a longtime course is se verely impaired b y both dilution and asymmetric cell division and remains one of the major concerns with cell tracking in dividing cells. Another major concer n with e xogenous MR contrast agents is the ability to monitor viab le cells alone. Once a cell has been labeled, viable or not, the cell will continue to produce contrast in an MR image until that contrast agent has been removed by macrophages or has been diluted sufficiently in the sur rounding tissue. This phenomenon ma y lead to incor rect obser vations of cell distributions. 82 The ability to deter mine viable from non viable cells in an MR image is of incredib le impor tance as, in most cases, the cells of interest to the in vestigator are indeed the viab le ones. For reliable tracking of viable and proliferating cells, the best strategy may be the use of an MR repor ter gene.99 Reporter genes such as LacZ, green fluorescent protein (GFP), and firefly luciferase have been used for many years

C

Figure 2. High-resolution ex vivo MR imaging of neonate shiverer mouse brain following intracerebroventricular transplantation of superparamagnetic iron oxide-labeled immortalized neural stem cells. The mismatch with conventional histology and loss of MR detectability following cell proliferation is apparent. A, 2 weeks after grafting, Feridex-labeled C17.2 cells migrated vast distances toward the outer cortical layers of the cerebrum and olfactory bulb as revealed by anti-β-gal staining. This is in sharp contrast with the MR imaging pattern (B) which shows hypointense cells centered in and around the ventricles but not the cortical layers. The merged histology/MR image (C) in which β-gal+ cells are red, and MRI hypointense cells yellow further illustrate this mismatch. Scale bar = 1 mm. Reproduced with permission from Walczak P et al.82

708

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

in vitro to measure genetic transcription acti vity. Recently, the approach has been transfer red to in vi vo imaging with nuclear and optic repor ters sho wing g reat promise. PET also has se veral types of repor ter genes/probes under development and v alidation.100 However, both optic and nuclear in vi vo imaging techniques ha ve their limitations. Optic imaging techniques are f aced with issues of tissue opacity, and nuclear imaging has been challenged with limitations in resolution.101,102 The use of reporter genes in MR can overcome both of these limitations with the superior three-dimensional (3D) spatial resolution of MRI providing detailed anatomic infor mation and cell tracking ability. By transfecting cells of interest with an MR reporter gene, one can o vercome the issues of contrast dilution in di viding cells and increase the odds of imaging viable cells. This section will re view the pro gress of recent candidate repor ter genes that ma y make this contrast mechanism a reality in the world of in vivo MR cell tracking.

FERRITIN/TRANSFERRIN Most MRI cell labeling techniques rely upon loading cells with iron o xides in vitro; therefore, it w as a v ery logical transition for developers to turn toward iron when considering possible reporter genes. Iron is an essential nutrient for all cells, par ticipating in man y metabolic pathw ays, and is required for proper function of numerous essential proteins.103–105 Evolution has pro vided a fe w specialized iron-binding proteins w hich aid in iron storage and bioavailability. Among these proteins, fer ritin, transferrin (Tf), and transfer rin receptor (Tf R) ha ve recei ved the most attention for MR cell tracking. Each of the pre viously mentioned proteins pla ys an important role in transpor ting iron either into the cell or into intracellular iron storage. In vertebrates, Tf is responsible for transpor ting iron throughout the b lood, and this iron is eventually internalized by cells following the binding of iron-loaded Tf to the TfR.106 With a little bit of help, cells are capable of expressing the TfR in incredibly high numbers, on the order of se veral million copies per cell. This overexpression can be used to increase iron uptak e and eventually iron accumulation in the cell and , due to the paramagnetic nature of fer ric iron, may be a possib le contrast mechanism for MRI. Using this rationale, the TfR was f irst used as an MRI repor ter gene by Koretsky and colleagues in 1996. 107,108 A tumorigenic f ibroblast cell line transfected with a DN A constr uct encoding for the human TfR (hTf R) w as used to produce tumors in mice. The overexpression of hTfR led to an approximately threefold increase in iron content in these cells, which was shown by electron microscop y to be concentrated in the ferritin protein. This increase in iron content manifested

itself in a signif icant reduction in signal in T2-weighted MR images. The work by Koretsky and colleagues def initely displayed a proof of concept for the use of repor ter genes in MRI; ho wever, the v ery high le vels of Tf R expression required in this study and the relati vely small changes in relaxation time ha ve proven to be a limitation in the application of this approach for cell tracking. To overcome this limitation, some g roups have given up the use of endo genous iron in conjunction with Tf and simply used Tf as a fer ry to tar get MR contrast agents to cells expressing the TfR. This approach has been successfully used to load oligodendroc ytes with USPIOs after USPIO conjugation to anti-Tf R antibodies. 70 In vivo targeting with USPIOs has also been achie ved as one g roup targeted rat mammar y tumors b y relying on the endo genously elevated expression of the TfR in malignancies. 109 Other groups have relied upon engineered TfRs in proliferating tumor cell lines and MION conjugated to Tf to detect the presence of TfR under MRI. 110,111 In each case, the groups have shown that it is possible to measure transgene expression noninvasively using MRI by targeting iron oxides to receptors found on the cell surf ace. An alter native to attacking iron accumulation from the outside is to sequester the iron inside the cell in a manner that would require the cell to upre gulate its iron transport naturally. Ferritin, an endogenous protein which envelopes an iron o xide core of up to 4,500 iron atoms, can ser ve e xactly that pur pose.112 Many in vestigations have also sho wn that fer ritin has an abnor mally high transverse relaxivity (R2), even at iron loadings as low as 13 to 14 iron atoms, and has a unique linear dependence of R2 on the magnetic f ield.113–116 All of these characteristics make the ferritin protein an exquisite choice for use as an MR reporter gene. The first time ferritin was used as an MR reporter gene, it was not actuall y in full for m but in a tr uncated version which relied upon the hea vy chain of fer ritin (h-fer ritin). Expression of h-fer ritin w as monitored in a tetrac yclineinducible C6 glioma tumor cell line as w ell as in specif ic tissues in transgenic mice. 117,118 Adenoviral infection w as used in mice brains to visualize transgene expression of the combination of both light (l-fer ritin) and hea vy chains of ferritin in neurons and glia. 119 Finally, the ferritin gene has even been used in combination with the TfR in mouse neural stem cells to take advantage of multiple points in the iron metabolism pathway.120 Moving away from traditional iron metabolic proteins, one may r un across the enzyme tyrosinase in search for means to accumulate iron within a cell. This enzyme is primarily responsible for re gulating the synthesis of the protein melanin, w hich man y kno w for its b lack coloration and possibly for it photoprotection abilities, but this protein

Cell Voyeurism Using Magnetic Resonance Imaging

also has an incredib ly high af finity and binding capacity for metal ions. As apposed to other iron contrast mechanisms w e ha ve discussed in this section, melanin is mainly responsib le for T1 hyperintensities instead of T2 hypointensities. In an y case, the contrast obtained b y natural melanin was enough for some to pursue the use of this tyrosinase-melanin system, in w hich tyrosinase is o verexpressed and leads to melanin synthesis, as a repor ter gene for MRI. Weissleder and colleagues 121 transfected human tyrosinase into mouse f ibroblasts and human embr yonal kidney cells, which led to constitutive expression of human tyrosinase and ele vated le vels of melanin production in transfected cells. The increased le vels of melanin w ere significant enough to produce detectab le increases in signal intensity b y MRI. Similar to the use of the fer ritin transgene, a group desired expression which could be regulated and de veloped a tetrac ycline-inducible tyrosinase system.122 As with all of the other contrast mechanisms we have discussed, there are a fe w potential pitf alls with using metalloprotein-based reporter genes in MRI cell tracking. With the e xception of tar geting the TfR with an e xogenous iron oxide, each of these methods tak es a considerable amount of time for detectab le iron accumulation to occur. Of course, this detectable iron level, in some cases, may ne ver occur . This method relies upon suf ficient endogenous iron being present in the surrounding tissues. Some g roups ha ve o vercome this limitation b y preincubating cells in an iron rich media before transfer; however, this will onl y provide a temporar y solution as these cells will e ventually dilute the iron stores as the y divide and again become reliant on endo genous iron. Finally, there is a linear increase in R 2 with increasing field strength, meaning the sensitivity of detection will be lower at clinical field strengths. Even with the possibility of these hindrances, there is still a g reat deal of e xcitement in the f ield due to the potential of these genes to provide temporal information for dividing cells that is not currently available with iron oxide labeling.

A

709

LYSINE-RICH PROTEIN/CHEMICAL EXCHANGE SATURATION TRANSFER In the previous section, it was mentioned that some groups had created inducible reporter genes. The rationale behind these inducible repor ter genes w as that contrast could be turned on or tur ned of f. Ho wever, it tak es considerab le time for iron to either accumulate or be remo ved from a cell. Thus, the temporal resolution of these inducib le systems drops drasticall y. The ability to tur n on or tur n of f contrast instantaneously would be of great utility. This type of rapid switching is possible with chemical exchange saturation transfer (CEST). Unlik e super paramagnetic substances that rel y on w ater relaxation to gain contrast and cannot be turned off, CEST relies upon the magnetization transfer that occurs between the bulk water protons and the macromolecular protons. Contrast can be tur ned on b y applying a saturation RF pulse at the macromolecular proton frequenc y or w hen no contrast is desired , and MR images can be obtained without perfor ming the selecti ve saturation pulse. One can choose at will w hen to attenuate the signal of the sur rounding bulk water. Our lab has recentl y be gun to use CEST contrast mechanisms for repor ter gene imaging with MRI. Gilad and colleagues 107 have successfully designed an endo genous CEST agent by creating a reporter gene that encodes a l ysine-rich protein (LRP). As the decrease in signal amplitude seen with CEST depends on the rate of exchange between the bulk w ater protons and the macromolecular protons, poly-L-lysine polypeptides were found to be ideal due to their rapidl y e xchanging amide protons,123 leading to the concept of LRP as CEST MR reporter gene. Extracts originating from LRP-e xpressing cells demonstrated a signif icant increase in MR contrast relative to e xtracts of control cells. The LRP protein w as also capab le of producing a signif icant CEST signal in vivo following injection of transfected 9L glioma cells into the brains of SCID mice (Figure 3). As mentioned previously, a significant benefit of CEST is the ability to tur n

B

Figure 3. A, In vivo anatomic image of mouse brain injected with lysine-rich protein (LRP)-expressing glioma cells in the left hemisphere and control tumor cells in the right hemisphere. B, Chemical exchange saturation transfer signal intensity–difference map overlaid on the anatomic image distinguishes LRP-expressing and control tumor xenografts. Note that in these initial studies, a proper adjustment of the field homogeneity was only achieved inside the brain, leading to some artifacts at the brain edges. Adapted from Gilad AA et al.107

710

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

the contrast on and of f at will, but the frequenc y dependence of CEST may also provide the possibility for multiple spectrall y resolv ed repor ters to be used to create varying “colors” of contrast. 124

An ideal research laborator y would have ready visual access to the bir th, migration, proliferation, dif ferentiation, and death of cells within the body. However, it is currently not possible to unobtrusively record every living moment of a gi ven cell or cell population. Of the many measurable attributes of cellular life, MRI is currently most applicab le for the nonin vasive obser vation of cellular position and mig ration in li ving specimens. Cells of the immune system, such as macrophages, DCs, T cells, and B cells, circulate throughout the body and periodically migrate toward sites of infection or tissue trauma. Similarl y, lineage restricted adult stem cells proliferate and mig rate in response to en vironmental cues to replace or repair damaged or dead tissue. MRI is highl y suited to nonin vasive tracking of immune cell and adult stem cell recr uitment at various locations in the body and pro vides impor tant insight regarding the po werful therapeutic proper ties of these dynamic cells. Having defined the need for noninvasive MR cell tracking and means for satisfying this demand, we no w delv e into the practical application of noninvasive MR cell tracking in cells of the immune system and in adult stem cells.

such as asthma and ar thritis. Until recentl y, the in vi vo study of immune cell populations w as hindered b y our lack of appropriate technolo gy. The studies of dynamic immune processes were limited to individual time points created from ex vivo histologic studies. As infor mative as these studies are, they consume a great deal of material and give only an indirect vie w of the masterful w ork that is constantly taking place within the body . With the development of in vi vo cellular imaging b y MRI, the repeated assessment of specif ic immune cells has become possible. The utility of in vi vo cellular imaging for studies of the immune system is enor mous. As new immunotherapies in the realm of cancer and autoimmune disease are becoming more and more pre valent every day, there is a growing need to study the role of immune cells in their endogenous en vironment. These cells also ha ve been found to play a large role in plaque formation, contribute to stok e patholo gy, and are the major pla yers in or gan rejection. A great deal of research has been dedicated to developing strate gies to label these cells for in vi vo imaging. To date, the most success has been in labeling the naturally phagocytic cells, such as macrophages, but progress has been made in labeling other immune cells which are not so for tunate to ha ve naturally phagocytic capabilities. For those involved in the f ield of immunology, the promise of tar geting a distinct population of cells in vi vo and studying their tissue distribution, proliferation, and sur vival nonin vasively is more than enough to w ait until this cur rent bump in appropriate technology is overcome.

IMMUNE CELLS

MACROPHAGE IMAGING

During early development, cells of the immune system develop “self ” recognition so that the body’ s own cells are not selected for immune attack. This sense of self is established early in life as the T cells migrate through the primary lymphoid tissues of the bone mar row (B cells) and thymus (T cells). Immunity and wound healing often occur in parallel and both these processes rel y on the ability of cells to mig rate. Immune cells circulate throughout the body and migrate through tissue in search of foreign material to isolate and destro y. For example, following a cut or bur n injury, nearby resident cells proliferate and mig rate to occup y the w ound site, and immune cells are recruited to eliminate invading bacteria and other microor ganisms. Once tissue repair is complete, the immune cells must deacti vate and lea ve the injury site to a void chronic inflammator y conditions

Macrophages are the sca vengers of our bodies, at some point sampling from nearly every foreign body that enters our circulation. F or this reason, the y w ere the f irst immune cell to be labeled for cellular MRI and have been the most e xtensively studied for this pur pose. K upffer cells, the resident phagoc ytes of the li ver, were found to be labeled upon intra venous administration of SPIO.125,126 This initial f inding providing the impetus to further study macrophage labeling for in vi vo cellular imaging in various other compartments of the body. New iron o xide for mulations, such as ultrasmall par ticles of iron o xides (USPIOs), w ere created to increase b lood half-life and to tar get macrophages in multiple compar tments including lymph nodes. 51,127,128 The in vivo cellular targeting of macrophages b y intravenous injection of USPIO results in a darkening of the lymph nodes on MR

MODEL SYSTEMS FOR NONINVASIVE MRI OF CELL MIGRATION

Cell Voyeurism Using Magnetic Resonance Imaging

images. Detection of tumor metastases is possib le using this technique as nonphagocytic metastatic tissues remain unchanged on USPIO-enhanced MR images. 129–131 Several studies ha ve demonstrated enhanced sensiti vity and specificity for l ymph node e valuation for pelvic, head and neck, chest, and breast malignancies after administration of USPIO .127,132–135 USPIO-enhanced MRI has even been shown to have a higher diagnostic accuracy for depicting lymph node metastasis than PET/CT.136 As great as MR has been for lymph node staging in cancer, its utility for cellular imaging of macrophage activity is not just limited to the l ymph node. Macrophage imaging has also found applications in many other disease models. One of the hallmarks of severe MS lesions is the presence of phagoc ytic macrophages. These inflammator y foci can be visualized in the brains of rats and mice with e xperimental autoimmune encephalom yelitis (EAE) using MRI and intravenous injection of iron o xides.22,137,138 The magnetic labels ma y either be phagoc ytosed by circulating monocytes or leak through the disr upted b lood-brain barrier where the y are then tak en up b y resident brain microglia. In some cases, MRI monitoring of macrophage inf iltration may provide an insightful platform to in vestigate the se verity of inflammator y demyelinating CNS diseases and pro vide a sur rogate imaging marker of treatment efficacy.138–140 Brain inflammation is also an impor tant pathomechanism of ischemic brain injur y acutel y follo wing strok e onset. P ostischemic inflammation is dominated b y mononuclear phagocytic cells including a dela yed inf iltration of hematogenous macrophages.141 The intensity of this inflammatory response can be monitored using USPIOenhanced MRI and has been shown noninvasively in animal models of ischemic strok e b y monitoring of macrophage infiltration.142–145 The success of e xperimental strok e lesions in visualizing macrophages b y USPIO-enhanced MRI has led to clinical pilot studies w here USPIOenhanced MRI has proven capab le of monitoring macrophage inf iltration in human ischemic strok e Saleh, 2004#147. As with MS, MR cellular imaging of macrophage inf iltration into earl y ischemic strok e lesions may help to more specif ically tar get anti-inflammator y therapies in humans. Another application of in vi vo cellular imaging of macrophages exists in visualizing or gan graft rejection. Detection of macrophage infiltration into grafts by labeling the cells with SPIO has been perfor med to characterize acute and chronic kidne y allo graft rejection in rats.146,147 However, this technique is not limited to laboratory animals as this approach has also been translated

711

to the clinic b y Hauger and colleagues 148 where the y have shown that USPIO-enhanced MRI can demonstrate infiltration of the kidneys by macrophages both in native and transplanted kidne ys. Hypointense MR signals following USPIO injection have also been found in heterotopic lung transplants in rats without an immunosuppressive regimen of cyclosporine A.149 Similar results have also been obtained for heterotopic hear t transplants,150 and recentl y, micrometer -sized iron oxides ha ve been used to visualize indi vidual macrophages involved in myocardial rejection.151 Finally, macrophages pla y a predominant role in plaque for mation and pro gression in atherosclerosis. Histologic studies ha ve identif ied “high risk” plaques as those with thin/eroded f ibrous caps w hich o verly lar ge necrotic lipid cores and ha ve an abundance of inflammatory cells (macrophages). 152 Therefore, plaque composition may be the best indicator for plaque stability and the risk of a clinical e vent. In light of this infor mation, there has been a signif icant push to visualize plaque composition in vivo.153–155 In fact, plaque composition can now be characterized noninvasively with MRI. 156,157 Animal studies of atherosclerosis ha ve shown that SPIO par ticles are taken up by inflamed plaques rich in macrophages as intracellular deposits that induce areas of signal loss on T2*-weighted MRI within the v essel w all.158–161 More recently, in vi vo human studies using the USPIO agent, Sinerem, have conf irmed these f indings and also ref ined optimal MRI parameters to detect inflamed plaques. 162–164 Studies of atherosclerosis have not been limited to USPIOs alone though as some groups have targeted receptors found on macrophages to deliver paramagnetic payloads.165

DC IMAGING Although macrophages ha ve recei ved the most attention for in vi vo cellular imaging, the y are not alone in their scavenging abilities. DCs are also highl y phagocytic (in an immature state) but until recently had not been studied for in vi vo cellular MRI. Cellular therapies using DCs are being increasingly applied to the study of tumor immunotherap y and other disease models. F or ef fective immunotherap y, ho wever, DCs must mig rate throughout the v ascular and l ymphatic system to present their antigens to T cells located within l ymph nodes. This mig ration has ne ver been visualized in vi vo and has presented itself as a major limitation in the de velopment and e valuation of therapeutic immunotherapies. Quantif iable inter mediate biologic endpoints are needed that can guide ef forts to improve immunologic potency.

712

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

For nearly all vaccines, a significant bottleneck in the afferent ar m of the immune response is the number of fully activated DCs that deliver captured antigen from the vaccine site to secondar y lymphoid tissues w here T cell priming occurs. The number of DCs that ultimatel y end up in the T cell zone has been sho wn to deter mine the magnitude of T cell proliferation and ef fector response.166 With this information in hand, our group has developed a no vel nonin vasive means to monitor and quantify this critical parameter using MRI. F or the f irst time, MR has been used to image antigen presenting cells that have been labeled by the physiologic process of antigen capture in vi vo in an attempt to elucidate the trafficking patterns of blood derived DCs (Figure 4). Using a SPIO-labeled g ranulocyte-macrophage colon y stimulating f actor tumor cell-based v accine (GVAX), our g roup has shown that it is possib le to quantitati vely assess the accumulation of blood derived antigen loaded DCs in the draining l ymph nodes. Antigen presenting cells w hich had captured SPIO w ere imaged over time as the y accumulated in lymph nodes and were shown to present tumor antigen to T cells following isolation in vitro. In addition, the method of nonin vasive quantif ication, w hich w as

A

E

B

validated by enumerating actual DC numbers, has suf ficient resolution to discriminate the impact of an adjuvant on DC deli very to l ymph nodes follo wing v accination. This study established MRI cell tracking as a useful tool to systematically evaluate key parameters relevant to the optimization of vaccine therapies. However, not all vaccines rely on endogenous DC production. Many groups have used DC vaccines for the purpose of immunotherapy. In these studies, DCs are matured ex vivo, loaded with candidate antigens, and injected back into the subject. The same hurdles are present with this vaccine approach as DCs must still reach draining l ymph nodes in suf ficient numbers to stimulate T cell proliferation. Independent studies have administered DCs by many different routes including intrader mally, subcutaneousl y, intravenously, and even intranodal. It is stillnot clear which route of administration is optimal. It is clear, however, that the ability to monitor DC traf ficking has the potential to facilitate in the design of optimal DC therap y. The f irst study that aimed at magneticall y labeling DCs for in vi vo MRI used receptor -mediated endocytosis of antibody-conjugated SPIO par ticles.167 Bone marrow-derived DCs and a fetal skin-deri ved DC line

C

D

G

Figure 4. Monitoring the trafficking of dendritic cells (DCs) that have taken up superparamagnetic iron oxide (SPIO) in vivo after intradermal injection of SPIO-labeled granulocyte-macrophage colony stimulating factor (GM-CSF) tumor cell vaccine into the footpad of mice. B, D, F, and H, Magnifications of the corresponding insets in (A, C, E, and G) respectively. Open arrows indicate the draining popliteal LN at the side of the footpad receiving SPIO-labeled GM-CSF vaccines. On the multi-gradient echo images, SPIO-containing LNs have decreased signal intensity indicating active trafficking of sentinel DCs from the footpad to the lymph node. At 12 hours (A and B) and 2 days (C and D) after injection, popliteal LNs do not yet show any evidence of homing of dendritic cells. At 3 (E and F) and 7 days (G and H) after injection, a decreased signal intensity, representative of DC homing, is apparent only in the draining LN on the side of SPIO-labeled vaccine injection.

Cell Voyeurism Using Magnetic Resonance Imaging

were incubated with SPIO conjugated to anti-CD11c monoclonal antibody . There w as an appro ximately 50-fold increase in uptake relative to DCs incubated with naked SPIO, y et the increased uptak e did not result in any signif icant changes in cell viability , immunological function, or phenotype of the labeled DCs, with the exception of a do wn re gulation of surf ace CD11c. Labeled DCs w ere injected into murine quadriceps and monitored in vivo for several days using MRI. Baumjohann and colleagues 168 have recentl y sho wn the ability to visualize the immig ration of SPIO-labeled DCs into the draining l ymph nodes of mice by MRI after subcutaneous administration. DCs w ere incubated with a mixture of protamine sulf ate and SPIO to enhance the SPIO uptak e. Mature SPIO-labeled DCs w ere then injected into the footpads of mice, and their accumulation in the draining popliteal l ymph nodes w as monitored in vivo b y MRI at 4.7T after 24 hours. Distinct signal reduction patterns were present that cor related nicely with the detection of SPIO-DCs found b y iron staining and immunohistology of dissected lymph nodes. The use of MRI for tracking SPIO-labeled DCs has even been used to monitor cellular therapy in humans.31 De Vries and colleagues labeled autologous immature DCs in culture with clinicall y appro ved SPIO par ticles after maturing and loading the DCs with tumor -derived antigenic peptides. These SPIO-labeled DCs, as w ell as 111Inlabeled DCs, w ere coinjected intranodall y into eight stage-III melanoma patients using ultrasound guidance. The higher spatial resolution of MRI allo wed the researchers to identify more l ymph nodes containing migrated DCs then the gamma scintig raphic method. Additionally, MRI revealed that even though the intranodal injections were done under ultrasound guidance, in more than half of the patients the DCs had not been injected into the lymph nodes cor rectly. Immunohistology and e x vivo MRI on the resected lymph nodes confirmed a colocalization of immig rated SPIO-DCs within the T-cell areas of successfully targeted lymph nodes. This study estab lished that MRI cell tracking using iron oxides appears to be w ell suited to monitor cellular therapy in humans. Even with the current success of iron oxides in labeling DCs, this contrast mechanism is not the onl y one being employed for DC cell tracking in MRI. As mentioned before in the contrast sections, a common limitation to all iron o xide-based contrast agents used for cellular MRI is the loss of signal in1H MR images caused by the indirect detection of iron par ticles. To overcome this limitation, Ahrens and colleagues have developed an advanced labeling method based on fluorinated nanoparticles.93,95 Initially, DCs were directly labeled using PFPE

713

and then, more recentl y, the lab has follo wed adoptively transferred T cells after being labeled with a PFPE nanoparticle composition (see above).

T CELL IMAGING Although most interest to date in the use of pre-labeled cells for cellular MRI has been in the use of stem and progenitor cells, the ability to monitorT cell trafficking is likely to ha ve a signif icant future impact on our understanding of the in vi vo immune response. T cells are incredibly active cells, both traf ficking and proliferating in response to immune activation. As cellular MRI allows serial monitoring of the same animal/tissue over time, the determination of the in vivo fate of adoptively transferred T cells, including their distribution, mig ration, and homing to targeted sites, may be possible. Tumor antigen-specific lymphocytes in particular have been used for adoptive transfer and treatment in lymphoma, melanoma, and other malignancies. 169–171 A major hurdle to accurate e valuation of treatment and antitumor ef fects has been the inability to track these l ymphocytes i n vivo with sufficiently high spatial and temporal resolutions. 172 The majority of studies attempting to follow the distribution of T cells in vivo over time have used either radiolabeled173 or bioluminescent cells. 174 These modalities pro vide lo w spatial resolution and, in the case of optic imaging, limited tissue penetration. For an imaging method to ultimately be clinically viab le and allo w the e valuation of both cell delivery and therapeutic effectiveness in patients, it must be noninvasive, nontoxic, and allow an accurate and quantitative determination of the cell-based therapy.175 MRI covers all of these bases b y offering the highest spatial resolution of all noninvasive imaging modalities. One of the major issues with labeling T cells is that they are not phagoc ytic by nature. Of course, the f irst step in imaging of these cells is labeling them with an appropriate contrast agent. Incubation of SPIO with transfection agents or conjugation with uptak efacilitating molecules has been e xperimented with. 176 Efficient intracellular labeling has been achie ved b y linking HIV -derived Tat peptides to CLIO par ticles.66,90,175,177 T cells ha ve also been labeled using other transfection methods such as electroporation (magnetoelectroporation)76 and comple xation to pol yL-lysine.178 Kircher and colleagues 175 were ab le to use their improved CLIO par ticles (CLIO-HD) to label l ymphocytes and develop a novel quantitative, noninvasive, highresolution imaging approach to follow the recruitment of antigen-specific CD8 + T cells to tar get tumors. Their

714

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 5. Early recruitment of cross-linked iron oxide-tat-labeled tumor specific (OT-1) CD8+ T cells is impaired in B16/OVA. CXCL12-high tumors. Axial MRI slices through mouse thighs show a significant and consistent signal reduction in B16/OVA.MSCV tumors compared with B16/OVA.CXCL12-high tumors, indicating that more labeled OT-1 T cells had been recruited into the B16/OVA.MSCV tumor. The intensity of T cell recruitment corresponding to dark areas (left panel) is also evident on the R2 map (right panel). The number of cells/voxel is indicated in the side bar. Adapted from Vianello F et al.177

findings showed that the labeled T cells remained viab le and maintained the same proliferation patter n and c ytotoxic capacity on tar get cells as the unlabeled l ymphocytes. Their g roup w as ab le to describe the 3D distribution of inf iltrating CD8+ T cells across the whole tumor while simultaneously assessing the tumor v olume over time as a function of T cell acti vity. Most recentl y, this group has used their in vivo MRI T cell tracking techniques to elucidate possib le mechanisms of tumor immune escape. 177 They have shown that a chemokine, CXCL12, may act as a chemorepellent to tumor -specific T cells and may be responsible for decreased T cell colocalization in certain tumor types (Figure 5). Thus, MRI represents a sophisticated tool for noninvasive imaging of the mig ration of a v ariety of immune cells in vivo. With the continuing development of advanced labeling techniques and the introduction of new MRI scanners at increasingl y higher f ield strengths, it may soon be possible for a clinical radiologist to track and quantify the cells in volved in cur rent immunotherapies. The infor mation gained with this technology will hopefull y aid in the creation of more effective cell therapies for cancer , MS, atherosclerosis, and a variety of other diseases.

THE ROOTS OF STEM CELL THERAPY Stem cells have significant potential in therapy for many clinical conditions because they possess mechanisms for overcoming natural cell replication limits and are y et capable of differentiation in response to e xtrinsic cues in vitro or in the transplant microen vironment. Stem cells can be isolated from di verse sources and are def ined functionally as cells with the capacity for self rene wal

and the ability to yield multiple, distinct, dif ferentiated cell types. Embr yonic stem cells are deri ved from the inner cell mass of the b lastocyst and are pluripotent— able to give rise to all the cell types of the deri vatives of all three embr yonic ger m la yers.179 Isolation of these cells from humans poses ethical concer ns, and due to a lack of control of their unlimited potential for proliferation and differentiation, risks of tumor formation or inappropriate dif ferentiation follo wing transplantation are present.16,180,181 In contrast, organ or tissue-specific stem cells are multipotent or unipotent pro genitors that contribute to the formation of fetal tissues and organs. In the adult, these cells continue to proliferate and pro vide regenerative capacity in cer tain tissues. Compared with fetal sources, adult stem cells can be isolated in g reater quantity for autologous or heterologous transplantation. In addition, stem cell lines can be estab lished for indef inite replication in tissue culture to pro vide large numbers of cells for systematic analysis and transplantation. One k ey advantage of stem cell line establishment is that the cells may be geneticall y engineered to impro ve their eng raftment or therapeutic suppor t at sites of transplantation.182–186 Equally possib le, gene deletion e xperiments can be performed to remove or suppress tumorigenic signaling pathways to enhance eng raftment and therapeutic potential. Increasing access to e xpandable sources of stem cells presents e xciting possibilities for transplantation research.187 A v ariety of medical conditions, such as myocardial infarction and cerebral ischemia, arise from a loss of cells due to trauma or degenerative disease, and treatment of these conditions b y cell transplantation to replace lost cells is intrinsicall y appealing. Until recently, transplantation research was confined to animal

Cell Voyeurism Using Magnetic Resonance Imaging

models but methods for isolation of various human stem cells have advanced signif icantly. Via systematic e xploration of w hich cells are optimal in a par ticular disease model it will be possible to determine which cell sources and which interventional protocols provide the best therapeutic outcome. As noninvasive MRI of cell transplants in animal models continues to re veal therapeutic possibilities, human stem cells will be readil y a vailable to translate this line of research into human trials.

MESENCHYMAL STEM CELLS Marrow is defined by Merriam Webster as “the choicest of food” or “the seat of animal vigor .” Not surprisingly, bone mar row is a source of hematopoietic stem and progenitor cells and mesench ymal stem cells (MSCs) for autologous cell transplantation. Systemic delivery of MSCs is minimall y in vasive and highl y suitab le for repeated administration of cell therap y. Adult MSCs possess the ability to dif ferentiate into chondroc ytes, osteocytes, adipoc ytes, and m yocytes188–190 and are most commonl y applied as cell therap y for inf arcted myocardium191–193 and focal CNS ischemia.180,182,183,194 MSCs are thought to be attracted by factors secreted by injured or diseased cardiac tissue 195 where, upon recruitment, the y themselv es may produce f actors that support angiogenesis, reduce cardiom yocyte apoptosis, and improve contractile activity.196,197 However, in contrast to mobilization of MSCs from bone mar row, cultured MSCs appear to become trapped upon direct systemic delivery while en route through the coronar y vasculature.193,198,199 Improved cardiac function w as obtained via injection of MSCs directly into the anterior wall of the left v entricle.200,201 Though MR tracking of SPIO-labeled MSC integration in cardiac tissue is quite challenging due to the rapid contraction rate of the rodent heart, MSCs w ere recently tracked using nonin191,193,202 vasive MR for one month after injection. Equivocal data on MR tracking of MSC in the hear t along with limited clinical ef ficacy in randomized clinical trials203–205 indicate that cell therapy using MSC for patients with ischemic hear t disease is not y et optimal. Ongoing de velopments in nonin vasive imaging of MSCs in heart will therefore pla y a prominent role in the optimization of MSC deli very, inte gration, and survival. MSC transplants have also been reported to attenuate functional def icits in rats follo wing CNS injury206,207 or middle cerebral ar tery occlusion model of cerebral ischemia/strok e.180,208 Rather than cellular differentiation and eng raftment, it is thought that

715

cytokine and growth factor secretion by MSCs supports the sur vival of cells in the lesion site. 209 While the mechanisms responsib le for functional impro vement remain to be identif ied, MSCs modif ied to overexpress and secrete f actors such as glial cell line deri ved trophic f actor,182 brain deri ved neurotrophic f actor,184,186 and placental g rowth f actor 185 exhibit improved neuroprotective capability following cerebral ischemia. It will be of interest to nonin vasively follow the mig ration of these geneticall y engineered MSC to sites of ischemic damage by MR. For example, following photochemical lesioning of a localized region in the cerebral cor tex, SPIO-labeled MSCs injected into the contralateral cerebral v entricle mig rated to the lesion site within 2 weeks of administration. 180

SING MORE SONGS Tissue-specific stem cells ensure proper organ function following therapeutic replacement of cells lost due to apoptosis, injury, or disease. Stem cells have long been considered absent from the adult CNS due to its limited capacity for re generation. Studies in canaries disco vered that neurogenesis is responsible for song behavior in these animals. 210 These f indings provided the impetus for other investigators to search the brains of mammals for similar neuro genic capacity . The adult subventricular zone (SVZ) of the lateral v entricles and subgranular zone (SGZ) of the dentate gyr us are tw o niches which exhibit the capacity to generate ne w neurons.211,212 Neurogenesis outside these “brain mar row” regions is either very limited or nonexistent in the normal adult ner vous system. 213 However, in patholo gic states or follo wing injury, the potential for neuro genesis outside the SVZ and SGZ increases possib ly due to increased trophic suppor t or a local reduction in inhibitory factors.214–220 In addition, complex environments,221 learning,222 and physical activity223 have all been shown to enhance adult neuro genesis in the SGZ whereas stress and sleep depri vation compromise this process.224,225 Neurogenesis in SVZ does not appear to be modulated by environmental factors, possibly due to the relative distance of stem cells in the SVZ from sites of neuronal activity.226,217 In rodents, neural stem cells in the SVZ gi ve rise to neurob lasts that mig rate along the rostral mig ratory stream and dif ferentiate into functional interneurons in the olfactory bulb. Following intracerebroventricular (ICV) injection of Bangs par ticles, the mig ration of neuronal precursors w as monitored noninvasively by in vivo MR for up to 3 weeks227 (Figure 6).

716

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 6. Noninvasive MR detection of neural stem cell migration in the rostral migratory stream of the rodent olfactory bulb. Following the addition of carboxyl groups, microspheres containing iron oxides particles were incubated with epidermal growth factor to promote receptor-mediated internalization by cells in the subventricular zone (SVZ) and stereotactically injected (1.5 × 107–1.5 × 108 particles in 5–50 µL) into the right lateral ventricle of young rats (6 weeks old). Serial MR imaging in two rats showed early signs of neural precursor cell migration from the anterior border of the SVZ (large hypointense region) after 1 week (B and F). Cell migration along the RMS continued into the second week (C and G) and cells infiltrated the olfactory bulb after 3 weeks (D and H). Adapted from Shapiro EM et al.227

In the adult SGZ, neural stem cells contrib ute new granule cell neurons. The pre vailing consensus is that these ne w neurons arise from cells of the radial glial lineage that e xpress glial f ibrillary acid protein, pol ysialylated for m of neural cell adhesion molecule, and proliferate in response to epider mal g rowth f actor (EGF) and basic fibroblast growth factor (bFGF).228–230 The ability to har ness the po wer of neural stem cells presents e xciting possibilities for nonin vasive MRI of cell therap y follo wing traumatic CNS injur y and in models of CNS disease w here cell attrition is implicated.

Neural stem cells isolated from either rodents or human divide continuously when cultured in a serum-free medium containing bFGF and EGF 231–233 and for m neurospheres that are ab le to gi ve rise to neurons, astrocytes, and oligodendroc ytes.25,234–238 Neurospheres can be injected into CNS parench yma for focal deli very or into the cerebral v entricles for deli very to g reater expanses of diseased CNS tissue. Transfection agent (dendrimer)-based methodology for labeling and in vi vo MR tracking of neural stem cells was established in 2001, allowing investigators to follow cells for at least 6 w eeks in vivo.67 This labeling approach w as readily applied to label human neural precursors 81,239 and human embr yonic stem cell (hESC) deri ved neurospheres 25 for MR detection in mouse CNS for up to 4 weeks. Though SPIO label generally appears to be well retained over long periods b y neural stem cells, depending on the cell type detection of SPIO-labeled stem cells beyond 2 w eeks may be impaired in vi vo w hen asymmetric di vision is present, ie, in the case of immor talized cells.81,82 In MS, subcor tical oligodendrocytes are lost due to attack by either autoimmune T cells, axonal loss, or both phenomena.240,241 Activated and possib ly nonacti vated lymphocytes can tra verse the endothelial cell la yer between blood and brain, kno wn as the b lood brain barrier (BBB), and make contact with CNS tissue even in the absence of inflammation. Following antigen recognition, as in the case of infection or autoimmune attack, T cells may initiate an inflammator y cascade that compromises the inte grity of the BBB allo wing entr y of g reater numbers of leuk ocytes and macrophages into the CNS parenchyma.242 Multipotent neural stem cells pro vide an opportunity to replace lost neurons and oligodendrocytes, and recent repor ts suggest that these cells possess immunomodulatory characteristics that may be favorable for cell therap y in MS. 235,236,243,244 The dynamic trio of CNS-immune, transplant-CNS, and transplant-immune interactions that are characteristics of rodent and primate models of MS is increasingl y addressed b y noninvasive MRI.138,140,245 ICV injection of neurospheres has been shown to decrease the se verity of e xperimental aller gic encephalomyelitis (EAE, model of MS) in mice.236,246 Tracking of SPIO-labeled hESCs follo wing ICV injection in mice with EAE re vealed that these cells mig rated in response to inflammatory cues within 1 week of transplantation and that migration distance cor related with severity of disease. 25 Current efforts are directed to ward noninvasive MR identif ication of optimal periods of intervention235,236 and MR tracking of neural stem cell delivery and mig ration toward sites of inflammation and demyelination in EAE. 25,26,28

Cell Voyeurism Using Magnetic Resonance Imaging

MRI OF EMBRYO DEVELOPMENT After conception, the zygote di vides to for m a ball of rapidly multipl ying cells called a b lastocyst. When the blastocyst ar rives in the uter us, it adheres and mig rates into the uterine w all so that a placenta ma y develop and nourish the embr yo. All the w hile, cells within the b lastocyst mig rate to for m ectoder mal, endoder mal, and mesodermal layers in a process called gastr ulation. Cells within these layers eventually migrate to their target destinations in the developing embryo where they are specified as components of liver, heart, bone, brain, and other developing organs. Noninvasive MRI of embryo development offers the oppor tunity to obser ve rapid and highl y patterned cell mo vements in a relati vely small FO V. Because serial high-resolution MRI of small mammalian embryos developing within the contractile uter us is very difficult, lar ge, e xternally de veloping amphibian and avian embr yos are prefer red for study. Noninvasive MR analyses of chick embr yo development are not y et available due to motion ar tifacts and signif icant postural changes associated with advanced stages of development. However, gentle cooling of chick embr yos (da y 12 to hatching at da y 20) immediatel y prior to imaging w as recently shown to reduce motion artifacts at 7T providing new oppor tunities for MRI of brain, li ver, and hear t development in ovo.247 Imaging of whole Xenopus laevis embryos is more advanced. Xenopus embryos can be isolated in space, and high-resolu tion MR images can be obtained at shor t intervals (5–10 min) due to small FO V (1 mm). Without contrast agent, investigators were able to apply MRI to follo w mitotic cell di vision in earl y blastomeres248 as well as initial cell di visions, gastr ulation, and the complete embr yonic development of a Xenopus laevis embryo in exquisite detail.249,250 In addition, injection of MR contrast agent into the cytoplasm of a specific cell in the embr yo cluster enables detection of cell pro geny within the de veloping embr yo.251 Finally, nonin vasive MRI of Xenopus embr yos was applied to indirectl y detect the e xpression of β galactosidase ( β-gal) mRNA using a gadolinium-based MR contrast agent that is activated by β-gal activity,252 offering the possibility to noninvasively track cell lineages using specif ic promoters to drive β-gal expression. Along with Xenopus, the zebra fish is a prime model organism for in vi vo MRI studies of v ertebrate development because the lar vae are nearl y transparent and the y develop in 60 hours from fer tilization to free-s wimming larvae. Recently, zebra f ish was noninvasively imaged at 7T with 47 micron spatial resolution following Dotarem® 253 injection at the one to four cell stage. Zebra f ish

717

embryos have under gone numerous mutant screens and other genetic analysis, which allows the role of candidate molecules to be tested in MRI experiments. Furthermore, stable transgenic lines that recapitulate the e xpression pattern of a specif ic gene of interest can be generated by raising fish to adulthood and screening founders for germ line transmission of the transgene.

TOWARD NONINVASIVE QUANTIFICATION OF CELL MIGRATION Though population anal yses of cells numbering in the thousands to millions are sufficient for ensuring delivery or migration of cells to specif ic anatomic sites, demand for more ref ined nonin vasive measurements of cell migratory responses to soluble factors or genetic manipulation is on the horizon. In addition, noninvasive MR analyses of stem cell mig ration and metastasis require the ability to track indi vidual cells or small g roups of cells. In addition to higher f ield strength, customdesigned g radient and RF coils and optimized pulse sequences (eg, 3D FIESTA) have improved the sensitivity of microimaging of SPIO-labeled cells in vi vo. MRI of SPIO-labeled cells g rown in vitro in collagen gels, gelatin, or agarose has been used to test the le vel of cell detection of clinical grade 1.5T254–257 and 3T258,259 scanners and commercially available gradients. However, it is not kno wn ho w w ell clinical MR scanners will detect individual cells in the presence of motion ar tifacts and larger FOV. In vivo limits of detection range from 100 to 500 cells containing 10 to 60 pico grams of SPIO per cell71,80,90,260 down to the single cell le vel using Bangs particles55,59 yielding 100 picograms of iron per cell. Basic parameters for the nonin vasive study of single cell migration include the quantification of cell displacement, v elocity of cell mig ration, and chemotactic responses to drugs, biologic factors, or genetic manipulation. Muscle cells and ger m cells ha ve been sho wn to migrate less than 10 µm/h261,262 and may traverse roughly two, 100 µm, voxels per day. In contrast, rapid mig rating cells such as T l ymphocytes (660 µm/h)263 and breast cancer cells (240 µm/h)264 may tra verse se veral v oxels during the course of an a verage scan and man y v oxels over the course of 1 day. It will be of interest to establish means to distinguish indi vidual cells over time by their relative spatiotemporal positions and demarcate cellular regions of interest to perfor m quantitative 3D anal ysis of cell mig ration in vi vo. To obtain detailed data describing cell position over time, higher magnetic field strength (3T or higher) and 3D MRI sequences that offer

718

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

resolution near or belo w 100 microns are necessar y and are often referred to as MR microscopy. Ben-Hur and colleagues25 have perfor med serial MRI of SPIO-labeled neurospheres in mice with EAE and from these data calculated an a verage speed of appro ximately 200 µm/day for the f irst 6 days. The proverbial “needle in a haystack,” periodic noninvasive sur veillance of a specif ic cell population in a vast in vi vo milieu poses signif icant technical challenges. Cur rent challenges in the application of noninvasive MR cell imaging include as follo ws: (1) obtaining cellular spatial resolution with reasonab le scan duration, (2) the ability to track the same cell across time, (3) impro ved temporal resolution for rapidly migrating cells, and (4) routine nonbiased methods for quantitati ve anal ysis of cell mig ration. MRI studies commonly apply 2D multislice analysis to determine the general position of labeled cells in the body . However, 2D multislice MRI sequences are often insufficient for quantitati ve anal ysis of cell mig ration because indi vidual cells are dif ficult to resolv e in the thick 2D slices (> 400 microns) necessar y to ensure the presence of adequate nuclear spins in each v oxel (signal-to-noise ratio). Because high-resolution imaging of a 3D slab of tissue significantly increases imaging time, even at high magnetic f ield strength, it is helpful to know, a priori, the location of the cells of interest within the subject so that the FOV can be adjusted to achieve a reasonable MR scan duration for li ve subjects. F or example, approximately 30 to 45 min of imaging time is needed to induce sedation, position the subject in the MR scanner , obtain se veral 2D multislice pilot scans, and a high-resolution 3D scan. For live subjects, motion artifacts are possib le and signal a veraging to impro ve signal to noise in high-resolution scans is very restricted due to increased periods of sedation and longer imaging time. Using anatomic landmarks or measured coordinates, the re gion of interest (R OI) is isolated for 3D MRI at re gular inter vals to follo w the position and migration of labeled cells o ver time. Because subject orientation and FOV placement can v ary between independent 3D MRI sessions, the image data must be digitally nor malized to a template for the distance of a cellular ROI from a point of reference to be determined. Finally, at present, e x vivo histological correlation with in vivo MR data is necessary to calibrate and confirm in vivo analyses of cell mig ration. Technical challenges associated with anal yses of cell migration by noninvasive MRI include the fact that the relative position of tw o or more h ypointense cellular ROIs may be dif ficult to distinguish w hen labeled cells are very close to one another or when foci of iron

rich er ythrocytes are present. In addition, depending upon the time interval between MRI sessions, it may be challenging to pinpoint the migratory path of a specific cell to calculate mig ration distance and rate. Also, injection of iron o xide-labeled cells can smear labeled cells along the injection path or r upture capillaries introducing iron-rich er ythrocytes w hich are also hypointense w hen deo xygenated. F inally, should the labeled cells die or become phagoc ytosed, contrast agent ma y be deposited nonspecif ically at sites of necrosis or carried distantly by phagocytic cells. Therefore, it is necessar y to conf irm mig ration of labeled cells upon completion of the in vi vo analysis using e x vivo high-resolution 3D MRI follo wed b y dual label immunohistochemistry for dextran75 that comprises the shell of the USPIO or SPIO nanopar ticle and a cellspecific mark er. In addition, visualization of cells expressing GFP or other histolo gic mark ers (ie, BrdU25) can be perfor med to conf irm the f inal location of transplanted cells.

THE FUTURE OF NONINVASIVE MR CELL TRACKING Methods for tagging cells for MR detection are rapidl y evolving, and the speed, sensitivity, and resolution of MRI devices is continually increasing. These advancements not only of fer di verse options for nonin vasive in vi vo cell tracking but will eventually enable facile, high-resolution MR pursuits of the mig ration of indi vidual cells at an y depth in li ving tissue. With the seamless inte gration of biologically compatib le tags, cell biolo gy, genetics, animal behavior and ph ysiology, pharmaceutical agents, tissue transplantation, e xperimental disease models, and imaging technology, noninvasive MRI of cell migration is the pinnacle of interdisciplinar y science. It is anticipated that e xtensive cross-fer tilization of betw een v arious research disciplines will further aid in the establishment of routine methods for the nonin vasive detection and quantification of cell migration by MRI. Moreover, the combination of MRI with nuclear medicine (PET and SPECT) or optic and bioluminescent imaging modalities w ould allow for further characterization of cell transplant viability, function, proliferation, and dif ferentiation. For example, myelin for mation and synapse for mation are highl y complex processes that may be regarded as stringent endpoints for judging dif ferentiation b y neural stem cells. However, it is not known how efficient these processes are executed by cell transplants. In future studies, it will be essential to nonin vasively deter mine the percentage of cells that recapitulate a particular biologic process and the rate and extent to which it happens.

Cell Voyeurism Using Magnetic Resonance Imaging

ACKNOWLEDGMENTS The authors wish to thank our colleagues for insightful discussions pertaining to nonin vasive imaging. Man y thanks to Drs Assaf Gilad, Mike McMahon, and Piotr Walczak.

REFERENCES 1. Dormann D , Weijer CJ . Imaging of cell mig ration. EMBO J 2006;25:3480–93. 2. Massoud TF, Gambhir SS. Molecular imaging in living subjects: seeing fundamental biolo gical processes in a ne w light. Genes De v 2003;17:545–80. 3. Massoud TF, Gambhir SS. Integrating noninvasive molecular imaging into molecular medicine: an evolving paradigm. Trends Mol Med 2007;13:183–91. 4. Paulmurugan R, Massoud TF, Huang J, Gambhir SS. Molecular imaging of dr ug-modulated protein-protein interactions in li ving subjects. Cancer Res 2004;64:2113–9. 5. Cyster JG. L ymphoid or gan de velopment and cell mig ration. Immunol Rev 2003;195:5–14. 6. Gerthoffer WT. Mechanisms of v ascular smooth muscle cell mig ration. Circ Res 2007;100:607–21. 7. Ghashghaei HT, Lai C, Anton ES. Neuronal mig ration in the adult brain: are we there yet? Nat Rev Neurosci 2007;8:141–51. 8. Trainor PA. Specification of neural crest cell formation and migration in mouse embryos. Semin Cell Dev Biol 2005;16:683–93. 9. Yamaguchi H, Wyckoff J, Condeelis J. Cell migration in tumors. Curr Opin Cell Biol 2005;17:559–64. 10. Halin C, Rodrigo Mora J , Sumen C, v on Andrian UH. In vi vo imaging of l ymphocyte traf ficking. Annu Re v Cell De v Biol 2005;21:581–603. 11. Hedberg EL, Kroese-Deutman HC, Shih CK, et al. Methods: a comparative analysis of radiography, microcomputed tomography, and histology for bone tissue engineering. Tissue Eng 2005;11:1356–67. 12. Perea H, Aigner J, Heverhagen JT, et al. Vascular tissue engineering with magnetic nanopar ticles: seeing deeper . J Tissue Eng Re gen Med 2007;1:318–321. 13. Terrovitis JV, Bulte JW , Sar vananthan S, et al. Magnetic resonance imaging of fer umoxide-labeled mesench ymal stem cells seeded on collagen scaffolds-relevance to tissue engineering. Tissue Eng 2006;12:2765–75. 14. Addis RC, Bulte JW, Gearhart JD. Special cells, special considerations: the challenges of bringing embryonic stem cells from the laboratory to the clinic. Clin Phar macol Ther 2008;83:386–9. 15. Dietz AB, Padley DJ, Gastineau D A. Infrastructure development for human cell therap y translation. Clin Phar macol Ther 2007;82:320–4. 16. Swijnenburg RJ, van der Bo gt KE, Sheikh AY, et al. Clinical hurdles for the transplantation of cardiom yocytes deri ved from human embr yonic stem cells: role of molecular imaging. Cur r Opin Biotechnol 2007;18: 38–45. 17. Levin EG. Cancer therap y through control of cell mig ration. Cur r Cancer Drug Targets 2005;5:505–18. 18. Modo M, Hoehn M, Bulte JW . Cellular MR imaging. Mol Imaging 2005;4:143–64. 19. Boretius S, Schmelting B, Watanabe T, et al. Monitoring of EAE onset and progression in the common mar moset monkey by sequential high-resolution 3D MRI. NMR Biomed 2006;19:41–9. 20. Jordan EK, McF arland HI, Le wis BK, et al. Serial MR imaging of experimental autoimmune encephalom yelitis induced b y human white matter or b y chimeric myelin-basic and proteolipid protein in the common mar moset. AJNR Am J Neuroradiol 1999;20: 965–76.

719

21. Nessler S, Boretius S, Stadelmann C, et al. Earl y MRI changes in a mouse model of multiple sclerosis are predictive of severe inflammatory tissue damage. Brain 2007;130(Pt 8):2186–98. 22. Xu S, Jordan EK, Brocke S, et al. Study of relapsing remitting experimental aller gic encephalom yelitis SJL mouse model using MION-46L enhanced in vivo MRI: early histopathological correlation. J Neurosci Res 1998;52:549–58. 23. Zhang J , Hutton G. Role of magnetic resonance imaging and immunotherapy in treating multiple sclerosis. Annu Re v Med 2005;56:273–302. 24. Ben-Hur T, Einstein O , Bulte JW. Stem cell therap y for m yelin diseases. Curr Drug Targets 2005;6:3–19. 25. Ben-Hur T, v an Hees wijk RB, Einstein O , et al. Serial in vi vo MR tracking of magneticall y labeled neural spheres transplanted in chronic EAE mice. Magn Reson Med 2007;57:164–71. 26. Bulte JW, Ben-Hur T, Miller BR, et al. MR microscop y of magnetically labeled neurospheres transplanted into the Le wis EAE rat brain. Magn Reson Med 2003;50:201–5. 27. Politi LS. MR-based imaging of neural stem cells. Neuroradiolo gy 2007;49:523–34. 28. Politi LS, Bacigaluppi M, Brambilla E, et al. Magnetic-resonancebased tracking and quantification of intravenously injected neural stem cell accumulation in the brains of mice with e xperimental multiple sclerosis. Stem Cells 2007;25:2583–92. 29. Meier DS, Guttmann CR. MRI time series modeling of MS lesion development. Neuroimage 2006;32:531–7. 30. Meier DS, Weiner HL, Guttmann CR. Time-series modeling of multiple sclerosis disease acti vity: a promising windo w on disease progression and repair potential? Neurotherapeutics 2007;4: 485–98. 31. de Vries IJ, Lesterhuis WJ, Barentsz JO , et al. Magnetic resonance tracking of dendritic cells in melanoma patients for monitoring of cellular therapy. Nat Biotechnol 2005;23:1407–13. 32. Frank JA, Miller BR, Arbab AS, et al. Clinicall y applicable labeling of mammalian and stem cells b y combining super paramagnetic iron oxides and transfection agents. Radiology 2003;228:480–7. 33. Frank JA, Zywicke H, Jordan EK, et al. Magnetic intracellular labeling of mammalian cells by combining (FDA-approved) superparamagnetic iron o xide MR contrast agents and commonl y used transfection agents. Acad Radiol 2002;9 Suppl 2:S484–7. 34. Toso C, Vallee, JP, Morel, P, et al. Clinical magnetic resonance imaging of pancreatic islet g rafts after iron nanopar ticle labeling. Am J Transplant 2008;8:701–6. 35. Zhu J, Zhou L, XingW u F. Tracking neural stem cells in patients with brain trauma. N Engl J Med 2006;355:2376–8. 36. Lauterbur PC. Image formation by induced local interactions. Examples employing nuclear magnetic resonance. 1973. Clin Or thop Relat Res 1989;244:3–6. 37. Brooks RA, Vymazal J, Bulte JW, et al. Comparison of T2 relaxation in blood, brain, and ferritin. J Magn Reson Imaging 1995;5:446–50. 38. Bulte JW, Vymazal J, Brooks RA, et al. Frequency dependence of MR relaxation times. II. Iron o xides. J Magn Reson Imaging 1993;3: 641–8. 39. Vymazal J, Bulte JW, Frank JA, et al. F requency dependence of MR relaxation times. I. P aramagnetic ions. J Magn Reson Imaging 1993;3:637–40. 40. Granot D, Kunz-Schughart LA, Neeman M. Labeling fibroblasts with biotin-BSA-GdDTPA-FAM for tracking of tumor -associated stroma b y fluorescence and MR imaging. Magn Reson Med 2005;54:789–97. 41. Aime S, Barge A, Cabella C, et al. Targeting cells with MR imaging probes based on paramagnetic Gd(III) chelates. Cur r Phar m Biotechnol 2004;5:509–18. 42. Anderson SA, Lee KK, F rank JA. Gadolinium-fullerenol as a paramagnetic contrast agent for cellular imaging. In vest Radiol 2006;41:332–8.

720

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

43. Biancone L, Crich SG, Cantaluppi V, et al. Magnetic resonance imaging of gadolinium-labeled pancreatic islets for e xperimental transplantation. NMR Biomed 2007;20:40–8. 44. Crich SG, Biancone L, Cantaluppi V, et al. Improved route for the visualization of stem cells labeled with a Gd-/Eu-chelate as dual (MRI and fluorescence) agent. Magn Reson Med 2004;51:938–44. 45. Crich SG, Lanzardo S, Alberti D, et al. Magnetic resonance imaging detection of tumor cells b y tar geting lo w-density lipoprotein receptors with gd-loaded lo w-density lipoprotein par ticles. Neoplasia 2007;9:1046–56. 46. Daldrup-Link HE, Rudelius M, Metz S, et al. Cell tracking with gadophrin-2: a bifunctional contrast agent for MR imaging, optical imaging, and fluorescence microscop y. Eur J Nucl Med Mol Imaging 2004;31:1312–21. 47. Giesel FL, Stroick M, Griebe M, et al. Gadofluorine m uptake in stem cells as a new magnetic resonance imaging tracking method: an in vitro and in vivo study. Invest Radiol 2006;41:868–73. 48. Vuu K, Xie J, McDonald MA, et al. Gadolinium-rhodamine nanoparticles for cell labeling and tracking via magnetic resonance and optical imaging. Bioconjug Chem 2005;16:995–9. 49. Aoki I, Takahashi Y, Chuang KH, et al. Cell labeling for magnetic resonance imaging with the T1 agent manganese chloride. NMR Biomed 2006;19:50–9. 50. Gilad AA, Walczak P, McMahon MT, et al. MR tracking oftransplanted cells with positi ve contrast using managnese o xide nanopar ticles. Magn Reson Med 2008. [In press] 51. Weissleder R, Elizondo G, Wittenberg J , et al. Ultrasmall superparamagnetic iron o xide: an intra venous contrast agent for assessing lymph nodes with MR imaging. Radiolo gy 1990;175: 494–8. 52. Weissleder R, Elizondo G, Wittenberg J, et al. Ultrasmall super paramagnetic iron o xide: characterization of a ne w class of contrast agents for MR imaging. Radiology 1990;175:489–93. 53. Bulte JW, Kraitchman DL. Iron oxide MR contrast agents for molecular and cellular imaging. NMR Biomed 2004;17:484–99. 54. Dunning MD, K ettunen MI, Ffrench Constant C, et al. Magnetic resonance imaging of functional Schw ann cell transplants labelled with magnetic microspheres. Neuroimage 2006;31: 172–80. 55. Heyn C, Ronald JA, Mackenzie LT, et al. In vivo magnetic resonance imaging of single cells in mouse brain with optical v alidation. Magn Reson Med 2006;55:23–29. 56. Heyn, C, Ronald J A, Ramadan SS, et al. In vi vo MRI of cancer cell fate at the single-cell le vel in a mouse model of breast cancer metastasis to the brain. Magn Reson Med 2006;56:1001–10. 57. Hill JM, Dick AJ, Raman VK, et al. Serial cardiac magnetic resonance imaging of injected mesench ymal stem cells. Circulation 2003;108:1009–14. 58. Hinds KA, Hill JM, Shapiro EM, et al. Highl y efficient endosomal labeling of progenitor and stem cells with large magnetic particles allows magnetic resonance imaging of single cells. Blood 2003;102:867–72. 59. Shapiro EM, Sharer K, Skrtic S, Koretsky AP. In vivo detection of single cells by MRI. Magn Reson Med 2006;55:242–9. 60. Shapiro EM, Skr tic S, Sharer K, et al. MRI detection of single par ticles for cellular imaging. Proc Natl Acad Sci USA 2004;101: 10901–6. 61. Weissleder R, Cheng HC, Bogdanova A, Bogdanov A Jr. Magnetically labeled cells can be detected b y MR imaging. J Magn Reson Imaging 1997;7:258–63. 62. Bulte JW, Arbab AS, Douglas T, Frank JA. Preparation of magnetically labeled cells for cell tracking b y magnetic resonance imaging. Meth Enzymol 2004;386:275–99. 63. Norman AB, Thomas SR, Pratt RG, et al. Magnetic resonance imaging of neural transplants in rat brain using asuperparamagnetic contrast agent. Brain Res 1992;594:279–83.

64. Hawrylak N, Ghosh P, Broadus J, et al. Nuclear magnetic resonance (NMR) imaging of iron o xide-labeled neural transplants. Exp Neurol 1993;121:181–92. 65. Bulte JW, Hoekstra Y, Kamman RL, et al. Specif ic MR imaging of human l ymphocytes b y monoclonal antibody-guided de xtranmagnetite particles. Magn Reson Med 1992;25:148–57. 66. Josephson L, Tung CH, Moore A, Weissleder R. High-ef ficiency intracellular magnetic labeling with novel superparamagnetic-Tat peptide conjugates. Bioconjug Chem 1999;10:186–91. 67. Bulte JW, Douglas T, Witwer B , et al. Magnetodendrimers allo w endosomal magnetic labeling and in vi vo tracking of stem cells. Nat Biotechnol 2001;19:1141–7. 68. Zhang RL, Zhang L, Zhang ZG, et al. Mig ration and dif ferentiation of adult rat sub ventricular zone progenitor cells transplanted into the adult rat striatum. Neuroscience 2003;116:373–82. 69. Zhang ZG, Jiang Q, Zhang R, et al. Magnetic resonance imaging and neurosphere therapy of stroke in rat. Ann Neurol 2003;53:259–63. 70. Bulte JW, Zhang S, v an Gelderen P, et al. Neurotransplantation of magnetically labeled oligodendroc yte pro genitors: magnetic resonance tracking of cell migration and myelination. Proc Natl Acad Sci USA 1999;96:15256–61. 71. Hoehn M, Küster mann E, Blunk J , et al. Monitoring of implanted stem cell mig ration in vi vo: a highl y resolv ed in vi vo magnetic resonance imaging in vestigation of e xperimental strok e in rat. Proc Natl Acad Sci USA 2002;99:16267–72. 72. Matuszewski L, Persigehl T, Wall A, et al. Cell tagging with clinically approved iron oxides: feasibility and effect of lipofection, particle size, and surf ace coating on labeling ef ficiency. Radiolo gy 2005;235:155–61. 73. Arbab AS, Yocum GT, Rad AM, et al. Labeling of cells with fer umoxides-protamine sulfate complexes does not inhibit function or differentiation capacity of hematopoietic or mesench ymal stem cells. NMR Biomed 2005;18:553–9. 74. Reynolds F, Weissleder R, Josephson L. Protamine as an eff icient membrane-translocating peptide. Bioconjug Chem 2005;16:1240–5. 75. Walczak P, Kedziorek DA, Gilad AA, et al. Instant MR labeling of stem cells using magnetoelectroporation. Magn Reson Med 2005;54:769–74. 76. Walczak P, Ruiz-Cabello J, Kedziorek DA, et al. Magnetoelectroporation: improved labeling of neural stem cells and leukocytes for cellular magnetic resonance imaging using a single FD Aapproved agent. Nanomedicine 2006;2:89–94. 77. Arbab AS, Bashaw LA, Miller BR, et al. Characterization of bioph ysical and metabolic properties of cells labeled with superparamagnetic iron o xide nanopar ticles and transfection agent for cellular MR imaging. Radiology 2003;229:838–46. 78. Bulte JW, Kraitchman DL, Mackay AM, Pittenger MF. Chondrogenic differentiation of mesenchymal stem cells is inhibited after magnetic labeling with fer umoxides. Blood 2004;104:3410–2; author reply 3412–3. 79. Kostura L, Kraitchman DL, Macka y AM, et al. F eridex labeling of mesenchymal stem cells inhibits chondro genesis but not adipogenesis or osteogenesis. NMR Biomed 2004;17:513–7. 80. Magnitsky S, Watson DJ, Walton RM, et al. In vivo and ex vivo MRI detection of localized and disseminated neural stem cell g rafts in the mouse brain. Neuroimage 2005;26:744–54. 81. Neri M, Mader na C, Cavazzin C, et al. Ef ficient in vitro labeling of human neural precursor cells with super paramagnetic iron o xide particles: relevance for in vivo cell tracking. Stem Cells 2007. 82. Walczak P, Kedziorek DA, Gilad AA, et al. Applicability and limitations of MR tracking of neural stem cells with asymmetric cell division and rapid turnover: the case of the shiverer dysmyelinated mouse brain. Magn Reson Med 2007;58:261–9. 83. Nitin N, LaConte LE, Zurkiya O, et al. Functionalization and peptidebased delivery of magnetic nanopar ticles as an intracellular MRI contrast agent. J Biol Inorg Chem 2004;9:706–12.

Cell Voyeurism Using Magnetic Resonance Imaging

84. Wunderbaldinger P , Josephson L, Weissleder R. Crosslink ed iron oxides (CLIO): a ne w platfor m for the de velopment of tar geted MR contrast agents. Acad Radiol 2002;9 Suppl 2:S304–6. 85. Schellenberger EA, Ho gemann D , Josephson L, Weissleder R. Annexin V-CLIO: a nanoparticle for detecting apoptosis by MRI. Acad Radiol 2002;9 Suppl 2:S310–1. 86. Kaufman CL, Williams M, Ryle LM, et al. Super paramagnetic iron o xide par ticles transacti vator protein-fluorescein isothiocyanate particle labeling for in vi vo magnetic resonance imaging detection of cell mig ration: uptak e and durability . Transplantation 2003;76:1043–6. 87. Pittet MJ, Swirski FK, Re ynolds F, et al. Labeling of immune cells for in vi vo imaging using magnetofluorescent nanopar ticles. Nat Protoc 2006;1:73–9. 88. Santra S, Yang H, Holloway PH, et al. Synthesis of water-dispersible fluorescent, radio-opaque, and paramagnetic CdS: Mn/ZnS quantum dots: a multifunctional probe for bioimaging. JAm Chem Soc 2005;127:1656–7. 89. Bullok KE, Gammon ST, Violini S, et al. Permeation peptide conjugates for in vi vo molecular imaging applications. Mol Imaging 2006;5:1–15. 90. Lewin M, Carlesso N, Tung CH, et al. Tat peptide-derivatized magnetic nanoparticles allow in vi vo tracking and reco very of pro genitor cells. Nat Biotechnol 2000;18:410–4. 91. Bulte JW. Hot spot MRI emerges from the background. Nat Biotechnol 2005;23:945–6. 92. Holland GN, Bottomley PA, Hinshaw WS. 19F magnetic resonance imaging. J Magn Reson 1977;28:133–6. 93. Ahrens ET, Flores R, Xu H, Morel PA. In vivo imaging platform for tracking immunotherapeutic cells. Nat Biotechnol 2005;23: 983–7. 94. Partlow KC, Chen J , Brant J A, et al. 19F magnetic resonance imaging for stem/progenitor cell tracking with multiple unique perfluorocarbon nanobeacons. FASEB J 2007;21:1647–54. 95. Srinivas M, Morel PA, Ernst LA, et al. Fluorine-19 MRI for visualization and quantif ication of cell mig ration in a diabetes model. Magn Reson Med 2007;58:725–34. 96. Ramachandran S, Desai NM, Goers TA, et al. Improved islet yields from pancreas preser ved in perflurocarbon is via inhibition of apoptosis mediated b y mitochondrial pathw ay. Am J Transplant 2006;6:1696–703. 97. Ricordi C, F raker C, Szust J , et al. Impro ved human islet isolation outcome from marginal donors following addition of o xygenated perfluorocarbon to the cold-storage solution. Transplantation 2003;75:1524–7. 98. Wakabayashi T, Tamura M, Nakamura T. P artial liquid v entilation with low-dose perfluorochemical and high-frequenc y oscillation improves oxygenation and lung compliance in a rabbit model of surfactant depletion. Biol Neonate 2006;89:177–82. 99. Gilad AA, Winnard PT Jr , van Zijl PC, Bulte JW . Developing MR reporter genes: promises and pitf alls. NMR Biomed 2007;20: 275–90. 100. Min JJ, Gambhir SS. Gene therapy progress and prospects: noninvasive imaging of gene therap y in li ving subjects. Gene Ther 2004;11:115–25. 101. Contag CH, Bachmann MH. Advances in in vi vo bioluminescence imaging of gene e xpression. Annu Re v Biomed Eng 2002;4: 235–60. 102. Gambhir SS, Bar rio JR, Phelps ME, et al. Imaging adeno viraldirected reporter gene expression in living animals with positron emission tomo graphy. Proc Natl Acad Sci USA 1999;96: 2333–8. 103. Arredondo M, Nunez MT. Iron and copper metabolism. Mol Aspects Med 2005;26:313–27. 104. Chiancone E, Ceci P, Ilari A, et al. Iron and proteins for iron storage and detoxification. Biometals 2004;17:197–202.

721

105. Siah CW, Trinder D , Ol ynyk JK. Iron o verload. Clin Chim Acta 2005;358:24–36. 106. Egyed A. Car rier mediated iron transpor t through er ythroid cell membrane. Br J Haematol 1988;68:483–6. 107. Gilad AA, McMahon MT, Walczak P, et al. Artificial reporter gene providing MRI contrast based on proton e xchange. Nat Biotechnol 2007;25:217–9. 108. Koretsky A, Lin Y-J, Schorle H, Jaenish R. Genetic control of MRI contrast by expression of the transferrin receptor. In: Proceedings of the International Society of Magnetic Resonance Medicine. 1996. 109. Kresse M, Wagner S, Pfefferer D, et al. Targeting of ultrasmall superparamagnetic iron oxide (USPIO) particles to tumor cells in vi vo by using transfer rin receptor pathw ays. Magn Reson Med 1998;40:236–42. 110. Moore A, Josephson L, Bhorade RM, et al. Human transferrin receptor gene as a mark er gene for MR imaging. Radiolo gy 2001;221:244–50. 111. Weissleder R, Moore A, Mahmood U, et al. In vi vo magnetic resonance imaging of transgene expression. Nat Med 2000;6:351–5. 112. Theil EC. Ferritin: at the crossroads of iron and oxygen metabolism. J Nutr 2003;133(5 Suppl 1):S1549–53. 113. Gossuin Y, Muller RN, Gillis P. Relaxation induced by ferritin: a better understanding for an improved MRI iron quantification. NMR Biomed 2004;17:427–32. 114. Gottesfeld Z, Neeman M. Ferritin effect on the transverse relaxation of w ater: NMR microscop y at 9.4 T. Magn Reson Med 1996;35:514–20. 115. Vymazal J, Brooks RA, Bulte JW, et al. Iron uptake by ferritin: NMR relaxometry studies at lo w iron loads. J Inor g Biochem 1998;71:153–7. 116. Vymazal J, Zak O , Bulte JW, et al. T1 and T 2 of fer ritin solutions: effect of loading factor. Magn Reson Med 1996;36:61–5. 117. Cohen B , Dafni H, Meir G, et al. F erritin as an endo genous MRI reporter for noninvasive imaging of gene expression in C6 glioma tumors. Neoplasia 2005;7:109–17. 118. Cohen B, Ziv K, Plaks V, et al. MRI detection of transcriptional regulation of gene e xpression in transgenic mice. Nat Med 2007;13:498–503. 119. Genove G, DeMarco U , Xu H, et al. A new transgene repor ter for in vivo magnetic resonance imaging. Nat Med 2005;11:450–4. 120. Deans AE, Wadghiri YZ, Bernas LM, et al. Cellular MRI contrast via coexpression of transfer rin receptor and fer ritin. Magn Reson Med 2006;56:51–9. 121. Weissleder R, Simono va M, Bo gdanova A, et al. MR imaging and scintigraphy of gene expression through melanin induction. Radiology 1997;204:425–9. 122. Alfke H, Stoppler H, Nock en F , et al. In vitro MR imaging of re gulated gene e xpression. Radiolo gy 2003;228:488–92. 123. Goffeney N , Bulte JW, Duyn J , et al. Sensiti ve NMR detection of cationic-polymer-based gene deli very systems using saturation transfer via proton exchange. J Am Chem Soc 2001;123:8628–9. 124. McMahon MT , Gilad AA, DeLiso MA, et al. Ne w multi-color polypeptide DIA CEST contrast agents for MR imaging. Magn Reson Med. [In press] 125. Stark DD, Weissleder R, Elizondo G, et al. Super paramagnetic iron oxide: clinical application as a contrast agent for MR imaging of the liver. Radiology 1988;168:297–301. 126. Weissleder R. Liver MR imaging with iron oxides: toward consensus and clinical practice. Radiology 1994;193:593–5. 127. Harisinghani MG, Barentsz J, Hahn PF, et al. Noninvasive detection of clinically occult l ymph-node metastases in prostate cancer . N Engl J Med 2003;348:2491–9. 128. Vassallo P, Matei C, Heston WD, et al. AMI-227-enhanced MR lymphography: usefulness for dif ferentiating reacti ve from tumor bearing lymph nodes. Radiology 1994;193:501–6.

722

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

129. Anzai Y, Blackw ell KE, Hirscho witz SL, et al. Initial clinical experience with de xtran-coated super paramagnetic iron o xide for detection of l ymph node metastases in patients with head and neck cancer. Radiology 1994;192:709–15. 130. Bellin MF, Roy C, Kink el K, et al. L ymph node metastases: safety and effectiveness of MR imaging with ultrasmall super paramagnetic iron o xide par ticles–initial clinical e xperience. Radiolo gy 1998;207:799–808. 131. Rogers JM, Jung CW , Le wis J , Groman EV . Use of USPIOinduced magnetic susceptibility ar tifacts to identify sentinel lymph nodes and lymphatic drainage patterns. I. Dependence of artifact size with subcutaneous Combide x dose in rats. Magn Reson Imaging 1998;16:917–23. 132. Anzai Y, Piccoli CW , Outw ater EK, et al. Ev aluation of neck and body metastases to nodes with fer umoxtran 10-enhanced MR imaging: phase III safety and ef ficacy study . Radiolo gy 2003;228:777–88. 133. Harada T, Tanigawa N, Matsuki M, et al. Ev aluation of lymph node metastases of breast cancer using ultrasmall super paramagnetic iron o xide-enhanced magnetic resonance imaging. Eur J Radiol 2007;63:401–7. 134. Koh DM, Brown G, Temple L, et al. Rectal cancer: mesorectal lymph nodes at MR imaging with USPIO v ersus histopatholo gic f indings—initial observations. Radiology 2004;231:91–9. 135. Memarsadeghi M, Riedl CC, Kaneider A, et al. Axillary lymph node metastases in patients with breast carcinomas: assessment with nonenhanced v ersus uspio-enhanced MR imaging. Radiolo gy 2006;241:367–77. 136. Choi SH, Moon WK, Hong JH, et al. L ymph node metastasis: ultrasmall super paramagnetic iron o xide-enhanced MR imaging v ersus PET/CT in a rabbit model. Radiolo gy 2007;242: 137–43. 137. Dousset V, Delalande C, Ballarino L, et al. In vivo macrophage activity imaging in the central ner vous system detected b y magnetic resonance. Magn Reson Med 1999;41:329–33. 138. Rausch M, Hiestand P, Baumann D, et al. MRI-based monitoring of inflammation and tissue damage in acute and chronic relapsing EAE. Magn Reson Med 2003;50:309–14. 139. Brochet B , Deloire MS, Touil T, et al. Earl y macrophage MRI of inflammatory lesions predicts lesion se verity and disease de velopment in relapsing EAE. Neuroimage 2006;32:266–74. 140. Dousset V, Brochet B , Deloire MS, et al. MR imaging of relapsing multiple sclerosis patients using ultra-small-par ticle iron o xide and compared with gadolinium. AJNR Am J Neuroradiol 2006;27:1000–5. 141. Stoll G, Jander S, Schroeter M. Inflammation and glial responses in ischemic brain lesions. Prog Neurobiol 1998;56:149–71. 142. Kleinschnitz C, Bendszus M, F rank M, et al. In vi vo monitoring of macrophage infiltration in experimental ischemic brain lesions by magnetic resonance imaging. J Cereb Blood Flo w Metab 2003;23:1356–61. 143. Nighoghossian N, Wiart M, Cakmak S, et al. Inflammatory response after ischemic strok e: a USPIO-enhanced MRI study in patients. Stroke 2007;38:303–7. 144. Rausch M, Sauter A, Frohlich J, et al. Dynamic patter ns of USPIO enhancement can be obser ved in macrophages after ischemic brain damage. Magn Reson Med 2001;46:1018–22. 145. Wiart M, Da voust N, Pialat JB , et al. MRI monitoring of neuroinflammation in mouse focal ischemia. Strok e 2007;38:131–7. 146. Beckmann N, Cannet C, Fringeli-Tanner M, et al. Macrophage labeling by SPIO as an early marker of allograft chronic rejection in a rat model of kidne y transplantation. Magn Reson Med 2003;49:459–67. 147. Zhang Y, Dodd SJ, Hendrich KS, et al. Magnetic resonance imaging detection of rat renal transplant rejection b y monitoring macrophage infiltration. Kidney Int 2000;58:1300–10.

148. Hauger O , Grenier N , Deminere C, et al. USPIO-enhanced MR imaging of macrophage inf iltration in nati ve and transplanted kidneys: initial results in humans. Eur Radiol 2007;17: 2898–907. 149. Kanno S, Lee PC, Dodd SJ , et al. A novel approach with magnetic resonance imaging used for the detection of lung allo graft rejection. J Thorac Cardiovasc Surg 2000;120:923–34. 150. Kanno S, Wu YJ, Lee PC, et al. Macrophage accumulation associated with rat cardiac allo graft rejection detected b y magnetic resonance imaging with ultrasmall superparamagnetic iron oxide particles. Circulation 2001;104:934–8. 151. Wu YL, Ye Q, Foley LM, et al. In situ labeling of immune cells with iron oxide particles: an approach to detect or gan rejection by cellular MRI. Proc Natl Acad Sci USA 2006;103:1852–7. 152. Stary HC, Chandler AB, Dinsmore RE, et al. A def inition of advanced types of atherosclerotic lesions and a histolo gical classification of atherosclerosis. A report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Circulation 1995;92:1355–74. 153. Fuster V, Fayad ZA, Moreno PR, et al. Atherothrombosis and high-risk plaque: P art II: approaches b y nonin vasive computed tomographic/magnetic resonance imaging. J Am Coll Cardiol 2005;46: 1209–18. 154. Fuster V, Moreno PR, F ayad ZA, et al. Atherothrombosis and highrisk plaque: part I: evolving concepts. J Am Coll Cardiol 2005;46: 937–54. 155. Moreno PR, F alk E, P alacios IF, et al. Macrophage inf iltration in acute coronar y syndromes. Implications for plaque r upture. Circulation 1994;90:775–8. 156. Fayad ZA, Fuster V. Clinical imaging of the high-risk or vulnerable atherosclerotic plaque. Circ Res 2001;89:305–16. 157. Fayad ZA, Fuster V. The human high-risk plaque and its detection by magnetic resonance imaging. Am J Cardiol 2001;88(2A): 42E–45E. 158. Litovsky S, Madjid M, Zarrabi A, et al. Superparamagnetic iron oxidebased method for quantifying recr uitment of monoc ytes to mouse atherosclerotic lesions in vivo: enhancement by tissue necrosis factor-alpha, interleukin-1beta, and interferon-gamma. Circulation 2003;107:1545–9. 159. Ruehm SG, Corot C, Vogt P, et al. Magnetic resonance imaging of atherosclerotic plaque with ultrasmall super paramagnetic par ticles of iron o xide in h yperlipidemic rabbits. Circulation 2001;103:415–22. 160. Schmitz SA, Coupland SE, Gust R, et al. Super paramagnetic iron oxide-enhanced MRI of atherosclerotic plaques in Watanabe hereditable h yperlipidemic rabbits. In vest Radiol 2000;35: 460–71. 161. Schmitz SA, Taupitz M, Wagner S, et al. Magnetic resonance imaging of atherosclerotic plaques using superparamagnetic iron oxide particles. J Magn Reson Imaging 2001;14:355–61. 162. Kooi ME, Cappendijk VC, Cleutjens KB, et al. Accumulation of ultrasmall super paramagnetic particles of iron o xide in human atherosclerotic plaques can be detected b y in vi vo magnetic resonance imaging. Circulation 2003;107:2453–8. 163. Trivedi RA, U-King-Im JM, Gra ves MJ, et al. In vi vo detection of macrophages in human carotid atheroma: temporal dependence of ultrasmall super paramagnetic par ticles of iron o xide-enhanced MRI. Stroke 2004;35:1631–5. 164. Trivedi RA, Malla warachi C, U-King-Im JM, et al. Identifying inflamed carotid plaques using in vi vo USPIO-enhanced MR imaging to label plaque macrophages. Arterioscler Thromb Vasc Biol 2006;26:1601–6. 165. Lipinski MJ, Amirbekian V, F rias JC, et al. MRI to detect atherosclerosis with gadolinium-containing immunomicelles tar geting the macrophage sca venger receptor . Magn Reson Med 2006;56:601–10.

Cell Voyeurism Using Magnetic Resonance Imaging

166. MartIn-Fontecha A, Sebastiani S, Hopken UE, et al. Regulation of dendritic cell mig ration to the draining l ymph node: impact on T l ymphocyte traf fic and priming. J Exp Med 2003;198: 615–21. 167. Ahrens ET, Feili-Hariri M, Xu H, et al. Receptor -mediated endocytosis of iron-o xide par ticles pro vides ef ficient labeling of dendritic cells for in vi vo MR imaging. Magn Reson Med 2003;49:1006–13. 168. Baumjohann D, Hess A, Budinsky L, et al. In vivo magnetic resonance imaging of dendritic cell mig ration into the draining l ymph nodes of mice. Eur J Immunol 2006;36:2544–55. 169. Hanson HL, Doner meyer DL, Ik eda H, et al. Eradication of established tumors by CD8+ T cell adoptive immunotherapy. Immunity 2000;13:265–76. 170. Wang RF, Rosenberg SA. Human tumor antigens for cancer vaccine development. Immunol Rev 1999;170:85–100. 171. Yee C, Thompson JA, Byrd D , et al. Adoptive T cell therap y using antigen-specific C D8+ T cell clones for the treatment of patients with metastatic melanoma: in vi vo persistence, mig ration, and antitumor effect of transfer red T cells. Proc Natl Acad Sci USA 2002;99:16168–73. 172. Yee C, Riddell SR, Greenberg PD. In vivo tracking of tumor-specific T cells. Curr Opin Immunol 2001;13:141–6. 173. Adonai N , Nguy en KN , Walsh J, et al. Ex vi vo cell labeling with 64Cu-pyruvaldehyde-bis(N4-methylthiosemicarbazone) for imaging cell trafficking in mice with positron-emission tomo graphy. Proc Natl Acad Sci USA 2002;99:3030–5. 174. Hardy J, Edinger M, Bachmann MH, et al. Bioluminescence imaging of l ymphocyte traf ficking in vi vo. Exp Hematol 2001;29:1353–60. 175. Kircher MF, Allport JR, Gra ves EE, et al. In vi vo high resolution three-dimensional imaging of antigen-specif ic c ytotoxic T-lymphocyte trafficking to tumors. Cancer Res 2003;63:6838–46. 176. Bulte JW, Laughlin PG, Jordan EK, et al. Tagging of T cells with superparamagnetic iron o xide: uptak e kinetics and relaxometr y. Acad Radiol 1996;3 Suppl 2:S301–3. 177. Vianello F, Papeta N, Chen T, et al. Murine B16 melanomas expressing high le vels of the chemokine stromal-deri ved f actor1/CXCL12 induce tumor -specific T cell chemorepulsion and escape from immune control. J Immunol 2006;176:2902–14. 178. Anderson SA, Shukaliak-Quandt J, Jordan EK, et al. Magnetic resonance imaging of labeled T-cells in a mouse model of multiple sclerosis. Ann Neurol 2004;55:654–9. 179. Thomson JA, Itskovitz-Eldor J, Shapiro SS, et al. Embr yonic stem cell lines deri ved from human b lastocysts. Science 1998;282: 1145–7. 180. Jendelova P, Her ynek V, Urdzik ova L, et al. Magnetic resonance tracking of transplanted bone mar row and embryonic stem cells labeled by iron oxide nanoparticles in rat brain and spinal cord. J Neurosci Res 2004;76:232–43. 181. McDonald JW, Liu XZ, Qu Y, et al. Transplanted embr yonic stem cells sur vive, dif ferentiate and promote reco very in injured rat spinal cord. Nat Med 1999;5:1410–2. 182. Horita Y, Honmou O, Harada K, et al. Intravenous administration of glial cell line-deri ved neurotrophic f actor gene-modif ied human mesenchymal stem cells protects against injur y in a cerebral ischemia model in the adult rat. J Neurosci Res 2006;84: 1495–504. 183. Kurozumi K, Nakamura K, Tamiya T, et al. Mesenchymal stem cells that produce neurotrophic f actors reduce ischemic damage in the rat middle cerebral ar tery occlusion model. Mol Ther 2005;11: 96–104. 184. Kurozumi K, Nakamura K, Tamiya T, et al. BDNF gene-modif ied mesenchymal stem cells promote functional reco very and reduce infarct size in the rat middle cerebral artery occlusion model. Mol Ther 2004;9:189–97.

723

185. Liu H, Honmou O, Harada K, et al. Neuroprotection b y PlGF genemodified human mesench ymal stem cells after cerebral ischaemia. Brain 2006;129(Pt 10):2734–45. 186. Nomura T, Honmou O, Harada K, et al. I.V. infusion of brain-derived neurotrophic factor gene-modified human mesenchymal stem cells protects against injur y in a cerebral ischemia model in adult rat. Neuroscience 2005;136:161–9. 187. Gottlieb DI. Lar ge-scale sources of neural stem cells. Annu Re v Neurosci 2002;25:381–407. 188. Jiang Y, Jahagirdar BN , Reinhardt RL, et al. Pluripotenc y of mesenchymal stem cells deri ved from adult mar row. Nature 2002;418:41–9. 189. Pittenger MF, Mackay AM, Beck SC, et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284:143–7. 190. Toma C, Pittenger MF, Cahill KS, et al. Human mesench ymal stem cells dif ferentiate to a cardiom yocyte phenotype in the adult murine heart. Circulation 2002;105:93–8. 191. Amsalem Y, Mardor Y, Feinberg MS, et al. Iron-o xide labeling and outcome of transplanted mesench ymal stem cells in the inf arcted myocardium. Circulation 2007;116(11 Suppl):I38–I45. 192. He G, Zhang H, Wei H, et al. In vi vo imaging of bone mar row mesenchymal stem cells transplanted into m yocardium using magnetic resonance imaging: a novel method to trace the transplanted cells. Int J Cardiol 2007;114:4–10. 193. Kraitchman DL, Heldman AW, Atalar E, et al. In vivo magnetic resonance imaging of mesenchymal stem cells in myocardial infarction. Circulation 2003;107:2290–3. 194. Shyu WC, Chen CP, Lin SZ, et al. Ef ficient tracking of non-ironlabeled mesenchymal stem cells with serial MRI in chronic stroke rats. Stroke 2007;38:367–74. 195. Takahashi T, Kalka C, Masuda H, et al. Ischemia- and c ytokineinduced mobilization of bone marrow-derived endothelial progenitor cells for neovascularization. Nat Med 1999;5:434–8. 196. Orlic D, Kajstura J, Chimenti S, et al. Bone mar row cells regenerate infarcted myocardium. Nature 2001;410:701–5. 197. Orlic D , Kajstura J , Chimenti S, et al. Mobilized bone mar row cells repair the infarcted heart, improving function and survival. Proc Natl Acad Sci USA 2001;98:10344–9. 198. Barbash, IM, Chouraqui P, Baron J, et al. Systemic delivery of bone marrow-derived mesench ymal stem cells to the inf arcted myocardium: feasibility , cell mig ration, and body distribution. Circulation 2003;108:863–8. 199. Kraitchman DL, Tatsumi M, Gilson WD, et al. Dynamic imaging of allogeneic mesenchymal stem cells trafficking to myocardial infarction. Circulation 2005;112:1451–61. 200. Kehat I, Khimovich L, Caspi O, et al. Electromechanical integration of cardiomyocytes derived from human embryonic stem cells. Nat Biotechnol 2004;22:1282–9. 201. Laflamme MA, Chen KY , Naumo va AV, et al. Cardiom yocytes derived from human embr yonic stem cells in pro-sur vival factors enhance function of infarcted rat hearts. Nat Biotechnol 2007;25: 1015–24. 202. Amado LC, Saliaris AP, Schuleri KH, et al. Cardiac repair with intramyocardial injection of allo geneic mesench ymal stem cells after m yocardial inf arction. Proc Natl Acad Sci USA 2005;102:11474–9. 203. Assmus B, Honold J, Schächinger V, et al. Transcoronary transplantation of pro genitor cells after m yocardial inf arction. N Engl J Med 2006;355:1222–32. 204. Schachinger V, Erbs S, Elsässer A, et al. Intracoronary bone marrowderived progenitor cells in acute m yocardial infarction. N Engl J Med 2006;355:1210–21. 205. Schachinger V, Erbs S, Elsässer A, et al. Improved clinical outcome after intracoronary administration of bone-mar row-derived progenitor cells in acute m yocardial infarction: f inal 1-year results of the REPAIR-AMI trial. Eur Heart J 2006;27:2775–83.

724

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

206. Lu D, Li Y, Wang L, et al. Intraar terial administration of mar row stromal cells in a rat model of traumatic brain injur y. J Neurotrauma 2001;18:813–9. 207. Urdzikova L, Jendelo va P, Glo garova K, et al. Transplantation of bone marrow stem cells as well as mobilization by granulocytecolony stimulating f actor promotes reco very after spinal cord injury in rats. J Neurotrauma 2006;23:1379–91. 208. Li Y, Chen J, Wang L, et al. Treatment of strok e in rat with intracarotid administration of mar row stromal cells. Neurolo gy 2001;56:1666–72. 209. Jendelova P , Her ynek V, DeCroos J , et al. Imaging the f ate of implanted bone marrow stromal cells labeled with superparamagnetic nanoparticles. Magn Reson Med 2003;50:767–76. 210. Goldman SA, Nottebohm F . Neuronal production, mig ration, and differentiation in a v ocal control nucleus of the adult female canary brain. Proc Natl Acad Sci USA 1983;80:2390–4. 211. Ming GL, Song H. Adult neuro genesis in the mammalian central nervous system. Annu Rev Neurosci 2005;28:223–50. 212. Sailor KA, Ming GL, Song H. Neurogenesis as a potential therapeutic strategy for neurodegenerative diseases. Expert Opin Biol Ther 2006;6:879–90. 213. Gould E, Reeves AJ, Graziano MS, Gross CG. Neurogenesis in the neocortex of adult primates. Science 1999;286:548–52. 214. Chen J , Maga vi SS, Macklis JD . Neuro genesis of cor ticospinal motor neurons e xtending spinal projections in adult mice. Proc Natl Acad Sci USA 2004;101:16357–62. 215. Curtis MA, Penney EB, Pearson AG, et al. Increased cell proliferation and neuro genesis in the adult human Huntington’ s disease brain. Proc Natl Acad Sci USA 2003;100:9023–7. 216. Magavi SS, Lea vitt BR, Macklis JD . Induction of neuro genesis in the neocortex of adult mice. Nature 2000;405:951–5. 217. Parent JM, Valentin VV, Lo wenstein DH. Prolonged seizures increase proliferating neurob lasts in the adult rat sub ventricular zone-olfactory bulb pathway. J Neurosci 2002;22:3174–88. 218. Parent JM, Yu TW, Leibowitz RT, et al. Dentate g ranule cell neurogenesis is increased b y seizures and contributes to aber rant network reor ganization in the adult rat hippocampus. J Neurosci 1997;17:3727–38. 219. Sanai N , Tramontin AD, Quiñones-Hinojosa A, et al. Unique astrocyte ribbon in adult human brain contains neural stem cells but lacks chain mig ration. Nature 2004;427:740–4. 220. Thored P, Arvidsson A, Cacci E, et al. P ersistent production of neurons from adult brain stem cells during recovery after stroke. Stem Cells 2006;24:739–47. 221. Kempermann G, Kuhn HG, Gage FH. More hippocampal neurons in adult mice li ving in an enriched en vironment. Nature 1997;386:493–5. 222. Gould E, Beylin A, Tanapat P, et al. Lear ning enhances adult neurogenesis in the hippocampal for mation. Nat Neurosci 1999;2:260–5. 223. van Praag H, Kempermann G, Gage FH. Running increases cell proliferation and neurogenesis in the adult mouse dentate gyr us. Nat Neurosci 1999;2:266–70. 224. Mirescu C, Peters JD, Noiman L, Gould E. Sleep deprivation inhibits adult neuro genesis in the hippocampus b y ele vating glucocor ticoids. Proc Natl Acad Sci USA 2006;103:19170–5. 225. Tanapat P, Galea LA, Gould E. Stress inhibits the proliferation of granule cell precursors in the developing dentate gyrus. Int J Dev Neurosci 1998;16:235–9. 226. Brown J, Cooper-Kuhn CM, K empermann G, et al. Enriched en vironment and ph ysical acti vity stimulate hippocampal but not olfactory bulb neurogenesis. Eur J Neurosci 2003;17:2042–6. 227. Shapiro EM, Gonzalez-P erez O , Manuel Garcia-V erdugo J , et al. Magnetic resonance imaging of the mig ration of zneuronal precursors generated in the adult rodent brain. Neuroimage 2006;32:1150–7.

228. Doetsch F, Garcia-Verdugo JM, Alvarez-Buylla A. Cellular composition and three-dimensional or ganization of the sub ventricular germinal zone in the adult mammalian brain. J Neurosci 1997;17:5046–61. 229. Garcia AD, Doan NB , Imura T, et al. GF AP-expressing progenitors are the principal source of constituti ve neuro genesis in adult mouse forebrain. Nat Neurosci 2004;7:1233–41. 230. Scheffler B, Walton NM, Lin DD , et al. Phenotypic and functional characterization of adult brain neuropoiesis. Proc Natl Acad Sci USA 2005;102:9353–8. 231. Craig CG, Tropepe V, Morshead CM, et al. In vi vo g rowth f actor expansion of endo genous subependymal neural precursor cell populations in the adult mouse brain. J Neurosci 1996;16: 2649–58. 232. Reynolds B A, Tetzlaff W, Weiss S. A multipotent EGF-responsi ve striatal embr yonic pro genitor cell produces neurons and astrocytes. J Neurosci 1992;12:4565–74. 233. Reynolds BA, Weiss S. Clonal and population anal yses demonstrate that an EGF-responsive mammalian embryonic CNS precursor is a stem cell. Dev Biol 1996;175:1–13. 234. Carpenter MK, Cui X, Hu ZY , et al. In vitro e xpansion of a multipotent population of human neural pro genitor cells. Exp Neurol 1999;158:265–78. 235. Einstein O, Fainstein N, Vaknin I, et al. Neural precursors attenuate autoimmune encephalom yelitis b y peripheral immunosuppression. Ann Neurol 2007;61:209–18. 236. Einstein O, Grigoriadis N, Mizrachi-Kol R, et al. Transplanted neural precursor cells reduce brain inflammation to attenuate chronic experimental autoimmune encephalom yelitis. Exp Neurol 2006;198:275–84. 237. Kukekov VG, Laywell ED, Suslov O, et al. Multipotent stem/progenitor cells with similar proper ties arise from tw o neuro genic regions of adult human brain. Exp Neurol 1999;156:333–44. 238. Uchida N, Buck D W, He D , et al. Direct isolation of human central nervous system stem cells. Proc Natl Acad Sci USA 2000;97: 14720–5. 239. Guzman R, Uchida N , Bliss TM, et al. Long-ter m monitoring of transplanted human neural stem cells in developmental and pathological conte xts with MRI. Proc Natl Acad Sci USA 2007;104:10211–6. 240. Bruck W. Inflammator y demyelination is not central to the pathogenesis of multiple sclerosis. J Neurol 2005;252 Suppl 5:v10–15. 241. Terajima K, Matsuzawa H, Tanaka K, et al. Cell-oriented anal ysis in vi vo using dif fusion tensor imaging for nor mal-appearing brain tissue in multiple sclerosis. Neuroimage 2007;37: 1278–85. 242. Hickey WF, Hsu BL, Kimura H. T-lymphocyte entry into the central nervous system. J Neurosci Res 1991;28:254–60. 243. Einstein O, Kar ussis D, Grigoriadis N , et al. Intra ventricular transplantation of neural precursor cell spheres attenuates acute experimental aller gic encephalom yelitis. Mol Cell Neurosci 2003;24:1074–82. 244. Pluchino S, Zanotti L, Rossi B , et al. Neurosphere-deri ved multipotent precursors promote neuroprotection b y an immunomodulatory mechanism. Nature 2005;436:266–71. 245. Floris S, Blezer EL, Schreibelt G, et al. Blood-brain bar rier permeability and monoc yte inf iltration in e xperimental aller gic encephalomyelitis: a quantitati ve MRI study . Brain 2004;127(Pt 3):616–27. 246. Pluchino S, Quattrini A, Brambilla E, et al. Injection of adult neurospheres induces recovery in a chronic model of multiple sclerosis. Nature 2003;422:688–94. 247. Bain MM, F agan AJ, Mullin JM, et al. Nonin vasive monitoring of chick development in ovo using a 7T MRI system from day 12 of incubation through to hatching. J Magn Reson Imaging 2007;26: 198–201.

Cell Voyeurism Using Magnetic Resonance Imaging

248. Papan C, Boulat B, Velan SS, et al. Time-lapse tracing of mitotic cell divisions in the earl y Xenopus embr yo using microscopic MRI. Dev Dyn 2006;235:3059–62. 249. Lee SC, Mietchen D , Cho JH, et al. In vi vo magnetic resonance microscopy of differentiation in Xenopus laevis embryos from the first cleavage onwards. Differentiation 2007;75:84–92. 250. Papan C, Boulat B , Velan SS, et al. Two-dimensional and threedimensional time-lapse microscopic magnetic resonance imaging of Xenopus gastrulation movements using intrinsic tissue-specif ic contrast. Dev Dyn 2007;236:494–501. 251. Jacobs RE, Fraser SE. Magnetic resonance microscopy of embryonic cell lineages and movements. Science 1994;263:681–4. 252. Louie AY, Huber MM, Ahrens ET, et al. In vivo visualization of gene expression using magnetic resonance imaging. Nat Biotechnol 2000;18:321–5. 253. Canaple L, Beuf O , Armenean M, et al. F ast screening of paramagnetic molecules in zebraf ish embr yos b y MRI. NMR Biomed 2007. 254. Bernas LM, F oster PJ, Rutt BK. Magnetic resonance imaging of in vitro glioma cell invasion. J Neurosurg 2007;106:306–13. 255. Foster-Gareau P, Heyn C, Alejski A, Rutt BK. Imaging single mammalian cells with a 1.5 T clinical MRI scanner. Magn Reson Med 2003;49:968–71. 256. Hsiao JK, Tai MF, Chu HH, et al. Magnetic nanopar ticle labeling of mesenchymal stem cells without transfection agent: cellular behavior and capability of detection with clinical 1.5 T magnetic resonance at the single cell le vel. Magn Reson Med 2007;58: 717–24.

725

257. Zhang Z, v an den Bos EJ , Wielopolski PA, et al. High-resolution magnetic resonance imaging of iron-labeled m yoblasts using a standard 1.5-T clinical scanner. MAGMA 2004;17:201–9. 258. Pinkernelle J, Teichgraber U, Neumann F , et al. Imaging of single human carcinoma cells in vitro using a clinical w hole-body magnetic resonance scanner at 3.0 T. Magn Reson Med 2005;53: 1187–92. 259. Zhang Z, van den Bos EJ, Wielopolski PA, et al. In vitro imaging of single living human umbilical v ein endothelial cells with a clinical 3.0-T MRI scanner. MAGMA 2005;18:175–85. 260. Stroh A, Faber C, Neuberger T, et al. In vivo detection limits of magnetically labeled embryonic stem cells in the rat brain using highfield (17.6 T) magnetic resonance imaging. Neuroimage 2005;24:635–45. 261. Knight B , Laukaitis C, Akhtar N , et al. Visualizing muscle cell migration in situ. Cur r Biol 2000;10:576–85. 262. Molyneaux KA, Stallock J, Schaible K, Wylie C. Time-lapse analysis of li ving mouse ger m cell mig ration. De v Biol 2001;240: 488–98. 263. Miller MJ, Wei SH, Cahalan MD, Parker I. Autonomous T cell trafficking examined in vivo with intravital two-photon microscopy. Proc Natl Acad Sci USA 2003;100:2604–9. 264. Farina KL, Wyckoff JB, Rivera J, et al. Cell motility of tumor cells visualized in living intact primary tumors using green fluorescent protein. Cancer Res 1998;58:2528–32.

45 TUMOR VASCULATURE AMBROS J. BEER, MD, GANG NIU, PHD, XIAOYUAN (SHAWN) CHEN, PHD, AND MARKUS SCHWAIGER, MD, PHD

Angiogenesis is an impor tant process taking par t in various ph ysiologic and patholo gic processes. As a physiologic process, it is required for ph ysical or gan development, wound repair, reproduction, and response to ischemia. Ho wever, it is also associated with pathologic conditions such as ar thritis, psoriasis, retinopathies, and cancer.1 Judah Folkman in 1971 first articulated the importance of angio genesis for tumor growth.2 He stated that the growth of solid tumors remains restricted to 2 to 3 mm in diameter until the onset of angio genesis. Although this h ypothesis f irst was strongly criticized, angiogenesis went on to become one of the most impor tant f ields of oncolo gic research in the follo wing y ears with subsequent in vestigations identifying more than 20 angio genic g rowth f actors, their receptors, and signal transduction pathways. Moreover, endo genous angio genesis inhibitors ha ve been discovered, and the cellular and molecular characterization of the angio genic phenotype in human cancers has been achieved.3,4 However, the results of the f irst clinical trials using angiogenesis inhibitors in oncology were disappointing, although most of the applied substances were ef fective in preclinical trials. 5,6 Only recentl y, encouraging results have been achieved with the vascular endothelial g rowth f actor (VEGF) antibody Avastin in combination with standard c ytotoxic chemotherap y first in metastasized colorectal cancer and subsequently in breast cancer and non-small-cell lung cancer (NSCLC).7,8 This success will spur the demand for imaging modalities used for assessment of the optimum dose of ne w antiangio genic agents and for response evaluation. Up to no w, clinical trials with con ventional cytotoxic chemotherapeutic agents ha ve mainl y used morphologic imaging to pro vide indices of therapeutic response, mostl y computed tomo graphy (CT) or

726

magnetic r esonance i maging ( MRI). B idimensional measurements of the maximum tumor e xtension ha ve mainly been used to estimate changes in response to the investigational therap y as compared with a baseline measure. Through standardization of these measurements by introducing the RECIST criteria in the year 2000, considerab le pro gress has been achie ved.9 However, as antiangio genic agents lead to a stop of tumor pro gression rather than to tumor shrinkage, the approach of measuring tumor response by a reduction of tumor size is not applicab le and might tak e months or years to assess. Therefore, there is g reat interest in reliable biomarkers of early tumor response to noncytotoxic drugs.10 Imaging techniques could potentiall y be used as such a biomarker and could provide an early indicator of ef fectiveness at a functional or molecular le vel. Because antiangiogenic therapies are designed to af fect the abnormal blood vessels found in tumors, changes in hemodynamic parameters such as blood flow, blood volume, or vessel permeability may be promising biomarkers for response evaluation. Sometimes these functional parameters are summarized using the ter m “perfusion,” sometimes the term “flo w” is used equi valent to “perfusion.” Ho wever, in the follo wing, the ter m “blood flow” (in mL/(min*mg)) will be used instead of more unspecific term “perfusion.” Current clinical trials use various imaging techniques for this purpose, mostly dynamic contrast-enhanced MRI (DCE-MRI), and less often ultrasound, positron emission tomo graphy (PET) (especially with [ 15O]water), and dynamic contrastenhanced CT (DCE-CT).11 In the future, the targeting of specific molecular markers of angiogenesis might also be used for response assessment of antiangiogenic therapies, for example, the VEGF pathway or cell surface markers such as the αvβ3 integrin.

Tumor Vasculature

In this chapter, various methods for assessment of tumor vasculature will be discussed. F irst, evaluation of the structure of angiogenic vasculature will be discussed, as with microCT and dedicated small animal ultrasound devices. Next, methods for the measurement of hemodynamic parameters of tumor vasculature will be presented, with a focus on DCE-MRI, which currently is the most widely applied method in clinical trials. F inally, imaging of molecular mark ers of tumor v asculature will be discussed, w hich has been v ery promising in preclinical studies and might pla y an impor tant role in the future clinical environment as well.

BIOLOGY OF ANGIOGENESIS Before discussing the different methods for imaging tumor vasculature, a shor t summary of the processes in volved in angiogenesis and dif ferent antiangio genic therapeutic strategies will be presented here. This should f acilitate the understanding of the v arious strategies used for molecular imaging of angiogenesis. Excellent detailed reviews on the mechanisms of angio genesis are also presented b y Carmeliet12 and Auguste and colleagues. 13 Each solid malignancy starts as a small population of transformed cells that do not initially have a blood supply of their o wn. Tumor cells are initiall y supplied b y diffusion, and tumor growth is limited by the lack of access to growth f actors, circulating o xygen, and nutrients. 14 Tumor cells ma y then inf iltrate nearby blood vessels to overcome these limitations to for m “mosaic v essel” which consists of nor mal endothelial cells mix ed with infiltrative tumor cells. This process, ho wever, can onl y supply the peripher y of the tumor with o xygen and nutrients, and tumor g rowth is still very slow. This phase of slow growth can last months or years. To create a blood supply of their o wn, tumors ha ve to s witch to an angiogenic phenotype, a phenomenon also called the “angiogenic s witch.” This s witch is not just a simple upregulation of angiogenic activity, but it is thought to be the result of a net balance of positi ve and negative regulating factors of angiogenesis with proangiogenic factors overcoming the ef fect of antiangio genic factors. Tumors produce a multitude of peptide angio genic f actors in response to tumor hypoxia, such as the VEGF, the acidic and basic f ibroblast g rowth f actors, and the plateletderived endothelial cell g rowth f actor.15 The angiogenic switch occurs when the tumors produce these angiogenic factors in excess of local angiogenesis inhibitors such as thrombospondin-1, endostatin, angiostatin, or antiangiogenic antithrombin III. Angiogenic growth factors diffuse toward nearb y pree xisting b lood v essels and bind to

727

receptors located on endothelial cells such as the receptors to VEGF (VEGFR-1/Flt-1, VEGFR2/KDR/Flk-1, VEGFR-3/KDR, Flt-1, VEGF-R2, VEGFR3/Flt-4, and VEGF-R4/neuropilin-1). 16 Receptor binding leads to receptor dimerization and acti vation of various signal transduction pathways, for example, phosphorylation of tyrosine kinases, protein kinases, and MAP kinases and consequentl y to activation of endothelial cells. 17–19 The original v essels undergo characteristic morphologic changes, including enlar gement of the diameter, basement membrane de gradation, a thinned endothelial cell lining, increased endothelial number , decreased pericyte number, and pericyte detachment. 20,21 In the ne xt step, at least four dif ferent mechanisms ma y lead to for mation of ne w tumor b lood v essels.22–25 The original v essels ma y retain their lar ge diameter and evolve into medium-sized arteries and veins by acquiring a smooth muscle and inter nal elastica, w hich can tak e from a fe w da ys to se veral months. Alternatively, the endothelium of a mother v essel may form smaller separate w ell-differentiated v essel channels b y projecting cytoplasmic s tructures i nto t he l umen w hich f orm translumenal bridges, which can take from several days to 3 w eeks. A third process is called intussusception and involves focal in vagination of connecti ve tissue pillars from within the mother v essel, which also can tak e from several da ys to se veral w eeks. F inally, endothelial cell sprouting may occur, which requires the focal dissolution of the basement membrane surrounding mother vessels.26 This is achie ved b y a number of proteol ytic enzymes, including matrix metalloproteinases (MMPs) and plasminogen activator, which enable endothelial cells to e xit the vessel. Activated angiogenic endothelial cells proliferate rapidl y and mig rate into the e xtracellular matrix (ECM) toward the angio genic stimulus. 27–29 Cell surface adhesion molecules such as the inte grins play an important role in endothelial cell migration, and in contact with the extracellular tumor matrix, facilitate cell survival. In this context, the role of the αvβ3 and αvβ5 integrins has been especially well examined.30–33 At the sprouting tips of g rowing vessels, endothelial cells secrete MMPs that facilitate degradation of the ECM and cell in vasion.34,35 Next, a lumen within an endothelial cell tubule has to be formed, w hich requires interactions betw een the ECM and the cell-associated surf ace proteins, among them are galectin-2, PECAM-1, and VE-cadherin.36,37 Finally, newly for med vessels are stabilized through the recr uitment of smooth muscle cells and peric ytes. During this process, the angiopoietin f amily plays a major role, such as angiopoietin-1 (Ang-1) which binds to the Tie-2 receptor on angiogenic endothelium.

728

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

STRUCTURAL IMAGING OF TUMOR VASCULATURE All imaging modalities can pro vide structural information, although the y ha ve dif ferent spatial resolution. The oldfashioned method of v ascular str ucture imaging is X-ra y angiography. However, it is difficult to provide microvasculature infor mation w hich is more impor tant in malignant diseases. Owing to impro vements in imaging equipment, contrast agents, data acquisition, and anal ysis techniques, more detailed vascular structure has been deciphered. Se veral modalities are a vailable for tumor micro vasculature imaging, including intra vital microscopy, CT angio graphy (CTA), contrast-enhanced ultrasound (CEU), and high-resolution magnetic resonance angio graphy (MRA). 38 In this section, we will focus on microCT angiography. For clinical CT, it has been repor ted that conventional tomographic analysis after intravenous infusion of contrast medium can potentiall y allow distinction betw een benign and malignant single pulmonary nodules.39 Blood flow patterns can also be evaluated with dynamic CT to delineate abnormal hemodynamic lesions in different organs.40,41 Multidetector CT angio graphy (MDCT A) allo ws one to scan larger volumes in less time with impro ved spatial and temporal resolution. In addition, the faster scanning makes it possib le to image dif ferent v ascular phases using the same contrast bolus, impacting the number of scans that can be done.42 CTA now provides a credible quick, noninvasive, low-dose alter native to con ventional digital subtraction angiography.43 CTA has the potential to pro vide invaluable information about sur rounding nonvascular str uctures and to image both ar terial and venous systems. With a subsecond single-section helical CT scanner and reduced collimation, a longitudinal resolution of 1.7 mm full width at half maximum can be reached. A little better spatial resolution can be achie ved with MDCT but is still not enough for microvasculature visualization.44 High-resolution X-ray CT systems (microCT) are capab le of visualizing e xtensive vascular netw orks but with limitations in f ield of vie w (FOV). For example, the MS8 cone-beam CT system manufactured by General Electric Medical Systems, designed for excised tissues such as rodent or gans, has a maximum FOV of 5 cm and a detector with 2,048 × 2,048 pixels. The system can be used for v ascular studies using contrast agents, and vessels down to 22 µm in diameter can be resolved.45 After perfusion of the v asculature with X-ra y absorbing agents, three-dimensional (3D) digital images can be generated with high contrast, which allows immediate computer processing to segment vessels from the tissue background and e xtract the branching str ucture and geometric mor phology of the vascular network.46 Jorgensen and colleagues 47 used a nearl y monochromatic desktop

X-ray source and optical magnif ication to visualize vasculature and soft tissue in isolated , f ixed, and stained rodent organs at a voxel size of 5 µm. In general, resolution of v essels imaged with benchtop µCT scanners is in the range of 10 to 20 µm (ie, the effective voxel size). The diameters of capillaries, however, are generally < 10 µm and cannot be resolved reliably because of partial volume effect. To improve the number of photons produced per unit area per second of benchtop X-ra y sources, synchrotron radiation µCT (SR µCT) using synchrotron light has been applied for better spatial and contrast resolution. Synchrotron light is electromagnetic radiation produced b y the acceleration of electrons that mo ve near the speed of light through magnetic f ields.48 After rat brain tissue w as perfused with a combination of contrast agent and a gelatin polymer, several mm3 of microvasculature including capillaries could be visualized with a combination of pure absorption and edge enhancement at a v oxel size of 1.4 µm.49 Vascular corrosion casts are a replication of the vessel lumen by filling the vasculature with a casting polymer. Heinzer and colleagues used intact cor rosion casts of whole mouse brains, a local SR µCT setup, and edge enhancement mode for visualizing and quantifying microvasculature in selected regions of interest (ROIs) at a voxel size of 1.4 µm. The same g roup also de veloped a method called hierarchical imaging for systematic imaging of large volumes of vasculature at any depth of the v ascular network. This method is based on the technique of modified v ascular cor rosion casting, desktop µCT, local SRµCT, and scanning electron microscop y imaging, following a hierarchic and strictly nondestructive approach.50 Although e x vi vo studies based on animal models offer some signif icant adv antages, in par ticular w hen going to high-resolution imaging, in vi vo experiments are the ultimate goal for clinical applications. µCT is a promising method to visualize the architecture of v asculature of normal organs and tumor, and importantly, to derive quantitative information from the images. However, a limitation of the method is that to optimally visualize the vasculature, the or gans under study such as kidne y or brain must be removed and prepared for e xamination. In addition, long scan periods and high X-ray doses for µCT make it unsuitable for repeated nonin vasive measurements of vessels. Kiessling and colleagues51 reported a prototype volumetric computed tomo graphy (VCT) apparatus for high-resolution imaging in e xperimental and preclinical applications. As an important technologic advancement for investigating fine structure of tissues in vivo, VCT combines the advantages of µCT and clinical CT scanners. Reconstr ucted VCT images have high resolution with isotropic voxels of < 4 × 10-4 mm3, and the minimum v essel diameter visualized b y VCT v aries betw een 40 and 50 µm (F igure 1).

Tumor Vasculature

A

729

B

Figure 1. A, Three-dimensional visualization showing VCT angiography of a tumor-bearing nude mouse and detailed image of the tumor after magnification (top right); dilatation of subcutaneous vessels (arrow on whole-body visualization), which interact with the tumor tissue, is clearly visible. B, On 3D volume rendering, larger vessels inside the tumor. Determined vessel diameters were as follows: 1 = 145 µm and 2 = 205 µm.52

Although small v essels with diameters less than 20 µm were not visualized with this modality, this technique permits the acquisition of a large volume of slices per rotation, with intrinsically higher resolution than is achievable with conventional CT. In comparison to small animal µCT, the VCT technolo gy of fers a lar ger FOV (~30 cm) and dramatically shor ter scanning times (2–8 s per rotation), which is desirable for animal imaging. 52 VCT fills the gap between clinical multislice computed tomography and preclinical µCT systems and is highl y suited for studying orthotopic and metastasizing tumor models. 53

FUNCTIONAL IMAGING OF TUMOR VASCULATURE Functional imaging of hemodynamic parameters such as blood flow, blood volume, and vascular permeability using PET, MRI or CT is currently widely used in study trials as an imaging biomarker of angiogenesis.54 Up to now, however, it has not been clear , which parameters and w hich imaging modalities would be optimal in this context. In the following passages, w e will discuss the dif ferent techniques used with a focus on MRI, as it has been the most commonly used technique for assessment of hemodynamic parameters of angiogenesis up to the present time.

MRI and CT Dynamic Contrast-Enhanced Imaging with Low Molecular Weight Contrast Agents

The techniques of DCE-CT and DCE-MRI will be discussed to gether, as the same principles appl y to both techniques, when low molecular weight contrast agents (LMWCA) a re u sed. An i mportant d ifference a nd

advantage of CT is, ho wever, that it is a quantitati ve technique because the measured Hounsf ield units are linearly cor related to the concentration of the contrast material in the tissue. 55 DCE-MRI and DCE-CT are nonin vasive methods of in vestigating micro vascular str ucture and w ork b y tracking the phar macokinetics of injected LMWCA as they pass through the tumor v asculature. Changes in vascular per meability, e xtracellular e xtravascular and vascular volumes, and in blood flow can be measured by these techniques. MRI has the advantage over CT that it does not in volve ionizing radiation, and that MRI contrast agents generally have a better toxicity profile than iodine contrast agents. In T1-weighted DCE-MRI, an intravenous bolus of Gadolinium (Gd) contrast agent enters tumor ar terioles, passes through the capillar y bed, and then drains via tumor veins. Gd ions are paramagnetic and interact with adjacent h ydrogen nuclei to shorten T 1-relaxation times in local tissue w ater which causes an increase in signal intensity on T1-weighted images to a v ariable extent within each v oxel. This signal enhancement is dependent on physiological and physical f actors, including tissue perfusion, capillar y surface area and per meability, and the v olume of the extracellular e xtravascular leakage space (EES). Ho wever, signal enhancement will also be af fected b y the native T1-relaxation time of each tissue, b y the contrast agent dose, and b y the chosen sequence parameters, which mak es quantitation of the results more dif ficult than with DCE-CT .56,57 DCE-MRI strate gies vary, but, in general, anatomic information to localize the tumor is acquired initially. Next, sequences that allow calculation of baseline tissue T1-values, before contrast agent administration, are perfor med to enab le subsequent analysis. Finally, dynamic data are acquired e very few seconds in T1-weighted images over a period of around

730

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

5 to 10 minutes. Fast T1-weighted spoiled gradient echo sequences are generall y used as the y allow good contrast medium sensitivity, high signal-to-noise ratio, adequate anatomic co verage, and rapid data acquisition. Several anal ysis methods can then be applied to the acquired data. A plain qualitative analysis of the signal intensity–time cur ve (e g, g radient, overall shape, time to 90% maximum enhancement) represents the most simple description of contrast agent distrib ution. Parameters that describe the shape of the contrast agent concentration–time cur ve represent a combination of flo w, blood v olume, v essel per meability, and EES v olume. One such parameter , the initial area under the contrast agent concentration–time cur ve (IAUC) is easy to calculate, reasonably reproducible, and routinely used as a biomarker in drugs trials. However, IAUC has a complicated relationship with underl ying tumor ph ysiology and represents a composite of ph ysiologic processes, including flo w, capillar y per meability, and b lood v olume. Therefore, the results are hard to inter pret in their physiologic meaning. A more meaningful and quantitative analysis can be achieved by applying pharmacokinetic models to contrast agent concentration data to enable estimates of ph ysiologic characteristics such as flow and capillar y endothelial per meability. Modeled parameters are not onl y more physiologically meaningful than simple qualitati ve cur ve anal ysis, such as IAUC, they are also in theory independent of the acquisition protocol and should onl y reflect tissue characteristics. Thus, the y are suitab le for multicenter studies with variation in image acquisition protocols and equipment.58 These parameters are the v olume transfer constant Ktrans (min-1), w hich describes the rate of flux of contrast agent into the e xtracellular extravascular space within a gi ven volume, the v olume of the e xtracellular extravascular space (EES) per unit v olume of tissue υe, and the rate constant for the back flux from the e xtracellular extravascular space to the vasculature κep (min-1). These ter ms are related to each other b y the equation κep = Ktrans/υe. The relationship of the dif ferent parameters Ktrans, κep, and υe can be expressed by the following equation:

or

dCt Ct trans =K × (C p − ) dt Ve

(1)

dCt = K trans × C p − kep × Ct , dt

(2)

where Ct is the tracer concentration in tissue, Cp is the tracer concentration in plasma, and t is time (in seconds).

Several other names have been used for these parameters in the literature. Ho wever, in a 1999 consensus publication, this set of ter ms w as recommended b y an international group of investigators developing DCE-MRI methodologies.56 The authors proposed that this set of kinetic parameters and symbols be used uni versally to describe the uptak e of lo w molecular w eight Gd-based contrast agents that are in clinical use toda y and related them to pre viously pub lished ter ms and symbols. Ktrans, κep, and υe are currently the most commonl y used hemodynamic parameters in clinical studies e valuating antiangiogenic therapies and are mainl y derived by DCE-MRI. However, the exact physiologic meaning of these parameters is complex and not related to a single process such as blood flo w or b lood v olume onl y. The inter pretation of Ktrans varies depending on the relationship betw een the blood flo w and the capillar y per meability–surface area product (PS). When tissue contrast deli very is unlimited and flow is high, Ktrans represents the PS per unit v olume of tissue, for transendothelial transport between plasma and EES. When tissue perfusion is limited , Ktrans represents the b lood flow per unit v olume of tissue. 56 In these simple models, both Ktrans and υe calculation are relatively stable but lack ph ysiologic specif icity. Extensions of this model are more comple x but enab le calculation of b lood plasma v olume (v p) and pro vide more accurate estimations of Ktrans and υe.59 More comprehensive models allow direct quantification of flow, extraction fraction, and mean capillary transit time. 60 Here it is possib le to separate blood flo w from capillar y per meability. Ho wever, successful application of this model requires a temporal resolution on the order of 1 second , w hich limits its application in clinical trials. In theor y, the models described above require direct measurement of an ar terial input function (AIF) along with the tumor contrast agent concentration–time course curve. These two functions are then used to quantify the passage of contrast agent through the tumor. Ideally, the AIF should be measured for each examination, as it varies between individuals and visits reflecting physiologic variation in cardiac output, v ascular tone, renal function, and injection timing. Unfortunately, AIF measurement is technicall y demanding, and a measurement from a nearb y large artery might differ from the v essel suppl ying the tumor . Therefore, many g roups use an idealized math ematic function instead, which does not attempt to reflect the tr ue blood supply to the tumor at each e xamination, but allo ws for greater fle xibility concer ning requirements for temporal resolution, slice positioning, and sequence choice. 61 Once one has acquired the data and chosen a specif ic model, in principle, tw o strate gies can be used for data

Tumor Vasculature

analysis. One can use a def ined ROI that encompasses all or part of the tumor. A single average enhancement curve can be extracted and used to generate values of parameters of interest (such as IAUC or Ktrans), and the same parameters can then be compared following therapy.62 However, this method ignores heterogeneity within the tumor. Alternatively, data can be extracted from each voxel within the ROI and summar y statistics such as the mean, median, and interquartile range ma y be calculated. 63 This second method can describe both nor mal and abnor mal data distributions and provides information regarding tumor vasculature heterogeneity. A limitation concerning DCE-MRI is that it is prone to artifacts. Therefore, significant motion artifacts, AIF and ROI definition, and signal-to-noise ratio should be assessed and if possible corrected. If correction is n ot p ossible, t hen c orrupted d ata s ets s hould b e removed from subsequent analysis.64 Concerning the reproducibility of these hemodynamic parameters, the maximum enhancement and the area under the cur ve appear to be the most reproducib le among the semiquantitative parameters. The slope or rate of enhancement appears to be the least reproducible, which may reflect the dependence of these measurements on the rate of injection of the contrast agent bolus as w ell as variations in cardiac input. Among the quantitative parameters, υe seems to have the least variability, with interpatient variability being much g reater than intrapatient v ariability. Ho wever, Ktrans and κep also have sufficient reproducibility to be useful for measuring changes over time. 65,66 Many investigators consider a change in Ktrans of > 40% as likely to represent a true difference in the parameter as some e vidence suggests it correlates with disease stability/response, but in practice, the conf idence interval for each parameter depends on the choice of model, AIF methods, and ROI definition.67 Therefore, many centers perfor m two baseline scans to measure reproducibility for each trial data set, in accordance with published guidelines.68 Evidence of dr ug ef ficacy has been demonstrated with DCE-MRI in several trials of antiangiogenic drugs. Significant reductions in Ktrans have been repor ted in patients with advanced breast cancer receiving bevacizumab alone. Ktrans reduction was increased following a further six c ycles of be vacizumab with con ventional chemotherapy. However, changes in Ktrans did not predict response rate. 69 Several tyrosine kinase inhibitors, such as AG-013736, BIBF1120, and AZD2171, have all shown dose-dependent reductions in Ktrans and the IAUC without a correlation to clinical response.70 Only a few trials up to now ha ve demonstrated a relationship betw een DCEMRI and clinical outcome. Correlation of Ktrans reduction and the response rate and pro gression-free sur vival has

731

been shown with B AY 43-9006. 71 Moreover, DCE-MRI has been successfully used to predict which patients progressed with gliob lastoma multifor me in a trial of PTK787/ZK222584 (PTK/ZK). 72 Three related trials of PTK/ZK repor ted promising earl y results with DCEMRI. Patients with colorectal carcinoma liver metastases showed a dose-dependent reduction in DCE-MRI parameters. Moreover, reduction in tumor enhancement predicted disease pro gression in these patients and w as positively correlated with reduction in tumor size. 73 Statistically signif icant dose dependent changes in DCEMRI from baseline w ere identif ied in tw o subsequent studies of PTK/ZK in patients with mixed solid tumors.54 In summar y, DCE-MRI is promising as an imaging biomarker in clinical trials, ho wever, data concer ning its performance especially for response assessment are not uniform and seem to depend strongl y on the therap y protocol and tumor type. Dynamic Contrast-Enhanced Imaging with Macromolecular Contrast Agents

Macromolecular c ontrast a gents ( MMCAs) g enerally range in molecular weight from 5 to 90 kDa, and both Gd- and iron oxide (IO)-containing agents are included in this category.74 MMCAs were initially designed for prolonged intravascular retention for use with MRA. Unlik e LMCA, they do not pass through nor mal endothelia and thus are potentially more suitable for selective imaging of the tumor neo vasculature, which tends to be highl y permeable. Thus, the increased size of MMCAs makes them less diffusible, and Ktrans values may more accurately reflect per meability within tumors. Moreo ver, the y are excellent blood pool agents, so the y can give more accurate estimates of tumor b lood v olume. MMCAs ha ve a delayed clearance from the body because the y are not cleared e xclusively through the kidne ys. This could , in theory, lead to to xicity from dissociation of the Gd ion from its chelate. Especiall y with nephro genic systemic fibrosis being recognized as a side ef fect of Gd-containing contrast agents, this prob lem deserves high attention and must be addressed in future to xicity studies.75 Albumin-(Gd-DTPA) is the prototype MMCA because of the ubiquity of albumin and its w ell-known properties but is not y et being commerciall y developed.76 A number of other macromolecular car rier compounds have been investigated, including MS-325, de xtran compounds, viral particles, dendrimers, perfluorocarbon emulsions, liposomes, and IO compounds. Most MMCA studies ha ve been perfor med onl y in animal models, although MS-325, dextrans, and IO compounds are being

732

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

tested in human clinical trials.55 Albumin-(Gd-DTPA) is helpful in characterizing the micro vessels of a wide range of tumors. It has been shown that albumin-(Gd-DTPA) correlates with microvessel density (MVD) and immunohistochemistry and is comparab le to LMCA DCE-MRI following treatment with SU6668 and SU11248 (an inhibitor of the VEGF, PDGF, KIT , and FL T3 tyrosine kinase receptors). 77,78 In one study , albumin-(Gd-DTPA)enhanced MRI cor related with the histolo gic g rade of breast cancer, whereas poor correlations occurred with LMCAs.79 MS-325 is an agent that re versibly binds albumin and was the first MMCA to be used in human trials. 80 Only the albumin-bound for m can be considered an MMCA, but the reversible nature of the Gd binding allows Gd to be more rapidly cleared from the body. MS-325 has been used in clinical trials as an MRA agent but has y et to be investigated for angiogenesis assessment in humans. 81 IO nanoparticles are also used as MMCAs. An advantage over Gd-MMCA is the lack of potential toxicity issues due to free Gd. Moreo ver, the body naturall y processes iron. However, iron-based compounds pose different problems. The length of time that iron is retained after contrast administration has led to concer ns about iron overload. In addition, iron nanoparticles have been shown to have their own to xicities, usuall y related to anaph ylaxis reactions. A number of different IO compounds have been developed as MMCA. IOs are superparamagnetic, and they predominantly act to shor ten T2* relaxation time to produce a “negative” enhancement. Superparamagnetic iron oxide (SPIO) particles are usually 50 to 150 nm in diameter. They have a polycrystalline core coated with either dextran or silica and are mainly taken up by phagocytic cells within the reticuloendothelial system (RES) and l ymphatic system. Ultrasmall SPIOs (USPIO) are 10 to 50 nm in diameter. Their smaller size means that they are less taken up by the RES. 82 USPIOs ha ve been used to image the angiogenic processes in murine breast cancer models, with results for Ktrans shown to cor relate with both tumor g rade and histologic MVD.83,84

spatial resolution of this method are limited, especially if the rate of b lood flo w is lo w, because the spin label is very shor t-lived. Nevertheless, ar terial spin labeling has been shown to be a reproducib le method for measuring blood flo w, e g, b y comparison with [ 123I]iodoamphetamine SPECT in the brain.85 Moreover, it does not require the use of contrast agents or arterial blood sampling. Therefore, this method is completel y nonin vasive and well suited for repeat measurements. Blood o xygen le vel-dependent (BOLD) imaging can detect changes in the o xygen saturation of the blood. By measuring signal changes in response to hypercapnia and hyperoxia, the vascular maturity can be detected as onl y mature v essels react to h ypercapnia.86 The underl ying ph ysiologic processes gi ving rise to measured BOLD signal changes include contrib utions from changes in blood flow, blood volume, and metabolic rate of o xygen consumption. Some preclinical studies show no correlation between parameters derived from DCE-MRI and BOLD MRI in a murine tumor xenograft. The authors conclude that the infor mation from both techniques is complementar y, b ut the e xact physiologic meaning of the BOLD signal in tumors remains complex and has to be fur ther evaluated.87

Arterial Spin Labeling and Blood Oxygen Level-Dependent Imaging

× mLtissue–1; where Ct is tissue concentration, mol Ci is influx concentration, mol × mLcarrier–1; Ce is eflux concentration, mol × mLcarrier–1; and F is flow, mLcarrier × min–1 × mLtissue–1. In literal terms, the amount of tracer cleared by the tissue over time t is the product of flo w F and tracer extraction. Flow is calculated by rearranging term (3) so that:

Water molecules can be labeled for MRI by inverting the nuclear spin of their hydrogen atoms by a radiofrequency pulse directed at ar terial blood before it enters the R OI. An absolute v alue for b lood flo w is deter mined b y the change in the magnetic resonance (MR) signal as the labeled w ater in the ar terial b loodstream ar rives in the ROI. Ho wever, the signal-to-noise ratio of ar terial spin labeling MRI is relati vely lo w and the sensiti vity and

Radiotracer Techniques A major advantage of the nuclear medicine techniques using radiotracers is that they are truly quantitative and that the tissue concentration Ct can be measured noninvasively. This mainly applies to PET, whereas SPECT is only rarely applied for blood flow measurements outside the brain, due to its limited spatial resolution and its limitations concer ning true quantification and attenuation correction. Fick f irst described in 1870 the central relationship between b lood flo w and tissue clearance of circulating tracers:

Ct = F × ∫ (Ci − Ce )dt ,

F=

Ct

∫ (Ci − Ce )dt

.

(3)

(4)

Some approaches such as indicator fractionation methods and indicator w ashout methods simplify the relationship such that Ci or Ce are equal to zero. 88 Indicator w ashout studies rel y on direct administration of tracer to the tissues or a lar ge ar tery, e g, the carotid artery for hemicerebral studies. F or this pur pose, [133Xe]xenon can be used because effectively all tracer is exhaled at f irst pass through the lungs without recirculation. Therefore, the arterial concentration Ca can be disregarded, and Ci is modeled as the concentration in tissue fluid ( Ct) and Ce as re gional v enous concentration ( Cv). However, this model is sensitive to assumptions including the prof ile of l ymphatic drainage and sequestration of the tracer in fat. Its main application in humans is the measurement of re gional cerebral b lood flow.89 Indicator fractionation methods model Ci as ar terial concentration ( Ca) and Ce as the concentration in tissue fluid (Ct), which is equal to zero, because effectively all tracer is trapped in capillaries. This can be achie ved by central arterial injection of radiolabeled [11C]microspheres of approximately 10 µm diameter . [ 11C]microspheres indicator methods are primaril y of interest as “gold standard” for flo w measurements or for v alidation of ne w imaging methods for flo w measurement. Most PET-flow measurements, ho wever, are no wadays performed using [ 15O]water. 15 O]oxygen is the With a half-life of 123 s, [ longest-lived positronic o xygen isotope. It can be further reacted with carbon or h ydrogen to produce [15O]CO2, [15O]CO, or [ 15O]H2O. [ 15O]H2O satisfies all the requirements for a flo w tracer in F ick’s model because it is biolo gically and metabolicall y iner t and freely diffusible into and out of tissue w ater. Thus “tissue w ater” can be modeled as a single compar tment including both tissue and its draining fluids (l ymphatics and veins). Two methods can be used for measuring flow with [ 15O]H2O: the steady-state method of F rack15 O]-dynamic w ater owiak and colleagues and the [ method by Lammertsma and Jones. 90 The latter is currently used most often for flow studies due to improved PET scanner technolo gy; therefore, w e will focus on this method. The tracer is administered by inhalation or peripheral venous b olus i njection. C ontinuous a rterial d ata a re obtained either by image-based AIFs (a large vessel like the aorta or the left v entricle) or b y peripheral sampling to a w ell-counter device. In the case of peripheral sampling, a cor rection needs to be made for dela y and dispersion of the recorded arterial curve, due to the length of the connecting tubes. The change in tissue concentration over time is modeled as follows:

Tumor Vasculature

733

dCt (t ) = F × Ca (t ) − ( F / VD + λ ) × Ct (t ), dt

(5)

where VD is the “volume of distribution,” the “proportion of the ROI in which the radioactive water is distributed,” mLblood/mLtissue = 1/ ρ(ρ is par tition coef ficient); Ct(t) is instantaneous tissue concentration of [ 15O]H2O at time t, Bq/mLtissue; Ca(t) is cor rected instantaneous ar terial concentration of [ 15O]H2O at time t, Bq/mLtissue. The mathematics for solving F and VD from the dynamic cur ves depends on con volution of the ar terial and tissue data sets. The expression for tissue concentration at each time t is given by the convolution integral:

Ct (t ) = ∫ F × Ca (T ) × e

− ( F /VD + λ ) × ( t − T )

dT (6)

or

Ct (t ) = F × Ca (t ) ⊗ e

− ( F /VD + λ )× t

,

(7)

where ⊗ is the operation of convolution. Ct(t) describes a biphasic curve with an initial peak followed by a longer tail of decay. F and VD can be determined from this curve using nonlinear least-squares f itting. The dynamic method has been shown to be less sensitive to tissue heterogeneity than the steady-state model. It is, ho wever, more sensiti ve to assumptions about constanc y of F and VD and free and instantaneous diffusion of [ 15O]H2O out of ar terial blood and through the tissues because equilibrium is not reached.91 This means that it is assumed that tissues exhibit neither tracer binding nor concentration g radients and that the arterial extraction fraction is uniform. The validity and reproducibility of this method was initially assessed for the brain and myocardium but subsequently also for tumors of pancreas, brain, breast, and li ver. The data are compatib le with those from diseases in vestigated with other methods, and the range of v alues repor ted b y PET for tumors is within the repor ted range for PET in other tissues. 92,93 However, the hetero geneity of tumor b lood suppl y on microscopic scales with such phenomena as ischemia, shunts, and necrosis mak es it clear that the assumption of a single ar terial input and equilibration of ar terial and tissue water is not fully in accordance with the physiology of tumor microcirculation. Ultimatel y, it has to be pro ven in patient studies that measurements of blood flow using PET provide re levant i nformation f or p atient m anagement, especially for response e valuation and dose optimization during chemotherap y. In locall y adv anced breast cancer , first results with dynamic [15O]H2O PET are promising, as tumor blood flow decreased in the responder g roup after chemotherapy, w hereas it increased in the nonresponder group.94

734

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

PET data can also be used to deri ve data on b lood volume and vascular permeability. Blood v olume imaging with PET uses [ 15O]CO or [11C]CO carbon monoxide. [15O]CO binds irreversibly with hemoglobin to form [15O]CO-Hb carboxyhemoglobin.88 Because [15O]CO-Hb remains exclusively within the vasculature, it can be used as a tracer of v ascular volume. A tissue concentration data set is obtained o ver a fur ther 5 to 6 minute, and an ar terial [ 15O]CO-Hb concentration cur ve is derived from a series of ar terial b lood samples o ver the same inter val. Tissue v ascular v olume can be def ined as follows:

Vv =

Vt × ∫ Ct R × ∫ Ca

,

(8)

where Vv is volume of vessels, mLvessels; Vt is volume of tissue within ROI on scan, mL tissue; R is ratio of small v essel to lar ge v essel hematocrit (assumed to be 1 in tumors); Ct(t) is tissue activity, Bq/(mL × min); and Ca(t) is arterial activity, Bq/(mL × min). Another method for b lood volume imaging is labeling red blood cells (RBCs) or albumin with radionuclides because both are too lar ge to leave normal blood vessels and are retained in the blood pool. In tumor vessels, leakage of these contrast agents into the tumor will occur, but this effect can be used to calculate the tumor v essel permeability, when dynamic imaging is performed. For PET, the tracer [ 68Ga]DOTA-albumin has been de veloped and showed f avorable results in f irst animal studies. 95 An advantage is that the radionuclide [68Ga] is generator produced and is therefore continuously available even to centers lacking an in-house cyclotron.

Ultrasound Ultrasound images are essentiall y maps of tissue echogenicity deri ved from the same principles as SONAR. The simplicity, ease of use, speed , and safety of ultrasound has led to a signif icant role in diagnosis, treatment assessment, follow-up, and guidance of therapy.96 The contrast of ultrasound is dependent on the sound speed , sound attenuation, backscatter , and the imaging algorithm. 97 Ultrasound can be used to image the microcirculation using both Doppler and microbubble methods.98 Power Doppler can be quantified to give an estimate of relati ve fractional v ascular v olume, whereas microbubbles can show blood flow down to the microcirculation le vel b y raising the signal from smaller vessels.

Ultrasound is also w ell estab lished as a means of measuring blood flow or, more precisel y, blood velocity using the Doppler principle. With color Doppler imaging ultrasound, signif icant changes in b lood flo w in renal artery feeding the murine renal cell carcinoma tumor were observed under treatment with a VEGFR inhibitor.99 It w as also demonstrated that b lood flo w measured b y color Doppler imaging ultrasound cor relates with the vessel density . Inno vations in the f ield of ultrasound imaging ha ve made it possib le to measure b lood flo w microvasculature smaller than 100 µm. At frequencies on the order of 50 MHz, Doppler processing allo ws direct assessment of flow dynamics in real time in arterioles as small as 15 µm.100 Microbubble contrast agents present a diameter from less than 1 to 5 µm and are composed of a shell of biocompatible materials, including proteins, lipids, or biopolymers, and a filling gas. Owing to their intrinsic compressibility (appro ximately 17,000 times more than water), microbubb les are v ery strong scatterers of ultrasound and can be detected at typical clinical frequencies of 1 to 13 MHz. There are three major advantages of contrast ultrasound over conventional ultrasound for the detection of the microcirculation. First, microbubble agents increase the po wer of the echo backscattered b y b lood b y 20 to 30 dB (Decibel) (a f actor of 100–1,000). Moreo ver, bubbles have a strong nonlinear response to sound, creating a unique signature that per mits echoes from the tr ue microcirculation to be differentiated from those of other tissues. Finally, microbubbles can be disrupted or “popped” with a controlled ultrasound pulse creating a quantitati ve means of probing perfusion of selected tissue re gions. In a rabbit model, microbubble contrast injection improved visualization of tumor neo vascularity. Earl y tumors had homo geneous vasculature but, with time, the centers became less vascular, w hile the peripher y increased. Neo vascularity was detected b y contrast injection before the tumor could be palpated or visualized by gray scale ultrasound.101 Niermann and colleagues102 evaluated the ability of dynamic microbubble contrast-enhanced sono graphy (MCES), in comparison with DCE-MRI and fluorodeo xyglucose positron emission tomo graphy (FDG-PET), to quantitatively characterize tumor perfusion in implanted murine tumors before and after treatment with a v ariety of re gimens. MCES revealed longitudinal decreases in tumor perfusion, blood volume, and microvascular velocity over the 5-day course of chemoradiotherap y, w hich is cor relative with DCE-MRI and FDG-PET . In a clinical study with 47 patients who underwent color Doppler US before and after intravenous injection of a microbubble contrast agent, carcinomas and benign lesions beha ved dif ferently in

Tumor Vasculature

degree, onset, and duration of Doppler US enhancement. However, high interindi vidual v ariability and temporal variations in the Doppler US signal still limit the v alue of these criteria for prospective diagnosis.103 Specialized contrast-specif ic US techniques ha ve been de veloped for impro ving image qualities. F or example, pulse inversion is the most kno wn phase modulation technique. This method combines the nonlinear properties of contrast agents and the ability of high intensity beams to destro y microbubb le agents. 104,105 Imaging begins with the transmission of a high-powered ultrasound beam into the tissue plane to be probed. This beam destroys the entire contrast agent in a slab of tissue the width of the ultrasound beam. Subsequently, images are made with lo w-powered beams in the same plane at different times. At short times, these images report only flow of contrast from adjacent tissue that enters the cleared plane via lar ger v essels with b lood flo wing at high velocity. Brightness in these images is propor tional to microvascular volume. This imaging method is particularly attractive because it can be readily applied to both animal models and human imaging. In po wer modulation, a train of tw o low transmit po wer US pulses with different amplitude is transmitted per image line, and the second pulse is half height amplitude relative to the first pulse. The received echoes are subtracted and result in effective remo val of tissue clutter . Cadence contrast pulse sequencing works by interrogating each scan line a number of times with pulses ha ving various amplitudes and phases. 106 Given the fundamental assumption that the relation betw een microbubb le concentration videointensity is linear up to the achie vement of a plateau phase, in animal models CEUS can quantify tumor v ascularity deter mined b y neoangio genesis.107–109 For the summary of different mathematic functions proposed to anal yze the replenishment kinetic of microbubb les, please refer to ar ticle b y F oster and colleagues. 100 For example, the negative exponential function has the form:

y = A(1 − e − α t ) + C ,

(9)

where A is the amplitude of the curve above baseline, α is the initial slope of the cur ve, and C is the intensity at baseline (the zero crossing point of the y-axis), and it is based on a single compar tment model. 110 The slope of the f irst ascending cur ve is related to microb ubble velocity, while the plateau phase is related to fractional blood v olume. Limitations of this model include the assumption of a constant concentration of microbubbles entering the ROI immediately after the destruction pulse and the neglecting of the different directions of the

735

vessels inside the examined ROI. Another sigmoid function model is based on the assumptions that microb ubble destr uction actuall y occurs in the feeding v essels that reach the ROI and that microbubbles present a constant concentration in the circulation.111 The multivessel model tak es into account the dif ferent directions and different amounts of blood flow that is found in v essels inside the ROI.112 After the microb ubble destruction by high transmitting power insonation, the replenishment is initially linear with time. Once the v essels in the R OI that are perpendicular to the US section and that demonstrate the maximum blood flow velocity of all vessels are completel y ref illed, a nonlinear increase in echosignal intensity occurs and is follo wed by saturation to maximum. The blood volume is then proportional to the maximum plateau of the replenishment cur ve. The use of ultrasound contrast agents and nonlinear processing pro vide access to the bulk proper ties of the microvascular compar tment, but the y do not of fer suf ficient resolution to obser ve the mor phology and detailed flow characteristics of the micro vasculature. High-frequency ultrasound (ultrasound biomicroscop y—UBM) refers to the use of frequencies appro ximately an order of magnitude higher than typical clinical frequencies. Because resolution is directl y propor tional to ultrasound frequency, it is possible to achieve resolution ranging from 15 to 100 µm in the 20 to 100 MHz range. Ho wever, it is subject to an inherent trade-off between image resolution and imaging depth. 113 For e xample, ultrasound in the 40 MHz range will pro vide resolution of appro ximately 60 µm but will be limited to penetration on the order of 6 mm. This probably precludes its use in the majority of clinical studies. Ho wever, high-frequenc y ultrasound ma y pro ve particularly suitable and may reveal unique patterns of neovascular growth and regression not observed with low-frequency ultrasound studies of deeper tumors for studies of ocular melanoma, malignant melanoma, basal cell carcinoma, and Kaposi’s sarcoma.

Optical Imaging Optical imaging is a relati vely low-cost method. In fluorescence imaging, excitation light illuminates the subject, and a charge-coupled device camera collects the emission light at a shifted wavelength.114 The fluorescent probe can be either injected or geneticall y engineered, and no substrate is required for its visualization. 115 A number of high-resolution microscopic imaging techniques ha ve recently been de veloped to study molecular e vents in vivo. In par ticular, intra vital fluorescence microscop y, confocal laser scanning microscop y, multiphoton laser

736

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

scanning microscop y (MPLSM), and in situ scanning force microscop y ha ve recentl y been introduced. 116–119 The major limitations of optical imaging are tissue light scattering and absor ption that af fect both image resolution and depth of light penetration of tissues. 120 In the ultraviolet and visib le re gions, tissue scattering and absorption of light is high, which limits its tissue penetration. At the near infrared re gion (NIR) betw een 700 and 900 nm, absor ption is low and allows light to penetrate much deeper into tissues, a depth that ma y be sufficient to practicall y image small animals and cer tain human cancers. Endoscopic NIR optical imaging should make it possib le to image cer tain inter nal or gans in patients. In addition, multiphoton microscopy has a much better signal-to-noise ratio, g reater imaging depth, and longer sample lifetimes. It pro vides a powerful approach to measure tumor angiogenesis and functional indexes of tumors.112 Optical imaging offers single cell resolution and realtime imaging to monitor tumor host interaction, w hich is a critical component in tumor development especially for tumor angiogenesis. GFP-labeled tumor cells in combination with the dorsal skinfold window chamber allow the obser vation of tumor g rowth from indi vidual tumor cells, w hereas earlier studies depended mostl y on immunohistochemistry of well-established tumors which does not allow similar observations of earlier angiogenic activities at equi valent spatial and temporal resolutions.121 The results have shown that tumor cells alter the host “normal vasculature” morphology in just a few days. The surrounding host “normal” vessels become dilated and torturous. In addition, when tumor cells reached the blood v essels, the y g row around the v essel and for m a cuff, and then g row along the v essels. This model can also be applied to moni tor t umor v ascular sur vival and v ascular r esponse t o t herapy s uch a s e xternal irradiation.122 Tumor vascular permeability, vessel size, and b lood flow are impor tant functional inde xes of tumor b lood vessels. Optical imaging has the ability to deter mine these parameters in vivo. Optical imaging offers great clarity for imaging b lood v essels, w hich has been elegantly demonstrated b y using fluorescent dy e or quantum dots. 123 Tumor v essels are generall y leak y with a heterogeneous per meability that depends on the tumor site. Multiphoton microscopy in combination with fluorescence-labeled molecules can be used to quantify the permeability of individual tumor blood vessels noninvasively deep inside li ving animals. 112 Using a MPLSM, 3D images of tumor v essels can be tak en deep inside tumors. When RBC v elocity measurements are

performed in tumor v essels in vi vo using MPLSM, velocities are quantif iable in the majority of v essels, with the e xception of the f astest flo w rates at the center of a fe w lar ge vessels. MPLSM also allo ws quantification of the per meability of indi vidual tumor vessels noninvasively deep inside living tumors (around 300 µm). When tetrameth ylrhodamine-BSA (TMRBSA) is injected for per meability measurement using MPLSM, e xtravasation from a chosen v essel can be monitored to deter mine its per meability. The extravasated TMR-BSA increases the fluorescence signal in the interstitium. The local permeability of the vessel to the molecule is then calculated from the dynamics of the accumulated fluorescence using the following equation. Fluorescently labeled molecules were injected into the b loodstream of an animal at t = 0, and the fluorescence signal in a single optical section containing a v essel of interest w as monitored. The per meability P of a v essel of radius r was then calculated. ∞

δ ∫r = R F ( r )rdr P = lim t →0 δ t ( F − F ) R v i

(10)

Here Fv is the fluorescence intensity from the plasma in the v essel and Fi is the fluorescence intensity immediately outside the vessel. The integral of F(r), the fluorescence intensity in the e xtravascular space, w as evaluated numerically along a line per pendicular to the flow axis of the v essel. This derivation assumes that the relationship between fluorescence signal and local concentration of fluorophore is uniform across the line of interest and there is no influx from adjoining v essels. Single-vessel per meability measurements can pro vide mechanistic insights into the causes and consequences of hyperpermeability in tumors. Ho wever, it is dif ficult to obtain o verall v olume-averaged v ascular per meability with this technique. Quantitative, 3D fluorescence-mediated tomo graphic (FMT) technique can be applied to measure angio genesis by using magnetofluorescent nanopar ticles.124 Because optical measurements can be obtained at multiple w avelengths in the near -infrared spectrum, FMT is par ticularly useful for serial and temporal imaging in the steady state. On nine different tumors, angiogenesis measurement from the optical measurements cor related w ell with CD31 microvascular density measurements ( r = 0.88) and MRI of v ascular v olume fraction (VVF) ( r = 0.95). The best correlation was observed between sensitive radiotracer measurements of tagged RBC and optical measurements

Tumor Vasculature

(r = 0.97). Another study perfor med b y the same g roup imaged and quantif ied VVF with 3D fluorescence molecular tomo graphy (FMT) using long-circulating near infrared fluorescent blood-pool agents optimized for two nonoverlapping excitation wavelengths (680 and 750 nm). With a known blood concentration of the tracers, tumoral VVFs were calculated as VVF = (Ct/Cb) × 100, where Ct is

737

the fluorochrome concentration in the tumor and Cb is the fluorochrome concentration in b lood. Fluorochrome concentrations deri ved from FMT measurements w ere reconstructed with an accurac y of ±10% at 680 nm and ±7% at 750 nm and in a depth-independent manner, unlike with reflectance imaging (F igure 2). FMT measurements of vascular fluorescent probes were linear with concentration

A

B

Figure 2. Serial coronal reconstructions (reconstruction voxel, 200 × 200 × 400 µm) of tumors. Imaging was performed in (A) ectopically and (B) orthotopically implanted colonic tumors. FMT enabled imaging of near-infrared fluorescent probes in tumor neovasculature in (A) superficial and (B) deep-seated tumors. Tumors had similar mean VVF in the two locations. Arrows = tumor, * = bladder.125

738

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

over several orders of magnitude ( r > 0.98). The natural course of angiogenesis and its inhibition could be reliab ly imaged and depicted seriall y in dif ferent e xperimental setups.125 Even though FMT -derived fluorochrome measurements are less depth dependent (0–8 mm) than those obtained with conventional reflectance imaging, it is difficult to apply to tumor tissue deeper than 1 cm.

IMAGING OF MOLECULAR MARKERS OF TUMOR VASCULATURE Although techniques like DCE-MRI and [15O]H2O PET for assessment of hemodynamic parameters are widel y used, the inter pretation of the results with re gard to their ph ysiologic meaning o ften remains dif ficult. Therefore, more specif ic markers of angio genic activity in tumors are necessar y for pretherapeutic assessment of angio genesis and response e valuation during therapy. One approach is to identify molecular markers of angio genesis such as receptors, enzymes, or ECM proteins and to use specific ligands to these targets conjugated with imaging probes for PET , SPECT , MRI, optical imaging, or ultrasound. One of the most promising targets in this respect is the integrin αvβ3, which, up to now, is one of the few markers of angiogenesis which has been successfully imaged in patients by using PET and SPECT techniques. 126,127

Radiotracer Techniques Imaging of Integrin αvβ3 Expression Using Monomeric and Multimeric Tracers

Integrins are heterodimeric transmembrane glycoproteins consisting of dif ferent α- and β-subunits which play an important role in cell-cell- and cell-matrix-interactions. Especially well examined is the integrin αvβ3 and its role in angio genesis and tumor metastasis b y f acilitating endothelial and tumor cell mig ration. It has been found that se veral ECM proteins lik e vitronectin, f ibrinogen, and fibronectin interact with integrins via the amino acid sequence ar ginine-glycine-aspartic acid or RGD in the single letter code. 128 Haubner and colleagues 129 developed the pentapeptide cyclo(-Arg-Gly-Asp-DPhe-Val-), which shows high affinity and selectivity for αvβ3. For the f irst evaluation of this approach, Haubner and colleagues synthesized radioiodinated RGD peptides which showed comparable affinity and selecti vity to the lead str ucture. In vi vo they re vealed receptor -specific tumor uptak e b ut also

predominantly hepatobiliar y elimination, resulting in high activity concentration in li ver and intestine, w hich is unf avorable for patient studies. 130 Consequently se veral strategies to improve the pharmacokinetics of radiohalogenated peptides ha ve been de veloped. The glycosylation approach is based on the introduction of sugar deri vatives which are conjugated to the ε-amino function of a cor responding lysine in the peptide sequence. By conjugating the RGD containing cyclic pentapeptide c yclo(-Arg-Gly-Asp-DPhe-Lys-) with glucose- or galactose-based sugar amino acids, [*I]Gluco-RGD and [18F]Galacto-RGD have been developed for PET and SPECT imaging. Both compounds demonstrated improved pharmacokinetics with predominantly renal tracer elimination and increased uptake and retention in a murine tumor model compared with the first generation peptides. 131 The conjuga tion of hydrophilic D-amino acids is another strategy to improve the pharmacokinetics of peptide-based tracers. 132 Again, tracer elimination could be shifted to the renal pathw ay, but tumor uptake of the compound [18F]DAsp3-RGD was lower than that found for [ 18F]Galacto-RGD. However, tumor/background ratios calculated from small animal PET images w ere comparab le due to the e ven f aster elimination. PEGylation is another way to improve many properties of peptides and proteins, lik e phar macokinetics, plasma stability, and immuno genicity.133 Chen and colleagues conjugated RGD-containing peptides with PEG moieties of dif ferent sizes, using dif ferent radiolabeling strategies. These studies revealed very different effects of PEGylation on the pharmacokinetics and tumor uptake and retention of RGD peptides, w hich seem to depend strongly on the nature of the lead str ucture and perhaps on the size of the PEG moiety.134 Besides radiohalogenated RGD peptides, a variety of radiometallated tracers ha ve been de veloped as w ell, including peptides labeled with [ 111In], [ 99mTc], [ 64Cu], [90Y], and [ 188Re]. Most of them are based on the c yclic pentapeptide and are conjugated via the ε-amino function of a lysine with different chelator systems, like DTPA, the tetrapeptide sequence H-Asp-L ys-Cys-Lys-OH, and DOTA. Although all these compounds ha ve shown high receptor affinity and selectivity and specif ic tumor accumulation, the pharmacokinetics of most of them still have to be impro ved.135 However, the compound [99mTc]NC100692 b y GE Healthcare has been successfully used for SPECT imaging in preclinical and clinical studies.127 Extensive preclinical e valuations concer ning monomeric compounds have been carried out using [*I]Gluco-RGD and [ 18F]Galacto-RGD.136,137 Initial in

Tumor Vasculature

vivo e valuation w as car ried out using the human melanoma M21 model, w hich is well characterized concerning αvβ3 expression.138 Using this model, uptake of [18F]Galacto-RGD and [ 125I]Gluco-RGD in the tumor 120 minute post-injection (p.i.) was 1.5 and 1.8% ID/g, respectively. Blocking e xperiments injecting 6 mg c(RGDfV) per kg mouse 10 minute prior to tracer injection reduced tumor accumulation to appro ximately 15% of control for [ 125I]Gluco-RGD and to appro ximately 35% of control for [ 18F]Galacto-RGD, w hich demonstrates receptor specif ic accumulation. Fur thermore, imaging studies with mice bearing melanoma tumors with increasing amounts of αvβ3-positive cells (produced by mixing M21 and M21-L cells) sho wed that there is a correlation between integrin expression and tracer accumulation.126 These data demonstrate that nonin vasive determination of αvβ3 expression and quantification with radiolabeled RGD peptides is feasib le with static emission scans. Moreover, animal PET studies with increasing amounts of c(RGDfV) w ere obtained that indicated that the dose-dependent b locking of tracer uptak e in the receptor positive tumor could be monitored. Additional experiments were car ried out with the RIPT ag model, a transgenic mouse model of carcinogenesis and angiogenesis w here the onco gene SV40 T antigen is e xpressed under the control of the rat insulin promoter .5 Immunohistochemical staining of pancreatic sections with a monoclonal antibody against the murine β3 subunit indicated that αvβ3 is e xclusively e xpressed on the v essels of the insulinoma and not on the tumor cells themselv es. Autoradiographic studies of pancreatic sections of mice sacrificed 2 hours after IV injection of [ 125I]Gluco-RGD showed high focal acti vity accumulation in the pancreas in RIP-Tag-positive mice.139 In contrast, for the RIP-Tagnegative mice, onl y lo w acti vity accumulation cor responding to the backg round was found. Moreover, there was a clear increase in acti vity accumulation in the pancreas of RIP-Tag-positive mice betw een 7 and 9 w eeks, which corresponds with the tumor differentiation. In contrast, for RIP-Tag-negative mice, tumor accumulation remained low over the whole observation period (Figure 3). In conclusion, these data suggest that αvβ3 expression can be quantified by radiolabeled RGD-peptides using static emission scans, and that αvβ3 expression on endothelial cells can be monitored sequentiall y during tumor -induced angio genesis with radiolabeled RGD-peptides. Up to no w, the onl y approach to imaging αvβ3 expression that has made the transition into the clinic is the radiotracer approach. [18F]Galacto-RGD was the first tracer applied in patients and could successfull y image

739

αvβ3 expression in human tumors with good tumor/ background ratios (F igure 4). In all patients, rapid , predominantly renal tracer elimination was observed, resulting in lo w backg round acti vity in most re gions of the body.140 High inter- and intra-individual variance in tracer accumulation in tumor lesions w as noted , suggesting great diversity of αvβ3 expression. Further biodistribution and dosimetry studies have confirmed rapid clearance of [18F]Galacto-RGD from the b lood pool and primaril y renal excretion. Background activity in lung and muscle tissue w as low, and the calculated ef fective dose found was approximately 19 µSv/MBq, which is very similar to an [ 18F]FDG scan. 141 Distribution v olume (D v) v alues, which are supposed to reflect the receptor concentration in the tissue, were on average four times higher for tumor tissue than for muscle tissue, suggesting specif ic tracer binding. In a recent study , dynamic emission scans o ver 60 minutes and kinetic modeling studies using the aorta as AIF were perfor med in patients with in vasive ductal breast cancer. We compared SUVs deri ved from the last 9 time frames with the D v values measured in normal tissue (breast, muscle) and in the tumors. The cor relation between both parameters increased continuousl y o ver time was best at ~55 minutes p.i. with r = 0.92. 142 This suggests that SUVs derived from static emission scans at ~60 minutes p.i. can be used for assessment of αvβ3 receptor density with reasonable accuracy. Scans should not be star ted much earlier , as unspecif ic ef fects lik e tracer acti vity in the b lood pool are lik ely to af fect the SUVs. Beer and colleagues also studied if [ 18F]GalactoRGD uptak e cor relates with αvβ3 expression. Nineteen patients with solid tumors (musculoskeletal system n = 10, melanoma n = 4, head and neck cancer n = 2, gliob lastoma n = 2, breast cancer n = 1) were examined with PET using [ 18]Galacto-RGD before sur gical remo val of the lesions.143 SUVs and tumor/b lood ratios w ere found to correlate significantly with the intensity of immunohistochemical staining as w ell as with the MVD . Moreover, immunohistochemistry conf irmed lack of αvβ3 expression in normal tissue and in the two tumors without tracer uptake. We are no w systematicall y e xamining dif ferent tumor entities with respect to their αvβ3 expression patterns as sho wn by [ 18F]Galacto-RGD PET. In squamous cell carcinoma of the head and neck (SCCHN), w e demonstrated good tumor/backg round ratios with [18F]Galacto-RGD PET, but again also a widel y varying intensity of tracer uptak e. Immunohistochemistr y demonstrated predominantl y v ascular αvβ3 expression, thus in SCCHN, [18F]Galacto-RGD PET might be used as a sur rogate parameter of angio genesis.144 We ha ve also compared the tracer uptak e of [ 18F]FDG and

740

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

B

C

A

Figure 3. Imaging of the angiogenic switch in RIP-Tag+ mice with [125I]Gluco-RGD. The autoradiographic sections (A) show tracer uptake in RIP-Tag+ mice at 12 weeks but not in RIP-tag- mice. Immunohistochemistry (B) shows β3-positive microvessels in the insulinomas but not in the normal islets. The graph shows the time course of tracer uptake in RIP-Tag+ and RIP-Tag- mice over time (C).

A

B

C

Figure 4. [18F]Galacto-RGD PET of a male patient with invasive ductal breast cancer on the left side (arrow, closed tip). The maximum intensity projection (A) and the axial section (B) show intense uptake in the tumor. Note the physiologic biodistribution of the tracer with predominantly renal elimination (kidneys, arrow, dotted line) and to a lesser extent hepatobiliary elimination (gallbladder, arrow, open tip). Moderate tracer uptake is notable in the liver, spleen, and intestine, whereas there is only low tracer uptake in lungs, mediastinum, and musculoskeletal system. Immunohistochemistry of αvβ3 expression shows intense staining predominantly of the neovasculature (C).

Tumor Vasculature

[18F]Galacto-RGD in patients with NSCLC ( n = 10) and various other tumors ( n = 8), because in case of a close correlation of the two tracers, there would probably be no need for a specif ic tracer lik e [ 18F]Galacto-RGD. The results sho wed no cor relation betw een the tw o tracers concerning all lesions ( r = 0.157). F or the subg roup of [18F]FDG-avid lesions and lesions in patients with NSCLC, there w as a slight trend to ward a higher [18F]Galacto-RGD uptake in more [18F]FDG-avid lesions (r = 0.337). However, the correlation coefficient was very low. Our results suggest that αvβ3 expression and glucose metabolism are not closel y cor related in tumor lesions and that consequentl y [ 18F]FDG cannot pro vide similar information as [ 18F]Galacto-RGD.145 Recently, the SPECT tracer [ 99mTc]NC100692 was introduced b y GE healthcare for imaging αvβ3 expression in humans and was first evaluated in breast cancer by Bach-Gansmo and colleagues127 19 of 22 tumors could be detected with this agent, which was safe and w ell tolerated by the patients. It is therefore e xpected that commercial agents for αvβ3 imaging will soon be available. However, lesion identif ication is still dif ficult in areas with ph ysiologic [ 18F]Galacto-RGD uptake, such as liver, spleen, and intestine. Variations in tracer design are supposed to fur ther impro ve the perfor mance of αvβ3 imaging, e g, using multimeric RGD peptides. Since the interaction betw een inte grin αvβ3 and RGDcontaining ECM-proteins in volves multi valent binding sites with clustering of integrins, the concept to improve the integrin αvβ3 binding affinity with multimeric cyclic RGD peptides could provide more effective antagonists with better tar geting capability and higher cellular uptake through the inte grin-dependent endoc ytosis pathway.146 A series of RGD peptides have been labeled with [ 18F] for PET imaging b y Chen and colleagues, using PEGylation and pol yvalency to impro ve the tumor-targeting ef ficacy and phar macokinetics. [18F]FB-E[c(RGDyK)]2 ([18F]FRGD2) showed predominantly renal excretion and almost twice as much tumor uptake in the same animal model compared with the monomeric tracer [ 18F]FB-c(RGDyK).147,148 Linear regression analysis of the dynamic microPET scans in six tumor xenograft models was carried out to cor relate the tumor uptak e with inte grin αvβ3 expression le vel measured b y SDS-PAGE autoradio graphy and sho wed an e xcellent cor relation. Moreover, at late time points when most of the nonspecific binding had been cleared, the tumor/backg round ratio had a linear relationship with tumor integrin αvβ3 expression. This demonstrates that quantif ication of αvβ3 expression is also feasib le with static emission scans, w hich facilitates translation

741

into the clinic. Multimeric RGD peptides have also been labeled with [ 64Cu] for PET imaging, using PEGylation and a multimeric approach to optimize the tumor-targeting efficacy and pharmacokinetics.149–151 The tetrameric 64 Cu]DOTARGD peptide-based tracer ,[ E[E[c(RGDfK)]2]2, sho wed signif icantly higher receptor binding af finity than the cor responding monomeric and dimeric RGD analogs and demonstrated rapid blood clearance, high metabolic stability , predominant renal excretion, and signif icant receptor -mediated tumor uptake with good contrast in x enograft-bearing mice. 152 Therefore, [64Cu]DOTA-E[E[c(RGDfK)]2]2 is a promising agent for peptide receptor radionuclide imaging as well as tar geted inter nal radiotherap y of αvβ3-positive tumors. Compared with tetramer , RGD octamer fur ther increased the integrin avidity by another 3-fold. In vivo microPET i maging s howed t hat [ 64Cu]DOTA-RGD octamer had slightl y higher initial tumor uptak e and much longer tumor retention in U87MG tumor that express high level of integrin (Figure 5). However, the octamer exhibited signif icantly higher tumor uptak e in mammary adeno carcinoma-bearing c- neu oncomice that e xpress medium le vel of inte grin. The high renal uptake of the octamer in both subcutaneous U87MG xenografts and mammar y adenocarcinoma–bearing cneu oncomice compared with the tetramer w as attributed mainly to the inte grin positivity of the kidne ys.153 Wester and K essler g roups ha ve also synthesized a series of monomeric, dimeric, tetrameric, and octameric RGD peptides.154–156 These compounds contain different numbers of c(RGDfE) peptides connected via PEG linker and lysine moieties, which are used as branching units. Ov erall, the multimerization approach leads to increased binding af finity and tumor uptak e as w ell as retention and can impro ve the phar macokinetics of peptide-based tracers. Nanoparticle-Based Radiotracers for α vβ3 Imaging

In general, the pur pose of nanopar ticle-based radiotracers for αvβ3 imaging is not to evaluate receptor expression levels but to provide guidance for integrin targeted drug delivery or therap y, which is a little dif ferent from pre viously described peptide- or antibody-based imaging. There have been onl y a fe w repor ts on PET imaging of tar geted nanoparticles to date. Cai and colleagues157 recently developed a QD-based probe for both NIR fluorescence (NIRF) and PET imaging. QD surf ace modif ication with RGD peptides allo ws for inte grin αvβ3 targeting and DO TA (1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid;

742

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

Figure 5. Small-animal PET studies of U87MG tumor-bearing mice and c-neu oncomice. A, Decay-corrected whole-body coronal images of athymic female nude mice bearing U87MG tumor at 30 min and at 1, 2, 6, and 20 h after injection of about 9 MBq of [64Cu]DOTA-RGD tetramer or [64Cu]DOTA-RGD octamer. B, Coronal images of U87MG tumor-bearing mice at 2 h after injection of [64Cu]DOTA-RGD tetramer or [64Cu]DOTA-RGD octamer without and with (denoted as “blocking”) coinjection of 10 mg of c(RGDyK) per kilogram of mouse body weight.156

a very effective chelator for many metal ions) conjugation enables PET imaging after [ 64Cu]-labeling. Using this dual-modality probe, it w as found that the majority of the probe in the tumor was within the tumor vasculature. Single-walled carbon nanotubes (SWNTs) e xhibit unique size, shape, and ph ysical proper ties that mak e them promising candidates for biolo gic applications.158,159 Liu and colleagues160 recently investigated the biodistribution of [64Cu]-labeled SWNTs in mice by PET, biodistribution, and e x vivo Raman spectroscopy.

It was found that properly PEGylated SWNTs have relatively long circulation half-life (a fe w hours) and low uptake by the RES. Ef ficient targeting of inte grin αvβ3-positive U87MG tumor in mice (~15% ID/g), among the highest of an y nanopar ticles ever repor ted, was also achie ved with RGD peptide conjugated SWNTs. The unique Raman signatures of SWNTs enabled direct measurement of SWNTs in various mice tissue which conf irmed the radionuclide-based results. Virtually no kidne y uptak e w as obser ved based on

Tumor Vasculature

Raman measurement of the tissue homo genate, although a small fraction of [ 64Cu] detached from the SWNT did gi ve appreciab le kidne y uptak e in PET imaging. Fur ther in vestigation of using SWNT as a nanoplatform, which has large surface areas that can be potentially functionalized in a variety of ways to attach therapeutic agents and other moieties, 157 for integrated multimodality imaging and molecular therap y is currently under way. Imaging of MMPs

MMPs are zinc endopeptidases which are responsible for the enzymatic degradation of connective tissue and thus facilitate endothelial cell migration during angiogenesis. From more than 18 members of the MMP f amily, the gelatinases MMP 2 and 9 are most consistentl y detected in malignancies and are therefore interesting for assessment of angiogenesis.161 Many strategies have been used for radiolabeling of MMP specif ic ligands. Via phage display techniques, the MMP specif ic decapeptide HCys-Thr-Thr-His-Trp-Gly-Phe-Thr-leu-Cys-OH (CTT) was found and could be labeled via the Iodogene method by adding a d-Tyr at the N-terminal end of this decapeptide. However, this tracer has unfavorable characteristics for in vi vo imaging because metabolic stability of the compound is low and lipophilicity is high. Another group optimized phar macokinetics b y adding a highl y hydrophilic and ne gatively char ged radiolabel, [111In]DTPA, to the N-ter minal residue distant from the HWGF motif ([111In]DTPA-CTT). In MMP-2 inhibition assays, [111In]DTPA-CTT significantly inhibited the proteolytic activity in a concentration-dependent fashion. In in vivo studies, [ 111In]DTPA-CTT showed low levels of radioactivity in the liver and kidney. Moreover, a significant cor relation w as obser ved betw een the accumulation in the tumor and tumor -to-blood ratio of [111In]DTPA-CTT and gelatinase acti vity.162 Therefore, this seems to be a promising candidate for imaging of gelatinase activity in metastatic tumors in vi vo. Another approach is labeling small molecule MMP inhibitors, which are also used as antiangio genic drugs. Different [ 18F]- and [ 11C]-labeled MMP inhibitors ha ve been synthesized and evaluated preclinically with mixed results.163 One of the more promising substances is based on a MMP inhibitor belonging to the f amily of N-sylfonylamino acid derivative. A [11C]-labeled analog was synthesized, which showed f avorable phar macokinetics in mice and metabolic stability up to 30 minutes after injection. Ho wever, none of these compounds has been used in patient studies up to no w. Another

743

approach for imaging MMPs is targeting tissue inhibitors of metalloproteinases (TIMPs) w hich represent a family of natural secreted and soluble (20–29 kDa) glycoproteins (TIMPs 1–4) that inhibit MMP activity by formation of a nonco valent binding with their catal ytic binding site. 164 N-TIMP-2 w as conjugated with the chelator DTP A and labeled with [ 111In] for SPECT imaging.165 A clinical imaging study w as perfor med in five patients with Kaposi sarcoma, however, none of the patients showed significant uptake in the lesions. Therefore, in general, significant improvements in pharmacodynamics and phar macokinetics are necessar y before use of MMP radiotracer imaging will translate into the clinic. Imaging of Extra Domain B of Fibronectin

Fibronectin is a large glycoprotein, which can be found physiologically in plasma and tissues. Ho wever, the extra domain B (EDB) of f ibronectin, consisting of 91 amino acids, is not present in the f ibronectin molecule under nor mal conditions, e xcept for the endometrium in the proliferati ve phase and some v essels of the ovaries. EDB is interesting as a marker of angiogenesis as it is expressed in a variety of solid tumors, as well as in ocular angio genesis and w ound healing. 166 The human antibody fragment scFv(L19) has been shown to efficiently localize on neo vasculature both in animal models and in cancer patients. In a study with patients suffering from v arious solid tumors, 16 of 20 tumor lesions could be identif ied by SPECT using [ 123I]scFv (L19). Whether the unidentif ied tumors w ere not detected because the y w ere either in a phase of slo w growth with lo w le vels of angio genesis or due to the technical limitations of SPECT imaging is not clear .167 No reports about PET tracers tar geting EDB are a vailable up to now. Radiotracer-Based Imaging of VEGF and Its Receptors

The VEGF family is composed of seven members with a common VEGF homology domain: VEGF-A, VEGF-B , VEGF-C, VEGF-D , VEGF-E, VEGF-F , and placenta growth f actor.168 VEGF-A is a dimeric, disulf ide-bound glycoprotein existing in at least se ven homodimeric isoforms, consisting of 121, 145, 148, 165, 183, 189, or 206 amino acids. Besides the dif ference in molecular weight, these isoforms also differ in their biologic properties such as the ability to bind to cell surface heparin sulfate proteoglycans. The angiogenic actions of VEGF are mainl y

744

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

mediated via two endothelium-specif ic receptor tyrosine kinases, Flt-1 (VEGFR-1) and Flk-1/KDR (VEGFR2).169 Both VEGFRs are lar gely restricted to v ascular endothelial cells, and all VEGF-A isoforms bind to both VEGFR-1 and VEGFR-2. It is no w generall y accepted that VEGFR-1 is critical for physiologic and developmental angio genesis and its function v aries with the stages of de velopment, the states of ph ysiologic and pathologic conditions, and the cell types in w hich it is expressed.170 VEGFR-2 is the major mediator of the mitogenic, angiogenic, and permeability-enhancing effects of VEGF. The critical role of VEGF-A in cancer progression has been highlighted b y the appro val of the humanized anti-VEGF monoclonal antibody be vacizumab (Avastin; Genentech) for f irst line treatment. 171 Development of VEGF- or VEGFR-targeted molecular imaging probes could serve as a new paradigm for the assessment of antiangio genic therapeutics and for better understanding the role and e xpression prof ile of VEGF/VEGFR in many angiogenesis-related diseases. With SPECT imaging, recombinant human VEGF121 was labeled with [111In] (t1/2 = 67.2 h) for identification of ischemic tissue in a rabbit model, w here unilateral hindlimb ischemia was created by femoral ar tery excision.172 VEGF121 has also been labeled with [ 99mTc] through an “Adapter/Docking” strategy, and the tracer w as tested in a murine mammar y carcinoma. 173,174 It was also applied to image tumor v asculature before and after dif ferent types of chemotherapy.175 [123I]VEGF165 has been reported as a potential tumor mark er.176 Despite the high receptor affinity of this tracer, biodistribution in A2508 melanoma tumor -bearing mice indicated lo w tumor -tobackground ratio, likely due to the lo w metabolic stability of the compound. Nonetheless, biodistribution, safety, and absorbed dose of [ 123I]VEGF165 were studied in nine patients with pancreatic carcinoma. 177 Following intravenous administration, sequential images w ere recorded during the initial 30 minutes p.i. Although the majority of primary pancreatic tumors and their metastases w ere visualized b y [ 123I]VEGF165 scans, the or gan with the highest absorbed dose was the thyroid due to severe deiodination. Another repor t e valuated the usefulness of [123I]VEGF165 for tumor localization in gastrointestinal cancer patients. 178 Dynamic acquisition w as initiated immediately after injection and car ried out until 30 minutes p.i. All patients then underw ent SPECT imaging at 1.5 hours p.i. Comparing the SPECT results with CT and MRI, the primar y and metastatic lesions w ere identif ied in some patients b y [ 123I]VEGF165 SPECT. Recently, [125I]-labeled VEGF121 and VEGF165 have also been used for biodistribution and autoradiography studies.179 As

with most other radioiodinated tracers, prominent activity accumulation in the stomach was observed due to deiodination. Interestingl y, [ 125I]VEGF121 accumulation in tumors decreased with increasing tumor volume suggesting that small tumors ha ve higher VEGFR e xpression than lar ge tumors. It w as also found that [ 125I]VEGF165 uptake w as g reater than that of [ 125I]VEGF121 in some organs (eg, kidney, heart, and lung) but lower in the other organs. The reason for dif ferent accumulation in these organs remains unclear although it has been hypothesized as related to the dif ferent expression levels of VEGFR-1 and VEGFR-2 in these or gans since VEGF165 binds to VEGFR-1 with higher affinity than VEGFR-2.180,181 A recombinant protein composed of VEGF165 fused through a fle xible G4S link er to the n-lobe of human transferrin (hnTf-VEGF) w as repor ted for imaging angiogenesis.182 The molecular w eight of hnTf-VEGF is 65 and 130 kDa for the monomer and dimer, respectively. [111In]hnTf-VEGF accumulated in U87MG human glioblastoma tumors (6.7% ID/g at 72 h p.i.) and the tumor uptak e decreased w hen coinjected with 100-fold excess of VEGF but not with apotransferrin. HnTf-VEGF represents a prototypic protein harboring the metal-binding site of transferrin for labeling with [111In] without the need to introduce metal chelators. Due to the soluble and more dynamic nature of VEGF, imaging VEGF expression is very difficult. Thus all the abovementioned reports used radiolabeled VEGF isoforms for SPECT imaging of VEGFR expression. Although the VEGF isoforms used in these studies all exist in nature and should have very strong binding affinity and specificity to VEGFRs, much research is needed in the future to improve the in vivo stability, target affinity/specificity, and pharmacokinetics of these radiopharmaceuticals. Another imaging modality, PET, may offer many advantages over SPECT, and the increasing popularity of the clinical PET and PET/CT scanners can significantly facilitate clinical translation of promising new tracers.183 A few radiolabeled antiVEGF antibodies have been reported. VG76e, an IgG1 monoclonal antibody that binds to human VEGF, w as labeled with [ 124I] for PET imaging of solid tumor xenografts in immune-def icient mice. 184 Whole-animal PET imaging studies revealed a high tumor-to-background contrast. Although VEGF specif icity in vi vo was demonstrated in this repor t, the poor immunoreacti vity (< 35%) of the radiolabeled antibody limits the potential use of this tracer. HuMV833, the humanized version of a mouse monoclonal anti-VEGF antibody MV833, w as also labeled with [ 124I], and the distribution and biolo gic ef fects of HuMV833 in patients in a phase I clinical trial w ere investigated.185 Patients with progressive solid tumors were

Tumor Vasculature

treated with various doses of HuMV833, and PET imaging using [124I]HuMV833 was carried out to measure the antibody distribution in and clearance from tissues. It w as found that antibody distribution and clearance w ere quite heterogeneous not onl y between and within patients but also between and within individual tumors, suggesting that intrapatient dose escalation approaches or more precisel y defined patient cohorts would be preferred in the design of phase I studies with antiangio genic antibodies lik e HuMV833. Cai and Chen 186 have labeled VEGF121 with [ 64Cu] for PET imaging of tumor angio genesis and VEGFR expression. DO TA-VEGF121 e xhibited nanomolar receptor binding affinity (comparable to VEGF121) in vitro. MicroPET imaging re vealed rapid , specif ic, and prominent uptak e of [ 64Cu]DOTA-VEGF121 (~15% ID/g) in highl y vascularized small U87MG tumor with high VEGFR-2 e xpression but signif icantly lo wer and sporadic uptake (~3% ID/g) in large U87MG tumor with low VEGFR-2 expression. Western blotting of tumor tissue l ysate, immunofluorescence staining, and b locking studies with unlabeled VEGF121 confirmed that the tumor uptake is VEGFR specific. This is the f irst report on PET imaging of VEGFR expression. This study also demonstrated the dynamic nature of VEGFR expression during tumor progression in that even for the same tumor model, VEGFR expression level varies by cancer types and animal/individual. Successful demonstration of the ability of [ 64Cu]DOTA-VEGF121 to visualize VEGFR expression in vivo should allow for clinical translation of this tracer to image tumor angio genesis and to guide VEGFR-targeted cancer therapy. In a follo w-up study, a VEGFR-2 specif ic fusion to xin VEGF121/rGel (composed of VEGF121 linked with a G4S tether to recombinant plant to xin gelonin) w as used to treat or thotopic glioblastoma in a mouse model. 187 Before initiation of 64 Cu]labeled treatment, microPET imaging with [ VEGF121/rGel was performed to e valuate the tumor targeting efficacy and the phar macokinetics. It w as found that [ 64Cu]DOTA-VEGF121/rGel exhibited high tumor accumulation/retention and high tumor -to-background contrast up to 48 hours after injection in gliob lastoma xenografts. Based on the in vivo phar macokinetics of [64Cu]DOTA-VEGF121/rGel, VEGF121/rGel was administered e very other da y for the treatment of or thotopic U87MG glioblastomas. Such study of tumor targeting efficacy and pharmacokinetics using radiolabeled dr ugs demonstrates the po wer of molecular imaging, w here cancer patients can also be selected for specif ic molecular cancer therapy based on pretreatment screening using a radiolabeled drug or drug analog. [64Cu] was also used

745

to site-specif ically label VEGF121, and it w as found that PEGylation sho wed considerab ly prolonged b lood clearance. Compared with [ 99mTc]-labeled analog where the tumor uptake (~2% ID/g) was lower than most of the normal organs and the kidne y uptak e w as about 120% ID/g, the PEGylated v ersion gave slightly higher tumor uptake (~ 2.5% ID/g) and lower kidney uptake at about 65% ID/g. Recentl y, the humanized monoclonal antibody be vacizumab that b locks VEGF-induced tumor angiogenesis has been labeled with the γ-emitting isotope [111In] and the PET isotope [89Zr] (t 1/2 = 78.4 h) for noninvasive in vivo VEGF visualization and quantif ication. [ 89Zr]-bevacizumab sho wed higher tumor uptak e determined by quantification of PET images than that of human [ 89Zr]IgG at 168 hours. The biodistribution of [89Zr]- and [ 111In]-bevacizumab at dif ferent time points was comparable.188 All VEGF-A isofor ms bind to both VEGFR-1 and VEGFR-2. In the imaging studies repor ted to date, specificity to either VEGFR-1 or VEGFR-2 has rarel y been achieved as most of the tracers are based on VEGF isoforms. Kidney has high VEGFR-1 expression which can tak e up VEGF-A based tracer and thus usuall y makes it the dose limiting or gan.189 Alanine-scanning mutagenesis, a widely used technique in the deter mination o f t he c atalytic o r f unctional r ole o f p rotein residues, has been used to identify a positi vely charged surface i n VEGF165 t hat m ediates t he b inding t o VEGFR-2.190 Arg82, L ys84, and His86, located in a hairpin loop, were found to be critical for binding VEGFR-2, w hile ne gatively char ged residues, Asp63, Glu64, and Glu67, were associated with VEGFR-1 binding. Mutations in the 63 to 67 re gion of VEGF exhibited only modest effects on VEGFR-2 binding but significant reduction in af finity with VEGFR-1. A VEGFR-2-specific PET tracer has been de veloped using the D63AE64AE67A mutant of VEGF121 (VEGFDEE) generated b y recombinant DN A technology. The renal uptak e of [ 64Cu]DOTA-VEGFDEE w as significantly lower than that of [ 64Cu]DOTA-VEGF121 as rodent kidneys expressed high levels of VEGFR-1 based on immunofluorescence staining. 191 With the de velopment of ne w tracers with better targeting efficacy and desirable pharmacokinetics, clinical translation will be critical for the maximum benefit of VEGF-based imaging agents. P eptidic VEGFR antagonists can be labeled with [ 11C] or [ 18F], and they may allow for higher throughput than antibody- or protein-based radiotracers, as 1 hour p.i. is usuall y sufficient for a peptide-based tracer to clear from the nontargeted organs and give high contrast PET images.

746

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

It can take several hours and even days before high contrast PET images can be obtained for antibody-based tracers.192

MRI MRI is widely used clinically to assess tumor growth and for response e valuation. Anatomic infor mation can be coregistered with functional and molecular information within a single imaging method. A fur ther adv antage compared with radiotracer techniques is that MRI does not use ionizing radiation and generall y is more widel y available than PET. MRI also of fers good depth penetration, and its resolution is usuall y higher than that of clinical PET scanners although this depends on the e xact protocol applied. A major disadv antage of MRI compared with radiotracer techniques is its lo wer sensitivity for the detection of targeted agents. Therefore, targeted molecular imaging agents for MRI ha ve not entered clinical trials y et, except for the f ibrin specif ic contrast agent EP2140R. 193 However, the prob lem of limited sensitivity might be overcome in the future by signal amplification strate gies that generate higher tar get to background contrast.194 The f irst MRI approach for molecular imaging of angiogenesis was imaging of αvβ3 expression. By using Gd3+-containing paramagnetic liposomes with a diameter of 300 to 350 nm and the αvβ3 specific antibody LM609 as a ligand , MRI of squamous cell carcinomas in a rabbit model w as successfull y achie ved.195 Peptidomimetic inte grin αvβ3 antagonist conjugated magnetic nanopar ticles were also used for MRI in a Vx-2 squamous cell carcinoma model with a common clinical MRI scanner at 1.5 T.196 The targeted nanoparticles increased the MR signal significantly in the periphery of the tumor at 2 hours p.i. By the same group, nude mice with human melanoma tumor xenografts were successfully imaged using αvβ3 integrin-targeted paramagnetic nanopar ticles.197 The authors claimed that v ery small regions of about 30 mm 3 of angiogenesis associated with melanoma tumor xenografts were visualized, which may enable characterizing and staging of early melanoma in a clinical setting. A star tup compan y named “K ereos” (http://www.kereos.de) is planning to use this approach commerciall y and in patient trials in the future. However, Gd3+ for enhancing the T1 contrast can only be reliably detected at millimolar le vels. SPIO nanopar ticles can be detected at a much lower concentration because of the high susceptibility induced b y this par ticles which leads to a decrease of the signal in T2 and especiall y

T2*-weighted sequences (“ne gative contrast”). 198 In a recent study, αvβ3 integrin–targeted USPIO nanopar ticles were used for nonin vasive dif ferentiation of tumors with high and lo wer area fractions of αvβ3-positive tumor v essels.199 After RGD-USPIO injection, T2*-weighted MRI identified the hetero geneous distribution of αvβ3-positive tumor v essels by an ir regular signal intensity decrease, whereas the signal intensity decreased more homo geneously in the control tumor with predominantly small and uniformly distributed v essels. Imaging based on T2w sequences is also possib le with IO-based dual modality nanoparticles, lik e RGD-CLIO-Cy5.5, w hich allo w both for MRI and fluorescence imaging.194 This approach combines the high sensitivity of optical imaging with the high spatial resolution of MRI. Our lab recentl y de veloped a bifunctional IO nanopar ticle probe for PET and MRI of tumor integrin αvβ3 expression. With a core size of 5 to 7 nm, pol y(aspartic acid) (P ASP)-coated IO nanopar ticles (PASP-IO) w ere coupled to c yclic RGD peptides and macrocyclic DOTA chelators to get a bifunctional probe— DOTA-IO-RGD conjugates. The probe bound specif ically to integrin αvβ3 in vitro. Both microPET and T2-weighted MRI s howed integrin-specific d elivery o f c onjugated RGD-PASP-IO n anoparticles. This bifunctional imaging approach may allow for earlier tumor detection with a high degree of accurac y and pro vide fur ther insight into the molecular mechanisms of cancer. Besides tar geting the inte grin αvβ3, a number of specific agents tar geted to acti vated endothelial tar gets have been tested with MRI. E-selectin is o verexpressed in proliferating endothelial cells, and as such can be used as a marker of angiogenesis.200 In a study by Kang and colleagues, 201 MRI and an anti-E-selectin antibody linked to IO nanopar ticles w ere used to demonstrate binding to human endothelial umbilical v ein cells (HUVEC) in vitro. MRI of the cells was able to reveal a significant T 2-weighted signal decrease for cells in the antibody tar geted g roup. Recentl y, this approach has also been used to image HUVEC cells in a mouse model in vivo.202 Although the VEGFR-2 is also a very promising potential target for MRI of angiogenesis, no reports about direct VEGFR-2 imaging with MRI in vi vo are available to date.

Ultrasound As described above, ultrasound has relati vely high spatial resolution (50–500 µm) yet it also has some disadvantages such as the relatively poor depth penetration (usually a few centimeters depending on the frequenc y used) and limited sensitivity.114 Further development of molecular imaging

Tumor Vasculature

with ultrasound will lik ely in volve the e xpansion of targeted disease states, impro vements in technolo gy for ligand attachment to microbubb les, characterization of the acoustic behavior of targeted contrast agents, and development of better methods for imaging tar geted ultrasound agents. Since acoustic destr uction of “pa yload-bearing” microbubbles can be used to deliver drugs or to augment gene transfection, angio genesis-targeted microb ubbles may also ha ve applications in site-specif ic therap y for ischemic tissues or tumors. 203 For integrin αvβ3 targeted CEU, the shell surf ace of microbubbles has been conjugated with echistatin, a peptide derived from the v enom of the viper Echis carinatus which bears the RGD motif. 204 Using g ray scale pulseinversion techniques, tumor b lood volume determined by CEU increased b y appro ximately 35% from day 14 to day 28, whereas microvascular blood velocity decreased, especially at the central por tions of the tumors. 205 In another study , antihuman inte grin αvβ3 antibody conjugated microbubbles and cyclic RGD peptide conjugated microbubbles have been prepared for in vitro ultrasonic analysis. Specif ic adhesion of these contrast agents to αvβ3-expressing cell monolayers was achieved in vitro, and acoustic studies illustrated a backscatter amplitude increase from monolayers exposed to the targeted contrast agents of up to 13-fold (22 dB) relati ve to enhancement due to control bubb les. Unfor tunately there w ere no in vivo CEU images a vailable.168 Tartis and colleagues 206 constructed drug delivery vehicles, referred to as acoustically active lipospheres (AALs), which are microbubbles surrounded by a shell of oil and lipid. In a re gion limited to the focal area of ultrasound application, circulating AALs are deflected by radiation force to a vessel wall and can subsequently be fragmented. RGD peptide conjugated AAL shell showed an increase in in vitro binding by 26.5fold over nontargeted agents. Toxicity assays demonstrate that paclitaxel-containing AALs exert a greater antiproliferative ef fect after insonation than free paclitax el at an equivalent concentration. The combination of ultrasound and molecular tar geting successfull y deli vered a model drug to the endothelium and interstitium of chorioallantoic membrane vasculature in vivo.206 In a human melanoma x enograft model, CEU measures of tumor neo vascularity were compared with the expression of molecular mark ers of angio genesis.207 After po wer Doppler and inter mittent pulse-in version harmonic imaging (PI-HI), the tumor tissues w ere surgically removed and sectioned in the same planes as the ultrasound images and immunohistochemical staining for VEGF were carried out. Although there is a trend of correlation betw een percent area stained with VEGF

747

and intermittent PI-HI results, no statistical significance was achieved. In a follo w-up study using a similar approach in tw o melanoma models, linear re gression analysis indicated statisticall y signif icant cor relations between percent area stained with VEGF and po wer Doppler and inter mittent PI-HI measures of tumor neovascularity.208 Power Doppler ultrasono graphy has been used to demonstrate the presence of b lood flow in small v essels, and it w as also found that the v ascular signal cor relates with histopatholo gic quantif ication of the v ascular density of syno vial tissue. 209 In all these studies, nonin vasive ultrasound imaging results w ere compared with ex vivo VEGF staining results. Although good correlation was observed in many cases, this is not noninvasive imaging of VEGF e xpression. It has not been until very recently that in vivo ultrasound imaging of VEGF/VEGFR expression was reported.210 Since most CEU imaging uses microb ubbles that are at least se veral micrometers in diameter , onl y the tumor endothelium can be tar geted as these microb ubbles are too lar ge to e xtravasate.201 Thus, VEGFR is an excellent candidate for tar geted ultrasound imaging since it is almost exclusively expressed on activated endothelial cells.211 In a mouse model of pancreatic adenocarcinoma, targeted microbubbles were used to image and quantify vascular effects of two different antitumor therapies in both subcutaneous and or thotopic pancreatic tumors. 201 Tumor-bearing mice w ere treated with anti-VEGF monoclonal antibodies and/or gemcitabine (a chemotherap y dr ug), and the localization of antibody-conjugated microbubb les to VEGFR-2 or VEGF-activated b lood v essels (the VEGF-VEGFR complex) was monitored by contrast ultrasound. Significant signal enhancement of tumor v asculature w as observed w hen compared with untar geted or control IgG-targeted microb ubbles. Video intensity from targeted microbubbles also cor related with the e xpression level of the tar get (VEGFR-2 or the VEGF-VEGFR complex) and with MVD in tumors under therapy. This study demonstrated that targeted microbubbles can be a novel and attracti ve tool for nonin vasive, v asculaturetargeted molecular imaging of tumor angio genesis and for in vivo monitoring of v ascular effects after therapy. In another report, Willmann and colleagues212 have imaged VEGFR-2 expression in two murine tumor models using anti-VEGFR2 monoclonal antibody conjugated microbubbles. CEU imaging using tar geted microbubbles sho wed signif icantly higher a verage video intensity compared with control microb ubbles in both tumor models, and video intensity w as significantly lo wer w hen b locked b y anti-VEGFR2

748

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

antibodies. Using CEU, the Gambhir group also imaged tumor angio genesis in murine tumor models with microbubbles targeted to VEGFR2. VEGFR2-tar geted microbubbles (MB VEGFR2) w ere de veloped specif ically to bind VEGFR2 (using anti-VEGFR2 antibodies and biotin-streptavidin chemistry). In cell culture, adherence of MB VEGFR2 on SVR cells (an immortalized murine endothelial cell line containing both SV40 lar ge T antigen and activated c-Ha-ras) was significantly higher than adherence of control microbubbles. In vivo CEU also showed signif icantly higher a verage video intensity when using MBVEGFR2 compared with control microbubbles for both angiosarcoma and malignant glioma tumors (Figure 6).

Optical Imaging Although optical imaging ma y not be widel y used in clinical settings, NIR (700–900 nm) approaches pro vide opportunities for rapid and cost-effective preclinical evaluation in small animal models before the more costl y radionuclide-based imaging studies. These approaches may also be translated into the clinic with fluorescencemediated tomo graphy (e g, for breast cancer imaging). Optical imaging has been used to study gene e xpression, tumor angiogenesis, physiologic function of tumors, and tumor metastasis.213,214 A way to impart molecular specificity into optical imaging agents is to synthesize reactive dye deri vatives and conjugate them to tar get-specific vehicles such as peptides, antibodies, or antibody fragments.215 Target-specific c yanine dy e conjugates with single chain antibodies directed against the angiogenesisspecific tar get protein ED-B-f ibronectin w ere repor ted by Neri and colleagues. 166 The fact that inte grin αvβ3 is expressed on both tumor v asculature and tumor cells makes it a prime target for molecular imaging, as extrava-

A

B

sation is not required to observe tumor signal. In the NIR region, the absorbance of all biomolecules reaches minima, providing a clear window for in vivo optical imagwn that NIR fluorescent ing.216 It has been sho dye–conjugated c yclic RGD peptide could be used to visualize s.c. inoculated integrin αvβ3-positive tumors.217 RGD containing peptides ha ve been conjugated to QD705 (emission maximum at 705 nm) and QD705RGD exhibited high affinity integrin αvβ3 specific binding in cell culture and e x vivo.218 In vivo NIRF imaging was successfully achieved in nude mice bearing subcutaneous integrin αvβ3-positive tumors. As neovasculature in m any t umor t ypes o verexpresses i ntegrin αvβ3, QD705-RGD has g reat potential as a uni versal NIRF probe for detecting tumor v asculature in li ving subjects. It is note worthy that the dy e-RGD peptide conjugate is small in size; therefore, it tar gets integrin αvβ3 on both tumor cells and tumor v asculature. For the quantum dotRGD peptide conjugate, it mainly targets integrin αvβ3 in the tumor vasculature because it does not extravasate well due to the relatively large size (≥ 20 nm). Choi and colleagues recentl y perfor med a systematic study of the effect of size and char ge on the in vi vo behavior of biocompatible quantum dots. It was observed that zwitterionic or neutral or ganic coatings of their par ticular quantum dots pre vented adsor ption of ser um proteins, which otherwise increased h ydrodynamic diameter . A final hydrodynamic diameter < 10 nm may be required for ef fective e xtravasation and size < 5.5 nm ma y be needed to minimize the RES uptak e and to induce rapid and efficient urinar y excretion and elimination of quantum dots from the body . Achilefu and colleagues 182 discovered that conjugating a presumab ly inacti ve linear hexapeptide GRDSPK with a NIR carbocyanine molecular probe yielded Cyp-GRD that tar gets inte grin αvβ3positive tumors. More experiments need to be carried out

C

Figure 6. Transverse color-coded ultrasound image of a subcutaneous 4.5 mm nude mouse angiosarcoma tumor (arrows) after intravenous injection of VEGFR2-targeted microbubbles (MBVEGFR2; before (A) and 30 min after (B) administration of anti-mouse VEGFR2 antibodies) and after injection of isotype-matched control IgG-labeled microbubbles (MBControl, (C)). The ultrasound signal is significantly reduced after blocking of VEGFR2. A very small signal was measured with control microbubbles before blocking with anti-mouse VEGF2 antibodies in this mouse (C).

Tumor Vasculature

to fully understand this surprising phenomenon, and docking study ma y re veal w hether Cyp-GRD actuall y binds to the RGD binding domain in inte grin αvβ3. The same group also synthesized and evaluated a series of multimeric RGD compounds constr ucted on a dicarboxylic acid–containing NIR fluorescent dy e cypate for tumor targeting.219,220 Tumor integrin αvβ3 expression in vivo also has been visualized b y using NIRF imaging of Cy5.5-linked cyclic RGD peptide in an or thotopic brain tumor model.221 Optimization of the spatial alignment of the RGD moieties through careful molecular design and library constr uction ma y induce multi valent ligandreceptor interactions useful for in vi vo tumor imaging and tumor targeted therapy. In a transgenic mouse model w here a VEGF promoter was chosen to dri ve a GFP repor ter gene, VEGF expression during w ound healing, and possib le impairment of wound healing due to collateral tissue damage, was imaged in vivo.204 Mice received two full thickness incisions in the dorsal skin: one with the free electron laser (FEL) and one with a scalpel. Afterwards, mice were imaged for GFP e xpression at multiple time points. It was found that GFP expression peaked at 2 to 3 weeks after sur gery and FEL lesions e xhibited more total GFP expression than scalpel lesions. This pioneering study demonstrated the feasibility of using transgenic mice car rying photoacti ve repor ter genes for studying cellular process in a nonin vasive manner . Human VEGF has also been conjugated to a self-assembled “dock and lock” system and retained its functional activities.222 After incor porating an additional c ysteine residue for site-specific modification, a NIR fluorescent dye Cy5.5 (maximum emission 696 nm) was conjugated and the resulting Cy5.5-VEGF was used for in vivo imaging. Although tumor contrast was observed after administration of the probe, no information was reported about the w hole body distrib ution of Cy5.5VEGF.223 This self-assembled “dock and lock” system may pro vide ne w oppor tunities of generating labeled functionally acti ve proteins for other biomedical purposes. Another component of optical imaging is bioluminescence imaging (BLI), w hich can be used to detect very low levels of signal because the emitted light is virtually backg round free. 224 BLI does not require an external light source. Instead, it detects light that is emitted from within the experimental animals through the action of an enzyme (usuall y luciferase) on its substrate (e g, Dluciferin). Noninvasive indirect imaging of VEGF expression with BLI in living transgenic mice has also been reported, w here a tw o-step transcriptional amplif ication

749

approach was used to augment the transcriptional acti vity of the relati vely w eak VEGF promoter .225 VEGF gene expression was imaged in both a wound-healing model and a subcutaneous mammar y tumor model. The BLI signal had good cor relation with the endo genous VEGF protein levels in the w ound tissue. This method provides another means for longitudinal monitoring of VEGF induction during wound healing and tumor pro gression. The transgenic mouse model developed in this study may also be useful in various other applications where noninvasive monitoring of VEGF gene expression is needed.

OUTLOOK A multitude of imaging techniques is available for assessment of tissue v asculature on a str uctural, functional, and molecular level. All these methods have been successfully used preclinically and will hopefully aid in antiangiogenic drug development in animal studies. Up to now, only imaging of functional hemodynamic parameters like Ktrans, blood flow, and blood volume is currently used in the clinical arena for e valuation of antiangiogenic and cytotoxic chemotherapies. However, results are often hard to interpret in their physiologic meaning. Macromolecular MRI and CT contrast agents will become clinically available and will probably facilitate the interpretation of these hemodynamic parameters. Concer ning imaging of molecular parameters of angiogenesis, only a few radiotracers have been used in humans up to no w, and their role in assessment of antiangiogenic therapies is still unsettled. Although αvβ3 is by f ar the most e xtensively studied angio genic factor for imaging, future trials still have to show which target structure is optimal for assessment of angio genic activity. Concer ning the optimum imaging technique, the radiotracer approach will probab ly be the f irst used on a wider scale in patients in the inter mediate ter m, due to its high sensitivity and low amounts of tracer that have to be used. Therefore, to xicity issues are of less importance compared with MRI or ultrasound imaging probes. In the long ter m, MRI might be a for midable alternative, due to its lack of ionizing radiation and high spatial resolution. However, it is likely that not one single parameter , tar get str ucture, or imaging technique will be used for assessment of angio genesis in the future, but rather a combination of parameters w hich will allow for evaluation of the angiogenic cascade in its full complexity (Figure 7). In summar y, assessment of the dif ferent aspects of angiogenesis at the functional and molecular le vels will hopefully become a reality in the not too distant future and

750

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 7. Multimodality assessment of angiogenesis and tumor biology in a patient with a non-small-cell lung cancer of the right hilum (arrow). In the DCE-MRI (inverted subtraction image, A) and in the ROI analysis of Gd-DTPA enhancement B, rapid tracer uptake with slow washout can be seen. The diffusion-weighted MRI (ADC map, C) shows restricted diffusion in the tumor. The CT image (D) clearly depicts the anatomy, whereas the [18F]FDG-PET/CT (E) demonstrates intense glucose metabolism and the [18F]Galacto-RGD PET (F) the level of αvβ3-expression.

will be implemented in therapy planning and response evaluation as par t of the concept of “personalized medicine.”

REFERENCES 1. Folkman J. Angiogenesis in cancer , v ascular, rheumatoid and other disease. Nat Med 1995;1:27–31. 2. Folkman J. Tumor angio genesis: therapeutic implications. N Engl J Med 1971;285:1182–6. 3. Ribatti D , Vacca A, Dammacco F . The role of the v ascular phase in solid tumor g rowth: a historical re view. Neoplasia 1999;1: 293–302. 4. Risau W. Mechanisms of angiogenesis. Nature 1997;386:671–4. 5. Bergers G, Ja vaherian K, Lo KM, et al. Ef fects of angio genesis inhibitors on multistage carcino genesis in mice. Science 1999;284:808–12. 6. Miller JC, Pien HH, Sahani D, et al. Imaging angiogenesis: applications and potential for dr ug de velopment. J Natl Cancer Inst 2005;97:172–87. 7. Kerbel RS. Antiangiogenic therap y: a uni versal chemosensitization strategy for cancer? Science 2006;312:1171–5. 8. Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leuco vorin for metastatic colorectal cancer. N Engl J Med 2004;350:2335–42. 9. Jaffe CC. Measures of response: RECIST, WHO, and new alternatives. J Clin Oncol 2006;24:3245–51.

10. Tortora G, Melisi D , Ciardiello F. Angiogenesis: a tar get for cancer therapy. Curr Pharm Des 2004;10:11–26. 11. Galbraith SM. Antivascular cancer treatments: imaging biomarkers in pharmaceutical drug development. Br J Radiol 2003;76:83–6. 12. Carmeliet P. Mechanisms of angio genesis and ar teriogenesis. Nat Med 2000;6:389–95. 13. Auguste P, Lemiere S, Larrieu-Lahargue F. Molecular mechanisms of tumor vascularization. Crit Rev Oncol Hematol 2005;54:53–61. 14. Holash J, Maisonpierre PC, Compton D, et al. Vessel cooption, regression, and growth in tumors mediated by angiopoietins and VEGF. Science 1999;284:1994–8. 15. Nguyen M. Angiogenic factors as tumor mark ers. Invest New Drugs 1997;15:29–37. 16. Veikkola T, Karkkainen M, Claesson-Welsh L, et al. Regulation of angiogenesis via v ascular endothelial g rowth f actor receptors. Cancer Res 2000;60:203–12. 17. Landgren E, Schiller P , Cao Y, Claesson-Welsh L. Placenta g rowth factor stimulates MAP kinase and mito genicity but not phospholipase C-gamma and migration of endothelial cells expressing Flt 1. Oncogene 1998;16:359–67. 18. D’Angelo G, Struman I, Martial J, Weiner RI. Activation of mitogenactivated protein kinases b y v ascular endothelial g rowth f actor and basic fibroblast growth factor in capillary endothelial cells is inhibited b y the antiangio genic f actor 16-kDa N-ter minal fragment of prolactin. Proc Natl Acad Sci U S A 1995;92:6374–8. 19. Nor JE, Christensen J , Mooney DJ, Polverini PJ. Vascular endothelial growth f actor (VEGF)-mediated angio genesis is associated with enhanced endothelial cell sur vival and induction of Bcl-2 e xpression. Am J Pathol 1999;154:375–84.

Tumor Vasculature

20. Paku S, P aweletz N . F irst steps of tumor -related angio genesis. Lab Invest 1991;65:334–46. 21. Pettersson A, Nagy J A, Brown L, et al. Hetero geneity of the angiogenic response induced in different normal adult tissues by vascular per meability f actor/vascular endothelial g rowth f actor. Lab Invest 2000;80:99–115. 22. Djonov V, Schmid M, Tschanz SA, Bur ri PH. Intussusceptive angiogenesis: its role in embr yonic v ascular network for mation. Circ Res 2000;86:286–92. 23. Patan S, Munn LL, Jain RK. Intussusceptive microvascular growth in a human colon adenocarcinoma xenograft: a novel mechanism of tumor angiogenesis. Microvasc Res 1996;51:260–72. 24. Metzger RJ, Krasnow MA. Genetic control of branching morphogenesis. Science 1999;284:1635–9. 25. Ausprunk DH, Folkman J. Migration and proliferation of endothelial cells in prefor med and newly formed blood vessels during tumor angiogenesis. Microvasc Res 1977;14:53–65. 26. Pepper MS, F errara N , Orci L, Montesano R. Vascular endothelial growth factor (VEGF) induces plasminogen activators and plasminogen acti vator inhibitor -1 in micro vascular endothelial cells. Biochem Biophys Res Commun 1991;181:902–6. 27. Denekamp J. Vascular attack as a therapeutic strategy for cancer. Cancer Metastasis Rev 1990;9:267–82. 28. Zetter BR. Migration of capillary endothelial cells is stimulated by tumour-derived factors. Nature 1980;285:41–59. 29. Asahara T, Chen D, Takahashi T, et al. Tie2 receptor ligands, angiopoietin-1 and angiopoietin-2, modulate VEGF-induced postnatal neovascularization. Circ Res 1998;83:233–40. 30. Maisonpierre PC, Suri C, Jones PF , et al. Angiopoietin-2, a natural antagonist for Tie2 that disr upts in vi vo angio genesis. Science 1997;277:55–60. 31. Friedlander M, Brooks PC, Shaffer RW, et al. Definition of two angiogenic pathways by distinct alpha v inte grins. Science 1995;270: 1500–2. 32. Brooks PC, Montgomer y AM, Rosenfeld M, et al. Inte grin alpha v beta 3 antagonists promote tumor re gression by inducing apoptosis of angiogenic blood vessels. Cell 1994;79:1157–64. 33. Nelson AR, Fingleton B, Rothenberg ML, Matrisian LM. Matrix metalloproteinases: biologic activity and clinical implications. J Clin Oncol 2000;18:1135–49. 34. Sang QX. Complex role of matrix metalloproteinases in angiogenesis. Cell Res 1998;8:171–7. 35. Nangia-Makker P , Honjo Y, Sar vis R, et al. Galectin-3 induces endothelial cell mor phogenesis and angio genesis. Am J P athol 2000;156:899–909. 36. Gamble J, Meyer G, Noack L, et al. B1 inte grin activation inhibits in vitro tube for mation: ef fects on cell mig ration, v acuole coalescence and lumen formation. Endothelium 1999;7:23–34. 37. Yang S, Graham J , Kahn JW , et al. Functional roles for PECAM-1 (CD31) and VEcadherin (CD144) in tube assemb ly and lumen formation in three-dimensional collagen gels. Am J P athol 1999;155:887–95. 38. Nakao N, Miura K, Takayasu Y, et al. CT angiography in hepatocellular carcinoma. J Comput Assist Tomogr 1983;7:780–7. 39. Littleton JT, Durizch ML, Moeller G, Herbert DE. Pulmonar y masses: contrast enhancement. Radiology 1990;177:861–71. 40. Miles KA, Ha yball MP, Dixon AK. Measurement of human pancreatic perfusion using dynamic computed tomo graphy with perfusion imaging. Br J Radiol 1995;68:471–5. 41. Sugano S, Okajima T, K ubo S, et al. Ev aluation of hepatic density change b y dynamic CT in health y humans and in patients with chronic liver diseases. Dig Dis Sci 1992;37:220–4. 42. Hiatt MD , Fleischmann D , Hellinger JC, Rubin GD . Angiographic imaging of the lower extremities with multidetector CT. Radiol Clin North Am 2005;43:1119–27, ix.

751

43. Grierson C, Uthappa MC, Uberoi R, Warakaulle D. Multidetector CT appearances of splanchnic ar terial patholo gy. Clin Radiol 2007;62:717–23. 44. Engelke C, Schaefer-Prokop C, Schirg E, et al. High-resolution CT and CT angio graphy of peripheral pulmonar y vascular disorders. Radiographics 2002;22:739–64. 45. Marxen M, Thornton MM, Chiarot CB , et al. MicroCT scanner performance and considerations for vascular specimen imaging. Med Phys 2004;31:305–13. 46. Bentley MD, Or tiz MC, Ritman EL, Romero JC. The use of microcomputed tomography to study microvasculature in small rodents. Am J Physiol Regul Integr Comp Physiol 2002;282:1267–79. 47. Jorgensen SM, Demirkaya O, Ritman EL. Three-dimensional imaging of vasculature and parenchyma in intact rodent organs with X-ray micro-CT. Am J Physiol 1998;275:1103–14. 48. Patterson BD, Abela R, Auderset H, et al. The materials science beamline at the Swiss Light Source: design and realization. Nucl Instrum Methods Phys Res A 2005;540:42–67. 49. Plouraboue F, Cloetens P, Fonta C, et al. X-ray high-resolution vascular network imaging. J Microsc 2004;215:139–48. 50. Heinzer S, Krucker T, Stampanoni M, et al. Hierarchical microimaging for multiscale anal ysis of lar ge vascular networks. Neuroimage 2006;32:626–36. 51. Kiessling F , Greschus S, Lich y MP , et al. Volumetric computed tomography (VCT): a new technology for noninvasive, high-resolution monitoring of tumor angiogenesis. Nat Med 2004;10: 1133–8. 52. Greschus S, Kiessling F, Lichy MP, et al. Potential applications of flatpanel volumetric CT in mor phologic and functional small animal imaging. Neoplasia 2005;7:730–40. 53. Wessels JT, Busse AC, Mahr t J, et al. In vi vo imaging in e xperimental preclinical tumor research—a re view. Cytometr y A 2007;71:542–9. 54. Kothari M, Guer mazi A, White D, et al. Imaging in antiangio genesis trial: a clinical trials radiolo gy perspecti ve. Br J Radiol 2003; 76:92–6. 55. Barrett T, Brechbiel M, Bernardo M, Choyke PL. MRI of tumor angiogenesis. J Magn Reson Imaging 2007;26:235–49. 56. Tofts PS, Brix G, Buckle y DL, et al. Estimating kinetic parameters from dynamic contrast-enhanced T(1)-weighted MRI of a dif fusable tracer: standardized quantities and symbols. J Magn Reson Imaging 1999;10:223–32. 57. Koh TS, Tan CK, Cheong LH, Lim CC. Cerebral perfusion mapping using a robust and efficient method for deconvolution analysis of dynamic contrast-enhanced images. Neuroimage 2006;32: 643–53. 58. Leach MO , Bo ggis CR, Dixon AK, et al; MARIBS study g roup. Screening with magnetic resonance imaging and mammo graphy of a UK population at high familial risk of breast cancer: a prospective multicentre cohor t study (MARIBS). Lancet 2005;365:1769–78. 59. Tofts PS. Modeling tracer kinetics in dynamic Gd-DTP A MR imaging. J Magn Reson Imaging 1997;7:91–101. 60. St La wrence KS, Lee TY. An adiabatic appro ximation to the tissue homogeneity model for w ater exchange in the brain: I. Theoretical derivation. J Cereb Blood Flow Metab 1998;18:1365–77. 61. Hylton N. Dynamic contrast-enhanced magnetic resonance imaging as an imaging biomarker. J Clin Oncol 2006;24:3293–8. 62. Dowlati A, Rober tson K, Coone y M, et al. A phase I phar macokinetic and translational study of the novel vascular targeting agent combretastatin a-4 phosphate on a single-dose intravenous schedule in patients with adv anced cancer . Cancer Res 2002;62: 3408–16. 63. Jayson GC, Zweit J, Jackson A, et al. Molecular imaging and biolo gical evaluation of HuMV833 anti-VEGF antibody: implications

752

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74. 75.

76.

77.

78.

79.

80.

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

for trial design of antiangio genic antibodies. J Natl Cancer Inst 2002;94:1484–93. Buonaccorsi GA, Rober ts C, Cheung S, et al. Comparison of the performance of tracer kinetic model-dri ven re gistration for dynamic contrast enhanced MRI using different models of contrast enhancement. Acad Radiol 2006;13:1112–23. Galbraith SM, Lodge MA, Taylor NJ , et al. Reproducibility of dynamic contrast-enhanced MRI in human muscle and tumours: comparison of quantitati ve and semi-quantitati ve analysis. NMR Biomed 2002;15:132–42. Padhani AR, Hayes C, Landau S, Leach MO. Reproducibility of quantitative dynamic MRI of nor mal human tissues. NMR Biomed 2002;15:143–53. Roberts C, Issa B, Stone A, et al. Comparative study into the robustness of compartmental modeling and model-free analysis in DCEMRI studies. J Magn Reson Imaging 2006;23:554–63. Padhani AR, Taylor NJ, d’Arcy JA, et al. Dynamic MRI evaluation of the triple receptor tyrosine kinase inhibitor BIBF 1120 in patients with adv anced solid tumours. Proc Int Soc Magn Reson Med 2006;14:765. Wedam SB, Low JA, Yang SX, et al. Antiangiogenic and antitumor effects of bevacizumab in patients with inflammator y and locally advanced breast cancer. J Clin Oncol 2006;24:769–77. O’Connor JP, Jackson A, Parker GJ, Jayson GC. DCE-MRI biomarkers in the clinical e valuation of antiangio genic and v ascular disrupting agents. Br J Cancer 2007;96:189–95. O’Dwyer PJ, Rosen M, Gallagher M, et al. Phar macodynamic study of BAY 43-9006 in patients with metastatic renal cell carcinoma. J Clin Oncol 2005;23:3005. Conrad C, Friedman H, Reardon D, et al. A phase I/II trial of singleagent PTK 787/ZK 222584 (PTK/ZK), a novel, oral angiogenesis inhibitor, in patients with recur rent gliob lastoma multifor me (GBM). J Clin Oncol 2004;22:1512. Morgan B, Thomas AL, Drevs J, et al. Dynamic contrast-enhanced magnetic resonance imaging as a biomark er for the phar macological response of PTK787/ZK 222584, an inhibitor of the vascular endothelial growth factor receptor tyrosine kinases, in patients with advanced colorectal cancer and liver metastases: results from tw o phase I studies. J Clin Oncol 2003;21: 3955–64. Padhani AR. MRI for assessing anti vascular cancer treatments. Br J Radiol 2003;76 Spec No 1:60–80. Thomsen HS, Marckmann P, Logager VB. Nephrogenic systemic fibrosis (NSF): a late adv erse reaction to some of the gadolinium based contrast agents. Cancer Imaging 2007;7:130–7. Schmiedl U, Brasch RC, Ogan MD , Moseley ME. Albumin labelled with Gd-DTP A. An intra vascular contrast-enhancing agent for magnetic resonance b lood pool and perfusion imaging. Acta Radiol Suppl 1990;374:99–102. Marzola P, Degrassi A, Calderan L, et al. Early antiangiogenic activity of SU11248 e valuated in vi vo by dynamic contrast-enhanced magnetic resonance imaging in an e xperimental model of colon carcinoma. Clin Cancer Res 2005;11:5827–32. Liu G, Rugo HS, Wilding G, et al. Dynamic contrast-enhanced magnetic resonance imaging as a phar macodynamic measure of response after acute dosing of AG-013736, an oral angio genesis inhibitor, in patients with advanced solid tumors: results from a phase I study. J Clin Oncol 2005;23:5464–73. Daldrup H, Shames D , Wendland M, et al. Cor relation of dynamic contrast-enhanced magnetic resonance imaging with histolo gic tumor grade: comparison of macromolecular and small molecular contrast media. AJR Am J Roentgenol 1998;171:941–9. Grist TM, Korosec FR, P eters DC, et al. Steady-state and dynamic MR angio graphy with MS-325: initial e xperience in humans. Radiology 1998;207:539–44.

81. Goyen M, Edelman M, P erreault P , et al. MR angiography of aortoiliac occlusi ve disease: a phase III study of the safety and effectiveness of the blood-pool contrast agent MS-325. Radiology 2005;236:825–33. 82. Weissleder R, Guillermo E, Wittenberg J, et al. Ultrasmall superparamagnetic iron o xide: an intra venous contrast agent for assessing lymph nodes with MRI. Radiology 1990;175:494–8. 83. Turetschek K, Huber S, Flo yd E, et al. MR imaging characterization of microvessels in experimental breast tumors by using a particulate contrast agent with histopathologic correlation. Radiology 2001;218:562–9. 84. van Dijk e CF, Brasch RC, Rober ts TP, et al. Mammar y carcinoma model: cor relation of macromolecular contrast-enhanced MR imaging characterizations of tumor micro vasculature and histologic capillary density. Radiology 1996;198:813–8. 85. Arbab AS, Aoki S, Toyama K, et al. Quantitati ve measurement of regional cerebral blood flow with flow sensitive alternating inversion recovery imaging: comparison with [iodine123]-iodoamphetamin single photon emission CT . AJNR Am J Neuroradiol 2002;23:381–8. 86. Neeman M, Dafni H, Bukhari O , et al. In vi vo BOLD contrast MRI mapping of subcutaneous vascular function and maturation: validation b y intra vital microscop y. Magn Reson Med 2001;45:887–98. 87. Baudelet C, Cron GO, Gallez B. Determination of the maturity and functionality of tumor vasculature by MRI: correlation between BOLD-MRI and DCE-MRI using P792 in e xperimental fibrosarcoma tumors. Magn Reson Med 2006. 88. Laking GR, Price PM. P ositron emission tomo graphic imaging of angiogenesis and vascular function. Br J Radiol 2003;76 Spec No 1:50–9. 89. Anderson RE. Cerebral blood flow xenon-133. Neurosurg Clin N Am 1996;7:703–8. 90. Lammertsma AA, Jones T. Low oxygen extraction fraction in tumours measured with the oxygen-15 steady state technique: effect of tissue heterogeneity. Br J Radiol 1992;65:697–700. 91. Blomqvist G, Lammer tsma AA, Mazoyer B, Wienhard K. Ef fect of tissue hetero geneity on quantif ication in positron emission tomography. Eur J Nucl Med 1995;22:652–63. 92. Wilson CB, Lammertsma AA, McKenzie CG, et al. Measurements of blood flo w and e xchanging w ater space in breast tumors using positron emission tomo graphy: a rapid and nonin vasive dynamic method. Cancer Res 1992;52:1592–7. 93. Anderson H, Price P. Clinical measurement of blood flow in tumours using positron emission tomo graphy: a re view. Nucl Med Commun 2002;23:131–8. 94. Tseng J , Dunnw ald LK, Schuber t EK, et al. 18F-FDG kinetics in locally advanced breast cancer: correlation with tumor blood flow and changes in response to neoadjuv ant chemotherap y. J Nucl Med 2004;45:1829–37. 95. Hoffend J, Mier W, Schuhmacher J, et al. Gallium-68-DOTA-albumin as a PET b lood pool mark er: e xperimental e valuation in vi vo. Nucl Med Biol 2005;32:287–92. 96. Bloch SH, Da yton PA, Ferrara KW. Targeted imaging using ultrasound contrast agents. Progess and opportunities for clinical and research applications. IEEE Eng Med Biol Mag 2004;23:18–29. 97. Wink MH, Wijkstra H, De La Rosette JJ, Grimbergen CA. Ultrasound imaging and contrast agents: a safe alter native to MRI? Minim Invasive Ther Allied Technol 2006;15:93–100. 98. Liang JD , Yang PM, Liang PC, et al. Three-dimensional po wer Doppler ultrasonography for demonstrating associated ar teries of hepatocellular carcinoma. J Formos Med Assoc 2003;102:367–74. 99. Drevs J, Hofmann I, Hugenschmidt H, et al. Ef fects of PTK787/ZK 222584, a specif ic inhibitor of vascular endothelial growth factor receptor tyrosine kinases, on primar y tumor , metastasis, v essel

Tumor Vasculature

100.

101.

102.

103.

104.

105.

106. 107.

108.

109.

110.

111.

112.

113.

114.

115. 116. 117. 118.

density, and b lood flow in a murine renal cell carcinoma model. Cancer Res 2000;60:4819–24. Foster FS, Bur ns PN, Simpson DH, et al. Ultrasound for the visualization and quantif ication of tumor microcirculation. Cancer Metastasis Rev 2000;19:131–8. Brown DL, Doubilet PM, Miller FH, et al. Benign and malignant ovarian masses: selection of the most discriminating g ray-scale and Doppler sonographic features. Radiology 1998;208:103–10. Niermann KJ, Fleischer AC, Huamani J, et al. Measuring tumor perfusion in control and treated murine tumors: cor relation of microbubble contrast-enhanced sonography to dynamic contrastenhanced magnetic resonance imaging and fluorodeo xyglucose positron emission tomo graphy. J Ultrasound Med 2007;26: 749–56. Huber S, Helbich T, Kettenbach J, et al. Effects of a microbubble contrast agent on breast tumors: computer-assisted quantitative assessment with color Doppler US—earl y e xperience. Radiolo gy 1998;208:485–9. Wilson SR, Burns PN, Muradali D, et al. Harmonic hepatic US with microbubble contrast agent: initial e xperience showing improved characterization of hemangioma, hepatocellular carcinoma, and metastasis. Radiology 2000;215:153–61. Burns PN, Wilson SR, Simpson DH. Pulse inversion imaging of liver blood flow: improved method for characterizing focal masses with microbubble contrast. Invest Radiol 2000;35:58–71. Quaia E. Microbubb le ultrasound contrast agents: an update. Eur Radiol 2007;17:1995–2008. Cornud F , Hamida K, Flam T, et al. Endorectal color doppler sonography and endorectal MR imaging features of nonpalpable prostate cancer: correlation with radical prostatectomy findings. AJR Am J Roentgenol 2000;175:1161–8. McCarville MB , Streck CJ , Dickson PV , et al. Angiogenesis inhibitors in a murine neurob lastoma model: quantitati ve assessment of intratumoral b lood flow with contrast-enhanced gray-scale US. Radiology 2006;240:73–81. Stieger SM, Bloch SH, F oreman O, et al. Ultrasound assessment of angiogenesis in a matrigel model in rats. Ultrasound Med Biol 2006;32:673–81. Weind KL, Maier CF , Rutt BK, Moussa M. In vasive carcinomas and fibroadenomas of the breast: comparison of microvessel distributions—implications for imaging modalities. Radiolo gy 1998;208:477–83. Lucidarme O, Franchi-Abella S, Cor reas JM, et al. Blood flo w quantification with contrast-enhanced US: “entrance in the section” phenomenon—phantom and rabbit study. Radiology 2003;228:473–9. Krix M, Kiessling F, Essig M, et al. Low mechanical index contrastenhanced ultrasound better reflects high arterial perfusion of liver metastases t han a rterial p hase c omputed t omography. I nvest Radiol 2004;39:216–22. Ferrara KW, Merritt CR, Burns PN, et al. Evaluation of tumor angiogenesis with US: imaging, Doppler , and contrast agents. Acad Radiol 2000;7:824–39. Massoud TF, Gambhir SS. Molecular imaging in living subjects: seeing fundamental biological processes in a new light. Genes Dev 2003;17:545–80. Weissleder R. Scaling do wn imaging: molecular mapping of cancer in mice. Nat Rev Cancer 2002;2:11–8. Jain RK, Munn LL, Fukumura D . Dissecting tumour pathoph ysiology using intravital microscopy. Nat Rev Cancer 2002;2:266–76. Zipfel WR, Williams RM, Webb WW. Nonlinear magic: multiphoton microscopy in the biosciences. Nat Biotechnol 2003;21:1369–77. Brown EB, Campbell RB, Tsuzuki Y, et al. In vivo measurement of gene e xpression, angio genesis and ph ysiological function in tumors using multiphoton laser scanning microscop y. Nat Med 2001;7:864–8.

753

119. Kunneke S, Janshoff A. Visualization of molecular recognition events on microstr uctured lipid-membrane compar tments b y in situ scanning force microscop y. Angew Chem Int Ed Engl 2002;41:314–6. 120. Weissleder R, Mahmood U . Molecular imaging. Radiolo gy 2001;219:316–33. 121. Li CY, Shan S, Huang Q, et al. Initial stages of tumor cell-induced angiogenesis: e valuation via skin windo w chambers in rodent models. J Natl Cancer Inst 2000;92:143–7. 122. Geng L, Donnell y E, McMahon G, et al. Inhibition of v ascular endothelial g rowth factor receptor signalling leads to re versal of tumor resistance to radiotherapy. Cancer Res 2001;61:2413–9. 123. Larson DR, Zipfel WR, Williams RM, et al. Water-soluble quantum dots for multiphoton fluorescence imaging in vi vo. Science 2003;300:1434–6. 124. Montet X, Ntziachristos V, Grimm J, Weissleder R. Tomographic fluorescence mapping of tumor targets. Cancer Res 2005;65:6330–6. 125. Montet X, F igueiredo JL, Alencar H, et al. Tomographic fluorescence imaging of tumor v ascular v olume in mice. Radiolo gy 2007;242:751–8. 126. Haubner R, Weber WA, Beer AJ, et al. Non-invasive visualization of the activated αvβ3 integrin in cancer patients by positron emission tomography and [ 18F]Galacto-RGD. PLoS Med 2005;2:e70. 127. Bach-Gansmo T, Danielsson R, Saracco A, et al. Inte grin receptor imaging of breast cancer: a proof-of-concept study to e valuate 99mTc-NC100692. J Nucl Med 2006;47:1434–9. 128. Ruoslahti E, Pierschbacher MD . New perspectives in cell adhesion: RGD and integrins. Science 1987;238:491–7. 129. Haubner R, Finsinger D, Kessler H. Stereoisomeric peptide libraries and peptidomimetics for designing selective inhibitors of the αvβ3 integrin for a new cancer therapy. Angew Chem Int Ed Engl 1997;36:1374–89. 130. Haubner R, Wester HJ, Reuning U, et al. Radiolabeled αvβ3 integrin antagonists: a new class of tracers for tumor targeting. J Nucl Med 1999;40:1061–71. 131. Haubner R, Wester HJ, Weber WA, et al. Nonin vasive imaging of αvβ3 integrin expression using 18F-labeled RGD-containing gl ycopeptide and positron emission tomo graphy. Cancer Res 2001;61:1781–5. 132. Haubner R. αvβ3-integrin imaging: a ne w approach to characterise angiogenesis? Eur J Nucl Med Mol Imaging 2006;33:54–63. 133. Harris JM, Mar tin NE, Modi M. P egylation: a no vel process for modifying pharmacokinetics. Clin Pharmacokinet 2001;40:539–51. 134. Chen X, Park R, Shahinian AH, et al. Pharmacokinetics and tumor retention of 125I-labeled RGD peptide are impro ved by PEGylation. Nucl Med Biol 2004;31:11–9. 135. van Hagen PM, Breeman WA, Ber nard HF, et al. Ev aluation of a radiolabelled c yclic DTPA-RGD analo gue for tumour imaging and radionuclide therapy. Int J Cancer 2000;90:186–98. 136. Haubner R, Wester HJ, Burkhar t F, et al. Gl ycosylated RGD-containing peptides: tracer for tumor tar geting and angio genesis imaging with improved biokinetics. J Nucl Med 2001;42:326–36. 137. Haubner R, Kuhnast B, Mang C, et al. [18F]Galacto-RGD: synthesis, radiolabeling, metabolic stability , and radiation dose estimates. Bioconjug Chem 2004;15:61–9. 138. Felding-Habermann B, Mueller BM, Romerdahl CA, Cheresh D A. Involvement of integrin αv gene expression in human melanoma tumorigenicity. J Clin Invest 1992;89:2018–22. 139. Pichler B, Braumueller H, Haubner R, et al. Monitoring of cellular immunotherapy in RIP1-Tag2 transgenic mice with radiolabelled RGD-peptides [abstract]. J Nucl Med 2002;43:122P. 140. Beer AJ, Haubner R, Goebel M, et al. Biodistribution and pharmacokinetics of the αvβ3 selective tracer 18F galacto-RGD in cancer patients. J Nucl Med 2005;46:1333–41.

754

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

141. Beer AJ, Haubner R, Wolf I, et al. PET-based human dosimetry of 18 F-galacto-RGD, a ne w radiotracer for imaging alpha v beta3 expression. J Nucl Med 2006;47:763–9. 142. Stangier I, Wester HJ, Schwaiger M, Beer AJ. Comparison of standardised uptak e v alues and distribution v olume for imaging of αvβ3 expression in breast cancer patients with [18F]galacto-RGD PET [abstract]. J Nucl Med 2007;48(S2):406. 143. Beer AJ, Haubner R, Sarbia M, et al. P ositron emission tomography using [ 18F]galacto-RGD identif ies the le vel of inte grin αvβ3 expresssion in man. Clin Cancer Res 2006;12:3942–9. 144. Beer AJ, Grosu AL, Carlsen J, et al. [18F]galacto-RGD PET for imaging of αvβ3 expression on neovasculature in patients with squamous cell carcinoma of the head and neck. Clin Cancer Res 2007;13:6610–6. 145. Beer AJ, Lorenzen S, Metz S, et al. Comparison of inte grin αvβ3 expression and glucose metabolism in primar y and metastatic lesions in cancer patients: a PET study using [ 18F]galacto-RGD and [18F]FDG. J Nucl Med 2008;49:22–9. 146. Boturyn D, Coll JL, Garanger E, et al.Template assembled cyclopeptides as multimeric system for integrin targeting and endocytosis. J Am Chem Soc 2004;126:5730–9. 147. Chen X, Tohme M, P ark R, et al. Micro-PET imaging of αvβ3-integrin expression with 18F-labeled dimeric RGD peptide. Mol Imaging 2004;3:96–104. 148. Zhang X, Xiong Z, Wu X, et al. Quantitative PET imaging of tumor integrin αvβ3 expression with 18F-FRGD2. J Nucl Med 2006;47:113–21. 149. Chen X, P ark R, Tohme M, et al. MicroPET and autoradio graphic imaging of breast cancer αv-integrin e xpression using 18F- and 64 Cu-labeled RGD peptide. Bioconjug Chem 2004;15:41–9. 150. Chen X, Hou Y, Tohme M, et al. P egylated Arg-Gly-Asp peptide: 64 Cu labeling and PET imaging of brain tumor αvβ3-integrin expression. J Nucl Med 2004;45:1776–83. 151. Chen X, Liu S, Hou Y, et al. MicroPET imaging of breast cancer αvintegrin expression with 64Cu-labeled dimeric RGD peptides. Mol Imaging Biol 2004;6:350–9. 152. Wu Y, Zhang X, Xiong Z, et al. MicroPET imaging of glioma αvintegrin expression using 64Cu-labeled tetrameric RGD peptide. J Nucl Med 2005;46:1707–18. 153. Li ZB, Cai W, Cao Q, et al. (64)Cu-labeled tetrameric and octameric RGD peptides for small-animal PET of tumor alpha(v)beta(3) integrin expression. J Nucl Med 2007;48:1162–71. 154. Thumshirn G, Hersel U, Goodman SL, Kessler H. Multimeric cyclic RGD peptides as potential tools for tumor tar geting: solid-phase peptide synthesis and chemoselecti ve oxime ligation. Chemistr y 2003;9:2717–25. 155. Poethko T, Schottelius M, Thumshirn G, et al. Two-step methodology for high-yield routine radiohalo genation of peptides: 18Flabeled RGD and octreotide analo gs. J Nucl Med 2004;45: 892–902. 156. Poethko T, Schottelius M, Thumshirn G, et al. Chemoselective pre-conjugate radiohalogenation of unprotected mono- and multimeric peptides via oxime formation. Radiochimica Acta 2004;92:317–27. 157. Cai W, Chen K, Li ZB, et al. Dual-function probe for PET and near infrared fluorescence imaging of tumor v asculature. J Nucl Med 2007;48:1862–70. 158. Balasubramanian K, Burghard M. Chemically functionalized carbon nanotubes. Small 2005;1:180–92. 159. Lacerda L, Bianco A, Prato M, K ostarelos K. Carbon nanotubes as nanomedicines: from to xicology to phar macology. Adv Dr ug Deliv Rev 2006;58:1460–70. 160. Liu Z, Cai W, He L, et al. In vivo biodistribution and highly efficient tumour targeting of carbon nanotubes in mice. Nat Nanotechnol 2007;2:47–52. 161. Hidalgo M, Eckhardt SG. De velopment of matrix metalloproteinase inhibitors in cancer therapy. J Natl Cancer Inst 2001;93:178–93.

162. Hanaoka H, Mukai T, Habashita S, et al. Chemical design of a radiolabeled gelatinase inhibitor peptide for the imaging of gelatinase activity in tumors. Nucl Med Biol 2007;34:503–10. 163. Furumoto S, Takashima K, K ubota K, et al. Tumor detection using 18F-labeled matrix metalloproteinase-2 inhibitor. Nucl Med Biol 2003;30:119–25. 164. Birkedal-Hansen H, Moore W, Bodden M, et al. Matrix metalloproteinases: a review. Crit Rev Oral Biol Med 1993;4:197. 165. Kulasegaram R, Giersing B , P age C, et al. In vi vo e valuation of 111In-DTPA-N-TIMP-2 in Kaposi sarcoma associated with HIV infection. Eur J Nucl Med 2001;28:756–61. 166. Neri D, Carnemolla B, Nissim A, et al. Targeting by affinity matured recombinant antibody fragments of an angio genesis associated fibronectin isoform. Nat Biotechnol 1997;15:1271–5. 167. Santimaria M, Moscatelli G, Viale GL, et al. Immunoscintigraphic detection of the ED-B domain of f ibronectin, a mark er of angiogenesis, in patients with cancer . Clin Cancer Res 2003;9: 571–9. 168. Dayton PA, Pearson D, Clark J, et al. Ultrasonic analysis of peptideand antibody-targeted microbubble contrast agents for molecular imaging of alpha vbeta3-expressing cells. Mol Imaging 2004;3: 125–34. 169. Cheung AM, Brown AS, Cucevic V, et al. Detecting vascular changes in tumour xenografts using micro-ultrasound and micro-ct following treatment with VEGFR-2 blocking antibodies. Ultrasound Med Biol 2007;33:1259–68. 170. Broumas AR, Pollard RE, Bloch SH, et al. Contrast-enhanced computed tomo graphy and ultrasound for the e valuation of tumor blood flow. Invest Radiol 2005;40:134–47. 171. Kennedy JE, ter Haar GR, Wu F, et al. Contrast-enhanced ultrasound assessment of tissue response to high-intensity focused ultrasound. Ultrasound Med Biol 2004;30:851–4. 172. Lu E, Wagner WR, Schellenberger U, et al. Targeted in vivo labeling of receptors for v ascular endothelial g rowth f actor: approach to identification of ischemic tissue. Circulation 2003;108:97–103. 173. Blankenberg FG, Mandl S, Cao YA, et al. Tumor imaging using a standardized radiolabeled adapter protein docked to vascular endothelial growth factor. J Nucl Med 2004;45:1373–80. 174. Backer MV, Levashova Z, Patel V, et al. Molecular imaging of VEGF receptors in angio genic v asculature with single-chain VEGFbased probes. Nat Med 2007;13:504–9. 175. Blankenberg FG, Back er MV, Le vashova Z, et al. In vi vo tumor angiogenesis imaging with site-specif ic labeled 99mTc-HYNICVEGF. Eur J Nucl Med Mol Imaging 2006;33:841–8. 176. Cornelissen B, Oltenfreiter R, Kersemans V, et al. In vitro and in vivo evaluation of [123I]-VEGF165 as a potential tumor mark er. Nucl Med Biol 2005;32:431–6. 177. Li S, P eck-Radosavljevic M, Kienast O , et al. Iodine-123-v ascular endothelial growth factor-165 (123I-VEGF165). Biodistribution, safety and radiation dosimetr y in patients with pancreatic carcinoma. Q J Nucl Med Mol Imaging 2004;48:198–206. 178. Li S, Peck-Radosavljevic M, Kienast O, et al. Imaging gastrointestinal tumours using v ascular endothelial g rowth f actor-165 (VEGF165) receptor scintigraphy. Ann Oncol 2003;14:1274–7. 179. Yoshimoto M, Kinuya S, Kawashima A, et al. Radioiodinated VEGF to image tumor angio genesis in a LS180 tumor x enograft model. Nucl Med Biol 2006;33:963–9. 180. Quinn TP, Peters KG, De Vries C, et al. Fetal liver kinase 1 is a receptor for v ascular endothelial g rowth f actor and is selecti vely expressed in v ascular endothelium. Proc Natl Acad Sci U S A 1993;90:7533–7. 181. Yamazaki Y, Morita T. Molecular and functional di versity of vascular endothelial growth factors. Mol Divers 2006;10:515–27. 182. Achilefu S, Bloch S, Markie wicz MA, et al. Syner gistic effects of light-emitting probes and peptides for tar geting and monitoring integrin expression. Proc Natl Acad Sci U S A 2005;102:7976–81.

Tumor Vasculature

183. Beyer T, Townsend DW, Brun T, et al. A combined PET/CT scanner for clinical oncology. J Nucl Med 2000;41:1369–79. 184. Collingridge DR, Car roll VA, Glaser M, et al. The development of [124I]iodinated-VG76e: a no vel tracer for imaging v ascular endothelial growth factor in vivo using positron emission tomography. Cancer Res 2002;62:5912–9. 185. Rehman S, Jayson GC. Molecular imaging of antiangiogenic agents. Oncologist 2005;10:92–103. 186. Cai W, Chen X. Anti-angiogenic cancer therap y based on inte grin alphavbeta3 antagonism. Anticancer Agents Med Chem 2006;6: 407–28. 187. Hsu AR, Cai W, Veeravagu A, et al. Multimodality molecular imaging of gliob lastoma g rowth inhibition with v asculature-targeting fusion toxin VEGF121/rGel. J Nucl Med 2007;48:445–54. 188. Nagengast WB, de Vries EG, Hospers GA, et al. In vi vo VEGF imaging with radiolabeled be vacizumab in a human o varian tumor xenograft. J Nucl Med 2007;48:1313–9. 189. Simon M, Rockl W, Hornig C, et al. Receptors of v ascular endothelial growth factor/vascular permeability factor (VEGF/VPF) in fetal and adult human kidney: localization and [125I]VEGF binding sites. J Am Soc Nephrol 1998;9:1032–44. 190. Keyt BA, Nguyen HV, Berleau L T, et al. Identif ication of v ascular endothelial growth factor determinants for binding KDR and FLT1 receptors. Generation of receptor-selective VEGF variants by site-directed mutagenesis. J Biol Chem 1996;271:5638–46. 191. Wang H, Cai W, Chen K, et al. A new PET tracer specific for vascular endothelial g rowth f actor receptor 2. Eur J Nucl Med Mol Imaging 2007. 192. Cai W, Chen K, He L, et al. Quantitative PET of EGFR expression in xenograft-bearing mice using 64Cu-labeled cetuximab, a chimeric anti-EGFR monoclonal antibody . Eur J Nucl Med Mol Imaging 2007;34:850–8. 193. Spuentrup E, Botnar RM. Coronar y magnetic resonance imaging: visualization of v essel lumen and the v essel wall and molecular imaging of arteriotrombosis. Eur Radiol 2006;16:1–14. 194. Jaffer FA, Weissleder R. Seeing within: molecular imaging of the cardiovascular system. Circ Res 2004;94:433–45. 195. Sipkins D A, Cheresh D A, Kazemi MR, et al. Detection of tumor angiogenesis in vivo by αvβ3-targeted magnetic resonance imaging. Nat Med 1998;4:623–6. 196. Winter PM, Caruthers SD, Kassner A, et al. Molecular imaging of angiogenesis in nascent Vx-2 rabbit tumors using a novel αvβ3targeted nanopar ticle and 1.5 tesla magnetic resonance imaging. Cancer Res 2003;63:5838–43. 197. Schmieder AH, Winter PM, Car uthers SD , et al. Molecular MR imaging of melanoma angiogenesis with αvβ3-targeted paramagnetic nanoparticles. Magn Reson Med 2005;53:621–7. 198. Thorek DL, Chen AK, Czupryna J, Tsourkas A. Superparamagnetic iron oxide nanopar ticle probes for molecular imaging. Ann Biomed Eng 2006;34:23–38. 199. Zhang C, Jugold M, Woenne EC, et al. Specif ic targeting of tumor angiogenesis by RGD-conjugated ultrasmall super paramagnetic iron o xide par ticles using a clinical 1.5-T magnetic resonance scanner. Cancer Res 2007;67:1555–62. 200. Bevilacqua MP, Pober JS, Mendrick DL, et al. Identif ication of an inducible endothelial-leuk ocyte adhesion molecule. Proc Natl Acad Sci U S A 1987;84:9238–42. 201. Kang HW, Josephson L, P etrovsky A, et al. Magnetic resonance imaging of inducible E-selectin expression in human endothelial cell culture. Bioconjug Chem 2002;13:122–7. 202. Kang HW, Torres D , Wald L, et al. Targeted imaging of human endothelial-specific marker in a model of adopti ve cell transfer . Lab Invest 2006;86:599–609. 203. Shohet RV, Chen S, Zhou YT, et al. Echocardiographic destruction of albumin microbubbles directs gene deli very to the m yocardium. Circulation 2000;101:2554–6.

755

204. Wierzbicka-Patynowski I, Nie wiarowski S, Marcinkie wicz C, et al. Structural requirements of echistatin for the reco gnition of alpha vbeta 3 and alpha 5beta 1 integrins. J Biol Chem 1999;274:37809–14. 205. Ellegala DB, Leong-Poi H, Carpenter JE, et al. Imaging tumor angiogenesis with contrast ultrasound and microb ubbles tar geted to alpha(v)beta3. Circulation 2003;108:336–41. 206. Tartis MS, McCallan J , Lum AF, et al. Therapeutic effects of paclitaxel-containing ultrasound contrast agents. Ultrasound Med Biol 2006;32:1771–80. 207. Forsberg F , Dick er AP, Thakur ML, et al. Comparing contrastenhanced ultrasound to immunohistochemical mark ers of angiogenesis in a human melanoma x enograft model: preliminar y results. Ultrasound Med Biol 2002;28:445–51. 208. Forsberg F, Ro RJ , Potoczek M, et al. Assessment of angio genesis: implications for ultrasound imaging. Ultrasonics 2004;42:325–30. 209. Taylor PC. Ser um vascular markers and vascular imaging in assessment of rheumatoid arthritis disease activity and response to therapy. Rheumatology (Oxford) 2005;44:721–8. 210. Korpanty G, Carbon JG, Grayburn PA, et al. Monitoring response to anticancer therap y by targeting microbubbles to tumor v asculature. Clin Cancer Res 2007;13:323–30. 211. Underiner TL, Ruggeri B, Gingrich DE. Development of vascular endothelial growth factor receptor (VEGFR) kinase inhibitors as anti-angiogenic agents in cancer therap y. Cur r Med Chem 2004;11:731–45. 212. Willmann JK, Paulmurugan R, Chen K, et al. Ultrasonic imaging of tumor angiogenesis with contrast microbubbles targeted to vascular endothelial growth factor type 2 receptor. Radiology 2008;246: 508–18. 213. Izzo AD, Mackanos MA, Beckham JT , Jansen ED . In vi vo optical imaging of e xpression of v ascular endothelial g rowth factor following laser incision in skin. Lasers Sur g Med 2001;29:343–50. 214. Lin PC. Optical imaging and tumor angio genesis. J Cell Biochem 2003;90:484–91. 215. Schirner M, Menrad A, Stephens A, et al. Molecular imaging of tumor angiogenesis. Ann N Y Acad Sci 2004;1014:67–75. 216. Frangioni JV. In vivo near-infrared fluorescence imaging. Curr Opin Chem Biol 2003;7:626–34. 217. Chen X, Conti PS, Moats RA. In vi vo near -infrared fluorescence imaging of inte grin alphavbeta3 in brain tumor x enografts. Cancer Res 2004;64:8009–14. 218. Cai W, Shin DW, Chen K, et al. P eptide-labeled near-infrared quantum dots for imaging tumor v asculature in li ving subjects. Nano Lett 2006;6:669–76. 219. Bloch S, Xu B , Ye Y, et al. Targeting Beta-3 inte grin using a linear hexapeptide labeled with a near -infrared fluorescent molecular probe. Mol Pharm 2006;3:539–49. 220. Ye Y, Bloch S, Xu B, Achilefu S. Design, synthesis, and evaluation of near infrared fluorescent multimeric RGD peptides for targeting tumors. J Med Chem 2006;49:2268–75. 221. Hsu AR, Hou LC, Veeravagu A, et al. In vi vo near-infrared fluorescence imaging of inte grin alphavbeta3 in an or thotopic glioblastoma model. Mol Imaging Biol 2006;8:315–23. 222. Backer MV, Patel V, Jehning BT, Backer JM. Self-assemb led “dock and lock” system for linking pa yloads to targeting proteins. Bioconjug Chem 2006;17:912–9. 223. Backer MV, Ga ynutdinov TI, P atel V, et al. Vascular endothelial growth f actor selectively tar gets boronated dendrimers to tumor vasculature. Mol Cancer Ther 2005;4:1423–9. 224. Negrin RS, Contag CH. In vi vo imaging using bioluminescence: a tool for probing g raft-versus-host disease. Nat Re v Immunol 2006;6:484–90. 225. Wang Y, Iyer M, Annala A, et al. Noninvasive indirect imaging of vascular endothelial g rowth factor gene e xpression using bioluminescence imaging in living transgenic mice. Physiol Genomics 2006;24:173–80

46 IMAGING HYPOXIA ASHLEY A. MANZOOR, BS, HONG YUAN, PHD, GREGORY M. PALMER, PHD, BENJAMIN L. VIGLIANTI, PHD, AND MARK W. DEWHIRST, DVM, PHD

Treatment of hypoxic tumors cur rently represents one of the major challenges in oncolo gy. F irst suspected in human tumors by Thomlinson and Gray in 1955, the presence of hypoxia has since become an increasingly important f actor in the treatment of cancer .1 Hypoxic tumors are more dif ficult to treat as the y are more resistant to both radiotherap y and man y chemotherapeutic dr ugs. Initially, predominant theories re garding the impor tance of oxygen with radiation therapy centered around the relative increase in unrepairab le DN A damage after treatment when oxygen was present, a phenomenon refer red to as “f ixation of radiation damage. ”2 With chemotherapy, it was thought that h ypoxic cells would be less proliferative and therefore more resistant to cell c ycle specific dr ugs. In addition, such cells ma y e xperience lower concentrations of dr ug as a result of being f arther from the tumor v asculature.3,4 However, in the past 5 to 10 y ears, molecular consequences of h ypoxia have also been shown to influence treatment resistance and may be more important than an y other f actor. Hypoxia has been shown to stimulate angio genesis and increase agg ressive traits, leading to the de velopment of a more malignant phenotype. Accordingly, a number of approaches are being used to target hypoxic tumor cells, including methods to increase deli very of o xygen, reduce o xygen consumption rates, or selecti vely tar get the h ypoxic gene response.5,6 Consequently, the ability to image h ypoxic areas of the tumor and the changes in tumor o xygenation over the course of treatment has become paramount. Clinically, hypoxia has been observed in a wide variety of human tumors, including gliomas, adenocarcinomas of the breast and pancreas, sarcomas, and squamous cell carcinomas.7 Oxygen probes, implanted directly into tumors to measure o xygen content, ha ve shown that the oxygenation status v aries across tumor types and within

756

the same tumor , with lar ge hetero geneity across e ven short tumor distances.8,9 With relation to the patient, hypoxia has been sho wn to contribute to poor pro gnosis in head and neck cancers, cer vical cancers, and earl ystage nonsmall cell lung cancers (NSCLC). 10–12 In these studies, Eppendorf oxygen probe measurements of partial pressure of oxygen (pO2) < 10 mm Hg cor relate with poor local tumor control, disease-free sur vival, and overall sur vival. Thus, h ypoxia imaging has become important in a number of scientif ic arenas; in the laboratory, hypoxia imaging aids in better understanding of the tumor en vironment and its af fects on cellular signaling and protein expression, while clinically hypoxia imaging can be used as a prognostic indicator for patient outcome, as a basis for deter mining treatment options, and as an indication of tumor response to v arious therapies. It has even been suggested that hypoxia mapping could be integrated with confor mal radiotherap y techniques to improve both target delineation and dose delivery. While initially most oxygen measurements were performed with oxygen probes, these devices permit only a single-site or single-line insight into tumor o xygenation, as w ell as requiring in vasive measurements. Hypoxia imaging attempts to provide more spatial information of the often hetero genous o xygen distribution over a re gion of interest. Better infor mation of the hypoxia status can lead to a better understanding of the underlying pathophysiology, providing numerous advantages for both research and clinical treatments. As there are a v ariety and wide e xtent of possib le uses for hypoxia imaging, one ideal imaging de vice may not exist; the requirements and interests for imaging the patient and those for in vi vo or in vitro imaging in the laboratory are widely different. For the patient, the ideal device for measuring h ypoxia should be ab le to perfor m

Imaging Hypoxia

repeated measurements of direct tissue pO 2 over the course of treatment noninvasively, with high sensitivity to oxygen heterogeneity. In the laborator y, the use of small animal imaging has pro vided a challenge as tumors are grown to a much smaller size than most clinical tumors, and hence, much higher resolution is often needed. On the other hand , these applications are not as limited b y clinical considerations such as to xicity. Table 1 provides a list of the h ypoxia imaging de vices discussed in this chapter and w hether the y are used clinicall y or preclinically. With the consideration of the abo ve applicationbased dif ferences, there are general qualities that are valuable to a hypoxia imaging device. These qualities fall into two categories: qualities that help define the hypoxic status of the tumor and qualities pertaining to the device’s applicability in the clinic. With re gards to the ph ysical measurement of o xygenation, a h ypoxia imaging de vice should ideally be able to (1) distinguish between hypoxia, anoxia, and necrosis; (2) identify cellular pO2 values as opposed to vascular pO2 values; and (3) be sensitive to pO2 values in clinicall y relevant levels (from 0–15 mm Hg). Other aspects of an ideal imaging modality that are equally important and relate more directly to its applicability in research and the clinical en vironment include (1) good spatial resolution on a scale similar to or less than the diffusion distance of O 2 (70–100 µm); (2) shor t acquisition time for good temporal resolution; (3) ability for repeatab le measurements; (4) nonin vasiveness; (5) applicability to an y tumor site (not depth-limited) with a wide spatial windo w for measurement; (6) to be simple to perform and nontoxic to the patient. To meet these man y qualities required of an ideal hypoxia imaging device, a multitude of approaches ha ve been de veloped. Including such di verse principles as magnetic r esonance ( MR), p hosphorescence, o ptical techniques, and fluorescence quenching, these methods vary not only in their use of different principles but also their approaches to ward imaging o xygenation. Some exploit the interactions of chemical compounds under oxygenated environments as in positron emission tomography (PET), while others such as magnetic resonance imaging (MRI) use magnetic influence on elements found in the body . The resulting form of measurement can thus vary greatly, from absolute measurements of pO2 or o xygen concentration to indirect measurements of a related parameter, such as hemo globin saturation. In this chapter, imaging methods are often refer red to via their ability to directly or indirectly quantify hypoxia. By this terminology, a direct imaging method is tak en to be one

757

Table 1. COMMONLY USED METHODS FOR EVALUATING HYPOXIA IN TUMORS AND WHETHER THEY ARE USED CLINICALLY Preclinical/Experimental

Clinical

TISSUE/MOLECULAR MARKERS Immunohistochemistry (EF5, pimonidazole, CAIX, HIF) Reporter constructs (fluorescence, luciferase) OPTICAL METHODS Oxygen-sensing electrodes (Oxylite, Eppendorf) Optical spectroscopy—diffuse reflectance Optical spectroscopy—diffuse optical topography and tomography Redox imaging Hyperspectral imaging Phosphorescence lifetime NUCLEAR MEDICINE PET imaging (FMISO, IAZA, FAZA, FETNIM, Cu-ATSM) SPECT imaging (IAZA, 99mTc-HL91) IAZG F-EF5/EF3 (clinical trials)

18

MAGNETIC RESONANCE Blood oxygen level dependent (BOLD) MRI Dynamic contrast-enhanced (DCE) MRI Magnetic resonance spectroscopy 31P Magnetic resonance spectroscopy 19F (clinical trials) Electron paramagnetic resonance imaging (EPRI) Proton-electron double resonance imaging (PEDRI) Methods spanning both columns are used both clinically and preclinically.

that can be calibrated to a pO 2 value, and an indirect method is one that can provide only correlation to a range of pO2 values or to relative measures of oxygenation. The differences in how hypoxia is measured also influence the terminology by which hypoxia is repor ted. The standard direct measurement is pO 2, repor ted in mm Hg or Torr, but other ter minology include percent o xygen (%O 2), oxygen concentration (µM), and oxygen saturation of hemoglobin. As these methods differ in their approach to measuring hypoxia, the y also dif fer in w hich of the abo ve desired imaging de vice qualities the y can achie ve. Table 2 compares some of the main h ypoxia imaging modalities that will be discussed further in this chapter. In general, hypoxia imaging devices f all into four broad cate gories: (1) tissue and molecular mark ers; (2) optical methods: optical spectroscopy, redo x imaging, and phosphorescence lifetime; (3) nuclear medicine: PET and single-photon emission

758

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

computed tomography (SPECT); (4) magnetic resonance: blood oxygen level-dependent (BOLD) MRI, 19F or 31P MR spectroscopy, dynamic contrast-enhanced MRI (DCEMRI), electron paramagnetic imaging (EPRI), and protonelectron double resonance imaging (PEDRI). It is important to note that while these imaging methods will be described in the context of tumor hypoxia, they are equally applicable to other tissues, and ha ve been used to assess h ypoxia in renal, brain, and cardiovascular sites.

BIOLOGY OF HYPOXIA To evaluate the ability of an imaging modality to assess the hypoxic status of a tumor , it is impor tant to understand the underlying biologic and physiologic causes of hypoxia. The ter m h ypoxia refers to par tial o xygen pressures below 10 mm Hg. This value is typically used as the h ypoxic threshold since cells e xposed to pO 2 below 10 mm Hg e xhibit increased radioresistance and changes in molecular e xpression.2,13 Additionally, tumor perfusion ma y change belo w this v alue, as the result of changes in red cell fluidity and increased blood viscosity.14 Hypoxia typically develops in solid tumors as a result of an imbalance between oxygen delivery by the vasculature and o xygen consumption by the tissue. 15 This interplay between oxygen delivery and consumption is present in all cells. Ho wever, in nor mal tissue these imbalances are cor rected for b y tight molecular re gulation of v essel tone, v ascular remodeling, and alterations in o xygen consumption.16 In tumor tissue, inadequacies in molecular signaling pathw ays prevent efficient response to changing o xygen concentrations. In addition, ill-formed tumor vasculature and low vascular density in expanding tumor areas often f ail to suppl y adequate oxygen to meet the metabolic demands of proliferating cancer cells. The result is that tumors often have large regions of hypoxia which vary spatially and temporally as the tumor attempts to adapt to cur rent oxygen levels. The o xygen concentration present in tissue is influenced b y tw o types of g radients15: (1) radial g radients, the result of O 2 diffusion limitations (appro ximately 70–100 µm), with highest pO 2 values closest to the blood vessels and decreasing radiall y a way from the v essel; (2) longitudinal g radients, arising from depletion of molecular oxygen from hemo globin as it tra verses from the arterial (afferent) input to the venous (efferent) egress. However, it is impor tant to note that these g radients are neither mutually exclusive nor independent of each other. For instance, as v essel O 2 concentration decreases, the radial diffusion distance of oxygen also decreases.

Since the o xygen concentration g radients relate directly to the v asculature, h ypoxia is influenced to a great extent by the or ganization of b lood vessels. While hypoxia is a voided in nor mal tissue b y or ganized and well-spaced blood vessels, the v ascular network present in tumors is often ill-formed, with abnormal structure and function.3,17 Tumor v asculature often suf fers from an uneven v essel distrib ution and str uctural abnor malities such as incongruous branching and tortuous vessel paths. These characteristics contribute to limited dif fusion distances and gi ve rise to the common conception of hypoxic areas being located f ar from v essels. However, changes in red blood cell flux and oxygen saturation also contribute to the availability of molecular oxygen and can greatly affect both the longitudinal and radial gradients.18 Vessels similar in size ma y exhibit different blood flow rates, with some str ucturally intact vessels exhibiting little or no throughput of red b lood cells. Tumors also have fewer arterial inputs and higher interstitial pressures than normal tissues, with steeper afferent longitudinal gradients that can rapidly deplete the oxygen concentration along a vessel. These factors can create such limited longitudinal and radial gradients that areas located immediately adjacent to blood vessels can also be hypoxic.15 Tumor vasculature is under a constant state of remodeling, causing oxygen gradient shifts that vary on a daily basis. Red cell flux es in micro vessels also fluctuate but 18 on a shor ter time scale of tens of minutes to hours. Since red blood cells contain > 95% of all o xygen in the blood, the fluctuations in red cell flux strongl y influence how much o xygen is deli vered to the tissue. F igure 1 illustrates ho w red b lood cell flux can influence local oxygen tension in tissues over time. Common conception typicall y identif ies h ypoxia as consisting of either an acute component caused b y interruptions in tumor microvascular flow or a chronic component resulting from f actors that influence the radial diffusion distance of o xygen; however, these conceptions do not reflect the true complexity or dynamic nature of the hypoxic state. In actuality , o xygen g radients in tumors have simultaneous spatial and temporal fluctuations. The size and severity of the h ypoxic subregions vary with the average magnitude of oxygen delivered and the instability in red cell flux of sur rounding microvessels. The complexity of pO 2 fluctuations and their ef fect on the h ypoxic subre gions can perhaps best be demonstrated by analogy to a small atoll in the ocean that is subjected to tidal v ariations as sho wn in F igure 2. In this analogy, the water represents the oxygen field within a tumor microregion. The size and height of the atoll that is observed above the water line is influenced by both the

Tissue

h

Markerdependent; 0–10 mm Hg for nitroimidazoles

Moderate

Vascular/ tissue pO2

Temporal resolution

Range

Sensitivity

Not applicable

Not applicable

min–h

Tissue

cm

Not applicable

Sensing depth

< 1 mm Hg

Probe dependent: Eppendorf less sensitive at low pO2; Oxylite highest at low pO2

s

Both; indistinguishable

Several cm

Invasive

Invasive

Invasive— requires biopsy and sectioning

Direct

Invasive/ noninvasive

Indirect

None

Fluorescence Eppendorf intensity of ionization accumulated of O2; proteins Oxylite: lifetime decay of fluorescent dye

Indirect

Direct/ Indirect

Exogenous reporter constructs

Reporter Constructs

Basis of Staining oxygen intensity measurement indicating amount of marker (ie, EF5); pimonidazole: % area stained

Extrinsic or endogenous markers

Chemical probe

Immunohistochemistry

Oxygen Sensing Probes

s

Vascular

mm–cm

Non-invasive

Optical absorbance of hemoglobin

Indirect

None

Optical Spectroscopy

< 1 mm Hg

High sensitivity to hemoglobin

Probe 0–100% Hb dependent saturation

s

Both

mm–cm

Minimally invasive

Decay lifetime of phosphor upon interaction with O2

Direct

Phosphors (external)

Phosphorescence Lifetime Imaging

Dependent on voxel spatial distribution of hypoxia

Markerdependent; 0–10 mm Hg 18F-EF5 and FMISO

min–h

Tissue

Whole body

Minimally invasive

Accumulated positron signal of retained drug

Indirect

Compound labeled with positron emitting isotope

PET

Dependent on voxel spatial distribution of hypoxia

0–10 mm Hg

min–h

Tissue

Whole body

Minimally invasive

Accumulated photon signal of retained drug

Indirect

Compound labeled with photon emitting isotope

SPECT

Table 2. COMPARISON OF HYPOXIA IMAGING MODALITIES

Not applicable

Relative pO2 based on reference image

s

Vascular; limited tissue

Whole body

Non-invasive

Magnetic state of deoxy-Hb in red blood cells

Indirect

None

BOLD MRI

Not applicable

Very weak correlation

s

Both

Whole body

Minimally invasive

Vascular perfusion/ permeability

Indirect

Contrast agent

DCE-MRI

F MRS

1–3 mm Hg

Large range; perfluorocarbon dependent

s–min

Tissue

Whole body

Minimally invasive

Change in resonance of probe upon interaction with O2

Direct

Perfluorocarbons

19

< 1 mm Hg

0–100 mm Hg

min–h

Tissue

mm–cm

Minimally invasive

Change in line width of probe upon interaction with O2

Direct

Paramagnetic species

EPRI

Possible with matched hypoxia markers

1

5

Monitors changes in hypoxia

Cost (1–5; low-high)

Availability (1–5; poor-wide)

2

3

Timing dependent on reporter half-life

Depends on device used for measurement

Reporter Constructs

4

2

Continuous

Point measurements

2

3

min–min

Sub mm–cm

2

1

Real time

mm–cm

Optical Spectroscopy

1

5

>1d

Several mm

PET

2

4

>1d

mm–cm

SPECT

s–s

3

5

Machine dependent; sub mm–mm

BOLD MRI

h–h

4

5

Sub mm

DCE-MRI

F MRS

1

5

Fairly continuous

mm–cm

19

1

3 (due to lower field strength)

min–min

mm

EPRI

BOLD MRI = blood oxygen level-dependent magnetic resonance imaging; DCE-MRI = dynamic contrast-enhanced magnetic resonance imaging; EPRI = electron paramagnetic imaging; MRS = magnetic resonance spectroscopy; PET = positron emission tomography; SPECT = single-photon emission computed tomography.

< Micrometer

Spatial resolution

Immunohistochemistry

Oxygen Sensing Probes

Phosphorescence Lifetime Imaging

Imaging Hypoxia

761

time

A B A D

C

50 mm Hg

B A D

C

B

D

25 mm Hg

C

0 mm Hg

pO2

pO2

Figure 1. Red blood cell flux influences oxygen tension in surrounding areas, leading to fluctuations in hypoxia. Areas (A–D) above demonstrate how tissue pO2 can fluctuate over time as a result of changing red cell flux. In region (A), the red cell flux constantly decreases over time, leading to a progressive decline in tissue oxygen levels. In regions (B) and (C), the surrounding red cell flux decreases and then increases, but only in some of the surrounding vasculature. This results in hypoxic gradients within regions (B) and (C). Region (D) shows a constantly hypoxic center, with a slightly fluctuating oxygen gradient as the red cell flux changes. Adapted from Cardenas-Navia LI et al.15

A A

B B

C C

D D

Figure 2. Atoll analogy of how hypoxia is influenced by changes in average pO2 and red cell flux. The size and height of the atoll is representative of hypoxic spatial extent and intensity, influenced by tidal (average pO2) and wave (red cell flux) oscillations. From (A)–(B), the average pO2 increases, decreasing both the hypoxic spatial extent and intensity. Similar to tidal effects, this is the predominant mechanism for altering tissue pO2. In (A) and (B), the wave size also oscillates, representative of red cell fluctuations. However, this occurs on a shorter time scale and has less overall influence on the extent of hypoxia. From (A)–(C) and (B)–(D), the average pO2 (tidal height) does not change, but greater changes in red cell flux (wave oscillations) have a more drastic effect on the extent of hypoxia, dependent on the wave amplitude size relative to the overall tidal height. In a tumor, the overall oxygenation of a tissue region may vary from day to day, based on the efficiencies of transport to the tumor on that day (the analogy here is the tide). Superimposed on this are the cycles of red cell flux that vary on a time scale of tens of minutes. This would be analogous to the waves.

762

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

tides and the amplitude of the w aves. The tides fluctuate on the order of twice daily, whereas the waves show much higher frequenc y of fluctuation. When the tide is high, small waves can strongl y influence the size of the atoll, whereas at lo w tide, the w aves have less of an ef fect on the size of the atoll, unless they are very large. Even large waves will only have a transient ef fect on the size of the atoll, relative to the effect of the tides. The shift in tidal height from (a) to (b) is analo gous to fluctuations of the a verage pO 2 that occur in a tumor area. The overall average pO 2 is influenced primaril y by the f actors listed abo ve and represents the main characteristic of the oxygen field, created by the overall balance between delivery and o xygen consumption. As the a verage pO2 in a vascular network increases, variations in red cell flux influence the spatial extent of hypoxia. However, this does not change the underl ying shape of the o xygen gradient, only its magnitude. In the atoll analogy, the size of the atoll is briefly influenced by the wave size, but the main influence remains the o verall effect of the tide. As Figure 2 demonstrates, hypoxia is not a static state, but a more dynamic state that shows spatial and intensity variation over multiple temporal fluctuations. In addition to the complexity of the physical effects that cause and contribute to h ypoxia, there are also cellular signaling and proteomic ef fects of h ypoxia that should be considered. Particular attention has been spent on h ypoxia inducib le f actor-1α (HIF-1α), a heterodimeric transcription f actor that has been associated with upregulation under hypoxic conditions. HIF-1 consists of two subunits: a constituti vely expressed HIF-1β subunit and the HIF-1 α subunit, w hich is tightl y regulated. HIF-1α expression is controlled through oxygen-independent synthesis and oxygen-dependent degradation.19 In the presence of o xygen, HIF-1α is targeted for degradation by a family of prolyl hydroxylases, which h ydroxylate the prol yl residues on the o xygendependent degradation domain (ODD) of HIF-1α. This activates HIF-1 α to bind to the v on Hippel-Lindau (VHL) E3 ubiquitin ligase complex, allowing HIF-1α to be degraded.20 Therefore, in high pO2, HIF-1α levels are kept low. However, when molecular oxygen is not readily available, HIF-1α accumulates and translocates to the nucleus, along with HIF-1 β. In the nucleus, these subunits dimerize and bind to gene tar gets ter med hypoxia responsive elements (HREs) to re gulate transcription. The dimerized HIF-1 transcription f actor thereb y controls the e xpression for a host of do wnstream genes, which to gether help the tumor cell adapt to h ypoxic conditions through enhanced gl ycolysis, promotion of cell sur vival, inhibition of apoptosis, inhibition of cell

differentiation, and angio genesis.7,21 This consortium of molecular events leads to an overall increased aggressive behavior of h ypoxic cells and may result in clonal expression of agg ressive traits from h ypoxic selection pressures. Due to its impor tance in both the h ypoxic environment and for increased tumor agg ressiveness, HIF-1 has become the tar get for man y emer ging anticancer drugs.22

TISSUE AND MOLECULAR MARKERS OF HYPOXIA While this chapter focuses on imaging methodology, it is also impor tant to understand some of the immunohistochemical methods for measuring hypoxia; these are often the standards with w hich the images are compared to determine cor relation of the imaging modality with hypoxic re gions. Immunohistoche mical staining for extrinsic mark ers is used e xtensively for h ypoxic correlation, par ticularly with administered 2-nitroimidazoles.23,24 Nitroimidazoles are reduced in the cell, for ming reacti ve species that can then either for m co valent bonds with cellular macromolecules or become reo xidized b y electron transfer to o xygen.25 When a vailable oxygen is low, the predominant reaction is the for mation of co valent bonds, thus trapping the compounds inside the cell. Subsequent biopsy , sectioning, and antibody treatment result in fluorescence labeling of the cellularl y trapped nitroimidazole. Images of stained slides can provide spatial identif ication of h ypoxic re gions of tissue and can also be combined with other tissue markers, such as CD31 (vascular marker) and Hoechst 33324 (perfusion marker). The most commonl y used nitroimidazoles for hypoxia staining are pimonidazole and EF5. Pimonidazole-stained slides are e valuated based on the percent area of tissue stained. F or EF5, however, the intensity of the stain can also be calibrated to pro vide an approximation of pO 2.25 Other dif ferences betw een these tw o nitroimidazoles are dependence on cell-line, dr ug exposure, and antibody concentration. EF5 binding is cell-line and e xposure dependent for a range betw een 0 to 100 mmHg, yet not dependent on antibody concentration. Pimonidazole, on the other hand, is generally considered to be independent of cell-line and e xposure time, with markedly increased uptak e below 10 mmHg, but dependent on antibody concentration. 178 However, it should be noted that these assumptions of dependence are still being debated. In addition to the nitroimidazoles, protein-based markers are also used to identify possib le hypoxic areas,

Imaging Hypoxia

such as carbonic anhydrase IX (CA-IX), HIF-1, and to a lesser extent, GLUT-1. While these endogenous proteins do not directl y interact with o xygen like the nitroimidazoles, as staining techniques they help to identify expression of impor tant hypoxia related proteins. Ho wever, the majority of research in this area has focused on cor relating these endo genous proteins with patient outcome; correlation betw een these mark ers and direct pO 2 measurements has been some what contro versial, if not more trending towards lack of cor relation.179,180 The main limitations to both these techniques are their in vasive nature that requires biopsy or tumor removal and their limited sample size. While the necessity for tumor removal limits identification of changes in hypoxia, Benne with and colleagues 26 have recentl y reported on matching hypoxia markers to designate areas of recent h ypoxic activity and those of longer time-integrated h ypoxia. This method consists of oral doses of pimonidazole given ad libitum in the drinking w ater of mice for a prolonged time period of 3 to 96 hours, with addition of an injected h ypoxia mark er, such as CCI103F, gi ven a fe w hours before e xcision. This method provides comparison of longer term and transient areas of hypoxia as labeled with pimonidazole, with the acute hypoxia experienced shortly before removal. This matching hypoxia marker technique has also been repeated b y other groups, with pimonidazole and EF5. 181 Another approach used preclinicall y that pro vides a larger spatial view and better temporal resolution is in vivo molecular imaging using exogenous “reporter constructs” or “reporter genes.” These reporter genes are often labeled with a reporter protein that can be imaged optically. Examples include g reen fluorescent protein (GFP), red fluorescent protein, and the f amily of luciferase enzymes. 4 Luciferase reacts with the substrate luciferin through an oxygen-dependent reaction generating a photon that can be detected externally and subsequently localized and quantified to reveal the amount of expressed luciferase present in the experimental subject.27 As used in hypoxia detection, the reporter genes are created to contain HREs and thus can aid in obser ving in vivo HIF-1 expression. HIF-1 expression leads to increased transcription of the reporter gene and hence increased fluorescence intensity which can be quantified for the whole body of a small animal. Reporter gene constructs have been compared with fluoromisonidazole (FMISO) PET uptak e in nude mice, as well as used as a method of hypoxia monitoring follo wing h ypoxia-targeted chemotherap y treatment.28,29 Other approaches to measure HIF-1 le vels have also been used, including the creation of reporter proteins with the proline-containing ODD .30 This mechanism

763

exploits HIF-1 o xygen-dependent de gradation, leading to posttranslational regulation of the reporter by oxygen. While often used in research as sur rogate markers for tissue h ypoxia, a main disadv antage to these approaches is that the repor ter proteins can be highl y sensitive to cellular o xygen concentrations. As mentioned previously, luciferase reacts in an oxygen-dependent manner with luciferin; in the absence of o xygen, luciferase has been shown to have marked signal loss.31 GFP also displays a similar requirement for an o xygen cofactor to generate an imaging signal. Additionally, care must be taken to understand how HRE or ODDs might be influenced b y changes in HIF-1 not link ed to hypoxia; w hile there are man y associations betw een HIF-1α and h ypoxic conditions, its e xpression is not always cor related with lo w pO 2. Genetic e vents can also influence HIF expression, particularly through loss of function of the VHL tumor suppressor protein.19,32 Cells that are def icient in VHL function have impaired degradation of HIF-1 α and can consequentl y ha ve increased le vels of HIF-1 α expression. Therefore, the HIF-1α expression in VHL-null cells may not correlate with h ypoxia. This ef fect has been obser ved in clear cell renal carcinomas and cerebellar hemangiomas with deficient VHL function.20,33 Additionally, Moeller and colleagues34 showed that increased HIF activation also occurs during reoxygenation following radiotherapy, the lik ely result of reacti ve o xygen species and stress granule depolymerization. This increase in HIF acti vation follo wing radiation therap y has also been paralleled in cer tain types of chemotherap y re gimens.182 Despite these disadvantages, exogenous reporter genes provide the only current insight into the molecular consequences of h ypoxia on a cellular le vel and can provide an important adjunct to other imaging modalities.

OPTICAL METHODS Oxygen-Sensing Electrodes Similar to the above immunohistochemistry methods of evaluating hypoxia, electrodes often provide a comparative pO 2 measurement for e valuating emer ging hypoxia imaging modalities. As such, it is impor tant to understand the principles behind these measurements as the y often constitute the “gold standard” b y w hich other modalities are measured. These electrodes are not without their o wn disadvantages, however. First, electrodes are in vasive, requiring inser tion of a probe within the region of interest. This may cause damage to the tissue at the point of inser tion, disrupting the local

764

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

oxygen supply. The invasiveness of these probes often limits much of their applicability to research procedures, or if used clinically, they are often limited to tumors that are superf icially accessible, and computed tomography (CT) or ultrasound guidance ma y be required to assure proper placement in the tumor . Second, the sensing v olume is f ixed based on the size of the sensiti ve element and can v ary betw een sensor types.35 The size of the probe sensing element can therefore deter mine whether the pO 2 reading is representative of the microvascular region, or a more global picture of oxygenation, for which these electrodes may not be as w ell suited. Third, since each electrode can generally only provide either a single point of measurement or a line of measurement points, there is a limited ability to provide spatial information. This lack of spatial infor mation limits inter pretation of the data; the point or points of measurement ma y not be representative of the tumor as a w hole. The major benef its of these devices are that they are relatively easy to implement and can be calibrated to quantify o xygen tension in a direct and absolute manner.

approximately 5 to 10 µm.35 There is a tradeoff, however, as less consumed o xygen results in a lo wer signal, and currents can fall in the pico-amp range, which means these devices must be well shielded.36 Eppendorf Microelectrode System

The commerciall y a vailable Eppendorf microelectrode system has g reatly f acilitated the application of polarographic electrodes to clinical and preclinical application, owing to its ease of use, and the accurac y with w hich measurements can be made. 36 The system is equipped with an automated stepper motor, which enables measurement of oxygen tensions along a linear ar ray of measurement sites. This coupled with automated digital recording of the oxygen tension greatly simplifies collection of data. The limitations of this system include the fact that because it consumes o xygen, it cannot mak e repeated measurements of the same site. This can be problematic when the dynamic response to a treatment is of interest. Fluorescence Lifetime Sensor

Polarographic Electrodes

These e lectrodes f unction b y m easuring t he c urrent formed at a cathode upon the ionization of O 2. Hence, these probes consume o xygen, so care must be tak en to avoid perturbing the local oxygen tension. The amount of oxygen consumed , and thus the cur rent generated , is dependent on the size of the electrode tip. The amount of current generated is linearl y proportional to the concentration of o xygen, w hich simplif ies calibration. On the other hand, it also means these de vices suffer from poor signal to noise at low oxygen tensions.36 Two designs will be discussed here, the commercially a vailable Eppendorf electrode system and the recessed-tip microelectrode. Other designs, such as the platinum wire, and Clark-type ha ve g reater prob lems with oxygen consumption, which limit their usefulness in tissue measurements.36 Recessed-Tip Design

To characterize oxygenation at the microvascular level, it is necessar y to minimize both sampling v olume and oxygen consumption. This ma y be achie ved using a recessed-tip design, w hich both limits o xygen diffusion through a specif ic area and reduces the amount of oxygen consumed.37 Typical tip diameters range from

A relati vely recent inno vation is a fluorescence lifetime sensor, introduced commerciall y as the Oxylite probe b y Oxford Optronics. This consists of a fluorescent dye at the tip of an optic fiber, which is clad in an oxygen permeable membrane. A pulsed laser source is used to excite this dye, and its deca y lifetime is characterized. Because the fluorescence of this dye is quenched by oxygen, the lifetime is shorter at high oxygen tensions. The nonlinear response of this electrode necessitates that each sensor is calibrated individually, and the y ha ve a limited shelf life. Oxylite probes have a relatively larger sampling volume (approximately 200 µm) compared with recessed-tip or Eppendorf electrodes,35 so the pO 2 is averaged over an area that is more likely to ha ve both interstitial and v ascular components than other electrode readings.They also exhibit minimal oxygen consumption compared with the Eppendorf probes, enabling continuous monitoring of oxygen tension in a single re gion. Another major adv antage of this approach is that the signal to noise is highest at lo w oxygen tensions.

Optical Spectroscopy Optical techniques represent an emer ging f ield for the characterization of h ypoxia. There are tw o primar y sources of intrinsic contrast for deter mining tissue hypoxia in vi vo, namel y hemo globin saturation and

Imaging Hypoxia

redox ratio. Optical spectroscop y is sensiti ve to hemoglobin saturation o wing to the dif fering absor ption spectra of o xygenated hemo globin (HbO 2) and deoxygenated hemoglobin (Hb). 38 There will typically be a mixture of HbO 2 and Hb present in the b lood. By characterizing the tissue absorbance as a function of wavelength, it is possible to determine what fraction of the heme-binding sites are bound to o xygen using the equation:

765

the visible and near-infrared (NIR), 44 so characterization of bulk tissue absorbance is still useful in determining the hemoglobin content and saturation. The four main applications of optical spectroscop y principles are diffuse reflectance, hyperspectral imaging, and diffuse optical topography and tomography. While diffuse reflectance and the diffuse optical topography and tomography methods are pre valent in both research and clinical uses, h yperspectral imaging is cur rently limited to preclinical research.

µ a(λ) = 2.303Hbtot ((1 − S)εHb (λ) + Sε HbO2 (λ)), where µa is the wavelength dependent absor ption coefficient, Hbtot is the total hemoglobin concentration, ε is the molar extinction coefficient, and S is the fractional o xygen saturation. This equation can be solv ed using least squares f itting, yielding the hemo globin oxygen saturation, defined as HbO 2/(HbO2 + Hb) × 100%. Knowing the hemoglobin saturation, it is possib le to relate this directl y to the tissue pO 2, by using the hemoglobin dissociation curve.39,40 This is problematic, however, because the af finity of hemoglobin for o xygen is affected by a variety of factors which must be considered, including pH, carbon dio xide, temperature, and 2,3diphosphoglycerate.41,42 These are adapti ve f actors that facilitate effective delivery of o xygen under a v ariety of conditions ( for e xample, u nloading m ore o xygen i n regions of high metabolic acti vity). With these considerations, an absolute calibration of tissue pO 2 is difficult and is generally avoided by presenting and analyzing data in ter ms of hemo globin saturation onl y. In addition, it is not possib le to measure o xygenation outside of b lood vessels, so optical spectroscop y only provides vascular pO2 and not tissue pO 2. Perhaps the best known and most widely applied use of optical spectroscop y is in the pulse o ximeter, w hich characterizes the hemo globin saturation of ar terial blood.43 This takes advantage of the f act that with each heart beat, the amount of material blood present in the tissue rises and f alls w hile the absor ption of sur rounding tissues is relati vely static. Thus, any dynamic change in absorbance can be assumed to be due to changes in ar terial b lood v olume, w hich g reatly simplif ies modeling. Unfortunately, this is of limited use in characterizing local tissue h ypoxia since it is not sensiti ve to the nonpulsatile flo w of the capillar y bed w here most gas exchange occurs. To ascer tain hemoglobin saturation of the entire vascular bed, the nonpulsatile absorbance of tissue must be characterized. F ortunately, even accounting for absor ption b y other tissue constituents, hemo globin still represents the dominant absorber in most tissues in

Diffuse Reflectance

The simplest approach to measuring hemo globin saturation in vi vo is to measure the dif fuse reflectance as a function of wavelength. In this method, the tissue is illuminated and the backscattered light is measured as a function of wavelength, with light typicall y coupled to a light source and detector via a f iber optic probe. The resulting dif fuse reflectance spectr um is then modeled and related to the underl ying absor ption and scattering properties of tissue, w hich can in tur n be related to hypoxia. For a single measurement site, which comprises a point measurement of tissue ph ysiology, a range of modeling solutions are a vailable, including empirical or simple analytical expressions,45–49 analytical (commonly diffusion-based) approximations of light transport,50–55 or Monte Carlo modeling of light transpor t.56,57 Twodimensional (2D) representations of tissue optical properties can then be obtained b y ha ving an ar ray of such measurement sites or b y sequentiall y mo ving the f iber optic probe to achie ve some spatial infor mation. The main disadvantage of this approach is that it has limited ability to obtain depth-resolved information. Hyperspectral Imaging

2D i maging o f h emoglobin s aturation h as a lso b een achieved in windo w chamber models. 58 This technique involves surgical attachment of a window chamber, within which a tumor can be implanted. This is commonly done in a dorsal skin fold59 but can also be performed orthotopically.60,61 This enables direct visualization of the tumor microvasculature with a commerciall y a vailable microscope. Equipping such a microscope with a tunab le optic filter enables measurement of the transmitted or reflected light as a function of wavelength, which can then be related to the w avelength dependent absor ption coefficient, as in Equation 1, to yield the hemo globin saturation. 62–64 The advantages of this approach are that it is capab le of high

766

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

resolution and can be used to image the o xygen saturation of indi vidual b lood v essels, thus enab ling modeling of oxygen transport in the microvasculature. Diffuse Optical Topography and Tomography

Finally, se veral emer ging technolo gies are being de veloped to obtain deeper tissue Hb saturation than the typical superf icial optical techniques described abo ve. Diffuse optical topography and tomography involve measurements comprising an ar ray of light sources/detectors spanning a relatively deep (tens of mm) tissue site.65 This depth is suf ficient to probe the cor tex, which is the primary application of diffuse optical topography, as well as probing the breast tissue and fetal brain.66 Diffuse optical tomography is similar to dif fuse optical topo graphy but uses more widely spaced sources and detectors to enab le even g reater penetration depth. Because of the need for deep-penetrating photons, NIR light is used. Tissue absorption is e xceptionally lo w in this spectral range, thus it is commonl y refer red to as the “NIR windo w.”67 Light in this windo w is absorbed primaril y by hemoglobin, water and lipids, with small amounts also absorbed by m yoglobin (Mb) found in muscle and c ytochrome c oxidase.68 The primary difficulty of these methods is that while absorption is low, tissue still scatters strongly, resulting in relatively poor resolution (~cm). The real adv antage of these optical techniques is the ability to pro vide functional information, and current research has been focused toward impro ving the functional signif icance of these methods, as opposed to attempting to achie ve the high resolution anatomic infor mation that is better addressed by other imaging modalities. 66 Diffuse optical tomo graphy has been widel y researched as a potential adjunct to mammography as tumors are generally more vascularized and hypoxic than normal tissues. Optical tomography has also been pursued as a means of detecting insuf ficient supply of oxygen to the fetus during labor. As many optical topo graphy and tomo graphy applications relate to functional brain imaging and rel y on the hemodynamic response, there are many correlates between this method and BOLD MRI. Advantages of this method over BOLD MRI include its potentially fast sampling rate, w hich has been demonstrated at 50 Hz, 69 as well as its por tability and relati vely low cost. Disadv antages include the lack of directl y core gistered anatomic information inherent to MRI methods (although multimodality measurements are possible), lower spatial resolution, and the shor ter penetration depth in the case of topography.

Novel Approaches to Improve Spatial Resolution

Finally, a number of novel approaches are being de veloped b y w hich the spatial resolution of optical techniques for measuring tissue h ypoxia ma y be improved. These include the use of ultrasound modulation70 and photoacoustic imaging. 71 Both of these techniques tak e adv antage of the interaction of light with ultrasound w aves to yield the functional information inherent to optical imaging, combined with the spatial resolution of ultrasound. Pump-probe techniques are also being de veloped to probe the dif fering time deca y prof iles of the e xcited states of o xy and deoxyhemoglobin using both tw o-photon absor ption microscopy72 and optical coherence tomo graphy. 73 These techniques are capab le of high (~ µm) resolution while providing quantitative infor mation regarding the hemoglobin saturation.

Redox Imaging Another source of intrinsic contrast is that of fluorescence redox imaging. This was pioneered by Chance and colleagues74 who demonstrated the fluorescence properties of the electron car riers, nicotinamide adenine dinucleotide (NAD), and fla vin adenine dinucleotide (F AD). Specifically, N AD is nonfluorescent, w hile its reduced form, N ADH, fluoresces w hen e xcited with ultra violet (UV) light. The re verse is tr ue of F AD, w hich is fluorescent in the o xidized (F AD), but not reduced (FADH2) form.74 Thus, by measuring the fluorescence of both these compounds, some measure of the redox status of a biologic sample can be obtained. Commonly, a fluorescence “redo x ratio” is calculated , usuall y def ined as FAD/(FAD+NADH) or alternatively as NADH/FAD. This thus provides an indirect measure of oxidative metabolism and hence h ypoxia. A number of g roups ha ve demonstrated the use of this metric for the diagnosis of cancer,75,76 and it has been proposed that it may be useful in quantifying h ypoxic stress. 77 The adv antage of this approach is its ability to probe the ph ysiologic response to hypoxia. However, it does not directly reflect oxygen tension. In addition, since these fluorophores are excited in the UV range/b lue, the penetration depth in tissue is limited to the order of millimeters. Another source of optic contrast is the absor ption of c ytochrome c, w hose absorption spectra is altered depending on its redox state. The use of this technique was pioneered by Jöbsis78 using NIR light to probe cerebral and cardiac tissue. The main limitation of this approach is that tissue absor ption is dominated b y hemo globin at these w avelengths, so

Imaging Hypoxia

separation of the absor ption due to c ytochrome c from the background absorption is challenging.

Phosphorescence Lifetime Another technolo gy recentl y pioneered b y Wilson and Cerniglia79 is that of phosphorescence lifetime imaging. This method is related to the fluorescence lifetime sensor described above, but rather than being encased in a f iber optic probe, the phosphors are encapsulated in a w atersoluble dendrimer and injected into the v asculature of the tissue re gion of interest. The dendrimer shields the phosphor, reducing its sensiti vity to the microen vironment. A light guide is used to focus the excited light from the phosphor to the surf ace of the tissue, w here it is then detected by a phosphorometer. The phosphor most commonl y used in vi vo for this method is Pcl-por phyrin, but other phosphors a vailable include Oxyphor G2 and Green 2W . These phosphors absorb in the NIR re gion of 620 to 1000 nm. 80 This spectral window proves to be advantageous since there is little absorbance from natural body pigments in this re gion, resulting in high specif icity for this method. 81 There are many advantages of phosphorescence imaging, including the ability to pro vide o xygen tension in absolute units with calibration and the ability to repor t on o xygen content throughout the entire tissue v olume. Additionally, this method has high accurac y in lo w pO 2 environments. With respect to resolution capabilities, light in the NIR range can penetrate tissue depths from a fe w millimeters to centimeters depending on the detection technique used. Also, the temporal resolution is on the order of seconds or less, enab ling almost “real time” data acquisition and ability for repeated measurements. 82 The phosphors can be detected using a v ariety of optical imaging modalities, including man y described above, enabling a wide range of capabilities. Phosphorescence imaging has become increasingl y impor tant for in vi vo applications and has successfull y provided 3D spatial re gistration using confocal imaging 83,84 and diffuse tomo graphy.85 The main limitations of this approach are the necessar y injection of the phosphor into the v asculature, the requirement of specialized light sources and imaging systems to characterize the lifetime of these compounds, and the specialized techniques required for preparation and handling of these probes, although some are commerciall y a vailable. Furthermore, it is impor tant to distinguish that this method typically is a reporter of vascular pO2, although there are phosphor probes that can penetrate into tissue, thereby also pro viding some infor mation as to the

767

tissue pO 2 status. When the probes are conf ined to the vasculature, however, comparisons with other modalities, par ticularly nonvasculature imaging methods, are required for robust inter pretation of the tr ue tumor hypoxic status.

PET/SPECT PET and SPECT are minimally invasive imaging techniques that are used to measure and map ph ysiologic and biologic processes using radiophar maceutical compounds. Both modalities rel y on radiotracers: either positron emitters for PET scanning or γ emitters for SPECT. In h ypoxia imaging, the radiotracers f irst are allowed to accumulate in h ypoxic cells. They are then detected b y PET or SPECT scanners. Because of the nature of the positron emitter, which emits two 511 keV γ rays in 180° opposite directions, PET usuall y has higher resolution and up to 100 times the sensiti vity of SPECT imaging. Both PET and SPECT ha ve become promising methods to measure tissue oxygenation, largely due to the development of a multitude of h ypoxia markers that can be labeled with either a positron emitter or a γ emitter. While the f irst nuclear medicine hypoxia marker was introduced in 1986, in the past 10 y ears the number of PET and SPECT imaging agents has risen dramatically as clinical studies ha ve sho wn the impor tance of tumor hypoxic status on treatment outcome (T able 3). A variety of radiomarkers are cur rently used, with 18F or 124I labels used for PET imaging and 122I, 125I, or 99mTc labels for SPECT imaging. These markers can be classified into two categories as nitroimidazoles or non-nitroimidazoles. Most h ypoxia mark ers are nitroimidazole-related compounds, with the exception of Cu-ATSM and 99mTc-HL91.

FMISO Among the potential PET/SPECT h ypoxia tracers, 18 F-FMISO is the most e xtensively studied and widel y used PET radiotracer for imaging tumor h ypoxia. First proposed in 1986 for use as a hypoxia marker, its mechanism of h ypoxia selectivity has since been w ell characterized and is referenced as the standard clinical PET hypoxia mark er.86,87 When 18F-FMISO dif fuses into cells, it is f irst reduced by nitroreductase enzymes to a radical form. Under aerobic conditions, the radical compounds will be reo xidized and dif fuse out of cells in a clearance process. Ho wever, under hypoxic conditions, these radicals will bind to intracellular macromolecules and accumulate inside the cell. Almost all nitroimidazole deri vative compounds share similar

768

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Table 3. MAIN HYPOXIA MARKERS FOR PET/SPECT IMAGING PET Hypoxia Markers

Abbreviation

Time Introduced

Clinical Trial (Tumor Type)

(18F) Fluoromisonidazole

18

198686

Head/neck cancer,91,93,94,99,153 NSCLC ,95,154 soft tissue sarcoma,155 renal tumor,97 brain tumor98,156,183,184

(18F) Fluoroazomycin arabinofuranoside

18

2005106

Head/neck,157,158

(18F) Fluoroetanidazole

18

1999159

(18F) Fluoroerythronitroimidazole

18

1995160

Head/neck161

(18F) [2-(2-nitro-1[H]-imidazol-1-yl)N-(2,2,3,3,3-pentafluoropropyl)acetamide]

18

2001101

Head/neck187,188

(124I) Iodoazomycin arabinoside

124

2007162

(124I) Iodoazomycin galactopyranoside

124

2004104

(60Cu, 61Cu, 62Cu, or 64Cu)-diacetylbis(N4-methylthiosemicarbazone)

60

(60Cu) (61Cu) (62Cu) (64Cu)

SPECT Hypoxia Markers

F-FMISO

F-FAZA F-FETA F-FETNIM F-EF5

I-IAZA I-IAZG

Cu-ATSM, 61 Cu-ATSM, 62 Cu-ATSM, 64 Cu-ATSM Abbreviation

2001,163 2005,164 1997,165 1999166

Time Introduced

(123I, 125I, or 131I) Iodoazomycin arabinoside

123

I-IAZA, 125 I-IAZA, 131 I-IAZA

(123I) 1992,168 (125I) 1991,102 (131I) 2005,169

(123I, 125I, or 131I) Iodoazomycin galactopyranoside

123

I-IAZG, 125 I-IAZG, 131 I-IAZG

(123I) 1996,171 (125I) 1998,172 (131I) 1998172

(99mTc) 4, 9-diaza-3, 3, 10, 10-tetramethyldodecan-2, 11-dione dioxime

99m

1997173

Tc-HL91

(62Cu) (60Cu) (60Cu) (60Cu)

Lung185 NSCLC,110 cervical cancer111,167 Rectal cancer186

Clinical Trial (Tumor Type) (123I) Multiple tumor sites168,170

Head/neck cancer115 NSCLC,114,115,174,175 multiple tumor sites,176 nasopharyngeal carcinoma177

NSCLC = nonsmall cell lung cancer; PET = positron emission tomography; SPECT = single-photon emission computed tomography.

mechanisms of retention and accumulation in h ypoxic tissue; their differences reside in their lipophilicity, pO2 dependency, and difficulty of synthesis. The synthesis of FMISO has been quite standardized and hence is relatively simplistic. The detailed procedure is outlined b y Grierson and colleagues. 88 Additionally, Chang and colleagues 89 have recentl y de veloped an automatic robotic system for FMISO synthesis, with radiochemical purity greater than 97%. This simplicity of synthesis is one of the main reasons for FMISO’s current clinical dominance. Studies cor relating direct pO 2 measurement with FMISO retention ha ve sho wn strong specif icity of FMISO for h ypoxic re gions, with binding occur ring mostly in pO 2 ranges betw een 2 and 10 mm Hg. 90,91 Furthermore, FMISO is sensiti ve to lo w pO 2 only in viable cells and not in necrotic tissue since a functioning electron transport chain is needed to reduce the nitroimidazole.7 Equally important, the delivery of this

imaging agent is not limited or influenced by perfusion after 2 hours circulation time. 92 The radiation dose required for this procedure is relati vely lo w, with approximately 250 MBq of acti vity used per patient. 7 Many clinical trials ha ve been perfor med with respect to FMISO PET imaging, for a vast number of cancer types including head/neck cancer ,93,94 NSCLC,95 soft tissue sarcoma, 96 renal tumor ,97 and brain tumor .98 Additionally, strong cor relations ha ve been repor ted between FMISO uptak e as a h ypoxia indicator and therapeutic outcomes.94,99 Furthermore, Lee and colleagues93 have perfor med a feasibility study using FMISO for image-modulated radiation therapy (IMRT) in patients with head and neck cancer . While FMISO has the adv antage of simplistic fabrication as well as extensive study and characterization, there are limitations to this PET hypoxia marker, including its slow blood clearance; this dela ys hypoxia-specific imaging for up to 2 to 3 hours after injection with

Imaging Hypoxia

relatively lo w tumor/b lood or tumor/muscle ratio, resulting in onl y modest o verall signal-to-noise ratio. Additionally, high uptake in the liver and accumulation in the bladder limit its use for imaging lesions of the li ver or those in the vicinity of the urinar y track. 18

F-EF5/EF3

18

F-EF5 is another nitroimidazole deri vative h ypoxia marker, proposed for PET imaging because of the superior hypoxia specificity of its unlabeled compound, EF5. EF5 is a standard h ypoxia marker in immunohistochemistry, with high hypoxia specificity independent of tumor type or perfusion ef fects.87,100 This rationale led to the de velopment of its 18F-labeled EF5 counterpart, which exhibits superior uniform distribution and specificity that could help it to become a powerful PET imaging mark er. Lik ewise, EF3, a deri vative of EF5, has also been de veloped with 18F as a PET h ypoxia marker. Ho wever, 18F-EF5 and EF3 ha ve had limited applicability due to their relati vely difficult synthesis procedure. Much effort has been put into improving the synthesis of 18F-EF5, and recently it has become practical to produce labeled EF5 in a reasonab le time with adequate yield.101 The main disadvantage to 18F-EF5 and EF3 is that, similarly with FMISO, their long halflife of around 12 hours can result in binding to aerobic cells and slo wer mark er clearance from the b lood and normal tissue, interfering with the o verall h ypoxia specificity of the image. 81 Both 18F-EF5 and EF3 have moved into the clinical arena, with initial biodistribution and phar macokinetics studies in head and neck cancer patients. 187,188

IAZA/IAZG/FAZA As mentioned above, both FMISO and 18F-EF5/EF3 suffer from long biolo gic half-li ves that af fect the o verall signal-to-noise ratio of the image. As a result, several compounds have been developed with faster blood clearance, most notab ly the sugar -coupled 2-nitroimidazole derivatives iodoazom ycin arabinoside (IAZA), Iodoazomycin galactopyranoside (IAZG), and fluoroazomycin arabinofuranoside (FAZA). Both of the iodine compounds can be labeled for SPECT imaging with either I-123 or I-125 and for PET imaging with I-124. The relatively long physical half-life of I-124 (4.2 days) enables imaging to be perfor med at times up to se veral da ys postinjection, allo wing adequate time for clearance of background acti vity and subsequentl y a higher tumor/muscle and tumor/blood ratio. This is not possible

769

with the shor t-lived 18F-labeled compounds (1.8 hours half-life). IAZA was first introduced in 1991 and showed good retention under h ypoxic conditions with much faster b lood clearance as used for SPECT imaging. 102 Similar to IAZA, IAZG is also e xcreted f aster than FMISO. Ho wever, in a recent study b y Riedl and colleagues,103 it was reported that the 18F-FMISO images are of superior diagnostic image quality compared with the 124 I-IAZG images in the Morris hepatoma McA-R-7777 tumor model; this is due to a lo wer tumor -to-normaltissue ratio with 124I-IAZG. In contrast, 124I-IAZG imaging was shown to be superior to FMISO when performed at 24 to 48 hours postinjection, w hen the w hole-body background had dissipated considerab ly.104 FAZA w as later de veloped with a str ucture similar to IAZA and IAZG but labeled with fluorine because of easy accessibility of the F-18 isotope. Sor ger and colleagues 105 compared 18F-FMISO with 18F-FAZA in rat carcinosarcoma tumors, both in vitro and in vivo, demonstrating that both compounds had similar accumulation in sites of h ypoxia on earl y PET imaging, but that 18F-FAZA was cleared much f aster from the b lood and muscle tissue. Another study sho wed that 18F-FAZA displa yed signif icantly higher tumor -to-muscle and tumor -to-blood ratios compared with 18F-FMISO in a mouse mammar y tumor model, indicating a f aster clearance of 18F-FAZA from normal tissues.106

FETNIM Studies have shown that misonidazole and its deri vative FMISO can produce deleterious side ef fects, such as peripheral s ensory n europathy. A m ore h ydrophilic hypoxia marker, fluoroerythronitroimidazole (FETNIM), was developed with the expectation that its poor diffusion would decrease penetration through the b lood-brain barrier and consequently reduce peripheral neuropath y. In a study comparing FETNIM and FMISO in murine mammary tumor models, it was found that while there was no significant difference in the intratumoral uptake between the two mark ers, 18F-FETNIM e xhibited lower radioactivity uptak e in nor mal brain tissue compared with 18 F-FMISO.107

Cu-ATSM In addition to the h ypoxia markers of the nitroimidazole group discussed above, metal-chelated compounds, such as 99mTc-HL91 and Cu-ATSM, have also been e valuated as h ypoxia mark ers. The use of copper -labeled

770

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

radiopharmaceuticals for PET is attractive because of the increasing availability of four positron-emitting radionuclides of copper ( 60Cu, 61Cu, 62Cu, and 64Cu). The preparation of positron-emitting 62Cu via a 62Zn/62Cu generator system has been reported b y se veral g roups.108,109 Cu-ATSM is a small molecule with high per meability to the cell membrane. It has much f aster b lood clearance than either EF5 or FMISO , resulting in a much shor ter imaging time. Intracellular Cu-A TSM is bioreduced and trapped in viab le cells under lo w cellular pO 2 of about 10 mm Hg. 86 Three clinical studies ha ve cor related the uptake of Cu-A TSM with poor treatment outcomes in cervix, rectal and lung cancer .110,111,186 However, the retention mechanism of this h ypoxia mark er remains unclear. Yuan and colleagues112 reported that the hypoxia specificity of Cu-ATSM is tumor type dependent, sho wing high hypoxia correlation in murine glioma and mammary adenocarcinoma tumors but not in f ibrosarcoma. O’Donoghue and colleagues113 similarly demonstrated this tumor dependent specif icity of Cu-A TSM; w hile human squamous cell carcinoma x enografts sho wed uptake of Cu-ATSM correlating with FMISO, they found discordance in a rat prostate cell line, with cor relation to FMISO occur ring onl y at later times postinjection (16–20 h). Additional studies that fur ther delineate the retention mechanism and Cu-A TSM’s application as a predictive assay will be needed. 99m

Tc-HL91

99m

Tc-HL91 is a SPECT imaging mark er for h ypoxia measurement. It w as f irst proposed as a h ypoxia marker in 1997 and has become a promising hypoxia marker for SPECT imaging due to its convenient 6 hour half-life and 99m Tc-HL91 good a vailability. In the synthesis of the compound, 99mTc(V) is chelated to the HL91 ligand (4,9-diaza-3,3,10,10-tetramethyldodecan-2,1-dione dioxime) for ming a macroc ycle in w hich the Tc atom occupies a central position. 99mTc-HL91 has been used in two clinical trials for head/neck 114 and NSCLC. 115 However, recent studies ha ve sho wn lo w tumor -to-normaltissue ratios in necro tic tumors 116 and lo w h ypoxia specificity in ischemic myocardium.117

MAGNETIC RESONANCE Nuclear imaging techniques can be classif ied into tw o separate sections depending on the subatomic par ticle being imaged. F or MRI, the subatomic par ticle of interest is the proton/atomic nuclei. Recentl y, electron

paramagnetic (“spin”) resonance (EPR or ESR) has been developed to center on electrons as the subatomic par ticle of interest. Both techniques allow for noninvasive imaging of a subject and are, from a strict inter pretation, indirect indicators of tissue h ypoxia. Ho wever, some of these methods, such as MRI 19F labeled molecules and EPR electron spin probes, ha ve physical proper ties that are directly affected by local oxygen tension, which render them a more direct indicator than other MRI methods. Unfortunately, the capacity of these probes to determine direct oxygen tension is often compromised by an inability to deliver them homogenously throughout the tumor tissue. For the pur poses of MRI of h ypoxia, three general indirect techniques have been developed that are used to noninvasively infer local pO 2: BOLD imaging, magnetic resonance spectroscopy (MRS) with 19F or 31P, and DCEMRI. Additionally, EPR methods include not onl y EPRI but also Overhauser-enhanced MRI (OMRI)/Proton Electron Double Resonance Imaging (PEDRI). MRI methods share the advantage of being able to provide coregistered functional information with anatomic information. While BOLD imaging is the onl y tr uly nonin vasive MRI method, the others are minimall y noninvasive, requiring only the injection of an e xogenous contrast agent or probe. It is impor tant to note that these methods are mainly indicators of vascular pO2, and the corresponding inference of tissue pO 2 is not always straightforward.

BOLD Imaging Currently, the use of MRI methodolo gy to infer tissue hypoxia has focused mostly on the BOLD technique, also referred to as intrinsic susceptibility weighted MRI. Although initially BOLD was used in functional MRI studies of the brain, its applications ha ve since expanded to tumor o xygenation e xperiments. This method tak es advantage of the intrinsic magnetic proper ties of hemoglobin contained within red b lood cells. As the o xygen saturation of the hemo globin protein increases, the iron contained within the heme subunit changes from a paramagnetic high spin state under lo w pO2 to a diamagnetic low spin state under high pO 2.118 This results in the ability to differentiate between paramagnetic deoxyhemoglobin and o xyhemoglobin, w hich is not paramagnetic. Deoxyhemoglobin alters the local magnetic field, thereby increasing the transv erse relaxation rate (R 2*) of w ater contained in b lood and pro ximal tissue. This alteration thus sensitizes the BOLD imaging technique to the pO 2

Imaging Hypoxia

of perfused b lood v essels and the tissue immediatel y surrounding those vessels.119 In terms of the MRI signal obtained, a decrease in the o xygenation of the blood will result in a corresponding decrease in the MRI image signal intensity. While flo w ef fects from deo xyhemoglobin are the main contributor to the BOLD image contrast, other static effects will also contribute, such as iron content found in muscle Mb or f ibrosis/ligamentous str uctures.7 This results in a need to decouple deo xyhemoglobin flo w effects from those of the static components. Decoupling is accomplished through application of g radient echo techniques and subsequent measurement of the T2* relaxation time. The transverse relaxation rate (R 2*) is related to this parameter through the equation R 2* = 1/T 2*. The change in 1/T 2* is linear with hemo globin saturation, enabling it to be quantif ied.118,120 Thulborn and colleagues118 were the first to measure this property change, demonstrating the dependence of T2* relaxation time with varying degrees of oxygenation. Absolute measurement of T2* is dependent upon both the percent hemo globin saturation and the absolute amount of hemo globin (ie, b lood v olume and hematocrit).7 Therefore, the distribution of b lood v olume should be kno wn to cor rectly inter pret the R 2* images. For instance, in a perfused area, there might be tw o different re gions; the re gion with the higher R 2* intensity can then be inferred to be relatively more hypoxic. In this way, areas of R 2* intensity can be compared to pro vide a measure of relative spatial hypoxia but not absolute pO 2 values. Therefore, changes in the BOLD signal (changes in R2*) are needed to evaluate the regions of interest and infer pO2 values. In functional MRI e xperiments, this change is due to perfusion in areas of the brain under going neural acti vation. To e valuate other tissues such as tumors, alteration of natural tissue perfusion through vasomodulation is often used. Vasomodulation is generall y car ried out through inspiration of gases with ele vated le vels of o xygen, known as hyperoxic gases. The most common hyperoxic gas used is carbo gen (95% O 2:5% CO 2). Carbo gen breathing induces multif arious ef fects in the tumor , including changes in o xygenation and v ascular physiology. In general, h yperoxic gases increase ar terial blood saturation, tumor b lood flow, and tumor b lood volume, causing increased pO 2 delivery to the tissue of interest.119 Increased b lood o xygenation yields a decrease in R2*. The intensity of the change in R 2* is g reatest in more v ascularized re gions that typicall y ha ve a lar ger blood volume. BOLD images taken before and after carbogen gas administration can be compared and used to

771

determine which areas are more susceptible to oxygenation and blood flow effects. Studies have shown that carbogen v asomodulation results in similarl y fluctuating tissue pO 2 values. Ho wever, other studies ha ve sho wn that this does not occur in all tumors as only slightly more than half of human tumors sho wed changes in the R2* BOLD signal following carbogen breathing.121 The inconsistent results obtained from hyperoxic gas modulation highlight the impor tance of understanding the parameters that af fect the BOLD signal. While the change in R2* typically correlates with expected changes following h yperoxic breathing, there are a v ariety of physiological effects that can confound cor rect interpretation of the BOLD MRI data, resulting in R2* changes that are smaller or e ven the re verse of that e xpected.119 Change in R 2* is influenced by the blood volume, blood flow, and arterial saturation and potentially by pH and altered glucose metabolism. The increased CO 2 burden in the b lood causes v asodilatation of the capillar y bed and aids in increasing o xygenation of the tumor, consistent with the increased hemo globin saturation and decreased R 2* mentioned abo ve. Ho wever, h yperoxic gases can affect noncapillary vasculature differently, significantly constricting arterioles, venules, and neovasculature.122 The ef fect of h yperoxia ma y also be tumor-dependent as some tumors respond b y reduced blood flow and vasoconstriction.123 The intensity of the h yperoxic response can also be influenced b y the micro vascular density , maturity and diameter of v essels, w hich modulates the o verall b lood volume. For instance, hypoxic regions with high R 2* signal and dense, lar ge diameter vessels may show a greater change in signal w hen e xposed to carbo gen due to the large blood volume and hence greater influence of perfusion in this re gion. Similarl y, another re gion of similar hypoxic fraction, yet situated in an area of lo w microvessel density or small v essel diameter, may show negligible change with carbogen modulation due to the lower impact of perfusion changes in the area. Tumors may contain vessels with little or no perfusion, and this may further lessen the effects of carbogen breathing. The maturity of blood vessels may also play a pivotal role, as mature b lood vessels better respond to vasodilation and constriction stimuli and thus are more lik ely to be influenced b y carbo gen vasomodulation than immature b lood v essels.7 In these ways, the blood flow and blood volume responses of carbogen breathing are coupled; impro ved b lood flo w is likely to be associated with an increase in b lood volume and therefore decrease R2*. However, the degree to which this effect causes a change in R 2* may be dependent on changes in oxygen consumption or utilization. 119

772

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Oxygen utilization in the tumor en vironment also affects the BOLD signal, particularly from changes in pH and oxygen metabolism. These effects may have a lar ge impact on the infer red pO 2 value that is often not tak en into account. For instance, increased blood flow may not correspond to increased oxygenation if oxygen consumption in nearb y tissues also increases. Such ef fects ma y result in smaller R 2* changes or e ven increases in R 2*. Carbogen breathing has been sho wn to decrease b lood pH.119 A decrease in blood pH results in slight acidif ication and shifting of the o xygen saturation cur ve, known as the Bohr ef fect. This right-shift causes o xygen to be more readil y released from hemo globin, and consequently, more oxygen is delivered to tissue. Overall, this effect could result in increased deoxyhemoglobin and therefore an increased R 2*. However, the extent to which this effect influences the overall R2* remains unknown. In addition to changes in pH, carbogen breathing has also been sho wn to ha ve up to a tw o-fold increase on blood glucose levels.124 Since tumors can use both oxidative and gl ycolytic metabolism, increased b lood glucose can contribute to unpredictab le changes in hemo globin saturation and R 2*. High glucose le vels can catal yze a shift in the tumor to a more gl ycolytic metabolism, referred to as the Crabtree ef fect. In this instance, R2* may decrease as less o xygen in the tissue is used. In contrast, elevated glucose levels may further stimulate oxidative metabolism, escalating depletion of tissue O2 and increasing R2*.125 Tumor response to varying glucose levels is unpredictab le and can fur ther complicate analysis of the BOLD signal. While the R 2* images reflect the presence of deoxyhemoglobin and other static components, it is important to note that these proper ties ma y also be affected b y changes in perfusion. Increased b lood flo w will typicall y lead to better o xygenation and less deoxyhemoglobin. This consequence of b lood flo w and blood v olume underscores the impor tant distinction of BOLD images as indicators of vascular pO 2. While some tissue pO2 may be reflected in the tissue immediately adjacent to blood vessels, this typically does not reflect o verall tissue pO2, most of which is located further from blood vessels and is more lik ely h ypoxic tissue. Additionally, hypoxic areas that are close to b lood vessels may still not be reflected b y BOLD images due to the unpredictable flow content in tumor vessels. Blood vessels often may be present but contain stagnant, little, or no b lood flow. Validation of the BOLD technique has been performed with immunohistochemistr y126 and EPR. 127 Others have used the inherent BOLD signal-to-noise ratio as a representati ve measurement of inter mittent h ypoxia

and/or a mark er of acute h ypoxia without the use of carbogen gas. 128 Overall, BOLD imaging has the adv antage of obtaining relative pO2 values noninvasively by using endogenous hemoglobin as a contrast agent, with good temporal and spatial resolution. This is in stark contrast with other modalities, such as PET , where the temporal and spatial resolutions are poor and the method requires an e xogenous radioacti ve tracer . Ho wever, BOLD imaging is not a direct method of quantifying pO2, providing information on relative changes in oxygenation but not absolute oxygen concentration. While image resolution qualities are high, the signal-to-noise contrast can be quite lo w. Fur thermore, one clinical disadv antage is the difficulty of implementing carbogen studies; patients can e xperience respirator y distress, and subsequent response of tumors is highly variable.7,121

MRS MRS techniques take advantage of performing proton (or other nuclei) spectroscop y measurements on a sample. These methods indicate the presence of chemical species containing the nuclei of interest and the concentration of that species within a re gion of interest. Most w ork has been in the area of 19F spectroscop y imaging, although some groups have also investigated the use of 31P.

19

F Spectroscopy

Fluorine imaging has significant advantages over the other MRI techniques, with the e xception that an e xogenous compound must be added for tissue contrast. The signal from 19F is 83% that from 1H with a similar magnetogyric ratio, and the lack of any endogenous 19F improves the signal-to-noise ratio, allowing higher sensitivity.129 There are two methods for 19F imaging: nitroimidazole-based compounds labeled with 19F that accumulate in hypoxic tissue in a reduction en vironment-dependent manner and e xogenous compounds that contain 19F and change their resonance frequenc y in an o xygendependent manner . Nitroimidazole-based compounds, also used in PET imaging, work as bioreducti ve agents that enter all cells. As described previously, they become trapped in the cell when exposed to reducing environments, such as found in h ypoxic regions.129 Their accumulation is then detected and repor ted as a measure of hypoxia. The adv antages of this technique are the systemic delivery of the compounds and their sensiti vity to hypoxia, making them potential clinical tools. Ho wever, heterogeneity of tumor perfusion and the presence of

Imaging Hypoxia

necrosis can hinder deli very and lo wer the o verall accumulation of the compound, limiting estimation of hypoxic fraction. Additionally, although nitroimidazoles accumulate in a reduced environment-dependent manner, the amount that accumulates is dif ficult to quantify based on pO 2. Hypoxia imaging with 19F is more commonly accomplished with perfluorocarbon-based compounds (PFCs). Oxygen measurements are obtained b y measuring the change in the spin-lattice relaxation rate (R 1) of the fluorocarbon probe. The relaxation rate is linear with dissolved oxygen concentration, enabling absolute measurement of pO 2.129 Many dif ferent PFCs ha ve been developed, such as perfluoro-15-cro wn-5-ether (15C5) and hexafluorobenzene (HFB). HFB is one of the most commonly used perfluorocarbons, due to its wide a vailability, low cost, high sensiti vity to o xygen, and lack of temperature dependency that other PFCs may exhibit. Perfluorocarbons ha ve the adv antage of measuring pO2 in the most direct manner for MRI and are generally unresponsive to other influences in the microen vironment, such as blood, pH, or common proteins. However, these compounds are e xtremely h ydrophobic and ha ve limited ability to be deli vered systemicall y. Emulsions have been used to combat these disadv antages but are generally being e xplored as b lood substitutes (e g, Oxygent, Fluosol, Therox, and Oxyphero) due to their high O2 solubility.129 Because of the dif ficulties with solubility, cur rent research with perfluorocarbons for pO2 measurement in vi vo has been perfor med with direct injection of the compound into the tissue of interest.129–133 The use of perfluorocarbons allo ws absolute quantification of o xygen concentration and per mits serial monitoring of tumor o xygenation status. The perfluorocarbons can be encapsulated into biocompatib le shields that retain o xygen per meability while enabling a longer biologic half-life; this f acilitates multiple readouts and repetitive measurements from the same re gion of interest.134 Additionally, 19F spectroscop y can be combined with 1H imaging to provide spatial registration of the hypoxic areas to the underlying anatomy. PFCs have been used to measure pO 2 in comparison to Eppendorf probes129 and have been shown to be retained in tissue both during the course of an e xperiment130,132,133 and while the tumor grows over time.131,133 31

P Imaging

Phosphate is used e xtensively in cellular metabolism, and some groups have looked at 31P and cellular pH in attempts

773

to estimate pO2.124,135,136 However, similar to the advantages and disadvantages of BOLD imaging, 31P is an endogenous marker of tumor metabolism but an indirect measure of pO2. Phosphate spectroscopy suffers from low signalto-noise ratio, lo w spatial resolution, and a length y scan time and requires v asomodulation and/or reference to normal tissue.

DCE-MRI DCE-MRI uses an e xogenous paramagnetic contrast agent to shor ten the T1 of protons, causing signal enhancement in a re gion of interest w here accumulation occurs. The contrast agent is normally a chelated gadolinium ion, injected intravenously. The change in T1 caused by the contrast agent can be quantified in a dynamic manner, yielding time-dependent concentration cur ves of the contrast agent for each pix el of imaging. The enhancement of a pixel can be represented as a two-compartment model: one compar tment for the v ascular space and the other for the e xtracellular space. 137 Using this tw ocompartment model and a v ascular input function (v ascular concentration time data), the e xtracellular volume fraction and the e xchange rate across the v ascular endothelium can be calculated on a pix el by pixel basis. Using this physiologic information, limited attempts have been made to cor relate these parameters to pO 2.138–140 Although this is a relati vely easy technique to implement and has direct clinical applications, several difficulties are present. For one, DCE-MRI is one of the most indirect methods for measuring pO2, as the signal obtained is based on v ascular perfusion, per meability, and e xtracellular volume fraction. 141 While these parameters ma y pla y a minor role in h ypoxia induction, the y fail to address some of the more pertinent parameters such as red blood cell flux or o xygen consumption. Additionally, due to the heterogeneity of tumor ph ysiology, inter pretation of these estimated ph ysiologic parameters is dif ficult.137 While some studies have shown correlation between pO2 and DCE-MRI signals, the linear cor relations have often been w eak, and the underlying physiologic basis suppor ting these cor relations is not well understood.141

EPRI and Spectroscopy EPRI is similar in principle to the MRI methods described previously, although there are a few key differences. For instance, both the type of radiation used and the required magnetic f ield strength are dif ferent. While MRI requires electromagnetic radiation in the

774

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

radiofrequency range, EPR uses radiation in the microwave region. EPR also requires a much lower magnetic f ield than MRI. Additionally, EPRI uses unpaired electrons to probe the tissue en vironment; this is in contrast to MRI, w hich uses h ydrogen or other elemental nuclei. Since the body does not possess suf ficient concentrations of stable unpaired electrons to perform native EPRI, EPRI of hypoxic environments requires the addition of an e xogenous compound similar to an MR contrast agent. 142,143 These paramagnetic probes can be fabricated either as an implantab le form that is insoluble or a soluble form that chemically interacts with but does not consume molecular oxygen.144 Upon imaging, these probes produce a spectr um analogous to MR spectroscop y. Each spin probe has a characteristic line width of its absor ption spectrum, with thinner or shar per lines pro viding g reater spatial resolution.143 When the probes are e xposed to molecular o xygen, the interaction of these tw o paramagnetic species alters the resonance beha vior of the spin probe, causing broadening of the characteristic line width. Thus, the line widths of the administered spin probes are sensiti ve to oxygen, and w hen quantif ied, the line broadening can determine pO 2. The relationship betw een molecular o xygen and line broadening is quantifiable, reproducible, and in most cases linear for each individual spin probe.142 With relation to common MR parameters, an increase in o xygen concentration causes an increased spin-spin relaxation rate (R 2) and consequentl y decreased spin-spin relaxation time (T2), which results in the broadening of the probe line width. 145 Soluble spin probes are based on the nitroxyl or triarymethyl/trityl radicals, although care must be tak en when using the nitro xyl radicals as the y are sensitive to reducing en vironments.142,143 While the solub le spin probes are v aluable for assessing the redo x environment of tissue, the y have limited ability to measure pO 2.142 In comparison, insoluble spin probes tend to e xhibit higher pO2 sensitivity and accurac y. They also ha ve the adv antage of a higher spin density and the ability to perfor m repeated measurements. The insoluble, or particulate, probes include lithium phthaloc yanine, lithium naphthalocyanine,146 coals,145 inks,147 and carbon b lacks.142 Both types of spin probes can either enter the cell or sta y in the extravascular space. They can be administered in a multitude of w ays, including intratumorall y,146 intravenously,148 intraperitoneally,143 intramuscularly,143 orally,145 or topically.145 EPR data ma y be acquired using either a “continuous wave” or pulse “time domain” spectrometer . Cur rently, most EPR studies are performed with the continuous wave

method w hich e xhibits less susceptibility to ar tifacts, particularly for lar ge objects. This method is also nonrestrictive in the selection of spin probes, as probes with both narrow and broad line widths can be detected. In contrast, time domain methods typicall y require spin probes with narrow line widths but ha ve the adv antage of impro ved temporal resolution. 81,143 Time domain methods are most commonly used for pharmacokinetic studies. Experimentally, EPR has been used to measure pO 2 in animal models with a variety of tumors, such as squamous cell carcinoma,149 fibrosarcoma,127 and RIF-1.146,147 While EPR is still a relati vely ne w h ypoxia imaging method, these studies ha ve demonstrated the ability of EPRI to measure h ypoxia, with cor responding independent validation and correlation of the EPR pO 2 measurement with Eppendorf o xygen probes, 148 BOLD imaging,127 or immunohistochemical hypoxia staining.150 EPRI has the adv antage of high specif icity, as there is little signal interference from the sur rounding tumor environment. Additionally, the probes are nontoxic and have reasonab le half-li ves, making them potentiall y translatable into the clinical environment. While this method ma y be used for repeatab le measurements, this can be limited b y the tissue clearance rate for the par ticular probe used. 142 Currently, the main limitations of this method are shallo w penetration depths on the order of millimeters, long data acquisition times of approximately 30 min, and hetero genous distribution of the paramagnetic probes, which may provide false absolute pO2 values in areas of lo w probe concentration. 151 However, improved technologies and emerging techniques may remedy some of these prob lems. For instance, the use of lower frequencies will increase penetration depth, and improvements in image reconstruction algorithms will result in shor ter data acquisition times. 152 Improved technologies such as OMRI/PEDRI (discussed belo w) ma y provide more insight into correcting for probe distribution. Currently this method has been limited to pre-clinical applications, in par t because the lack of anatomical information results in pO2 values that have little spatial reference to underlying structures. One future direction that ma y aid in solving this limitation w as demonstrated by Matsumoto et al, in which they were able to combine imaging platforms for EPRI and MRI by exploiting a common radio frequency between the tw o modalities at dif ferent f ield strengths. 189 This novel platform allows co-registration of pO2 values with anatomical infor mation. While currently this requires a 7 Tesla MRI magnet for the anatomical information, finding ways to incor porate this platfor m on a more clinicall y relevant 1.5 Tesla machine may allow for translation of this modality into the clinical environment.190

Imaging Hypoxia

PEDRI Recently, there has been an ef fort to impro ve EPRI of hypoxia through combination with 1H MRI, known either as OMRI or as PEDRI. OMRI/PEDRI is a doub le resonance technique that uses unpaired electrons and protons together. Similarly to EPRI, a paramagnetic probe must first be administered. A strong EPR pulse is then given to saturate the unpaired electrons of the spin probe. Saturation of the unpaired electrons results in dipole coupling with the surrounding protons, leading to an increase in proton polarization and subsequent enhancement of the MRI signal intensity .143,148 This is kno wn as the Ov erhauser effect, and the intensity of this effect is influenced by the presence of paramagnetic o xygen molecules. Unlike EPR, the spatial resolution is not governed by the line width of the spin probe but instead to the de gree of saturation. The Ov erhauser enhancement is the increase in amplitude of the proton’s magnetization relative to a ther mal equilibrium sample magnetization. The intensity of this enhancement is dependent both on the degree of electron spin saturation and on the line width of the spin probe, which is based on oxygenation. Consequently, if two images are taken with different EPR pulse energies, then the difference in Overhauser enhancement from the tw o images can be used not onl y to deter mine the oxygenation of the tissue but also the concentration of the spin probe. 143,148 This is par ticularly advantageous in regions of the tumor which have little perfusion and/or distribution of the spin probe. Therefore, this technique can provide more infor mation than the traditional EPRI method, correcting for areas that are falsely thought to be hypoxic due to heterogenous spin probe distribution. The in vi vo application of this technique has been demonstrated b y Subramanian and colleagues 143 and Krishna and colleagues, 148 with spatial and temporal resolutions of 1 mm and 2 min, respecti vely.

CONCLUSION Hypoxia imaging modalities ma y be used to deter mine applicable treatment, as a diagnostic tool, or to follo w patient response. Fur thermore, technologic advances in the field of both hypoxia and oncology have led to future visions of combinational therapies that are onl y no w starting to be realized. For instance, the increasing incorporation of image-guided radiation therap y (IMRT) into clinical radiation oncolo gy has enab led oncolo gists to deliver more spatiall y confor mal isodoses of radiation. Often refer red to as “dose-painting, ” the application of IMRT could per mit oncolo gists to incor porate special

775

targeting of hypoxic tumor areas into the treatment planning process, providing biologic rationale for dose escalation in these areas. This possibility is further facilitated b y the increasing numbers of dedicated PET/CT dual modality scanners, w hich enab le o verlay of the hypoxic areas intensif ied b y PET onto the anatomical CT images used in the treatment planning process. However, the usefulness of such combinational radiation planning is still some what uncer tain. Studies of patient benefit following hypoxia targeted radiation treatments are sparse, and the reoxygenation of tumors following radiation ma y limit the usefulness of incor porating hypoxia planning in the earl y stages of treatment. Ho wever, serial h ypoxia monitoring in the clinic o ver the course of treatment ma y still elucidate o verall tumor response to therapy. Technological advances have provided a wide variety of modalities for hypoxia imaging. Yet the differences between these modalities are g reat and must be understood to adequatel y inter pret the acquired data. Often, modalities with a more direct measurement of pO2 require an exogenous compound to be injected and interact with the oxygen available in the tissue. Yet these injected compounds are limited b y the perfusion and distrib ution effects in the tissue. Modalities that do not require an injected compound usually rely on more indirect measurements of tissue hypoxia, often only reflecting relative and not absolute values, or being indicative only of vascular pO 2. PET and BOLD MRI are often considered the “gold standards” for h ypoxia imaging, but PET’ s limited resolution and the lack of absolute data as w ell as limited tissue pO2 information of BOLD imaging provide impetus for pursuing better methods for imaging h ypoxia. The multitude of imaging modalities has also caused some confusion and lack of congruity in research or clinical trials. Each method relies on a unique principle to assess the o xygen content, with the result of h ypoxia being repor ted in a v ariety of dif ferent conte xts. This necessitates that each modality be rigorousl y v alidated against the other methods to ensure that the measure being used can provide a meaningful interpretation of the tumor oxygenation status. As the role of h ypoxia in cancer treatment becomes further elucidated, the ability to image the tissue oxygen content will become e xceedingly impor tant. Hypo xia imaging is becoming more widely used in the clinical environment, and more attention is turning to the in vivo molecular ef fects of h ypoxia in preclinical research models. In response, many novel imaging strategies have been in vestigated, each with their o wn adv antages and limitations. It is doubtful that one modality will fulfill all

776

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

the necessar y requirements for ideal h ypoxia imaging, particularly with the v ast amount of dif ferent applications. Therefore, fur ther pro gress in h ypoxia imaging may not stem from the identif ication of one ideal device but instead the reco gnition of how each imaging de vice may offer a unique insight into the hypoxic environment of the tumor.

19. 20.

21. 22.

REFERENCES 1. Thomlinson RH, Gray LH. The histological structure of some human lung cancers and the possib le implications for radiotherap y. Br J Cancer 1955;9:539–49. 2. Hall EJ, Giaccia AJ. Radiobiology for the radiologist. 6th ed. Philadelphia (PA): Lippincott Williams and Wilkins; 2006. 3. Brown JM. The hypoxic cell: a tar get for selecti ve cancer therap y— eighteenth Br uce F. Cain Memorial Award lecture. Cancer Res 1999;59:5863–70. 4. Graves EE, Giaccia AJ. Imaging tumoral hypoxia: oxygen concentrations and be yond. Oncolo gy (W illiston P ark) 2007;21:368–76; discussion 377–8, 384. 5. Dings RP, Loren M, Heun H, et al. Scheduling of radiation with angiogenesis inhibitors anginex and Avastin improves therapeutic outcome via v essel nor malization. Clin Cancer Res 2007; 13:3395–402. 6. Moeller BJ, Richardson RA, De whirst MW. Hypoxia and radiotherapy: oppor tunities for impro ved outcomes in cancer treatment. Cancer Metastasis Rev 2007;26:241–8. 7. Padhani AR, Krohn KA, Lewis JS, Alber M. Imaging oxygenation of human tumours. Eur Radiol 2007;17:861–72. 8. Kallinowski F, Zander R, Hoeck el M, Vaupel P. Tumor tissue o xygenation as e valuated b y computerized-pO2-histo graphy. Int J Radiat Oncol Biol Phys 1990;19:953–61. 9. Brizel DM, Rosner GL, Har relson J, et al. Pretreatment o xygenation profiles of human soft tissue sarcomas. Int J Radiat Oncol Biol Phys 1994;30:635–42. 10. Hockel M, Schlenger K, Aral B , et al. Association betw een tumor hypoxia and malignant pro gression in adv anced cancer of the uterine cervix. Cancer Res 1996;56:4509–15. 11. Nordsmark M, Bentzen SM, Rudat V, et al. Prognostic value of tumor oxygenation in 397 head and neck tumors after primar y radiation therapy. An inter national multi-center study . Radiother Oncol 2005;77:18–24. 12. Le QT, Chen E, Salim A, et al. An evaluation of tumor o xygenation and gene e xpression in patients with earl y stage non-small cell lung cancers. Clin Cancer Res 2006;12:1507–14. 13. Jiang BH, Semenza GL, Bauer C, Mar ti HH. Hypo xia-inducible factor 1 levels vary exponentially over a physiologically relevant range of O2 tension. Am J Physiol 1996;271(4 Pt 1):C1172–80. 14. Kavanagh BD, Coffey BE, Needham D, et al. The effect of flunarizine on erythrocyte suspension viscosity under conditions of e xtreme hypoxia, lo w pH, and lactate treatment. Br J Cancer 1993; 67:734–41. 15. Cardenas-Navia LI, Richardson RA, De whirst MW. Targeting the molecular ef fects of a h ypoxic tumor microen vironment. F ront Biosci 2007;12:4061–78. 16. Pries AR, Reglin B, Secomb TW. Structural adaptation of micro vascular networks: functional roles of adaptive responses. Am J Physiol Heart Circ Physiol 2001;281:H1015–25. 17. Thomlinson RH. Hypo xia and tumours. J Clin P athol 1977; 11:105–13. 18. Lanzen JL, Br uan RD , Klitzman B , et al. Direct demonstration of instabilities in oxygen concentrations within the extravascular

23.

24.

25.

26.

27. 28.

29.

30.

31.

32. 33.

34.

35.

36.

37.

38.

39. 40.

compartment of an e xperimental tumor . Cancer Res 2006; 66:2219–23. Semenza GL. Targeting HIF-1 for cancer therap y. Nat Re v Cancer 2003;3:721–32. Maxwell PH, Wiesener MS, Chang GW, et al. The tumour suppressor protein VHL tar gets h ypoxia-inducible f actors for o xygendependent proteolysis. Nature 1999;399:271–5. Powis G, Kirkpatrick L. Hypoxia inducible factor-1α as a cancer drug target. Mol Cancer Ther 2004;3:647–54. Belozerov VE, Van Meir EG. Hypoxia inducible factor-1: a novel target for cancer therapy. Anticancer Drugs 2005;16:901–9. Bussink J , Kaanders JH, v an der K ogel AJ. Tumor h ypoxia at the micro-regional le vel: clinical rele vance and predicti ve v alue of exogenous and endo genous h ypoxic cell mark ers. Radiother Oncol 2003;67:3–15. Williams KJ , P arker CA, Stratford IJ . Exo genous and endo genous markers of tumour o xygenation status: def initive mark ers of tumour hypoxia? Adv Exp Med Biol 2005;566:285–94. Koch CJ. Measurement of absolute o xygen levels in cells and tissues using o xygen sensors and 2-nitoimidazole EF5. Methods Enzymol 2002;352:3–31. Bennewith K L, R aleigh J A, D urand R E. O rally a dministered pimonidazole to label h ypoxic tumor cells. Cancer Res 2002; 62:6827–30. Contag PR, Olomu IN , Stevenson DK, Contag CH. Bioluminescent indicators in living mammals. Nat Med 1998;4:245–7. Wen B, Burgman P, Zanzonico P, et al. A preclinical model for noninvasive imaging of h ypoxia-induced gene e xpression; comparison with an exogenous marker of tumor hypoxia. Eur J Nucl Med Mol Imaging 2004;31:1530–8. Liu J, Qu R, Ogura M, et al. Real-time imaging of hypoxia-inducible factor-1 activity in tumor xenografts. J Radiat Res (Tokyo) 2005; 46:93–102. Safran M, Kim WY, O’Connell F, et al. Mouse model for noninvasive imaging of HIF prolyl hydroxylase activity: assessment of an oral agent that stimulates er ythropoietin production. Proc Natl Acad Sci U S A 2006;103:105–10. Cecic I, Chan D A, Sutphin PD, et al. Oxygen sensiti vity of repor ter genes: implications for preclinical imaging of tumor hypoxia. Mol Imaging 2007;6:219–28. Kaelin WG Jr. The v on hippel-lindau tumor suppressor protein: an update. Methods Enzymol 2007;435:371–83. Zhong H, De Marzo AM, Laughner E, et al. Ov erexpression of hypoxia-inducible factor 1α in common human cancers and their metastases. Cancer Res 1999;59:5830–5. Moeller BJ, Cao Y, Li CY, Dewhirst MW. Radiation activates HIF-1 to regulate vascular radiosensitivity in tumors: role of reoxygenation, free radicals, and stress g ranules. Cancer Cell 2004;5:429–41. Braun RD , Lanzen JL, Sn yder SA, De whirst MW. Comparison of tumor and nor mal tissue o xygen tension measurements using OxyLite or microelectrodes in rodents. Am J Ph ysiol Hear t Circ Physiol 2001;280:H2533–44. Dewhirst MW, Klitzman B, Braun RD, et al. Review of methods used to study oxygen transport at the microcirculatory level. Int J Cancer 2000;90:237–55. Schneiderman G, Goldstick TK. Oxygen electrode design criteria and performance characteristics: recessed cathode. J Appl Ph ysiol 1978;45:145–54. Zijlstra WG, Buursma A, Meeuwsen-van der Roest WP. Absorption spectra of human fetal and adult o xyhemoglobin, de-o xyhemoglobin, carbo xyhemoglobin, and methemo globin. Clin Chem 1991;37:1633–8. Severinghaus JW. Simple, accurate equations for human b lood O2 dissociation computations. J Appl Physiol 1979;46:599–602. Leow MK-S. Conf iguration of the hemo globin o xygen dissociation curve dem ystified: a basic mathematical proof for medical and

Imaging Hypoxia

41. 42. 43. 44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58. 59. 60.

61.

biological sciences under graduates. Adv Ph ysiol Educ 2007; 31:198–201. Ganong WF. Review of medical ph ysiology. 13th ed. Norw alk (CT): Appleton & Lange; 1987. Grippi MA. Pulmonar y pathophysiology. Philadelphia (PA): Lippincott; 1995. Webster JG, Clark JW . Medical instr umentation: application and design. 3rd ed. New York: Wiley; 1998. Patterson MS, Wilson BC, Wyman DR. The propagation of optical radiation in tissue. II: optical proper ties of tissues and resulting fluence distributions. Lasers Med Sci 1991;6:379–90. Pfefer TJ, Matchette LS, Bennett CL, et al. Reflectance-based determination of optical properties in highly attenuating tissue. J Biomed Opt 2003;8:206–15. Yang Y, Celmer EJ , K outcher J A, Alfano RR. DN A and protein changes caused by disease in human breast tissues probed b y the Kubelka–Munk Spectral Functional. Photochem Photobiol 2002; 75:627–32. Amelink A, Sterenborg HJ, Bard MP, Burgers SA. In vivo measurement of the local optical proper ties of tissue b y use of dif ferential pathlength spectroscopy. Opt Lett 2004;29:1087–9. Mourant JR, Bigio IJ, Jack DA, et al. Measuring absorption coefficients in small volumes of highly scattering media: source-detector separations for which path lengths do not depend on scattering proper ties. Appl Opt 1997;36:5655–61. Chance B, Leigh JS, Miyak e H, et al. Comparison of time-resolv ed and -unresolved measurements of deoxyhemoglobin in brain. Proc Natl Acad Sci U S A 1988;85:4971–5. Farrell TJ, Patterson MS, Wilson B. A diffusion theory model of spatially resolved, steady-state diffuse reflectance for the noninvasive determination of tissue optical properties in vivo. Med Phys 1992; 19:879–88. Zonios G, P erelman LT, Backman V, et al. Dif fuse reflectance spectroscopy of human adenomatous colon pol yps in vi vo. Appl Opt 1999;38:6628–37. Ghosh N, Mohanty SK, Majumder SK, Gupta PK. Measurement of optical transport properties of normal and malignant human breast tissue. Appl Opt 2001;40:176–84. Finlay JC, F oster TH. Hemo globin o xygen saturations in phantoms and in vivo from measurements of steady-state diffuse reflectance at a single, shor t source-detector separation. Med Ph ys 2004; 31:1949–59. Cerussi A, Hsiang D, Shah N, et al. Predicting response to breast cancer neoadjuvant chemotherapy using diffuse optical spectroscopy. Proc Natl Acad Sci U S A 2007;104:4014–9. Shah N, Gibbs J , Wolverton D, et al. Combined dif fuse optical spectroscopy and contrast-enhanced magnetic resonance imaging for monitoring breast cancer neoadjuv ant chemotherapy: a case study. J Biomed Opt 2005;10:051503. Palmer GM, Ramanujam N . Monte Carlo-based in verse model for calculating tissue optical proper ties. Part I: theor y and validation on synthetic phantoms. Appl Opt 2006;45:1062–71. Thueler P, Char vet I, Be vilacqua F, et al. In vi vo endoscopic tissue diagnostics based on spectroscopic absor ption, scattering, and phase function properties. J Biomed Opt 2003;8:495–503. Pittman R, Duling B. Measurement of percent o xyhemoglobin in the microcirculation. J Appl Physiol 1975;38:321–7. Algire G. Adaptation of transparent-chamber technique to mouse. J Natl Cancer Inst 1943;4:1–11. Yuan F, Salehi HA, Boucher Y, et al. Vascular permeability and microcirculation of gliomas and mammar y carcinomas transplanted in rat and mouse cranial windows. Cancer Res 1994;54:4564–8. Shan S, Sorg B, Dewhirst MW. A novel rodent mammary window of orthotopic breast cancer for intravital microscopy. Microvasc Res 2003;65:109–17.

777

62. Shonat RD, Wachman ES, Niu W, et al. Near-simultaneous hemoglobin saturation and o xygen tension maps in mouse brain using an AOTF microscope. Biophys J 1997;73:1223–31. 63. Sorg BS, Moeller BJ, Donovan O, et al. Hyperspectral imaging of hemoglobin saturation in tumor micro vasculature and tumor hypoxia development. J Biomed Opt 2005;10:044004. 64. Styp-Rekowska B , Disassa NM, Re glin B , et al. An imaging spectroscopy approach for measurement of oxygen saturation and hematocrit during intra vital microscop y. Microcirculation 2007; 14:207–21. 65. Koizumi H, Yamamoto T, Maki A, et al. Optical topography: practical problems and new applications. Appl Opt 2003;42:3054–62. 66. Gibson AP, Hebden JC, Arridge SR. Recent advances in diffuse optical imaging. Phys Med Biol 2005;50:R1–43. 67. Jöbsis-vanderVliet FF. Discovery of the near-infrared window into the body and the earl y de velopment of near -infrared spectroscop y. J Biomed Opt 1999;4:392–6. 68. Boushel R, Langber g H, Olesen J , et al. Monitoring tissue o xygen availability with near infrared spectroscop y (NIRS) in health and disease. Scand J Med Sci Spor ts 2001;11:213–22. 69. Franceschini MA, F antini S, Thompson JH, et al. Hemodynamic evoked response of the sensorimotor cor tex measured nonin vasively with near -infrared op tical imaging. Psychoph ysiology 2003;40:548–60. 70. Wang L, Jacques SL, Zhao X. Continuous-w ave ultrasonic modulation of scattered laser light to image objects in turbid media. Opt Lett 1995;20:629. 71. Wang X, P ang Y, K u G, et al. Nonin vasive laser -induced photoacoustic tomography for structural and functional in vivo imaging of the brain. Nat Biotechnol 2003;21:803–6. 72. Fu D, Ye T, Matthews TE, et al. High-resolution in vivo imaging of blood vessels without labeling. Opt Lett 2007;32:2641–3. 73. Applegate BE, Izatt JA. Molecular imaging of endogenous and exogenous chromophores using ground state recovery pump-probe optical coherence tomography. Opt Express 2006;14:9142–55. 74. Chance B, Schoener B, Oshino R, et al. Oxidation-reduction ratio studies of mitochondria in freeze-trapped samples. N ADH and fla voprotein fluorescence signals. J Biol Chem 1979;254:4764–71. 75. Ramanujam N, Richards-Kortum R, Thomsen S, et al. Low temperature fluorescence imaging of freeze-trapped human cer vical tissues. Opt Express 2001;8:335–43. 76. Drezek R, Sokolov K, Utzinger U, et al. Understanding the contributions of NADH and collagen to cervical tissue fluorescence spectra: modeling, measurements, and implications. J Biomed Opt 2001;6:385–96. 77. Chance B, Nioka S, Warren W, Yurtsever G. Mitochondrial NADH as the bell wether of tissue O2 deli very. Adv Exp Med Biol 2005; 566:231–42. 78. Jöbsis FF. Noninvasive, infrared monitoring of cerebral and m yocardial o xygen suf ficiency and circulator y parameters. Science 1977;198:1264–7. 79. Wilson DF, Cer niglia GJ . Oxygenation of tumors as e valuated b y phosphorescence imaging. Adv Exp Med Biol 1994;345:539–47. 80. Wilson DF, Vinogradov S, Lo LW, Huang L. Oxygen dependent quenching of phosphorescence: a status report. Adv Exp Med Biol 1996;388:101–7. 81. Vikram DS, Zw eier JL, K uppusamy P . Methods for nonin vasive imaging of tissue h ypoxia. Antioxid Redo x Signal 2007; 9:1745–56. 82. Vinogradov SA, Grosul P, Rozhkov V, et al. Oxygen distributions in tissue measured b y phosphorescence quenching. Adv Exp Med Biol 2003;510:181–5. 83. Plant RL, Burns DH. Quantitative, depth-resolved imaging of oxygen concentration b y phosphorescence lifetime measurement. Appl Spectrosc 1993;47:1594–9.

778

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

84. Koo YE, Cao Y, Kopelman R, et al. Real-time measurements of dissolved oxygen inside li ve cells b y organically modif ied silicate fluorescent nanosensors. Anal Chem 2004;76:2498–505. 85. Apreleva SV, Wilson DF, Vinogradov SA. Tomographic imaging of oxygen b y phosphorescence lifetime. Appl Opt 2006; 45:8547–59. 86. Jerabek PA, Patrick TB, Kilbourn MR, et al. Synthesis and biodistribution of 18F-labeled fluoronitroimidazoles: potential in vivo markers of h ypoxic tissue. Int J Rad Appl Instr um [A] 1986; 37:599–605. 87. Ljungkvist AS, Bussink J, Kaanders JH, v an der Kogel AJ. Dynamics of tumor h ypoxia measured with bioreducti ve h ypoxic cell markers. Radiat Res 2007;167:127–45. 88. Grierson JR, Link JM, Mathis CA, et al. A radiosynthesis of fluorine18 fluoromisonidazole. J Nucl Med 1989;30:343–50. 89. Chang CW, Chou TK, Liu RS, et al. A robotic synthesis of [18F]fluoromisonidazole ([18F]FMISO). Appl Radiat Isot 2007;65:682–6. 90. Rasey JS, Nelson NJ, Chin L, et al. Characteristics of the binding of labeled fluoromisonidazole in cells in vitro. Radiat Res 1990; 122:301–8. 91. Gagel B, Reinar tz P, Dimar tino E, et al. pO(2) P olarography versus positron emission tomo graphy ([(18)F] fluoromisonidazole, [(18)F]-2-fluoro-2ʹ′-deoxyglucose). An appraisal of radiotherapeutically relevant hypoxia. Strahlenther Onkol 2004;180:616–22. 92. Rajendran JG, Krohn KA. Imaging h ypoxia and angio genesis in tumors. Radiol Clin North Am 2005;43:169–87. 93. Lee NY, Mechalakos JG, Nehmeh S, et al. Fluorine-18-labeled fluoromisonidazole positron emission and computed tomo graphyguided intensity-modulated radiotherap y for head and neck cancer: a feasibility study. Int J Radiat Oncol Biol Ph ys 2007. 94. Rajendran JG, Schw artz DL, O’Sulli van J , et al. Tumor h ypoxia imaging with [F-18] fluoromisonidazole positron emission tomography in head and neck cancer . Clin Cancer Res 2006; 12:5435–41. 95. Eschmann SM, P aulsen F, Reimold M, et al. Pro gnostic impact of hypoxia imaging with 18F-misonidazole PET in non-small cell lung cancer and head and neck cancer before radiotherapy. J Nucl Med 2005;46:253–60. 96. Rajendran JG, Mankoff DA, O’Sullivan F, et al. Hypoxia and glucose metabolism in malignant tumors: e valuation b y [18F]fluoromisonidazole and [18F]fluorodeo xyglucose positron emission tomography imaging. Clin Cancer Res 2004;10:2245–52. 97. Lawrentschuk N, Poon AM, Foo SS, et al. Assessing regional hypoxia in human renal tumours using 18F-fluoromisonidazole positron emission tomography. BJU Int 2005;96:540–6. 98. Bruehlmeier M, Roelck e U, Schubiger PA, Ametamey SM. Assessment of hypoxia and perfusion in human brain tumors using PET with 18F-fluoromisonidazole and 15O-H2O. J Nucl Med 2004; 45:1851–9. 99. Rischin D, Hicks RJ, Fisher R, et al. Prognostic significance of [18F]misonidazole positron emission tomo graphy-detected tumor hypoxia in patients with adv anced head and neck cancer randoml y assigned to chemoradiation with or without tirapazamine: a substudy of Trans-Tasman Radiation Oncolo gy Group Study 98.02. J Clin Oncol 2006;24:2098–104. 100. Evans SM, Hahn SM, Magarelli DP, Koch CJ. Hypoxic heterogeneity in human tumors: EF5 binding, vasculature, necrosis, and proliferation. Am J Clin Oncol 2001;24:467–72. 101. Dolbier WR Jr, Li AR, Koch CJ, et al. [18F]-EF5, a mark er for PET detection of hypoxia: synthesis of precursor and a new fluorination procedure. Appl Radiat Isot 2001;54:73–80. 102. Mannan RH, Soma yaji VV, Lee J , et al. Radioiodinated 1-(5-iodo-5-deoxy-β-D-arabinofuranosyl)-2-nitroimidazole (iodoazomycin arabinoside: IAZA): a no vel mark er of tissue hypoxia. J Nucl Med 1991;32:1764–70. 103. Riedl CC, Brader P , Zanzonico P, et al. Tumor hypoxia imaging in orthotopic liver tumors and peritoneal metastasis: a comparati ve

104.

105.

106.

107.

108.

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

119.

120.

121.

study featuring dynamic (18)F-MISO and (124)I-IAZG PET in the same study cohort. Eur J Nucl Med Mol Imaging 2007. Zanzonico P, O’Donoghue J, Chapman JD, et al. Iodine-124-labeled iodo-azomycin-galactoside imaging of tumor h ypoxia in mice with serial microPET scanning. Eur J Nucl Med Mol Imaging 2004;31:117–28. Sorger D, Patt M, Kumar P, et al. [18F]Fluoroazomycinarabinofuranoside (18F AZA) and [18F]Fluoromisonidazole (18FMISO): a comparative study of their selecti ve uptake in h ypoxic cells and PET imaging in e xperimental rat tumors. Nucl Med Biol 2003;30:317–26. Piert M, Machulla HJ , Picchio M, et al. Hypo xia-specific tumor imaging with 18F-fluoroazomycin arabinoside. J Nucl Med 2005; 46:106–13. Gronroos T, Bentzen L, Marjamaki P , et al. Comparison of the biodistribution of tw o h ypoxia mark ers [18F]FETNIM and [18F]FMISO in an e xperimental mammar y carcinoma. Eur J Nucl Med Mol Imaging 2004;31:513–20. Takahashi N, Fujibayashi Y, Yonekura Y, et al. Copper-62 ATSM as a hypoxic tissue tracer in m yocardial ischemia. Ann Nucl Med 2001;15:293–6. Mathias CJ, Margenau WH, Brodack JW, et al. A remote system for the synthesis of copper -62 labeled Cu(PTSM). Int J Rad Appl Instrum [A] 1991;42:317–20. Dehdashti F, Mintun MA, Le wis JS, et al. In vi vo assessment of tumor hypoxia in lung cancer with 60Cu-ATSM. Eur J Nucl Med Mol Imaging 2003;30:844–50. Dehdashti F , Grigsb y PW , Mintun MA, et al. Assessing tumor hypoxia in cervical cancer by positron emission tomography with 60Cu-ATSM: relationship to therapeutic response-a preliminar y report. Int J Radiat Oncol Biol Ph ys 2003;55:1233–8. Yuan H, Schroeder T, Bowsher JE, et al. Inter tumoral differences in hypoxia selectivity of the PET imaging agent 64Cu(II)-diacetylbis(N4-methylthiosemicarbazone). J Nucl Med 2006;47:989–98. O’Donoghue J A, Zanzonico P , Pugache v A, et al. Assessment of regional tumor h ypoxia using 18F-fluoromisonidazole and 64Cu(II)-diacetyl-bis(N4-methylthiosemicarbazone) positron emission tomo graphy: comparati ve study featuring microPET imaging, Po2 probe measurement, autoradio graphy, and fluorescent microscopy in the R3327-AT and FaDu rat tumor models. Int J Radiat Oncol Biol Phys 2005;61:1493–502. Van De WC, Versijpt J, Dierckx RA, et al. 99Tc(m) labelled HL91 versus computed tomogaphy and biopsy for the visualization of tumour recurrence of squamous head and neck carcinoma. Nucl Med Commun 2001;22:269–75. Li L, Yu J, Xing L, et al. Serial hypoxia imaging with 99mTc-HL91 SPECT to predict radiotherap y response in nonsmall cell lung cancer. Am J Clin Oncol 2006;29:628–33. Yao Z, Qu W, Zhou Y, et al. 99mTc-HL91 “hot spot” imaging of mice bearing human carcinoma b y gamma camera and the ef fects of tumor necrosis on imaging. Chin Med J (Engl) 2000;113:404–7. Wu YW, Lee CM, Huang YH, et al. 99mTc-HL91 is inferior to 201Tl in scintigraphic detection of chronic m yocardial ischaemia. Nucl Med Commun 2005;26:1119–23. Thulborn KR, Waterton JC, Matthe ws PM, Radda GK. Oxygenation dependence of the transv erse relaxation time of w ater protons in whole b lood at high f ield. Biochim Bioph ys Acta 1982;714: 265–70. Howe FA, Robinson SP, McIntyre DJ, et al. Issues in flo w and oxygenation dependent contrast (FLOOD) imaging of tumours. NMR Biomed 2001;14:497–506. Ogawa S, Lee TM, Kay AR, Tank DW. Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci U S A 1990;87:9868–72. Taylor NJ, Baddeley H, Goodchild KA, et al. BOLD MRI of human tumor o xygenation during carbo gen breathing. J Magn Reson Imaging 2001;14:156–63.

Imaging Hypoxia

122. Neeman M, Dafni H, Bukhari O , et al. In vi vo BOLD contrast MRI mapping of subcutaneous vascular function and maturation: v alidation b y intra vital microscop y. Magn Reson Med 2001; 45:887–98. 123. Dunn TJ, Braun RD, Rhemus WE, et al. The effects of hyperoxic and hypercarbic gases on tumour b lood flo w. Br J Cancer 1999; 80:117–26. 124. Robinson SP, Howe FA, Stubbs M, Grif fiths JR. Ef fects of nicotinamide and carbo gen on tumour o xygenation, blood flow, energetics and blood glucose levels. Br J Cancer 2000;82:2007–14. 125. Stubbs M, Robinson SP, Rodrigues LM, et al. The effects of host carbogen (95% o xygen/5% carbon dio xide) breathing on metabolic characteristics of Mor ris hepatoma 9618a. Br J Cancer 1998; 78:1449–56. 126. Hoskin PJ, Carnell DM, Taylor NJ, et al. Hypoxia in prostate cancer: correlation of BOLD-MRI with pimonidazole immunohistochemistry-initial obser vations. Int J Radiat Oncol Biol Ph ys 2007; 68:1065–71. 127. Elas M, Williams BB, Parasca A, et al. Quantitative tumor oxymetric images from 4D electron paramagnetic resonance imaging (EPRI): methodology and comparison with b lood o xygen leveldependent (BOLD) MRI. Magn Reson Med 2003;49:682–91. 128. Baudelet C, Ansiaux R, Jordan BF , et al. Ph ysiological noise in murine solid tumours using T2*-weighted gradient-echo imaging: a mark er of tumour acute h ypoxia? Ph ys Med Biol 2004;49: 3389–411. 129. Yu JX, Kodibagkar VD, Cui W, Mason RP. 19F: a versatile reporter for non-invasive physiology and pharmacology using magnetic resonance. Curr Med Chem 2005;12:819–48. 130. Hunjan S, Zhao D , Constantinescu A, et al. Tumor o ximetry: demonstration of an enhanced dynamic mapping procedure using fluorine-19 echo planar magnetic resonance imaging in the Dunning prostate R3327-AT1 rat tumor. Int J Radiat Oncol Biol Phys 2001;49:1097–108. 131. Mason RP, Antich PP, Babcock EE, et al. Non-in vasive determination of tumor oxygen tension and local variation with growth. Int J Radiat Oncol Biol Phys 1994;29:95–103. 132. Mason RP, Rodbumrung W, Antich PP. Hexafluorobenzene: a sensitive 19F NMR indicator of tumor o xygenation. NMR Biomed 1996;9:125–34. 133. Noth U, Rodrigues LM, Robinson SP, et al. In vi vo determination of tumor o xygenation during g rowth and in response to carbo gen breathing using 15C5-loaded alginate capsules as fluorine-19 magnetic resonance imaging oxygen sensors. Int J Radiat Oncol Biol Phys 2004;60:909–19. 134. Robinson SP, Griffiths JR. Current issues in the utility of 19F nuclear magnetic resonance methodologies for the assessment of tumour h ypoxia. Philos Trans R Soc Lond B Biol Sci 2004; 359:987–96. 135. Howe FA, Robinson SP, Griffiths JR. Modification of tumour perfusion and o xygenation monitored b y g radient recalled echo MRI and 31P MRS. NMR Biomed 1996;9:208–16. 136. Stubbs M. Application of magnetic resonance techniques for imaging tumour physiology. Acta Oncol 1999;38:845–53. 137. Tofts PS, Brix G, Buckle y DL, et al. Estimating kinetic parameters from dynamic contrast-enhanced T(1)-weighted MRI of a dif fusable tracer: standardized quantities and symbols. J Magn Reson Imaging 1999;10:223–32. 138. Ceelen W, Smeets P , Back es W, et al. Nonin vasive monitoring of radiotherapy-induced micro vascular changes using dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) in a colorectal tumor model. Int J Radiat Oncol Biol Ph ys 2006; 64:1188–96. 139. Cooper RA, Car rington BM, Loncaster J A, et al. Tumour oxygenation le vels cor relate with dynamic contrast-enhanced magnetic resonance imaging parameters in carcinoma of the cervix. Radiother Oncol 2000;57:53–59.

779

140. Egeland TA, Gaustad JV, Vestvik IK, et al. Assessment of fraction of radiobiologically hypoxic cells in human melanoma xenografts by dynamic contrast-enhanced MRI. Magn Reson Med 2006; 55:874–82. 141. Lyng H, Vorren AO, Sundfor K, et al. Assessment of tumor o xygenation in human cer vical carcinoma b y use of dynamic Gd-DTPA-enhanced MR imaging. J Magn Reson Imaging 2001; 14:750–6. 142. Gallez B, Baudelet C, Jordan BF. Assessment of tumor oxygenation by electron paramagnetic resonance: principles and applications. NMR Biomed 2004;17:240–62. 143. Subramanian S, Matsumoto K, Mitchell JB , Krishna MC. Radio frequency continuous-wave and time-domain EPR imaging and Overhauser-enhanced magnetic resonance imaging of small animals: instr umental de velopments and comparison of relati ve merits for functional imaging. NMR Biomed 2004;17:263–94. 144. Swartz HM. Using EPR to measure a critical but often unmeasured component of oxidative damage: oxygen. Antioxid Redox Signal 2004;6:677–86. 145. Ilangovan G, Li H, Zw eier JL, K uppusamy P. In vi vo measurement of tumor redox environment using EPR spectroscopy. Mol Cell Biochem 2002;234–235:393–8. 146. Ilangovan G, Bratasz A, Kuppusamy P. Non-invasive measurement of tumor oxygenation using embedded microparticulate EPR spin probe. Adv Exp Med Biol 2005;566:67–73. 147. O’Hara JA, Goda F, Demidenko E, Swartz HM. Effect on regrowth delay in a murine tumor of scheduling split-dose ir radiation based on direct pO2 measurements b y electron paramagnetic resonance oximetry. Radiat Res 1998;150:549–56. 148. Krishna MC, English S, Yamada K, et al. Overhauser enhanced magnetic resonance imaging for tumor o ximetry: core gistration of tumor anatomy and tissue o xygen concentration. Proc Natl Acad Sci U S A 2002;99:2216–21. 149. Devasahayam N, Subramanian S, Mur ugesan R, et al. Strate gies for improved temporal and spectral resolution in in vi vo o ximetric imaging using time-domain EPR. Magn Reson Med 2007; 57:776–83. 150. Mahy P , De Bast M, Gallez B , et al. In vi vo colocalization of 2-nitroimidazole EF5 fluorescence intensity and electron paramagnetic resonance o ximetry in mouse tumors. Radiother Oncol 2003;67:53–61. 151. Kuppusamy P, Shankar RA, Zweier JL. In vivo measurement of arterial and v enous oxygenation in the rat using 3D spectral-spatial electron paramagnetic resonance imaging. Ph ys Med Biol 1998;43:1837–44. 152. Ahmad R, Clymer B, Deng Y, et al. Optimization of data acquisition for EPR imaging. J Magn Reson 2006;179:263–72. 153. Hicks RJ , Rischin D , F isher R, et al. Utility of FMISO PET in advanced head and neck cancer treated with chemoradiation incorporating a hypoxia-targeting chemotherapy agent. Eur J Nucl Med Mol Imaging 2005;32:184–91. 154. Gagel B, Reinartz P, Demirel C, et al. [18F] fluoromisonidazole and [18F] fluorodeo xyglucose positron emission tomo graphy in response e valuation after chemo-/radiotherap y of non-small-cell lung cancer: a feasibility study. BMC Cancer 2006;6. 155. Rajendran JG, Wilson DC, Conrad EU , et al. [(18)F]FMISO and [(18)F]FDG PET imaging in soft tissue sarcomas: cor relation of hypoxia, metabolism and VEGF expression. Eur J Nucl Med Mol Imaging 2003;30:695–704. 156. Cher LM, Murone C, Lawrentschuk N, et al. Correlation of hypoxic cell fraction and angio genesis with glucose metabolic rate in gliomas using 18F-fluoromisonidazole, 18F-FDG PET , and immunohistochemical studies. J Nucl Med 2006;47:410–8. 157. Grosu AL, Souvatzoglou M, Roper B , et al. Hypo xia imaging with FAZA-PET and theoretical considerations with re gard to dose painting for individualization of radiotherapy in patients with head and neck cancer. Int J Radiat Oncol Biol Ph ys 2007;69:541–51.

780

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

158. Souvatzoglou M, Grosu AL, Roper B , et al. Tumour hypoxia imaging with [18F]FAZA PET in head and neck cancer patients: a pilot study. Eur J Nucl Med Mol Imaging 2007;34:1566–75. 159. Rasey JS, Hofstrand PD , Chin LK, Tewson TJ. Characterization of [18F]fluoroetanidazole, a ne w radiophar maceutical for detecting tumor hypoxia. J Nucl Med 1999;40:1072–9. 160. Yang DJ, Wallace S, Cherif A, et al. Development of F-18-labeled fluoroerythronitroimidazole as a PET agent for imaging tumor hypoxia. Radiology 1995;194:795–800. 161. Lehtio K, Eskola O, Viljianen T, et al. Imaging perfusion and hypoxia with PET to predict radiotherapy response in head-and-neck cancer. Int J Radiat Oncol Biol Ph ys 2004;59:971–82. 162. Reischl G, Dorrow DS, Cullinane C, et al. Imaging of tumor h ypoxia with [124I]IAZA in comparison with [18F]FMISO and [18F]FAZA—first small animal PET results. J Phar m Phar m Sci 2007;10:203–11. 163. Chao KS, Bosch WR, Mutic S, et al. A novel approach to overcome hypoxic tumor resistance: Cu-ATSM-guided intensity-modulated radiation therapy. Int J Radiat Oncol Biol Phys 2001;49:1171–82. 164. Laforest R, Dehdashti F , Le wis JS, Schw arz SW . Dosimetr y of 60/61/62/64Cu-ATSM: a h ypoxia imaging agent for PET . Eur J Nucl Med Mol Imaging 2005;32:764–70. 165. Fujibayashi Y, Taniuchi H, Yonekura Y, et al. Copper -62-ATSM: a new hypoxia imaging agent with high membrane permeability and low redox potential. J Nucl Med 1997;38:1155–60. 166. Lewis JS, McCar thy D W, McCar thy TJ, et al. Ev aluation of 64Cu-ATSM in vitro and in vivo in a hypoxic tumor model. J Nucl Med 1999;40:177–83. 167. Grigsby PW, Malyapa RS, Higashikubo R, et al. Comparison of molecular mark ers of h ypoxia and imaging with (60)Cu-A TSM in cancer of the uterine cer vix. Mol Imaging Biol 2007;9:278–83. 168. Parliament MB, Chapman JD, Urtasun RC, et al. Non-invasive assessment of human tumour h ypoxia with 123I-iodoazom ycin arabinoside: preliminar y repor t of a clinical study . Br J Cancer 1992; 65:90–5. 169. Kumar P, McQuarrie SA, Zhou A, et al. [131I]Iodoazom ycin arabinoside for lo w-dose-rate isotope radiotherap y radiolabeling, stability, long-ter m w hole-body clearance and radiation dosimetr y estimates in mice. Nucl Med Biol 2005;32:647–53. 170. Groshar D , McEw an AJ, P arliament MB , et al. Imaging tumor hypoxia and tumor perfusion. J Nucl Med 1993;34:885–8. 171. Chapman JD , Coia LR, Stobbe CC, et al. Prediction of tumour hypoxia and radioresistance with nuclear medicine mark ers. Br J Cancer Suppl 1996;27:S204–8. 172. Iyer RV, Kim E, Schneider RF, Chapman JD. A dual hypoxic marker technique for measuring oxygenation change within individual tumors. Br J Cancer 1998;78:163–9. 173. Okada RD, Johnson G III, Nguyen KN, et al. 99mTc-HL91. Effects of lo w flo w and h ypoxia on a ne w ischemia-a vid m yocardial imaging agent. Circulation 1997;95:1892–9. 174. Li L, Yu JM, Xing LG, et al. Hypoxic imaging with 99mTc-HL91 single photon emission computed tomo graphy in advanced nonsmall cell lung cancer. Chin Med J (Engl) 2006;119:1477–80. 175. Li L, Yu JM, Sun XD , et al. [Pro gnostic v alue of 99mTc-HL91 SPECT h ypoxia imaging in patients with adv anced NSCLC]. Zhonghua Zhong Liu Za Zhi 2007;29:127–30.

176. Cook GJ , Houston S, Bar rington SF , F ogelman I. Technetium99m-labeled HL91 to identify tumor hypoxic correlation with fluorine-18-FDG. J Nucl Med 1998;39:99–103. 177. Zheng YJ, F an W, Zhao C, et al. [Clinical application of 99mTcHL91 hypoxia imaging in nasophar yngeal carcinoma]. Ai Zheng 2006;25:378–81. 178. Koch CJ. Importance of antibody concentration in the assessment of cellular h ypoxia b y flo w c ytometry: EF5 and pimonidazole. Radiat Res 2008;169:677–88. 179. Mayer A, Hockel M, Vaupel P. Endogenous hypoxia markers: case not proven! Adv Exp Med Biol 2008;614:127–36. 180. Kappler M, Taubert H, Holzhausen HJ, et al. Immunohistochemical detection of HIF-1 alpha and CAIX in adv anced head-and-neck cancer: prognostic role and correlation with tumor markers and tumor o xygenation parameters. Strahlenther Onk ol 2008; 184(9):491. 181. Sun X, Russell J, Xing L, et al. Changes in tumor hypoxia induced by mild temperature h yperthermia as assessed b y dual-tracer immunohistochemistry. Radiother Oncol 2008;88:269–76. 182. Viola RJ, Provenzale JM, Li F, et al. In vi vo bioluminescence imaging monitoring of h ypoxia-inducible f actor 1alpha, a promoter that protects cells, in response to chemotherap y. AJR Am J Roentgenol 2008;191:1779–84. 183. Swanson KR, Chakrabor ty G, Wang CH, et al. Complementar y but distinct roles for MRI and 18F-fluoromisonidazole PET in the assessment of human glioblastomas. J Nucl Med 2009;50:36–44. 184. Spence AM, Muzi M, Sw anson KR, et al. Re gional h ypoxia in glioblastoma multifor me quantif ied with [ 18F] fluoromisonidazole positron emission tomo graphy before radiotherap y: cor relation with time to pro gression and sur vival. Clin Canc Res 2008; 14:2623–30. 185. Wong TZ, Lacy JL, P etry NA, et al. PET of h ypoxia and perfusion with 62Cu-ATSM and 62C-PTSM using a 62Zn/62Cu gernerator. Am J Roentgenol 2008;190:427–32. 186. Dietz DW, Dehdashti F, Grigsby PW, et al. Tumor hypoxia detected by positron emission tomography with 60Cu-ATSM as a predictor of response and sur vival in patients under going Neoadjuv ant chemoradiotherapy for rectal carcinoma: a pilot study . Dis Colon Rectum 2008;51:1641–8. 187. Komar G, Seppanen M, Esk ola O, et al. 18F-EF5: a new PET tracer for imaging hypoxia in head and neck cancer . J Nucl Med 2008; 48:1944–51. 188. Mahy P , Geets X, Lonneux M, et al. Deter mination of tumour hypoxia with [ 18F]EF3 in patients with head and neck tumours: a phase I study to assess the tracer phar macokinetics, biodistribution and metabolism. Eur J Med Mol Imaging 2008;35:1282–9. 189. Matsumoto S, Hy odo F, Subramanian S, et al. Lo w-field paramagnetic resonance imaging of tumor o xygenation and gl ycolytic activity in mice. J Clin Invest 2008;118:1965–73. 190. Manzoor AA, Schroeder T, De whirst MW. One-stop-shop tumor imaging: buy h ypoxia, get lactate free. J Clin In vest 2008; 118:1616–9.

47 MOLECULAR IMAGING INTERACTIONS

OF

PROTEIN–PROTEIN

TARIK F. MASSOUD, MD, PHD, RAMASAMY PAULMURUGAN, PHD, PRITHA RAY, PHD, ABHIJIT DE, PHD, CARMEL CHAN, PHD, HUA FAN-MINOGUE, PHD, AND SANJIV S. GAMBHIR, MD, PHD

In the last few years, there has been a veritable explosion in the f ield of repor ter gene imaging, with the aim of determining the location(s), duration, and extent of gene expression within living subjects. An important application of this is in molecular imaging of interacting protein par tners, an area that could pave the way to functional proteomics in living animals and provide a tool for whole-body evaluation of new phar maceuticals tar geted to modulate protein–protein interactions (PPIs). We re view the three general methods currently available for imaging PPIs in living subjects using reporter genes, namel y, a modif ied mammalian tw o-hybrid system, a bioluminescence resonance ener gy transfer (BRET) system, and split repor ter protein complementation and reconstitution strategies.

SIGNIFICANCE OF PROTEIN–PROTEIN INTERACTIONS Proteins perfor m cellular functions primaril y as components of complexes. We now fully appreciate that the cell is not a simple aqueous solution but instead a dense gel of interacting proteins forming the basis of phenomena at almost e very le vel of cell function, in the str ucture of subcellular organelles, the transport machinery across the various biological membranes, packaging of chromatin, the netw ork of submembrane f ilaments, muscle contraction, signal transduction, and re gulation of gene expression to name a few.1,2 Other interacting protein complexes w ork as components of cellular machines, such as ribosomes that read genetic infor mation and synthesize proteins. Indeed , a frequent theme per vading biological investigation is that the great majority of proteins generally operate as constituents of comple xes that contain other macromolecules to car ry out specif ic

biological functions and that netw orks of interactions (interactomes) connect multiple, dif ferent cellular processes.3 Protein–protein interactions (PPIs) ha ve been the object of intense research for many years because of their importance in de velopment and disease. Man y human diseases can be traced to aberrant PPIs either through the loss of an essential interaction or through the for mation of an abnormal protein complex at an inappropriate time or location involving endogenous proteins, proteins from pathogens, or both. 4 A meticulous characterization of PPIs is necessar y for a thorough understanding of cell function. This characterization includes, but is b y no means limited to, the determination of the three-dimensional str uctures of these molecules. 5 Examples of the str uctures of a fe w protein interaction motifs involved in cell signaling provide an idea of the beauty and di versity of protein str ucture. In addition to structural considerations, the dynamic and ener getic proper ties of these systems re veal the exquisite subtlety in volved in biolo gical specif icity and control.5 Noninvasive molecular imaging of PPIs in li ving subjects of fers another dimension for in vestigating and characterizing all these impor tant intracellular events.

BIOPHYSICAL FEATURES OF PPIS Royer describes more full y the impor tant bioph ysical characteristics that govern PPIs.5 It emerges from studies of PPIs that nature has used in man y instances a strategy of mixing and matching of protein domains that specify 781

782

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

particular classes of PPIs b y modifying the amino acid sequence to confer specif icity for par ticular tar get proteins. Thus, specificity and, for example, the strength of signal transduction, are encoded b y the e xact amino acid sequence of the domain, and it is this relationship between sequence, structure, dynamics, energetics, and function that constitutes the fundamental issues for the biophysics of PPIs. It has to be said that thorough quantitative ther modynamic and kinetic studies of man y PPIs remain to be carried out. One reason for this lack of information is that priority has been gi ven over the past fe w years to the identification of these important proteins and to the determination of their three-dimensional str ucture, both of which are prerequisites to a full understanding of function. Nonetheless, understanding the mechanisms underlying t he f unction o f t hese p roteins l ikewise requires the characterization of their ener getic and dynamic properties, even if this presents certain practical experimental difficulties. Another level of re gulation of cell function brought about b y the interactions of these proteins is delicatel y balanced by the relati ve affinities of the v arious protein partners and the modulation of these af finities b y the binding of ligands, other proteins, nucleic acids, ions such as Ca++, and covalent modification, such as specific phosphorylation or acetylation reactions. Generall y, enough dif ferences are present within each protein sequence and str ucture to preclude inappropriate protein pairs from for ming. This implies that the relati ve affinities betw een these v arious protein par tners ha ve been tuned throughout e volution such that protein pairing occurs onl y between par ticular par tners and onl y under the appropriate conditions.

CATEGORIES OF PPIS There are four different types of protein–protein complexes, as outlined by Jones and Thornton6: (1) homodimeric proteins, (2) heterodimeric proteins, (3) enzyme-inhibitor complexes, and (4) antibody-protein complexes. Communication at the level of the organism or the cell requires the translation of ph ysical or chemical information signals (e g, the presence of a par ticular chemical factor, the chemical modification of existing substances, a change in pH or ion concentration) from one compartment to another. In large part, this communication relies on the specif ic interaction betw een particular heterologous proteins, in response to particular chemical or ph ysical signals. Homocomple xes are usually per manent and optimized (e g, the homodimer cytochrome c’), although an enor mous number of

enzymes, car rier proteins, scaf folding proteins, transcriptional re gulatory f actors, etc also function as homo-oligomers. Heterocomplexes can also ha ve such properties, or the y can be nonob ligatory, being made and brok en according to the en vironment or e xternal factors and in volve proteins that must also e xist independently. Trimers are relati vely rare compared with tetramers. The number of str uctures in higher multimeric states fall markedly, with the obvious exception of the viral coat proteins, w hich contain numbers of subunits (e g, up to 240). In nature, man y of the most important biological functions involve huge multicomponent complexes, for example, the ribosome. The different types of comple xes have different biological roles. Most homodimers are onl y observed in the multimeric state, and it is often impossible to separate them without denaturing the indi vidual monomer str uctures. Man y homodimers also ha ve tw ofold symmetr y. Many enzyme-inhibitor complexes are also strongly associated, y et these molecules also e xist independentl y as stable entities in solution. Similarl y, antibody-protein complexes and man y heterocomplexes are composed of molecules that have an independent existence. From an e volutionary perspective, the homodimers, enzyme-inhibitors, and the heterocomple xes ha ve presumably all e volved over time to optimize the interface to suit their biological function, whether this requires strong or w eaker binding. In contrast, the antibodyprotein interactions occur relati vely speaking “b y chance” and are selected principally by the strength of the binding constant, without being subject to e volutionary optimization over many years. Protein complexes having relatively strong interactions are those with a dissociation constant ( Kd) in the submicromolar range. Indeed , PPIs where the Kd is ≤ 10−8 M are roughl y considered to be high-affinity interactions.7 There are several fundamental properties that characterize a protein–protein interf ace. These proper ties dictate the principles governing the interactions involved in protein–protein reco gnition, namel y, size and shape, complementarity betw een surf aces, residue interf ace propensities, h ydrophobicity including h ydrogen bonding, se gmentation and secondar y str ucture, and conformational changes upon complex formation.6

ROLE OF PPIS IN SIGNAL TRANSDUCTION Owing to their biological importance and the growing interest in the mechanism of their function, the protein–protein recognition surf aces of some of the proteins in volved in intracellular signal transduction merit a separate brief

Molecular Imaging of Protein–Protein Interactions

mention. These have become of g reat interest due to their role in the control of a m yriad of cellular acti vities. Please see Royer5 for a more detailed account. Protein domains in volved in signal transduction include Src homolo gy domains 2 and 3, commonl y referred to as SH2 and SH3 domains. 8 The SH2 domains recognize tyrosine phosphor ylated proteins, especiall y autophosphorylated growth factor receptors. They are found in growth factor receptor-binding proteins that are present in signal transduction do wnstream of the receptors, but upstream of Ras; they have also been found in docking proteins. On the other hand, SH3 domains bind to proline rich sequences in their tar get protein par tners. SH3 domains are also found in proteins involved in signal transduction, such as the protein tyrosine kinases. They recognize pol yproline type II helical str uctures (PXXP motifs) in cell-signaling proteins. Unlike the SH2 domains, the phosphopeptide interacting domains, or PI/phosphotyrosine binding (PTB) domains, reco gnize the sequence that is N-ter minal, rather than C-ter minal, to the phosphotyrosine of the receptor.9 These adapter proteins, once tyrosine phosphorylated by the receptor, become capable of binding to the Grb2/Sos complex, thus coupling the receptor to the Ras signaling pathway. PDZ domains (also ter med GLFG repeats or DHR domains) were f irst identif ied as 90 amino acid se gments in three proteins, PSD-95, DlgA, and ZO-1 (hence the acronym, PDZ domain), which are all guanylate kinases.10 Proteins containing PDZ domains have been implicated in ion channel receptor clustering, receptor/enzyme coupling, and a variety of other protein associations. LIM domains are zinc f inger cysteine rich domains found in many homeodomain proteins involved in development and in nonhomeodomain proteins in volved in differentiation, associated with the c ytoskeleton or in cellular senescence.11 These domains bind to PDZ motifs, helix-loop-helix (bHLH) transcription f actors, as well as other LIM domains, and LIM-binding proteins and thus mediate PPIs implicated in impor tant cell functions. An incomplete list of other protein interaction motifs includes the pleckstrin homolo gy domain that binds to acidic domains in signal transduction proteins as w ell as to phosphoinositdes, 12 the WW domain, a semiconser ved region of 38 to 40 amino acids termed WW because of the two conserved tryptophan residues spaced 20 amino acids apart and w hich interacts with proteins in volved in cell signaling,13 the WSXWS amino acid motif in c ytokine receptors,14 and the WD repeat (WD-repeat-containing proteins are those that contain four or more copies of the

783

WD-repeat [ tryptophan-aspartate r epeat], a s equence motif approximately 31 amino acids long, w hich encodes a structural repeat).15

METHODS TO STUDY PPIS It is useful to generate multiple different classes of information about proteins. 16 For an y gi ven protein, these classes of knowledge would include the following: (1) the structural and sequence proper ties, (2) the e volutionary history and patter n of conser vation, (3) the e xpression profile, (4) the intracellular localization, (5) the for ms of post-translational regulation to which a protein is subject, and (6) the other cellular proteins with w hich the protein associates. All the f irst f ive points together contribute to the deter mination of the sixth, and deter mination of the profile of PPIs is an extremely important step toward the ultimate goal of identifying the functional significance of the activity of any given protein in a cell. 16 Techniques to provide classes of information regarding PPIs fall into three cate gories:16 First, there are techniques to identify e very possib le interacting set of proteins for a given protein of interest. Current research aims to isolate and str ucturally characterize all the proteins that exist in the cell. Importantly, PPIs are now considered to be so vital to cellular function that one of the first experiments performed on a protein may be a search for its interaction par tners.1 As of October 2008, 26,069 of 49,279 proteins in the Protein Data Bank w ere of known protein–protein comple xes; this being up from about 12,000 kno wn str uctures in April 2004, from a variety organisms, of assemb lies involving two or more protein chains. 17 Just ho w man y comple xes e xist in a particular proteome is not easy to deduce because of the different component types (e g, proteins, nucleic acids, nucleotides, metal ions) and the v arying life spans of the protein comple xes (e g, transient PPIs, such as those involved in signaling, and stab le interactions, such as in the ribosome). Until recentl y, the most comprehensi ve information about PPIs w as available for the y east proteome, consisting of approximately 6,200 proteins.17,18 In yeast, there are about nine protein par tners per protein, although not necessaril y all direct or interacting at the same time. The human proteome may be an order of magnitude more complex than the yeast cell.17,19 The second set of techniques is used in circumstances where interacting proteins have been defined, and the goal is to detail the biological function and impact of their interactions, that is, to establish physiological significance.16 In this case, it is essential to be ab le to study the interaction

784

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

under conditions that cor respond as closely as possible to the endogenous situation. In its cur rent role, nonin vasive strategies for molecular imaging of PPIs in living subjects fit within this second cate gory of techniques, and the advantages of these approaches are discussed belo w. Finally, there are techniques to de vise high-throughput methodologies to identify agents that modulate a kno wn and well-characterized interaction in desirable ways.20 Several technologies, grouped together under the ter m “proteomics” (a ter m introduced in 1995 b y Wasinger and colleagues21), have emerged with the common objecti ve of studying protein function at the scale of an entire pathway, a whole cell, or e ven a w hole organism. Proteomic anal yses encompass lar ge-scale studies of PPIs or comple xes to establish comprehensive protein interactomes, the global examination of protein e xpression prof iles and , more recently, of protein post-translational modifications.22 Many experimental techniques, for example, co-immunoprecipitation, bimolecular fluorescence complementation, fluorescence resonance ener gy transfer (FRET), label transfer , yeast two-hybrid screen, in vivo crosslinking of protein complexes using photoreacti ve amino acid analo gs, tandem affinity purif ication, chemical crosslinking, quantitati ve immunoprecipitation combined with knockdo wn, dual polarization interferometr y, protein–protein docking, static light scattering, matrix assisted laser desorption/ionization (MALDI) mass spectrometr y, strep-protein interaction experiments, surface plasmon resonance, fluorescence correlation spectroscopy (many of these techniques are reviewed else where2,23), ha ve been de veloped and studied using intact cells and cell extracts to study PPIs and to facilitate these proteomic endea vors. Each of these anal ytical systems has its own merits and demerits, as reviewed previously.23,24 Clearly, one aim of proteomics is to identify which proteins interact. Although the molecular imaging and study of individual PPIs (that have already been identified as such) might appear to fall outside the scope of proteomic endea vors, it is also impor tant to note their complementar y roles. Indeed, a prime challenge in the future is to conduct targeted studies of proteins of interest (including noninvasive molecular imaging anal ysis) while considering the lar ger context of w hole or ganismal function and con versely to carefull y validate systematic large-scale models of organismal function through individual test cases.16

task of adapting them so that signals can be nonin vasively detected from the exterior of living subjects upon the cellular or subcellular interaction of tw o proteins of interest. Only o ver the last 6 to 7 y ears has it been possib le to develop such methods as a result of the tr ue explosion in availability of noninvasive small animal imaging technologies and the rapidly expanding field of molecular imaging, allowing signal detection from deep tissues within a li ving subject (Table 1). We previously reviewed in detail the many advantages afforded by molecular imaging in li ving subjects (such as assessment of w hole-body phenomena, repeatability, functionality, and quantif ication).25 One subset of molecular imaging techniques comprises repor ter gene expression imaging. This represents an “indirect” imaging method in volving multiple components, entailing the use of a pretargeting molecule (an imaging reporter gene) that is subsequentl y activated upon occur rence of a specific molecular event. Following this, a molecular probe (a substrate or a ligand) specif ic for the activated pretargeting molecule (an enzyme or receptor) is often needed (but not for fluorescent repor ter proteins) and used to image its activation.26 An important feature of reporter gene imaging techniques is their par ticular versatility, which allows them to be adapted for imaging diverse PPIs in intact living subjects, as outlined belo w, and as also re viewed b y our group,24,27 and others recently.28–34 The ability to noninvasively image PPIs has important implications for a wide v ariety of biolo gical research endeavors, drug discovery, and molecular medicine. In particular, the visual representation, characterization, quantification, and timing of these biological processes in living subjects could create unprecedented opportunities to complement available in vitro or cell culture methodologies, in order (1) to accelerate the e valuation in living subjects of novel dr ugs that promote or inhibit acti ve homodimeric, heterodimeric, or multimeric protein assembly and (2) to characterize more full y kno wn PPIs (e g, the reasons for and the factors that drive their association) in the context of whole-body physiologically authentic environments.35

ADVANTAGES OF NONINVASIVE MOLECULAR IMAGING OF PPIS

It has been estimated that more than 50% of all protein interactions described in the literature ha ve been detected using the y east tw o-hybrid system. 36,37 In a tw o-hybrid assay, two proteins are expressed in yeast with one fused to a DN A-binding domain (BD) and the other fused to a transcription activation domain (AD). If the tw o proteins

The overall modif ication of e xisting in vitro and cell culture-based experimental assays to study PPIs in living small animal models of disease is dependent on the challenging

TECHNIQUES FOR NONINVASIVE REPORTER GENE IMAGING OF PPIS Modified Mammalian Two-Hybrid System

Molecular Imaging of Protein–Protein Interactions

785

Table 1. CHRONOLOGICAL DEVELOPMENTS OF MOLECULAR IMAGING STRATEGIES FOR DETECTION OF PPIS IN LIVING ANIMALS Year

Authors

Area of PPI Study

Imaging Assay

Reporter Used

Reference

2002

Ray et al.

Development of new assay

Two-hybrid system

Fluc

38

2002

Luker et al.

Development of new assay

Two-hybrid system

TK and Fluc

42

2002

Paulmurugan et al.

Development of new assay

Split reporter complementation & reconstitution

Fluc

52

2003

Luker et al.

PPI between P53 and large T antigen of SV40 virus

Two-hybrid system

TK and Fluc

44

2004

Paulmurugan et al.

Rapamycin modulation of FRB and FKBP12 PPI

Split reporter complementation

Rluc

77

2004

Massoud et al.

Homodimeric PPIs

Split reporter complementation

Rluc

35

2004

Kim et al.

Protein nuclear transport

Split reporter reconstitution

Rluc

65

2004

Luker et al.

Development and applications of new split reporter fragments

Split reporter complementation

Fluc

69

2005

Paulmurugan et al.

Development of self complementing split reporter fragments

Split reporter complementation

Fluc

24

2005

Paulmurugan et al.

Development of a fusion protein approach to image drug modulation of PPIs

Split reporter complementation

Rluc

24

2005

De and Gambhir

Development of a new assay

BRET

Rluc (& GFP variant)

47

2005

Kim et al.

Detection of stress-related corticosterone level increases

Split reporter reconstitution

Rluc

66

2006

Kanno et al.

Detection of protein release from mitochondria to cytosol

Split reporter reconstitution

Rluc

68

2007

De et al.

Further developments of an assay

BRET

Rluc (& GFP variant)

48

2007

Zhang et al.

Detection of Akt kinase activity

Split reporter complementation

Fluc

73

2007

Paulmurugan and Gambhir

Detection of multiprotein PPIs

Split reporter complementation

Fluc & Rluc

83

2007

Massoud et al.

Development of new split reporter

Split reporter complementation

TK

87

2008

Choi et al.

PPI between HIF-1 α and VHL

Split reporter complementation

Fluc

71

2008

Chan et al.

Detection of HSP90 inhibitors

Split reporter complementation

Rluc

81

2008

Chan et al.

Detection of protein phosphorylation mediated by protein kinases

Split reporter complementation

Fluc

72

2008

Pichler et al.

PPI between P53 and large T antigen of SV40 virus in a transgenic mouse

Two-hybrid system

Fluc

45

2008

Zhang et al.

Enhanced detection of Akt kinase activity

Split reporter complementation

Fluc

74

2009

De et al.

Further developments of an assay

BRET

Rluc (& GFP variant)

50

Akt = enzymes that are members of the serine/threonine-specific protein kinase family. Akt was originally identified as the oncogene in the transforming retrovirus, AKT8; BRET = bioluminescence resonance energy transfer; FKBP12 = FK506-binding protein; Fluc = firefly luciferase; FRB = FK506-binding protein (FKBP12) rapamycin-binding domain; GFP = green fluorescent protein; HIF-1 α = hypoxia-inducible factor-1 α; HSP90 = heat shock protein 90; PPI = protein–protein interaction; Rluc = Renilla luciferase; TK = herpes simplex virus type-1 thymidine kinase; VHL = von Hippel–Lindau tumor suppressor.

786

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

interact, they activate transcription of one or more reporter genes that are distal to binding sites for the BD . Studied first in y east, this classical tw o-hybrid system w as later adapted for mammalian cells, using dif ferent expression plasmids but similar assa y principles. Our g roup has adapted this system fur ther (F igure 1) using pBIND-Id (which contains the yeast GAL4 DNA-BD fused to the Id protein) and pACT-MyoD (which contains the herpes simplex vir us (HSV) VP16 AD fused to a se gment of the murine My oD protein) as the tw o hybrid proteins. 38 We had been using the GAL4-VP16 system to de velop novel strategies for tw o-step transcriptional acti vation of reporter gene expression for several years39 and therefore were readily able to adapt them for the two-hybrid strategy. Id and MyoD are the members of the helix-loop-helix family of nuclear proteins and are known to strongly interact in vivo during myogenic differentiation.40 To inducibly modulate the expression of these tw o hybrid proteins, we replaced the c ytomegalovirus (CMV) promoter of pBIND-Id and pA CT-MyoD with TNF-α inducible NFκB response elements. TNF- α is a pleiotropic cytokine secreted by lipopolysaccharide (LPS)-stimulated macrophages that induces a variety of cell specific events, including NFκB activation, and causes tumor necrosis in

vivo w hen injected in tumor -bearing mice. 41 Finally, we used pG5-luc (w hich contains f ive GAL4-binding sites upstream of a minimal TATA box, followed by the f irefly luciferase gene [ Fluc]) as a tw o-hybrid repor ter. In cells treated with TNF-α, activated endogenous NF κB moves to the nucleus and activates transcription of the two hybrid proteins: they in turn interact and activate transcription of the f irefly luciferase (Fluc) repor ter. We studied this system in detail in cell culture and in 293T cells implanted in mice using cooled char ge-coupled device (CCD) camera bioluminescence imaging while using TNF-α to modulate the system.38 Separately, but following a similar design, Luker and colleagues42 developed a tetrac ycline (or do xycycline)inducible, bidirectional v ector car rying in one direction the tumor suppressor P53 gene fused with Gal4 and in the other direction the T antigen (TAg) of SV 40 fused with VP16. Expression of the P53 and TAg hybrid proteins is induced by doxycycline, resulting in their interaction and for mation of a VP16-Gal4 transacti vator complex. This comple x binds to the Gal4 binding sequences in the promoter of a HSV1-sr39tk-GFP (green fluorescent p rotein) r eporter f usion p rotein. Tumor xenografts of HeLa cells stab ly expressing both the

A Promoter A Promoter B

gBl4 vp16

6000

x Inducible expression

y

Binding and transcriptional activation y

5000 4000

x

x

7000

y

3000

vp16

0h

GAL4

8h

2 0h

30 h

B

2000 7000 6000

vp16

5000 GAL4

4000

Firefly luciferase 5 3 GAL4 bs

Minimal promoter

3000

0h

8h

2 0h

30 h

2000

Figure 1. Imaging protein–protein interaction in living mice with a modified yeast two-hybrid strategy. Schematic diagram of the system for imaging the interaction of proteins X and Y. The first step involves the vectors pA-gal4-x and pB-vp16-y, which are used to drive transcription of gal4-x and vp16-y through use of promoters A and B. In the second step, the two fusion proteins GAL4-X and VP16-Y interact because of the specificity of protein X for protein Y. Subsequently, the GAL4-X-Y-VP16 binds to GAL4-binding sites (five GAL4-binding sites [bs] are available) on a reporter template. This leads to VP16-mediated transactivation of firefly luciferase reporter gene expression under the control of GAL4 response elements in a minimal promoter. Transcription of the firefly luciferase reporter gene leads to firefly luciferase protein, which, in turn, leads to a detectable visible light signal in the presence of the appropriate substrate (D-Luciferin), ATP, Mg2+, and oxygen. The NFκB promoter was used for either pA or pB and TNF-α-mediated induction. In vivo optical CCD imaging of mice carrying transiently transfected 293T cells for induction of the yeast two-hybrid system. All images shown are the visible light image superimposed on the optical CCD bioluminescence image with a scale in p/s/cm2/steridian (sr). Mice in top row were imaged after injection of D-Luciferin but with no TNF-α-mediated induction. Mice in bottom row were imaged after injection of D-Luciferin after TNF-α-mediated induction, showing marked gain in signal from the peritoneum over 30 h. Reproduced with permission from Ray P et al.38

Molecular Imaging of Protein–Protein Interactions

reporter plasmid and the bidirectional tw o-hybrid expression plasmid w ere implanted in li ving mice, and doxycycline-induced PPIs w ere imaged b y micro positron emission tomo graphy (microPET) at dif ferent time points. MicroPET imaging w as perfor med using a Fluorine-18 positron labeled analo g of penciclo vir, which is trapped in cells e xpressing the repor ter fusion (through action of the sr39 th ymidine kinase (TK) protein).43 The fusion protein showed correlated increase in expression as detected b y fluorescence microscop y (for the GFP component) and by microPET (for the TK component) when induced with increasing doses of do xycycline both in cells and in tumor x enografts of li ving mice. A subsequent study def ined quantitative (relative differences in amounts of interacting proteins and expression of repor ter gene) and kinetic (time inter val between induction of interacting proteins and detection of reporter activity in vivo) parameters pertaining to this system.44 Mapping of b iomolecular c omplexes w ithin t he cellular en vironment o ver biolo gically rele vant time scales w ould be desirab le in an attempt to more full y understanding the functional comple xity of the protein interactome. Therefore, ne wer approaches to imaging PPIs in vi vo w ould allo w the study of functional proteomics of human biology and disease within the context of li ving animals. Pichler and colleagues 45 described a universal transgenic reporter mouse strain that expresses Fluc under the regulatory control of a concatenated Gal4 promoter (Tg(G4F(±))). Using an adenovirus to deliver a fused binding-domain-acti vator chimera (Gal4BDVP16), induction of bioluminescence in Tg(G4F(±)) tissues of up to four orders of magnitude was observed in fibroblasts, liver, respiratory epithelia, muscle, and brain. The Tg(G4F( ±)) repor ter strain allo wed nonin vasive detection of viral infectivity, duration of the infection, as well as viral clearance in v arious tissues in vi vo. To demonstrate PPIs in li ve mice, the w ell-characterized interaction betw een tumor suppressor P53 (fused to Gal4BD) and large T antigen (TAg) (fused to VP16) was visualized in vi vo by using a tw o-hybrid strategy. Hepatocytes of Tg(G4F( ±)) mice transfected with P53/TAg demonstrated 48-fold g reater induction of Fluc expression in vi vo than noninteracting pairs. Fur thermore, to demonstrate the feasibility of monitoring e xperimental therapy with siRN A in vi vo, tar geted knockdo wn of P53 resulted in markedly reduced light output, whereas use of a control siRN A had no ef fect on protein interaction-dependent induction of Fluc. Thus, this highl y inducible Gal4/Fluc conditional repor ter strain should facilitate imaging studies of PPIs, signaling cascades,

787

viral dissemination, and therapy within the physiological context of the whole animal. The modif ied mammalian tw o-hybrid system is derived from the yeast two-hybrid system. As such, advantages of this system are that it is fairly simple, rapid and the strategies developed should be generally applicable to many protein partners as long as the interaction complex can lead to transactivation of the reporters of choice. The main limitation, however, is that these strate gies are con ventionally limited temporally (on the order of hours), and it can only be used to detect interacting proteins in the nucleus or that require nuclear translocation of h ybrid fusion proteins, thereby often limiting their analytical value.

Bioluminescence Resonance Energy Transfer Bioluminescence Resonance Energy Transfer (BRET) technology in volves the nonradioacti ve transfer of ener gy between donor and acceptor molecules by the Förster mechanism46 (Figure 2). The energy transfer primarily depends on the following: (1) an overlap between the emission and excitation spectra of the donor and acceptor molecules, respectively and (2) the pro ximity of < 100 Å betw een the donor and the acceptor entities. As FRET/BRET-based technologies are assuming more prominent roles in the f ield of studying PPIs, manuf acturers are continuall y de veloping new instr umentations for measuring FRET/BRET ratios, which are, in general, low-intensity signals. BRET measurements are usuall y obtained with a microplate reader equipped with specif ic filter sets for detection of the donor and acceptor emission peaks. This cellular assay has been applied to real-time imaging of cells, high-throughput screening of drugs, and small animal and plant models.We47 used the BRET2 system (Biosignal P ackard Montreal, Canada), which involves renilla luciferase ( Rluc) as a bioluminescent donor and mutant GFP2 as a fluorescent acceptor, adapted for e xpression in mammalian cells and characterized by a signif icantly red-shifted Stokes shift that emits transfer red energy at 508 nm. The resonance ener gy transfer from the reaction of the reconstr ucted Rluc protein with its substrate Deep Blue Coelenterazine (DBC) excites the GFP2 protein, when interaction of the two fused proteins Id and My oD or FKBP12 and FRB in the presence of a small molecule mediator (rapam ycin) occurs. We demonstrated the ability to detect signal from PPIs in cultured cells, as well as from the surface and deeper tissues of small living animals with implanted cells over expressing the fusion constructs (Figure 2). We recently showed that the BRET 2 assay sensitivity can be significantly improved by using Rluc mutants with

788

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A A No Interaction

Interaction Deep Blue C GFP 2

hRlUC

Deep Blue C

Rapamycin

FRB FKPB 12

Rapamycin

Energy transfer

B B With Rapamycin

Without Rapamycin P/s/cm2/sr 10e+04

P/s/cm2/sr 30000 25000

8e+04

20000

6e+04

15000 4e+04 10000 2e+04 GFP2 Emission

5000 DBC Emission

GFP2 Emission

DBC Emission

Figure 2. A, Schematic showing small molecule-mediated protein–protein interaction (PPI) leading to BRET. FKBP12 is fused to the N-terminus of RLUC donor protein, and a FRB is fused to the C-terminus of GFP2 acceptor protein. When the genes encoding for both of these two fusion proteins are expressed inside cells and rapamycin is present to mediate FRB-FKBP12 interaction, then resonance energy transfer occurs. This BRET signal can be detected by using the DBC substrate for RLUC. B, Detection of in vivo BRET2 signal from specific PPI. Dorsal view of a nude mouse implanted SQ with 5 × 106 293T cells either transiently transfected with pBRET 2 (L) or with pFKBP12-hRluc (LL) alone or cotransfected with pFKBP12-hRluc and pGFP 2-FRB (LR) in presence (right panel) or absence (left panel) of rapamycin. Mice that received the small molecule mediator drug rapamycin (5 mg/kg) were injected IP immediately after cell implantation. The scan was performed 7 h after drug administration. Mice were scanned for 5 min integration time using either GFP2 or DBC filters in succession by injecting with 25 µg DBC intravenously. Reproduced with permission from De and Gambhir.47

improved quantum ef ficiency and/or stability (e g, Rluc8 and RlucM) as a donor .48,49 To e xtend the time of light measurement, we also developed CLZ400 (also known as bisdeoxycoelenterazine) analo gs, sho wing that signal from our impro ved BRET 2 vector can be monitored for up to 6 hours. This approach, cur rently undergoing continued validation, should have important implications for the study of PPIs in cells maintained in their natural environment, par ticularly if it can be ef fectively applied for the evaluation of new pharmaceuticals. Further advances in this f ield have led us latel y to develop a highl y photon ef ficient, self-illuminating fusion p rotein c ombining a m utant r ed f luorescent protein (mOrange) and a mutant Rluc (Rluc8). 50 This new BRET fusion protein (BRET3) exhibits several fold improvement in light intensity in comparison to e xisting BRET fusion proteins. BRET 3 also e xhibits the most red-shifted light output (564 nm peak wavelength) of any reported bioluminescence protein that uses its natural coelenterazine substrate, a benefit of which can be demonstrated at v arious tissue depths in small animals.

The imaging utility of BRET3 at the single cell level was demonstrated using an intramolecular sensor incorporating tw o mT OR pathw ay proteins (FKBP12 and FRB) that dimerize only in the presence of rapamycin. With its increased photon intensity, red-shifted light output, and 3 good spectral resolution (~85 nm), BRET showed improved spatial and temporal resolution for measuring intracellular events in single cells and when using living small animal models. The de velopment of fur ther BRET3-based assays will allow imaging of PPIs using a single assay directly scalable from intact living cells to small living subjects, allo wing for potential accelerated drug discovery. The main adv antages of BRET w hen compared to FRET arise through its high sensiti vity for measurement of interactions avoiding the consequences of the required excitation of the donor with an e xternal light source. BRET assays show no photo b leaching or photoisomerization of the donor protein, no photodamage to cells, and no light scattering or autofluorescence from cells or microplates (when used in vitro), which can be caused by

Molecular Imaging of Protein–Protein Interactions

incident excitation light. In addition, one main advantage of BRET over FRET is the lack of emission arising from direct excitation of the acceptor . This reduction in background should permit detection of interacting proteins at much lower concentrations than it is possib le for FRET. Adapting BRET technolo gy for imaging PPIs in li ving subjects is currently being validated with a view to effective application in the evaluation of new pharmaceuticals.

Split Reporter Strategies In certain circumstances, functional proteins can assemble from one or more pol ypeptide fragments, with the occurrence and ef ficiency of assemb ly commandeered into a strate gy to measure real-time PPIs. Indeed , synthetically separated fragments of some enzymes can reconstitute functionally active protein particularly if the interaction is helped along by fusion of the enzyme fragments to strongly interacting moieties. Thus, in the splitprotein strate gy, a single repor ter protein/enzyme is cleaved into N-ter minal and C-ter minal segments; each segment is fused to one of tw o interacting proteins (X and Y). Ph ysical interactions betw een the tw o proteins X and Y reconstitute the functional repor ter protein leading to signal generation that can be measured. This split-protein strategy can work either through protein-fragment complementation assa ys (PCA) or intein-mediated reconstitution assa ys. In the for mer, noncovalent assembly of the repor ter protein occurs and in the latter case reconstitution of the full reporter protein occurs through covalent bonding. To date, several reporter proteins (e g, β-lactamase, β-galactosidase, ubiquitin, dihydrofolate reductase (DHFR), Fluc, Rluc, GFP) ha ve been adapted for split-protein strate gies b y finding various split sites for each repor ter protein. 51–54 If a full-length reporter can be imaged in living subjects, and this reporter can be appropriately split, then the split reporter assa y could possib ly be used to nonin vasively image PPIs. The appropriate split point should lead to two fragments that do not ha ve signif icant af finity for each other and yet when brought together (through interaction of the two proteins being studied for their mutual affinity) lead to detectable signal. The principle of the PCA strate gy for detecting PPIs was first demonstrated by Pelletier and colleagues using the enzyme DHFR,55 following inspiration from a 1994 ar ticle b y Johnsson and Varshavsky56 describing what they called the “ubiquitin split protein sensor .” In all PCAs, splitting a specif ic repor ter protein into tw o distinct fragments abolishes its function. Bringing the two fragments back to gether in a controlled manner

789

then restores functional acti vity57 (Figure 3). Selected fragments of man y proteins can associate to produce functional bimolecular comple xes58; the PCA system can, therefore, be generalized for a number of enzymes for detection of PPIs, examples including DHFR, glycinamide ribonucleotide (GAR) transfor mylase, aminoglycoside and h ygromycin B phosphotransferases, all reviewed by Michnick and colleagues,57 Escherichia coli TEM-1 β-lactamase,53,59 GFP and its variants,58 and the molecular imaging repor ters Fluc52 and Rluc. 54 Intein-Mediated Reconstitution Assays

These assa ys are based on the restoration of the full reporter protein through co valent bonding. Inteins ha ve been def ined as protein sequences embedded in-frame within a precursor protein sequence and e xcised during a maturation process termed protein splicing.60 Protein splicing is a post-translational e vent involving precise excision of the intein sequence and concomitant ligation of the flanking sequences (N- and C-exteins) by a normal peptide bond.61 Inteins are intervening DNA sequences that are not present in the mature gene product as a result of splicing at the protein le vel instead of at the RN A level. In 1998, it was discovered that the gene for the catal ytic α subunit of the replicative DNA polymerase III from Synechocystis sp. PCC6803 (Ssp) is encoded in tw o se gments dnaE-n and dnaE-c.62 Inteins represent a potentially powerful means of protein manipulation because tw o peptide bonds are broken and a ne w peptide bond is for med during the protein splicing process. Protein splicing is an e xceedingly complex self-catalyzed process that requires neither cof actors nor auxiliary enzymes. It requires no source of metabolic energy and therefore in volves bond rear rangements rather than bond clea vage followed by resynthesis. The elucidation of the reaction steps involved in protein splicing has made it possib le to modulate the reactions b y mutations and to design proteins that can under go highl y specif ic self-cleavage and protein ligation reactions. An intein can be viewed as an enzyme w hose substrate is the adjacent amino acid residues in the two exteins to which it is linked. Ozawa and colleagues 63 initially demonstrated that Fluc can be split betw een amino acid positions 437 and 438 and used with inteins (DnaE) in a reconstitution strategy to detect insulin-induced interaction of phosphorylated insulin receptor substrate-1 (IRS-1) and its tar get N-terminal SH2 domain of PI 3-kinase in a cell culture assay (Figure 4). Upon interaction of the two proteins, the two DnaE fragments are brought close enough to fold together and initiate splicing and linking of the tw o Fluc halves with a peptide bond. The Fluc gene has to be

790

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

AA Inactive X

y

Rfluc

Dfluc

Active X

BB

Un-induced

y

Induced 0 hrs

p/s/cm2/sr 20

A

C

B

D

A

C

B

15 10

D

3 104

5

18 hrs

p/s/cm2/sr 6 5 4 3

3 106

2 1

36 hrs

p/s/cm2/sr 20 15 3 106 10 5

A. B. C. D.

Mock Transfected cells Plasmid PD Plasmid PG Plasmids PD 1 PG

Figure 3. Schematic diagram of the split reporter-based complementation strategy used to optically image PPIs in living mice. The N-terminal half of firefly luciferase is attached to protein X through a short peptide FFAGYC, and the C-terminal half of firefly luciferase is connected to protein Y through the peptide CLKS. Interaction of proteins X and Y recovers Fluc activity through protein complementation (A, top panel). In vivo optical cooled CCD imaging of mice carrying transiently transfected 293T cells for the induction of the complementation-based split-luciferase system. All images shown are the visible light image superimposed on the optical CCD bioluminescence image with a scale in p/s/cm2/sr. Mice were imaged in a supine position after IP injection of D-Luciferin. (B, lower panel) A set of nude mice was repetitively imaged after SQ implantation of 293T cells transiently transfected with various plasmids as described in reference Paulmurugan et al.52 One group of mice was induced with TNF-α, and the other group was not induced. The images are from one representative mouse from each group immediately after implanting cells (0 h) and 18 and 36 h after TNF-α induction. The induced mouse showed higher Fluc signal at site D (where interacting proteins result in reporter protein complementation) when compared with the mouse not receiving TNF-α. The Fluc signal significantly increases after receiving TNF-α. Reproduced with permission from Paulmurugan R et al.52

rationally dissected so that each half of Fluc is inacti ve. After ligating the Fluc fragments to gether, the resultant mature Fluc recovers its bioluminescence activity.64 We subsequently reasoned that it ma y be possible to split Fluc and use split reporter complementation without inteins. We therefore studied PCA and intein-mediated reconstitution of Fluc fragments and found that a complementation strate gy w as as sensiti ve as the inteinmediated reconstitution strate gy under the conditions tested.52 Thus, we demonstrated for the first time the feasibility of imaging PPIs using split repor ters in small living animals. We studied a PCA based on split Fluc (cleaved into two fragments nFluc: residues 1–437; and cFluc: residues 438–550), using the interaction of Id and MyoD as test proteins 52 (Figure 3). Also recentl y, Kim and colleagues 65 developed a genetically encoded bioluminescence indicator for monitoring and imaging the nuclear traf ficking of target proteins in vitro and in vi vo. The principle is based on reconstitution of split fragments of Rluc b y protein splicing with a DnaE intein. A target cytosolic protein fused to the N-ter minal half of Rluc is e xpressed in mammalian cells. If the protein translocates into the nucleus, the Rluc moiety meets the C-ter minal half of Rluc and full-length Rluc is reconstituted b y protein splicing (F igure 5A). They demonstrated quantitati ve cell-based in vitro sensing and in vi vo imaging of ligand-induced translocation of androgen receptor (AR), which allowed high-throughput screening of exogenous and endogenous agonists and antagonists of this receptor (Figure 5B). The same authors used a similar approach to noninvasive molecular imaging of ph ysical and emotional stress by developing a method for detecting physiological increases in the endo genous cor ticosterone caused by exogenous and endogenous stress in living animals.66 They constr ucted a pair of geneticall y encoded indicators composed of cDNAs of glucocorticoid receptor (GR), split Rluc, and a DnaE intein. The GR fused with C-terminal halv es of Rluc and DnaE w as localized in the cytosol, whereas a fusion protein of N-terminal halves of Rluc and DnaE w as localized in the nucleus. If cor ticosterone induces GR translocation into the nucleus, the C-terminal Rluc meets the N-terminal one in the nucleus and full-length Rluc is reconstituted b y protein splicing with DnaE. Cell-based methods provided a quantitati ve bioluminescence assa y of the extent of GR translocation into the nucleus. The authors further demonstrated that the indicator enabled noninvasive imaging in mice subjected to two different types of imposed stress: a forced s wimming and metabolic perturbation caused b y 2-deo xy-D-glucose. This stress

Molecular Imaging of Protein–Protein Interactions

A

791

C-terminal domain (C_Luc)

N-terminal domain (N_Luc)

B B

Cys HS dna En

H N

dna Ec

A

O

B

HS N H NH2

O

C-Luc

O Asn

N-Luc

S dna En NH2

A

dna Ec

O

B

O S

N O

NH2

Figure 4. Illustrations of the split-luciferase system. A, Three-dimensional structure of firefly luciferase. Amino-terminal (cyan, 1–437 amino acids) and carboxy-terminal (yellow, 438–544 amino acids) halves of luciferase are shown. B, Principle of the split-luciferase system. N- and C-dnaE are connected to the amino- and carboxy-terminal halves of luciferase, respectively. Partner proteins A and B are linked to opposite ends of those dnaEs. Interactions between the two proteins accelerate the folding of N- and C-dnaE and protein splicing occurs. The amino- and carboxy-terminal halves of luciferase are linked together by a peptide bond to recover its bioluminescent activity. Reproduced with permission from Ozawa and Umezawa.64

indicator should be v aluable for screening phar macological compounds and in studying mechanisms of physiological stress. A similar cell culture approach for detection of nucleocytoplasmic transport was again adopted b y Kim and colleagues 67 for detecting phosphor ylation- or proteolysis-induced nuclear transpor ts of a tar get

protein. Two model proteins, signal transducer and activator of transcription 3 (ST AT3) and sterol-re gulatory element binding protein-2 (SREBP-2), were exemplified as phosphor ylation- and proteol ysis-induced nuclear transport, respectively. Each STAT3 or SREBP-2 is connected with C-terminal halves of RLuc and DnaE. If the protein translocates into the nucleus, the C-ter minal

792

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Nuclear membrane cytosol DHT DnaEn

Nucleus

DnaEn

Interaction between DnaEs

DnaEc Protein splicing

AR DnaEc

N-Luc

N-Luc

Luciferase (Luminescence)

C-Luc

C-Luc

Figure 5A. When the AR is bound to 5α-dihydrotestosterone (DHT), it translocates into the nucleus, and brings the N- and C-terminal halves of DnaEs close enough to fold correctly, thereby initiating protein splicing to link the concomitant Rluc halves with a peptide bond. The C-terminal half of split Rluc was located beforehand in the nucleus by a fused NLS. The cells containing this reconstituted Rluc allow one to monitor nuclear translocation of AR with its luminescence by coelenterazine as the substrate. Reproduced with permission from Umezawa et al. 2005.32

(1)

(2)

(3)

(4)

3.0 2.5 2.0 1.5 1.0

DMSO (control)

B

DHT

DHT and procymidone

photons/sec/cm2 (x104)

A A

DHT and PCB

photons/sec/cm2 (x104)

9.0 8.0 7.0 6.0 5.0 4.0 DMSO (control)

DHT

DHT and procymidone

DHT and PCB

Figure 5B. An inhibitory effect of procymidone or polychlorinated biphenyls (PCB) on the bioluminescence developed by DHT (10 mg/kg of body weight) in living mice. A, The inhibitory effect of chemicals on AR translocation into the nucleus in the mouse brain. The COS-7 cells transiently cotransfected with pcRDn-NLS and pcDRc-AR were implanted in the forebrain of the nude mice at a depth of 3 mm through a 1-mm burrhole. Of mouse groups 1–4, groups 1 and 2 were stimulated with 1% DMSO, whereas groups 3 and 4 were stimulated with procymidone (10 mg/kg body weight) and PCB (10 mg/kg of body weight), respectively. Two hours after the stimulation, mouse groups 2–4 were then stimulated with DHT (10 µg/kg of body weight). Two hours after DHT stimulation, the mice were imaged in 2-min intervals until reaching the maximum photon counts after intercerebral injection of coelenterazine (1.4 mg/kg of body weight). B, The average of photon counts from each implanted site in (A) (n = 3). The averages of three mice were (5.53 ± 0.53) × 104 (group 1), (7.68 ± 0.91) × 104 (group 2), (5.07 ± 0.23) × 104 (group 3), and (4.18 ± 0.55) × 104 (group 4) (p/s/cm2). Reproduced with permission from Kim et al. 2004.65

fragment of RLuc meets the N-ter minal fragment of RLuc and full-length RLuc is reconstituted b y protein splicing in the nucleus. The indicator with SREBP-2 enabled them to quantify the intracellular concentrations of cholesterol. The indicator with STAT3 quantif ied the extent of the nuclear transport induced by representative cytokines.

Kanno and colleagues68 also developed a genetically encoded bioluminescence indicator for monitoring the release of proteins from the mitochondria in living cells. The principle of this method is based on reconstitution of split Rluc fragments b y protein splicing with a DnaE intein. A target mitochondrial protein connected with an N-ter minal fragment of Rluc and an

Molecular Imaging of Protein–Protein Interactions

N-terminal fragment of DnaE is e xpressed in mammalian cells. If the tar get protein is released from the mitochondria toward the cytosol upon stimulation with a specif ic chemical, the N-ter minal Rluc meets the C-terminal Rluc connected with C-ter minal DnaE in the cytosol, and thereb y, the full-length Rluc is reconstituted by protein splicing. The extent of release of the target fusion protein was evaluated by measuring activities of the reconstituted Rluc. To test the feasibility of this m ethod, t he a uthors m onitored t he r elease o f Smac/DIABLO protein from mitochondria during apoptosis in li ving cells and mice. Their method allowed high-throughput screening of an apoptosisinducing reagent, staurosporine, and imaging of the Smac/DIABLO release in cells and in living mice. This rapid analysis may be used for screening and assa ying chemicals that w ould increase or inhibit the release of mitochondrial proteins in living cells and animals. The split-intein system generally suffers from slow kinetic response rates posing problems for quantitative interrogation of reversible biochemical reactions, druginduced protein associations, or shifts in equilibrium states of interacting proteins. The tr uly impor tant aspects of studying PPIs in living cells or animals lie in the ability to do so in real time. The inevitable delay in the ability to detect an interaction using the split-intein strategy can be attrib uted to the time required for the splicing reaction. Although this may not be a factor for slow reactions occurring over long-time frames, numerous dr ugs, chemicals, and natural ligands e xert their effects in seconds to minutes. The system also exhibits a high false-positive rate on account of the split-intein fragments being so small that at times it is belie ved they act merely as simple link er proteins, thus limiting the quantif ication of protein interactions PCAs Using Split Firefly Luciferase

The most commonl y used bioluminescence repor ter gene for research purposes has been the luciferase from the North American firefly (Photinus pyralis; Fluc). Fluc (61 kDa) catal yzes the transfor mation of its substrate D-Luciferin into oxyluciferin in a process dependent on ATP, Mg2+, and O2, leading to self emission of light from green to y ellow wavelengths (560–610 nm, peak emission at 562 nm). In 2002, w e demonstrated for the f irst time the feasibility of imaging PPIs using split repor ters in small li ving animals. 52 We studied a PCA based on split Fluc (clea ved into tw o fragments, nFluc: residues 1–437; and cFluc: residues 438–550), using the interaction of Id and MyoD as test proteins (Figure 3).

793

Subsequently, Luk er and colleagues 69 described a systematic tr uncation librar y yielding alter native complementary N- and C-fragments of Fluc (nFluc: residues 2–416; and cFluc: residues 398–550). These fragments were used to monitor rapam ycin-mediated interaction of rapam ycin-binding proteins FRB and FKBP12 (see below). We similarly used the Fluc fragments previously tested with Id and My oD to study rapam ycinmediated interactions and found the complementation to be too weak for optical imaging in li ving animals using the CCD camera (unpublished data). Further studies on repor ter complementation assays for imaging of PPIs in li ving subjects led us to use a combinatorial strate gy to identify a no vel split site for Fluc with impro ved imaging characteristics o ver previously published split sites.70 A combination of fragments with g reater absolute signal and near -zero backg round signal was achieved by screening 115 different combinations. The identified fragments were further characterized by using f ive different interacting protein par tners and an intramolecular folding strate gy (see below). Cell culture studies and imaging in living mice were performed to v alidate the ne w split sites. In addition, the signal generated by the newly identified combination of fragments (nFluc 398/cFluc 394) w as compared with different split-luciferase fragments cur rently in use for studying PPIs and w as shown to be mark edly superior with a lo wer self-complementation signal and equal or higher postinteraction absolute signal. This study also identified man y dif ferent combinations of nono verlapping and overlapping Fluc fragments that can be used for studying dif ferent cellular events such as subcellular localization of proteins, cell–cell fusion, and e valuating cell delivery vehicles, in addition to PPIs, both in cells and small living animals. The developed split Fluc system w as recently used to study the crucial role of tumor hypoxia in tumorigenesis.71 Under hypoxia, hypoxia-inducible factor-1 α (HIF-1 α) re gulates acti vation of genes promoting malignant pro gression. Under nor moxia, HIF-1 α is hydroxylated on prolines 402 and 564 and is targeted for ubiquitin-mediated degradation by interacting with the von Hippel–Lindau protein complex (pVHL). We developed a no vel method of studying the interaction between HIF-1 α and pVHL using the split Fluc complementation-based bioluminescence system in w hich HIF-1 α and pVHL are fused to amino-terminal and carboxy-terminal fragments of the luciferase, respecti vely. We demonstrate that hydroxylation-dependent interaction between the HIF-1 α and pVHL leads to complementation of the tw o luciferase fragments, resulting in

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

bioluminescence in vitro and in vi vo. Complementation-based bioluminescence is diminished when mutant pVHLs with decreased affinity for binding HIF-1 α are used (Figure 6). This method represents a new approach for studying interaction of proteins in volved in the regulation of protein degradation. In another recent application, protein phosphor ylation mediated by protein kinases was studied using a genetically encoded , generalizab le split Fluc-assisted complementation system. 72 This was developed for noninvasive monitoring of phosphor ylation events and ef ficacies of kinase inhibitors in cell culture and in small living subjects by optical bioluminescence imaging. The serine/threonine kinase Akt mediates mito genic and antiapoptotic responses that result from acti vation of multiple signaling cascades. It is considered a k ey determinant of tumor agg ressiveness and is a major tar get for anticancer drug development. An Akt sensor (AST) w as constructed to monitor Akt phosphor ylation and the effect of different PI-3K and Akt inhibitors (Figure 7A).

Specificity of AST was determined using a nonphosphorylable mutant sensor containing an alanine substitution (ASA). It w as found that the PI-3K inhibitor L Y294002 and Akt kinase inhibitor perifosine led to temporal- and dose-dependent increases in complemented Fluc acti vities in 293T human kidney cancer cells stably expressing AST (293T/AST) but not in 293T/ASA cells. Inhibition of endogenous Akt phosphorylation and kinase activities by perifosine also cor related with increase in complemented Fluc acti vities in 293T/AST cells but not in 293T/ASA cells. Treatment of nude mice bearing 293T/AST x enografts with perifosine led to a tw ofold increase in complemented Fluc activities compared to that of 293T/ASA x enografts (F igure 7B). This system was used to screen a small chemical library for novel modulators of Akt kinase activity. It is foreseen that this generalizable approach for nonin vasive monitoring of phosphorylation events will accelerate the disco very and validation of novel kinase inhibitors and modulators of phosphorylation events.

A A

BB 3 105 (photons/sec/cm2/sr)

794

400

200

3 101

300

100

15

10

5

0 WT

P564G

400

3 101

300

200

100

3 105 (photons/sec/cm2/sr)

D D

C C

20

10

0 WT VHL

F119S

Figure 6. Studying HIF-1 α and pVHL interaction in vivo using bioluminescence optical imaging. A, 293T cells cotransfected with VHL-carboxy-terminal firefly luciferase fragment (CLUC) and amino-terminal firefly luciferase fragment (NLUC)-HIF-1 α wild-type (WT) (amino acids 556–603) (left flank) or NLUC-HIF-1 α P564G (right flank) were implanted subcutaneosuly into nude mice. B, In vivo bioluminescence signals were quantified (p/s/cm2/sr), and the means were plotted. Error bars indicate ± 1 SD (n = 9 mice); p < .001. C, 293T cells cotransfected with NLUC-HIF-1 α (amino acids 556–603) and WT VHL-CLUC (right abdomen) or VHL F119S-CLUC (left abdomen) were implanted subcutaneously into nude mice. D, In vivo bioluminescence signals were quantified (p/s/cm2/sr), and the means were plotted. Error bars indicate ±1 SD (n = 8 mice); p < .001. Reproduced with permission from Choi CY et al.71

Molecular Imaging of Protein–Protein Interactions

AA (G4S)3

(G4S)3

PI-3K/Akt Inhibitors

P

795

PI-3K/Akt-P Activation hv 1 O2, Mg21 , ATP & D-Luciferin

AST

OP

OH Akt-P

Phosphorylable AKMT in AST: RKRDRLGTLGI

Akt-P

Non-phosphorylable AKMA in ASA: RKRDRLGALGI

CC 140

140

120

120 Norm. FL Activity (1 /2 S.E.)

Norm. FL Activity (1 /2 S.E.)

B B

100 80 60 40

ASA

20

AST

0

RKRDRLGTLGI RKRDRLGALGI Without PDGF

With PDGF *

100 80 60 40 20 0

0

15

30

45

60

ASA

AST

(LY294002) (mM) Figure 7A. A, Schematic diagram for Akt kinase sensor (AST). Inhibition of Akt kinase activity by PI-3K/Akt inhibitors leads to decreased phosphorylation of AST at the Akt kinase motif (AKMT) and interaction with the phosphothreonine-binding domain (FHA2). This leads to increased complementation of split FL fragments (N-FL and C-FL) and light production in the presence of the FL substrate D-Luciferin. On the other hand, activation of Akt kinase activity (by PI-3K/AKT-P) leads to phosphorylation of AST at the AKMT and increased interaction with FHA2, thus hinders complementation between N-FL/C-FL. The ASA sensor with a nonphosphorylable AKMA motif served as a negative control. B, Inhibition of Akt kinase activity led to increase in complemented FL activity in BT474 cells transiently transfected with AST. BT474 cells were transiently transfected with ASA or AST for 24 h in the presence of the PI-3K inhibitor LY294002 or carrier control prior to bioluminescence imaging of intact cells upon addition of D-Luciferin. Total flux (complemented FL activities) was normalized for transfection efficiency using RL activities, protein content, and to carrier control treated cells (100%). LY led to a dose-dependent increase in complemented FL activities in BT474 cells transiently transfected with AST compared to that of carrier control-treated cells. *p < .05 relative to carrier control treated cells. C, Activation of Akt kinase activity by platelet-derived growth factor led to decrease in complemented FL activity. 293T cells transiently transfected with ASA or AST for 24 h were treated with PDGF or carrier control for 30 min prior to analysis of complemented FL activity as described in 1(B). In 293T cells transiently transfected with AST, PDGF led to decrease in complemented FL activity. On the other hand, in 293T cells transiently transfected with ASA, PDGF did not lead to significant decrease in complemented FL activity. *p < .05 relative to carrier control treated cells. Reproduced with permission from Chan CT et al.72

Zhang and colleagues 73 have also described a ne w reporter molecule whose bioluminescence activity within live cells and in mice can be used to measure Akt activity. Akt acti vity in cultured cells and tumor x enografts w as monitored quantitatively and dynamicall y in response to activation or inhibition of receptor tyrosine kinase, inhibition of phosphoinositide 3-kinase, or direct inhibition of Akt (Figure 8A,B). The results pro vided unique insights into the phar macokinetics and phar macodynamics of agents that modulate Akt activity, revealing the usefulness of this repor ter for rapid dose and schedule optimization

in the dr ug de velopment process. Ha ving constr ucted a genetically engineered hybrid bioluminescent Akt reporter (BAR) molecule that repor ts on Akt serine/threonine kinase acti vity (containing an Akt consensus substrate peptide, consisting of a domain that binds phosphorylated amino acid residues (FHA2) flank ed by nFluc and cFluc reporter domains), the same authors subsequentl y described a modif ied v ersion of this repor ter molecule (myristoylated and palmito ylated bioluminescent Akt reporter [MyrPalm-BAR]), which is membrane bound and whose bioluminescence acti vity can be used to monitor

796

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A A

0 hr 2

15 hrs 2

2

1

2

1

4

3

AST

2

1

2

1

2

3 106

ASA

1

39 hrs

6 hrs

Normalized FL Activity (Treated/Untreated) (1 /2 S.E.)

B B

27 hrs

200 180 160 140 120 100 80 60 40

ASA

20

AST

0

0

10

20

30

40

50

Time Post Perifosine Treatment (Hours)

Figure 7B. Inhibition of Akt kinase activity by perifosine in living mice. A, The efficacy of perifosine in inhibition of Akt kinase activity in living mice was determined by bioluminescence imaging of nude mice bearing 293T/ASA (left) and 293T/AST (right) xenografts. Upon determination of baseline signals at time 0 h, mice were treated with 30 mg/kg of perifosine or carrier control and reimaged at 6, 15, 27, and 39 h upon IP injection of D-Luciferin. B, Max photons at each time point were normalized to that of carrier control treated mice and expressed as average normalized max photons ± SEM for each xenograft. *p < .05 relative to mice bearing 293T/ASA xenografts. Reproduced with permission from Chan CT et al.72

Akt activity at the cell membrane.74 This was based on the fact that Akt is recruited to the plasma membrane upon activation. Using changes in Akt acti vation status with small molecule inhibitors of Akt, they demonstrated that the membrane-tar geted Akt repor ter w as more sensiti ve and quantitative. In addition, inhibition of upstream signaling kinases such as epider mal g rowth f actor receptor and phosphatidylinositol 3-kinase acti vity resulted in changes in Akt activity that were quantitatively monitored by bioluminescence imaging. Based on these results, the authors proposed that the membrane-associated Akt

reporter ma y be better suited for high-throughput screening and identif ication of no vel compounds that modulate the Akt pathway. In a slight v ariation on the theme of imaging PPIs, w e identified different fragments of Fluc based on the cr ystal structure of Fluc; these split repor ter genes, w hich encode fragments distinctly different from those cur rently used for studying PPIs, can self-complement and pro vide Fluc enzyme activity in different cell lines in culture and in living mice.75 The comparison of the fragment complementationassociated reco very of Fluc acti vity with intact Fluc w as

Molecular Imaging of Protein–Protein Interactions

expressing nFluc (1–475) and reco vering signal, that the complementing fragments could be ef ficiently used for screening macromolecule delivery vehicles. These complementing fragments should be useful for many reporter-based assays including intracellular localization of proteins, studying cellular macromolecule deli very v ehicles, studying cell–cell fusions, and also developing intracellular phosphorylation sensors based on fragment complementation.

C-Luc

Substrate peptide

FHA2

P

FHA2

Phosphatase

C-Luc

N-Luc

Akt

L N-

Kinase OFF, Light ON

uc

Kinase ON, Light OFF

Figure 8A. The proposed mechanism of action for the BAR reporter involves Akt-dependent phosphorylation of the Aktpep domain (thick line), which results in its interaction with the FHA2 domain (right). In this form, the reporter has minimal bioluminescence activity. In the absence of Akt activity, the N-Luc and C-Luc domains reassociate, restoring bioluminescence activity (left). Reproduced with permission from Zhang et al. 2007.73

Before

After 10 8

20 mg/kg API-2 6 40 mg/kg API-2

30 mg/kg perifosine

4 2

Photon counts (106/sec/cm2/steradian)

Control

Max

797

Figure 8B. Tumor-bearing mice were treated with vehicle control (20% DMSO in PBS), API-2 (20 mg/kg or 40 mg/kg), or perifosine (30 mg/kg). Images of representative mice are shown before treatment, during maximal luciferase signal upon treatment (Max), and after treatment. Reproduced with permission from Zhang et al. 2007.73

estimated for dif ferent fragment combinations and ranged from 0.01 to 4% of the full Fluc acti vity. Using a cooled optical CCD camera, the analysis of Fluc fragment complementation in transiently transfected subcutaneous 293T cell implants in li ving mice sho wed signif icant detectab le enzyme activity upon injecting D-Luciferin, especially from the combinations of fragments identif ied (nFluc and cFluc are the N and C fragments of Fluc, respecti vely): nFluc (1–475)/cFluc (245–550), nFluc (1–475)/cFluc (265–550), and nFluc (1–475)/cFluc (300–550). The cFluc (265–550) fragment, upon expression with the nuclear localization signal (NLS) peptide of SV40, sho wed reduced enzyme activity when the cells were cotransfected with the nFluc (1–475) fragment e xpressed without NLS. We also pro ved in this study, b y deli vering TAT-cFluc (265–550) to cells stab ly

PCAs Using Split Renilla Luciferase

The enzyme Rluc (or the synthetically mutated humanized version, hRluc), is a 311-residue, 36 kDa monomeric bioluminescence imaging repor ter protein, being the smallest optical repor ter protein identif ied to date for studying PPIs in a PCA strate gy.54,76 This PCA strate gy, using N and C ter minal halves of split Rluc functions in both cell culture and in li ving animals and has been demonstrated with se veral dif ferent protein par tners. We used fragments generated b y splitting betw een residues 229 and 230 to study rapamycin-induced interaction of the human proteins FRB and FKBP12 77 (see below). Moreover, protein interaction betw een IRS-1 and the SH2 domain of PI3K in the insulin signaling pathw ay w as located in living mammalian cells using Rluc split between residues 91 and 92. 54 Rluc is capab le of generating a flash of b lue light (460–490 nm, peak emission at 482 nm) upon reaction with its substrate coelenterazine. One limitation associated with the use of Rluc is its relati vely rapid reaction kinetics, requiring earl y time-point measurements. 78 Nevertheless, this split reporter system appears highly suitable for studying PPIs in cells and in li ving animals owing to its optical bioluminescence nature and its signal that is amplifiable through an enzymatic process. Further, the complementation strategies based on Rluc fragments, with smaller fragment size than Fluc, have less hindrance with interacting protein par tners and w ork more ef ficiently with dif ferent imaging assa ys. Another clear advantage of using the split Rluc system when compared to the split Fluc is that the for mer’s enzymatic reaction is ATP independent and therefore could be used in specif ic situations where the PPI under study itself requires ATP, for e xample, the binding of Hsp90 to the co-chaperone protein p23 (see below). Unfortunately, several coelenterazines ha ve been found to be substrates for the ef flux transporter MDR1 P-glycoprotein, including coelenterazine f, h, and hcp.79 This raises some general concern for the indiscriminant use of coelenterazine and Rluc reporters in li ve cell assa ys and nonin vasive w holeanimal imaging. The photon output of the reporter can be

798

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

affected by changes in P-gl ycoprotein transpor t activity that alter substrate a vailability within the cells thereb y introducing signal ar tifacts not related to the biolo gical process under in vestigation, that is, PPIs. Fur thermore, coelenterazine cannot be used in experimentation involving transport across the blood–brain barrier because brain capillaries are rich in outw ardly directed P-glycoprotein, effectively e xcluding coelenterazine from the central nervous system. To date, the split Rluc system has been used in several molecular imaging applications, including PPIs, smallmolecule-induced PPIs, small molecule-mediated inhibition of PPIs, and protein homodimerization (see belo w). More recentl y, w e de veloped a no vel fusion protein approach for studying rapam ycin-mediated interaction of fused FRB and FKBP12 with either split hRluc or split enhanced GFP to achie ve a system with g reater sensitivity for detecting lo wer le vels of dr ug-mediated PPIs in vivo.80 These applications are described more fully below. Networks of protein interactions mediate cellular responses to environmental stimuli and direct the execution of m any d ifferent c ellular f unctional p athways. S mall molecules synthesized within cells or recr uited from the external en vironment mediate man y protein interactions. The study of small molecule-mediated interactions of proteins is impor tant to understand abnor mal signal transduction pathw ays in cancer and in dr ug de velopment and validation. In one study , w e used split hRluc protein fragment-assisted complementation to evaluate heterodimerization of the human proteins FRB and FKBP12 mediated by the small molecule rapamycin77 (Figure 9). The concentration of rapam ycin required for ef ficient dimerization and that of its competiti ve binder ascom ycin required for dimerization inhibition w ere studied in cell lines. The system w as dually modulated in cell culture at the transcription level, by controlling nuclear factor κB promoter/enhancer elements using tumor necrosis factor-α, and at the interaction level, by controlling the concentration of the dimerizer rapam ycin. The rapam ycin-mediated dimerization of FRB and FKBP12 also w as studied in living mice by locating, quantifying, and timing the hRluc complementation-based bioluminescence imaging signal using a cooled CCD camera. It was found that this split reporter system can be used to ef ficiently screen small molecule drugs that modulate PPIs and also to assess drugs in living animals. Both are essential steps in the preclinical evaluation of candidate phar maceutical agents tar geting PPIs, including signaling pathways in cancer cells. In another study adopting a different strategy, we evaluated the rapam ycin-mediated interaction of the human

No interaction FKBP12

FRB

N-hRLUC

Rapamycin

C-hRLUC

Interaction FRB

FKBP12

N-hRLUC C-hRLUC

p/sec/cm2/sr 2.5 2.0

0h

3 104 1.5 1.0

2.5 24 h

2.0 3 104 1.5 1.0 2.5 2.0

48 h

3 104 1.5

R2

R1

1.0

Figure 9. Schematic diagram of rapamycin-mediated synthetic Renilla luciferase (hRLUC) protein fragment-assisted complementation strategy. In this strategy, N-terminal and COOH-terminal portions of hRLUC fragments are attached to proteins X and Y, respectively, through a short peptide linker GGGGSGGGGS. The N and C portions of hRLUC fragments are closely approximated by the dimerization of proteins FRB and FKBP12 only in the presence of the small molecule rapamycin, and this, in turn, leads to recovered activity of the hRLUC protein. Optical CCD imaging of living mice carrying IV injected 293T cells transiently cotransfected with Nhrluc-FRB and FKBP12-Chrluc. The animals not receiving rapamycin showed only a mean background signal of 4 ± 1 × 103 p/s/cm2/sr at all of the time points studied. The animals receiving repeated injections of rapamycin emitted signals, originating from the region of the liver, that were threefold (mean, 1.6 × 104 p/s/cm2/sr) and fivefold (mean, 3.0 × 104 p/s/cm2/sr) higher than background (p < .05) at 24 and 48 h after the injection of rapamycin, respectively. (R −, animals not receiving rapamycin; R+, animals receiving rapamycin). Reproduced with permission from Paulmurugan R et al.77

Molecular Imaging of Protein–Protein Interactions

proteins FK506-binding protein (FKBP12), rapam ycinbinding domain (FRB), and FKBP12 by constructing a fusion of these proteins with a split-Rluc or a splitenhanced green fluorescent protein (split-EGFP) such that complementation of the repor ter fragments occurs in the presence of rapam ycin80 (Figure 10A). Dif ferent link er peptides in the fusion protein w ere evaluated for the ef ficient maintenance of complemented reporter activity. This system was studied in both cell culture and x enografts in living animals. We found that peptide link ers with two or four EAAAR amino acid repeats showed higher PPI-mediated signal with lo wer backg round signal compared with having no link er or link ers with amino acid sequences GGGGSGGGGS, A CGSLSCGSF, and A CGSLSCGSFACGSLSCGSF. A 9 ± 2-fold increase in signal intensity both in cell culture and in li ving mice was seen compared with a system that e xpresses both repor ter fragments and the interacting proteins separatel y (F igure 10B). In this fusion system, rapam ycin-induced heterodimerization of the FRB and FKBP12 moieties occur red rapidly even at very lower concentrations (0.00001 nmol/L) of rapamycin. For a similar fusion system employing split-EGFP, flow cytometry analysis showed significant level of rapamycin-

A

799

induced complementation. We also e valuated small molecule-mediated inhibition of PPIs. 81 Heat shock protein90α (Hsp90α)/p23 and Hsp90β/p23 interactions are crucial for proper folding of proteins in volved in cancer and neurode generative diseases. Small molecule Hsp90 inhibitors b lock Hsp90 α/ p23 and Hsp90 β/p23 interactions in par t b y pre venting ATP binding to Hsp90. The impor tance of isofor mselective Hsp90 α/p23 and Hsp90 β/p23 interactions in determining the sensitivity to Hsp90 was examined using 293T human kidne y cancer cells stab ly expressing split Rluc reporters (Figure 11A). Interactions between Hsp90α/p23 and Hsp90β/p23 in the presence of split Rluc reporters led to complementation of Rluc acti vity, which was determined b y bioluminescence imaging of intact cells in cell culture and li ving mice using a cooled CCD camera. The three geldanam ycin-based and se ven purinescaffold Hsp90 inhibitors led to dif ferent levels of inhibition of complemented Rluc activities (10–70%). However, there was no isofor m selectivity to either class of Hsp90 inhibitors in cell culture conditions. The most potent Hsp90 inhibitor, PU-H71, ho wever, led to a 60 and 30% decrease i n R luc a ctivity ( 14 h ) i n 2 93T x enografts

B

Figure 10A. A, Schematic diagram of a fusion protein–based approach for studying PPIs using a rapamycin-mediated complementation strategy with either split-RLuc or split-EGFP protein fragments. In this strategy, the NH2-terminal and COOH-terminal fragments of a reporter protein (RLuc or EGFP) are each attached to one of two interacting proteins (FRB and FKBP12) through a short peptide linker (GGGGSGGGGS). The two interacting proteins (FRB and FKBP12) are in turn attached with a different peptide linker that has the property of thermodynamically favoring a conformation such that the split reporter moieties of the fusion protein are kept away from each other. Ideally, this thermodynamic barrier is low enough that when given a suitable signal for PPI to occur (rapamycin in this example), the two interacting proteins cause conformational changes such that the split reporter moieties can interact and complement themselves. B, Schematic representation of different plasmid constructs made and used in this study. Shown are the components of the genes (N-rluc, NH2-terminal fragment of Renilla luciferase; C-rluc, COOH-terminal fragment of Renilla luciferase; N-egfp, NH2-terminal fragment of EGFP; C-egfp, COOH-terminal portion of EGFP; FRB and FKBP12 are the rapamycin-binding proteins). CMV is the early promoter/enhancer element. (G4S)2 is the amino acid sequence for the linkers between the reporter fragments and the interacting proteins. Reproduced with permission from Paulmurugan R et al.80

800

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Day 1

Without Rapamycin

Day 2

Day 3

Day 20

Day 4

A

p/sec/cm2/sr

B C

9.0 7.0 5.0 3.0

With Rapamycin

3 103

p/sec/cm2/sr 2.5 2.0 1.5 3 106 1.0 0.5 0

A: Mock B: pcDNA-Nrluc-FRB+pcDNA-FKPB12-Crluc C: pcDNA-Nrluc-FRB-E4-FKPB12-Crluc

Figure 10B. Optical imaging of living mice carrying two million SQ injected 293T cells stably transfected with pcDNA-Nrluc-FRB-E4-FKBP12-Crluc (site B), 4.5 million cells cotransfected with pcDNA-Nrluc-FRB + pcDNA-FKBP12-Crluc (site C), and mock-transfected cells (site A).The animals were imaged immediately (time 0) and at 24, 48, 96 h, and 20 d after injecting repeated doses of 25 µg rapamycin (IP) or DMSO. A significant increase in complemented Renilla luciferase signal was observed only from the group that had received repeated doses of rapamycin and only in the site implanted with cells stably transfected with the fusion protein–based Renilla luciferase complementation system. Reproduced with permission from Paulmurugan R et al.80

expressing Hsp90α/p23 and Hsp90β/p23 split reporters, respectively, relati ve to car rier control-treated mice (Figure 11B). Molecular imaging of isofor m-specific Hsp90α/p23 and Hsp90β/p23 interactions and ef ficacy of different classes of Hsp90 inhibitors in living subjects have been achie ved with a no vel geneticall y encoded repor ter gene strategy that should help in accelerating development of potent and isoform-selective Hsp90 inhibitors. Homodimeric protein interactions are potent re gulators of cellular functions, but are par ticularly challenging to study in vivo. We also used a split hRluc complementationbased bioluminescence assay to study homodimerization of herpes simple x vir us type 1 TK in mammalian cells and in living mice 35 (Figure 12). We quantif ied and imaged homodimerization of TK chimeras containing N-terminal (n-hRluc) or C-ter minal (c-hRluc) fragments of hRluc in the upstream and do wnstream positions, respectively (tail-to-head homodimer). This w as monitored using luminometry (68-fold increase, and w as significantly ( p < .01) abo ve backg round light emission) and b y CCD camera imaging of living mice implanted with ex vivo transfected 293T cells (2.7-fold increase and w as signif icantly (p < .01) above background light emission). We also made a mutant-TK to generate n-hRluc mutant TK and mutant TKc-hRluc by changing a single amino acid at position 318 from arginine to cysteine, a key site that has previously been reported to be essential for TK homo-dimerization, to support the specif icity of the hRluc complementation signal from TK homodimerization. Ex vivo substrate (8- 3H Penciclovir) accumulation assays in 293T cells expressing the TK protein chimeras sho wed acti ve TK enzyme. We also

devised an experimental strategy by constructing variant TK chimeras (possessing extra n-hRluc or c-hRluc “spacers”) to monitor incremental lack of association of the tail-to-head TK homodimer. Application of this potentially generalizable assay to screen for molecules that promote or disr upt ubiquitous homodimeric PPIs could serve not only as an invaluable tool to understand biological networks but could also be applied to drug discovery and validation in living subjects. As an of fshoot to our designing of strate gies for PPI imaging, we used the split Renilla system to help de vise newer approaches to high-throughput anal ysis of interactions betw een v arious hor mones and dr ugs with the estrogen receptor (ER). 82 These are crucial for accelerating the understanding of ER biolo gy and phar macology. Through careful anal yses of the cr ystal str uctures of the human ER (hER) ligand-binding domain (hER-LBD) in complex with dif ferent ligands, w e h ypothesized that the hER-LBD intramolecular folding patter n could be used to distinguish ER agonists from selecti ve ER modulators and pure antiestrogens. We therefore constr ucted and validated intramolecular folding sensors encoding various hER-LBD fusion proteins that could lead to split Rluc/Fluc repor ter complementation in the presence of the appropriate ligands (Figure 13A). A mutant hER-LBD with low affinity for circulating estradiol was also identif ied for imaging in li ving subjects. Cells stably expressing the intramolecular folding sensors e xpressing wild-type and mutant hER-LBD w ere used for imaging ligand-induced intramolecular folding in living mice (F igure 13B). This is the f irst hER-LBD intramolecular folding sensor suited for high-throughput quantitative analysis of interactions between hER with hor-

Molecular Imaging of Protein–Protein Interactions

A

C

801

B

D

Figure 11A. SRL-PFAC was sensitive and specific for indirect monitoring of Hsp90/p23 interactions. A, Schematic diagram for monitoring Hsp90/p23 interaction using the SRL-PFAC. The two interacting proteins p23 and Hsp90 were fused to the NRL (amino acids 1–229) and CRL (amino acids 230–311) portion of the RL through an EF and (G4S)2 peptide linker, respectively (top). In the presence of ATP, p23/Hsp90 interactions brought NRL and CRL in close proximity and led to complementation of RL enzyme activity and photon production in the presence of the substrate coelenterazine. Binding of Hsp90 inhibitors (I ) to Hsp90-CRL leads to a conformation change and prevents ATP from binding, thus diminishing the interaction between NRL-p23 and Hsp90-CRL and complementation of RL activity. B, Structure of geldanamycin-based Hsp90 inhibitors used in this study (left). Structure (top right) and Hsp90-binding affinity, SKBr3 cell growth inhibition, and IC50 values for Her2 degradation (bottom right) of purine-scaffold Hsp90 inhibitors used in this study. a–e, determined as described by Luker and Piwnica-Worms,31 Fields and Song,36 Ray et al.,38 Galarneau et al.53 Kaihara et al.,54 respectively, as displayed in Chan CT et al.81 C, Complementation of Hsp90/p23 split RL reporter was orientation specific. 293T cells were transiently cotransfected with plasmids expressing the Hsp90β1.4 and p23 fused to NRL and CRL in eight possible orientations. FL was transfected to control for transfection efficiency. RL and FL activities were determined by luminometer assays 24 h post-transfection and normalized for protein content. Cotransfection of NRL-p23/Hsp90β1.4-CRL led to RL activity that was significantly higher than that of pcDNA vector control–transfected cells (p < .05). 293T cells transfected with full-length RL were used as a positive control. Columns, mean; bars, SE. *p < .05; **p < .005. D, Specificity of SRL-PFAC for monitoring Hsp90β1.4/p23 interaction was determined using noninteracting protein partners (NRL-p23/MyoD-CRL and NRL-Id/Hsp90β1.4-CRL) and a p23 (F103A) mutant that does not interact with Hsp90(NRL-p23[F103A]/Hsp90β1.4-CRL). RL and FL activities were determined as described in (B). Columns, mean; bars, SE. *p < .05; **p < .005 relative to 293T cells transiently cotransfected with NRL-p23/Hsp90β1.4-CRL (normalized to 100%). p < .005 relative to 293T cells transiently transfected with pcDNA. Reproduced with permission from Chan CT et al.81

mones and drugs using cell lysates, intact cells, and molecular imaging of small living subjects. The strategies developed can also be extended to study and image other important protein intramolecular folding systems. In a recent interesting combined application of both split Fluc and split Rluc systems, we demonstrated the fea-

sibility of imaging ER-ligand mo dulated multiprotein interactions (human estro gen receptors [ER- α/β], TP53 tumor suppressor protein and the human equi valent of mouse doub le minute 2 [HDM2]). 83 The backg round impetus to this arises from knowledge that cancers develop when accumulated genetic defects cause cells to proliferate

802

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

TK Monomers

N-hrluc

C-hrluc

TK Homodimer

Protein-Protein Interaction

B Complementation

Active hrluc

Figure 11B. Disruption of Hsp90α /p23 and Hsp90β/p 23 interactions by Hsp90 inhibitors in living mice. 5 × 106 293T cells stably transfected with NRL(M185V)-p23/Hsp90α2.2-CRL (left) or NRL(M185V)-p23/Hsp90β2.2-CRL (right) were implanted SQ in the lower flanks of each female nude mouse to allow xenograft establishment for 2 weeks, respectively. A, Mice (n = 10) were imaged at 0 h to determine the RL activities in the implanted cells by optical bioluminescence imaging using a cooled CCD camera with an acquisition time of 3 min immediately after tail vein injection of 30 µg of coelenterazine. Mice (n = 5) were subsequently IP injected with 75 mg/kg of purine-scaffold Hsp90 inhibitor PU-H71 (n = 5) diluted in 6.7% DMSO in PBS (carrier control) IP in a final volume of 200 µL. Control mice (n = 5) were treated IP with equal volume of carrier control. RL activity was determined at 6 and 14 h post-treatment. Representative images from two mice from each treatment group at each time point are shown with the optical bioluminescence image superimposed on the visible light image. Representative images from two mice were shown here in the PU-H71 and carrier control–treated group. B, Quantitation of the maximum photons from site implanted with NRL(M185V)-p23/Hsp90α2.2-CRL (left) or NRL(M185V)-p23/Hsp90β2.2-CRL (right). Points, average of normalized maximum photons at 0 h; bars, SE. *, p < .05 relative to carrier control–treated mice; NS, p > .05. Reproduced with permission from Chan CT et al.81

unchecked and escape from programmed cell death. P53, HDM2, and ERs are a fe w among se veral prime tar gets that generate genetic abnormalities that lead to cancer. These proteins function independentl y and interact together by forming a ternary complex that regulates the expression level of dif ferent proteins required for nor mal cell death/g rowth. The adv antages of these tar gets o ver other proteins are that they can be controlled by small-molecule-ligands that bind to an y one of these three (P53/HDM2/ER) to modulate the functional role of all three proteins. The authors showed for the first time simultaneous interaction of these multiproteins and their modulation in response to dif ferent ER-ligands. Plasmids

Figure 12. Schematic diagram showing the protein-fragmentassisted complementation strategy using split synthetic Renilla luciferase (here abbreviated as hRLUC) to monitor the TK homodimeric protein–protein interaction. The N-terminal portion of hRLUC is attached to one TK monomer through a peptide linker, and the C-terminal portion of hRLUC is similarly attached to another TK monomer. Dimerization of the two TK monomers restores hRLUC activity through protein complementation and produces light in the presence of the substrate coelenterazine. Reproduced with permission from Massoud TF et al.35

expressing fusion-proteins nFluc-ER- α/β-chRluc, P53cFluc, and nhRluc-HDM2 w ere transientl y cotransfected in 293T cells, and complemented Fluc (ER/ P53-interaction) and Rluc ( P53/HDM2-interaction) acti vities w ere determined before and after e xposure to 13 dif ferent ligands that bind to ER. The results showed significant levels of modulation by ER-ligands for the Fluc signals (ER/P53interaction). There was no signif icant change in the Rluc signal ( P53/HDM2-interaction). Studies with both ER- α and β showed similar le vels of interaction with P53. The system with the ER/P53 interaction showed significant levels of re versibility upon remo val of the ligand. These results demonstrated that the interaction betw een ER and P53 was ligand-dependent. Studying this multiprotein interaction system with ligands specific to P53 or/to HDM2, in the presence and the absence of ER-ligands, may further identify the complexity behind the interactions between these three proteins. This was the first demonstration of multiprotein interactions studied with multiplesplit-reporters and this strate gy should be useful for also studying other similar types of comple x interactions. Of note, the split-protein strategies described above are based on absolute stereospecific and regiospecific requirements for comple x for mation among interacting sequences. Although no head-to-head comparison is available, this strate gy w ould therefore appear more

A

B N-rluc

C N-rluc

C-rluc

Molecular Imaging of Protein–Protein Interactions

N-rluc

C-rluc c rlu C-

Agonist

Antagonist

PartialAgonist/ Antagonist

Low Complementation

Complementation

No Complementation

Figure 13A. Schematic representation of the hypothetical model of ligand-induced intramolecular folding of ER that leads to split RLUC complementation. The N- and C-terminal fragments of split RLUC were fused to the N and C terminus, respectively, of the hERα of various lengths (amino acids 281–549 and 281–595). Binding of ER ligands to the intramolecular folding sensor (N-RLUC-hER-C-RLUC) induces different potential folding patterns in the LBD based on the type of ligand. This folding leads to split RLUC complementation for ER antagonist (B) (H12 and ligands are colored green), low complementation for ER agonist (A) (H12 and ligands are colored blue), and no complementation for partial ER agonist/antagonist (C) (H12 and ligands are colored gold) with the selective folding sensor. Even though the distance between the N- and C-RLUC fragments after binding with partial agonist (C) is smaller than that of agonists (B), this model depicts the importance of the orientations of the split RLUC fragments in complementation. The yellow spheres are hydrophobic amino acids located between helix 3 and helix 5 of LBD. Reproduced with permission from Paulmurugan and Gambhir.82

specific than the modified yeast two-hybrid system, which suf fers from man y f alse-positive outcomes, at least in its standard (nonimaging) laborator y use. Moreover, the split-protein strate gies can be used to image interacting proteins anywhere in the cell. PCAs Using Other Split Reporters and Other Reporter Complementation Strategies

Although not yet at a stage of application in living subjects, Kim and colleagues 84 have recently evaluated the Click beetle luciferase (CBluc) and its luminescence signal as a bioanal ytical index reporting the magnitude of a signal transduction of interest. CBluc is insensiti ve to pH, temperature, and heavy metals and emits a stable, highly tissue-transparent red light with luciferin in

803

physiological circumstances. They v alidated a singlemolecule-format complementation system of split CBluc to study signal-controlled PPIs. First, they generated 10 pairs of N- and C-ter minal fragments of CBluc to examine whether a signif icant recovery of the acti vity occurs through intramolecular complementation. The ligand binding domain of andro gen receptor (AR LBD) was connected to a functional peptide sequence through a fle xible link er. The fusion protein w as then sandwiched between the dissected N- and C-terminal fragments of CBluc. Androgen induces the association between AR LBD and a functional peptide and the subsequent complementation of N- and C-ter minal fragments of split CBluc inside the single-molecule-for mat probe, which restores the activities of CBluc. The examination about the split sites of CBluc re vealed that the dissection positions ne xt to the amino acids D412 and I439 can admit a stable recovery of CBluc activity through an intramolecular complementation. The ligand sensitivity and kinetics of the single molecular probe with split CBluc were studied in various cell lines and in different protein-peptide-binding models. The probe may be applicab le to de veloping biotherapeutic agents relevant to AR signaling and for screening adv erse chemicals that possib ly influence the signal transduction of proteins in li ving cells or animals, although the latter setting is yet to be verified. G-protein coupled receptors (GPCRs) are a v ersatile and ubiquitous family of membrane receptors that transmit extracellular signals to mammalian cells and constitute the most important class of drug targets. Yet, sensitive and specific methods are lacking that w ould allo w quantitati ve comparisons of pharmacologic properties of these receptors in physiological or pathological settings in live animals. von Degenfeld and colleagues 85 sought to overcome these limitations by employing low affinity, reversible β-galactosidase complementation (ie, a sensing system not based on split reporter technology, but instead based on the principle of lacZ intracistronic complementation, pre viously described by the same authors 86) to quantify GPCR acti vation via interaction with β-arrestin. A panel of cell lines was engineered e xpressing dif ferent GPCRs to gether with the reporter system. In vitro e valuation revealed highly sensitive, dynamic, and specif ic assessment of GPCR agonists and antagonists. F ollowing implantation of the cells into mice, it was possible for the f irst time to monitor phar macological GPCR acti vation and inhibition in their ph ysiological c ontext b y n oninvasive F luc b ioluminescence imaging in li ving animals. This technolo gy has unique advantages that ma y enab le no vel applications in the

804

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A.

B.

Figure 13B. Bioluminescence imaging of ER antagonist-induced intramolecular folding in a mouse model. A, Shown is optical CCD camera imaging of 293T cells stably expressing intramolecular folding sensors N-RLUC-hER281–549-C-RLUC and N-RLUC-mutant-hER281–549/G521TC-RLUC in living female nude mice before and after treatment with antagonist raloxifene (0.5 mg per mouse) and the corresponding quantitative graph. B, Similar imaging conducted by using the same sensors with the split FLUC fragment system (N-FLUC-hER281–549-C-FLUC and N-FLUC-mutant-hER281–549/G521T-C-FLUC). The site implanted with the cells expressing the intramolecular folding sensor with the mutant hER (G521T) shows a higher RLUC complementation signal after raloxifene treatment compared with that of the wild-type hER. Reproduced with permission from Paulmurugan and Gambhir.82

functional in vestigation of GPCR modulation in li ve animals in biological research and drug discovery. Importantly, there is a pressing need to develop better techniques for nonin vasive imaging of PPIs using split reporters. We recently described the molecular engineering rationale and constr uction of a no vel PET -based reporter (the her pes simplex vir us type 1-TK) split into two fragments betw een Thr-265 and Ala-266 after demonstration of preserved enzymatic activity (85.2%, as compared to intact TK) in a circularl y per muted variant based on this cleavage site.87 We used this split TK in a PCA to quantitatively measure PPIs in mammalian 293T cells using an in vitro [8-3H]Penciclovir cell uptake assay. We showed a greater extent of TK fragment complementation w hen using FRB/FKBP12 than Id/My oD as test proteins. We deter mined that coe xpression of nTK-FRB together with FKBP12-cTK ga ve the optimal orientation

of chimeras for evaluation in this PCA assay. We also demonstrated the use of this split TK in a PCA to quantitatively microPET image PPIs in 293T cells subcutaneously implanted in li ving mice. F or this, w e prepared 293T cells stab ly e xpressing nTK-FRB and FKBP12cTK in a single v ector. Cell uptak e studies using these stably transfected cells demonstrated that the competitive inhibitor ascom ycin (FK506) pre vented rapam ycininduced TK activity in a dose-dependent manner. Prior to imaging we also estab lished that adequate le vels of protein expression were present by Western blot. The designing of this novel split TK reporter and its application in an in vi vo PET-based PCA potentiall y allo ws for the f irst time a more precise fully quantitative and tomographic localization of c ytosolic or nuclear PPIs in preclinical small and lar ge animal models of disease then has been possible to date.

Molecular Imaging of Protein–Protein Interactions

FUTURE OUTLOOK AND CONCLUSIONS We ha ve re viewed the three general methods cur rently available for imaging PPIs in li ving subjects. Se veral areas of in vestigation are required to fur ther ref ine the use of the modif ied mammalian tw o-hybrid system for noninvasive imaging of PPIs. As well as further quantitative and kinetic evaluations (eg, characterizing the ability to follow interactions over time based on the half-life of the reporter protein(s) that are transactivated), studies are needed to optimize the choice of transactivator as well as the choice of promoters and le vels of fusion proteins. Mathematical models to correlate and study these effects are currently under development. The emission light of current BRET systems is still suboptimal in the conte xt of small animal imaging because of the shor t wavelength nature of the emission light and the strong attenuation of biolo gical tissue to photons of wavelength less than 600 nm. Therefore, for in vivo molecular imaging of PPIs, an optimized technique using resonance transfer methods w ould use a single chain biosensor (to f ix molar ratios of donor and acceptor) with a bioluminescence donor in conjunction with a red-shifted fluorophore. In our laborator y, mutation studies on Rluc have led to the development of several Rluc variants that exhibit significantly enhanced light output and stabili ty compared to the nati ve enzyme, as well as red-shifted light output, which yields a green-peaked emission spectrum. The combination of such luciferases with red fluorescent proteins is being investigated currently. Unlike fluorescence microscop y-based techniques, studies of the kinetics of PPIs, including anal ysis of complementation reversibility, are not possible at present, although this will be an area of future active investigation. These future experiments will also require assessment in several cell lines, as well as with a greater variety of protein partners of different sizes and interaction af finities (w eak transient to strong ob ligate), to establish the general widespread applicability of this technique. The oppor tunity to measure tw o dif ferent protein interactions at the same time b y spectrall y unmixing output colors will be useful in attempts to multiple x image protein interaction netw orks.88 Recent adv ances in processing of tw o color imaging no w allows for the total spectral deconvolution of multicolored bioluminescent i mages, a ssuming t he s pectra a re d ifferent enough to reliab ly calculate the contrib ution of each individual emitter within each detection windo w, based

805

on their published spectra. Simultaneous imaging of multiple interactions should allo w decon volution of complex protein interactions and , e ventually, protein interactomes. The high sensiti vity of these assa ys for detecting, locating, and quantifying PPIs, combined with the advantages of doing so in a living subject environment, should make them of potential v alue in man y areas of biological in vestigation and future clinical molecular medicine applications. Indeed , endpoints in molecular imaging of PPIs can be quantif ied and therefore are particularly useful for translational research. Ultimately, we foresee innovative molecular imaging tools, such as the one presented , enhancing our appreciation of entire biological pathway systems and their pharmacological re gulation and accelerating the achie vement of a “systems biolo gy” understanding of biolo gical complexity.89

REFERENCES 1. Mayer BJ. Protein-protein interactions in signalling cascades. Methods Mol Biol 2006;332:79–99. 2. Droit A, P oirier GG, Hunter JM. Experimental and bioinfor matic approaches for interrogating protein-protein interactions to determine protein function. J Mol Endocrinol 2005;34:263–80. 3. Cusick ME, Klitgord N, Vidal M, Hill DE. Interactome: gateway into systems biology. Hum Mol Genet 2005;14:R171–81. 4. Ryan DP , Matthe ws JM. Protein-protein interactions in human disease. Curr Opin Struct Biol 2005;15:441–6. 5. Royer C. Protein-protein interactions. 2004. Available from: http://www.biophysics.org/education/croyer.pdf 6. Jones S, Thornton JN. Principles of protein-protein interactions. Proc Natl Acad Sci U S A 1996;93:13–20. 7. Figeys D, McBroom LD, Moran MF. Mass spectrometry for the study of protein-protein interactions. Methods 2001;24:230–9. 8. Engen JR, Wales TE, Hochrein JM, et al. Structure and dynamic regulation of Src-f amily kinases. Cell Mol Life Sci 2008; 65:3058–73. 9. Uhlik MT, Temple B, Bencharit S, et al. Str uctural and e volutionary division of phosphotyrosine binding (PTB) domains. J Mol Biol 2005;345:1–20. 10. Jemth P, Gianni S. PDZ domains: folding and binding. Biochemistr y 2007;46:8701–8. 11. Grunewald TG, Butt E. The LIM and SH3 domain protein f amily: structural proteins or signal transducers or both? Mol Cancer 2008;7:31. 12. Lemmon MA. Pleckstrin homology (PH) domains and phosphoinositides. Biochem Soc Symp 2007;74:81–93. 13. Ilsley JL, Sudol M, Winder SJ. The WW domain: linking cell signalling to the membrane cytoskeleton. Cell Signal 2002;14:183–9. 14. Weidemann T, Höf inger S, Müller K, Auer M. Be yond dimerization: a membrane-dependent activation model for interleukin-4 receptor-mediated signaling. J Mol Biol 2007;366:1365–73. 15. Smith TF. Di versity of WD-repeat proteins. Subcell Biochem 2008;48:20–30. 16. Golemis E. Toward an understanding of protein interactions. In: Golemis, E, editor . Pr otein-protein interactions: a m olecular cloning manual. Woodbury NY: Cold Spring Harbor Laborator y Press; 2002, pp. 1–6.

806

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

17. Russell RB, Alber F, Aloy P, et al. A structural perspective on proteinprotein interactions. Curr Opin Struct Biol 2004;14:313–24. 18. Ho Y, Gruhler A, Heilbut A, et al. Systematic identification of protein complexes in Saccharom yces cere visiae b y mass spectrometr y. Nature 2002;415:180–3. 19. Abbott A. Proteomics: the society of proteins. Nature 2002;417:894–6. 20. Gentleman R, Huber W. Making the most of high-throughput proteininteraction data. Genome Biol 2007;8:112. 21. Wasinger VC, Cordwell SJ, Cerpa-Poljak A, et al. Progress with geneproduct mapping of the Mollicutes: Mycoplasma genitalium. Electrophoresis 1995;16:1090–4. 22. Colland F, Daviet L. Inte grating a functional proteomic approach into the target discovery process. Biochimie 2004;86:625–32. 23. Golemis E. Protein-protein Interactions: A Molecular Cloning Manual. Woodbury NY: Cold Spring Harbor Laborator y Press; 2002. 24. Paulmurugan R, Ray P, De A, et al. Imaging protein-protein interactions in living subjects. Trends Anal Chem 2005;24:446–58. 25. Massoud TF, Gambhir SS. Molecular imaging in living subjects: seeing fundamental biological processes in a new light. Genes Dev 2003;17:545–80. 26. Gambhir SS, Barrio JR, Herschman HR, Phelps ME. Assays for noninvasive imaging of repor ter gene e xpression. Nucl Med Biol 1999;26:481–90. 27. Massoud TF, P aulmurugan R, De A, et al. Repor ter gene imaging of protein-protein interactions in li ving subjects. Cur r Opin Biotechnol 2007;18:31–7. 28. Haberkorn U , Altmann A. Noninvasive imaging of protein-protein interactions in living organisms. Trends Biotechnol 2003;21:241–3. 29. Umezawa Y. Seeing w hat w as unseen: ne w anal ytical methods for molecular imaging. Chem Rec 2003;3:22–8. 30. Luker GD , Shar ma V, Piwnica-Worms D . Visualizing protein-protein interactions in living animals. Methods 2003;29:110–22. 31. Luker KE, Piwnica-Worms D. Optimizing luciferase protein fragment complementation for bioluminescent imaging of protein-protein interactions in li ve cells and animals. Methods Enzymol 2004;385:349–60. 32. Umezawa Y. Genetically encoded optical probes for imaging cellular signaling pathways. Biosens Bioelectron 2005;20:2504–11. 33. Ozawa T. Designing split repor ter proteins for anal ytical tools. Anal Chimica Acta 2006;556:58–68. 34. Villalobos V, Naik S, Piwnica-W orms D. Cur rent state of imaging protein-protein interactions in vi vo with geneticall y encoded reporters. Annu Rev Biomed Eng 2007;9:321–49. 35. Massoud TF, P aulmurugan R, Gambhir SS. Molecular imaging of homodimeric p rotein-protein i nteractions i n l iving s ubjects. FASEB J 2004;18:1105–7. 36. Fields S, Song O. A novel genetic system to detect protein-protein interactions. Nature 1989;340:245–6. 37. Fields S, Ster nglanz R. The two-hybrid system: an assa y for proteinprotein interactions. Trends Genet 1994;10:286–92. 38. Ray P, Pimenta H, Paulmurugan R, et al. Noninvasive quantitative imaging of protein-protein interactions in living subjects. Proc Natl Acad Sci U S A 2002;99:3105–10. 39. Zhang L, Adams JY, Billick E, et al. Molecular engineering of a twostep transcription amplif ication (TST A) system for transgene delivery in prostate cancer. Mol Ther 2002;5:223–32. 40. Benezra R, Da vis RL, Lockshon D , et al. The protein Id: a ne gative regulator of helix-loop-helix DN A binding proteins. Cell 1990, 61:49–59. 41. Boland MP, O’Neill LA. Ceramide acti vates NFkappaB b y inducing the processing of p105. J Biol Chem 1998;273:15494–500. 42. Luker GD, Sharma V, Pica CM, et al. Noninvasive imaging of proteinprotein interactions in li ving animals. Proc Natl Acad Sci U S A 2002;99:6961–6. 43. Gambhir SS, Bauer E, Black ME, et al.A mutant herpes simplex virus type 1 thymidine kinase reporter gene shows improved sensitivity

44.

45.

46. 47.

48.

49.

50.

51.

52.

53.

54.

55.

56. 57.

58.

59.

60.

61. 62.

63.

64.

for imaging repor ter gene e xpression with positron emission tomography. Proc Natl Acad Sci U S A 2000;97:2785–90. Luker GD, Sharma V, Pica CM, et al. Molecular imaging of proteinprotein interactions: controlled e xpression of p53 and lar ge T-antigen fusion proteins in vivo. Cancer Res 2003;63:1780–8. Pichler A, Prior JL, Luk er GD , Piwnica-Worms D . Generation of a highly inducible Gal4/Fluc universal reporter mouse for in vi vo bioluminescence imaging. Proc Natl Acad Sci U S A 2008;105: 15932–7. Förster T. Intermolecular energy transference and fluorescence. Ann Phys 1948;2:54–75. De A, Gambhir SS. Nonin vasive imaging of protein-protein interactions from live cells and living subjects using bioluminescence resonance energy transfer. FASEB J 2005;19:2017–9. De A, Loening AM, Gambhir SS. An improved bioluminescence resonance energy transfer strategy for imaging intracellular events in single cells and living subjects. Cancer Res 2007;67:7175–83. Levi J, De A, Cheng Z, Gambhir SS. Bisdeo xycoelenterazine derivatives for improvement of bioluminescence resonance ener gy transfer assays. J Am Chem Soc 2007;129:11900–1. De A, Ray P, Loening AM, Gambhir SS. BRET3: A red-shifted bioluminescence resonance ener gy transfer (BRET) based inte grated platform for imaging protein-protein interactions from single li ve cell and living animals. FASEB J 2009 (in press). Wehrman T, Kleaveland B, Her JH, et al. Protein-protein interactions monitored in mammalian cells via complementation of beta-lactamase enzyme fragments. Proc Natl Acad Sci U S A 2002;99:3469–74. Paulmurugan R, Umeza wa Y, Gambhir SS. Nonin vasive imaging of protein-protein interactions in li ving subjects b y using repor ter protein complementation and reconstitution strate gies. Proc Natl Acad Sci U S A 2002;99:15608–13. Galarneau A, Primeau M, Trudeau LE, Michnick SW. Beta-lactamase protein fragment complementation assa ys as in vi vo and in vitro sensors of protein-protein interactions. Nat Biotechnol 2002;20:619–22. Kaihara A, Kawai Y, Sato M, et al. Locating a protein-protein interaction in li ving cells via split Renilla luciferase complementation. Anal Chem 2003;75:4176–81. Pelletier JN , Campbell-V alois F , Michnick SW . Oligomerization domain-directed reassemb ly of acti ve dih ydrofolate reductase from rationally designed fragments. Proc Natl Acad Sci U S A 1998;95:12141–6. Johnsson N, Varshavsky A. Split ubiquitin as a sensor of protein interactions in vivo. Proc Natl Acad Sci U S A 1994;91:10340–4. Michnick SW , Rem y I, Campbell-V alois FX, et al. Detection of protein-protein interactions by protein fragment complementation strategies. Methods Enzymol 2000;328:208–30. Hu CD, Kerppola TK. Simultaneous visualization of multiple protein interactions in living cells using multicolor fluorescence complementation analysis. Nat Biotech 2003;21:539–45. Spotts JM, Dolmetsch RE, Greenber g ME. Time-lapse imaging of a dynamic phosphor ylation-dependent protein-protein interaction in mammalian cells. Proc Natl Acad Sci U S A 2002;99:15142–7. Martin DD, Xu MQ, Evans TC Jr. Characterization of a naturally occurring trans-splicing intein from Synechoc ystis sp. PCC6803. Biochemistry 2001;40:1393–402. Paulus H. Protein splicing and related for ms of protein autoprocessing. Annu Rev Biochem 2000;69:447–96. Wu H, Hu Z, Liu XQ. Protein trans-splicing by a split intein encoded in a split DnaE gene of Synechoc ystis sp. PCC6803. Proc Natl Acad Sci U S A 1998;95:9226–31. Ozawa T, Kaihara A, Sato M, et al. Split luciferase as an optical probe for detecting protein-protein interactions in mammalian cells based on protein splicing. Anal Chem 2001;73:2516–21. Ozawa T, Umezawa Y. Detection of protein-protein interactions in vivo based on protein splicing. Cur r Opin Chem Biol 2001;5:578–83.

Molecular Imaging of Protein–Protein Interactions

65. Kim SB, Ozawa T, Watanabe S, Umezawa Y. High-throughput sensing and nonin vasive imaging of protein nuclear transpor t b y using reconstitution of split Renilla luciferase. Proc Natl Acad Sci U S A 2004;101:11542–7. 66. Kim SB, Ozawa T, Umezawa Y. Genetically encoded stress indicator for nonin vasively imaging endo genous cor ticosterone in li ving mice. Anal Chem 2005;77:6588–93. 67. Kim SB, Takao R, Ozawa T, Umezawa Y. Quantitative determination of protein nuclear transport induced by phosphorylation or by proteolysis. Anal Chem 2005;77:6928–34 68. Kanno A, Ozawa T, Umezawa Y. Genetically encoded optical probe for detecting release of proteins from mitochondria toward cytosol in living cells and mammals. Anal Chem 2006;78:8076–81. 69. Luker KE, Smith MC, Luker GD, et al. Kinetics of regulated proteinprotein interactions revealed with firefly luciferase complementation imaging in cells and li ving animals. Proc Natl Acad Sci U S A 2004;101:12288–93. 70. Paulmurugan R, Gambhir SS. Combinatorial librar y screening for developing an impro ved split-f irefly luciferase fragmentassisted complementation system for studying protein-protein interactions. Anal Chem 2007;79:2346–53. 71. Choi CY, Chan D A, P aulmurugan R, et al. Molecular imaging of hypoxia-inducible factor 1 α and von Hippel-Lindau interaction in mice. Mol Imaging 2008, 7:139–46. 72. Chan CT, Paulmurugan R, Ree ves RE, et al. Molecular Imaging of Phosphorylation Ev ents for Dr ug De velopment. Mol Imaging Biol 2008 Dec 2. [Epub ahead of print] 73. Zhang L, Lee KC, Bhojani MS, et al. Molecular imaging of Akt kinase activity. Nat Med 2007;13:1114–9. 74. Zhang L, Bhojani MS, Ross BD, Rehemtulla A. Enhancing Akt imaging through tar geted repor ter e xpression. Mol Imaging 2008;7:168–74. 75. Paulmurugan R, Gambhir SS. F irefly luciferase enzyme fragment complementation for imaging in cells and li ving animals. Anal Chem 2005;77:1295–302. 76. Paulmurugan R, Gambhir SS. Monitoring protein-protein interactions using split synthetic renilla luciferase protein-fragment-assisted complementation. Anal Chem 2003;75:1584–9. 77. Paulmurugan R, Massoud TF, Huang J, Gambhir SS: Molecular imaging of dr ug-modulated protein-protein interactions in li ving subjects. Cancer Res 2004;64:2113–9. 78. Bhaumik S, Gambhir SS. Optical imaging of Renilla luciferase reporter gene e xpression in li ving mice. Proc Natl Acad Sci U S A 2002;99:377–82.

807

79. Pichler A, Prior J, Piwnica-Worms D. Imaging reversal of multidrug resistance in li ving mice with bioluminescence: MDR1 P-glycoprotein transpor ts coelenterazine. Proc Natl Acad Sci USA 2004;101:1702–7. 80. Paulmurugan R, Gambhir SS. Novel fusion protein approach for efficient high-throughput screening of small molecule-mediating protein-protein interactions in cells and li ving animals. Cancer Res 2005;65:7413–20. 81. Chan CT, Paulmurugan R, Gheysens OS, et al. Molecular imaging of the efficacy of heat shock protein 90 inhibitors in li ving subjects. Cancer Res 2008;68:216–26. 82. Paulmurugan R, Gambhir SS. An intramolecular folding sensor for imaging estrogen receptor-ligand interactions. Proc Natl Acad Sci U S A 2006;103:15883–8. 83. Paulmurugan R, Gambhir SS. Imaging ER-ligand modulated multiprotein interactions [(human estro gen receptors (ER-alpha/beta), p53 tumor suppressor protein and the human equi valent of mouse double minute 2 (HDM2)] with a split-luciferase system. Presented at the Joint Molecular Imaging Conference, Pro vidence, Rhode Island, September 2007. 84. Kim SB, Otani Y, Umezawa Y, Tao H. Bioluminescent indicator for determining protein-protein interactions using intramolecular complementation of split click beetle luciferase. Anal Chem 2007;79:4820–6. 85. von Degenfeld G, Wehrman TS, Hammer MM, Blau HM. A universal technology for monitoring G-protein-coupled receptor acti vation in vitro and noninvasively in live animals. FASEB J 2007;21:3819–26. 86. Rossi F, Charlton CA, Blau HM. Monitoring protein-protein interactions in intact eukaryotic cells by beta-galactosidase complementation. Proc Natl Acad Sci U S A 1997;94:8405–10. 87. Massoud TF, P aulmurugan R, Gambhir SS. A no vel split HSV1thymidine kinase for PET repor ter fragment complementation imaging of protein-protein interactions. Presented at the Joint Molecular Imaging Conference, Pro vidence, Rhode Island , September 2007. 88. Gammon ST, Leevy WM, Gross S, et al. Spectral unmixing of multicolored bioluminescence emitted from hetero geneous biological sources. Anal Chem 2006;78:1520–7. 89. Massoud TF, Gambhir SS. Integrating noninvasive molecular imaging into molecular medicine: an evolving paradigm. Trends Mol Med 2007;13:183–91.

48 FLUORESCENCE READOUTS OF BIOCHEMISTRY IN LIVE CELLS AND ORGANISMS ROGER Y. TSIEN, PHD

This chapter discusses the most common general principles b y w hich fluorescence is used to measure important biochemical species or monitor biochemical pathways in li ving cells, tissues, and or ganisms. Fluorescence is a uniquely powerful and complex imaging modality because it combines high spatial resolution, time resolution, sensiti vity, and spectroscopic modulation. Classical microscopy per mits spatial resolution do wn to about 200 nm, but single fluorescent particles can now be localized down to ~1 nm precision, 1–3 and spectroscopic techniques, such as fluorescence resonance ener gy transfer (FRET), can repor t yet smaller distance changes. The time resolution of fluorescence intrinsicall y e xtends at least to the nanosecond domain, that is, f ar f aster than most biochemical events. Fluorescence recordings of li ve cell physiology typically range from nanoseconds to many hours (Figure 1A). In cell-free systems, single fluorescent molecules can be reliab ly detected ,4,5 which for man y applications is the ultimate sensiti vity limit. Finally, fluorescence can be usefull y modulated b y a wide v ariety of molecular mechanisms. This en vironmental sensiti vity stands in stark est contrast to radioacti ve decay, which is completely indif ferent to the chemical en vironment. Of course, fluorescence also has some fundamental limitations. Strongly fluorescent tags are full-sized molecules in their own right, so attachment to a biolo gical molecule of interest can per turb the latter signif icantly. Strong illumination of fluorophores e ventually bleaches them and can also damage sur rounding molecules and cells. Fluorescence imaging inside most intact tissues and or ganisms rapidly loses spatial resolution and sensiti vity as depths increase from tens of microns to millimeters or centimeters (Figure 1B). In most systems from li ve cells to intact organisms, endogenous background fluorescence (autofluorescence), rather than instr umental sensiti vity, 808

determines the minimum detectab le concentration of exogenous fluorophores. Space limitations forbid detailed e xplanations of the basic ph ysical principles of fluorescence or of sophisticated biophysical techniques mainly applicable to purified molecules in vitro. Standard te xtbooks should be consulted for such topics. 6 I f irst discuss the general mechanisms by which fluorescence can report biological signals in live cells and organisms. Then I review how some of the more impor tant pathw ays ha ve actuall y been imaged. Again because of space limitations, I mainl y cite recent reviews, rather than describe biological results in detail or credit all the original pioneers.

MAJOR MECHANISMS FOR FLUORESCENCE RESPONSES IN LIVE CELLS AND ORGANISMS Fluorophore as Spectroscopically Passive Tag for Macromolecule of Interest Many applications of fluorescence mak e no use of its potential en vironmental sensiti vity. Instead , the fluorophore is attached to a molecule of interest merel y to make the latter visib le and trackab le. Such tagging is required because few biologically relevant molecules have useful intrinsic fluorescence. Among the e xceptions7 are reduced p yridine nucleotides (N ADH and N ADPH), flavins, tetrap yrroles, such as protopor phyrin IX and chlorophyll, and aging-related pigments, such as lipofuscins. Many proteins are somewhat fluorescent at ultraviolet (UV) wavelengths due to their tr yptophan content, but such fluorescence is mainl y useful for studying purif ied proteins rather than intact cells because the e xcitation wavelengths (~300 nm) are phototo xic and too shor t for

Fluorescence Readouts of Bioc hemistry in Live Cells and Or ganisms

most microscopes and because no one protein stands out sufficiently from all the others. Exogenous Added Dyes, Peptides, Proteins

Traditionally, peptides and proteins are tagged in vitro b y reacting free amino or thiol groups of the purified macromolecule with reacti ve deri vatives of fluorescent dy es. Isothiocyanates and N-h ydroxysuccinimide (NHS) esters are probably the most popular amine-reacti ve derivatives, whereas maleimides and iodoacetyl g roups are the dominant thiol-reacti ve deri vatives. The most common fluorophores are xanthenes (fluoresceins and rhodamines)

A A 3

PET BL

Log (time/s)

2 Fixation EM, AFM

1 0

OCT

MRI US

Fluorescence

2 Quick-freeze EM

3 6

B B 5

5

4 3 2 1 0 1 Log (lateral dimension/mm)

Log (depth/mm)

3 2 1 0

2

3

AFM TIRFM

4

Quick-freeze EM Fluorescence

Fixation EM OCT BL MEG

1 2

US MRI

for emission maxima of up to about 600 nm, w hereas indolenine c yanines star t around 550 nm and dominate above 600 nm. Many other fluorophores have advantages for specif ic applications and are tabulated in catalo gs.8 Although longer w avelengths are usuall y biolo gically advantageous because of deeper penetration and reduced background fluorescence and phototo xicity, the cor responding fluorophores are chemicall y less con venient, being less r ugged, less w ater-soluble, and generall y less tractable than their cousins operating at shor ter w avelengths. The fundamental limitations on this approach are the need to isolate and purify reasonab le quantities of the starting macromolecule, to devise a labeling protocol that confers suf ficient fluorescence without destro ying the essential biochemical proper ties, and to reintroduce the labeled macromolecule to the appropriate location in the cell or or ganism, which is par ticularly challenging if that location is intracellular. Fluorescent Proteins

MEG

1

809

PET

3 Figure 1. Comparison of the lateral dimensions, time scales, and depth penetrations of major imaging modalities. Note that all scales are logarithmic and that all boundaries are fuzzy and can be breached by unusual examples or technological improvements. A, Time scales vs lateral dimensions. B, Depth penetration vs lateral dimensions. EM, electron microscopy; AFM, atomic force microscopy; OCT, optical coherence tomography; BL, bioluminescence; US, ultrasound; MRI, magnetic resonance imaging; PET, positron emission tomography; MEG, magnetoencephalography; TIRFM, total internal reflection fluorescence microscopy. The boundary for each modality roughly matches the hue of its label.

Protein labeling and in vi vo fluorescence have been revolutionized b y the disco very, cloning, and mutagenic improvement of natural proteins with strong visib le fluorescence.9–11 Thus f ar, all such fluorescent proteins (FPs) have originated from marine coelenterates, such as jell yfish and corals, are c ylindrical 11-stranded β barrels of ~2.4 nm diameter and 4 nm length, and contain at least 200 amino acids. The light-absorbing unit or chromophore is a hydroxybenzylideneimidazolinone, generated by spontaneous cyclization and autoxidation of a few amino acids on a helix running up the center of the cylinder. Many variations of chromophore structure (Figure 2) have been generated b y ecolo gical di versification and ar tificial reengineering and are responsible for a wide range in colors. The cr ucial advantage of FPs is their genetic encodability, so that cells e xpressing the gene encoding the FP become fluorescent. In-frame fusion of the FP gene with the gene for the protein of interest generates a chimeric protein in situ, w hich hopefully (and often actuall y) functions lik e the nati ve protein e xcept that it is fluorescent. Thus, the protein never has to be purified in vitro, the location of the tag within the primar y sequence is accuratel y specified, and the normal biosynthetic machinery of the cell can be used to tar get the chimera to vir tually any desired location. After much engineering, well-behaved FPs9–11 are now available with e xcitation maxima from about 380 to 598 nm, emission maxima from 440 to 650 nm, fairly good resistance to photobleaching and other en vironmental factors, very little phototo xicity, and lack of oligomerization. Ob ligate dimer or tetramer for mation is a feature

810

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 2. Structures of the chromophores of representative fluorescent proteins, ranging from the shortest to the longest emission wavelengths. Side chains that significantly influence the fluorescence are shown in gray.

of many wild-type FPs, especiall y the longer -wavelength representatives from corals. 12 Such oligomerization can be problematic when the FP is fused to a protein of interest because the latter is forced to join in the agg regation, which can readil y cause mistraf ficking, precipitation, or toxicity.13 Oligomerization is less frequentl y a prob lem when the FP is left unfused simpl y to mark whole cells or tissues, but it ma y explain why several attempts to mak e transgenic mice e xpressing tetrameric red FPs (DsRed varieties) f ailed.14–16 This prob lem was eventually solv ed by the introduction of monomeric or tandem-dimeric mutants of red FPs.13,17,18 (A tandem dimer is a concatenation of two copies of the protein such that the dimeric protein-protein interaction is satisfied intramolecularly). Even the original jell yfish green fluorescent protein (GFP) and all color variations are weakly dimeric (K d ~ 0.1 mM) 19,20 unless the dimer interf ace has been deliberatel y destroyed by mutagenesis. This weak association can cause problems when monitoring protein-protein interactions, especially at high local concentrations or w hen the proteins are anchored to membranes. 20,21 The abo ve discussion has focused on FPs as passi ve tags, but FP variants ha ve also been evolved in which the fluorescence spontaneously changes color or can be usefull y s witched on or of f b y different wavelengths of light or environmental factors (see sections “Single FP: P erturbation of Chromophore Protonation” and “Protein Trafficking and Degradation”).

Of course, FPs ha ve their own limitations. The most fundamental are that their size cannot be signif icantly reduced, that they require some O2 to generate their internal fluorophores, and that H 2O2 is a stoichiometric by-product of these reactions.9,22 An early claim of a halfsized FP pro ved to be an ar tifact.23 The O 2 dependence should be kept in mind when using FPs as readouts under hypoxic conditions, but for tunately the auto xidation can proceed at quite lo w partial pressures of O 2 (~0.1 ppm24 or about 3 µM), and sometimes the readout can be postponed until after O 2 can be readmitted to f inish the maturation process. 25 Current shor tcomings that might be ameliorated b y fur ther protein engineering include the lack of bright FPs with reall y long w avelengths (excitation maxima > 600 nm to a void the hemo globin absorbance edge, emission maxima > 650 nm) and photostability that is still not as good as the best small molecules or quantum dots. Translocation Assays26,27 Intracellular signaling pathways often in volve translocation of a transducer protein from one subcellular compartment to another, for example, from cytosol to the plasma membrane or into the nucleus, or the re verse. Examples include recr uitment of protein kinase C from cytosol to the plasma membrane by generation of Ca 2+ and diacylglycerol, recr uitment of β-arrestin from c ytosol to acti vated G-protein-coupled receptors, release of pleckstrin homolo gy domains from plasma

Fluorescence Readouts of Bioc hemistry in Live Cells and Or ganisms

membrane to cytosol upon hydrolysis of phosphatidylinositol-4,5-bisphosphate to inositol 1,4,5-trisphosphate (Ins(1,4,5)P3), translocation of the nuclear f actor of activated T cells (NF-AT) into the nucleus upon dephosphorylation by calcineurin, etc. Such activation steps can usually be monitored simply by fusing the translocating protein to an FP and imaging the subcellular distribution of the chimera.26,27 Such assa ys are usuall y simple to estab lish and quite robust, though rarel y calibrated in ter ms of the concentration of triggering ligand. Their disadvantages are that high spatial resolution is required and that the speed of response may be limited by diffusion kinetics. Genetically Targeted Small Molecules

For man y pur poses, it w ould be desirab le to combine genetic tar geting with the phenotypic v ersatility and small size of man-made dy es and probe molecules. 28,29 The most compact products w ould be achie ved b y hijacking specif ic stop codons to encode single unnatural (eg, fluorescent) amino acids. This requires reengineering transfer RNAs (tRNA) and tRNA-synthetases to incorporate the desired unnatural amino acid. Also, the hijacked codon has to be inser ted into the rele vant mRNA, but the concer n remains that other endo genous mRNAs that happen to use that stop codon will translate into proteins also incor porating the unnatural amino acid plus C-ter minal extensions encoded b y previously untranslated 3 ʹ′sequences. Therefore, this strate gy has not yet progressed enough to be relied upon for molecular imaging of intact cells or or ganisms, though this may well change due to acti ve cur rent research. 30 The next step up in size is to devise a short (< 20 aa) peptide sequence that can be incor porated into the protein of interest and that can be labeled in situ with small organic molecules. The earliest example was a tetracysteine motif, −CCXXCC−, w hich can be labeled inside live cells with membrane-permeant dyes containing two arsenic atoms at the right spacing to plug into the tw o pairs of sidechain thiol g roups.31 Antidotes that prevent the biarsenical dye from poisoning endogenous proteins with cysteine pairs need to be co-administered but seem quite effective. Later, it w as discovered that the amino acids outside of and betw een the c ysteines could be optimized, yielding the cur rently preferred sequence −FLNCCPGCCMEP−.32 Advantages of the tetrac ysteine-biarsenical system include the v ery small size of targeting sequence, here 12 aa, and the abilities to detect newly synthesized peptide chains rapidl y (< 1 min), 33,34 to measure the age of proteins b y pulse-chase labeling with tw o dif ferent dy es on a time scale of hours to

811

days,35,36 to cor relate their fluorescence with electron microscopy,35,37 to inacti vate the protein’ s function in seconds to minutes, 38 and to monitor protein unfolding in situ.39 Subsequent approaches ha ve been analo gously introduced to label he xahistidine motifs with Ni 2+ or Zn2+ complexes,40–42 to enzymaticall y attach biotin or lipoic acid analogs to short acceptor peptides,43–45 or to enzymatically replace a C-ter minal –LPXTG motif b y any tag with three glycines (GGG–) at its N-terminus.46 Also, several full-sized protein domains can bind small molecules with suf ficient specif icity for in situ labeling, though in such cases there is no major size advantage o ver FPs. Three e xamples include the following: (1) O-alkylguanine transferase or a mutant reacts ir reversibly with fluorescent benzyl deri vatives of guanine or c ytosine, respecti vely;47 (2) antibodies raised against dyes as haptens can then bind those dyes and enhance their fluorescence; 48 and (3) a ph ytochrome protein from c yanobacteria has been mutated so that binding of its linear tetrap yrrole cof actor, ph ycocyanobilin, produces a brightly fluorescent holoprotein instead of the native nonfluorescent light-triggered histidine kinase. The long w avelength e xcitation and emission maxima (648 and 672 nm, respecti vely) and respectable e xtinction coef ficient (measure of ho w strongly a molecule absorbs light, in this case 73,000 M−1cm−1) and quantum yield (probability than an excited molecule will reemit a photon, in this case > 0.1) could be of considerab le interest for in vi vo imaging.49,50 In bacteria and plants, the phycocyanobilin can be added from outside or biosynthesized from heme via enzymes that ha ve been cloned. Unfor tunately, there are no pub lications showing this system w orks in animal cells, and preliminary experiments with exogenous phycocyanobilin or transfection of the biosynthetic enzyme ha ve not been promising (X. Shu and R.Y . Tsien, unpublished).

Sequestration of Critical Lone Pair of Electrons The most common mechanism by which inorganic cations can directly affect the absorbance and fluorescence of synthetic small-molecule indicators is by electrostaticall y sequestering a lone pair of electrons on an o xygen or nitrogen atom within the chromophore. 51,52 Examples are shown in Figure 3. In each case, cation binding shifts the absorbance and excitation spectra to shor ter wavelengths. When the cation binds to an aniline sur rounded by additional coordination sites, the electrostatic sequestration is

812

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

O

O

HO

O

OOC

O COO

OOC

COO COO

O

COO

H

Protonated BCECF (excitation peak at 482 nm)

Deprotonated BCECF (excitation peak at 503 nm) OOC COO COO N

OOC

ʹ′ −

OOC N COO OOC COO Ca2 O N

N O

O

O Ca2 O

O N

N O

O

COO

COO

Fura-2 preferentially excited at 380 nm

Ca2 complex of Fura-2 preferentially excited at 340 nm

Figure 3. Structures of popular, representative small molecule indicators for which cation binding sequesters a crucial lone pair of electrons. The key lone pairs are indicated by electron clouds protruding from the relevant oxygen or nitrogen atoms. Top: the pH indicator BCECF. Bottom: the Ca2+ indicator fura-2.

supplemented b y twisting of the nitro gen-to-aryl bond , further disconnecting the lone pair from the rest of the chromophore. The most popular long-w avelength metal indicators, such as fluo-3, fluo-4, Calcium Green, OregonGreen-BAPTA, and their analo gs for other cations, onl y increase their fluorescence intensity upon metal binding, with v ery little or no w avelength shift. Here, the metal binding site is remote from y et connected to the actual chromophore. Excitation of the chromophore consists of promoting an electron from the highest occupied molecular orbital (HOMO) to the lo west unoccupied molecular orbital (LUMO). If the aniline is cation-free, it is electronrich enough to donate an electron to fill the vacancy in the HOMO, thus stranding the electron in the LUMO and quenching fluorescence. If the aniline is cation-bound, it is no longer electron-rich enough to donate to the HOMO, so fluorescence is restored. 51,52 Although such indicators are the easiest w ay to get absorbance and emission maxima > 500 nm, their signals are hard (b ut not impossible53) to calibrate because changes in anal yte concentration are hard to distinguish from changes in amount of dy e, motion artifacts, fluctuations in excitation intensity, etc. In principle, excited state lifetime measurements could overcome this ambiguity b y resolving metal-free from metal54 bound indicator molecules. Unfortunately, the

metal-free species is often undetectab le because its emission is too weak and short-lived. In some older metallochromic indicators, such as calcein or arsenazo III, the cr ucial phenols are surrounded by additional coordination sites to which the metal primarily binds, then forcing the phenol to deprotonate. Because the metal is fur ther away from and less tightly bound to the phenolate o xygen than the proton had been, the metal shifts the spectr um from that of the neutral phenol toward that of the phenolate anion. These indicators have the severe disadvantage that their metal responsiveness is limited to a nar row range of pH because of two intrinsic pH-sensitivities. At low pH, the metal has increasing dif ficulty displacing the proton, whereas at high pH, the phenol has already lost its proton, so metal binding has little spectral ef fect.

Modulating of Quenching by Dye Aggregation Another important way to quench fluorophores is simply to pack them to gether at high ef fective concentrations, typically in the millimolar range. One explanation is that identical fluorophores often lik e to stack side-b y-side with their flat f aces in contact to minimize h ydrophobic

Fluorescence Readouts of Bioc hemistry in Live Cells and Or ganisms

interactions with w ater. Any e xcitation of the dimer immediately relaxes to a lo wer-energy composite quantum state in which the excitation is spread between both fluorophores but w hose transition dipoles are antiparallel. This state is nonfluorescent because destr uctive interference betw een the oppositel y-oriented transition dipoles hinders photon emission during an y attempt to return to the g round state. 55 Many other e xplanations have also been proposed. 56 Whatever the quantum mechanical e xplanation, quenching by crowding has been empiricall y useful as a way of sensing an y biochemical process that allo ws packed fluorophores to disperse and escape quenching. Thus, if membrane-imper meant fluorescent dy es are loaded at high concentration inside v esicles, leakage of those vesicles or fusion with unloaded vesicles dilutes the dye molecules and increases fluorescence per molecule. Dyes attached at high density to a v esicular membrane likewise dequench when that membrane fuses with unlabeled membranes, per mitting dilution b y lateral dif fusion.57 Dyes closel y pack ed on pol ymers or proteins dequench w hen the pol ymers or proteins are disassembled, proteolyzed, or even unfolded by applied tension. 56 This effect has become a very popular method for visualizing local proteol ysis, both in culture and in li ve animals.58–61 Another v ariation is to image l ysosomal proteolysis after ligand endocytosis.62 However, one fundamental ambiguity should be kept in mind w hen imaging local dequenching by disassembly or proteolysis: the fluorescence increase should not be sustained but should reverse as fluorophores dif fuse a way from their site of release.

Single FP: Perturbation of Chromophore Protonation As shown in F igure 2, most emission from FPs stems from deprotonated chromophores, so an y biochemical signal that af fects the de gree of deprotonation will affect fluorescence. The most obvious signal is pH. The earliest disco vered FP , Aequorea Green FP , contains about 6:1 protonated:deprotonated chromophores in its wild-type ground state, 63 explaining why the e xcitation spectrum has a major peak at ~395 nm and only a minor peak at 475 nm. This ratio is sur prisingly indifferent to external pH betw een 4 and 11, impl ying that the fluorophore is w ell insulated from e xternal protons until extreme pH’s begin to denature the protective β-barrel.64 However, absorption of a photon strongly favors deprotonation of the fluorophore, so that both e xcitation peaks give a common emission peak at 510 nm. 65 The

813

ground state protonation ratio at pH 7 can be forced to full protonation (e xcitation onl y near 400 nm) v ersus deprotonation (e xcitation onl y at 475 nm, with much greater amplitude) by different mutations, for e xample, T203I versus S65T, respectively.9,66 For most pur poses, it is preferab le to put the maximum amplitude into the longer-wavelength e xcitation peak, so most routine applications of GFP as a passive tracer now incorporate S65T or equi valent mutations. Ho wever, S65T will reversibly protonate with a pK a of 6.15, and other mutants give both higher and lo wer pK a’s, so appropriate FPs geneticall y tar geted to dif ferent subcellular locations are amongst the best w ays to measure pH in those compar tments.67 Most secretor y g ranules are quite acidic inter nally, pH ~5, so e xocytosis can be monitored as pH-sensiti ve FPs tar geted to the v esicle lumen go from such acidic pH to pH 7.4 in the extracellular medium.68 In other mutants of GFP, the protonation state of the chromophore has been accidentall y or deliberatel y engineered to respond to a v ariety of signals other than pH, including halide ion concentrations, 69 thiol-disulfide redox potentials, 70,71 hydrogen pero xide,72 singlet o xygen, Ca 2+ ions,73–78 and even membrane potential. 79 It is often helpful to do circular permutation of the FP, that is, connect the original N- and C-termini by a peptide linker and generate new N- and C-ter mini at a ne w location on the β-barrel (Figure 4). The resulting per mutants generally seem a little more confor mationally fle xible than FPs with wild-type topolo gy, making it easier for other protein domains fused to the ne w N- or C-ter mini to affect the protonation state of the inter nal fluorophore. 73 A generic caution for all these sensors based on chromophore protonation is that the y are still some what or very sensitive to pH over certain ranges.

FRET FRET is a quantum-mechanical interaction betw een a donor fluorophore and an acceptor chromophore that can absorb some of the w avelengths at w hich the fluorophore would normally emit.6,80 The acceptor chromophore may be, but does not ha ve to be fluorescent. If the fluorophore and chromophore are within a fe w nanometers of each other and are appropriatel y oriented with respect to each other , then the e xcited state of the donor fluorophore can directl y transfer its ener gy to the acceptor chromophore without emission and reabsor ption of a photon. The resulting excited acceptor behaves just as if it had gotten its energy from a real photon, so if it is fluorescent, it can reemit at y et longer wavelengths.

814

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Old N and C terminii KYL VM

MGHT

G EDM HT

ED

GGKYL GGT S VM

Circular permutation

Y145

M (new N terminus) New C terminus

Figure 4. Schematic showing how circular permutation rearranges the topology of a fluorescent protein, such as GFP. Left: α-carbon backbone and chromophore of GFP before permutation. Colors of the α-carbon backbone range from blue at the N-terminus through the visible spectrum to magenta at the C-terminus simply to help the eye trace the sequence of strands. Amino acids MV at the N-terminus and THGMDELYK at the C-terminus are written as text because they are too disordered to have coordinates in the crystal structure. Right: GFP circularly permuted with new N- and C-termini created at Tyr-145 and the former N- and C-termini connected with a GGTGGS linker. See Baird GS et al.73 for further details on the construction and properties of this circular permutant.

Because FRET depends strongl y on donor -acceptor distance and orientation, it can be a sensiti ve readout of association equilibria and confor mational changes. 81 In intact cells and or ganisms, the donor and acceptor are nowadays almost al ways FPs or geneticall y tar geted small molecules, so that they can be precisely attached to the host protein(s) and so that the resulting chimera can be targeted to the appropriate location(s). FRET can be monitored b y many different spectroscopic techniques. 82 The most common is to use a fluorescent acceptor and monitor the ratio of emissions at wavelengths characteristic of the acceptor and the donor , respectively. This ratio increases as FRET increases, b ut quantitative calibration of this emission ratio in ter ms of FRET ef ficiency requires mathematical cor rections for the extent to w hich the donor emits within the acceptor’ s emission band, the e xtent to w hich the donor e xcitation wavelength also directl y e xcites the acceptor and the quantum yields of the isolated donor and acceptor .83–85 A mathematically simpler procedure is to obser ve ho w much the donor gets brighter w hen the acceptor is selectively photob leached, but this calibration is destr uctive, relatively slow, and requires that the amount of donor in each pixel remain unchanged during such destr uction of the acceptor .86 FRET can be ele gantly and nondestr uctively quantif ied b y its diminution of the e xcited state

lifetime of the donor ,87 but the instr umentation to mak e such measurements is e xpensive and rare, and the signal acquisition times are much longer than for collection of emission ratios. F inally, FRET can occur to some e xtent between identical fluorophores due to the modest o verlap between e xcitation and emission spectra. Such “homoFRET” is particularly useful for assessing homodimerization but can onl y be detected b y a reduction in the polarization of the emission. 88 The fundamental advantages of FRET are its generality, w ell-defined quantitati ve basis, sensiti vity to small changes in distance or orientation, instantaneous kinetics (within the nanosecond e xcited state lifetime), and complete reversibility. Also, FRET does not require that donor and acceptor touch each other , and it does not introduce any e xtra attraction or repulsion betw een them. On the other hand, FRET becomes v ery weak beyond 6 to 8 nm, which is too shor t a range for man y interacting protein complexes. Even within that distance range, FRET usually causes quite subtle changes in the composite emission spectrum, f ar from an all-or -nothing signal, so careful quantitation is required. One almost never knows the relative orientation of the donor and acceptor fluorophores, and the traditional simplifying assumption of random orientation is unreliable for fluorophores as large as FPs,89 so orientation becomes a signif icant unknown perturbant.

Fluorescence Readouts of Bioc hemistry in Live Cells and Or ganisms

A useful relati ve of FRET is bioluminescence resonance energy transfer (BRET), in w hich the donor is a bioluminescent protein. 90–92 Generation of the e xcited state b y chemical decomposition rather than photon absorption makes no difference to the subsequent interaction with a neighboring acceptor . BRET shares the abo ve features of FRET and furthermore avoids the problems of autofluorescence background and dif ficult penetration of excitation light. On the other hand , BRET shares the generic weaknesses of bioluminescence, such as requirement for cof actors, lo w light output per molecule, and inability to apply tricks based on e xcitation control, such as e xcited state lifetime measurements, polarization measurements, and multiple e xcitation locations and beam directions to aid tomographic reconstruction.

Other Proximity Assays FP Complementation

Many proteins can be expressed as two separate fragments that if held in close pro ximity can refold to gether and reconstitute the function of the nati ve protein. 93,94 FPs might seem too par ticularly hard to reconstitute from separate fragments because the β-barrel looks so monolithic. However, the tolerance of FPs to circular per mutation73 suggested that fragment complementation might w ork, which proved to be tr ue. Thus, co-expression of residues (1 to x) of GFP fused to protein X, plus residues (x to 238) of GFP fused to protein Y, where x = 155 or 173, generates fluorescence only if X and Y form a comple x.95–97 A further extension is to mix and match fragments derived from cyan fluorescent protein (CFP) and yellow fluorescent protein (YFP); the resulting h ybrid proteins sho w spectra intermediate between those of their parents. 98 By contrast, fragments consisting of amino acids 1 to 214 and 215 to 230 of a superfolder GFP , each indi vidually nonfluorescent, spontaneously reassociate to gi ve fluorescence without requiring an y par tners to splint them to gether.99 Presumably these fragments ha ve a much higher association constant than those split at 155 or 173, but these association parameters ha ve not been measured. The g reat advantage of such complementation methods (also applicable to bioluminescent proteins and man y enzymes 93,94) is their all-or-nothing response, so that it is easy to qualitatively detect lo w levels of interaction. On the other hand , there is little or no quantif ication of ho w strong or stab le the interaction between X and Y must have been to generate the observed signal. False negative results could result if the X-Y comp lex does not put the FP fragments into the correct orientation and close proximity, but this possibility

815

has not yet been documented. Folding and reconstitution of intact function is slow and irreversible for FPs. Activatable Cell-Penetrating Peptides

Another type of pro ximity assa y is based on cellpenetrating peptides 100 (CPPs, also kno wn as protein transduction domains or PTDs), w hich are pol ycationic peptides or oligomers that strongl y adhere to the ne gatively char ged surf aces (par ticularly proteo glycans) of cells and are then inter nalized, mostl y b y endoc ytosis. Payloads, such as imaging or therapeutic moieties, attached to the CPPs are thus taken up into cells, though much or most ma y remain trapped inside or ganelles. A polyanionic peptide fused to the CPP through a link er vetoes uptak e, probab ly b y for ming a hair pin with the polycation and neutralizing the latter’ s positi ve char ge. Cleavage of the link er, for e xample, b y proteol ysis, releases the pol yanion, remo ves the inhibitor y constraint, and frees the CPP . Thus, such acti vatable cellpenetrating peptides (A CPPs) localize CPP-mediated retention and uptak e of pa yloads in the immediate vicinity of e xtracellular proteases or other acti vities (eg, reduction of a disulfide) that can cut whatever linker connects the pol yanion to the pol ycation.101–103 This mechanism is functionally somewhat analogous to FRET because the pol ycation and pol yanion are analo gous to donor and acceptor , respecti vely, and pro ximity quenches cell uptak e rather than fluorescence. In some ways ACPPs are the in verse of FP complementation, in that the interaction of the two peptide segments is antagonistic rather than synergistic, so the assay detects dissociation rather than association. The biggest advantage is that the payload does not ha ve to be fluorescent but can be species as di verse as magnetic resonance imaging contrast agents or bacteriophages. 104,105 The only known limitation on the pa yload is that it must not mask and neutralize the pol ycation. Thus, nak ed oligonucleotides are probab ly poor pa yloads because of their multiple negative charges.

PATHWAYS COMMONLY MONITORED BY FLUORESCENCE Transcription and Translation106 Reporter FPs

The earliest 107 and probab ly still most common applications of FPs are as transcriptionall y controlled repor ter genes.108 Fluorescence can thus monitor w hich cells ha ve

816

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

been transfected , w hich cells e xpress a tissue-specif ic promoter, which promoter regions are activated, etc. Typically, the FP is left unfused, but recently the highest sensitivity has been obtained b y fusing a bright YFP to a membrane protein and thus restricting Bro wnian motion.109 Transcriptional generation and translation of a single messenger RN A (mRN A) molecule generated bursts of fluorescence in an Escherichia coli cell as the reporter YFP fusions matured their chromophores then bleached in the intense laser e xcitation. In this regime, the temporal fidelity with which translation can be reported is limited by the maturation time of the FP. Proteolytic degradation of the FP is ir relevant because photob leaching is much quicker. However, most uses of FPs to repor t transcription and translation are done with large populations of molecules per cell obser ved at much lo wer e xcitation intensities. Under these conditions, the intrinsic maturation time of the FP may not be limiting if it tak es much longer simply to build up enough FP to be seen over background. Also the fluorescence as a function of time is governed not just b y protein synthesis but also b y de gradation. FPs themselves are relatively long-lived and resistant to proteolysis. The instantaneous fluorescence more f aithfully reflects the rate of translation if the FP is destabilized b y appending degradation-promoting sequences,110,111 though the ine vitable trade-of f is a decrease in sensiti vity. This trade-off should be a voidable if the biolo gical system and instrumentation allow higher e xcitation intensities so that the FP can be simultaneousl y read out and destro yed b y photobleaching or photoconversion to a different color. For tracking the f ates of multiple cells in the same organism, the sizab le number of distinctl y colored individual FPs can be combinatorially multiplied by targeting to different subcellular locations as well as ingenious permutations of genetic recombinations controlled b y the Cre/lox system. More than a hundred color combinations can be distinguished. 112,114 When the focus is on the rate of translation in isolation from transcription, v ariations in transcription or mRNA stability can be cor rected for if the same v ector additionally encodes an FP of a dif ferent color , ideall y under the control of an inter nal ribosome entry sequence (IRES).114–116 The ratio of signals from the test and reference FPs isolates translational modulation of the for mer. The above readouts focus on the rates of transcription and translation. For investigations on the subcellular locations of these processes, the rele vant DN As or mRNAs can be indirectly visualized by inclusion of multiple tandem repeats of polynucleotide sequences (eg, lac operons for DN A117,119 or MS2 118,119 for mRN A) for which high-affinity specific binding proteins (eg, lacI or

MS2-binding protein) are kno wn. Co-e xpression of those binding proteins fused to FPs of dif ferent colors causes fluorescent dots to appear at the rele vant loci of the DN A or mRN A. Such techniques ha ve limitations, such as the lar ge bulk and potential per turbations introduced by the multiple nucleic acid repeats and binding protein-FP fusions. But with appropriate controls, the y have yielded much biological insight. Likewise, selected protein constituents of transcription comple xes or ribosomes can be fused to FPs for intact cell localization or in vitro biophysical measurements. 120 Reporter Enzymes with Fluorogenic Substrates106,121

An additional amplification stage can be incorporated if the reporter protein is not an FP but rather an enzyme that rapidly converts nonfluorescent substrates into fluorescent products or changes the color of the substrate. The costs of this increased sensiti vity are the need to synthesize and administer the substrate and the loss in spatial and temporal resolution. Man y such fluoro genic repor ter enzymes have been introduced , including alkaline phosphatase, 122 β-galactosidase (the product of the lacZ gene),123 RTEM β-lactamase (the product of the ampicillin-resistance gene),124 and uropor phyrinogen III-meth yltransferase (the product of the cobA gene).125 The first two enzymes generate fluorescence b y clea ving phosphate and galactose respectively of f phenolic g roups within dy es. Because phosphates and galactosides are much more h ydrophilic than free phenols, the enzyme substrates are much less membrane-permeant than the products. Therefore, it is difficult both to get the nonfluorescent substrates into li ving cells and to retain the fluorescent products there. Therefore, it may be better to put the reporter enzyme on the outside of cells, w here imper meant substrates ha ve relati vely free access, then trap the fluorescent product either b y precipitation122 or acetoxymethyl (AM) ester h ydrolysis.123 β-lactamase substrates are more inherentl y suitab le for imaging of single li ve cells because enzyme e xpressed in the c ytosol attacks the β-lactam link er, splitting tw o dy e molecules apar t, disr upting FRET, and increasing the net charge on the products. The loss of FRET gives a very large increase in blue/green emission ratio, enabling quantitative and v ery sensiti ve measurements on full y viab le single cells, especiall y b y flo w c ytometry.124,126,127 The unique advantages of uropor phyrinogen III-meth yltransferase is that its substrate is the ubiquitous inter mediate in heme biosynthesis, uropor phyrinogen III, and that its product (sirohydrochlorin) is red fluorescent. Despite its unique ability to generate long-wavelength fluorescence without an

Fluorescence Readouts of Bioc hemistry in Live Cells and Or ganisms

exogenous substrate, this repor surprisingly little-used.

ter enzyme

125

has been

Protein Trafficking and Degradation Nowadays, the dominant way to image protein trafficking and degradation is to fuse the protein of interest to an FP, preferably one engineered to delete an y tendenc y to oligomerize.13,21 In principle, the FP can be geneticall y appended to either end of the host protein. If both termini of the host protein are sacrosanct, then the FP can often be inserted within the host protein sequence, typicall y at an inter nal loop. Assuming a fusion can e ventually be found that has a similar function and fate as the untagged protein, and that the FP is not split of f before the entire chimera is e ventually de graded, imaging sho ws directly the spatial distribution of the protein of interest at or approaching steady state. The f ate of the protein can be probed with g reater spatiotemporal resolution b y choosing FPs w hose fluorescence properties can be locally or kinetically varied.128 The simplest case are “fluorescent timers, ” FPs w hose color gradually changes from green to red due to intrinsically slow kinetics of maturation, so that the a verage age of the FP or fusion can be judged b y the color .129,130 Unfortunately, existing fluorescent timers are still ob ligate tetramers. Also, the time resolution is limited b y the gradual nature of the color change at the population level or the stochastic variation in maturation time at the single molecule level. Better time resolution is obtainab le with an e xternal stimulus. Some earl y FPs, including wildtype Aequorea GFP, refused to become fluorescent if expressed above a per missive temperature, so one could selectively image those copies synthesized during a period of lo wer temperature. 131 Because temperature steps have so man y nonspecif ic biological effects, such modulation was soon superseded b y photochemical triggering, that is, use of actinic light to s witch fluorescence on or off or to change its color. Such modulation can now be either reversible or irreversible depending on the exact FP and conditions of ir radiation. Because light can be localized with g reat precision, local photochemical triggering also enab les monitoring of the mo vement of fusions between spatial compar tments. For further information on this comple x topic, the reader is refer red to recent reviews.128,132–134 Permeant small molecules can also be used to trigger the recording of the age and tur nover of proteins. One approach is via the geneticall y encoded tags discussed in section “Genetically Targeted Small Molecules.” Assuming the incor poration of the small-molecule fluorophore into

817

the peptide tag is saturating and irreversible, administration of one species of small molecule labels all the fusion proteins made up to that point. If e xcess unbound dy e is washed out, then a fluorophore of a dif ferent color is administered, it labels only the newer copies of protein, that is, those synthesized after w ashout of the f irst dy e.35,36,47 Although small molecules cannot be delivered and removed as precisel y as photons can, appropriate small-molecule tags per mit imaging b y nonoptical modalities, such as electron microscop y,35,37 and tracking of posttranslational modifications, such as glycosylation.135 A different form of pharmacological triggering has more potential to w ork in intact animals. Here, the protein of interest is geneticall y fused via a cis-acting protease to any easy-to-detect label, such as an epitope tag or FP. Once the polypeptide emerges from the ribosome, the constituti vely acti ve cis-acting protease cleaves itself from both its flanking par tners, lea ving the protein of interest untagged and unper turbed. But once a small-molecule inhibitor of the protease is administered, cleavage stops, so all newer copies of the protein of interest remain tagged and can be imaged by retrospective histology.136 Currently, the prefer red cisacting protease is that of hepatitis C vir us because this protease is monomeric, str ucturally and biochemicall y well characterized, quite specific in its substrate preferences, and inhibited b y membrane-per meant dr ugs already developed by the pharmaceutical industry.

Exocytosis and Endocytosis137 Techniques for monitoring rapid e xocytosis and endocytosis have been the focus of special attention because of the impor tance of these traf ficking e vents in synaptic transmission and endocrine secretion. Small amphiphilic styryl dyes with two permanent positive charges, such as FM1-43 and FM4-64, intercalate tightl y to the outer leaflet of the plasma membrane but do not flip-flop across it. Membrane-bound dy e is highl y fluorescent, whereas aqueous dye is practically nonfluorescent. Thus, exocytosis can be monitored either b y the increase in fluorescence as previously unstained vesicular membrane is added to the plasma membrane and becomes e xposed to free dye or by the subsequent washout of fluorescence as pre viously stained and endoc ytosed v esicular membrane is once again e xocytosed into dye-free medium. 138 More recently, pH-sensitive FPs ha ve provided a genetically encoded alternative.139 If an appropriate FP is fused to the luminal f ace of a v esicular membrane protein, its fluorescence is quenched by acidic pH (~5) in the lumen, switched on b y nor mal e xtracellular pH (~7.4) upon

818

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

exocytosis, then requenched g radually upon endoc ytic recycling and reacidif ication. This fluorescence readout can be used either to study synaptic biolo gy or to monitor in vivo neuronal activity.137,140

RNA Trafficking Unfortunately, no direct equi valent of FPs is kno wn for RNA, that is, a nucleotide sequence that folds up and spontaneously generates an internal fluorophore. In analogy to the peptide tags of section “Geneticall y Targeted Small Molecules,” RN A aptamers are kno wn that bind membrane-permeant nonfluorescent dyes, such as Malachite Green with submicromolar dissociation constants and enhance the dy e fluorescence se veral 1000-fold , attaining respectab le quantum yields. 141 Unfortunately, the same dyes also become fluorescent upon nonspecif ic binding to other intracellular constituents, so these aptamers have not y et enabled imaging of RN As within live cells. Distant membrane-per meant analogs of Malachite Green do not sho w signif icant fluorescence upon contact with mammalian cells (S.R. Adams, J . Babendure, and R.Y. Tsien, unpublished data), so the po werful techniques of in vitro RN A evolution should be ab le to generate tight-binding, fluorescence-enhancing aptamers against these more suitab le dy es. Meanw hile, the main approach currently available is to fuse the RN A of interest to multiple copies of a 19-nucleotide stem-loop sequence that binds specif ically and tightl y to a dimeric coat protein from bacteriophage MS2. This MS2 protein is co-e xpressed as a fusion to an FP , w hich therefore lights up the RNA of interest. 119 Once multiple copies of the stem-loop ha ve attracted equal numbers of MS2-FP fusions, many hundreds of kilodaltons ha ve been added to the RN A of interest, but much biolo gy has been learned ne vertheless. To reduce backg round fluorescence, e xcess copies of MS2-FP fusion should be avoided, or tw o RN A sequences each binding a different protein should be juxtaposed. In tur n, the tw o RNA-binding proteins are fused to complementar y fragments of a split FP, so that fluorescence only develops after the tw o RN A-binding proteins f ind each other on the same piece of RNA.142

Inorganic Ion Concentrations H+

Cytosolic pH can be measured with a range of smallmolecule fluorescent probes, perhaps the most popular of which is a fluorescein deri vative, BCECF, designed to

have an optimized pK a, improved cellular retention, and excitation ratiometric readout.143 The obvious genetically encoded alter natives are pH-sensiti ve FPs, w hich are especially advantageous if targeting to specific organelles or subcellular locations is desired. Most FPs become brighter overall as the pH increases, until e xtreme alkalinity causes ir reversible denaturation. The pK a’s cover a wide range; alkaline pKa’s are most readily available from circularly permuted FPs. 73 A few FPs have been evolved to ha ve ratiometric readouts, that is, pH-dependent changes in peak excitation or emission wavelength.144,145 Ca2+

Techniques for measurement and imaging of intracellular Ca2+ have under gone intensi ve de velopment due to the ubiquitous biological importance of this ion in cell signaling. In addition, Ca2+ signals are often the most convenient readout for cell acti vation in neuronal netw orks146 and pharmaceutical drug screening. The chemical challenge in sensing intracellular Ca 2+ is that the rele vant range of free concentrations is 10−8 to 10−7 M in quiescent cells, rising to 10−6 to 10−5 M in activated cells, sometimes for only a few milliseconds. Meanw hile, both intracellular Mg 2+ and extracellular Ca2+ are roughly 10−3 M, so the indicator must have at least f ive orders of magnitude selecti vity for Ca 2+ over Mg2+, and it must be introduceable into the cytoplasm without causing an y significant leakage in the plasma membrane. The organic chemical solutions 51,143 to these challenges are mostl y deri vatives of B APTA, 1,2-bis(2-aminophenoxy)ethane-N,N,Nʹ′,Nʹ′-tetraacetic acid, whose metal-binding site f its well-around Ca2+ but is too large for Mg 2+ to f it snugly. Ca 2+ alters the spectra b y sequestering the lone-pair electrons on nitro gen (see section “Sequestration of Critical Lone Pair of Electrons”). ties, including Ca2+ affinities and spectral proper excitation/emission wavelengths from UV to near-infrared (IR), are tuned b y the choice of substituents on the tw o aromatic rings of BAPTA. These polycarboxylic acids can be esterif ied with AM g roups to generate unchar ged, lipophilic derivatives, which can diffuse across the plasma membrane. The esters are then hydrolyzed by intracellular esterases to re generate the pol yanions, w hich are thus trapped inside the cell, mostl y in the c ytoplasm, though under some circumstances accumulation in or ganelles can also occur. Though the by-products of AM ester hydrolysis are acetic acid and for maldehyde, these are generated slowly enough and at lo w enough concentration not to cause noticeable toxicity except in especially sensitive tissues, such as retina, w hich require coadministration of formaldehyde antidotes. 51,147 AM esters can e ven load

Fluorescence Readouts of Bioc hemistry in Live Cells and Or ganisms

large numbers of neurons within intact brain to per mit imaging of f iring patter ns in neuronal netw orks.148,149 Higher loadings can buffer Ca2+ during excitotoxicity.150 A greater number of genetically encoded indicators have been created for Ca 2+ than for an y other analyte.148,151 Most are based on the endo genous Ca 2+ sensor calmodulin (CaM), w hich binds four Ca 2+ then wraps itself around a tar get peptide, of w hich the most commonly chosen is M13, derived from skeletal muscle myosin light chain kinase. Ca 2+ affinities can be tuned by mutation of the binding sites within CaM, for e xample, by changing glutamates to glutamines. Cross-reactivity with ubiquitous endogenous CaM or CaM-binding domains can be reduced b y introducing compensating mutations into both CaM and M13 152 or b y s witching the Ca2+ sensor to troponin C, 153 which is endogenously abundant only in skeletal muscle. The f inal mechanism for fluorescence responsi veness can be modulation 73–78 either of the protonation state of a single FP (see section “Single FP: P erturbation of Chromophore Protonation”) or the mutual orientation and FRET efficiency betw een a donor FP and an acceptor FP (see section “FRET”). 89 Although the geneticall y encoded indicators gi ve f ar g reater precision of localization anywhere within the transfected cell or organism and are self-rene wing, the y still gi ve much smaller fractional changes in fluorescence intensity or ratio and slower reaction kinetics than their small-molecule counterparts.146 For the specialized challenge of measurement of f ast Ca 2+ transients in nanodomains, such as within nanometers of the mouth of Ca 2+ channels, a h ybrid solution is possib le: small-molecule Ca2+ indicators with biarsenical substituents, w hich localize to tetracysteine motifs (see section “Genetically Targeted Small Molecules”) geneticall y placed within the channel sequence. 154 Miscellaneous Metal Cations

BAPTA and its deri vatives also bind man y other divalent and tri valent cations, for e xample, Sr 2+, Ba 2+, Mn2+, Zn 2+, Cd 2+, Pb 2+, and lanthanides. 51 Fortunately, these ions are nor mally at ne gligible free concentrations in cells, allo wing the dy es to measure Ca 2+. In some tumor cell types 155 or w hen nor mal cells are challenged with unusual le vels of hea vy metals, 156–159 signals from hea vy metals become signif icant and can be disentangled from Ca 2+ signals if suf ficient care is taken. Selecti vities can be shifted to ward f irst ro w “hard” transition metals b y replacing carbo xylates b y pyridine ligands. 160 To sense mono valent cations, the

819

multiple carbo xylate anions are replaced b y neutral macrocyclic rings, for example, diazacrown ethers. Na+ favors 15-membered, pentacoordinating rings, w hereas K+ prefers 18-membered , he xacoordinating rings. 161 Protein-based indicators for Zn 2+ have been based on Zn-fingers,17 metallothionein,162 carbonic anh ydrase,163 and de novo cysteines and histidines causing FP dimerization.164 FRET sensors for Cu+ and Zn2+ can be engineered from copper chaperones. 165 Unfortunately, few if any protein domains have been characterized that under go lar ge confor mational changes selectively in response to metal ions other than Ca 2+, Cu+, and Zn 2+. When such domains are isolated , it should be possible to incorporate them into genetically encoded sensors b y per turbation of FP chromophore protonation or by FRET. CI– and Orthophosphate

A few fluorescent dy es containing quater nized heterocyclic cations are quenched b y encounter with Cl –, permitting their use as chloride indicators. 166,167 They are some what dif ficult to use because of shor t w avelengths, easy b leaching, and poor trappability inside cells. Geneticall y encoded indicators became possib le with the disco very that some but not all YFPs have a cavity near the chromophore into which halide ions can fit.69 The bound anion electrostatically promotes protonation of the chromophore, w hich diminishes fluorescence, an e xample of the mechanism discussed in section “Single FP: P erturbation of Chromophore Protonation.” This quenching can be con verted into a ratiometric readout b y fusion of a pH-insensiti ve CFP to the halide- and pH-sensiti ve YFP.168,169 Cl– binding decreases both the spectral o verlap of YFP e xcitation with CFP emission, w hich FRET requires, as w ell as the quantum yield for YFP reemission. Orthophosphate (H 2PO4– and HPO 4–2) can be sensed by fusing a bacterial phosphate-binding protein betw een CFP and YFP and monitoring the change in FRET from the CFP to YFP moieties.170

Small Organic Molecules Amino Acids and Sugars

Fluorescent indicators for a v ariety of impor tant small or ganic molecules including amino acids (eg, glutamate and tryptophan) and sugars have been created by the same strategy, that is, inser tion of a periplasmic analyte-binding protein betw een CFP and YFP.171 Glutamate

820

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

has recei ved the most biolo gical attention because of its importance as the major e xcitatory neurotransmitter in the mammalian CNS.172,173 The original E. coli glutamate binding protein GltI has too high an af finity to sense the peak concentrations around mammalian synapses, b ut it is relatively easy to reduce af finities mutagenically. Optimization of the FRET response required systematic searching of sequence space, truncation of 0 to 15 amino acids from the N-terminus, and 0 to 10 amino acids from the C-terminus of GltI (16 × 11 = 176 combinations). The f itness landscape, that is, plot of response v ersus number of amino acids tr uncated from the N- and C-termini, was surprisingly jagged, in that the best perfor mer, with eight and f ive deletions from N- and C-termini, respectively, was surrounded on all sides by v ery poorl y perfor ming neighbors. This e xperience suggests that high-throughput generation and testing of systematic variations may be required to optimize FP-based FRET reporters.172 Cyclic Nucleotides

Cyclic nucleotides, c yclic adenine monophosphate (cAMP) and c yclic guanine monophosphate (cGMP), are important second messengers that necessarily bind to and activate several endogenous sensor proteins during their signal transduction cascades. These natural transducers can been converted into fluorescent indicators. 174 The cAMP-induced dissociation of catal ytic (C) and regulatory (R) subunits of protein kinase A can be monitored b y loss of FRET betw een fluorescein and rhodamine labels on C and R, respecti vely.175 This composite indicator produced impor tant biolo gical insights into cAMP signaling in in vertebrate neurons.176,177 However, the difficulties of attaching dy es to recombinant proteins in vitro, reconstitution of holoenzyme, and microinjection back into cells, w ere primar y motivations for the original de velopment of FPs with different colors suitab le for FRET. After much trial and error, viab le fusions of C and R to GFP and BFP and later to YFP and CFP were developed.178,179 Such fusions made introduction into transfectab le cells much easier , but left the prob lems of balancing the e xpression level of two separate gene products, and potential cross-reaction with endo genous unlabeled R and C subunits. Therefore, alter native FRET-based indicators for cAMP ha ve been created from a cAMP-acti vated guanine nucleotide e xchange factor, Epac. 180–182 Indicators for cGMP can lik ewise be engineered from cGMPdependent protein kinase (PKG), either sandwiched between donor and acceptor FPs 183,184 or fused to a single circularl y-permuted FP.185 Both Epac and PKG

change confor mation upon binding cAMP or cGMP respectively but do not dissociate into separate subunits as protein kinase A does. Cyclic nucleotide signaling can also be monitored at the ne xt do wnstream mechanistic stage, the activation of endogenous protein kinase A or G (see section “Kinase and Phosphatase Activities”). Ins(1,4,5)P3

Several indicators for the Ca 2+-mobilizing messenger Ins(1,4,5)P3 have been based either on translocation of tagged pleckstrin homolo gy domains from plasma membrane to the cytosol, or on FRET between FPs flanking portions of the Ins(1,4,5)P 3 receptor.186 Guanosine Triphosphate/Guanosine Diphosphate Status of G Proteins

Because of the widespread impor tance of both heterotrimeric and small guanosine triphosphate (GTP)binding proteins (G proteins) in signal transduction, much ef fort has been de voted to imaging their state of activation in live cells, that is, w hether they are bound to GTP versus GDP. One direct strate gy is to fuse an FP to the G protein of interest, for e xample, Ras, microinject GTP tagged with a small dy e Bodip y-TR, and obser ve FRET from the FP to the dy e, preferably at the le vel of single molecules so that colocalization of the tw o fluorophores and their dif fusional trajectories can also be resolved.187 One concern is that h ydrolysis of the bound Bodipy-TR-GTP might occur and might not be spectroscopically obvious, in which case the observations might include inacti ve Ras. Another approach, w hich a voids microinjection and can be w holly genetically encoded, is to fuse an FP to a protein domain kno wn to bind onl y to one for m of the G protein. 188,189 For e xample, the Rasbinding domain from the ef fector protein Raf-1 binds Ras-GTP, not Ras-GDP. Such binding can be detected by intermolecular FRET, for example, from CFP-tagged Ras to YFP-tagged Ras-binding domain. Alternatively, the G protein, the effector domain, and the donor and acceptor FPs can be fused in various orders into four-part chimeras in w hich intramolecular FRET reflects acti vation status of the e xogenous G protein. 189,190 Activation of endo genous unlabeled G proteins at par ticular subcellular locations, for e xample, plasma membrane v ersus Golgi, can be qualitati vely imaged b y translocation of FP-tagged binding domains from the c ytosol to those sites. 188 Results with heterotrimeric G proteins ha ve been more complex and controversial. Whereas the classical view is that βγ subunits bind to α-guanosine diphosphate and

Fluorescence Readouts of Bioc hemistry in Live Cells and Or ganisms

dissociate from α-GTP, there is increasing evidence from FRET and BRET experiments that in some cases, activation causes onl y a confor mational change, not o vert dissociation.191

Redox Potential; Reactive Oxygen, and Nitrogen Species Most of the redo x couples in cells are not in equilibrium with each other, which is fortunate because ultimate equilibrium with atmospheric O 2 would require most biolo gical molecules to burn up. Therefore, many separate redox potentials could be def ined. Thiol-disulfide e xchange is one of the few biochemically relevant redox reactions that does not require enzymatic catal ysis, so thiol-disulf ide redox potential is the first for which fluorescent indicators have become a vailable. They consist of YFP192 or GFP mutants70,71 bearing two cysteines on adjacent strands of the β-barrel, placed so that for mation of the intramolecular disulf ide linkage f avors protonation of the FP chromophore. Such protonation quenches the fluorescence of the YFP but enhances the 390 to 400 nm e xcitation peak of the GFP at the e xpense of the 470 to 490 nm peak. Thus, the YFP-based indicator (rxYFP) is intensity-onl y, whereas the GFP-based indicators (roGFPs) are ratiometric in e xcitation. As e xpected from electrostatic effects, placement of positi ve charges near the c ysteines speeds the reaction kinetics and mak es the midpoint potentials more oxidizing.193,194 Equilibration of rxYFP with oxidized/reduced glutathione (GSSG/GSH) is strongly catal yzed b y glutaredo xin, so fusion of rxYFP with glutaredoxin makes the chimera kineticall y specif ic for glutathione redo x status. 195 The FP-based indicators allow continuous nondestr uctive monitoring of thioldisulfide ratios in single cells or subcellular compar tments and indicate that c ytosol and mitochondria are much more reducing 70,196 than deduced from pre vious destructive measurements of GSH and GSSG. Although GSH is relati vely easy to assa y destr uctively, GSSG is equally important, yet far scarcer in cells and more dif ficult to assa y because GSH must f irst be destro yed; also, much of GSSG may be compartmentalized and releasable only upon cell disruption. The abo ve thiol-disulf ide redo x indicators can indirectly respond to reacti ve oxygen species, especiall y when the latter are g rossly elevated, but the linkages are complex and biolo gically interesting in their o wn right. Therefore, indicators specif ic for indi vidual reacti ve oxygen and nitro gen species w ould be v aluable. Smallmolecule fluorescent sensors for H 2O 2 have been created by e xploiting the ability of H 2O2 to con vert

821

arylboronic acids into phenols and have now been able to detect phagoc ytosis-associated H 2O2 production in macrophages.197 The f irst geneticall y encoded H 2O2 indicator is HyPer,72 in which a circularly permuted YFP has been inser ted into the re gulatory domain of a prokaryotic H 2O2 sensor. This molecule has detected the much smaller H 2O2 transients associated with apoptotic and g rowth f actor stimuli. Quenched fluorescein and rhodamine dyes that brighten upon reaction with singlet oxygen198 or other highl y reacti ve o xygen species (hydroxyl radical, pero xynitrite, and h ypochlorite199,200) have been described , though minimal biolo gical results have been repor ted y et. GFPs that respond specif ically and irreversibly to singlet oxygen with an excitation ratio change have been developed (C. Dooley and R.Y. Tsien, unpublished). F or signals related to nitric o xide (NO), one should distinguish betw een the free radical NO versus derivatives of the nitrosyl cation NO +. A genetically encoded sensor for NO was based201 on fusing GFP to the heme-binding re gion of solub le guanylyl cyclase, which presumab ly becomes loaded with endo genous heme. Exposure to NO ir reversibly increased the GFP fluorescence by 14%, so this chimera is an inte grator of NO exposure, not a re versible indicator of instantaneous levels. This indicator re vealed the limited spatial spread and frequenc y dependence of NO generation from cerebellar synapses from parallel f ibers onto Purkinje neurons.201 NO+ can react with both thiols and aromatic amines instead of heme iron. Small-molecule dy es (fluoresceins and near -IR c yanines) containing o-phenylenediamines respond to NO + by for ming triazoles, pre venting photo-induced electron transfer as described in section “Sequestration of Critical Lone P air of Electrons” and thereb y dequenching fluorescence. 202 The multiple cysteines in metallothioneins have been the basis for a genetically encoded FRET sensor for nitrosating species, such as nitrosothiols. 203

Endoprotease Activities It is easier to deliver fluorescent or fluorogenic substrates to proteases that are extracellular or in the lumen of endocytic or ganelles than to proteases in other intracellular locations. Therefore, the former class of protease activities can be monitored b y fluorescent suicide substrates (for mechanism see section “Fluorophore as Spectroscopically Passive Tag for Macromolecule of Interest”), release of free amines and lone pair electrons (see section “Sequestration of Critical Lone P air of Electrons”), dequenching due to disagg regation (see section “Modulating of Quenching b y Dye Aggregation”), disruption of

822

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

FRET (see section “FRET”), and acti vation of cellpenetrating peptides (see section “ Activatable CellPenetrating P eptides”), w hereas c ytosolic endoprotease activities are most cleanly imaged by disruption of FRET between FPs. Proteases w hose specif icity is go verned mainly by the amino acids on the ac yl side of the peptide bond to be clea ved (the S1, S2, S3 … positions) can be assayed destr uctively with a suicide substrate ( …S3-S2S1-warhead) or nondestr uctively with a clea vable substrate (…S3-S2-S1-dye). Attack of the active-site serine or cysteine on the w arhead results in co valent labeling and inactivation of the enzyme. 204,205 The fragment that remains attached to the enzyme car ries a fluorescent tag. If a por tion of the w arhead is displaced b y the active-site serine or c ysteine, the lea ving g roup could car ry a quencher, so that the fluorescence is dequenched upon successful reaction with the enzyme. 205 This mechanism facilitates molecular identif ication of the rele vant proteases, but it is most suited to serine or cysteine proteases, sacrifices enzymatic amplif ication, and potentiall y perturbs any biology downstream of the active protease. In cleavable substrates, h ydrolysis of the amide bond between S1 and dy e releases the free amine for m of the latter, including the lone pair of electrons on the amine nitrogen. The dy e is typicall y an aminocoumarin or aminoxanthene, in w hich the delocalization of the lone pair of electrons into the rest of the chromophore increases the intensity and the w avelength of the fluorescence compared with the intact substrate. Mechanisms described in sections “Modulating of Quenching b y Dye Aggregation,” “FRET,” and “Activatable Cell-Penetrating Peptides,” accommodate natural amino acids on the amine side of the peptide bond to be cleaved (the S1ʹ′, S2ʹ′, S3ʹ′… positions etc) and are equall y applicable to all four major classes of proteases (serine, c ysteine, aspar tic, and metallo-). F or these mechanisms, the clea vable peptides respectively link multiple fluorescent dy es to a polymeric backbone (see section “Modulating of Quenching by Dye Aggregation”), or donor to acceptor fluorophores (see section “FRET”), or pol ycationic to pol yanionic sequences (see section “Activatable Cell-Penetrating Peptides”). Inside cells, FPs are usuall y the most con venient fluorophores, which are very easy to genetically combine into donor-linker-acceptor chimeras. Fortunately, the FPs themselves are generall y quite resistant to endoproteases and proteolytic cleavage of the linker allows the donor and acceptor to dif fuse apar t, so that the FRET decrease is maximal. Caspase activation in apoptosis is a par ticularly popular application, w here imaging readil y re veals that the g radual response seen in cell populations is actuall y composed of transitions that are step-lik e in each individual cell but desynchronized between neighboring cells. 206

Ubiquitination and Proteasomal Degradation, Especially of Cell Cycle Proteins Decoration of a protein by ubiquitin is a cr ucial signal to trigger degradation of that protein in proteasomes. Such ubiquitination has been imaged b y FRET from a de gradation-prone GFP to a YFP-ubiquitin fusion, w hich is evidently accepted b y the ubiquitination machiner y. The YFP had been mutated to reduce its emission quantum yield to near zero, w hereas its absorbance w as kept high to preserve its ability to accept FRET and thereby reduce the fluorescence lifetime of the GFP donor.207 The final degradation of the polyubiquitinated protein in proteasomes leads to destr uction of an y fused FP. Surprisingly, FPs seem to vary significantly in their resistance to such de gradation. Relati vely de gradable g reen and orange-red FPs ha ve been fused to geminin and Cdt1, nuclear proteins that respecti vely accumulate in S/G 2/M and G 1 phases before being destroyed in late M/G 1 versus S/G2 phases of the cell cycle. Therefore, in cells transfected with both constr ucts, nuclei in S/G 2/M phase glow green, whereas nuclei in G1 are scarlet. This combination enables visualization and tracking of the mitotic status of every cell in a culture, xenograft, or transgenic mammal. 208

Kinase and Phosphatase Activities The most robust fluorescence repor ters of the balance between endo genous protein kinases and phosphatases are chimeras in w hich donor and acceptor FPs brack et a linker containing a peptide substrate for phosphor ylation/dephosphorylation and a protein domain that binds the relevant phosphor ylated amino acid , for e xample, an SH2 domain for phosphotyrosine or an FHA1 domain for phosphothreonine. When the substrate becomes phosphorylated, it presumab ly is captured b y the neighboring domain in some w ay like the folding of a jackknife. This conformational change can either increase or decrease FRET betw een the flanking FPs; either direction is acceptable provided that the change is sufficiently lar ge and reproducib le. The design and application of kinase activity repor ters has been recently re viewed.209 The specificity for par ticular protein kinases is go verned b y the substrate sequence and optional subcellular tar geting. If the phosphorylation-recognizing domain has too high an affinity for the phosphorylated amino acid, it may protect the latter so thoroughl y that phosphatases cannot attack, rendering the indicator irreversible.210 Most kinase activity repor ters are re versible, so the y actuall y signal a balance betw een kinase and phosphatase acti vities, though w e kno w comparati vely little about w hich

Fluorescence Readouts of Bioc hemistry in Live Cells and Or ganisms

phosphatases are involved. One phosphatase with a relatively well-defined role is calcineurin, w hich dephosphorylates and thereb y acti vates an impor tant f amily of transcription f actors, the nuclear f actors of acti vated T-cells (NFAT). Recently, a FRET sensor for calcineurin has been constructed211 by sandwiching a tr uncated regulatory domain from NFAT1 between CFP and a circularly permuted YFP.

Protein-Protein Interactions The detection of protein-protein interactions is an enormous topic for w hich recent re views should be consulted.94,212 Please also see Chapter 47, “Molecular Imaging of Protein-Protein Interactions” for additional discussion. When the proteins of interest are fused either to donor and acceptor FPs or to fragments of FPs or of fluorogenic reporter enzymes, successful FRET or complementation (see sections “FRET” and “FP Complementation”) give strong e vidence for pro ximity within a few nanometers. However, a lack of observable FRET or complementation is usually not good evidence for lack of interaction, so these methods are best used to quantify the spatiotemporal dynamics of kno wn interacting par tners rather than as primary screens to f ind novel partners. For larger comple xes w here the distance betw een tags exceeds 10 nm, the best cur rent approaches are probably to demonstrate that the tags remain colocalized even as they both dif fuse randomly (fluorescence crosscorrelation spectroscopy)213,214 or as the spatial resolution is sharpened by single-molecule imaging.3,215,216 Whereas most of the techniques described abo ve require that the partners of interest be geneticall y fused and transfected, two-color super -resolution co-localization of antibodies should detect proximities of endogenous unfused proteins in histological sections.215

SOME FUTURE PERSPECTIVES The examples in section “Pathways Commonly Monitored by Fluorescence” suggest that FPs have been at the core of most new fluorescent sensors for biochemical pathw ays. Synthetic dy es and quantum dots ha ve their indi vidual domains of superiority , but the genetic encodability and targetability of FPs gi ve them much wider applicability . The most direct route to translating the man y FP-based sensors from tissue culture microscop y to w hole-mouse imaging w ould be the de velopment of bright FPs with excitation maxima abo ve 600 nm to get past heme absorbances and minimize interference from autofluorescence. FRET par tners or complementing fragments

823

working at such long w avelengths w ould be e ven more po werful (albeit challenging to create). Highquantum-yield FPs that are ef ficient BRET acceptors might pro vide alter natives in man y cases (Chapter 22, “Nanochemistry for Molecular Imaging”). Fur ther improvements in instr uments and algorithms to collect and localize long-w avelength fluorescence from deep inside scattering tissues would be of complementary value, though be yond the scope of this re view (Chapter 11, “Fluorescence Tomography,” Chapter 14, “Dif fuse Optical Tomography and Spectroscopy”).

REFERENCES 1. Hell SW. Far-field optical nanoscopy. Science 2007;316:1153–8. 2. Toprak E, Selvin PR. New fluorescent tools for watching nanometerscale confor mational changes of single molecules. Annu Re v Biophys Biomol Struct 2007;36:349–69. 3. Michalet X, Lacoste TD, Weiss S. Ultrahigh-resolution colocalization of spectrall y separab le point-lik e fluorescent probes. Methods 2001;25:87–102. 4. Park H, Toprak E, Selvin PR. Single-molecule fluorescence to study molecular motors. Q Rev Biophys 2007;40:87–111. 5. Michalet X, Weiss S, Jager M. Single-molecule fluorescence studies of protein folding and conformational dynamics. Chem Rev 2006; 106:1785–813. 6. Lakowicz JR. Principles of fluorescence spectroscop y. Ne w York: Springer; 2006. 7. Schenke-Layland K, Riemann I, Damour O, et al. Two-photon microscopes and in vivo multiphoton tomographs—powerful diagnostic tools for tissue engineering and dr ug deli very. Adv Dr ug Deli v Rev 2006;58:878–96. 8. Haugland RP. Handbook of fluorescent probes and research chemicals. Eugene (OR): Molecular Probes; 2002. 9. Tsien RY. The g reen fluorescent protein. Annu Rev Biochem 1998; 67:509–44. 10. Shaner NC, Steinbach PA, Tsien RY. A guide to choosing fluorescent proteins. Nat Methods 2005;2:905–9. 11. Lukyanov KA, Chudak ov DM, F radkov AF, et al. Disco very and properties of GFP-lik e proteins from nonbioluminescent anthozoa. Methods Biochem Anal 2006;47:121–38. 12. Baird GS, Zacharias D A, Tsien RY. Biochemistry, mutagenesis, and oligomerization of dsRed , a red fluorescent protein from coral. Proc Natl Acad Sci U S A 2000;97:11984–9. 13. Campbell RE, Tour O, Palmer AE, et al. A monomeric red fluorescent protein. Proc Natl Acad Sci U S A 2002;99:7877–82. 14. Zhu H, Wang G, Li G, et al. Ubiquitous e xpression of mRFP1 in transgenic mice. Genesis 2005;42:86–90. 15. Long JZ, Lackan CS, Hadjantonakis AK. Genetic and spectrall y distinct in vivo imaging: embr yonic stem cells and mice with widespread expression of a monomeric red fluorescent protein. BMC Biotechnol 2005;5:20. 16. Shuen JA, Chen M, Gloss B, Calakos N. Drd1a-tdTomato BAC transgenic mice for simultaneous visualization of medium spin y neurons in the direct and indirect pathw ays of the basal ganglia. J Neurosci 2008;28:2681–5. 17. Shaner NC, Campbell RE, Steinbach P A, et al. Impro ved monomeric red, orange and yellow fluorescent proteins derived from Discosoma sp. red fluorescent protein. Nat Biotechnol 2004;22:1567–72. 18. Merzlyak EM, Goedhar t J, Shcherbo D, et al. Bright monomeric red fluorescent protein with an e xtended fluorescence lifetime. Nat Methods 2007;4:555–7.

824

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

19. Phillips GN . Str ucture and dynamics of g reen fluorescent protein. Curr Opin Struct Biol 1999;7:821–7. 20. Zacharias DA, Violin JD, Newton AC, Tsien RY. Partitioning of lipidmodified monomeric GFPs into membrane microdomains of li ve cells. Science 2002;296:913–6. 21. Zacharias D A. Stick y ca veats in an otherwise glo wing repor t: oligomerizing fluorescent proteins and their use in cell biolo gy. Sci STKE 2002;2002:PE23. 22. Zhang L, P atel HN , Lappe JW , Wachter RM. Reaction pro gress of chromophore biogenesis in green fluorescent protein. J Am Chem Soc 2006;128:4766–72. 23. Wiedenmann J, Elke C, Spindler KD , Funke W. Cracks in the beta-can: fluorescent proteins from Anemonia sulcata Anthozoa, Actinaria). Proc Natl Acad Sci U S A 2002;99:13358. 24. Hansen MC, P almer RJ Jr , Udsen C, et al. Assessment of GFP fluorescence in cells of Streptococcus gordonii under conditions of lo w pH and lo w o xygen concentration. Microbiolo gy 2001; 147:1383–91. 25. Zhang C, Xing XH, Lou K. Rapid detection of a gfp-mark ed Enterobacter aerogenes under anaerobic conditions b y aerobic fluorescence recovery. FEMS Microbiol Lett 2005;249:211–8. 26. Meyer T, Teruel MN . Fluorescence imaging of signaling netw orks. Trends Cell Biol 2003;13:101–6. 27. O’Rourke NA, Meyer T, Chandy G. Protein localization studies in the age of ‘Omics’. Curr Opin Chem Biol 2005;9:82–7. 28. Chen I, Ting AY. Site-specif ic labeling of proteins with small molecules in live cells. Curr Opin Biotechnol 2005;16:35–40. 29. O’Hare HM, Johnsson K, Gautier A. Chemical probes shed light on protein function. Curr Opin Struct Biol 2007;17:488–94. 30. Wang L, Xie J , Schultz PG. Expanding the genetic code. Annu Rev Biophys Biomol Struct 2006;35:225–49. 31. Griffin B A, Adams SR, Tsien R Y. Specif ic co valent labeling of recombinant protein molecules inside li ve cells. Science 1998; 281:269–72. 32. Martin BR, Giepmans BN , Adams SR, Tsien RY. Mammalian cellbased optimization of the biarsenical-binding tetrac ysteine motif for impro ved fluorescence and af finity. Nat Biotechnol 2005; 23:1308–14. 33. Rice MC, Czymmek K, Kmiec EB . The potential of nucleic acid repair in functional genomics. Nat Biotechnol 2001;19:321–6. 34. Rodriguez AJ, Shenoy SM, Singer RH, Condeelis J . Visualization of mRNA translation in living cells. J Cell Biol 2006;175:67–76. 35. Gaietta G, Deerinck TJ, Adams SR, et al. Multicolor and electron microscopic imaging of conne xin trafficking. Science 2002;296: 503–7. 36. Ju W, Morishita W, Tsui J, et al. Activity-dependent regulation of dendritic synthesis and trafficking of AMPA receptors. Nat Neurosci 2004;7:244–53. 37. Gaietta GM, Giepmans BN , Deerinck TJ, et al. Golgi twins in late mitosis revealed by genetically encoded tags for live cell imaging and cor related electron microscop y. Proc Natl Acad Sci U S A 2006;103:17777–82. 38. Tour O , Meijer RM, Zacharias D A, et al. Geneticall y tar geted chromophore-assisted light inacti vation. Nat Biotechnol 2003; 21:1505–8. 39. Ignatova Z, Gierasch LM. Monitoring protein stability and agg regation in vivo by real-time fluorescent labeling. Proc Natl Acad Sci U S A 2004;101:523–8. 40. Guignet EG, Hovius R, Vogel H. Reversible site-selective labeling of membrane proteins in li ve cells. Nat Biotechnol 2004; 22:440–4. 41. Kapanidis AN, Ebright YW, Ebright RH. Site-specific incorporation of fluorescent probes into protein: he xahistidine-tag-mediated fluorescent labeling with (Ni(2+): nitrilotriacetic Acid (n)-fluorochrome conjugates. J Am Chem Soc 2001;123:12123–5. 42. Hauser CT, Tsien RY. A hexahistidine-Zn2+-dye label reveals STIM1 surface exposure. Proc Natl Acad Sci U S A 2007;104:3693–7.

43. Fernandez-Suarez M, Bar uah H, Mar tinez-Hernandez L, et al. Redirecting lipoic acid ligase for cell surface protein labeling with small-molecule probes. Nat Biotechnol 2007;25:1483–7. 44. Howarth M, Takao K, Hayashi Y, Ting AY. Targeting quantum dots to surface proteins in living cells with biotin ligase. Proc Natl Acad Sci U S A 2005;102:7583–8. 45. Slavoff SA, Chen I, Choi YA, Ting AY. Expanding the substrate tolerance of biotin ligase through exploration of enzymes from diverse species. J Am Chem Soc 2008;130:1160–2. 46. Mao H, Har t SA, Schink A, P ollok B A. Sor tase-mediated protein ligation: a new method for protein engineering. J Am Chem Soc 2004;126:2670–1. 47. Gautier A, Juillerat A, Heinis C, et al. An engineered protein tag for multiprotein labeling in living cells. Chem Biol 2008;15:128–36. 48. Szent-Gyorgyi C, Schmidt BA, Creeger Y, et al. Fluorogen-activating single-chain antibodies for imaging cell surf ace proteins. Nat Biotechnol 2008;26:235–40. 49. Fischer AJ, Lagarias JC. Harnessing phytochrome’s glowing potential. Proc Natl Acad Sci U S A 2004;101:17334–9. 50. Murphy JT, Lagarias JC. The phytofluors: a new class of fluorescent protein probes. Curr Biol 1997;7:870–6. 51. Shu X, Ro yant A, Lin MZ, Aguilera TA, Lev-Ram V, Steinbach PA, Tsien RY. Mammalian expression of infrared fluorescent proteins engineered from a bacterial phytochrome. Science 2009, in press. 52. Adams SR. How calcium indicators work. In Imaging in Neuroscience and De velopment. eds. Yuste,R. & K onnerth,A. (Cold Spring Harbor: Cold Spring Harbor Laboratory Press), 2005; pp. 239–44. 53. Kao JPY , Harootunian AT, Tsien R Y. Photochemicall y generated cytosolic calcium pulses and their detection b y fluo-3. J Biol Chem 1989;264:8179–84. 54. Lakowicz JR, Szmacinski H, Johnson ML. Calcium imaging using fluorescence lifetimes and long-w avelength probes. J Fluoresc 1992;2:47–62. 55. Mason SF. Color and the electronic state of or ganic molecules. In: Venkataraman K, editor. The chemistry of synthetic dy es. Vol III. New York: Academic Press; 1970. p. 169–221. 56. Zhuang X, Ha T, Kim HD, et al. Fluorescence quenching: a tool for single-molecule protein-folding study. Proc Natl Acad Sci U S A 2000;97:14241–4. 57. Blumenthal R, Gallo SA, Viard M, et al. Fluorescent lipid probes in the study of viral membrane fusion. Chem Ph ys Lipids 2002; 116:39–55. 58. Bremer C, Bredow S, Mahmood U , et al. Optical imaging of matrix metalloproteinase-2 acti vity in tumors: feasibility study in a mouse model. Radiology 2001;221:523–9. 59. Bremer C, Tung CH, Weissleder R. In vivo molecular target assessment of matrix metalloproteinase inhibition. Nat Med 2001;7:743–8. 60. Funovics M, Weissleder R, Tung CH. Protease sensors for bioimaging. Anal Bioanal Chem 2003;377:956–63. 61. Mahmood U , Weissleder R. Near -infrared optical imaging of proteases in cancer. Mol Cancer Ther 2003;2:489–96. 62. Hama Y, Urano Y, Koyama Y, et al. A target cell-specif ic activatable fluorescence probe for in vivo molecular imaging of cancer based on a self-quenched a vidin-rhodamine conjugate. Cancer Res 2007;67:2791–9. 63. Brejc K, Sixma TK, Kitts PA, et al. Structural basis for dual excitation and photoisomerization of the Aequorea victoria green fluorescent protein. Proc Natl Acad Sci U S A 1997;94:2306–11. 64. Ward WW, Prentice HJ, Roth AF, et al. Spectral per turbations of the Aequorea green-fluorescent protein. Photochem Photobiol 1982; 35:803–8. 65. Chattoraj M, King BA, Bublitz GU, Boxer SG. Ultra-fast excited state dynamics in green fluorescent protein: multiple states and proton transfer. Proc Natl Acad Sci U S A 1996;93:8362–7. 66. Tsien RY, Prasher DC. Molecular biolo gy and mutation of GFP . In: Chalfie M, Kain S, editors. GFP: green fluorescent protein strategies and applications. New York: Wiley & Sons; 1998. pp. 97–118

Fluorescence Readouts of Bioc hemistry in Live Cells and Or ganisms

67. Llopis J, McCaffery JM, Miyawaki A, et al. Measurement of cytosolic, mitochondrial and golgi pH in single li ving cells with g reen fluorescent protein. Proc Natl Acad Sci U S A 1998;95:6803–8. 68. Yuste R, Miller RA, Holthof f K, et al. Synapto-pHluorins: chimeras between pH-sensiti ve mutants of g reen fluorescent protein and synaptic v esicle membrane proteins as repor ters of neurotransmitter release. Methods Enzymol 2000;327:522–46. 69. Jayaraman S, Haggie P, Wachter RM, et al. Mechanism and cellular applications of a g reen fluorescent protein-based halide sensor . J Biol Chem 2000;275:6047–50. 70. Dooley CM, Dore TM, Hanson GT , et al. Imaging dynamic redo x changes in mammalian cells with green fluorescent protein indicators. J Biol Chem 2004;279:22284–93. 71. Hanson GT, Aggeler R, Oglesbe D , et al. In vestigating mitochondrial redox potential with redo x-sensitive g reen fluorescent protein indicators. J Biol Chem 2004;279:13044–53. 72. Belousov VV, Fradkov AF, Lukyanov KA, et al. Geneticall y encoded fluorescent indicator for intracellular h ydrogen pero xide. Nat Methods 2006;3:281–6. 73. Baird GS, Zacharias D A, Tsien RY. Circular per mutation and receptor insertion within green fluorescent proteins. Proc Natl Acad Sci U S A 1999;96:11241–6. 74. Nagai T, Sawano A, Park E, Miyawaki A. Circularly permuted green fluorescent proteins engineered to sense Ca 2+. Proc Natl Acad Sci U S A 2001;98:3197–202. 75. Nakai J, Ohkura M, Imoto K. A high signal-to-noise Ca2+ probe composed of a single green fluorescent protein. Nat Biotechnol 2001; 19:137–41. 76. Mao T, O’Connor DH, Scheuss V, et al. Characterization and subcellular tar geting of GCaMP-type geneticall y-encoded calcium indicators. PLoS ONE 2008;3:e1796. 77. Ohkura M, Matsuzaki M, Kasai H, et al. Geneticall y encoded bright Ca2+ probe applicab le for dynamic Ca2 + imaging of dendritic spines. Anal Chem 2005;77:5861–9. 78. Reiff DF , Ihring A, Guer rero G, et al. In vi vo perfor mance of genetically encoded indicators of neural acti vity in flies. J Neurosci 2005;25:4766–78. 79. Guerrero G, Siegel MS, Roska B, et al. Tuning FlaSh: redesign of the dynamics, v oltage tange, and color of the geneticall y encoded optical sensor of membrane potential. Biophys J 2002;83:3607–18. 80. Piston DW, Kremers GJ. Fluorescent protein FRET: the good, the bad and the ugly. Trends Biochem Sci 2007;32:407–14. 81. Tsien RY. Indicators based on fluorescence resonance energy transfer. In: Yuste R, K onnerth A, editors. Imaging in neuroscience and de velopment. Cold Spring Harbor: Cold Spring Harbor Laboratory Press; 2005. p. 549–56. 82. Jovin TM, Arndt-Jovin DJ. Luminescence digital imaging microscopy. Annu Rev Biophys Biophys Chem 1989;18:271–308. 83. Gordon GW, Berry G, Liang XH, et al. Quantitative fluorescence resonance ener gy transfer measurements using fluorescence microscopy. Biophys J 1998;74:2702–13. 84. Sorkin A, McClure M, Huang F, Carter R. Interaction of EGF receptor and g rb2 in li ving cells visualized b y fluorescence resonance energy transfer (FRET) microscopy. Curr Biol 2000;10:1395–8. 85. Erickson MG, Alseikhan BA, Peterson BZ, Yue DT. Preassociation of calmodulin with v oltage-gated Ca(2+) channels re vealed b y FRET in single living cells. Neuron 2001;31:973–85. 86. Miyawaki A, Tsien RY. Monitoring protein conformations and interactions by fluorescence resonance energy transfer between mutants of green fluorescent protein. Methods Enzymol 2000;327:472–500. 87. Bastiaens PI, Squire A. Fluorescence lifetime imaging microscop y: spatial resolution of biochemical processes in the cell.Trends Cell Biol 1999;9:48–52. 88. Tramier M, Coppe y-Moisan M. Fluorescence anisotrop y imaging microscopy for homo-FRET in li ving cells. Methods Cell Biol 2008;85:395–414.

825

89. Nagai T, Yamada S, Tominaga T, et al. Expanded dynamic range of fluorescent indicators for Ca(2+) b y circularl y per muted yellow fluorescent proteins. Proc Natl Acad Sci U S A 2004; 101:10554–9. 90. Hamdan FF , P ercherancier Y, Breton B , Bouvier M. Monitoring protein-protein interactions in li ving cells b y bioluminescence resonance ener gy transfer (BRET). Cur r Protoc Neurosci 2006, Chapter 5, unit 5.23. 91. Bacart J, Corbel C, Jock ers R, et al. The BRET technolo gy and its application to screening assays. Biotechnol J 2008;3:311–24. 92. Massoud TF, Paulmurugan R, De A, et al. Reporter gene imaging of protein-protein interactions in living subjects. Curr Opin Biotechnol 2007;18:31–7. 93. Michnick SW, Ear PH, Manderson EN , et al. Universal strategies in research and dr ug discovery based on protein-fragment complementation assays. Nat Rev Drug Discov 2007;6:569–82. 94. Villalobos V, Naik S, Piwnica-W orms D. Cur rent state of imaging protein-protein interactions in vi vo with geneticall y encoded reporters. Annu Rev Biomed Eng 2007;9:321–49. 95. Ghosh I, Hamilton AD, Regan L. Antiparallel leucine zipper-directed protein reassembly: application to the g reen fluorescent protein. J Am Chem Soc 2000;122:5658–9. 96. Hu C-D, Chinenov Y, Kerppola TK. Visualization of interactions among bZIP and Rel family proteins in living cells using bimolecular fluorescence complementation. Mol Cell 2002; 9:789–98. 97. Zhang S, Ma C, Chalf ie M. Combinatorial marking of cells and organelles with reconstituted fluorescent proteins. Cell 2004; 119:137–44. 98. Hu CD , K erppola TK. Simultaneous visualization of multiple protein interactions in li ving cells using multicolor fluorescence complementation analysis. Nat Biotechnol 2003;21:539–45. 99. Cabantous S, Terwilliger TC, Waldo GS. Protein tagging and detection with engineered self-assemb ling fragments of g reen fluorescent protein. Nat Biotechnol 2005;23:102–7. 100. Langel U, editor. Handbook of cell-penetrating peptides. Boca Raton (FL): Taylor and Francis; 2007. 101. Jiang T, Olson ES, Nguy en QT, et al. Tumor imaging b y means of proteolytic activation of cell-penetrating peptides. Proc Natl Acad Sci U S A 2004;101:17867–72. 102. Goun EA, Shinde R, Dehner t KW, et al. Intracellular car go delivery by an octaar ginine transpor ter adapted to tar get prostate cancer cells through cell surf ace protease acti vation. Bioconjug Chem 2006;17:787–96. 103. Zhang Y, So MK, Rao J . Protease-modulated cellular uptak e of quantum dots. Nano Lett 2006;6:1988–92. 104. Olson ES, Jiang T, Scadeng M, Ellies LG, Nguyen QT, Aguilera TA, Tsien RY. ACPPs attached to dendrimeric nanoparticles: development of dual probes for fluorescence and magnetic resonance imaging of proteases in vivo. 105. Whitney M, Olson ES, Aguilera TA, Crisp JL, Gross LA, Ellies LG, Tsien RY. Parallel in vivo and in vitro selection using phage display identifies protease dependent tumor targeting peptides. 106. Dolmetsch RE, Gomez-Ospina N , Green E, Nigh EA. Imaging gene expression in live cells and tissues. In:Yuste R, Konnerth A, editors. Imaging in neuroscience and de velopment. Cold Spring Harbor: Cold Spring Harbor Laboratory Press; 2005. p. 605–17. 107. Chalfie M, Tu Y, Euskirchen G, et al. Green fluorescent protein as a marker for gene expression. Science 1994;263:802–5. 108. Barth AL. Visualizing circuits and systems using transgenic reporters of neural activity. Curr Opin Neurobiol 2007;17:567–71. 109. Yu J, Xiao J, Ren X, et al. Probing gene e xpression in li ve cells, one protein molecule at a time. Science 2006;311:1600–3. 110. Li X, Zhao X, F ang Y, et al. Generation of destabilized g reen fluorescent protein as a transcription repor ter. J Biol Chem 1998; 273:34970–5.

826

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

111. Dantuma NP, Lindsten K, Glas R, et al. Short-lived green fluorescent proteins for quantifying ubiquitin/proteasome-dependent proteolysis in living cells. Nat Biotechnol 2000;18:538–43. 112. Lichtman JW, Livet J, Sanes JR. A technicolour approach to the connectome. Nat Rev Neurosci 2008;9:417–22. 113. Livet J, Weissman TA, Kang H, et al. Transgenic strategies for combinatorial e xpression of fluorescent proteins in the ner vous system. Nature 2007;450:56–62. 114. Babendure JR, Babendure JL, Ding JH, Tsien R Y. Control of mammalian translation by mRNA structure near caps. RNA 2006; 12:851–61. 115. Bouabe H, Fassler R, Heesemann J . Improvement of repor ter activity b y IRES-mediated pol ycistronic repor ter system. Nucleic Acids Res 2008;36:e28. 116. Weinberger LS, Bur nett JC, Toettcher JE, et al. Stochastic gene expression in a lenti viral positive-feedback loop: HIV-1 Tat fluctuations drive phenotypic diversity. Cell 2005;122:169–82. 117. Straight AF, Marshall WF, Sedat JW, Mur ray AW. Mitosis in li ving budding yeast: anaphase A but no metaphase plate. Science 1997; 277:574–8. 118. Kumaran RI, Spector DL. A genetic locus tar geted to the nuclear periphery in living cells maintains its transcriptional competence. J Cell Biol 2008;180:51–65. 119. Querido E, Char trand P. Using fluorescent proteins to study mRN A trafficking in living cells. Methods Cell Biol 2008;85:273–92. 120. Elf J, Li GW, Xie XS. Probing transcription f actor dynamics at the single-molecule level in a living cell. Science 2007;316:1191–4. 121. Daunert S, Barrett G, Feliciano JS, et al. Genetically engineered wholecell sensing systems: coupling biolo gical recognition with repor ter genes. Chem Rev 2000;100:2705–38. 122. Huang Z, Terpetschnig E, You W, Haugland RP . 2-(2’-Phosphor yloxyphenyl)-4(3H)-quinazolinone derivatives as fluoro genic precipitating substrates of phosphatases. Anal Biochem 1992; 207:32–9. 123. Kamiya M, Kobayashi H, Hama Y, et al. An enzymatically activated fluorescence probe for tar geted tumor imaging. J Am Chem Soc 2007;129:3918–29. 124. Zlokarnik G, Ne gulescu PA, Knapp TE, et al. Quantitation of transcription and clonal selection of single living cells using β-lactamase as reporter. Science 1998;279:84–8. 125. Wildt S, Deuschle U. cobA, a red fluorescent transcriptional reporter for Escherichia coli, yeast, and mammalian cells. Nat Biotechnol 1999;17:1175–8. 126. Knapp T, Hare E, Feng L, et al. Detection of beta-lactamase reporter gene expression by flow cytometry. Cytometry 2003;51A:68–78. 127. Raz E, Zlokar nik G, Tsien RY, Drie ver W. Beta-Lactamase as a marker for gene e xpression in li ve zebraf ish embr yos. Dev Biol 1998;203:290–4. 128. Wolff M, Wiedenmann J, Nienhaus GU, et al. Novel fluorescent proteins for high-content screening. Dr ug Disco v Today 2006; 11:1054–60. 129. Terskikh A, Fradkov A, Ermakova G, et al. ‘‘Fluorescent timer’’: protein that changes color with time. Science 2000;290:1585–8. 130. Verkhusha VV, Chudakov DM, Gurskaya NG, et al. Common pathway for the red chromophore for mation in fluorescent proteins and chromoproteins. Chem Biol 2004;11:845–54. 131. Kaether C, Gerdes HH. Visualization of protein transpor t along the secretory pathw ay using g reen fluorescent protein. FEBS Lett 1995;369:267–71. 132. Lippincott-Schwartz J, Patterson GH. Fluorescent proteins for photoactivation experiments. Methods Cell Biol 2008;85:45–61. 133. Shaner NC, P atterson GH, Da vidson MW. Advances in fluorescent protein technology. J Cell Sci 2007;120:4247–60. 134. Stark DA, K ulesa PM. An in vi vo comparison of photoacti vatable fluorescent proteins in an a vian embr yo model. De v Dyn 2007;236:1583–94.

135. Baskin JM, Prescher JA, Laughlin ST, et al. Copper-free click chemistry for dynamic in vi vo imaging. Proc Natl Acad Sci U S A 2007;104:16793–7. 136. Lin MZ, Glenn JS, Tsien RY. A drug-controllable tag for visualizing newly synthesized proteins in cells and w hole animals. Proc Natl Acad Sci U S A 2008;105:7744–9. 137. Schweizer FE, Ryan TA. The synaptic v esicle: cycle of e xocytosis and endocytosis. Curr Opin Neurobiol 2006;16:298–304. 138. Brumback AC, Lieber JL, Angleson JK, Betz WJ. Using FM1-43 to study neuropeptide g ranule dynamics and e xocytosis. Methods 2004;33:287–94. 139. Miesenböck G. A practical guide: Synapto-pHluorins— Geneticall y encoded reporters of synaptic transmission. In: Yuste R, Konnerth A, editors. Imaging in neuroscience and de velopment. Cold Spring Harbor: Cold Spring Harbor Laborator y Press; 2005. p. 599–603. 140. Bozza T, McGann JP, Mombaerts P, Wachowiak M. In vivo imaging of neuronal acti vity b y tar geted e xpression of a geneticall y encoded probe in the mouse. Neuron 2004;42:9–21. 141. Babendure JR, Adams SR, Tsien RY. Aptamers s witch on fluorescence of triphen ylmethane dy es. J Am Chem Soc 2003;125: 14716–7. 142. Rackham O, Brown CM. Visualization of RNA-protein interactions in li ving cells: FMRP and IMP1 interact on mRN As. EMBO J 2004;23:3346–55. 143. Tsien RY. Fluorescent indicators of ion concentrations. Methods Cell Biol 1989;30:127–56. 144. Miesenböck G, De Angelis DA, Rothman JE. Visualizing secretion and synaptic transmission with pH-sensiti ve g reen fluorescent proteins. Nature 1998;394:192–5. 145. McAnaney TB, Shi XH, Abbyad P, et al. Green fluorescent protein variants as ratiometric dual emission pH sensors. 3. Temperature dependence of proton transfer. Biochemistry 2005;44:8701–11. 146. Knopfel T, Diez-Garcia J , Akemann W. Optical probing of neuronal circuit dynamics: genetically encoded versus classical fluorescent sensors. Trends Neurosci 2006;29:160–6. 147. Ratto GM, P ayne R, Ow en WG, Tsien RY. The concentration of cytosolic free calcium in v ertebrate rod outer segments measured with fura-2. J Neurosci 1988;8:3240–6. 148. Garaschuk O, Milos RI, Grienber ger C, et al. Optical monitoring of brain function in vi vo: from neurons to netw orks. Pflugers Arch 2006;453:385–96. 149. Ohki K, Chung S, Kara P, et al. Highly ordered ar rangement of single neurons in orientation pinwheels. Nature 2006;442:925–8. 150. Tymianski M, Spigelman I, Zhang L, et al. Mechanism of action and persistence of neuroprotection b y cell-per meant Ca 2+ chelators J Cereb Blood Flow Metab 1994;14:911–23. 151. Kotlikoff MI. Geneticall y encoded Ca 2+ indicators: using genetics and molecular design to understand comple x physiology. J Physiol 2007;578:55–67. 152. Palmer AE, Giacomello M, K ortemme T, et al. Ne w Ca 2+ indicators based on computationall y-redesigned calmodulin-peptide pairs. Chem Biol 2006;13:521–30. 153. Garaschuk O, Griesbeck O , Konnerth A. Troponin C-based biosensors: a ne w f amily of geneticall y encoded indicators for in vi vo calcium imaging in the ner vous system. Cell Calcium 2007; 42:351–61. 154. Tour O, Adams SR, Kerr RA, et al. Calcium Green FlAsH as a genetically targeted small-molecule calcium indicator . Nat Chem Biol 2007;3:423–31. 155. Arslan P, Di Virgilio F, Beltrame M, et al. Cytosolic Ca 2+ homeostasis in Ehrlich and Yoshida carcinomas. J Biol Chem 1985; 260:2719–27. 156. Tomsig JL, Suszkiw JB . Pb 2+-induced secretion from bo vine chro2+ maffin cells: fura-2 as a probe for Pb . Am J Ph ysiol 1990; 259:C762–8.

Fluorescence Readouts of Bioc hemistry in Live Cells and Or ganisms

157. Hinkle PM, Shanshala EDI, Nelson EJ. Measurement of intracellular cadmium with fluorescent dy es. Further evidence for the role of calcium channels in cadmium uptak e. J Biol Chem 1992; 267:25553–9. 158. Marchi B, Burlando B, Panfoli I, Viarengo A. Interference of hea vy metal cations with fluorescent Ca 2+ probes does not af fect Ca 2+ measurements in living cells. Cell Calcium 2000;28:225–31. 159. Atar D, Backx PH, Appel MM, et al. Excitation-transcription coupling mediated by zinc influx through v oltage-dependent calcium channels. J Biol Chem 1995;270:2473–7. 160. Burdette SC, Walkup GK, Spingler, B, et al. New fluorescent sensors for Zn2+ based on a fluorescein platform: synthesis, properties and intracellular distribution. J Am Chem Soc 2001;123:7831–41. 161. Minta A, Tsien R Y. Fluorescent indicators for c ytosolic sodium. J Biol Chem 1989;264:19449–57. 162. Pearce LL, Gandle y RE, Han W, et al. Role of metallothionein in nitric o xide signaling as re vealed b y a ne w g reen fluorescent fusion protein. Proc Natl Acad Sci U S A 2000;97:477–82. 163. Bozym RA, Thompson RB , Stoddard AK, F ierke CA. Measuring picomolar intracellular e xchangeable zinc in PC-12 cells using a ratiometric fluorescence biosensor . ACS Chem Biol 2006; 1:103–11. 164. Evers TH, Appelhof MA, Graaf-Heuv elmans PT, et al. Ratiometric detection of Zn(II) using chelating fluorescent protein chimeras. J Mol Biol 2007;374:411–25. 165. van Dongen EM, Ev ers TH, Dekkers LM, et al. Variation of link er length in ratiometric fluorescent sensor proteins allo ws rational tuning of Zn(II) af finity in the picomolar to femtomolar range. J Am Chem Soc 2007;129:3494–5. 166. Biwersi J, Farah N, Wang Y-X, et al. Synthesis of cell-imper meable Cl-sensitive fluorescent indicators with impro ved sensitivity and optical properties. Am J Physiol 1992;262:C243–50. 167. Verkman AS. Development and biolo gical applications of chloridesensitive fluorescent indicators. Am J Physiol 1990;259:C375–88. 168. Kuner T, Augustine GJ. A genetically encoded ratiometric indicator for chloride: capturing chloride transients in cultured hippocampal neurons. Neuron 2000;27:447–59. 169. Markova O, Mukhtarov M, Real E, et al. Geneticall y encoded chloride indicator with impro ved sensiti vity. J Neurosci Methods 2008;170:67–76. 170. Gu H, Lalonde S, Okumoto S, et al. A novel analytical method for in vivo phosphate tracking. FEBS Lett 2006;580:5885–93. 171. Medintz IL. Recent progress in developing FRET-based intracellular sensors for the detection of small molecule nutrients and ligands. Trends Biotechnol 2006;24:539–42. 172. Hires SA, Zhu Y, Tsien RY. Optical measurement of synaptic glutamate spillover and reuptak e by linker optimized glutamate-sensitive fluorescent repor ters. Proc Natl Acad Sci U S A 2008; 105:4411–6. 173. Dulla C, Tani H, Okumoto S, et al. Imaging of glutamate in brain slices using FRET sensors. J Neurosci Methods 2008;168:306–19. 174. Lohse MJ, Bunemann M, Hof fmann C, et al. Monitoring receptor signaling b y intramolecular FRET . Cur r Opin Phar macol 2007; 7:547–53. 175. Adams SR, Bacskai BJ , Taylor SS, Tsien RY. Optical probes for cyclic AMP. In: Mason WT, editor. Fluorescent probes for biological activity of living cells — a practical guide. Ne w York: Academic Press; 1993. p. 133–49. 176. Bacskai BJ, Hochner B, Mahaut-Smith M, et al. Spatiall y resolved dynamics of cAMP and protein kinase A subunits in Aplysia sensory neurons. Science 1993;260:222–6. 177. Hempel CM, Vincent P, Adams SR, et al. Spatio-temporal dynamics of cAMP signals in an intact neural circuit. Nature 1996; 384:166–9. 178. Zaccolo M, De Gior gi F, Cho CY , et al. A geneticall y encoded , fluorescent indicator for cyclic AMP in living cells. Nat Cell Biol 2000;2:25–9.

827

179. Zaccolo M, P ozzan T. Discrete microdomains with high concentration of cAMP in stimulated rat neonatal cardiac m yocytes. Science 2002;295:1711–5. 180. DiPilato LM, Cheng X, Zhang J . Fluorescent indicators of cAMP and Epac activation reveal differential dynamics of cAMP signaling within discrete subcellular compartments. Proc Natl Acad Sci U S A 2004;101:16513–8. 181. Nikolaev VO, Gambaryan S, Engelhardt S, et al. Real-time monitoring of the PDE2 activity of live cells: hormone-stimulated cAMP hydrolysis is f aster than hor mone-stimulated cAMP synthesis. J Biol Chem 2005;280:1716–9. 182. van der Krogt GN, Ogink J, Ponsioen B, Jalink K. A comparison of donor-acceptor pairs for geneticall y encoded FRET sensors: application to the Epac cAMP sensor as an e xample. PLoS ONE 2008;3:e1916. 183. Honda A, Adams SR, Sawyer CL, et al. Spatiotemporal dynamics of guanosine 3 ʹ′,5ʹ′-cyclic monophosphate re vealed b y a geneticall y encoded, fluorescent indicator . Proc Natl Acad Sci U S A 2001; 98:2437–42. 184. Nikolaev VO, Gambaryan S, Lohse MJ. Fluorescent sensors for rapid monitoring of intracellular cGMP. Nat Methods 2006;3:23–5. 185. Nausch LW, Ledoux J , Bone v AD, et al. Dif ferential patter ning of cGMP in v ascular smooth muscle cells re vealed by single GFPlinked biosensors. Proc Natl Acad Sci U S A 2008;105:365–70. 186. Matsu-ura T, Michika wa T, Inoue T, et al. Cytosolic inositol 1,4, 5-trisphosphate dynamics during intracellular calcium oscillations in living cells. J Cell Biol 2006;173:755–65. 187. Murakoshi H, Iino R, K obayashi T, et al. Single-molecule imaging analysis of Ras activation in living cells. Proc Natl Acad Sci U S A 2004;101:7317–22. 188. Bivona TG, Quatela S, Philips MR. Analysis of Ras activation in living cells with GFP-RBD. Methods Enzymol 2006;407:128–43. 189. Hodgson L, Pertz O, Hahn KM. Design and optimization of genetically encoded fluorescent biosensors: GTP ase biosensors. Methods Cell Biol 2008;85:63–81. 190. Kitano M, Naka ya M, Nakamura T, et al. Imaging of Rab5 acti vity identifies essential re gulators for phagosome maturation. Nature 2008;453:241–5. 191. Lohse MJ, Nikolaev VO, Hein P, et al. Optical techniques to analyze real-time activation and signaling of G-protein-coupled receptors. Trends Pharmacol Sci 2008;29:159–65. 192. Ostergaard H, Henriksen A, Hansen FG, Winther JR. Shedding light on disulfide bond formation: engineering a redox switch in green fluorescent protein. EMBO J 2001;20:5853–62. 193. Hansen RE, Oster gaard H, Winther JR. Increasing the reacti vity of an ar tificial dithiol-disulf ide pair through modif ication of the electrostatic milieu. Biochemistry 2005;44:5899–906. 194. Cannon MB , Remington SJ . Re-engineering redo x-sensitive g reen fluorescent protein for improved response rate. Protein Sci 2006; 15:45–57. 195. Bjornberg O, Ostergaard H, Winther JR. Mechanistic insight provided by glutaredoxin within a fusion to redo x-sensitive yellow fluorescent protein. Biochemistry 2006;45:2362–71. 196. Lopez-Mirabal HR, Winther JR. Redox characteristics of the eukaryotic cytosol. Biochim Biophys Acta 2008;1783:629–40. 197. Srikun D , Miller EW , Domaille D W, Chang CJ . An ICT -based approach to ratiometric fluorescence imaging of h ydrogen peroxide produced in li ving cells. J Am Chem Soc 2008; 130:4596–7. 198. Tanaka K, Miura T, Umeza wa N , et al. Rational design of fluorescein-based fluorescence probes. Mechanism-based design of a maximum fluorescence probe for singlet oxygen. J Am Chem Soc 2001;123:2530–6. 199. Kenmoku S, Urano Y, Kojima H, Nagano T. Development of a highly specific rhodamine-based fluorescence probe for h ypochlorous acid and its application to real-time imaging of phagoc ytosis. J Am Chem Soc 2007;129:7313–8.

828

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

200. Koide Y, Urano Y, Kenmoku S, et al. Design and synthesis of fluorescent probes for selecti ve detection of highl y reacti ve o xygen species in mitochondria of li ving cells. J Am Chem Soc 2007; 129:10324–5. 201. Namiki S, Kakizawa S, Hirose K, Iino M. No signalling decodes frequency of neuronal acti vity and generates synapse-specif ic plasticity in mouse cerebellum. J Physiol 2005;566:849–63. 202. Sasaki E, K ojima H, Nishimatsu H, et al. Highl y sensiti ve near infrared fluorescent probes for nitric o xide and their application to isolated organs. J Am Chem Soc 2005;127:3684–5. 203. St Croix CM, Stitt MS, Watkins SC, Pitt BR. Fluorescence resonance energy transfer-based assays for the real-time detection of nitric oxide signaling. Methods Enzymol 2005;396:317–26. 204. Cravatt BF, Wright AT, Kozarich JW. Activity-based protein prof iling: from enzyme chemistr y to proteomic chemistr y. Annu Rev Biochem 2008;77:383–414. 205. Blum G, v on De genfeld G, Merchant MJ , et al. Nonin vasive optical imaging of cysteine protease activity using fluorescently quenched acti vity-based probes. Nat Chem Biol 2007; 3:668–77. 206. Rehm M, Dussmann H, Janick e RU, et al. Single-cell fluorescence resonance energy transfer analysis demonstrates that caspase activation during apoptosis is a rapid process. Role of caspase-3. J Biol Chem 2002;277:24506–14. 207. Ganesan S, Ameer-Beg SM, Ng TT, et al. A dark y ellow fluorescent protein (YFP)-based Resonance Ener gy-Accepting Chromoprotein (REACh) for Forster resonance energy transfer with GFP. Proc Natl Acad Sci U S A 2006;103:4089–94.

208. Sakaue-Sawano A, K urokawa H, Morimura T, et al. Visualizing spatiotemporal dynamics of multicellular cell-c ycle progression. Cell 2008;132:487–98. 209. Zhang J, Allen MD. FRET-based biosensors for protein kinases: illuminating the kinome. Mol Biosyst 2007;3:759–65. 210. Zhang J , Hupfeld CJ , Taylor SS, et al. Insulin disr upts betaadrenergic signalling to protein kinase A in adipoc ytes. Nature 2005;437:569–73. 211. Newman RH, Zhang J. Visualization of phosphatase activity in living cells with a FRET-based calcineurin activity sensor. Mol Biosyst 2008;4:496–501. 212. Lalonde S, Ehrhardt D W, Loque D , et al. Molecular and cellular approaches for the detection of protein-protein interactions: latest techniques and current limitations. Plant J 2008;53:610–35. 213. Briddon SJ, Hill SJ. Pharmacology under the microscope: the use of fluorescence correlation spectroscopy to determine the properties of ligand-receptor comple xes. Trends Phar macol Sci 2007;28: 637–45. 214. Haustein E, Schwille P. Fluorescence correlation spectroscopy: novel variations of an established technique. Annu Rev Biophys Biomol Struct 2007;36:151–69. 215. Bates M, Huang B , Dempse y GT , Zhuang X. Multicolor super resolution imaging with photo-s witchable fluorescent probes. Science 2007;317:1749–53. 216. Shroff H, Galbraith CG, Galbraith JA, et al. Dual-color superresolution imaging of geneticall y e xpressed probes within indi vidual adhesion comple xes. Proc Natl Acad Sci U S A 2007;104: 20308–13.

49 IMAGING

OF

SIGNALING PATHWAYS

MAHAVEER S. BHOJANI, PHD, BRIAN D. ROSS, PHD, AND ALNAWAZ REHEMTULLA, PHD

Advances in cellular and molecular biology have resulted in the identification of key components of signaling pathways in normal and in pathophysiology. Among the identified biomarkers, kinases and proteases for m the largest groups of proteins with k ey roles in a number of processes ranging from embryogenesis to death. Furthermore, dysre gulated kinase and proteol ytic acti vity is often observed in diseases such as cancer , viral and bacterial pathogenesis, cardiovascular and neurolo gical disorders, and asthma. The ability to monitor these enzymes noninvasively would provide novel insights into their role in the abo ve-mentioned pathologies and in nor mal biology. In this chapter , we describe no vel molecular imaging–based tools for monitoring kinases and proteases. These repor ters allow real-time, dynamic, and quantitative imaging of protein kinase or protease acti vity in cell culture and li ving animals. These technologies influence the process of dr ug discovery by enabling the v alidation of dr ug–target interaction as w ell as in the preclinical determination of optimal dr ug dosage, schedule, and optimization of combination therapies.

PROTEIN KINASES Protein kinases are enzymes that control the plurality of cellular decisions and dynamic beha vior of eukar yotic cells.1–6 These enzymes posttranslationally modify substrate proteins b y catalyzing the co valent addition of a negatively charged phosphate group from ATP to a specific amino acid. 3,5 In eukaryotes, protein kinases phosphorylate predominantl y serine or threonine (protein Ser/Thr kinases) as w ell as tyrosine residues (protein Tyr kinase). 3–6 Although phosphorylation at its residues is a cr ucial component of signaling in prokar yotes and lower eukaryotes, there are very few reports of its phos7,8 phorylation in mammalian signaling cascades. Phosphorylation of target residues in proteins results in

changes in substrate acti vity, subcellular localization, and/or interaction with other macromolecules. 9 These changes mediate the bulk of signaling in normal eukaryotic cells and are v ery tightl y controlled b y inhibitor y and re gulatory constraints that act as a safe guard for aberrant kinase activation.3–6,10 Protein phosphorylation by kinases is the most common type of cellular re gulation that influences the coordination of essentially every biological processes such as cell g rowth and di vision, cell c ycle pro gression, membrane transpor t, motility , transcription, translation, c ytoskeletal rear rangement, learning, memor y, and apoptosis. 2,11–15 Furthermore, kinases pla y a critical role in inter - and intracellular communication in ph ysiological responses and homeostasis and in the ef ficient functioning of v arious systems including digesti ve, ner vous, and immune systems.16–21 Sequencing of the human genome has shown the existence of at least 518 protein kinases that constitute around 1.7% of all proteins, most of which fall into a single super family.2,22 Interestingly, more than 30% of the total proteins in the cell are estimated to be phosphor ylated transiently suggesting that a single kinase ma y have multiple substrates.23 Conversely, man y proteins are phosphor ylated by more than one kinase (for example, β-catenins are reported to be phosphorylated by different kinases namely CK-1α,24–26 GSK-3β,27 protein kinase G,28 Akt,29 Nek230). Therefore, for proper control of the comple x inte grated signal transduction pathw ays, kinases must be specif ic and phosphorylate only a defined subset of cellular target proteins. This precision in specif icity is extremely essential for maintaining the inte grity of signal transduction process. Although our understanding of ho w such specificity is generated is patch y, some basic concepts ha ve emerged from decades of tar geted biochemical studies and recent proteomic and phosphoproteomic studies. 31–42 It appears that a par t of protein kinase specif icity is 829

830

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

generated by the existence of a consensus sequence motif within the substrate protein that is recognized by the active site of the kinase catal ytic domains. 43,44 However, additional f actors that ma y influence the specif icity of the protein kinases include subcellular compartmentalization, co-localization via anchoring proteins and scaf folds, substrate capture b y noncatal ytic interaction domains (eg, SH2 domains), temporal and cell type–specif ic coexpression, kinase-docking motifs within substrates and regulatory subunits.45–52 Phosphorylation of the consensus sequence within the substrate proteins is also controlled b y protein phosphatases, which catalytically excise the phosphate g roup at a specif ic sequence. 53 The human genome is repor ted to contain more than 130 protein phosphatases that e xert a tight and re versible control on protein phosphor ylation.54–56 On the basis of their specif icity, both protein kinases and phosphatases can be subdi vided into tw o groups one that is specific for tyrosine residues and other that is serine/threonine specif ic.57–59 In addition, some possess dual specificity for both tyrosine and serine/threonine, and a fe w members of the phosphatidylinositol kinase family also exhibit protein-serine/threonine kinase activity.54 Although protein phosphatases pla y an important role in the remo val of phosphate g roups added b y kinases, they have been relegated to the role of a second fiddle to kinases. Onl y recently the onco genic potential of the protein phosphatases have been reported, and these group of enzymes are being tar geted for cancer therapy.33,55,58

Protein Kinases in Cancer Dysregulation and mutation of kinases has been reported to play a causal role in man y human diseases such as cancer , rheumatoid arthritis, cardiovascular and neurological disorders, asthma, and psoriasis. 10,60–64 Oncological signaling has received a lion’s share of research attention and has benefited the most from rapid technolo gical developments in molecular profiling of pathological samples and corroborative de velopment in bioinfor matics.65–69 This has led to identification of a v ariety of potential pro gnostic biomarkers. Protein kinases, among the biomarkers identified, have gained much attention for the de velopment of agonists and antagonists with therapeutic potentials. Therefore, it is no surprise that the first molecularly targeted drug approved by the Food Drug Administration for treatment of cancers such as chronic myelogenous leukemia (CML) and gastrointestinal stromal tumors (GISTs) is a tyrosine kinase inhibitor specific for ab l (the Abelson protoonco gene), c-kit, and

PDGF-R (platelet-deri ved g rowth f actor receptor). 70 Erlotinib, Avastin, Lapatinib, and Herceptin (all kinase inhibitors) are other high-prof ile b lockbuster molecular drugs that are no w available in the clinician’ s arsenal for therapeutic intervention of human malignancies. Additionally, a number of other kinase inhibitors are at v arious stages of preclinical or clinical e valuation. However, only 5% of cancer dr ugs entering clinical trials reach mark eting approval71 and man y potential cancer tar gets remain undrugged. Fur thermore, it is w ell documented that with the exception of Imatinib, that most kinase inhibitors such as Erlotinib, Lapatinib, Cetuximab, ABX-EGF, Be vacizumab evaluated for control of human malignanc y have relatively modest acti vity as a single agent. 72,73 This suggests that b lockade of a single kinase alone might not be sufficient to achieve a clinical benefit and that multitargeted kinase inhibitors might be more promising than selecti ve agents, both with respect to ef ficacy and pre vention of resistance. Successful treatments need to not only optimize the combination and dosage of the therapeutic agents but also the schedule of the indi vidual agents. For example, in a preclinical model of head and neck cancer , it was shown that gemcitabine follo wed by gef itinib but not the re verse resulted in enhanced ef ficacy.74 Such understanding in the clinical setting has been lacking and has had its toll in failed therapeutic trials. The use of tar geted agents in oncolo gy would therefore significantly benefit from the use of molecular imaging in the preclinical setting with the goal of identifying the most ef ficacious agents that modulate the desired pathways in a targeted manner with minimal off-target ef fects. These model systems w ould also pro vide the ability to screen lar ge number of molecules so that the safest and most biologically relevant derivative is ultimately translated to the clinic.

Molecular Imaging of Protein Kinases Screening and identif ication of new kinase inhibitors from a library of chemical compounds requires robust methods. Existing methodologies to monitor kinase activity use techniques such as w estern b lotting or immunoc ytochemistry using a phospho-specific target antibody, and radioactive in vitro kinase assa ys. However, these methods are in vasive, cumbersome, and only provide a snapshot view of the activity at a specif ic time point. Additionally, some of these assays are suitable only for in vitro studies and may not be able to reproduce the in vi vo specif icity and acti vity of a kinase. Recent discoveries in the field of molecular imaging may aid in o vercoming such limitations of con ventional methodologies. The f ield of molecular imaging

Imaging of Signaling Pathways

encompasses the nonin vasive visual representation of biological processes at the cellular and molecular le vel in the whole organism and the modalities and instrumentation to suppor t the visualization and measurement of these processes.75 These imaging technolo gies are an attempt to bridge the gap between discovery of causal disease markers and identif ication of their inhibitor for potential therapeutic use.76 In the last decade, at least three dif ferent molecular imaging technologies have been used for the understanding of disease biomark ers, dr ug de velopment, or monitoring therapeutic outcome. They are (1) optical imaging (bioluminescence and fluorescence imaging) (2) magnetic resonance imaging (MRI), and (3) nuclear imaging (eg, single photon emission computed tomo graphy (SPECT) and positron emission tomo graphy (PET)), which are described in detail else where in this book and other re views.76 These imaging technolo gies ha ve emerged as po werful tools that enab le real-time, dynamic, and quantif iable monitoring of gene e xpression, signal transduction, protein–protein interaction, and cell traf ficking in intact cells and li ving animals. Thus, molecular imaging allo ws a direct cell-based measurement of signaling cascades and its pla yers, thereby eliminating the need for time-consuming dissection and histological methods for tissue anal ysis. Fur thermore, these tools allow noninvasive imaging and quantif ication of kinase activity in living cells and subjects, w hich will aid in preclinical determination of drug dosage, schedule, and combination. In addition, imaging of kinases that play a k ey role in cell proliferation and g rowth, such as Akt, could be e xploited as a sur rogate marker for monitoring real-time response of cancer cells to therapeutic regimens. The applications of molecular imaging assa ys in vitro also provide an excellent platform for cell-based high-throughput screening. Inhibitors identif ied by such assays typically pass the barriers of solubility, membrane permeability, and toxicity, thus enhancing their chance of being successful in e xperimental and clinical in vestigation. In summary, the development of molecular imaging tools for kinases w ould signif icantly enhance our understanding of the biology of cancer and assist in the de velopment and evaluation of novel therapeutic agents.

Platform for Bioluminescent Kinase Imaging We recently reported the development of a bioluminescent reporter to monitor Akt/PKB,77 a serine/threonine kinase, which plays a crucial role in tumor initiation, progression,

831

neoangiogenesis, and resistance to cancer therap y in a cross-section of human malignancies. This repor ter w as based on confor mation-dependent complementation of firefly luciferase. Protein complementation assa ys ha ve their roots in protein engineering strate gies w herein a monomeric repor ter is split into tw o separate inacti ve components in such a w ay that w hen these components are brought into close proximity they reconstitute the original repor ter activity. Protein complementation has been extensively exploited to understand protein–protein interaction. Historicall y, the y east tw o h ybrid system f irst described by Stanley Fields78 is based on protein complementation of GAL4, a transcriptional acti vator protein. This discovery revolutionized the understanding of signaling cascades in eukaryotic organisms. Although a number of interacting par tners in a v ariety of signaling cascades were discovered, this system has limited utility in the context of a living cell or subject. A number of reporters routinely used in understanding mammalian biolo gy w ere engineered for complementation studies. These include fluorescent proteins (GFP and YFP), bioluminescent enzymes (f irefly luciferase and renilla luciferase), β-galactosidase, dih ydrofolate reductase (DHFR), and TME-1 β-lactamase.9,79–85 Of the dif ferent complementation assa ys, the bioluminescence repor ter has emer ged as a useful technique for small animal imaging. Luciferase is a photoprotein that modif ies the substrate (luciferin) b y releasing photons in the presence of o xygen and ATP. The light emitted by f irefly luciferase appears blue to yellow green in color with an emission spectra that peaks at wavelength between 490 and 620 nm. 86 There are more than 30 luciferase–luciferin systems of independent origin, but the most used luciferase for in vivo molecular imaging is the ATP-dependent firefly (Photinus pyralis) luciferase.87 This is chiefly due to the f act that 30% of the light generated by firefly luciferase has an emission spectra above 600 nm, a re gion w here the signal attenuation b y the absorption and scattering properties of mammalian tissue is minimal.87,88 This is a major advantage compared with other optical imaging systems such as fluorescence imaging w herein the e xcitation light can also e xcite other naturally occur ring fluorescent molecules in the body, w hich ma y result in a high le vel of backg round autofluorescence. An optimized f irefly luciferase protein fragment complementation w as de veloped b y screening incremental tr uncation libraries of N- and C-ter minal fragments of luciferase. 89 The N-ter minal and C-ter minal luciferase fragments were fused with FRB of the mammalian target of rapamycin and FK506-binding protein

832

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

12 (FKBP), respecti vely. The optimized pair of FRB–N-Luc/C-Luc–FKBP reconstituted luciferase activity upon single-site binding of rapam ycin in an FK506-competitive manner. By using this strate gy, the investigators monitored protein–protein interaction, such as the phosphor ylation-dependent interaction between human Cdc25C and 14-3-3 ε in vivo. Paulmurugan and Gambhir 90 designed a complementationbased assa y using renilla luciferase b y e xploiting the strong interaction of My oD and Id. Recentl y, a nonATP–dependent Gaussia princeps luciferase enzyme was engineered for protein complementation assay that showed a crosstalk between insulin and TGFβ signaling pathways.83 With the goal of de veloping a prototype repor ter for kinase activity, we created a bioluminescent Akt reporter (BAR), which is a recombinant chimeric protein consisting of an Akt consensus substrate peptide and the phosphorylated amino acid binding domain from FHA2, flanked by the amino- (N-Luc) and carboxyl- (C-Luc) terminal domains of the f irefly luciferase repor ter molecule (Figure).77 In the presence of Akt kinase acti vity, phosphorylation of the Akt consensus substrate sequences within the reporter results in its interaction with the FHA2 domain, thus stearicall y pre venting reconstitution of a functional luciferase repor ter molecule. In the absence of Akt kinase activity, release of this stearic constraint allows reconstitution of the luciferase repor ter molecule w hose

activity can be detected nonin vasively b y bioluminescence. In contrast to the FRET -based repor ters, w hich allow reporter activity to be monitored in single cells, the BAR and its adaptation for other kinase acti vites allows imaging in not onl y live cells but also animals in a quantitative, dynamic, and noninvasive manner.

Utility of Bioluminescent Kinase Reporter A kinase repor ter may be used to study the specif icity of the drug in inhibiting the target kinase. For example, in the case of B AR, treatment of cells with an Akt inhibitor (API-2) and a PI-3K inhibitor (perifosine) resulted in an increase of bioluminescence acti vity in a time- and dose-dependent manner , which indicates that BAR provides a sur rogate for Akt acti vity in ter ms of quantity and dynamics. This f inding was cor roborated with conventional western b lotting using antibodies to phosphorylated Akt. 77 Bioluminescent kinase repor ters also aid in in vestigating upstream signaling e vents that impinge on the kinase acti vity. To demonstrate this point, w hen w e treated the B AR repor ter–expressing cells with EGF, changes in BAR bioluminescent activity was detected.77 This suggested that activation of EGFR, which has been documented to feed into the Akt cascade, can be monitored b y Akt imaging and that the BAR activity can act as a surrogate for EGFR signaling. Utilization of the BAR reporter for monitoring upstream

Figure 1. Diagrammatic representation of the structure of the AKT kinase reporter and the basis of Akt kinase reporter activity. A, The chimeric Akt kinase reporter is a fusion of N-Luc and C-Luc components of luciferase complementation system89 separated by the Aktpep domain that harbors a consensus Akt substrate sequence.124 On either side of the substrate sequence, flexible linker sequence was included (GGSGG), and at the amino-terminal of the Aktpep domain, the yeast FHA2 phospho-Ser/Thr binding domain (reisdues 420 to 582)125 was included. B, Phosphorylation of the Akt peptide within the reporter results in interaction with the FHA2 phospho-peptide binding domain causing stearic constraints on the C-luc and N-luc. Inhibition of the Akt kinase results in decreased binding of phospho-peptide and phospho-peptide binding domain enabling the N-Luc and C-Luc interaction to restore bioluminescence.

Imaging of Signaling Pathways

signaling was further examined by the use of an EGFR inhibitor, Erlotinib. Cells treated with Erlotinib sho wed a significant and reproducible increase in BAR bioluminescent acti vity.77 Differential acti vation of the B AR reporter w as obser ved in Erlotinib-sensiti ve and Erlotinib-resistant cell lines in response to Erlotinib, further conf irming the utility of this kinase repor ter in sensing specific upstream signals. A signif icant adv antage of bioluminescent kinase imaging is that it provides the ability to monitor signaling pathways in li ve animals, w hich provides a unique understanding of phar macokinetics and bioa vailability of specific drugs. For example, at 40 mg/kg API-2 treatment, inhibitor y levels of the compound w ere detected for up to 24 hours but decreased thereafter. On the contrary, when 20 mg/kg was delivered, peak inhibition was detected at 12 hours, and b y 24 hours, little inhibitor y activity was detected (Figure 2). Unlike API-2 for which published phar macokinetics data are not a vailable, the pharmacokinetics of perifosine has been e xtensively studied. Pub lished data sho wed that high plasma concentrations of the dr ug could be detected for as long as 7 da ys posttreatment. Results obtained from the B AR reporter studies in live animals confirm this observation because high le vels of Akt inhibitor y acti vity w as detected within 2 hours of treatment and remained elevated for 7 da ys. Such studies estab lish a platfor m for identification of in vivo drug–target interaction. Finally,

A

B

833

these investigations can be adapted to monitor for optimization of dose, combination, and schedule in animals. This platfor m for imaging of kinase acti vity also holds g reat promise in the conte xt of high-throughput screening of compound libraries and siRN A libraries. Such cell-based screens provide a major advantage, in that only compounds that interact with the target in the correct cellular compar tment and under nor mal cellular ph ysiological conditions of that compar tment (pH, concentrations of specif ic ions, etc) w ould be identif ied. In addition, because the assay involves live cells, the reporter enables one to monitor the kinase in question in context of other signaling pathw ays. F or e xample, the impact of PTEN status (a phosphatase that inhibits Akt activation) on Akt activity can be monitored using the B AR reporter but would not be possible using traditional biochemical assays because of their invasive nature. Lastly, in contrast to other cell-based repor ter screens, w hich are fraught with false positives, the kinase repor ter described here is “gain-of-function assay” wherein the inhibition of kinase activity results in increase in bioluminescence. F or example, compounds that are c ytotoxic (and thus result in loss of signal) or those that inhibit the repor ter directl y (e g, luciferase inhibitors) may show up as false positives in traditional kinase assays but would not do so when using the BAR reporter platform. Such carefully designed screening methodologies w ould enab le one to nar row do wn the number of hits to a smaller g roup of “true positives.”

C

Figure 2. Imaging of pharmacodynamics of two distinct Akt kinase inhibitors in live animals. A, Mice transplanted with D54 cells stably expressing bioluminescent Akt reporter (BAR) were treated with vehicle control (20% DMSO in PBS), API-2 (20 mg/kg or 40 mg/kg), or perifosine (30 mg/kg). Images of representative mice are shown before treatment, during maximal luciferase signal upon treatment (Max), and after treatment. B, Tumor-specific bioluminescence activity of D54 cells stably expressing BAR, treated with either the vehicle control (20% DMSO in PBS) or API-2 (20 mg/kg or 40 mg/kg), was monitored at various times. Fold induction of signal intensity over pretreatment values was plotted as mean ±SEM for each of the groups. C, Bioluminescence activity in tumor-bearing mice before treatment and in response to treatment with 30 mg/kg perifosine, plotted as fold induction over pretreatment values (±SEM) for each of the groups.

834

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Adaptation of Bioluminescent Kinase Reporter Specificity for a number of kinases is dependent upon its subcellular localization. For example, Akt is recruited to the plasma membrane b y PI-3 kinase–generated D3-phosphorylated phosphoinositides that bind to the Akt PH domain and induce the translocation. 91,92 At the cell membrane, phosphoinositide-dependent kinase-1 co-localizes and phosphor ylates within the acti vation loop of Akt.91,92 Therefore, it is plausible that membrane recruitment of the BAR should enhance the sensitivity of the repor ter. We ha ve indeed tested this h ypothesis b y constructing a membrane-tar geted bioluminescent Akt reporter b y fusing 10 amino-terminal residues of L yn kinase responsib le for m yristoylation and palmito ylation93 to B AR (F igure 3). Sensiti vity of the MyrP almBAR reporter was doubled when compared with BAR in reporting Akt signaling (data not pub lished). Thus, the next generation of reporters should also the consider the subcellular location of the acti vity in question as these distinct locations harbor the highest concentration of the kinase acti vity and should , thus, result in increased reporter activity. The bioluminescent kinase reporter can be adapted for other protein kinases including receptor or nonreceptor Tyr and Ser/Thr kinases by using a suitable substrate and phosphorylation recognition domains. To aid in identifying of the tar get phosphor ylation site, there are a number of resources and methodolo gies described in literature 32–42

and on the websites such as , , . In addition, a database of experimentally v erified phosphorylation site is a vailable at , w hich contains 4,026 substrate proteins.42,94,95 For prediction of phosphor ylation, w ebsites such as or may be useful.

MOLECULAR IMAGING OF PROTEASES Many biolo gical processes are re gulated through the actions of proteases. This can in volve clea vage of the nascent polypeptide chain, posttranslational clea vage of inactive enzymes to yield functional enzymes and at the level of proteol ytic degradation of an enzyme w hen its activity is not needed. Proteol ytic processing is modulated both temporally and positionally, and it contributes to protein activation and subcellular localization. Cleavage b y proteases is a ubiquitous, comple x, and tightl y regulated process with a critical role in se veral physiological processes ranging from embr yogenesis to cell death. Proteases can be subdi vided into e xopeptidases, whose enzymatic activity leads to cleavage of amino- or carboxy-terminal peptide bonds, or endopeptidases, enzymes that cleave peptide bonds inter nally. Endopeptidases are fur ther subg rouped as aspar tic, c ysteine, serine, and metalloproteases based on the presence of unique amino acid(s) at the acti ve site. F or e xample, aspartic proteases contain tw o aspar tic acid residues at

Figure 3. Membrane-targeted bioluminescent Akt reporter. MyrPalm-BAR is generated by adding 10 amino acids from the N-terminus of Lyn kinase to BAR plasmid. The proposed basis of reporter activity for the MyrPalm-BAR reporter remains the same as that for BAR alone, which is described in Figure 1. This involves (A) Akt-dependent phosphorylation of the Aktpep domain that results in its interaction with the FHA2 domain. In this form (Akt-ON), the reporter has minimal bioluminescence activity (Light OFF). B, In the absence of Akt activity (Akt-OFF), association of the N-Luc and C-Luc domains restores bioluminescence activity (Light-ON).

Imaging of Signaling Pathways

the active site that are in volved in proteol ysis and c ysteine proteases contain a c ysteine residue at the acti ve site that is in volved in for ming a co valent inter mediate with the substrate. These proteases and their molecular mechanisms ha ve been e xtensively described. 96–99 This section is dedicated to the description of the molecular imaging strategies that were developed by our laboratory and others to nonin vasively monitor protease acti vity in cells and animals.

Role of Proteases in Normal and Pathophysiology As mentioned abo ve, cellular proteases perfor m di verse critical functions crucial to proper execution of physiological processes, including but not limited to de velopment, hormone maturation, immunity, blood clotting, pathogenesis of viral and bacterial diseases, and pro grammed cell death. Proteases are v ery tightl y controlled and are involved in balancing processes that determine a cells fate: for example, they play a critical role in the coordination of proliferation and pro grammed cell death (apoptosis), which are essential for nor mal physiology. An imbalance of these two opposing processes results in various diseases including AIDS, neurode generative disorders, m yelodysplastic syndromes, ischemia/reperfusion injur y, cancer , autoimmune disease. Therefore, quantitati ve nonin vasive imaging of proteases w ould be a signif icant advance for monitoring disease progression and in screening and v alidation of experimental therapeutic agents.

835

involved in the clea vage of k ey cellular proteins and the concomitant appearance apoptotic mor phology.101,105,106 We described in 2002 the de velopment of a no vel chimeric luciferase repor ter molecule that w hen expressed in mammalian cells has attenuated le vels of bioluminescent activity.107 This fusion protein harbored an estro gen receptor re gulatory domain (ER) both at the N- and C-ter minus of luciferase (F igure 4) that silenced the bioluminescent acti vity of this enzyme. The inclusion of a protease clea vage site for caspase-3 (DEVD) at the junction of luciferase and ER domains allowed for protease-mediated activation of the reporter molecule after separation from the silencing domain (ie, ER). In cells under going apoptosis, caspase3–dependent clea vage of the recombinant product occurred resulting in the restoration of luciferase activity, which could be detected in li ve animals using bioluminescence imaging. 107 Furthermore, in vi vo studies using x enografts of this cell line sho wed that caspase-3 acti vation b y TRAIL treatment could be imaged noninvasively using bioluminescence. The ability to image caspase-3 activation noninvasively showed a unique tool for the e valuation of therapeutic ef ficacy of individual experimental agents or their combination. We showed using bioluminescent caspase-3 repor ter that a combina tion of 5-fluorouracil and TRAIL/Apo2L leads to an increased ef ficacy in treatment of mice transplanted with D54 cells. These results suggest a potential application for this technolo gy in investigating ef ficacy of a therapeutic re gimen (dose,

Molecular Imaging of Apoptotic Proteases Apoptosis, or programmed cell death, is characterized by cell shrinkage, membrane blebbing, chromatin condensation, DNA fragmentation, and selective activation of caspases with a subsequent clea vage of specif ic tar get proteins.100–103 Apoptosis may be initiated extrinsically by the acti vation of a death receptor b y ligands, such as tumor necrosis factor α-related apoptosis-inducing ligand (TRAIL), or intrinsicall y b y inhibitors of cellular pathways, such as staurosporine. 103,104 Caspases, a f amily of cysteine-dependent aspar tate-directed proteases, are prominent players both in extrinsic and intrinsic pathways as molecular initiators and e xecutors of apoptosis.101,103–106 These proteases are synthesized as inacti ve zymogens that become acti vated b y scaf fold-mediated transactivation or b y cleavage via upstream proteases in an intracellular cascade.101,103–106 The converging point of the protease cascade in both intrinsic and e xtrinsic apoptotic pathw ays is the acti vation of caspase-3, w hich is

Figure 4. The strategy for bioluminescent imaging of apoptosis based on attenuation of luciferase activity. Recombinant reporter was created by fusing ER (residues 281 to 599 of the modified mouse estrogen receptor sequence; {Littlewood, 1995 #307}) at both N- and C-terminus of luciferase separated by DEVD sequence on both ends. This reporter ER–DEVD–Luc–DEVD–ER showed attenuated levels of luciferase activity, which was restored by caspase-3–dependent cleavage of the ER domains from the luciferase.

836

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

combination, or schedule of dr ugs) and real-time monitoring of the outcome. 108 To impro ve the signal-to-noise ratio of the repor ter described abo ve, three additional repor ters w ere constructed. The f irst w as based on the principle of f irefly complementation89 where we constr ucted a recombinant protein wherein peptideA–N-Luc (ANLuc) and peptideB–C-Luc (BCLuc) are fused with an inter vening caspase-3 cleavage site. Peptide A and B are a pair of peptides that ha ve been repor ted to interact with each other with strong affinity.109,110 This chimeric luciferase reporter has signif icantly reduced backg round luciferase acti vity as the N-Luc and C-Luc are unab le to complement w hen expressed as fusion protein. Ho wever, in presence of caspase-3 activity, the reporter is cleaved, and pepA and pepB associate through a high-affinity interaction and facilitate complementation of N-Luc and C-Luc domains, w hich reconstitutes luciferase (F igure 5). We ha ve recentl y shown that this apoptosis repor ter system is a highl y sensitive, dynamic, and quantitati ve repor ter of caspase-3 activity both in vitro and in vi vo.111 Additionally, this reporter system allo wed in vi vo optimization of dose, combination, and schedule of no vel therapies in a dynamic, noninvasive manner. The second strate gy to minimize the backg round activity was based on the principle that used a reporter that

requires oligomerization for maximal acti vity. Therefore, the addition of ER re gulatory domains w ould not onl y inhibit the interaction of the substrate with the enzyme but also hinder the oligomerization. This silencing will ef fectively abolish the background activity that is typically seen with the monomeric enzymes lik e luciferase. We used β-galactosidase, a tetrameric enzyme consisting of four identical pol ypeptides,112 to constr uct an ER– LacZ–ER reporter (Figure 6). This repor ter was more sensiti ve and has a better signal-to-noise ratio w hen compared with the ER–Luc–ER (manuscript in preparation). Additional advantages of constr ucting a β-galactosidase–based protease sensing repor ter are the a vailability of a v ariety of substrates that are either fluoro genic, para-magnetic, radioactive or chemiluminiscent. 113–116 Furthermore, β-galactosidase is very stable and active in a range of conditions, of fering a robust and multimodality molecular imaging technology. The third strate gy to improve signal-to-noise ratio in imaging of apoptotic cells was based on differential localization of a recombinant chimeric single chain antibody . This non-bioluminescent repor ter was a chimeric protein constructed from a single chain antibody (harboring signal peptide, HA and m yc tags, and a transmembrane domain), and a Golgi retention signal separated b y caspase-3 reco gnition and clea vage sequence. 75 In g rowing

A.

B.

Figure 5. Strategy for noninvasive imaging of caspase-3 utilizing the luciferase complementation. A, Schematic representation of the bioluminescent caspase-3 reporter (AN-Luc-BC-Luc). Apoptosis imaging reporter constitutes the split luciferase (N-Luc and C-Luc) domains fused to interacting peptides, pepA and pepB, with an intervening caspase-3 cleavage motif. B, On induction of apoptosis, the reporter molecule is proteolytically cleaved by caspase-3 at the DEVD motif. This cleavage enables interaction between pepANLuc and pepBCLuc, thus reconstituting luciferase activity.

Imaging of Signaling Pathways

cells, the chimeric single chain antibody resides in Golgi bodies; w hereas in cells under going apoptosis in a caspase-3–dependent manner, the recombinant single chain antibody translocates to cell surf ace allowing an accessible means to image dying cells (F igure 7). This apoptosis imaging technology may overcome the limitation of tw odimensional imaging with luciferase reporter and enable a true three-dimensional imaging of apoptosis a possibility. For e xample, phOx, the hapten for the single chain antibody, when coupled to nanopar ticles embedded with magnetic resonance or computed tomo graphy contrast agents ma y allo w for the visualization of tumor cells undergoing apoptosis b y a caspase-3–dependent increase in contrast enhancement. Similarly coupling of phOx with either near infra red fluorochrome or radioacti ve agent would allo w detection of apoptosis b y optical fluorescence imaging or b y nuclear imaging modalities such as positron emission tomo graphy (PET) or single photon emission computed tomo graphy (SPECT), respecti vely. Alternately, a multimodality imaging strate gy ma y be envisaged based on the creation of nanosized delivery system–harboring reagents that allo w detection b y nuclear , optical, and magnetic imaging, w hich w ould pro vide

837

distinct advantages by improving the ability to quantify , interpret, and locate caspase activation with high sensitivity and specif icity. In summar y, this section highlights the no vel imaging tools our laborator y has generated to image caspase3 activity. These models ser ve as proof of principle for the more general application of real-time imaging of intracellular proteases. The ability to image apoptosis noninvasively and dynamically over time provides a new opportunity for high-throughput screening of pro- and antiapoptotic compounds and for target validation.

Molecular Imaging of Organelle-specific Proteases Proteases in volved in maturation of secretor y proteins typically reside in the trans-Golgi network (TGN), where they proteolytically process newly formed proteins from the endoplasmic reticulum before packaging into secretory v esicles. This process of proprotein maturation involves reco gnition and clea vage of unique sequences in tar get protein prodomains b y specif ic proteases. 117 TGN-resident proteases include carbo xypeptidases,

A

B Substrate

Proteolytic cleavage

ER mediated inhibition of tetramerization

Active tetramer

Figure 6. Tetrameric LacZ–based reporter for monitoring cell death. A, Recombinant beta-galactosidase–based apoptosis reporter was created by fusing ER (See Figure 4) at both N- and C-terminus of LacZ separated by DEVD sequence on both ends. B, Addition of the ER to both end of LacZ had lead to inhibitory effect on the activity of ER–DEVD–LacZ–DEVD–ER: It inhibited the tetramerization of the enzyme, which is critical for its activity, leading to increased signal-to-noise ratio.

838

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

prohormone convertase (PC) family members, and β-site amyloid precursor protein (APP)-clea ving enzyme (BACE) family members.118–121

Although there ha ve been e xtensive efforts to study TGN protease biolo gy, a v ersatile and sensiti ve assa y system to monitor the acti vity noninvasively in real-time

Figure 7. Strategy for imaging of apoptosis based on conditional expression of single chain antibody. A, Constitutive imaging construct (CIC) was constructed from a single chain antibody, signal peptide, HA and myc tags, and a transmembrane domain. When expressed in cells, this fusion protein would localize to cell surface. B, The inducible imaging construct (IIC) was CIC engineered to contain a Golgi retention signal, separated by caspase-3 recognition and cleavage sequence. The DEVD sequence was strategically placed between transmembrane domain and Golgi retention signal so that this caspase-3 cleavage sequence is loops in the cytoplasm (see inset). When expressed in cells, this chimeric protein localizes to Golgi bodies. Induction of apoptosis leads to caspase-3 activation and cleavage of the chimeric protein resulting in its translocation to cell surface.

Imaging of Signaling Pathways

has been lacking. Studying TGN enzymes using noninvasive assays allows preservation of the unique intracellular TGN environment (low pH and high Ca 2+) that is readily perturbed by invasive assays. Fur thermore, strate gies to monitor TGN enzyme activity directly in its native physiological environment facilitate discovery of novel pharmaceutical agents that can tra verse plasma and Golgi membranes and retain inhibitor y acti vity within the microenvironment of TGN. Identif ication of such inhibitors ma y ha ve a profound influence on the treatment of diseases such as AD and cancer and may help to combat lethal viral and bacterial infections.

839

In an effort to develop a cell-based assay to report on TGN-resident enzyme acti vity, w e constr ucted a h ybrid reporter protein, GRAP , consisting of three functional domains: (1) secreted alkaline phosphate (SEAP), (2) a Golgi protease-specif ic recognition and clea vage site, and (3) the cytoplasmic and transmembrane domains from BACE that retain the repor ter within the TGN (Figure 8). This fusion protein, w hen expressed in cells, localizes to TGN until it is clea ved by a specif ic TGN protease, after which SEAP is secreted into the e xtracellular medium. Therefore, SEAP levels present in the medium are indicative of intracellular TGN protease acti vity. Decrease in

Figure 8. Strategy for noninvasive monitoring of TGN proteases. A, Fusion reporter protein consisting of three functional domains, (1) secreted alkaline phosphatase (SEAP), (2) a Golgi protease–specific recognition and cleavage site, and (3) Golgi retention signal from BACE that retains the reporter within the TGN, was created. When expressed in cells, this reporter localizes to TGN until it is cleaved by a specific TGN protease, after which SEAP is secreted into the extracellular media. Note: Because of the strategically positioned Golgi retention signal, the protease-specific recognition and cleavage sequence is located within the lumen of the Golgi to respond to specific TGN proteases residing in this compartment of the Golgi bodies (see inset).

840

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

SEAP le vels signif ies a loss of protease acti vity and allows positi ve identif ication of protease inhibitors. We constructed GRAPfurin, which contains a 10–amino acid recognition and clea vage site (GLSARNRQKR ↓) from the furin substrate Stromol ysin-3 (ST3), and sho wed that overexpression of furin increased processing of GRAPfurin leading to a reduction in unprocessed protein and an increase in processed protein w as obser ved. There w as concomitant increase in the SEAP acti vity in the media. The specificity of this system was shown by construction of GRAPfurinmut in w hich the furin tar get reco gnition and cleavage sequence was mutated to GLSAANAQAA↓, rendering this repor ter nonresponsive to furin proteol ytic activity. This led to a reduction in processed protein in both the l ysate and extracellular medium and a concomitant decrease in SEAP acti vity in media. 122,123 Additionally, a notable decrease in SEAP activity was observed when CHO-GRAPfurin cells w ere treated with 25 µM of the furin inhibitor , dec-R VKR-CMK, w hereas similar treatment with dec-R VKR-CMK caused no decrease in SEAP acti vity in the control CHO cell line e xpressing SEAP constitutively.123 In summary, using this technology, we are ab le to specif ically monitor furin protease activity.123 Furin plays a critical role in processing a m yriad of proteins containing the RX(K/R)R ↓ cleavage domain found in ser um proteins (proalbumin), coagulation f actors (pro-von Willebrand factor), growth factors, and hormones (pro- β–nerve g rowth f actor and bone morphogenic f actor-4), cell surf ace receptors (insulin proreceptor), and matrix metalloproteases (stromolysin-3 and MT1-MMP). Processing by furin and other PC f amily members contributes to development of se veral diseases, such as Alzheimer’s disease (AD) and arthritis, and also enhances in vasion and proliferation of cancer cells. In addition, furin activity is necessary for propagation of viruses such as HIV -1, ebola, and a vian influenza and activation of virulent bacterial pathogens such as anthrax, pseudomonas, and diphtheria. The remarkab ly cr ucial role of furin and f amily members in viral infections is emphasized by the disco very that H5N1 a vian influenza virus patho genicity is attributed to the acquisition of a polybasic tract in hemagglutinin gl ycoprotein, w hich is effectively cleaved by furin and other PCs. Acquisition of this furin cleavage domain may have extended the virus’s ability to infect humans. H5N1 is e xtremely lethal because ubiquitousl y e xpressed furin allo ws systemic infection of all or gans. In addition, furin acti vation of anthrax to xin is necessar y for infection. The usage of anthrax to xin as a biolo gical w eapon necessitates the development of sensitive assays that lead to identification

of novel furin inhibitors that ha ve potential to treat a vian influenza, anthrax, and other furin-related diseases. Several protein/peptide-based furin inhibitors ha ve been reported, including α1-PDX, D6R, and D9R. Unfortunately, the use of these inhibitors as pharmaceutical agents is hampered b y their lar ge size, lack of stability , and/or to xicity. Another inhibitor dec-R VKR-CMK is a potent inhibitor of furin with an IC50 in the picomolar range but is extremely toxic. To date, the only non-protein/peptide inhibitor of furin is a naturall y occur ring neoandrographolide and its succinoyl ester derivatives, with IC50 values ranging from high micromolar to low millimolar values. Although copper comple xes of ter pyridine deri vatives show inhibitor y concentrations in the micromolar range, there cur rently is no documentation of ef ficacy against furin-related diseases in vi vo. There is an ur gent need for new inhibitors, and the cell-based assay described here will expedite dr ug disco very because librar y screening will allow for identification of inhibitors with desired characteristics in context to toxicity, solubility, and ability to interact with the protein tar get in its appropriate subcellular compartment. We adapted the cell-based assay system described above for high-throughput screening (HTS) and identif ied furin inhibitors from a screen of 39,000 molecules. CCG 8294, one of the major hits in the high-throughput assa y, has shown great promise as a furin inhibitor with high efficacy in cells and has also shown a dose-dependent inhibition of furin-mediated processing of pol ypeptides within the secretory pathway.123 The long-ter m goal of this study is to use these novel assays to better understand the biolo gy of Golgiresident proteases, their re gulation, and their role in specific patholo gies. The data presented here ha ve described the utility of these assa ys in identif ication of novel modulators of protease acti vity and can be extended to studies of other Golgi proteases.

CONCLUDING REMARKS Integration of genetically encoded imaging reporters into cells and animals has pro vided a unique oppor tunity to monitor molecular, biochemical, and cellular pathways in vivo. It will greatly facilitate the process of target validation and dose and schedule optimization as w ell as providing a w ay to identify lead compounds from a library using cell-based, high-throughput screening.

ACKNOWLEDGMENTS We thank Steven Kronenberg for designing the f igures, Christin Hamilton and Hyma Rao for critical reading of

Imaging of Signaling Pathways

the manuscript. This w ork w as suppor ted b y US National Institutes of Health g rants P01CA85878, P50CA93990, R24CA83099, R01RCA129623A and a grant from the John and Suzanne Munn Endo wed Research Fund of the Uni versity of Michigan Comprehensive Cancer Center (to MSB).

REFERENCES 1. Manning G, Plowman GD, Hunter T, Sudarsanam S. Evolution of protein kinase signaling from y east to man. Trends Biochem Sci 2002;27:514–20. 2. Manning G, Whyte DB, Martinez R, et al. The protein kinase complement of the human genome. Science 2002;298:1912–34. 3. Edelman AM, Blumenthal DK, Krebs EG. Protein serine/threonine kinases. Annu Rev Biochem 1987;56:567–613. 4. Hubbard SR, Till JH. Protein tyrosine kinase str ucture and function. Annu Rev Biochem 2000;69:373–98. 5. Hunter T, Cooper J A. Protein-tyrosine kinases. Annu Rev Biochem 1985;54:897–930. 6. Tsygankov AY. Non-receptor protein tyrosine kinases. F ront Biosci 2003;8:S595–635. 7. Besant PG, Tan E, Attwood PV. Mammalian protein histidine kinases. Int J Biochem Cell Biol 2003;35:297–309. 8. Tan E, Lin Zu X, Yeoh GC, et al. Detection of histidine kinases via a filter-based assa y and re verse-phase thin-la yer chromato graphic phosphoamino acid analysis. Anal Biochem 2003;323:122–6. 9. Zhang L, Bhojani MS, Ross BD, Rehemtulla A. Molecular imaging of protein kinases. Cell Cycle 2008;7:314–7. 10. Blume-Jensen P, Hunter T. Oncogenic kinase signalling. Nature 2001; 411:355–65. 11. Hunter T. Signaling—2000 and beyond. Cell 2000;100:113–27. 12. Werlen G, Hausmann B, Naeher D, Palmer E. Signaling life and death in the thymus: timing is everything. Science 2003;299:1859–63. 13. Montell DJ. A kinase gets caspases into shape. Cell 2006;126:450–2. 14. McGaugh JL. Memor y—a centur y of consolidation. Science 2000; 287:248–51. 15. Sutton MA, Schuman EM. Dendritic protein synthesis, synaptic plasticity, and memory. Cell 2006;127:49–58. 16. Raddatz D , Ramadori G. Carboh ydrate metabolism and the li ver: actual aspects from ph ysiology and disease. Z Gastroenterol 2007;45:51–62. 17. Skalhegg BS, Funder ud A, Henanger HH, et al. Protein kinase A (PKA)—a potential tar get for therapeutic inter vention of dysfunctional immune cells. Curr Drug Targets 2005;6:655–64. 18. Okkenhaug K, Vanhaesebroeck B. PI3K-signalling in B- and T-cells: insights from gene-tar geted mice. Biochem Soc Trans 2003; 31:270–4. 19. Takeuchi O , Akira S. Reco gnition of vir uses b y innate immunity . Immunol Rev 2007;220:214–24. 20. Liu B, Neufeld AH. Activation of epider mal growth factor receptors in astrocytes: from development to neural injur y. J Neurosci Res 2007;85:3523–9. 21. Runeberg-Roos P , Saar ma M. Neurotrophic f actor receptor RET : structure, cell biolo gy, and inherited diseases. Ann Med 2007; 39:572–80. 22. Johnson SA, Hunter T. Kinomics: methods for deciphering the kinome. Nat Methods 2005;2:17–25. 23. Kobe B , Kampmann T, F orwood JK, et al. Substrate specif icity of protein kinases and computational prediction of substrates. Biochim Biophys Acta 2005;1754:200–9. 24. Peters JM, McKay RM, McKay JP, Graff JM. Casein kinase I transduces Wnt signals. Nature 1999;401:345–50.

841

25. Sakanaka C, Leong P, Xu L, et al. Casein kinase iepsilon in the wnt pathway: regulation of beta-catenin function. Proc Natl Acad Sci U S A 1999;96:12548–52. 26. Polakis P. Casein kinase 1: a Wnt’er of disconnect. Cur r Biol 2002; 12:R499–501. 27. Harwood AJ. Re gulation of GSK-3: a cellular multiprocessor . Cell 2001;105:821–4. 28. Li H, Pamukcu R, Thompson WJ. beta-Catenin signaling: therapeutic strategies in oncology. Cancer Biol Ther 2002;1:621–5. 29. Tian Q, F eetham MC, Tao WA, et al. Proteomic anal ysis identif ies that 14-3-3zeta interacts with beta-catenin and f acilitates its activation by Akt. Proc Natl Acad Sci U S A 2004;101:15370–5. 30. Bahmanyar S, Kaplan DD , Deluca JG, et al. beta-Catenin is a Nek2 substrate in volved in centrosome separation. Genes De v 2008; 22:91–105. 31. de Graauw M, Hensber gen P, v an de Water B . Phospho-proteomic analysis of cellular signaling. Electrophoresis 2006;27:2676–86. 32. Cohen P, Knebel A. KESTREL: a powerful method for identifying the physiological substrates of protein kinases. Biochem J 2006; 393:1–6. 33. Murray JT, Campbell DG, Peggie M, et al. Identification of filamin C as a ne w physiological substrate of PKBalpha using KESTREL. Biochem J 2004;384:489–94. 34. Peng C, Knebel A, Morrice NA, et al. Pim kinase substrate identif ication and specif icity. J Biochem 2007;141:353–62. 35. Huang HD, Lee TY, Tzeng SW, Horng JT. KinasePhos: a web tool for identifying protein kinase-specific phosphorylation sites. Nucleic Acids Res 2005;33:W226–9. 36. Senawongse P, Dalby AR, Yang ZR. Predicting the phosphor ylation sites using hidden Markov models and machine learning methods. J Chem Inf Model 2005;45:1147–52. 37. Wong YH, Lee TY, Liang HK, et al. KinasePhos 2.0: a web server for identifying protein kinase-specific phosphorylation sites based on sequences and coupling patter ns. Nucleic Acids Res 2007;35:W588–94. 38. Niedner RH, Buzko OV, Haste NM, et al. Protein kinase resource: an integrated en vironment for phosphor ylation research. Proteins 2006;63:78–86. 39. Petretti C, Prigent C. The protein kinase resource: e verything y ou always wanted to kno w about protein kinases but w ere afraid to ask. Biol Cell 2005;97:113–8. 40. Smith CM. The protein kinase resource and other bioinfor mation resources. Prog Biophys Mol Biol 1999;71:525–33. 41. Smith CM, Shindyalo v IN , Veretnik S, et al. The protein kinase resource. Trends Biochem Sci 1997;22:444–6. 42. MacDonald JA, Mackey AJ, Pearson WR, Haystead TA. A strategy for the rapid identif ication of phosphor ylation sites in the phosphoproteome. Mol Cell Proteomics 2002;1:314–22. 43. Obenauer JC, Cantle y LC, Yaffe MB. Scansite 2.0: proteome-wide prediction of cell signaling interactions using shor t sequence motifs. Nucleic Acids Res 2003;31:3635–41. 44. Puntervoll P , Linding R, Gemund C, et al. ELM ser ver: a ne w resource for in vestigating shor t functional sites in modular eukaryotic proteins. Nucleic Acids Res 2003;31:3625–30. 45. Linding R, Jensen LJ, Pasculescu A, et al. NetworKIN: a resource for exploring cellular phosphor ylation networks. Nucleic Acids Res 2008;36:D695–9. 46. Remenyi A, Good MC, Bhattacharyya RP, Lim WA. The role of docking interactions in mediating signaling input, output, and discrimination in the yeast MAPK network. Mol Cell 2005;20:951–62. 47. Bhattacharyya RP, Remen yi A, Good MC, et al. The Ste5 scaf fold allosterically modulates signaling output of the yeast mating pathway. Science 2006;311:822–6. 48. Barsyte-Lovejoy D, Galanis A, Shar rocks AD. Specif icity deter minants in MAPK signaling to transcription f actors. J Biol Chem 2002;277:9896–903.

842

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

49. Tanoue T, Adachi M, Moriguchi T, Nishida E. A conserved docking motif in MAP kinases common to substrates, activators and regulators. Nat Cell Biol 2000;2:110–6. 50. Vinciguerra M, Vivacqua A, F asanella G, et al. Dif ferential phosphorylation of c-Jun and JunD in response to the epidermal growth factor is determined by the structure of MAPK targeting sequences. J Biol Chem 2004;279:9634–41. 51. Biondi RM, Nebreda AR. Signalling specif icity of Ser/Thr protein kinases through docking-site-mediated interactions. Biochem J 2003;372:1–13. 52. Tanoue T, Nishida E. Molecular recognitions in the MAP kinase cascades. Cell Signal 2003;15:455–62. 53. Cohen P. The structure and regulation of protein phosphatases. Annu Rev Biochem 1989;58:453–508. 54. Mustelin T. A brief introduction to the protein phosphatase f amilies. Methods Mol Biol 2007;365:9–22. 55. Mumby M. PP2A: un veiling a reluctant tumor suppressor . Cell 2007;130:21–4. 56. Ray D, Kiyokawa H. CDC25A phosphatase: a rate-limiting oncogene that determines genomic stability. Cancer Res 2008;68:1251–3. 57. Ducruet AP, Vogt A, Wipf P, Lazo JS. Dual specif icity protein phosphatases: therapeutic tar gets for cancer and Alzheimer’s disease. Annu Rev Pharmacol Toxicol 2005;45:725–50. 58. Easty D, Gallagher W, Bennett DC. Protein tyrosine phosphatases, new targets for cancer therap y. Cur r Cancer Dr ug Targets 2006; 6:519–32. 59. Sridhar R, Hanson-P ainton O, Cooper DR. Protein kinases as therapeutic targets. Pharm Res 2000;17:1345–53. 60. Westra J, Limburg PC. p38 mitogen-activated protein kinase (MAPK) in rheumatoid arthritis. Mini Rev Med Chem 2006;6:867–74. 61. Kumar R, Singh VP, Baker KM. Kinase inhibitors for cardio vascular disease. J Mol Cell Cardiol 2007;42:1–11. 62. Mueller BK, Mack H, Teusch N. Rho kinase, a promising dr ug target for neurological disorders. Nat Rev Drug Discov 2005;4:387–98. 63. Blease K. Targeting kinases in asthma. Exper t Opin In vestig Dr ugs 2005;14:1213–20. 64. Ben-Bassat H. Biological activity of tyrosine kinase inhibitors: novel agents for psoriasis therap y. Cur r Opin In vestig Dr ugs 2001; 2:1539–45. 65. Tschoep K, Kohlmann A, Schlemmer M, et al. Gene e xpression profiling in sarcomas. Crit Rev Oncol Hematol 2007;63:111–24. 66. Margalit O, Somech R, Amariglio N, Rechavi G. Microar ray-based gene e xpression prof iling of hematolo gic malignancies: basic concepts and clinical applications. Blood Rev 2005;19:223–34. 67. Alizadeh A, Eisen M, Davis RE, et al. The lymphochip: a specialized cDNA microarray for the genomic-scale anal ysis of gene expression in nor mal and malignant l ymphocytes. Cold Spring Harb Symp Quant Biol 1999;64:71–8. 68. Nishiu M, Yanagawa R, Nakatsuka S, et al. Microar ray anal ysis of gene-expression prof iles in diffuse large B-cell lymphoma: identification of genes related to disease pro gression. Jpn J Cancer Res 2002;93:894–901. 69. Morris DS, Tomlins SA, Rhodes DR, et al. Inte grating biomedical knowledge to model pathways of prostate cancer progression. Cell Cycle 2007;6:1177–87. 70. Radford IR. Imatinib . No vartis. Cur r Opin In vestig Dr ugs 2002; 3:492–9. 71. Collins I, Workman P. New approaches to molecular cancer therapeutics. Nat Chem Biol 2006;2:689–700. 72. Melnikova I, Golden J. Targeting protein kinases. Nat Rev Drug Discov 2004;3:993–4. 73. Melnikova I, Golden J . Apoptosis-targeting therapies. Nat Re v Drug Discov 2004;3:905–6. 74. Chun PY, Feng FY, Scheurer AM, et al. Syner gistic effects of gemcitabine and gefitinib in the treatment of head and neck carcinoma. Cancer Res 2006;66:981–8.

75. Bhojani MS, Hamstra D A, Chang DC, et al. Imaging of proteol ytic activity using a conditional cell surf ace receptor . Mol Imaging 2006;5:129–37. 76. Bhojani MS, Laxman B , Ross BD , Rehemtulla A. Molecular imaging in cancer. In: Debatin KM, Fulda S, editors. Apoptosis and cancer therapy. Weinheim, Ger many: Wiley-VCH; 2006. p. 37–59. 77. Zhang L, Lee KC, Bhojani MS, et al. Molecular imaging of Akt kinase activity. Nat Med 2007;13:1114–9. 78. Fields S, Song O . A novel genetic system to detect protein-protein interactions. Nature 1989;340:245–6. 79. Ghosh I, Hamilton AD, Re gan L. Antiparallel leucine zipper directed protein reassembly: application to the green fluorescent protein. J Am Chem Soc 2000;122:5658–9. 80. Luker KE, Piwnica-Worms D. Optimizing luciferase protein fragment complementation for bioluminescent imaging of protein-protein interactions in li ve cells and animals. Methods Enzymol 2004;385:349–60. 81. Michnick SW, Ear PH, Manderson EN , et al. Uni versal strategies in research and dr ug discovery based on protein-fragment complementation assays. Nat Rev Drug Discov 2007;6:569–82. 82. Remy I, Ghaddar G, Michnick SW. Using the beta-lactamase proteinfragment complementation assay to probe dynamic protein-protein interactions. Nat Protoc 2007;2:2302–6. 83. Remy I, Michnick SW. A highly sensitive protein-protein interaction assay based on Gaussia luciferase. Nat Methods 2006;3:977–9. 84. Remy I, Michnick SW . Application of protein-fragment complementation assa ys in cell biolo gy. Biotechniques 2007; 42:137, 139, 141 passim. 85. Stefan E, Aquin S, Berger N, et al. Quantification of dynamic protein complexes using Renilla luciferase fragment complementation applied to protein kinase A activities in vivo. Proc Natl Acad Sci U S A 2007;104:16916–21. 86. McCaffrey A, Kay MA, Contag CH. Advancing molecular therapies through in vi vo bioluminescent imaging. Mol Imaging 2003; 2:75–86. 87. Greer LF III, Szalay AA. Imaging of light emission from the expression of luciferases in li ving cells and or ganisms: a review. Luminescence 2002;17:43–74. 88. Contag CH, Ross BD . It’ s not just about anatom y: in vi vo bioluminescence imaging as an e yepiece into biolo gy. J Magn Reson Imaging 2002;16:378–87. 89. Luker KE, Smith MC, Luker GD, et al. Kinetics of regulated proteinprotein interactions revealed with f irefly luciferase complementation imaging in cells and living animals. Proc Natl Acad Sci U S A 2004;101:12288–93. 90. Paulmurugan R, Gambhir SS. Monitoring protein-protein interactions using split synthetic renilla luciferase protein-fragment-assisted complementation. Anal Chem 2003;75:1584–9. 91. Chan TO, Rittenhouse SE, Tsichlis PN . AKT/PKB and other D3 phosphoinositide-regulated kinases: kinase acti vation b y phosphoinositide-dependent phosphor ylation. Annu Re v Biochem 1999;68:965–1014. 92. Cantley LC. The phosphoinositide 3-kinase pathw ay. Science 2002;296:1655–7. 93. Kovarova M, Tolar P , Arudchandran R, et al. Str ucture-function analysis of Lyn kinase association with lipid rafts and initiation of early signaling events after Fcepsilon receptor I agg regation. Mol Cell Biol 2001;21:8318–28. 94. Blom N, Kreegipuu A, Brunak S. PhosphoBase: a database of phosphorylation sites. Nucleic Acids Res 1998;26:382–6. 95. Kreegipuu A, Blom N , Br unak S. PhosphoBase, a database of phosphorylation sites: release 2.0. Nucleic Acids Res 1999; 27:237–9. 96. Zhivotovsky B . Caspases: the enzymes of death. Essa ys Biochem 2003;39:25–40.

Imaging of Signaling Pathways

97.

98. 99. 100.

101. 102. 103. 104.

105. 106. 107. 108.

109.

110. 111.

112.

Earnshaw WC, Martins LM, Kaufmann SH. Mammalian caspases: structure, activation, substrates, and functions during apoptosis. Annu Rev Biochem 1999;68:383–424. Ho PK, Hawkins CJ. Mammalian initiator apoptotic caspases. FEBS J 2005;272:5436–53. Strasser A, O’Connor L, Dixit VM. Apoptosis signaling. Annu Rev Biochem 2000;69:217–45. Ameisen JC. On the origin, evolution, and nature of programmed cell death: a timeline of four billion y ears. Cell Death Dif fer 2002;9:367–93. Nicholson DW, Thornberry NA. Apoptosis. Life and death decisions. Science 2003;299:214–5. Hengartner MO . The biochemistr y of apoptosis. Nature 2000; 407:770–6. Bhojani MS, Rossu BD , Rehemtulla A. TRAIL and anti-tumor responses. Cancer Biol Ther 2003;2:S71–8. Bhojani MS, Ross BD , Rehemtulla A. TRAIL in cancer therapy. In: El-Deiry WS, editor. Death receptor in cancer therapy. Totowa (NJ): Humana; 2004. p. 263–80. Shi Y. Apoptosome: the cellular engine for the activation of caspase9. Structure 2002;10:285–8. Nicholson DW, Thornberry NA. Caspases: killer proteases. Trends Biochem Sci 1997;22:299–306. Laxman B, Hall DE, Bhojani MS, et al. Noninvasive real-time imaging of apoptosis. Proc Natl Acad Sci U S A 2002;99:16551–5. Lee KC, Hamstra D A, Bhojani MS, et al. Nonin vasive m olecular imaging sheds light on the syner gy between 5-fluorouracil and TRAIL/Apo2L for cancer therap y. Clin Cancer Res 2007; 13:1839–46. Thormeyer D, Ammerpohl O, Larsson O, et al. Characterization of lacZ complementation deletions using membrane receptor dimerization. Biotechniques 2003;34:346–50, 352–5. Zhang Z, Zhu W, Kodadek T. Selection and application of peptidebinding peptides. Nat Biotechnol 2000;18:71–4. Coppola JM, Ross BD, Rehemtulla A. Noninvasive imaging of apoptosis and its application in cancer therapeutics. Clin Cancer Res 2008;14:2492–501. Matthews BW. The structure of E. coli beta-galactosidase. C R Biol 2005;328:549–56.

843

113. Van P oucke SO , Nelis HJ . Rapid detection of fluorescent and chemiluminescent total coliforms and Escherichia coli on membrane filters. J Microbiol Methods 2000;42:233–44. 114. Olesen CE, Yan YX, Liu B, et al. Novel methods for chemiluminescent detection of repor ter enzymes. Methods Enzymol 2000; 326:175–202. 115. Celen S, Deroose C, de Groot T, et al. Synthesis and e valuation of 18F- and 11C-labeled phen yl-galactopyranosides as potential probes for in vi vo visualization of LacZ gene e xpression using positron emission tomo graphy. Bioconjug Chem 2008; 19:441–9. 116. Smith AD, Trempe JP. Luminometric quantitation of photinus pyralis firefly luciferase and Escherichia coli beta-galactosidase in bloodcontaminated organ lysates. Anal Biochem 2000;286:164–72. 117. Halban PA, Ir minger JC. Sor ting and processing of secretor y proteins. Biochem J 1994;299(Pt 1):1–18. 118. Reznik SE, Fricker LD. Carboxypeptidases from A to z: implications in embr yonic development and Wnt binding. Cell Mol Life Sci 2001;58:1790–804. 119. Molloy SS, Anderson ED, Jean F, Thomas G. Bi-c ycling the furin pathway: from TGN localization to patho gen acti vation and embryogenesis. Trends Cell Biol 1999;9:28–35. 120. Vassar R. Beta-secretase (B ACE) as a dr ug tar get for Alzheimer’s disease. Adv Drug Deliv Rev 2002;54:1589–602. 121. Vassar R, Bennett BD, Babu-Khan S, et al. Beta-secretase clea vage of Alzheimer’s amyloid precursor protein by the transmembrane aspartic protease BACE. Science 1999;286:735–41. 122. Coppola JM, Bhojani MS, Ross BD , Rehemtulla A. A smallmolecule furin inhibitor inhibits cancer cell motility and invasiveness. Neoplasia 2008;10:363–70. 123. Coppola JM, Hamilton CA, Bhojani MS, et al. Identif ication of inhibitors using a cell-based assa y for monitoring Golgi-resident protease activity. Anal Biochem 2007;364:19–29. 124. Yaffe MB, Leparc GG, Lai J , et al. A motif-based prof ile scanning approach for genome-wide prediction of signaling pathw ays. Nat Biotechnol 2001;19:348–53. 125. Violin JD, Zhang J, Tsien RY, Newton AC. A genetically encoded fluorescent repor ter re veals oscillator y phosphor ylation b y protein kinase C. J Cell Biol 2003;161:899–909.

50 MOLECULAR AND FUNCTIONAL IMAGING THE TUMOR MICROENVIRONMENT

OF

KRISTINE GLUNDE, PHD, ROBERT J. GILLIES, PHD, MICHAL NEEMAN, PHD, AND ZAVER M. BHUJWALLA, PHD

A malignant tumor consists of cancer cells and the tumor microenvironment (TME) that pro vides the infrastr ucture for cancer cells to sur vive and g row. This microenvironment, w hich consists of the v asculature, stromal cells, and the e xtracellular matrix (ECM), to gether with the abnor mal ph ysiologic en vironments, such as hypoxia and acidic e xtracellular pH (pHe) that e xist in tumors, influences a range of phenotypic traits of cancer including invasion, metastasis, and response to therap y. A schematic of the influences of some of the dif ferent components of the TME are outlined in Figure 1, demonstrating its comple xity and the wide range of ef fects it can have on the cancer cells. As the interactions between cancer cells and the TME change continuall y with growth or therapy, the ability to noninvasively image the TME is criticall y impor tant. In this chapter , w e ha ve described recent developments in the use of noninvasive imaging modalities to image the TME, with an emphasis on magnetic resonance (MR) imaging (MRI) and spectroscopy (MRS). Multimodal and multiparametric molecular and functional imaging pro vide unprecedented oppor tunities for imaging the TME and imaging interactions between cancer cells and stromal cells. Se veral of these applications ha ve the potential for clinical translation allowing bench to bedside applications. These imaging capabilities can also be exploited for no vel therapeutic strate gies tar geting the TME. The development of novel imaging probes to visualize the acti vity of de gradative enzymes, the ability to image lysosomes, and the ability to image specific cellular and physiologic compar tments, and characteristics of the TME are some of the adv ances already a vailable to characterize the TME that are described here. In this chapter , we outline e xciting new developments, such as molecular 844

targeting of specif ic microenvironments or stromal compartments and image-guided pro-dr ug enzyme therap y, that can be used to target the TME and conclude with challenges and developments for the future of this f ield.

IMAGING THE PHYSIOLOGIC ENVIRONMENT The ph ysiologic microen vironment of cancers is an important component in the pro gression and the e volution of cancers, as well as their response to treatment. Of primary importance are the obser vations that tumors are often associated with a microen vironment that is both acidic and h ypoxic. Hypo xic v olumes result from poor perfusion and are generally characterized as either dif fusion-limited wherein tumor cells are > 150 µm away from a blood vessel1,2 or perfusion-limited wherein blood flow is periodic and cells experience transient ischemia.3,4 Acidosis is the result of poor perfusion coupled to increased glucose metabolism, 5,6 which is often visualized in the 18 F-deoxyglucose clinic as increased trapping of (FDG).7,8 It has been estimated that approximately half of the acid load in tumors comes from v olatile acids, for example, CO2, and half from nonvolatile acids, for example, lactic acid.9–11

Imaging Hypoxia Imaging Hypoxia with Non-MR Approaches

Electrodes The “Gold Standard” for o xygen measurements in tumors requires the use of small o xygen-sensing mini-electrodes, usuall y those made b y Eppendorf.

Molecular and Functional Ima ging of the Tumor Microenvironment

845

Immune System

Vascularization and Angiogenesis

Extracellular Matrix Degradative Enzymes Parenchymal and Stromal Cells

Physiological parameters

Figure 1. Schematic of the influences of the tumor microenvironment on cancer cells (marked C), superimposed on a 3D representation of immunofluorescent-labeled section from a human breast cancer xenograft, stained for endothelial cells (red), cell nuclei (blue), and cell proliferation (green). 3D confocal image provided by A. P. Pathak.

Numerous studies ha ve sho wn positive relationships between Eppendorf measurements and treatment outcomes in electrode-accessib le human tumors. It has been estimated that at least 20 independent measurements are necessary in order to reliably classify tumors as hypoxic.12 Eppendorf measurements in mice ha ve also demonstrated that tumors are more lik ely to metastasize if the y are hypoxic13,14 and that this ma y be related to acutel y fluctuating h ypoxia.15 Eppendorf par tial pressure of o xygen (pO2) measurements are generall y accepted as a Gold Standard. Although there are poor cor relations with other oxygen measuring techniques, 16–23 the Eppendorf is clinically predictive.24 Nonetheless, a major concern is that this technique is ab le to measure acute, but not necessaril y chronic, hypoxia, which is better measured b y alternative techniques. Nitroimidazoles Although Eppendorf measurements are infor mative, they are in vasive and are thus not a vailable to be routinely applied across patients. A widely used alternative is the use of 2-nitroimidazoles that reducti vely precipitate and can be assessed b y positron emission tomography (PET) or immunohistochemistr y (IHC). 25 Nitroimidazole markers are targeted selectively to viable, hypoxic cells through a bioreductive mechanism in which the marker is acti vated b y one-electron reductases (eg, c ytochrome P450 reductase). A series of reduction reactions occur as o xygen tension decreases and ultimately produce a h ydroxylamine derivative that is highly reactive and binds covalently to the original marker.26 For IHC measurements, the most widely used mark ers are EF-5 and pimonidazole. F or PET imaging, most studies have used 18F-fluoromisonidazole (F-MISO).

EF-5, a fluorinated derivative of etanidazole, is highly lipophilic, w hich is adv antageous in humans, w here the drug half-life ma y last longer than that of other tracers. EF-5 distribution is generall y observed with fluorescence microscopy using antibodies against EF-5 dr ug adducts.27,28 EF-5 binding can be con verted into a pO 2 value based on kno wn proper ties of EF-5 binding. These characteristics are consistent in vitro among a number of human cell lines deri ved from patient tumors. pO 2 values are also computed as a fraction of maximum EF-5 binding in tissue samples.29 EF-5 binding predicts the relative resistance of indi vidual tumors, 16 and IHC with fluorescent antibodies shows that the time to recur rence is shor test in tumors that are the most hypoxic. An advantage of EF-5 is its ability to quantify pO 2 levels, and results using this tracer have generally shown moderate h ypoxia in tumors, around 20 mm Hg, w hich is consistent with needle electrode values. On the basis of the success of EF-5 as an IHC tracer, an 18F-labeled tracer has been de veloped for PET imaging,30 although it has only had limited use. 31 F-MISO is readily adaptable to existing radiolabeling technolo gy and is accumulated b y ischemic myocardium, intestinal anaerobes, and h ypoxic tumors. Unbound tracer is rapidl y e xcreted, impro ving image contrast but can be confounded b y dif ferences in blood.32 F-MISO also suffers from relatively low signalto-background, with typical tumor-blood ratio cutoffs of ca.1.2. Thus, contrast is not as clear as that obser ved with FDG-PET images. In list-mode acquisition, w hich is becoming standard on all PET instr uments, the entire time course of F-MISO uptak e and retention can be followed, w hich should theoreticall y increase the

846

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

specificity of the technique, although this is rarely done in humans and 1.5 to 2.0 hours are needed for h ypoxiadependent partition. Although serial measurements can be made with 18F-FMISO PET, the shor test durations have been 1 da y apar t. Thus, the ability of 18F-FMISO PET to monitor temporal heterogeneity is limited. Optical imaging Optical imaging can detect o xygen saturation in tissues. Oxygen-saturation imaging is possible through near-infrared (NIR) tomography and has successfully been applied in combination with MRI in patients.33–35 The technique can quantify the o xy- and deoxy-hemoglobin (Hb) of breast lesions in vivo. Light at wavelengths of about 750 and 830 nm is transmitted through human breast tissue, and time spectra of the diffused light can be detected through the tissue. 36 The optical imaging technique measures absorption and scattering coefficients.36 The absor ption coef ficients at the tw o wavelengths are related to o xygen concentration and blood volume.36 Using this technique, o xygen saturation was observed to be lo wer in the tumor as compared with surrounding healthy tissue, which was in good agreement with the e xistence of h ypoxia in lar ger solid tumors. 33–35 Breast cancers contained decreased oxygen saturation and higher blood concentration than benign lesions and normal tissue. The average Hb concentration ([H]) of cancers was 0.130 ± 0.100 mM, and the average Hb saturation (Y) was 60 ± 9% compared with [H] = 0.018 ± 0.005 mM and Y = 69 ± 6% of background tissue.33 Imaging Hypoxia by MRI and MRS

Blood-Oxygen level dependent imaging Deoxy-Hb is paramagnetic, w hich per turbs local magnetic f ields, leading to increases in the transv erse relaxi vity, R2* (equal to 1/T 2*). Oxy-Hb is diamagnetic and thus does not lead to local susceptibility dif ferences. These effects can be measured with b lood-oxygen le vel dependent (BOLD) MRI. 37 The oxygenation of Hb is propor tional to the pO 2 of blood, which is assumed to be in equilibrium with tissue pO 2. However, the Hb o xygen binding curve (% bound vs pO 2) is signif icantly af fected b y a multitude of other f actors, including b lood flo w, CO 2, pH, and bisphospho glycerate le vels. Additionally, BOLD-MRI, as measured with g radient-echo, is dependent on the total concentration of deo xy-Hb, not on the ratio, and is thus af fected b y hematocrit, which can be highl y variable in tumors. 38 Nonetheless, BOLD-MRI is completel y nonin vasive and can yield time-dependent information about fluctuations in deoxyHb content of tumor v asculature.39 BOLD-MRI signals can be quantified using multiecho techniques, which can

be used to deri ve an R2* map. 40,41 Notably, R2* can be correlated to the e xtent of pimonidazole staining in human tumors and appears to be most sensitive for identifying hypoxic regions within tumors. 42 R2* maps also correlate with CA-IX staining, but not with GLUT -1, in rectal cancers. 43 Such maps reflect the local concentration of deo xy-Hb and minimize inflo w ef fects w hile retaining sensitivity to oxygenation changes that may be caused by alterations in flow. Typically, BOLD signals are measured in response to hyperoxic or h ypercapnic challenge, often deli vered as carbogen (95% O 2: 5% CO 2). Hypercapnia can induce vasodilation and changes in b lood flow, as w ell as acute changes in pH, which can change the Hb oxygen binding curve.37,38,44 In rodent tumors, changes in the BOLD signal in response to carbo gen mir ror changes in pO2,40,41,45–47 so that the BOLD response to carbo gen can be used to predict radiosensiti vity.48 These changes can be used to identify patients w ho may benef it from vasomodulation prior to radiotherap y.49 BOLD-MRI has a high failure rate in humans e xposed to carbo gen, due to hypercapnic-induced respirator y distress, w hich may be overcome with lo wer pCO 2 levels.50 Even without v asomodulation, R2* maps ma y be useful for radiotherap y planning and for monitoring the ef fects of anti vascular agents. Dynamic contrast-enhanced-MRI Because permeability in tumors is relatively high and thus not limiting, dynamic contrast-enhanced (DCE) MRI using smallmolecular-weight contrast agents is sensiti ve to b lood flow.51 DCE-MRI using higher molecular -weight contrast agents are more sensiti ve to per meability, and this can reflect activity of vascular endothelial growth factor (VEGF), which is induced by hypoxia.52–55 Gadoliniumlabeled albumin (Gd-albumin) is the most commonl y used high-molecular -weight contrast agent b ut is synthesized ad hoc and thus lacks interin vestigator consistency. Smaller agents with lar ge h ydrodynamic radii, such as P-792 and dendrimers, can beha ve as lar ger agents and pro vide more quality control. 56,57 Although high-molecular-weight contrast per meability can be correlated to h ypoxia,58,59 factors other than VEGF can affect v ascular per meability and f actors other than hypoxia induce VEGF expression. Vascular per meability strongly correlates with aggressiveness or metastatic potential.60,61 Electron paramagnetic resonance Unlike MR, which detects nuclei with magnetic moments, electron paramagnetic resonance (EPR) detects species with unpaired electrons, such as free radicals and transition metal complexes. Because electrons have strong magnetic

Molecular and Functional Ima ging of the Tumor Microenvironment

moments and transitions are more intense, EPR is much more sensitive than MR at a given magnetic field strength. A limitation to the use of EPR is that it requires the use of nontoxic stable free radicals for signal generation. Ho wever, these signals can be strongly perturbed by molecular oxygen, w hich has tw o unpaired electrons. Infusib le probes can accumulate in e xtracellular spaces and allo w EPR detection of pO2 levels throughout tumors. Typically, EPR is detected spectroscopicall y, with o xygen strongly increasing the spectral line width.62–64 EPR can also be adapted for imaging with appropriate f ield g radients.65 Overhauser-enhanced imaging (OMRI) has been used to monitor response to carbo gen e xposure and has been correlated with Eppendorf measurements and BOLD-MRI.66–70 An exciting extension of EPR imaging is the combination with h yperpolarized 13C.71,72 If a label could be identif ied that indicated h ypoxia based on the biochemistry of hypoxic tumors, imaging using h yperpolarized 13C-based tracers could be feasible. 19 F MRI/MRS The 19F nucleus has a lar ge gyromagnetic ratio and thus has an MR sensiti vity comparable to that of 1H. As there are little or no endo genous 19 F-containing molecules in the body , exogenous agents must be deli vered. To date, the most widel y used 19 F probes for measurement of pO 2 have been perfluorocarbon emulsions (PFCs), w hose relaxi vity is strongl y dependent on the concentration of dissolv ed molecular oxygen.73 These are typically not used for tumor hypoxia imaging, ho wever, as the y are generall y asymmetrical, highly temperature sensiti ve, and intravascular. Hexafluorobenzene (HFB) is a symmetric PFC, w hose relaxivity is relati vely insensiti ve to temperature and has been developed for oxygen imaging of tumors. 74 HFB oximetry compares favorably to optical and electrode measurements.75–77 A disadv antage to HFB is that it is not infusible and must be deli vered b y direct injection into the tumor. This also limits detection to the dif fusion distance along injection tracks.78 Nonetheless, this approach is no more in vasive than the Eppendorf electrode and allows simultaneous and longitudinal measurements of pO2 throughout the tumor o ver a time course of man y hours.

Imaging Tumor pH by MRI/MRS Tumor pH is generall y acidic 6 and can be predicted b y glycolytic or metabolic rates. pH is independent of hypoxia but ma y serve as a sur rogate for FDG uptak e.79 In addition, imaging can directl y measure pH; with hypoxia, there is some equi vocation about w hat is being

847

measured. It is lik ely that chronic h ypoxia co-localizes with low pH in the presence of glucose and that pH is low in areas that are highly glycolytic. Low pH also is dependent on poor perfusion. When protons dif fuse into the vasculature, they accumulate if poor perfusion interferes with a route of escape. Lo w pH stimulates in vasion in vitro,80 and pH can be manipulated in vi vo by chronic or acute treatment with bicarbonate. 81 When a spontaneous metastatic model is treated with bicarbonate, the number of metastases is reduced b y a third. 82 As might be expected, necrotic v olumes within a tumor e xhibit high pH, most likely because no metabolism occurs. The binding of pimonidazole, w hich is positi vely charged,83 with oxygen inter mediates depends g reatly on pH, w hereas EF-5 binding does not. 31 P MRS One of the earliest in vi vo uses of nuclear magnetic resonance was the measurement of er ythrocyte pH by 31P MRS. 84 Since then, the resonance frequenc y of endogenous inorganic phosphate, P i, has been v erified and validated as a robust indicator of intracellular pH in both humans and rodents.85,86 In 2004, these measurements were extended to include the simultaneous deter mination of extracellular pH using the exogenous and diffusible indicator, 3-aminopropylphosphonate, 3-APP.87,88 These measurements have proven useful for understanding pH re gulation in e xperimental tumors and ha ve consistentl y shown that the e xtracellular pH (pHe), of tumors is acidic, w hile the intracellular pH (pHi) is neutral-to-alkaline. 31P MRS has been used to measure pHi in human cancers, 89 yet it is unlikely that 3-APP will be clinicall y useful, as it can be neurotoxic if it crosses the blood-brain barrier. 1 H magnetic resonance spectroscopic imaging Although 31P MRS can be spatiall y localized , the lo w gyromagnetic ratio limits voxel sizes to ca. 8 × 8 × 8 mm3. Higher resolution can be obtained with other nuclei, such as 1H and 19F,90–93 which have been used to measure localized tumor pH using magnetic resonance spectroscopic imaging (MRSI) with spatial resolution approaching 1 × 1 × 1 mm.91,94,95 The most widely used 1H-labeled pH indicators are imidazoles. 90,96 One of these, the cellimpermeant 2-imidazole-1-yl-etho xy carbon yl propionic acid has been successfull y used to image pHe across a wide variety of tumors, with the consistent observation of an acidic and hetero geneous pHe, 91,94,97 as sho wn in Figure 2A. A major advantage of MR methods is the ability to obtain co-localized multiparametric data sets. An example of combined v ascular imaging using macro molecular DCE-MRI with pHe imaging is sho wn in Figure 2B. This ability can pro vide an understanding of the relationship betw een pHe, v ascular v olume, and

848

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

B

sensitive by using pH-sensiti ve paramagnetic lanthanide chelates (P araCEST),105–107 this approach still suf fers from insensitivity, requiring > 10 mM contrast agent and, like pH-dependent relaxometr y, also requires kno wledge of contrast agent concentration. Nonetheless, this is an area of acti ve in vestigation and it is possib le that approaches for impro ving sensiti vity will be developed.108

IMAGING THE ECM

Figure 2. A, (i) Dynamic contrast-enhanced magnetic resonance (DCE-MR) image of an MDA-MB-435 tumor and (ii) a corresponding extracellular pH (pHe) map acquired after injection of 2-imidazole-1-yl-ethoxy carbonyl propionic acid (IEPA) (obtained from Gillies RJ). B, 3D-reconstructed maps obtained from an MDA-MB-231 tumor. (i) vascular volume map in red, (ii) permeability map in green, (iii) pHe map generated from the magnetic resonance spectroscopic imaging data set following an intraperitoneal injection of IEPA, (iv) fused color image of vascular volume in red, permeability in green, and pHe in blue. Adapted with permission from Bhujwalla ZM et al.91

permeability in tumors with dif ferent phenotypic characteristics, such as increased metastatic ability or drug resistance. pH-Dependent relaxation Although MRSI approaches pro vided unparalleled resolution, dra wbacks include long acquisition times e ven at high f ield (ca. 40 minutes at 4.7 T). In the early 2000s, pH-dependent T1 relaxometry was introduced, with the promise that it could generate pH maps with higher spatio-temporal resolution. A number of pH-dependent contrast reagents ha ve been synthesized.98,99 These measurements, ho wever, require simultaneous pixel-by-pixel cor rection for concentration. This was f irst achieved using a dual-injection approach, where a pH-independent agent w as injected f irst, and the pharmacokinetics of this agent were used to correct for the concentration of the subsequentl y injected pH-dependent agent.95,100 An alternative may be an R1/R2 “ratiometric” method, y et this ma y onl y be v alid for lar ger contrast agents with correlation times > 1 ns. 101 Chemical exchange saturation transfer An alternative approach is to use the rate of acid-catal yzed exchange of e xogenous or endo genous amide hydrogens with bulk water to measure pH. 102–104 A drawback of this approach is that chemical e xchange saturation transfer (CEST) is relati vely insensiti ve, requiring > 50 mM exchangeable amides. Although it has been made more

An impor tant f actor in the TME is the ECM, w hich profoundly af fects proliferation, angio genesis, adhesion, migration, invasion, and metastasis. 109–111 Cancer cells are surrounded by a modif ied ECM, w hich is composed of a complex meshw ork of collagens, f ibrillar gl ycoproteins, and proteo glycans. Cancer cells interfere with ECM biosynthesis and thereby modify its str ucture and composition. Extensive ECM remodeling in tumors can proceed through de gradation of pre-e xisting ECM molecules and the neosynthesis of ECM components not present in the ECM of nor mal tissues. The ECM in tumors can displa y reduced integrity, which can influence drug delivery112 and cancer cell dissemination. 110 Proteases of e very known class ha ve been link ed to ECM remodeling in tumors, and in tur n to in vasion and metastasis of tumor cells. 113 Proteases are enzymes that cleave proteins at specific sites based on a specif ic amino acid sequence. They participate in de grading and remodeling the ECM and basement membranes. F or e xample, overexpression of cysteine, aspartic proteases, and matrix metalloproteases (MMPs) in se veral cancers has been associated with ECM remodeling, tumor agg ressiveness, and poor clinical outcome, respecti vely.114,115 Proteases degrade the ECM directl y or acti vate other proteases, thereby for ming a protease cascade w hose end result is ECM proteolysis.113 Lysosomal cathepsins are kno wn to either initiate this cascade or directly degrade the ECM in tumors.113 Only a limited number of proteases ha ve the capability to initiate the cleavage of fibrillar collagens due to their rigid and compact str uctures.116 These include lysosomal cathepsins, which have particularly been implicated in the collagen clea vage that occurs in acidic pH environments,116 as well as in laminin117 and fibronectin118 degradation. Hyaluronan de gradation is, among others, mediated b y l ysosomal h yaluronidase-2,119,120 which has been implicated in tumor de velopment.121 Heparanase participates in ECM de gradation and remodeling b y heparan sulfate degradation and is localized in lysosomes as well.122

Molecular and Functional Ima ging of the Tumor Microenvironment

Imaging ECM-Modifying Proteases The ability of cancer cells to in vade tissue and metastasize to other or gans represents tw o of the most lethal phenotypic traits of cancer. The initial escape of a tumor cell from its primary site requires loss of cell-cell attachment, follo wed b y in vasion of sur rounding connecti ve tissue, initial in vasion of a v essel, ar rest of circulating tumor cells and adhesion, e xtravasation from the v essel, and f inally fur ther in vasion at the site of metastatic lesion resulting in for mation of a metastatic nodule. 123 Cancer cells e xpress and secrete proteol ytic enzymes, several of w hich are secreted as inacti ve profor ms that become acti vated b y components in the ECM or are produced by stromal cells.124 The ability to noninvasively image the activity of degradative enzymes would be useful to characterize the in vasiveness of the tumor and to follow the action of proteol ytic enzyme inhibitors for therapy.125,126 Optical Imaging

The most e xtensively e xplored technique for imaging ECM-modifying proteases in cancer has been optical imaging of enzyme-acti vated probes. Optical imaging is an emer ging nonin vasive diagnostic imaging modality , which confers adv antages, such as high sensiti vity, lo w cost, and possibility of real-time image-guided sur gical procedures.127,128 The use of NIR optical imaging for the detection of such enzyme-acti vated probes has been explored to minimize tissue autofluorescence, photon attenuation, and light scattering. These probes contain quenched fluorophores, w hich fluoresce after being released from a car rier follo wing clea vage of the probe by a specific enzyme. 129 These “smar t” optical contrast agents ha ve been de veloped for MMP-2 detection114 and for NIR optical imaging of cathepsin B-sensitive probes in breast tumor models. 115 Although activation of the cathepsin B probe w as observed in both well-differentiated and highly invasive metastatic tumors, the fluorescence signal intensity w as higher in the metastatic tumor model, consistent with a higher cathepsin B protein expression or activity in the more metastatic tumor.115 It may be possible to use such an agent clinically to discer n cancers with higher proteol ytic acti vity that may as a result be more metastatic; NIR tomo graphic imaging of intrinsic contrast is already being tested in clinical trials.33,130 Multiple proteases including the dif ferent types of cathepsins are sequestered in l ysosomes, w hich are cellular or ganelles responsible for the de gradation and

849

removal of unnecessar y macromolecules. L ysosomes carry multiple proteases including the different types of cathepsins and therefore play a key role in cancer in vasion and metastasis. 131,132 Noninvasive NIR optical imaging of 6-O ʹ′-glucosamine-labeled NIR fluorescent probes w as recentl y de veloped to image l ysosomes in breast cancer cell cultures (F igure 3A), as w ell as following systemic administration in breast tumor models (Figure 3B). 133,134 MRI

Recently, MRI has been used to detect proteolytic enzymes, such as transglutaminases and h yaluronidase.125,126 Transglutaminases are a class of enzymes that mediate tissue remodeling by catalyzing covalent cross-links between proteins of the ECM.125 A low-molecular-weight MRI reporter was recently developed by using a peptide with a specif ic recognition sequence for transglutaminase link ed to Gd-diethylenetriaminepentaacetate (Gd-DTPA). Transglutaminase activity-dependent MRI contrast was generated by the alteration of the spin-lattice relaxation rate (1/T 1) of 125 water molecules in contact with the contrast agent. A similar mechanism was exploited to detect hyaluronidase activity by using a contrast agent that combines hyaluronan with Gd-DTPA and nontoxic agarose beads. Degradation of hyaluronan by hyaluronidases produces f actors that stimulate endothelial cell proliferation and promote neovascularization126; secretion of h yaluronidase by cancer cells ma y contribute to their in vasiveness.126 Hyaluronidase acti vity was detected noninvasively in cell culture and in an ovarian cancer xenograft model by changes induced by the contrast agent in the spin-lattice and the spin-spin relaxation rates (1/T2) of water.126

Imaging ECM Structure and Integrity Optical Imaging

Optical imaging techniques, such as dif ferential interference contrast (DIC) microscop y and second-har monic generation (SHG) microscop y,135 have been applied to study the ECM. Optical imaging using DIC optics generates contrast by means of different optical path length gradients passing through a Nomarski prism. 136 Such DIC microscopy has been used to dynamicall y track cellinduced ECM remodeling.136 SHG microscopy measures a signal from a nonlinear optical process, w hich requires an environment without a center of symmetr y, such as an interfacial re gion. This optical contrast mechanism has been used to image endogenous structural proteins, such as

850

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

A

C

B

Figure 3. A, Representative fluorescence microscopy images of human MDA-MB-231 breast cancer cells that were treated with 20 µM IR-2 for 24 h, fixed, and followed by immunofluorescence staining with CD-63 antibody. The near-infrared (NIR) fluorescence is displayed in red and CD-63 immunofluorescence in green. IR-2 co-localized well with the lysosomal marker CD-63 as evident by the yellow color. Scale bar, 15 µm. These images were acquired with a 100× oil-immersion lens. Adapted with permission from Li C et al.134 B, Representative NIR fluorescence image (top) and the corresponding superimposed photography and color-coded fluorescence image (bottom) of a MDA-MB-435 tumor-bearing mouse at 24 h postinjection of 100 nmol IR-2. Arrows point out the position of the tumor xenograft. Adapted with permission from Li C et al.134 C, Second-harmonic generation signal (SHG) with highlighted vessels in a Mu89 melanoma grown in the dorsal skin-fold chamber of a severe combined immunodeficiency mouse. Vessels were highlighted with an intravenous injection of 0.1 mL tetramethylrhodamine-dextran (10 mg/mL; red pseudocolor). SHG signal, green pseudocolor. There was no colocalization of SHG signal with the borders of blood vessels. The image shown is 275 µm in width. Adapted with permission from Brown E et al.138

collagen-rich la yers within the der mis.137 As sho wn in Figure 3C, 138 it is possib le to visualize collagen I f ibers using intrinsic contrast, in this instance, in a melanoma xenograft grown in a windo w chamber model in a se vere combined immunodef icient mouse mouse. With this optical imaging technique, Bro wn and colleagues 138 observed that enzymatic modif ication of tumor collagen b y relaxin can improve diffusive transport in tumors, which is important for drug delivery processes in tumors. Collagen matrix reorganization by tumor cells into a radially aligned pattern was sho wn to f acilitate local in vasion b y these cells. 139 Two-photon fluorescence cor relation microscop y w as applied in vivo and revealed that diffusion in the ECM of tumors displa ys bi-phasic beha vior.112 Optical coherence tomography (OCT) has the ability to detect high-resolution, cross-sectional microscopic images of tissue. 140 This technique uses low-coherence interferometry to produce a two-dimensional image of optical scattering from tissue microstructures in a w ay that is analo gous to ultrasonic pulse-echo imaging. OCT has a spatial resolution of a fe w micrometers. Endobronchial OCT w as ab le to identify abnormal areas in the lung and may be used as method for early endoscopic diagnosis. 140 MRI

Although optical imaging has been the most frequentl y used modality for assessing ECM structure, MRI methods can be applied to characterize ECM integrity by assessing

the diffusive hindrance of macromolecular contrast agents through the ECM. Since proteolytic enzymes degrade the ECM, its integrity or porosity can be used to characterize the invasiveness of the tumor and potentially design treatments that ma y alter ECM inte grity for impro ved dr ug delivery to the tumor interstitium. Understanding f actors, which influence the delivery, movement, and clearance of macromolecules through the ECM and supporting stroma of solid tumors, is also impor tant to delineate critical mechanisms in cancer invasion and metastasis. MRI of the macromolecular contrast agent albumin-Gd-DTP A w as used to characterize the e xtravascular transport of macromolecules through the ECM of solid tumors in vi vo.110 Macromolecular transpor t was characterized b y draining or pooling re gions of the agent. Macromolecular drain mostly occurred convectively through the ECM since the contrast agent w as rarel y obser ved within the tumoral lymphatic vessels and most lik ely followed paths of least resistance through the ECM. 110 Increased drainage area was obser ved in the more in vasive tumors that had a greater capacity for degrading ECM.110 MRI of l ymph nodes has been applied follo wing administration of the no vel contrast agent biotinBSA-Gd-DTPA-FAM/ROX to assess heparanasedependent changes and to allo w for earl y detection of metastatic dissemination. 141 This approach pro vides a combined anal ysis of b lood v olume, v ascular per meability, and interstitial con vection and therefore indirectly probes ECM str ucture and integrity.141

Molecular and Functional Ima ging of the Tumor Microenvironment

IMAGING TUMOR-ASSOCIATED TME CELLS The TME attracts man y nontumor cells, accounting in some cases for a signif icant fraction of the tumor mass. These cells f acilitate tumor pro gression b y assisting in escape from the immune system b y providing the vasculature necessar y for tumor perfusion and metastasis, b y contributing to the deposition of an ECM suppor tive of tumor and stroma cell proliferation, b y secretion of enzymes to f acilitate cell in vasion, and b y pro viding growth factors that induce tumor and stroma cell proliferation including induction of angio genesis and l ymphangiogenesis. Noninvasive imaging of TME cells has been demonstrated by a number of approaches.

Reporter Genes The first approach includes detection of endogenous TME cells via e xpression of detectab le repor ter genes. In this approach, the host stroma cells can be visualized either in spontaneous tumors or in tumor x enografts through the use of geneticall y modif ied mice e xpressing detectab le reporter genes. F or e xample, dual-fluorescence imaging of green fluorescent protein (GFP)-expressing mice bearing RFP-e xpressing tumors sho wed the recr uitment of neovasculature and immune cells, including l ymphocytes and dendritic cells to the tumor.142,143

Cell Surface Markers In the absence of easil y detectable genetically encoded reporters, endo genous TME cells can be detected through the use of tar geted contrast agents, aimed toward specif ic TME cell surf ace mark ers. Such contrast media can be administered systemicall y allowing detection of their retention b y stromal cells residing in the tumor. This approach was particularly successful in detecting tumor-associated endothelial cells, b y targeting endothelial cell surf ace receptors, and detecting tumor macrophages through their endoc ytosis of systemically administered nanopar ticles. Detection of other stromal cells b y this approach is limited b y the delivery of contrast agents.

Ex Vivo Labeling Tracking the recruitment of exogenous TME precursor (or stem) cells can be achie ved using cells labeled e x vi vo through induction of the expression of reporter genes or by loading with detectable reporter probes. A major advantage

851

of repor ter genes is that the y can be used for dynamic analysis of changes in TME cell number since label intensity is not diluted b y cell proliferation. Fur thermore, the label remains relatively specific and selective to the target cell. Reporter genes can also be used for imaging complex cell fate processes including induction of specific differentiation pathways. Reporter probes, on the other hand , can be used relatively safely with any cell type, can be cleared by the body , and require no genetic alteration, imposing less prolonged effects on the labeled cells. Upon administration of the labeled cells, it is possible to follow their mig ration and incor poration in the tumor . Noninvasive cellular imaging has been applied for tracking the recr uitment of multiple TME pro genitor cells to tumors, such as endothelial, stromal f ibroblast, and immune cells, with bone marrow cells already being used in clinical studies. 144 While fluorescence and bioluminescence imaging of reporter genes and tar geted contrast agents pro vide an attractive tool for mapping tumor TME cells in smallanimal models of cancer, these imaging modalities offer limited sensitivity and spatial resolution for deep-tissue detection, and thus cannot be easil y translated for clinical use. On the other hand , MRI is a method of choice for detection and in vivo tracking of labeled cells in pre-clinical and clinical imaging.

Vascular Endothelial Cells and Endothelial Progenitor Cells Receptors on endothelial cells lining the w alls of tumor blood vessels offer the most accessible cellular target for in situ labeling and imaging. Targeted contrast agents, particularly macromolecular-based or nanoparticle-based contrast agents, can be administered intra venously and retained by tumor endothelial cells. Targets for imaging include antibodies or ligand analo gs that are directed toward adhesion molecules, such as the inte grin-αvβ3, and receptors, such as the neural cell adhesion molecule (NCAM). The e xpression of the αvβ3-integrin is ele vated in endothelial cells during angio genesis. Targeted contrast agents allo w detection of tumor angio genesis b y MRI, using Gd-based and iron o xide-based par ticles.145–148 Similarly, imaging agents tar geting αvβ3 have also been developed for PET imaging 149 and for fluorescence imaging.147,150–153 The expression of NCAM is also elevated in tumor endothelial cells. To visualize this cell surface receptor, NCAM w as tar geted using a biotin ylated peptide. The bound peptide w as detected using streptavidin-labeled Gd-loaded apoferritin.154

852

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Expansion of the v asculature in tumors occurs predominantly through the process of sprouting angiogenesis involving proliferation, migration, and tube formation originating from e xisting local b lood v essels. Ho wever, bone mar row-derived stem cells and b lood-borne endothelial progenitor cells (EPC) can also contribute to a varying degree to the tumor vasculature. The role of EPC to tumor angio genesis can be studied using dynamic imaging-based tracking of prelabeled cells. Using the clinically approved contrast agent FePro, derived from ferumoxide (Feridex), and protamine sulfate as a transfection agent,155 EPC were effectively labeled allowing MRI mapping of their recr uitment to the angio genic rim of glioma tumors.156–158 Microscopic analysis conf irmed the incorporation of the labeled cells within tumor b lood vessels. Dual labeling of EPC for detection b y fluorescence microscopy and MRI w as achie ved with Gd-HPDO3A and Eu-HPDO3A sho wing sustained h yperintensity of cells injected in a matrigel plug. 159 Labeling EPC with magnetic par ticles allo wed their detection b y MRI and also allo wed the manipulation of their distribution, through the use of e xternal magnets, within a matrigel plug following inoculation.160

Fibroblasts, Myofibroblasts, and Mesenchymal Tumor Stroma Cells Mesenchymal stem cells, f ibroblasts, and m yofibroblasts are recr uited b y tumors, w here the y can constitute the major fraction of the TME cells, sometimes contributing to over half of the tumor mass. These cells par ticipate in the deposition of the ECM and the secretion of angiogenic and tumor-promoting g rowth f actors. Recr uitment of f ibroblasts to tumors w as demonstrated in tumor sections using GFP fluorescent cells. 161 In vivo detection of the recr uitment of f ibroblasts by tumors w as recently demonstrated using MRI (F igure 4) and NIR imaging. 162,163 Fibroblasts were labeled for MRI by active uptake mediated by caveolae of biotin-BSA-Gd-DTP A or b y endoc ytosis of SPIO (Feridex) and for NIR imaging b y labeling with DiR (Molecular Probes). Two-photon microscopy showed that the recr uited cells infiltrated the desmoplastic tumor and populated the vascular areas surrounding tumor nodules.

Tumor Immune Cells Tumor-associated macrophages pla y an impor tant role in tumor pro gression and angio genesis. Monoc ytes and macrophages can be easily labeled either ex vivo or in situ through their endo genous endoc ytic pathways.164–166 Iron

oxide particles delivered systemically are cleared from the circulation through endocytosis by circulating monocytes. These labeled cells populate normal lymph nodes allowing indirect detection, b y MRI, of tumor metastatic spread to lymph nodes through e xclusion of the administered par ticles.167 In established tumors, these particles can be used at early time points as b lood pool agents to visualize tumor vasculature, while at longer periods, MRI contrast primarily detects the retention of the contrast material in tumor macrophages.166 A similar labeling approach can also be used for ex vivo labeling of cells prior to their administration. Homing of macrophages to or thotopic C6 glioma tumors in Wistar rats was detected by MRI through ex vivo labeling of monoc ytes with lar ge (0.9 µm) MPIO iron oxide fluorescent par ticles.164 This labeling approach provides the possibility of single cell detection b y MRI.168 The homing of immune cells, including monoc ytes, dendritic cells, and T cells to tumors provides an attractive target for tumor-targeted cellular therapy and for exploiting cellular recr uitment for tar geted deli very of therap y. MRI provides a par ticularly impor tant tool for labeling and tracking the redistribution of these cells after delivery. Imaging the recr uitment of e xogenously labeled immune cells w as demonstrated b y a number of approaches.155,169–171 Notably, MRI w as recentl y applied for monitoring the accurac y of deli very and mig ration of cells in a clinical study of iron o xide-labeled autologous dendritic cell therapy of patients with melanoma. 169,171 A cross-link ed iron o xide nanopar ticle (CLIO-HD) was developed for cell labeling allo wing near single cell detection by MRI.172 These particles were used for tracking the recruitment of labeled T cells to tumors. Dynamic imaging of serial administration of the T cells sho wed these cells populating different regions of the tumor. The ability to label T cells using CLIO-tat iron oxide particles was used for MRI tracking ofT-cell recruitment to tumors and the role of high e xpression of CXCL12 in the repulsion of T cells and immune escape of tumors. Recruitment of the labeled T cells w as manifested b y significant hypointensity of MRI signal in control tumors relative to CXCL12 overexpressing tumors.173 Alternative, noniron oxide-based approaches for labeling immune cells for MRI detection of tumor tar geting include the use of T1-enhancing MnCl2174 and 19F MRI for tracking dendritic cells labeled with perfluoropolyether.170 Significant progress has been achieved in imaging and tracking TME-associated cells for pre-clinical and clinical applications. Ne vertheless, the cur rent sensiti vity and specificity leave room for impro vement. Labeling of cells with e xogenous repor ter probes is in variably associated with loss of label follo wing cell proliferation. No vel

Molecular and Functional Ima ging of the Tumor Microenvironment

853

R1 (S–1)

5

fibroblasts

0

systemic i.p control

labeled

Figure 4. Systemic recruitment of fibroblasts to subcutaneous ovarian carcinoma tumors. Fibroblasts labeled with biotin-BSA-Gd-DTPA (top right; MRI analysis of R1) or with iron oxide particles (Feridex; bottom right; Prussian Blue histological analysis) could be detected in the tumor, where they contributed to the angiogenic stroma rim. Adapted with permission from Granot D et al.163

methods for in vi vo imaging of repor ter genes might enable stable labeling of proliferating cells. Specif icity of the detected signal is limited not onl y b y sensiti vity of detection and the number of cells but also b y interference from backg round signals. Moreo ver, specif icity can also be lost due to secondar y enhancement b y transfer of the label to phagoc ytic cells. Despite these limitations, currently available noninvasive cellular imaging tools provide multiple options for detection and tracking of TME cells in pre-clinical tumor models and in human cancer.

IMAGING THE INTERACTION BETWEEN THE TME AND THE CANCER CELLS The TME, especiall y h ypoxia, can ha ve a multitude of effects on tumor function. Nonin vasive molecular and functional imaging is proving useful in understanding the relationship between the TME and the tumor phenotypic characteristics.

Imaging Cancer Cell-Stromal Cell Interactions in Cancer Cell Invasion Interactions between cancer cells and the ECM have been investigated e x vi vo using a MR-compatib le in vasion assay. This assay can dynamically track and quantify the invasion of cancer cells into ECM gel o ver time and simultaneously characterize o xygen tensions, as w ell as physiology and metabolism. 175 This system w as used to determine the ef fects of h ypoxia on prostate cancer cell invasion and metabolism. 176 By inser ting a la yer of human v ascular endothelial cells betw een the ECM gel layer and the cancer cells, this assa y w as used to investigate the contribution of endothelial cells to the invasive process under o xygenated and h ypoxic conditions.176 The data demonstrated that endothelial cells f acilitated in vasion under h ypoxic conditions and underlined the impor tance of functional imaging assa ys to deter mine the ef fect of microen vironmental changes on cancer cell invasion.176

854

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Tumor Vasculature MRI techniques ha ve been used for se veral decades to characterize functional aspects of the tumor vasculature, using BOLD-MRI and DCE-MRI as described in detail in Section “Imaging Hypo xia b y MRI and MRS. ” BOLD-MRI has been used to detect v ascular g rowth, vessel maturation, acute hypoxia, and treatment-induced hemodynamic changes. 59 Low-molecular-weight Gdbased agents that are clinicall y used can be used to derive tracer kinetic parameters: K trans (min−1), the v olume transfer constant between the blood plasma and the extravascular extracellular space (EES); k ep (min−1), the rate constant between the EES and blood plasma; and ve (%), the volume of the EES per unit volume of tissue. 177 For pre-clinical studies, a wide variety of contrast agents, macromolecular as w ell as tar geted, are a vailable for characterizing tumor vasculature. MRI of the macromolecular intra vascular contrast agent albumin-Gd-DTP A has been used to characterize tumor vascular volume and permeability surface area product (PSP).178 These vascular imaging capabilities in combination with molecular imaging repor ters can be used to understand the relationship betw een, for e xample, h ypoxia and tumor vasculature. In a recent study , MRI of albumin-GdDTPA w as combined with optical imaging to understand the relationship betw een v ascular parameters and hypoxia.178 These studies w ere perfor med using a PC-3 human prostate cancer xenograft stably expressing GFP under control of a h ypoxia-response element (HRE). Core gistered maps of v ascular v olume, per meability, and GFP demonstrated that h ypoxic re gions were characterized b y lo w v ascular v olume, and frequently by high permeability because hypoxia increased the e xpression of the potent per meability f actor VEGF.178 Consistent with these obser vations, VEGFoverexpressing PC-3 tumors contained re gions of high vascular v olume that w ere also more per meable.179,180 Control tumors exhibited few co-localized areas (yellow) of high v ascular v olume (red) and high per meability (green), whereas most of the vasculature in VEGF-overexpressing tumors was highly permeable to albumin-GdDTPA.

Tumor Metabolism Metabolic imaging either detects and visualizes endo genous molecules to generate “probe-free contrast” or uses isotope-labeled metabolite precursors that are being imaged. F or nuclear metabolic imaging, radiolabeled tracer compounds are administered that are analo gs of

endogenous metabolites. For MRS or MRSI, 1H and 31P MRS or MRSI can be used to detect endogenous metabolite le vels. Carbon-13 MRS has been applied for metabolic tracer studies follo wing administration of 13 C-labeled substrates. Functional Lactate MRSI

Typical metabolic features detected in vir tually all tumors are the activation of glycolysis and active suppression of the tricarboxylic acid c ycle.181 This activation of gl ycolysis in tumors has two components, one of which is due to genetic alterations acquired b y cancer cells that modify their biochemical pathways and result in abnor mal tumor metabolism. The second component is the TME, as h ypoxia activates glycolytic enzymes through transcriptional activation by the h ypoxia-inducible factor-1 (HIF-1). 182,183 Activation of gl ycolytic genes b y HIF-1 182,183 is considered critical for metabolic adaptation to h ypoxia through increased con version of glucose to p yruvate and subsequently to lactate. In a recent study , lactate-edited MRSI was perfor med on 38 patients with high-g rade gliomas before surgical diagnosis. 184 In this study , elevated lactate levels confirmed the lack of oxygenation within regions of compromised vascular integrity.184 Quantified lactate levels increased with the tumor grade in a recent study, which was consistent with pro gression from h ypoxia to necrosis. 185 However, the potential of assessing tumor h ypoxia through functional lactate MRSI needs to be fur ther v alidated in additional pre-clinical and clinical studies. Functional Total Choline MRSI

Increased total choline (tCho) le vels are typicall y observed in tumors and occur mostly due to increased PC levels,186 which have been link ed to onco genic ras signaling187 and onco genic transfor mation.188 In a recent study, the relationship between hypoxia, choline metabolites, and choline kinase (Chk) w as demonstrated in a human prostate cancer model using a combination of optical imaging and MRSI. 189 In this study , h ypoxic tumor re gions co-localized with re gions of ele vated tCho,189 as sho wn in F igure 5A and B . Molecular e vidence was provided that putative HREs within a putative chk-α promoter region functioned in vitro. 189 HIF-1 was shown to directly bind to a region of the endogenous chkα promoter in h ypoxic prostate cancer cells. 189 The heterogeneous h ypoxic TMEs ma y therefore contribute to the hetero geneous distribution of the tCho signal frequently observed in pre-clinical and clinical cancers.

Molecular and Functional Ima ging of the Tumor Microenvironment

A A

855

B

overlay of tCho in red and EGFP in green

corresponding EGFP fluorescence

tCho MRSI

pixel intensity correlation diagram

colocalized pixels in white on overlay

sum of pixel intensity correlation diagrams (n = 18) 255

green pixel intensity

warped brightfield

green pixel intensity

255

0 0

0 0

red pixel intensity

red pixel intensity

255

255

Figure 5. A, The warped bright-field image (top left) was used to guide the concurrent warping of the corresponding enhanced green fluorescent proteins (EGFP) fluorescence image (top center). Thus, the hypoxia-induced EGFP fluorescence and the corresponding total choline (tCho) magnetic resonance spectroscopic imaging (MRSI) data set (top right) were coregistered, as described in Glunde K et al.189 In the resulting overlay image (bottom left), the tCho MRSI image has been rendered in red while EGFP fluorescence is shown in green so that colocalization becomes yellow. Colocalization of tCho (red) and EGFP (green) was quantified by generating pixel intensity correlation diagrams of both images (bottom center). Here pixels within the yellow rectangular region and bounded by the red lines correspond to the pixels displayed in white in the overlay image shown in the bottom right panel. Therefore, an EGFP fluorescence and tCho MRSI point were only considered as co-localized if their respective intensities were higher than the threshold intensity, for example, those points with intensities that ranged between 50 to 255 (yellow rectangular), plus an additional intensity ratio threshold of 50% that is bounded by the two red lines. B, The sum of all pixel intensity correlation diagrams was generated to quantify the colocalization of the EGFP fluorescence (green) with the tCho MRSI signal and vice versa in all 18 tumors. The high number of pixels localized on what would be a straight line defined by y = x indicates good colocalization of EGFP fluorescence and tCho signal. This is indicative of increased tCho levels in tumor regions with high EGFP expression. Adapted with permission from Glunde K et al.189

TARGETING THE TME FOR THERAPY Current adv ances in multimodality imaging, contrast development, and molecular biolo gy are re volutionizing the applications of imaging in cancer therapy. The identification of specif ic targets in cancer is dri ving advances in novel image-guided platfor ms, such as liposomes and microencapsulation de vices, to deliver small interfering RNA (siRNA) to down regulate specif ic targets and pathways. Another exciting development is in the development, synthesis, and application of a no vel prototype agent for image-guided prodr ug therap y. These de velopments ma y be used to target specific components of the TME.

Targeting Hypoxia For se veral decades, the h ypoxic en vironment of solid tumors has pro ven to be a major obstacle for successful radiation and chemotherapy of cancer.190 Although hypoxic

environments occur in ischemic diseases, cells in these environments do not contribute to the recurrence of disease and are generall y destined to die. In contrast, cells in the hypoxic environments of solid tumors, have adapted to this environment,191–193 are the most resistant to radiation and chemotherapy and most lik ely lead to recur rence of the disease.194–197 Targeting hypoxic cells in solid tumors is critically important for successful treatment outcome. 194–197 To target hypoxia effectively, it is also necessar y to ha ve the ability to image it, and as described earlier, there are several noninvasive imaging techniques cur rently a vailable to image hypoxia. Several targeting strategies using hypoxia to drive gene therap y ha ve been pre viously repor ted.198–207 Many of these are based on the e xpression of an enzyme that converts a prodrug to its active form under hypoxia. For instance, horseradish peroxidase has been used to generate the acti ve prodr ug indole-3-acetic acid , and a bacterial nitroreductase enzyme has been used to generate the prodrug CB1954.208,209 These approaches, however, rely on

856

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

the diffusion of the prodrug to the site of enzyme synthesis, and prodrugs can be tak en up and acted upon b y any cell. Another strate gy tar gets h ypoxic cells using plasmids encoding suicidal-gene-lik e Bax or Caspase 3 that are selectively e xpressed under h ypoxic conditions. 201,205 The activity of these genes will be, to some e xtent, still dependent on the endo genous genetic makeup of the cells. Winnard and colleagues 210 have recently described the use of HRE-controlled e xpression of a proapoptotic gene. The mature form of human endonuclease G (EndoG) is used as the toxic gene.211 MDA-MB-435 cells that stably expressed hypoxia-inducible EndoG were found to undergo enhanced cell death compared with wild-type MD A-MB-435 cells following treatment with the h ypoxia-mimetic cobalt chloride. Tumors deri ved from MD A-MB-435-5 × HREODD-EndoG cells exhibited a retarded g rowth rate. In the future, this 5 × HRE-ODD-EndoG system can be used to target hypoxic regions using a lentiviral construct by incorporating the expression of the HRE-regulated proapoptotic gene into a lenti viral v ector. The lenti viral v ector can be constructed to e xpress an imaging repor ter to detect the expression of the proapoptotic gene in tumors.

Image-Guided Prodrug Therapy Damage to nor mal tissue is a major limiting f actor in chemotherapy and radiation therapy, and numerous strategies to protect nor mal tissue while maximizing damage to cancer cells ha ve been acti vely pursued. Prodr ug enzyme activation systems, where enzymes delivered to the tumor convert a nontoxic prodrug to a c ytotoxic drug are one of the most attractive of these strategies. Imaging the delivery of the enzyme w ould be impor tant for incor porating such a therapeutic strategy to target the TME. We have synthesized a prototype agent consisting of a cancer therapeutic prodrug enzyme labeled with multimodal MR and optical imaging reporters (Figure 6A). The prodrug enzyme, cytosine deaminase, converts a nontoxic prodrug 5-fluorocytosine (5-FC) to 5-fluorouracil (5-FU). The prodr ug 5-FC and its con version to 5-FU can be detected nonin vasively by 19F MRS. The conjugate demonstrates high relaxi vity, low c ytotoxicity, impro ved enzymatic specif icity to prodrug, efficient cell uptake, and high enzymatic stability in serum and in breast cancer cell culture. This prototype agent has been used to demonstrate the feasibility of

A = Biotin = Rhodamine = Gd3+-DOTA

O O N H

H N

N

NH2 NH2

O

S O

NH2

bCD-PLL

D 1900 D 1800

B

1700

T1 value (ms)

1800

C

Preinjection

5 min

60 min

24 h

1600

1700

1500 0

1600

40 80 Time (min)

120

1500 0

300

600 900 Time (min)

1200

1500

Figure 6. Example of image-guided therapy. In vivo MRI of the delivery of bacterial cytosine deaminase conjugated to poly-L-lysine (bCDPLL) in a wild-type MDA-MB-231 tumor. A, Schematic of bCD-PLL conjugate. Representative high-resolution T1-weighted MR images (B) and quantitativeT1 maps of a tumor (192 mm3) (C) before and after administration of bCD-PLL (1,000 mg/kg, intravenous). D, In vivo time-dependent changes in T1 values of the tumor (n = 4) before and after injection of bCD-PLL. Inset, changes in T1 within the first 2 h after injection of enzyme. Adapted with permission from Li C et al.213

Molecular and Functional Ima ging of the Tumor Microenvironment

image-guided prodrug enzyme therapy using MRI (Figure 6B–D); imaging can be used to time the administration of the prodrug when the enzyme has cleared from normal tissue but is still present in the tumor. Such an approach minimizes c ytotoxic side-ef fects. We are e xpanding this platform to targeted delivery of the prodrug enzyme using targeting peptides. We are also using this prototype platform to combine prodrug enzyme delivery and the delivery of siRNA to target, for instance, Chk. The optical reporter is especially useful to track the conjugate in cells and tissue

857

using microscopy while the MR reporter provides potential clinical translatability of this approach. These adv ances can be used to tar get specif ic pathw ays, microen vironments, and cell types within tumors.

CHALLENGES AND FUTURE DIRECTIONS The physiologic TME, interactions between cancer cells and stromal cells, such as endothelial cells, f ibroblasts,

C A B

E

F

D

Figure 7. “Mouse to man” multimodality imaging capabilities to investigate the tumor microenvironment. A, Mouse whole-body SPECT/CT imaging can be used to image tracers. B, Mouse optical imaging of a fluorescent reporter in a tumor. C, Clinical combined nearinfrared (NIR) optical/MR imaging. Photographs of the combined NIR optical/MRI scanner prototype including NIR interface placed inside the MRI breast coil (top row). Second row shows T1 MR images with dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) of the tumor. NIR images from the tumor MRI plane of hemoglobin, oxygen saturation, water fraction, and scatterer particle size and number density (bottom row). Adapted with permission from Carpenter CM et al.214 D, Proton MRI and MRSI of a 41-year-old patient with infiltrating ductal carcinoma of the breast. (a) Postcontrast T1-weighted images of the breast lesion. (b) MRSI images of water, choline, and lipids. Representative spectrum of elevated Choline (SNR = 10.6) within the lesion (c) and magnified (×50) region (d) demonstrates a detectable choline signal. Adapted with permission from Jacobs MA et al.215 E, Localized proton spectra from a human prostate cancer xenograft (PC-3) in a severe combined immunodeficient mouse, obtained with an in-plane spatial resolution of 1 mm × 1 mm and a slice thickness of 4 mm. Signal intensities of individual peaks, such as total choline and lipid triglycerides, in the spectrum obtained from each voxel can be converted into an image (top, total choline image). F, MRI of a macromolecular contrast agent albumin-Gd–DTPA to detect extracellular matrix integrity. Adapted with permission from Pathak AP et al.110,111 (Schematic of mouse and man adapted with permission from Shultz LD et al.216 red box in mouse represents activation of red fluorescent protein reporter in organ, and green circle represents activation of green fluorescent protein reporter in tumor; Figure 7 designed with assistance from A. P. Pathak.)

858

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

and macrophages, the ECM, and a multitude of secreted factors and cytokines influence progression, aggressiveness, and response of the disease to treatment.While the complexity of the TME poses a major challenge for effective treatment of cancer, it also provides unique targets to exploit. Multimodality imaging can play a role in understanding the interaction betw een the TME and cancer cells and in targeting the TME. Challenges associated with multimodality imaging will be accurate coregistration of images and image anal yses. Ne w directions for the future include tar geting specif ic microenvironments or stromal compar tments using image-guided incor poration of nanode vices and microdevices into tumors 212 for slo w release of therapeutic agents, gene delivery, and image-guided prodr ug enzyme therapy. With exciting developments in siRN A technology, the image-guided deli very of siRN A to a tumor, visualization of this deli very via liposome technology, and detection of a response are all achie vable with current imaging capabilities. Imaging cancer stem cells and imaging and targeting permissive or preventive microenvironmental niches for cancer stem cells are other areas that will ha ve signif icant impact on cancer research and treatment. Achieving successful image-guided TME tar geting will rel y hea vily on the molecular disco very and characterization of cell-type specific marker molecules, w hich should preferentiall y be localized on the cell surf ace. Other associated challenges, especially for MR methods are the amplification strategies that will be required to o vercome relati vely low sensitivity of detection. As shown in Figure 7, multimodality imaging of fers a v ast ar ray of techniques to probe the TME. Man y of these are at the pre-clinical stage (examples marked in red) and are v aluable methods to understand cancer . Ho wever, a major direction for the future will be to translate pre-clinical discoveries to the clinic, to noninvasively characterize and target the TME. Some of these are already a vailable for clinical translation (examples marked in yellow). The availability of multimodality imaging systems, such as PET/CT, optical/MR, and PET/MR, that combine the strengths and the capabilities of each modality should signif icantly facilitate this translation and allow bench to bedside applications.

REFERENCES 1. Thomlinson RH, Gray LH. The histological structure of some human lung cancers and the possib le implications for radiotherap y. Br J Cancer 1955;9:539–49. 2. Helmlinger G, Yuan F, Dellian M, Jain RK. Interstitial pH and pO2 gradients in solid tumors in vi vo: high-resolution measurements reveal a lack of cor relation. Nat Med 1997;3:177–82.

3. Brown JM. Evidence for acutely hypoxic cells in mouse tumours, and a possible mechanism of reoxygenation. Br J Radiol 1979;52:650–6. 4. Trotter MJ, Chaplin DJ, Durand RE, Oli ve PL. The use of fluorescent probes to identify re gions of transient perfusion in murine tumors. Int J Radiat Oncol Biol Ph ys 1989;16:931–4. 5. Gillies RJ, Schor nack PA, Secomb TW, Raghunand N . Causes and effects of hetero geneous perfusion in tumors. Neoplasia 1999;1:197–207. 6. Schornack PA, Gillies RJ . Contributions of cell metabolism and H+ diffusion to the acidic pH of tumors. Neoplasia 2003;5:135–45. 7. Gambhir SS, Czernin J, Schwimmer J, et al. A tabulated summary of the FDG PET literature. J Nucl Med 2001;42:1S–93. 8. Kelloff GJ, Hoffman JM, Johnson B , et al. Pro gress and promise of FDG-PET imaging for cancer patient management and oncolo gic drug development. Clin Cancer Res 2005;11:2785–808. 9. Griffiths JR, McIntyre DJ , Ho we FA, Stubbs M. Why are cancers acidic? A carrier-mediated diffusion model for H+ transport in the interstitial fluid. Novartis Found Symp 2001;240:46–62. 10. Guppy M, Leedman P, Zu X, Russell V. Contribution by different fuels and metabolic pathways to the total ATP turnover of proliferating MCF-7 breast cancer cells. Biochem J 2002;364:309–15. 11. Zu X, Gupp y M. Cancer metabolism: f acts, f antasy, and f iction. Biochem Biophys Res Commun 2008;313:459–65. 12. Doll CM, Milosevic M, Pintilie M, et al. Estimating hypoxic status in human tumors: a simulation using eppendorf oxygen probe data in cervical cancer patients. Int J Radiat Oncol Biol Ph ys 2003;55:1239–46. 13. Hill RP, De Jaeger K, Jang A, Cairns R. pH, hypoxia and metastasis. Novartis Found Symp 2001;240:154–65. 14. De Jae ger K, Ka vanagh MC, Hill RP . Relationship of h ypoxia to metastatic ability in rodent tumours. Br J Cancer 2001; 84:1280–5. 15. Cairns RA, Kalliomaki T, Hill RP. Acute (cyclic) hypoxia enhances spontaneous metastasis of KHT murine tumors. Cancer Res 2001;61:8903–8. 16. Evans SM, Judy KD, Dunphy I, et al. Comparati ve measurements of hypoxia in human brain tumors using needle electrodes and EF5 binding. Cancer Res 2004;64:1886–92. 17. Hedley D, Pintilie M, Woo J, et al. Carbonic anh ydrase IX expression, h ypoxia, and pro gnosis in patients with uterine cer vical carcinomas. Clin Cancer Res 2003;9:5666–74. 18. Le QT, Kovacs MS, Dorie MJ , et al. Comparison of the comet assa y and the oxygen microelectrode for measuring tumor o xygenation in head-and-neck cancer patients. Int J Radiat Oncol Biol Ph ys 2003;56:375–83. 19. Haugland HK, Vukovic V, Pintilie M, et al. Expression of h ypoxiainducible f actor-1α in cer vical carcinomas: cor relation with tumor o xygenation. Int J Radiat Oncol Biol Ph ys 2002; 53:854–61. 20. Airley R, Loncaster J , Davidson S, et al. Glucose transpor ter glut-1 expression correlates with tumor hypoxia and predicts metastasisfree sur vival in adv anced carcinoma of the cer vix. Clin Cancer Res 2001;7:928–34. 21. Loncaster J A, Har ris AL, Da vidson SE, et al. Carbonic anh ydrase (CA IX) expression, a potential ne w intrinsic marker of hypoxia: correlations with tumor o xygen measurements and pro gnosis in locally adv anced carcinoma of the cer vix. Cancer Res 2001;61:6394–9. 22. Milosevic M, Fyles A, Hill R. Interstitial fluid pressure in cer vical cancer: guide to tar geted therap y. Am J Clin Oncol 2001;24:516–21. 23. Aquino-Parsons C, Luo C, Vikse CM, Olive PL. Comparison between the comet assay and the o xygen microelectrode for measurement of tumor hypoxia. Radiother Oncol 1999;51:179–85. 24. Kavanagh MC, Tsang V, Chow S, et al. A comparison in indi vidual murine tumors of techniques for measuring o xygen levels. Int J Radiat Oncol Biol Phys 1999;44:1137–46.

Molecular and Functional Ima ging of the Tumor Microenvironment

25. Chapman JD, F ranko AJ, Shar plin J. A mark er for h ypoxic cells in tumours with potential clinical applicability . Br J Cancer 1981;43:546–50. 26. Chapman JD , Coia LR, Stobbe CC, et al. Prediction of tumour hypoxia and radioresistance with nuclear medicine mark ers. Br J Cancer Suppl 1996;27:S204–8. 27. Lord EM, Harw ell L, K och CJ. Detection of h ypoxic cells b y monoclonal antibody reco gnizing 2-nitroimidazole adducts. Cancer Res 1993;53:5721–6. 28. Koch CJ, Evans SM, Lord EM. Oxygen dependence of cellular uptake of EF5 [2-(2-nitro-1H-imidazol-1-yl)-N-(2,2,3,3,3-pentafluoropropyl)acet amide]: anal ysis of dr ug adducts b y fluorescent antibodies vs bound radioactivity. Br J Cancer 1995;72:869–74. 29. Koch CJ. Measurement of absolute o xygen levels in cells and tissues using oxygen sensors and 2-nitroimidazole EF5. Methods Enzymol 2002;352:3–31. 30. Ziemer LS, Ev ans SM, Kachur AV, et al. Nonin vasive imaging of tumor h ypoxia in rats using the 2-nitroimidazole 18FEF5. Eur J Nucl Med Mol Imaging 2003;30:259–66. 31. Koch CJ, Evans SM. Non-in vasive PET and SPECT imaging of tissue hypoxia using isotopically labeled 2-nitroimidazoles. Adv Exp Med Biol 2003;510:285–92. 32. Martin GV, Caldwell JH, Rase y JS, et al. Enhanced binding of the hypoxic cell mark er [3H]fluoromisonidazole in ischemic myocardium. J Nucl Med 1989;30:194–201. 33. Ntziachristos V, Yodh AG, Schnall MD , Chance B . MRI-guided diffuse optical spectroscopy of malignant and benign breast lesions. Neoplasia 2002;4:347–54. 34. Srinivasan S, P ogue BW, Jiang S, et al. Inter preting hemo globin and water concentration, o xygen saturation, and scattering measured in vivo by near-infrared breast tomography. Proc Natl Acad Sci U S A 2003;100:12349–54. 35. Srinivasan S, Pogue BW, Carpenter C, et al. Developments in quantitative oxygen-saturation imaging of breast tissue in vivo using multispectral near -infrared tomo graphy. Antioxid Redo x Signal 2007;9:1143–56. 36. Zhu Q, Conant E, Chance B . Optical imaging as an adjunct to sonograph in dif ferentiating benign from malignant breast lesions. J Biomed Opt 2000;5:229–36. 37. Howe FA, Robinson SP, McIntyre DJ , et al. Issues in flo w and o xygenation dependent contrast (FLOOD) imaging of tumours. NMR Biomed 2001;14:497–506. 38. Neeman M, Dafni H, Bukhari O , et al. In vi vo BOLD contrast MRI mapping of subcutaneous vascular function and maturation: validation by intravital microscopy. Magn Reson Med 2001;45:887–98. 39. Baudelet C, Ansiaux R, Jordan BF , et al. Ph ysiological noise in murine solid tumors using T2*-weighted gradient echo imaging: a marker for tumor acute h ypoxia? Ph ys Med Biol 2004;49:3389–411. 40. Robinson SP, Collingridge DR, Ho we FA, et al. Tumour response to hypercapnia and hyperoxia monitored by FLOOD magnetic resonance imaging. NMR Biomed 1999;12:98–106. 41. Howe FA, Robinson SP, Rodrigues LM, Grif fiths JR. Flow and oxygenation dependent (FLOOD) contrast MR imaging to monitor the response of rat tumors to carbo gen breathing. Magn Reson Imaging 1999;17:1307–18. 42. Carnell DM, Smith RE, Taylor NJ, et al. Evaluation of hypoxia within human prostate carcinoma using quantif ied BOLD MRI and pimonidazole immunohistochemical mapping. Br J Cancer 2003;88:S12. 43. Tatum JL, Kelloff GJ, Gillies RJ, et al. Hypoxia: importance in tumor biology, nonin vasive measurement b y imaging, and v alue of its measurement in the management of cancer therap y. Int J Radiat Biol 2006;82:699–757. 44. Neeman M, Provenzale JM, Dewhirst MW. Magnetic resonane imaging applications in the e valuation of tumor angio genesis. Semin Radiat Oncol 2001;11:1–13.

859

45. Maxwell RJ , Robinson SP , McIntyre DJ , et al. Comparison of changes in g radient echo 1H MR image intensity and pO2 in rodent tumours in response to carbo gen breathing. Proceedings of the 7th Inter national Society of Magnetic Resonance in Medicine; 1999;495–495. 46. Al Hallaq HA, Ri ver JN, Zamora M, et al. Cor relation of magnetic resonance and oxygen microelectrode measurements of carbogeninduced changes in tumor o xygenation. Int J Radiat Oncol Biol Phys 1998;41:151–9. 47. Al Hallaq HA, F an X, Zamora M, et al. Spectrall y inhomogeneous BOLD contrast changes detected in rodent tumors with high spectral and spatial resolution MRI. NMR Biomed 2002;15:28–36. 48. Robinson SP, Rodrigues LM, Ojugo AS, et al. The response to carbogen breathing in e xperimental tumour models monitored b y gradient-recalled echo magnetic resonance imaging. Br J Cancer 1997;75:1000–6. 49. Taylor NJ, Baddeley H, Goodchild KA, et al. BOLD MRI of human tumor o xygenation during carbo gen breathing. J Magn Reson Imaging 2001;14:156–63. 50. Rijpkema M, Kaanders JH, Joosten FB , et al. Ef fects of breathing a hyperoxic hypercapnic gas mixture on blood oxygenation and vascularity of head-and-neck tumors as measured b y magnetic resonance imaging. Int J Radiat Oncol Biol Ph ys 2002;53:1185–91. 51. Padhani AR, Dzik-Jurasz A. P erfusion MR imaging of e xtracranial tumor angiogenesis. Top Magn Reson Imaging 2004;15:41–57. 52. Knopp MV, Giesel FL, Marcos H, et al. Dynamic contrast-enhanced magnetic resonance imaging in oncology. Top Magn Reson Imaging 2001;12:301–8. 53. Roberts TP. Ph ysiologic measurements b y contrast-enhanced MR imaging: e xpectations and limitations. J Magn Reson Imaging 1997;7:82–90. 54. Degani H, Chetrit-Dadiani M, Bo gin L, Fur man-Haran E. Magnetic resonance imaging of tumor v asculature. Thromb Haemost 2003;89:25–33. 55. Schirner M, Menrad A, Stephens A, et al. Molecular imaging of tumor angiogenesis. Ann N Y Acad Sci 2004;1014:67–75. 56. de Lussanet QG, Langereis S, Beets-Tan RG, et al. Dynamic contrastenhanced MR imaging kinetic parameters and molecular w eight of dendritic contrast agents in tumor angio genesis in mice. Radiology 2005;235:65–72. 57. Daldrup-Link HE, Brasch RC. Macromolecular contrast agents for MR mammography: current status. Eur Radiol 2003;13:354–65. 58. Dafni H, Israel y T, Bhujwalla ZM, et al. Ov erexpression of v ascular endothelial growth factor 165 drives peritumor interstitial convection and induces l ymphatic drain: magnetic resonance imaging, confocal microscop y, and histolo gical tracking of triple-labeled albumin. Cancer Res 2002;62:6731–9. 59. Baudelet C, Cron GO , Gallez B . Deter mination of the maturity and functionality of tumor v asculature by MRI: cor relation between BOLD-MRI and DCE-MRI using P792 in e xperimental fibrosarcoma tumors. Magn Reson Med 2006;56:1041–9. 60. Bhujwalla ZM, Artemov D, Natarajan K, et al. Vascular dif ferences detected b y MRI for metastatic v ersus nonmetastatic breast and prostate cancer xenografts. Neoplasia 2001;3:143–53. 61. He Q, Xu RZ, Shkarin P , et al. Magnetic resonance spectroscopic imaging of tumor metabolic markers for cancer diagnosis, metabolic phenotyping, and characterization of tumor microen vironment. Dis Markers 2003;19:69–94. 62. Dinguizli M, Jeumont S, Beghein N, et al. Development and evaluation of biocompatible films of polytetrafluoroethylene polymers holding lithium phthaloc yanine cr ystals for their use in EPR o ximetry. Biosens Bioelectron 2006;21:1015–22. 63. Charlier N, Beghein N, Gallez B. Development and evaluation of biocompatible inks for the local measurement of o xygen using in vivo EPR. NMR Biomed 2004;17:303–10. 64. Gallez B, Swartz HM. In vivo EPR: when, how and why? NMR Biomed 2004;17:223–5.

860

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

65. Subramanian S, Matsumoto K, Mitchell JB , Krishna MC. Radio frequency continuous-w ave and time-domain EPR imaging and overhauser-enhanced magnetic resonance imaging of small animals: instrumental developments and comparison of relative merits for functional imaging. NMR Biomed 2004;17:263–94. 66. Krishna MC, Subramanian S, K uppusamy P, Mitchell JB . Magnetic resonance imaging for in vi vo assessment of tissue o xygen concentration. Semin Radiat Oncol 2001;11:58–69. 67. Krishna MC, English S, Yamada K, et al. Ov erhauser enhanced magnetic resonance imaging for tumor o ximetry: core gistration of tumor anatomy and tissue o xygen concentration. Proc Natl Acad Sci U S A 2002;99:2216–21. 68. Afeworki M, Miller NR, Devasahayam N, et al. Preparation and EPR studies of lithium phthaloc yanine radical as an o xymetric probe. Free Radic Biol Med 1998;25:72–8. 69. Braun RD, Lanzen JL, Dewhirst MW. Fourier analysis of fluctuations of oxygen tension and b lood flow in R3230Ac tumors and muscle in rats. Am J Physiol 1999;277:H551–68. 70. Brurberg KG, Graf f B A, Rofstad EK. Temporal hetero geneity in oxygen tension in human melanoma x enografts. Br J Cancer 2003;89:350–6. 71. Golman K, Ardenkjaer-Larsen JH, P etersson JS, et al. Molecular imaging with endogenous substances. Proc Natl Acad Sci U S A 2003;100:10435–9. 72. Ardenkjaer-Larsen JH, Fridlund B, Gram A, et al. Increase in signalto-noise ratio of > 10,000 times in liquid-state NMR. Proc Natl Acad Sci U S A 2003;100:10158–63. 73. Parhami P, Fung BM. Fluorine-19 relaxation study of perfluoro chemicals as o xygen car riers. J Ph ys Chem 1983; 87:1928–31. 74. Mason RP, Hunjan S, Constantinescu A, et al. Tumor oximetry: comparison of 19F MR EPI and electrodes. Adv Exp Med Biol 2003;530:19–27. 75. Mason RP, Constantinescu A, Hunjan S, et al. Re gional tumor o xygenation and measurement of dynamic changes. Radiat Res 1999;152:239–49. 76. Kodibagkar VD, Wang X, Mason RP. Physical principles of quantitative nuclear magnetic resonance o ximetry. F ront Biosci 2008;13:1371–84. 77. Xia M, K odibagkar V, Liu H, Mason RP . Tumour oxygen dynamics measured simultaneousl y b y near -infrared spectroscop y and 19F magnetic resonance imaging in rats. Ph ys Med Biol 2006;51:45–60. 78. Zhao D, Constantinescu A, Hahn EW, Mason RP. Differential oxygen dynamics in tw o diverse dunning prostate R3327 rat tumor sublines (MAT-Lu and HI) with respect to g rowth and respirator y challenge. Int J Radiat Oncol Biol Ph ys 2002;53:744–56. 79. Pauwels EK, Ribeiro MJ , Stoot JH, et al. FDG accumulation and tumor biology. Nucl Med Biol 1998;25:317–22. 80. Martinez-Zaguilan R, Seftor EA, Seftor RE, et al.Acidic pH enhances the invasive behavior of human melanoma cells. Clin Exp Metastasis 1996;14:176–86. 81. Raghunand N , Gillies RJ . pH and chemotherap y. No vartis F ound Symp 2001;240:199–211. 82. Raghunand N, Gatenby RA, Gillies RJ. Microenvironmental and cellular consequences of altered b lood flow in tumours. Br J Radiol 2003;76 Suppl 1:S11–22. 83. Vordermark D , Bro wn JM. Endo genous mark ers of tumor h ypoxia predictors of clinical radiation resistance? Strahlenther Onk ol 2003;179:801–11. 84. Moon RB, Richards JH. Deter mination of intracellular pH b y 31-P magnetic resonance. J Biol Chem 1973;248:7276–8. 85. Gillies RJ, Alger JR, Den Hollander J A, Shulman RG. Intracellular pH measured b y NMR: methods and results. In: Nuccitelli R, Deamer DW, editors. Intracellular pH. Ne w York, NY: Alan R. Liss; 1982. p. 79–104.

86. Griffiths JR. Are cancer cells acidic? Br J Cancer 1991;64:425–7. 87. Gillies RJ, Liu Z, Bhujw alla ZM. 31P-MRS measurements of e xtracellular pH of tumors using 3-aminoprop ylphosphonate. Am J Physiol 1994;267:C195–203. 88. Gillies RJ, Raghunand N, Garcia-Martin ML, Gatenby RA. pH imaging. A re view of pH measurement methods and applications in cancers. IEEE Eng Med Biol Mag 2004;23:57–64. 89. Negendank W. Studies of human tumors b y MRS: a re view. NMR Biomed 1992;5:303–24. 90. Gil S, Zaderenzo P , Cr uz F, et al. Imidazol-1-ylalkanoic acids as extrinsic 1H NMR probes for the deter mination of intracellular pH, e xtracellular pH and cell v olume. Bioor g Med Chem Lett 1994;2:305–14. 91. Bhujwalla ZM, Artemov D, Ballesteros P, et al. Combined v ascular and e xtracellular pH imaging of solid tumors. NMR Biomed 2002;15:114–9. 92. Ojugo AS, McSheehy PM, McIntyre DJ , et al. Measurement of the extracellular pH of solid tumours in mice b y magnetic resonance spectroscopy: a comparison of e xogenous (19)F and (31)P probes. NMR Biomed 1999;12:495–504. 93. Mason RP . Transmembrane pH g radients in vi vo: measurements using fluorinated vitamin B6 deri vatives. Cur r Med Chem 1999;6:481–99. 94. van Sluis R, Bhujwalla ZM, Raghunand N, et al. In vi vo imaging of 1 extracellular pH using H MRSI. Magn Reson Med 1999;41:743–50. 95. Garcia-Martin ML, Martinez GV, Raghunand N, et al. High resolution pH(e) imaging of rat glioma using pH-dependent relaxi vity. Magn Reson Med 2006;55:309–15. 96. Gil MS, Cr uz F, Cerdan S, Ballesteros P . Imidazol-1-ylalkanoate esters and their cor responding acids. A novel series of e xtrinsic 1 H NMR probes for intracellular pH. Bioor g Med Chem Lett 1992;2:1717–22. 97. Garcia-Martin ML, Herigault G, Remy C, et al. Mapping extracellular pH in rat brain gliomas in vi vo b y 1H magnetic resonance spectroscopic imaging: comparison with maps of metabolites. Cancer Res 2001;61:6524–31. 98. Zhang S, Wu K, Sher ry AD. A no vel pH-sensiti ve MRI contrast agent. Angew Chem Int Ed Engl 1999;38:3192–4. 99. Aime S, Delli CD , Terreno E. No vel pH-repor ter MRI contrast agents. Angew Chem Int Ed Engl 2002;41:4334–6. 100. Raghunand N, Howison C, Sherry AD, et al. Renal and systemic pH imaging b y contrast-enhanced MRI. Magn Reson Med 2003;49:249–57. 101. Aime S, Fedeli F, Sanino A, Terreno E. A R2/R1 ratiometric procedure for a concentration-independent, pH-responsi ve, Gd(III)-based MRI agent. J Am Chem Soc 2006;128:11326–7. 102. Ward KM, Balaban RS. Deter mination of pH using w ater protons and chemical e xchange dependant sturation transfer (CEST). Magn Reson Med 2000;44:799–802. 103. Ward KM, Aletras AH, Balaban RS. A new class of contrast agents for MRI based on proton chemical exchange dependent saturation transfer (CEST). J Magn Reson 2000;143:79–87. 104. Zhou J, Payen JF, Wilson DA, et al. Using the amide proton signals of intracellular proteins and peptides to detect pH effects in MRI. Nat Med 2003;9:1085–90. 105. Aime S, Bar ge A, Delli CD, et al. P aramagnetic lanthanide(III) complexes as pH-sensiti ve chemical e xchange saturation transfer (CEST) contrast agents for MRI applications. Magn Reson Med 2002;47:639–48. 106. Woods M, Woessner DE, Sherry AD. Paramagnetic lanthanide complexes as PARACEST agents for medical imaging. Chem Soc Rev 2006;35:500–11. 107. Cai S, Seu C, K ovacs Z, et al. Sensiti vity enhancement of multidimensional NMR experiments by paramagnetic relaxation effects. J Am Chem Soc 2006;128:13474–8.

Molecular and Functional Ima ging of the Tumor Microenvironment

108. Yoo B, Pagel MD. An overview of responsive MRI contrast agents for molecular imaging. Front Biosci 2008;13: 1733–52. 109. Lochter A, Bissell MJ . In volvement of e xtracellular matrix constituents in breast cancer. Semin Cancer Biol 1995;6:165–73. 110. Pathak AP, Artemov D , Neeman M, Bhujw alla ZM. L ymph node metastasis in breast cancer xenografts is associated with increased regions of extravascular drain, lymphatic vessel area, and invasive phenotype. Cancer Res 2006;66:5151–8. 111. Pathak AP, Artemov D, Ward BD, et al. Characterizing e xtravascular fluid transpor t of macromolecules in the tumor interstitium b y magnetic resonance imaging. Cancer Res 2005;65:1425–32. 112. Alexandrakis G, Brown EB, Tong RT, et al. Two-photon fluorescence correlation microscopy reveals the tw o-phase nature of transpor t in tumors. Nat Med 2004;10:203–7. 113. Skrzydlewska E, Sulk owska M, K oda M, Sulk owski S. Proteol yticantiproteolytic balance and its regulation in carcinogenesis. World J Gastroenterol 2005;11:1251–66. 114. Bremer C, Bredow S, Mahmood U, et al. Optical imaging of matrix metalloproteinase-2 acti vity in tumors: feasibility study in a mouse model. Radiology 2001;221:523–9. 115. Bremer C, Tung CH, Bo gdanov A Jr, Weissleder R. Imaging of differential protease e xpression in breast cancers for detection of aggressive tumor phenotypes. Radiology 2002;222:814–8. 116. Song F, Wisithphrom K, Zhou J, Windsor LJ. Matrix metalloproteinase dependent and independent collagen de gradation. F ront Biosci 2006;11:3100–20. 117. Zheng WQ, Looi LM, Cheah PL. Cor relation between laminin and cathepsin D e xpressions in breast carcinoma. Tumori 2002;88:296–9. 118. Guinec N, Dalet-Fumeron V, Pagano M. “In vitro” study of basement membrane degradation by the cysteine proteinases, cathepsins B, B-like and L. Digestion of collagen IV , laminin, f ibronectin, and release of gelatinase acti vities from basement membrane fibronectin. Biol Chem Hoppe Seyler 1993;374:1135–46. 119. Lepperdinger G, Mullegger J, Kreil G. Hyal2—less active, but more versatile? Matrix Biol 2001;20:509–14. 120. Lepperdinger G, Strobl B, Kreil G. HYAL2, a human gene e xpressed in many cells, encodes a lysosomal hyaluronidase with a novel type of specificity. J Biol Chem 1998;273:22466–70. 121. Bertrand P, Courel MN, Maingonnat C, et al. Expression of HY AL2 mRNA, h yaluronan and h yaluronidase in B-cell non-Hodgkin lymphoma: relationship with tumor agg ressiveness. Int J Cancer 2005;113:207–12. 122. Zetser A, Levy-Adam F, Kaplan V, et al. Processing and activation of latent heparanase occurs in l ysosomes. J Cell Sci 2004;117: 2249–58. 123. Chambers AF, Groom AC, MacDonald IC. Dissemination and growth of cancer cells in metastatic sites. Nat Rev Cancer 2002;2:563–72. 124. Mueller MM, Fusenig NE. F riends or foes—bipolar ef fects of the tumour stroma in cancer. Nat Rev Cancer 2004;4:839–49. 125. Mazooz G, Mehlman T, Lai TS, et al. De velopment of magnetic resonance imaging contrast material for in vivo mapping of tissue transglutaminase activity. Cancer Res 2005;65:1369–75. 126. Shiftan L, Israely T, Cohen M, et al. Magnetic resonance imaging visualization of h yaluronidase in o varian carcinoma. Cancer Res 2005;65:10316–23. 127. Shah N , Cer ussi A, Ek er C, et al. Nonin vasive functional optical spectroscopy of human breast tissue. Proc Natl Acad Sci U S A 2001;98:4420–5. 128. Koyama Y, Hama Y, Urano Y, et al. Spectral fluorescence molecular imaging of lung metastases tar geting HER2/neu. Clin Cancer Res 2007;13:2936–45. 129. Weissleder R, Tung CH, Mahmood U, Bogdanov A Jr. In vivo imaging of tumors with protease-acti vated near -infrared fluorescent probes. Nat Biotechnol 1999;17:375–8.

861

130. Pogue BW, Poplack SP, McBride TO, et al. Quantitative hemoglobin tomography with diffuse near-infrared spectroscopy: pilot results in the breast. Radiology 2001;218:261–6. 131. Glunde K, Guggino SE, Solaiyappan M, et al. Extracellular acidif ication alters l ysosomal trafficking in human breast cancer cells. Neoplasia 2003;5:533–45. 132. Glunde K, F oss CA, Takagi T, et al. Synthesis of 6 ʹ′-O-lissaminerhodamine B-glucosamine as a no vel probe for fluorescence imaging of l ysosomes in breast tumors. Bioconjug Chem 2005;16:843–51. 133. Li C, Greenwood TR, Bhujwalla ZM, Glunde K. Synthesis and characterization of glucosamine-bound near -infrared probes for optical imaging. Org Lett 2006;8:3623–6. 134. Li C , Greenwood TR, Glunde K. Glucosamine-bound near-infrared fluorescent probes with l ysosomal specif icity for breast tumor imaging. Neoplasia 2008;10:389–98. 135. Friedl P. Dynamic imaging of cellular interactions with e xtracellular matrix. Histochem Cell Biol 2004;122:183–90. 136. Petroll WM, Ma L. Direct, dynamic assessment of cell-matrix interactions inside f ibrillar collagen lattices. Cell Motil Cytosk eleton 2003;55:254–64. 137. Campagnola PJ, Millard AC, Terasaki M, et al. Three-dimensional high-resolution second-har monic generation imaging of endo genous str uctural proteins in biolo gical tissues. Bioph ys J 2002;82:493–508. 138. Brown E, McKee T, diTomaso E, et al. Dynamic imaging of collagen and its modulation in tumors in vivo using second-harmonic generation. Nat Med 2003;9:796–800. 139. Provenzano PP, Eliceiri KW, Campbell JM, et al. Collagen reor ganization at the tumor -stromal interf ace f acilitates local in vasion. BMC Med 2006;4:38. 140. Tsuboi M, Hayashi A, Ikeda N, et al. Optical coherence tomo graphy in the diagnosis of bronchial lesions. Lung Cancer 2005;49:387–94. 141. Dafni H, Cohen B, Ziv K, et al. The role of heparanase in lymph node metastatic dissemination: dynamic contrast-enhanced MRI of Eb lymphoma in mice. Neoplasia 2005;7:224–33. 142. Yang M, Li L, Jiang P, et al. Dual-color fluorescence imaging distinguishes tumor cells from induced host angio genic v essels and stromal cells. Proc Natl Acad Sci U S A 2003; 100:14259–62. 143. Hoffman RM, Yang M. Color -coded fluorescence imaging of tumor host interactions. Nat Protoc 2006;1:928–35. 144. Callera F, de Melo CM. Magnetic resonance tracking of magnetically labeled autologous bone marrow CD34+ cells transplanted into the spinal cord via lumbar puncture technique in patients with chronic spinal cord injur y: CD 34+ cells’ migration into the injured site. Stem Cells Dev 2007;16:461–6. 145. Sipkins D A, Cheresh D A, Kazemi MR, et al. Detection of tumor angiogenesis in vi vo by αvβ3-targeted magnetic resonance imaging. Nat Med 1998;4:623–6. 146. Winter PM, Car uthers SD, Kassner A, et al. Molecular imaging of angiogenesis in nascent Vx-2 rabbit tumors using a no vel ανβ3targeted nanoparticle and 1.5 Tesla magnetic resonance imaging. Cancer Res 2003;63:5838–43. 147. Mulder WJ, Strijk ers GJ, Habets JW, et al. MR molecular imaging and fluorescence microscopy for identification of activated tumor endothelium using a bimodal lipidic nanopar ticle. F ASEB J 2005;19:2008–10. 148. Zhang C, Jugold M, Woenne EC, et al. Specif ic targeting of tumor angiogenesis b y RGD-conjugated ultrasmall super paramagnetic iron o xide par ticles using a clinical 1.5-T magnetic resonance scanner. Cancer Res 2007;67:1555–62. 149. Chen X, Park R, Shahinian AH, et al. 18F-labeled RGD peptide: initial evaluation for imaging brain tumor angio genesis. Nucl Med Biol 2004;31:179–89.

862

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

150. Chen X, Sie vers E, Hou Y, et al. Inte grin αvβ3-targeted imaging of lung cancer. Neoplasia 2005;7:271–9. 151. Hsu AR, Hou LC, Veeravagu A, et al. In vivo near-infrared fluorescence imaging of integrin αvβ3 in an or thotopic glioblastoma model. Mol Imaging Biol 2006;8:315–23. 152. Achilefu S, Bloch S, Markie wicz MA, et al. Syner gistic effects of light-emitting probes and peptides for tar geting and monitoring integrin expression. Proc Natl Acad Sci U S A 2005;102:7976–81. 153. Wang W, Wu Q, Pasuelo M, et al. Probing for inte grin αvβ3 binding of RGD peptides using fluorescence polarization. Bioconjug Chem 2005;16:729–34. 154. Geninatti Crich S, Bussolati B, Tei L, et al. Magnetic resonance visualization of tumor angio genesis by targeting neural cell adhesion molecules with the highly sensitive gadolinium-loaded apoferritin probe. Cancer Res 2006;66:9196–201. 155. Frank JA, Miller BR, Arbab AS, et al. Clinicall y applicable labeling of mammalian and stem cells b y combining super paramagnetic iron oxides and transfection agents. Radiology 2003;228:480–7. 156. Anderson SA, Glod J , Arbab AS, et al. Nonin vasive MR imaging of magnetically labeled stem cells to directl y identify neo vasculature in a glioma model. Blood 2005;105:420–5. 157. Arbab AS, Bashaw LA, Miller BR, et al. Intracytoplasmic tagging of cells with fer umoxides and transfection agent for cellular magnetic resonance imaging after cell transplantation: methods and techniques. Transplantation 2003;76:1123–30. 158. Arbab AS, Pandit SD, Anderson SA, et al. Magnetic resonance imaging and confocal microscop y studies of magneticall y labeled endothelial pro genitor cells traf ficking to sites of tumor angiogenesis. Stem Cells 2006;24:671–8. 159. Crich SG, Biancone L, Cantaluppi V, et al. Impro ved route for the visualization of stem cells labeled with a Gd-/Eu-chelate as dual (MRI and fluorescence) agent. Magn Reson Med 2004;51:938–44. 160. Wilhelm C, Bal L, Smirnov P, et al. Magnetic control of vascular network formation with magnetically labeled endothelial pro genitor cells. Biomaterials 2007;28:3797–806. 161. Ishii G, Sangai T, Ito T, et al. In vivo and in vitro characterization of human fibroblasts recruited selectively into human cancer stroma. Int J Cancer 2005;117:212–20. 162. Granot D , K unz-Schughart LA, Neeman M. Labeling f ibroblasts with biotin-BSA-GdDTPA-FAM for tracking of tumor-associated stroma b y fluorescence and MR imaging. Magn Reson Med 2005;54:789–97. 163. Granot D, Addadi Y, Kalchenko V, et al. In vi vo imaging of the systemic recr uitment of f ibroblasts to the angio genic rim of o varian carcinoma tumors. Cancer Res 2007;67:9180–9. 164. Valable S, Barbier EL, Ber naudin M, et al. In vi vo MRI tracking of exogenous monocytes/macrophages targeting brain tumors in a rat model of glioma. Neuroimage 2007;37 Suppl 1:S47–58. 165. Bulte JW, Kraitchman DL. Iron o xide MR contrast agents for molecular and cellular imaging. NMR Biomed 2004;17:484–99. 166. Corot C, Petry KG, Trivedi R, et al. Macrophage imaging in central nervous system and in carotid atherosclerotic plaque using ultrasmall superparamagnetic iron oxide in magnetic resonance imaging. Invest Radiol 2004;39:619–25. 167. Weissleder R, Elizondo G, Josephson L, et al. Experimental l ymph node metastases: enhanced detection with MR l ymphography. Radiology 1989;171:835–9. 168. Hinds KA, Hill JM, Shapiro EM, et al. Highl y efficient endosomal labeling of progenitor and stem cells with large magnetic particles allows magnetic resonance imaging of single cells. Blood 2003;102:867–72. 169. de Vries IJ, Lesterhuis WJ, Barentsz JO , et al. Magnetic resonance tracking of dendritic cells in melanoma patients for monitoring of cellular therapy. Nat Biotechnol 2005;23:1407–13. 170. Ahrens ET, Flores R, Xu H, Morel P A. In vivo imaging platfor m for tracking immunotherapeutic cells. Nat Biotechnol 2005;23:983–7.

171. Verdijk P, Scheenen TW, Lesterhuis WJ, et al. Sensitivity of magnetic resonance imaging of dendritic cells for in vi vo tracking of cellular cancer vaccines. Int J Cancer 2007;120:978–84. 172. Kircher MF, Allport JR, Gra ves EE, et al. In vi vo high resolution three-dimensional imaging of antigen-specif ic c ytotoxic T-lymphocyte trafficking to tumors. Cancer Res 2003;63:6838–46. 173. Vianello F, Papeta N, Chen T, et al. Murine B16 melanomas expressing high le vels of the chemokine stromal-deri ved f actor1/CXCL12 induce tumor -specific T cell chemorepulsion and escape from immune control. J Immunol 2006;176:2902–14. 174. Aoki I, Takahashi Y, Chuang KH, et al. Cell labeling for magnetic resonance imaging with the T1 agent manganese chloride. NMR Biomed 2006;19:50–9. 175. Ackerstaff E, Gimi B, Artemov D, Bhujwalla ZM. Anti-inflammatory agent indomethacin reduces in vasion and alters metabolism in a human breast cancer cell line. Neoplasia 2007;9:222–35. 176. Ackerstaff E, Artemov D, Gillies RJ, Bhujwalla ZM. Hypoxia and the presence of human v ascular endothelial cells af fect prostate cancer cell invasion and metabolism. Neoplasia 2007;9:1138–51. 177. Baudelet, C., Cron, G.O., and Gallez, B. (2006). Determination of the maturity and functionality of tumor v asculature by MRI: correlation between BOLD-MRI and DCE-MRI using P792 in e xperimental fibrosarcoma tumors. Magn Reson Med 56, 1041–9. 178. Laking GR, West C, Buckley DL, et al. Imaging vascular physiology to monitor cancer treatment. Crit Re v Oncol Hematol 2006;58:95–113. 179. Raman V, Artemov D, Pathak AP, et al. Characterizing vascular variables in h ypoxic re gions: a combined magnetic resonance and optical imaging study of a human prostate cancer model. Cancer Res 2006. [in press] 180. Raman V, Artemov D, Pathak AP, et al. Characterizing v ascular parameters in h ypoxic re gions: a combined magnetic resonance and optical imaging study of a human prostate cancer model. Cancer Res 2006;66:9929–36. 181. Raman V, Pathak AP, Glunde K, et al. Magnetic resonance imaging and spectroscopy of transgenic models of cancer . NMR Biomed 2007;20:186–99. 182. Kim JW, Tchernyshyov I, Semenza GL, Dang CV . HIF-1-mediated expression of pyruvate dehydrogenase kinase: a metabolic switch required for cellular adaptation to h ypoxia. Cell Metab 2006;3:177–85. 183. Semenza GL. HIF-1 and tumor pro gression: pathoph ysiology and therapeutics. Trends Mol Med 2002;8:S62–7. 184. Semenza GL, Artemov D , Bedi A, et al. ‘The metabolism of tumours’: 70 y ears later [discussion]. No vartis F ound Symp 2001;240:251–64. 185. Lupo JM, Cha S, Chang SM, Nelson SJ. Analysis of metabolic indices in re gions of abnor mal perfusion in patients with high-g rade glioma. AJNR Am J Neuroradiol 2007;28:1455–61. 186. Howe FA, Barton SJ, Cudlip SA, et al. Metabolic prof iles of human brain tumors using quantitati ve in vi vo 1H magnetic resonance spectroscopy. Magn Reson Med 2003;49:223–32. 187. Ackerstaff E, Glunde K, Bhujwalla ZM. Choline phospholipid metabolism: a target in cancer cells? J Cell Biochem 2003;90:525–33. 188. Ramirez de Molina A, Penalva V, Lucas L, Lacal JC. Re gulation of choline kinase acti vity b y ras proteins in volves Ral-GDS and PI3K. Oncogene 2002;21:937–46. 189. Aboagye EO, Bhujwalla ZM. Malignant transfor mation alters membrane choline phospholipid metabolism of human mammar y epithelial cells. Cancer Res 1999;59:80–4. 190. Glunde K, Shah T, Winnard PT Jr , et al. Hypo xia regulates choline kinase expression through hypoxia-inducible factor-1alpha signaling in a human prostate cancer model. Cancer Res 2008;68:172–80. 191. Young SD, Marshall RS, Hill RP. Hypoxia induces DNA overreplication and enhances metastatic potential of murine tumor cells. Proc Natl Acad Sci U S A 1988;85:9533–7.

Molecular and Functional Ima ging of the Tumor Microenvironment

192. Graeber TG, Osmanian C, Jacks T, et al. Hypoxia-mediated selection of cells with diminished apoptotic potential in solid tumours. Nature 1996;379:88–91. 193. Reynolds TY, Rockwell S, Glazer PM. Genetic instability induced by the tumor microenvironment. Cancer Res 1996;56:5754–7. 194. Yuan J , Nara yanan L, Rockw ell S, Glazer PM. Diminished DN A repair and ele vated mutagenesis in mammalian cells e xposed to hypoxia and low pH. Cancer Res 2000;60:4372–6. 195. Brizel DM, Scully SP, Harrelson JM, et al. Tumor oxygenation predicts for the lik elihood of distant metastases in human soft tissue sarcoma. Cancer Res 1996;56:941–3. 196. Harris AL. Hypoxia—a key regulatory factor in tumour g rowth. Nat Rev Cancer 2002;2:38–47. 197. Rofstad EK. Microen vironment-induced cancer metastasis. Int J Radiat Biol 2000;76:589–605. 198. Sutherland RM. Tumor hypoxia and gene e xpression—implications for malignant pro gression and therap y. Acta Oncol 1998;37:567–74. 199. Dachs GU, Patterson AV, Firth JD, et al. Targeting gene expression to hypoxic tumor cells. Nat Med 1997;3:515–20. 200. Dachs GU, Tozer GM. Hypo xia modulated gene e xpression: angiogenesis, metastasis and therapeutic e xploitation. Eur J Cancer 2000;36:1649–60. 201. de Groot FM, Damen EW , Scheeren HW . Anticancer prodr ugs for application in monotherap y: targeting hypoxia, tumor-associated enzymes, and receptors. Cur r Med Chem 2001;8:1093–122. 202. Harada H, Hiraoka M, Kizaka-K ondoh S. Antitumor effect of TAToxygen-dependent de gradation-caspase-3 fusion protein specif ically stabilized and acti vated in hypoxic tumor cells. Cancer Res 2002;62:2013–8. 203. Naughton DP. Dr ug targeting to h ypoxic tissue using self-inacti vating bioreducti ve deli very systems. Adv Dr ug Deli v Re v 2001;53:229–33. 204. Peters KB , Bro wn JM. Tirapazamine: a h ypoxia-activated topoisomerase II poison. Cancer Res 2002;62:5248–53. 205. Ruan H, Su H, Hu L, et al. A h ypoxia-regulated adeno-associated virus v ector for cancer -specific gene therap y. Neoplasia 2001;3:255–63.

863

206. Ruan H, Wang J, Hu L, et al. Killing of brain tumor cells by hypoxiaresponsive element mediated e xpression of B AX. Neoplasia 1999;1:431–7. 207. Shibata T, Giaccia AJ, Brown JM. Development of a hypoxia-responsive v ector for tumor -specific gene therap y. Gene Ther 2000;7:493–8. 208. Wouters BG, Weppler SA, Koritzinsky M, et al. Hypo xia as a tar get for combined modality treatments. Eur J Cancer 2002;38:240–57. 209. Greco O, Marples B, Dachs GU, et al. Novel chimeric gene promoters responsi ve to h ypoxia and ionizing radiation. Gene Ther 2002;9:1403–11. 210. Shibata T, Giaccia AJ, Brown JM. Hypoxia-inducible regulation of a prodrug-activating enzyme for tumor-specific gene therapy. Neoplasia 2002;4:40–8. 211. Winnard PTJ , Botlagunta M, Kluth JB , et al. Hypo xia induced human endonuclease G expression suppresses tumor growth in a xenograft model. Cancer Gene Ther 2008. [in press] 212. Li LY, Luo X, Wang X. Endonuclease G is an apoptotic DNase w hen released from mitochondria. Nature 2001;412:95–9. 213. Gimi B, Leong T, Gu Z, et al. Self-assembled three dimensional radio frequency (RF) shielded containers for cell encapsulation. Biomed Microdevices 2005;7:341–5. 214. Li C, Penet MF, Winnard PT Jr, et al. Image-guided enzyme/prodrug cancer therapy. Clin Cancer Res 2008;15:515–22. 215. Carpenter CM, Pogue BW, Jiang S, et al. Image-guided optical spectroscopy provides molecular-specific infor mation in vi vo: MRIguided spectroscop y of breast cancer hemo globin, w ater, and scatterer size. Opt Lett 2007;32:933–5. 216. Jacobs MA, Bark er PB, Bottomley PA, et al. Proton magnetic resonance spectroscopic imaging of human breast cancer: a preliminary study. J Magn Reson Imaging 2004;19:68–75. 217. Shultz LD, Ishikawa F, Greiner DL. Humanized mice in translational biomedical research. Nat Rev Immunol 2007;7:118–30.

51 NOVEL MR AND PET IMAGING IN THE RT PLANNING AND ASSESSMENT OF RESPONSE OF MALIGNANT GLIOMAS CHRISTINA TSIEN, MD

Metabolic magnetic resonance imaging (MRI) and positron emission tomography (PET) enables the analysis of tumor tissue proper ties including chemical composition, v asculature and perfusion, w ater mobility, per meability, hypoxia, and tumor proliferation. Metabolic MRI techniques including proton spectroscopic imaging, blood v olume and flo w imaging, v ascular per meability imaging, and dif fusion and dif fusion tensor imaging are readily available. These techniques are impor tant in providing information regarding biologic processes that are not adequately met b y mor phologic imaging alone. The validity of these techniques for tar get volume def inition and early prediction of treatment response andtumor progression are currently being investigated. Thus, this chapter will briefl y describe the moder n treatment of high-grade gliomas and then focus on the potential v alue of metabolic MRI and PET biomark ers in the clinical management of malignant gliomas.

CURRENT TREATMENT OF HIGH-GRADE GLIOMA Approximately 20,000 ne w cases of primar y malignant gliomas are diagnosed each y ear in the United States. 1 Glioblastoma multifor me (GBM) is the most common and the most agg ressive type of primar y malignant glioma. Despite advances in imaging, sur gery, and radiation, the prognosis for this disease remains dismal with a 2-year sur vival rate of less than 10%. 2 Until recently, the traditional standard treatment for GBM is sur gical resection, if feasib le, follo wed b y adjuv ant radiation.

864

Radiation therap y (R T) is an impor tant f actor in prolonging sur vival of primar y high-g rade gliomas. Doses of up to 60 Gy yielded a dose-related increase in o verall survival.3,4 An EOR TC/NCIC Phase III randomized trial for patients with ne wly diagnosed gliob lastoma demonstrated that the addition of concur rent temozolomide (TMZ), a radio sensitizer , during and after con ventional radiotherapy (60 Gy) impro ved median sur vival from 12.1 months (11.2–13.0) to 14.6 months (13.2–16.8) compared with RT alone.5 This has led to a new standard of care using combined chemo-RT with TMZ after resection when possible. In an accompanying paper, Hegi and colleagues6 also sho wed that O6-alk ylguanine-DNA alkyltransferase meth ylation status is associated with improved sur vival among gliob lastoma patients treated with TMZ. Although these f indings require fur ther validation in prospective clinical trials, these studies suggest molecular biomarkers may have potential prognostic or predictive value in assessing response to therapy.

ROLE OF RADIATION IN HIGH-GRADE GLIOMAS Although R T has been sho wn to prolong sur vival, patients treated with R T typicall y pro gress within the radiation f ield.7–9 The addition of concur rent and adjuvant TMZ to standard-dose radiation (60 Gy) impro ves survival, but the majority of failures continue to be local. Prior dose-escalation studies with radiation alone suggest that patter ns of f ailure ma y be altered with

Novel MR and PET Ima ging

sufficiently high doses of radiation. In a single institution study using accelerated proton boost to 90 cobalt gray equivalent (CGE), only 1/23 patients was found to have central recurrence.10 Tanaka and colleagues,11 in a retrospective series, demonstrated an impro vement in overall survival in patients treated with 90 Gy with conformal radiation compared with historical controls treated with 60 Gy. Furthermore, a decrease in the proportion of local f ailures w as noted in the 90 Gy patients. However, this apparent increase in local control is also associated with a higher rate of to xicity. Patients from prior brach ytherapy and stereotactic radiosurgery (SRS) boost trials had higher rates of symptomatic radiation necrosis in comparison to patients treated to the cur rent standard of 60 Gy.12–13 Prior dose-escalation studies ha ve relied on standard contrast-enhanced MRI to define the extent of disease. Con ventional tar get v olumes, ho wever, result in the treatment of lar ge v olumes of normal brain with high doses or radiation, and thus cause high complication rates of symptomatic radiation necrosis. Therefore, new imaging techniques are needed to better def ine target volumes that are at the highest risk for recur rence and differentiate normal brain.

3D CONFORMAL RADIATION TREATMENT PLANNING Traditionally, 3D conformal radiotherapy is planned using a single set of computed tomography (CT) scans acquired prior to radiation treatment. Ho wever, in addition to contrast-enhanced CT , data from other imaging modalities such as MRI and PET are becoming increasingl y important in precisely defining target volumes at risk. 14,15 Image re gistration per mits accurate mapping of imaging data from other modalities to a single coordinate system, typicall y the treatment planning CT scan. 16 In this way, the geometric and electron density infor mation required for accurate dose calculations from the CT scan is combined with anatomic or functional data obtained from the MRI or PET studies. Gross tumor v olume (GTV) is typicall y def ined on individual CT slices. A clinical tar get v olume (CTV) is then def ined that includes a mar gin around the GTV to account for areas of potential microscopic disease. The CTV is then typicall y further expanded by the addition of a uniform margin to create a planning target volume (PTV) (Figure 1). This margin is used to account for uncertainties associated with treatment setup and to assure consistent

865

dose delivery to the CTV. By using a 3D rendering of the patient’s anatom y reconstr ucted from the CT dataset, tumor and normal structures are displayed in a beam’s eye view fashion (anatomy is displayed as from the perspective of the radiation source). 17 This allows the careful selection of optimal beam directions that allow the best separation of the target from the normal structures (Figure 2).

INTENSITY-MODULATED RT Recent advances in radiation treatment planning include intensity-modulated radiation therap y (IMR T) using

Figure 1. Target volume definition for radiation planning for malignant high-grade gliomas is typically based on magnetic resonance imaging (MRI) using T1-weighted post-gadolinium. Gross target volume (GTV) (green) is defined on MRI, the clinical target volume (CTV) (blue) defined using a uniform 1.5 cm expansion of the GTV, and the planning target volume (red) is defined to account for patient set-up uncertainties using a uniform expansion of the CTV.

Figure 2. A beam’s eye view display of the planning target volume (pink) as well as the optic nerves (red), optic chiasm (yellow), and brainstem (white) in a patient with a malignant high-grade glioma. A five field noncoplanar intensity modulated radiation therapy plan was designed to improve tumor coverage while reducing the dose to the critical normal tissue structures.

866

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

inverse planning based on computer iterati ve optimization. Radiation beam is typically subdivided into a series of finite-sized beams (beamlets) of 1 × 1 cm or 5 × 5 mm2 (Figure 3). Based on predef ined clinical treatment goals, the intensity (fluence) of each beamlet is then optimized. Computer search algorithms are used to vary the individual beam intensities (beam w eights) to meet the desired treatment plan. The quality of the plan is then scored based on its ability to achie ve the predef ined clinical treatment goals. After the optimization is complete, a beamlet dose distribution is computed, and final intensity maps are then sequenced into a set of deli verable beam segments. For a more complete review, please refer to the IMRT Collaborative Working Group paper.18 IMRT with dose optimization has signif icantly improved our ability to maximize tar get dose w hile reducing the dose to normal critical nor mal str uctures such as the optic chiasm, brainstem, and spinal cord (F igure 4).19

FUNCTIONAL AND METABOLIC MRI AND PET Unlike CT, MRI provides superior soft tissue contrast and thus it is routinely used in the tar get volume definition of brain tumors. 20 Physiologic MRI reveals metabolic information in addition to mor phological imaging and can serve as a sur rogate marker for biologic processes. These techniques allow for the anal ysis of tumor tissue properties including chemical composition, v asculature, perfusion, and water mobility within tissue architecture. 21 With a number of tracer compounds a vailable, PET imaging provides unique infor mation regarding cellular and physiologic processes that ma y improve target volume definition in radiotherapy planning.22,23 These techniques ma y allo w for nonin vasive measurements of tumor hypoxia, proliferation index, and markers of apoptosis.24,25 Developing reliable biologic maps may permit us to test the h ypothesis that dif ferentially directing higher radiation doses to areas of g reater tumor burden or functional radioresistance would improve outcome.

METABOLIC MRI

Figure 3. Beam intensity map of a single treatment intensitymodulated radiation therapy field showing the varying 1 × 1 cm beamlet intensities including the low-dose gradient (shown in dark grey) near the critical structure, such as the brainstem.

With ne w de velopments in tar geted therapies and advancements in radiation treatment planning, it is becoming more impor tant to improve target volume definition and deter mine the most agg ressive tumor regions for potential dose intensif ication. Ev aluating tumor response earl y during the course of treatment ma y also allow for indi vidualized adjustments of therap y prior to

Figure 4. Proton spectra in a patient with newly diagnosed glioblastoma multiforme WHO grade 4. The spectra from the contrast enhanced lesion (voxel 1) demonstrate an elevated ratio of the choline to N-acetylaspartate (NAA) peak consistent with tumor in comparison to the spectra from the contra-lateral normal brain (voxel 3). A voxel beyond the contrast-enhanced lesion also demonstrated an elevated ratio of choline to NAA peak consistent with tumor (voxel 2).

Novel MR and PET Ima ging

the completion of treatment. In the future, these needs may be better met by new biological and genetic markers. Although significant progress has been made, these techniques are not yet readily available in the clinic. Metabolic MR provides an in vivo quantitative analysis of tumor tissue properties, including chemical composition, vasculature, perfusion, water mobility, and vascular permeability. Metabolic MRI techniques include proton spectroscopic imaging, b lood volume and flow imaging, v ascular permeability imaging, and dif fusion and dif fusion tensor imaging.20 The clinical usefulness of these techniques for target volume definition, differentiating tumor versus radiation effects, and early prediction of treatment response or tumor progression is currently being investigated.

MAGNETIC RESONANCE SPECTROSCOPY Proton magnetic resonance spectroscop y (MRS) detects proton metabolites in tissue in vi vo. It provides information regarding tumor proliferation, cell membrane breakdown, neuronal acti vity, and tumor necrosis. Concentrations of each of these chemical metabolites including choline, creatine, N-acetylaspar tate (N AA), lactate, and lipid can be mapped and the data can be obtained in either a 2D or a 3D acquisition. Cor responding high peaks in choline-containing compounds and reductions in N AA are noted within malignant brain tumors compared with the normal brain tissue. Increased choline in malignant tumors cor relates to tumor cell proliferation from cell membrane phospholipid tur nover.26 MRS has been initiall y studied in brain tumors to improve target volume delineation and also for evaluation of treatment response. Prior MRS studies have also confirmed that the spatial e xtent of the ele vated choline has been found beyond the contrast-enhanced lesion as defined on postgadolinium T 1-weighed MRI (see F igure 4). More recently, studies ha ve sho wn that 3D magnetic resonance spectroscopic imaging (MRSI) ma y ser ve as an adjunct to conventional MRI in delineating the extent of disease in high-g rade gliomas and combining standard MRI with MRSI volumes leads to improved coverage of areas that subsequently progress.27–29 Graves et al30 demonstrated a trend for improved survival outcome in patients treated with stereotactic radiosurgery for recur rent gliomas w hen the pretreatment MRS metabolic abnor mality (ele vated choline signal) was included within the radiosur gical tar get v olumes. Chan et al also demonstrated an impro ved sur vival in recurrent g rade IV glioma patients treated with SRS when the target volume overlapped with the pretreatment

867

MRS metabolic lesion. In the latter study , patients w ere divided into two groups based on the percentage of overlap betw een the radiosur gical tar get and the metabolic lesion. Median sur vival w as 15.7 months for patients with a g reater than 50% o verlap versus 10.4 months for those with less than 50% overlap. This difference was statistically significant (p < .01). Therefore, there is a potential that MRS ma y be used to impro ve tar get v olume definition in malignant gliomas and impro ve patient outcome.31

MR PERFUSION Cerebral blood volume (CBV) images help assess for tumor recurrence, evaluate early response to treatment including radiation, and predict outcome (F igure 5). 31–33 Estimation of CBV is obtained by a quick scan using dynamic acquisition of T2*-weighted images follo wing intravenous injection of a bolus of Gd-DTP A. CBV map is obtained b y integrating the area under the dynamic contrast uptak e curve and provides a relative measure of CBV.20 Significant tumor hetero geneity in CBV has been noted in previous studies. 32–34 Using a median or a mean CBV, anal yses a veraged o ver the w hole tumor v olume tends to ignore the hetero geneity of perfusion characteristics within high-grade gliomas. Treatments such as radiation may affect the regions of high or low CBV within a tumor differently. Galban et al analyzed changes in MR perfusion in 48 primar y high-g rade glioma patients using a no vel voxel-by-voxel method of quantifying hemodynamic alterations following therapy, parametric response map analysis. (PRM) Although onl y modest changes w ere observed in mean rCBV anal ysis, fractional v olume of decreasing rCBV (PRM rCBV) at week 1 and week 3 during treatment was an independent predictor of outcome in high-g rade gliomas. 35 In a study of 23 patients with primary high-g rade glioma, the pretreatment fractional tumor volume with high CBV w as a predictor of poor overall sur vival. Fur thermore, changes in CBV during treatment w ere also predicti ve of outcome. CBV ma y potentially identify re gions of the tumor that de velop resistance to therapy that may benefit from further treatment intensification.32

MR DIFFUSION Diffusion MRI is another technique to measure the mobility of water within tissue at the cellular le vel. Diffusion MRI appears to be sensitive and early indicator of both treatment response and overall survival (Figure 6).36

868

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 5. Cerebral blood volume (CBV) map pretreatment (middle panel) demonstrates areas of high CBV (red), approximately five times higher than of normal white matter, and extends beyond the contrast-enhancing tumor rim on magnetic resonance (MR) T1-weighted postgadolinum image (far right panel). Repeat MR perfusion imaging performed 3 weeks following radiation showed a substantial decrease in the initial high CBV (far left panel) suggestive of response to treatment.

Figure 6. Bifrontal glioblastoma noted on postcontrast T1-weighted magnetic resonance imaging (left panel) with calculated apparent diffusion coefficient map (ADC) (right panel) showing corresponding area of heterogenous high signal intensity.

Functional dif fusion maps (FDM) demonstrate that regional changes in apparent dif fusion coefficients can be quantif ied within a tumor .37 When patients with primary high-grade gliomas were stratif ied based on FDM changes from w eek 3 during treatment compared to baseline, they were found to have signif icantly different time to progression and overall survival.38 The potential for dif fusion-weighted MRI to pro vide a nonin vasive method of monitoring early therapeutic response to treatment in human malignant brain tumors has been demonstrated.37,38 Further v alidation of these anal yses with

sufficient patholo gical and clinical end points will be performed prior to its incor poration into routine clinical practice.

MR DYNAMIC IMAGING OF VASCULAR PERMEABILITY Angiogenesis pla ys an impor tant role in tumor g rowth, infiltration, and in vasion of malignant gliomas. Tumor neovascularization is closely linked to tumor growth, and therefore, characterizing the functional proper ties of

Novel MR and PET Ima ging

tumor v asculature ma y aid in better delineation of the most aggressive region of the tumor.39 Dynamic contrast-enhanced (DCE) MRI using T1-weighted imaging allo ws quantitati ve assessment of tumor vascular per meability. Estimation of v ascular permeability can be obtained by a quantitative estimate of the transport constant (k trans) of Gd-DTPA from blood plasma to the e xtravascular e xtracellular space b y using a pharmacokinetic model and dynamic imaging acquisition. 20 Compartmental modeling with arterial input function and dynamic imaging acquisition are required. These studies are therefore more complex than CBV estimation. DCE MRI allo ws for quantitati ve assessment of the functional characteristics of the v ascular microen vironment in gliomas. The effectiveness of therap y including radiation, chemotherap y, or immunotherap y is strongl y dependent on the func tion of the vasculature. For example, few chemotherapy drugs are capab le of passing the tight endothelial junctions that mak e up the b lood-tumor barrier (BTB). The time course of BTB opening follo wing radiation may be used to optimize scheduling of concurrent chemotherap y as w ell as radiation sensitizers. Cao and colleagues suggested that the optimal time window is typically follo wing 30 Gy of radiation and peaking at approximately 1 month following completion of RT. Furthermore, radiation appears to affect the BTB more readily than b lood-brain bar rier (BBB). This has impor tant implications in the design of clinical trials using chemotherapy in combination with RT.40 Another application of DCE MRI is as a potential biomarker of response to anti-angiogenic drugs including bevacizumab. Desjardins and colleagues demonstrated that decreases in ktrans by at least 50% occurred as early as 1 da y follo wing therap y in 6/13 patients and 12/13

869

patients following 1 c ycle of be vacizumab and CPT-11. Decrease in K trans was highly correlated with the percent decrease in contrast enhancement at end of one c ycle (Pearson correlation = 0.82; p = .0006). However, in this small g roup of patients, changes did not cor relate with patient outcome. 41 Similarly, another phase I trial found that decreases in v ascular per meability cor responded to levels of dr ug e xposure.42 DCE MRI ma y be used to assess tumor v ascular per meability changes induced b y antivascular drugs, radiation, or tumor progression.42,43

MR T1-WEIGHTED IMAGING MRI T1-weighted imaging precontrast and postintra venous bolus administration of a single dose of contrast material MR is routinely obtained on standard clinical MR protocols (Figure 7). T1-weighted signal intensity (SI) changes reflect interactions between mobile water protons and macromolecular protons within tissue. Based on preclinical data, reduced T1 SI may be associated with increased tissue density, possibly reflecting response to treatment. 44,45 Several studies have suggested distinct differences in the growth factor expression and vascular properties in the regions of the contrast-enhancing rim and non-enhancing core.46,47 Biopsy studies performed in patients with GBM at surgical resection using contrast-enhanced MRI appear to cor relate with imaging characteristics demonstrating increased v ascular per meability and v essel density in regions of contrast enhancement compared to nonenhancing areas. 46 Tsien et al 48 analyzed separatel y the contrast-enhancing tumor rim as compared to the nonenhanced (NE) tumor core using quantitati ve MR T1-SI changes to assess re gional response to radiation and provide additional infor mation in predicting outcome.

Figure 7. Precontrast and postcontrast T1 magnetic resonance imaging is shown on the right and central panel, respectively. Using a calculated image from the natural logarithmic ratio of the pair of post to pre contrast T1-weighted MR images, two distinct regions were defined: a contrast enhancing rim (red) and the non-enhancing tumor core (blue). These regions were generated automatically using the region of the normal brain receiving < 10 Gy.

870

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Imaging biomark ers ma y pro vide complementar y information to clinical, pro gnostic indicators that ha ve been v alidated as useful predictors of patient sur vival including R TOG recursi ve par titioning anal ysis (RP A) classification, tumor g rade, age, pretreatment tumor size, Karnofsky performance scale, and e xtent of tumor resection. In this study , multi variate anal ysis sho wed that SI changes in the NE tumor core at 1 month following treatment was predictive of overall survival. When both RTOG RPA classification and imaging parameters were included in a multivariate model, SI changes of the NE tumor core at 3 week RT (p = .05) remained statisticall y signif icant, whereas RTOG RPA classification did not.48 This suggests that additional information provided by the imaging parameters regarding outcome is not solel y predicted b y pretreatment factors.

Future Directions Metabolic MRI of fers the potential to not onl y evaluate differences in efficacy between patients but also to assess the heterogeneity of response within an individual tumor. The development of quantitative MRI sequences, including dif fusion-weighted, perfusion, DCE MR, and functional brain imaging methods such as MRS, may increase our ability to determine early response to treatment compared with conventional MR techniques. Earl y detection of regional response to treatment ma y allow an oppor tunity to adjust therapy on an indi vidual basis and identify patients who may benef it from fur ther treatment intensification to a specific tumor region or a change in therapy. Further work is needed to deter mine w hether these techniques can specif ically identify areas within the target that are at risk for failure. In the future, it seems likely that the complementary information obtained with multiple imaging modalities will lead to the most ef ficient delineation of the appropriate tar get volume. Identifying coincident re gions from multiple imaging modalities of leaky v asculature, mark ers of h ypoxia, and increased tumor cellular proliferation 49 may be useful in deter mining the most aggressive regions of the tumor for radiation treatment planning and monitoring response to therap y.

METABOLIC PET IMAGING Functional PET imaging has also been sho wn to more accurately reflect tumor extent than morphologic imaging in malignant gliomas. 18F-fluoro-deoxy-D-glucose-PET is able to detect brain tumors by a nonspecific assessment of increased cellular glucose metabolism in tumors. However, its use in target definition is complicated by the

high le vel of intrinsic glucose uptak e in the brain. 50 11C methionine PET scan reflects metabolic acti vity through increased transport mediated by type-L amino acid carriers that are highly expressed in malignant tumors and low uptake in nor mal tissue. 51–53 Increased 11C methionine uptake ma y reflect tumor proliferation as e videnced b y stereo-tactic biopsies cor related with higher PCN A expression, micro vessel density , and Ki-67 protein expression.54,55 Another PET tracer , 18F-fluorothymidine (FLT), is a thymidine analogue that is incorporated exclusively into DN A. It measures the acti vity of cellular thymidine kinase and increases several fold as cells enter the S phase and be gin DNA synthesis. 56 Increased FLT uptake and therefore th ymidine kinase le vels pro vide a direct measure of the cellular proliferation rate. FL T tumor uptake is also cor related with Ki-57 staining and has been sho wn as a potential biomark er in predicting treatment response in recurrent glioma.57 Thus, this chapter will discuss the potential v alue of metabolic PET imaging in def ining target volumes in the radiation planning of high-grade gliomas. 11 Metabolic imaging studies such C methionine PET (MET-PET) ma y impro ve our ability to identify target volumes at highest risk of local failure. MET-PET imaging demonstrates increased metabolic acti vity due to increased acti vation of the type L-mediated amino acid transpor t at the le vel of the BBB in glioma cells compared to nor mal brain. 56 There is an e vidence that MET-PET may be useful for identifying residual tumor after resection and identifying recur rent gliomas. 57,58 Nuutinen and colleagues 59 have in vestigated the v alue of MET-PET in radiation treatment planning and monitoring response in lo w-grade gliomas. Grosu and colleagues ha ve completed se veral studies demonstrating the benef it of 11C MET and 123I-α-methyl-tyrosine-single-photon computed tomo graphy (IMT -SPECT) in radiation treatment planning for recur rent and primar y gliomas.23,60,61 However, there are no prospecti ve studies that assessed w hether MET-PET prior to treatment ma y be used to predict sites of subsequent treatment f ailures. It is hypothesized that areas of increased MET-PET activity would be at higher risk for recurrence (Figure 8). In order to test this hypothesis, Lee and colleagues prospectively acquired MET-PET scans to compare initial areas of increased methionine uptak e with subsequent site of failure. Patterns of f ailure were determined by comparing recurrence volume with the delivered dose distribution. Recur rence v olume (rV OI) w as def ined b y a neuroradiologist on post-gadolinium T1-weighted MRI. Patterns of f ailure w ere then classif ied as central,

Novel MR and PET Ima ging

871

Figure 8. 31-year-old female with left temporal glioblastoma. 11C MET PET (left panel) demonstrates area of metabolic uptake extending beyond the T1 post-contrast-enhanced MRI (right panel) along the white matter tract. MR FLAIR (right panel) also demonstrates peritumoral edema. 11C MET PET provides additional information compared to conventional MR regarding appropriate target volume definition.

in-field, marginal, or distant according to the proportion of the rVOI contained within the 95% prescription isodose surface (IDL), that is, central failures were defined as ≥ 95% of rV OI encompassed b y the 95% IDL, in-field failure 80 to 95%, marginal failure 20 to 80%, and distant < 20% (Figure 9). 62 Pretreatment 11C MET-PET revealed areas of abnor mality not apparent b y contrastenhanced MRI. All patients with suboptimal co verage of the PET-GTV developed noncentral failures, whereas the majority of patients with adequatel y covered PETGTV had central f ailures. There was a statisticall y significant cor relation betw een increased MET -PET uptake outside the high-dose re gion and a subsequent noncentral f ailure (see F igure 8). 62 The cor relation between the pattern of failure and the adequacy of coverage of the pretreatment MET-PET supports the incorporation of MET-PET into radiation treatment planning, although additional prospective data are clearl y needed to cor roborate these f indings. These f indings suggest that it w ould be reasonab le to test a strate gy of using MET-PET to identify focal areas for a confor mal radiation boost. Previous studies ha ve also demonstrated that MET PET acti vity ma y e xtend be yond area of the contrast61 enhancing lesion on MRI. Grosu and colleagues retrospectively analyzed patients with primar y GBM w ho underwent MET -PET prior to radiation. In that study , which was limited to patients w ho had under gone resection, in 29 (74%) of 39 patients, the area of MET uptak e was larger than the contrast-enhancing Gd v olume. Focal MET uptake was detected up to 4.5 cm be yond the contrast-enhanced MRI and up to 4 cm on the T2-weighted MRI.62 However, this study did not include data on subsequent patter n of f ailure, so it w as unclear w hether the

MET-PET acti vity w ould be impor tant in identifying regions at high risk for progression. There is sufficient data to assess in a prospecti ve trial testing whether selectively increasing the radiation dose to MET-PET avid regions may improve treatment outcome. The University of Michigan has initiated an initial prospective study using 11C MET PET for target volume definition in primary GBM patients. By limiting the radiation boost volume, higher doses may be safely achieved, so focal dose escalation may be more successful than uniformly escalating dose across the entire contrast-enhancing MRI volume. Grosu et al demonstrated that treatment planning based on biologic imaging such as PET/SPECT imaging was associated with an improved survival compared with treatment planning using CT/MRI in recur rent gliomas; median sur vival of 9 months v ersus 5 months ( p = .03, lo g rank). Ho wever, in a multi variate model, combined radiation treatment with TMZ w as more important than the inclusion of biologic imaging (p = .04 and p = .3, respectively).61 In recurrent gliomas re-treated with radiation, careful tar get v olume delineation is important to reduce risk of nor mal central ner vous system toxicity. Biologic treatment planning ma y be important in def ining tar get v olumes for re-ir radiation b y decreasing the lik elihood of geo graphic miss and decreasing the volume of normal brain irradiated. The main limitation of 11C MET is the short half-life of approximately 20 minutes, limiting its a vailability to centers with an on-site c yclotron. Ho wever, an amino acid analo g 18F-labeled O-(2) fluoroeth yl-L-tyrosine (FET) radiolabeled with fluorine-18 is a similar isotope with a longer half-life that demonstrates similar results in tumor delineation. Stereotactic biopsies w ere performed to conf irm f indings. Sensitivity of conventional

872

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 9. Recurrence volume of interest (rVOI) was compared to the delivered dose distribution. 95% isodose line is shown in yellow. Central failures were defined as ≥ 95% of rVOI encompassed by the 95% isodose surface, in-field failure 80 to 95%, marginal failure 20 to 80%, and distant < 20%.

P

Central

In-field

Marginal

Well-covered PET-GTV

9

1

1

Inadequately covered PETGTV

0

2

3

0.004 (Fisher’s exact test)

Figure 10. All patients with suboptimal coverage of the PET-gross tumor volume (GTV) developed noncentral failures, whereas the majority of patients with adequately covered PET-GTV had central failures. A statistically significant correlation was noted between increased MET-PET uptake outside the high-dose region and a subsequent noncentral failure.

MRI for detection of tumor tissues w as 96% but specificity w as onl y 53%. The addition of 18F-FET PET substantially improved the specif icity to 94%. 63 A small exploratory study used FLT PET in recurrent gliomas as an earl y imaging biomark er of response to bevacizumab and irinotecan treatment in patients with recurrent high-grade gliomas was recently reported. Multivariate anal ysis suggested that FL T response w as an important predictor of survival compared with other clinical variables tested. Baseline FL T SUVs w ere not predictive of sur vival. FL T PET response of g reater than 25% reduction in tumor FLT uptake was found to be the threshold with the best predicti ve power of o verall survival. These changes were most predictive at 6 weeks but

changes as earl y as 1 to 2 w eeks during treatment w ere among the tw o most po werful independent predictors of survival among all variables tested including MRI. 58 Functional PET imaging shows promise in identifying areas of increased metabolic acti vity that ma y extend be yond area of contrast enhancement on MRI. Post-operative functional PET imaging ma y be ab le to identify residual tumor v olume that signif icantly correlates in predicting sur vival. Increased areas of MET PET uptake may also correspond to areas at highest risk of recurrence. Thus, MET PET uptake may significantly modify tar get v olumes b y focusing on the re gion of increased metabolic activity and sparing volume of normal brain treated. Functional PET imaging ma y also be an impor tant biomark er of response to therap y especially in the era of tar geted therapy and antiangio genic agents. Man y tar geted agents are c ytostatic or demonstrate response mainly in the contrast-enhancing lesion, and metabolic PET imaging ma y pro vide additional information to con ventional MRI to deter mine responses to therapy.

REFERENCES 1. CBTRUS. Statistical repor t: primar y brain tumors in the United States, 1998–2002. 2005. 2. Mahaley MS, Mettlin C, Natarajan N , et al. National sur vey of patterns of care for brain-tumor patients. J Neurosur g 1989;71:826–36. 3. Walker MD, Alexander E Jr, Hunt WE, et al. Evaluation of BCNU and/or radiotherap y in the treatment of anaplastic gliomas. A cooperative clinical trial. J Neurosurg 1978;49:333–43.

Novel MR and PET Ima ging

4. Walker MD, Green SB , Byar DP, et al. Randomized comparisons of radiotherapy and nitrosoureas for the treatment of malignant glioma after surgery. N Engl J Med 1980;303:1323–9. 5. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherap y plus concomitant and adjuv ant temozolomide for gliob lastoma. N Engl J Med 2005;352:987–96. 6. Hegi ME, Diserens AC, Gorlia T, et al. MGMT gene silencing and benefit from temozolomide in gliob lastoma. N Engl J Med 2005;352:997–1003. 7. Wallner KE, Galicich JH, Krol G, et al. P atterns of f ailure following treatment for gliob lastoma multiforme and anaplastic astroc ytoma. Int J Radiat Oncol Biol Phys 1989;16:1405–9. 8. Chan J, Lee S, F raass B, et al. Sur vival and f ailure patterns of highgrade gliomas after three-dimensional confor mal radiotherapy. J Clin Oncol 2002;20:1635–42. 9. Lee S, Fraass B, Marsh LH, et al. Patterns of failure following high dose 3D conformal radiotherapy for high grade astrocytomas: a quantitative dosimetric study. Int J Radiat Oncol Biol Ph ys 1999;43:79–88. 10. Fitzek MM, Thornton AF, Rabinov JD, et al. Accelerated fractionated proton/photon ir radiation to 90 cobalt g ray equi valent for glioblastoma multiforme: results of a phase II prospecti ve trial. J Neurosurg 1999;92:251–60. 11. Tanaka M, Ino Y, Nakagawa K, et al. High-dose conformal radiotherapy for supratentorial malignant glioma: a historical comparison. Lancet Oncol 2005;6:953–60. 12. Sneed PK, Lambor n KR, Larson D A, et al. Demonstration of brachytherapy boost dose-response relationships in gliob lastoma multiforme. Int J Radiat Oncol Biol Ph ys 1996;35:37–44. 13. Shrieve DC, Alexander E III, Black PM, et al. Treatment of patients with primar y gliob lastoma multifor me with standard postoperative radiotherapy and radiosur gical boost: pro gnostic factors and long-term outcome. J Neurosurg 1999;90:72–7. 14. Nelson SJ, Graves E, Pirzkall A, et al. In vi vo molecular imaging for planning radiation therapy of gliomas: an application of 1H MRSI. J Magn Reson Imaging 2002;16:464–76. 15. Zakian KL, K outcher JA, Ballon D , et al. De velopments in nuclear magnetic resonance imaging and spectroscop y: application to radiation oncology. Semin Radiat Oncol 2001;11:3–15. 16. Meyer CR, Boes JL, Kim B , et al. Demonstration of accurac y and clinical v ersatility of mutual infor mation for automatic multimodality image fusion using af fine and thin plate spline w arped geometric deformations. Med Image Anal 1997;1:195–206. 17. Galvin JM, Ezzell G, Eisbruch A, et al. Implementing IMRT in clinical practice: a joint document of the American Society for Therapeutic Radiology and Oncology and the American Association of Physicists in Medicine. Int J Radiat Oncol Biol Ph ys 2004;58:1616–34. 18. Intensity Modulated Radiation Therapy Collaborati ve Working Group. Intensity-modulated radiotherap y: cur rent status and issues of interest. Int J Radiat Oncol Biol Phys 2001;51:880–914. 19. Narayana A, Yamada J, Berry S, et al. Intensity-modulated radiotherapy in high-g rade gliomas: clinical and dosimetric results. Int J Radiat Oncol Biol Phys 2006;64:892–7. 20. Thornton AF Jr, Hegarty TJ, TenHaken RK, et al. Three dimensional treatment planning of astrocytomas: a dosimetry study of cerebral irradiation. Int J Radiat Oncol Biol Ph ys 1991;20:1309–15. 21. Cao Y, Sundgren PC, Tsien CI, et al. Physiologic and metabolic magnetic resonance imaging in gliomas. J Clin Oncol 2006;24: 1228–35. 22. Pardo F, Aronen H, Kennedy D, et al. Functional cerebral imaging in the evaluation and radiotherapeutic treatment planning of patients with malignant glioma. Int J Radiat Oncol Biol Ph ys 1994;30: 663–9. 23. Grosu AL, Feldmann H, Dick S, et al. Implications of IMT-SPECT for postoperati ve radiotherap y planning in patients with gliomas. Int J Radiat Oncol Biol Ph ys 2002;54:842–54.

873

24. Chao KS, Bosch WR, Mutic S, et al. A novel approach to o vercome hypoxic tumor resistance: Cu-ATSM-guided intensity-modulated radiation therapy. Int J Radiat Oncol Biol Phys 2001;49:1171–82. 25. Barthel H, Cleij MC, Colling ridge DR, et al. 3 ʹ′-deoxy-3ʹ′-[18F]fluorothymidine as a ne w mark er for monitoring tumor response to antiproliferative therapy in vivo with positron emission tomography. Cancer Res 2003;63:3791–8. 26. Preul MC, Caramanos Z, Collins DL, et al. Accurate, noninvasive diagnosis of human brain tumors by using proton magnetic resonance spectroscopy. Nat Med 1996;2:323–5. 27. Park I, Tamai G, Lee MC, et al. P atterns of recur rence anal ysis in newly diagnosed glioblastoma multiforme after three-dimensional conformal radiation therapy with respect to pre-radiation therap y magnetic resonance spectroscopic f indings. Int J Radiat Oncol Biol Phys 2007;69:381–9. 28. Pirzkall A, Li X, Oh J , et al. 3D MRSI for resected high-g rade gliomas before RT: tumor e xtent according to metabolic acti vity in relation to MRI. Int J Radiat Oncol Biol Phys 2004;59:126–37. 29. Pirzkall A, McKnight TR, Graves EE, et al. MR-spectroscopy guided target delineation for high-grade gliomas. Int J Radiat Oncol Biol Phys 2001;50:915–28. 30. Graves EE, Nelson SJ, Vigneron DB, et al. Serial proton MR spectroscopic imaging of recurrent malignant gliomas after gamma knife radiosurgery. AJNR Am J Neuroradiol 2001;22:613–24. 31. Chan AA, Lau A, Pirzkall A, et al. Proton magnetic resonance spectroscopy imaging in the evaluation of patients undergoing gamma knife sur gery for g rade IV glioma. J Neurosur g 2004;101: 467–75. 32. Cao Y, Tsien CI, Nagesh V, et al. Clinical in vestigation survival prediction in high-g rade gliomas b y MRI perfusion before and during earl y stage of R T. Int J Radiat Oncol Biol Ph ys 2006;64:876–85. 33. Kremer S, Grand S, Rem y C, et al. Cerebral b lood volume mapping by MR imaging in the initial evaluation of brain tumors. J Neuroradiol 2002;29:105–13. 34. Law M, Yang S, Wang H, et al. Glioma g rading: sensitivity, specificity, and predictive values of perfusion MR imaging and proton MR spectroscopic imaging compared with con ventional MR imaging. AJNR Am J Neuroradiol 2003;24:1989–98. 35. Galbán CJ, Chenevert TL, Meyer CR. Functional perfusion mapping of gliomas: an MRI biomark er of earl y treatment response. Nat Med 2008. [Submitted] 36. Chenevert TL, Stegman LD, Taylor JM, et al. Diffusion magnetic resonance imaging: an early surrogate marker of therapeutic efficacy in brain tumors. J Natl Cancer Inst 2000;92:2029–36. 37. Moffat BA, Chenevert TL, Lawrence TS, et al. Functional diffusion map: a nonin vasive MRI biomark er for earl y stratif ication of clinical brain tumor response. Proc Natl Acad Sci U S A 2005; 102:5524–9. 38. Hamstra DA, Chenevert TL, Moffat BA, et al. Ev aluation of the functional diffusion map as an earl y biomark er of time-to-pro gression and overall survival in high-grade glioma. Proc Natl Acad Sci U S A 2005;102:16759–64. 39. Cao Y, Nagesh V, Hamstra D, et al. The extent and severity of vascular leakage as e vidence of tumor agg ressiveness in high-g rade gliomas. Cancer Res 2006;66:8912–7. 40. Cao Y, Tsien CI, Shen Z, et al. Use of magnetic resonance imaging to assess blood-brain/blood-glioma barrier opening during conformal radiotherapy. J Clin Oncol 2005;23:4127–36. 41. Desjardins A, Barboriak DP , Her ndon JE, et al. Dynamic contrastenhanced magnetic resonance imaging (DCE-MRI) e valuation in glioblastoma (GBM) patients treated with be vacizumab (BEV) and irinotecan (CPT-11), ASCO Annual Meeting Proceedings Part I, [Abstract No 2029]. J Clin Oncol 2007;25 Suppl 18. 42. Rugo HS, Herbst Rs, Liu G, et al. Phase I trial of the oral anti-angiogenesis agent AG-013736 in patients with adv anced solid

874

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

tumors: phar macokinetic and clinical results. J Clin Oncol 2005;23:5474–83. Just M, Thelen M. Tissue characterization with T1, T2, and proton density values: results in 160 patients with brain tumors. Radiolo gy 1988;169:779–85. Steen RG, Koury BSM, Granja CI, et al. Effect of ionizing radiation on the human brain: white matter and gray matter T1 in pediatric brain tumor patients treated with conformal radiation therapy. Int J Radiat Oncol Biol Phys 2001;49:79–91. Lundbom N, Brown RD III, Koenig SH, et al. Magnetic f ield dependence of 1/T1 of human brain tumors; correlations with histology. Invest Radiol 1990;25:1197–205. Cha S, Jonson G, Wadghir Y, et al. Dynamic contrast-enhanced MRI in mouse gliomas: cor relation with histopatholo gy. Magn Reson Med 2003;49:848–55. Englund E, Br un A, Larsson EM, et al. Tumours of the central nervous system. Proton magnetic resonance relaxation times T1 and T2 and histopatholo gic cor relates. Acta Radiol Diagn (Stockh) 1986;27:653–9. Tsien C, Gomez-Hassan D, Chenevert TL, et al. Predicting outcome of patients with high-g rade gliomas after radiotherap y using quantitative anal ysis of T1-weighted magnetic resonance imaging. Int J Radiat Oncol Biol Ph ys 2007;67:1476–83. Lupo JM, Cha S, Chang SM, Nelson SJ . Analysis of metabolic indices in re gions of abnor mal perfusion in patients with highgrade glioma. AJNR Am J Neuroradiol 2007;28:1455–61. Herholz K, Pietrzyk U , Voges J, et al. Cor relation of glucose consumption and tumor cell density in astroc ytomas: a stereotactic PET study. J Neurosurg 1993;79:853–8. Jager PL, Vaalburg W, Pruim J, et al. Radiolabeled amino acids: basic aspects and clinical applications in oncolo gy. J Nucl Med 2001;42:432–45. Langen KJ, Muhlensiepen H, Holschbach M, et al. Transport mechanisms of 3-[ 123I]iodo-alpha-methyl-L-tyrosine in a human glioma cell line: comparison with [3H]methyl]-L-methionine. J Nucl Med 2000;41:1250–5. Torii K, Tsuyuguchi N , Ka wabe J , et al. Cor relation of amino-acid uptake using methionine PET and histolo gical classif ications in various gliomas. Ann Nucl Med 2005;19:677–83.

54. Sato N, Suzuki M, K uwata N, et al. Ev aluation of the malignanc y of glioma using 11C-methionine positron emission tomo graphy and proliferating cell nuclear antigen staining. Neurosur g Rev 1999; 22:210–4. 55. Kracht L W, F riese M, Herholz K, et al. Meth yl-[11C]-lmethionine uptak e as measured b y positron emission tomo graphy correlates to microvessel density in patients with glioma. Eur J Nucl Med Mol Imaging 2003;30:868–73. 56. Rasey JS, Grierson JR, Wiens LW, et al. Validation of FLT uptake as a measure of thymidine kinase-1 activity in A549 carcinoma cells. J Nucl Med 2002;43:1210–7. 57. Shields AF, Grierson JR, Dohmen BM, et al. Imaging proliferation in vivo with [F-18]FLT and positron emission tomography. Nat Med 1998;4:1334–6. 58. Chen W, Delalo ye S, Silv erman DH, et al. Predicting treatment response of malignant gliomas to be vacizumab and irinotecan b y imaging proliferation with [ 18F] fluoroth ymidine positron emission tomography: a pilot study. J Clin Oncol 2007;25:4714–21. 59. Nuutinen J, Sonninen P, Lehikoinen P, et al. Radiotherap y treatment planning and long-term follow-up with [(11)C]methionine PET in patients with low-grade astrocytoma. Int J Radiat Oncol Biol Phys 2000;48:43–52. 60. Grosu AL, Weber WA, F ranz M, et al. Reir radiation of recur rent high-grade gliomas using amino acid PET (SPECT)/CT/MRI image fusion to deter mine g ross tumor v olume for stereotactic fractionated radiotherapy. Int J Radiat Oncol Biol Ph ys 2005;63: 511–9. 61. Grosu AL, Weber WA, Riedel E, et al. L-(meth yl-11C) methionine positron emission tomo graphy for tar get delineation in resected high-grade gliomas before radiotherap y. Int J Radiat Oncol Biol Phys 2005;63:64–74. 62. Lee L, Piert M, Gomez-Hassan D, et al. Association of 11C-Methionine PET uptake with site of failure after concurrent temozolomide and radiation for primar y gliob lastoma multifor me (accepted to Int J Rad Biol Phys 4/2/08). [In press] 63. Pöpperl G, Kreth FW , Mehrk ens JH. 18F-fluoro-L-thymidine and 11C-methylmethionine as markers of increased transport and proliferation in brain tumors. Eur J Nucl Med Mol Imaging 2007;34:1933–42.

52 PET DIAGNOSIS AND RESPONSE MONITORING IN ONCOLOGY RODNEY J. HICKS, MD, FRACP AND RICHARD L. WAHL, MD, FACR

Necessarily, a discussion of the role of positron emission tomography (PET) in oncology conf ined to a single chapter must be conceptual rather than comprehensi ve, philosophical rather than enc yclopedic. Consequentl y, the following re view of the use of PET and PET/Computed Tomography (CT) in the diagnosis and evaluation of cancer will focus on general principles and illustrati ve studies rather than to summarize the entire cur rent evidence base. Indeed, whatever might be considered an up-to-date synopsis would certainly be already outdated by the time of publication; such is the pace at w hich new studies are being published re garding ne w and e volving applications in oncology. Many excellent, and sometimes seminal, studies performed by researchers throughout the world will necessarily go unrecognized but the principles that they have elucidated are, we hope, reflected in this summary. For a recent summary of the collated literature, readers are encouraged to read the 2007 supplement of the Jour nal of Nuclear Medicine dealing with PET/CT 1 and man y other more recently published reviews that have appeared in almost all general and specialist imaging and oncology journals dealing with the application of PET to specif ic cancer types. The majority of this re view will focus on the principles under pinning the increasing use of the glucose analog, fluorine-18 fluoro-deo xyglucose (FDG), since this is the workhorse for most oncological PET and the tracer for which there is the strongest body of evidence. However, an e ver-increasing ar ray of PET tracers is being evaluated in oncolo gy. These offer the potential to provide unique insights into cancer biolo gy.2 Some of these new tracers will also be briefly discussed in the context of the role of PET pro viding new imaging biomarkers of pro gnosis and prediction of therapeutic response. It is impor tant to state at the outset that it is our view that PET is a complementary technique in the

imaging ar mamentarium and should be considered as an adjunct to other modalities in the e valuation of cancer and not as a competing technolo gy. Indeed , the development of hybrid imaging, as represented by combined PET/CT scanners, 3 has fused function and for m in the minds of oncolo gists in a manner akin to the image fusion that the y provide. The perception of cancer imaging will ne ver be the same. While anatomic imaging has been the historical standard for cancer staging and treatment response assessment, it is clear that the combination of anatom y and function elucidated b y PET/CT pro vides more infor mation than either method alone, in the v ast majority of cases.

WHAT IS THE ROLE OF IMAGING IN CANCER MANAGEMENT? Diagnosis, staging, treatment planning, therapeutic monitoring, and surveillance are the key processes involved in evaluation of appropriate management of patients with cancer. Imaging, because of its nonin vasive nature, plays a key role in all these phases of cancer e valuation and is currently highl y dependent upon anatomical imaging techniques, particularly including CT and magnetic resonance imaging (MRI). These techniques involve identifying, localizing, counting, and measuring the size of mass lesions or “lumps.” However, as discussed below, reliance on anatomy alone is fundamentall y at odds with moder n molecular medicine concepts. Being based primarily on size criteria, tumor inf iltration of normal structures such as lymph nodes can be difficult to detect on radiologic techniques. Similarly, benign processes ma y create v ery sizeab le lesions that ma y be mistaken for malignant tumors, or ma y coe xist with cancers, giving an incor rect impression of distant spread. 875

876

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Incidental reacti ve l ymph nodes, pulmonar y nodules, renal lesions, hepatic hemangiomas or c ysts, and adrenal adenomas masquerading as metastases are relatively common examples of this. Accordingly, biopsy or patholo gic examination of sur gical material has increasingl y been required for assessment of abnor mal anatomical imaging findings. We are often led to belie ve that such pathologic confirmation is the “gold standard” for deter mining the diagnostic performance of an imaging test. While it is true that patholo gy can conf irm f alse-negative results that diminish sensiti vity and f alse-positive results that reduce specif icity, the assumption must be that all lik ely and par ticularly, all suspected sites of metastatic disease can be, and have been, adequately sampled and examined under the microscope. Were biopsy consistentl y cor rect, no patient with pathologically complete resection and negative nodes would ever die of local failure. Unfortunately, however, such patients do quite commonl y relapse. Furthermore, the dif ferentiation of dysplastic lesions from frankly malignant lesions may be dependent on the for tuitous discovery of a focus of in vasion in a biopsy specimen since cellular mor phologic characteristics ma y be indistinguishable between these tw o entities. Ev en when the pathologic diagnosis is clear-cut, the natural history of many malignancies is highl y variable despite apparentl y similar histologic characteristics. The method of stage g rouping is also cur rently inconsistent with moder n thinking about the molecular biology of cancer and the range of a vailable therapeutic options. Ha ving characterized , often inaccuratel y, the extent of disease, the current process of stage assignment reflects an era w hen the onl y treatment of cancer that offered any hope of cure w as surgery. In moder n oncology, a plethora of therapies exists, and these therapies are seldom applied in isolation. Surgery is often preceded by chemotherapy (neoadjuvant) or followed by it (adjuvant), and radiotherapy is increasingly applied with concur rent chemotherapy. Chemotherapy has long in volved combination of multiple drugs. Current staging categories have been rendered e ven less rele vant to cur rent treatment options b y an increasing range of therapeutic agents, often reliant for efficacy on the expression of a particular molecular tar get. This has fur ther complicated management decisions beyond simple assessment of the extent of disease. A prototypical example of the limitations of the current paradigm is the staging of non-small-cell lung cancer (NSCLC). Despite genomic differences both between and within the v arious histopathologic subgroups, which influence their natural history (prognosis) and potentially their response to therap y,4 patients are fur ther g rouped

into clinical stages from I to IV based on a combination of the staging features of the primar y tumor (T -stage), regional l ymph nodes (N-stage), and distant metastatic sites (M-stage). Treatments are then selected and deli vered on the basis of this stage. Subject to comorbidities, patients with stage I to IIA disease are generall y treated surgically, whereas patients with stage IIB to IIIB disease are treated b y combined modalities, often including chemoradiotherapy, and patients with stage IV disease are generall y treated with chemotherap y or palliati ve radiotherapy. To understand how far such an approach is from being based on the biolo gic characteristics of these cancers, we need only to consider stage IIIB disease. This stage is represented by either T4N0M0 or T1N3M0 classifications using the TNM system. These different TNM classifications are lo gically g rouped in the cur rent staging system by being unsuitable for sur gery and having a low likelihood of cure. Ho wever, it is clear that the f irst of these two categories has demonstrated a propensity for local g rowth without a pre-dilection for metastatic spread, whereas the second has an agg ressive metastatic phenotype as e xemplified b y the presence of distant nodal disease at presentation despite a small and well-localized primary lesion. F rom f irst principles, one would reason that aggressive local therapy would be most appropriate in the for mer g roup of patients, w hereas systemic treatment capable of eradicating or reducing the likelihood of fur ther distant microscopic metastases would be important for the latter g roup. Anatomical g rouping ignores dif fering genomic characteristics that influence biolo gic behavior and provide new therapeutic targets. For example, availability of new biologic agents that target the epithelial growth factor receptor (EGFR), overexpressed in certain adenocarcinomas of the lung and predicti ve of response to these agents,5 emphasizes the need to personalize cancer care beyond only definition of disease extent. Although tissue genomics are lik ely to gi ve signif icant clues to major therapeutic tar gets in the future, these techniques are subject to the tissue that can be sampled patholo gically. A biopsy from one site is not necessarily reflective of all disease sites since genomic instability , microen vironmental f actors or selecti ve pressure related to therapies that have previously been delivered, can result in significant tumor heterogeneity within and betw een subpopulations of cancer cells that comprise indi vidual disease sites.6 In addition, the y ma y be subject to processing techniques. Indeed , estimates of 20% “inaccurac y” of testing for expression of Her-2 neu has been repor ted in a recent review.7 It is even more problematic when there are no obvious masses to biopsy or when the masses that

PET Diagnosis and Response Monitoring in Oncolo gy

can be identified are comprised of a combination of scar tissue, inflammatory cells, necrotic material, and viab le tumor follo wing therapeutic inter vention. Molecular imaging offers the potential to characterize the nature of tissues based on biochemical and biolo gic features and to guide biopsy to the sites most likely to yield representative tissue. Thus, the infor mation pro vided is fundamentally dif ferent to that pro vided b y anatomical imaging studies and complementar y to patholo gic and genomic characterization. Nevertheless, within cur rent management paradigms for most cancers, def inition of anatomic stage is a major prognostic f actor and deter mines most therapeutic decisions. Thus, the first step required to judge the clinical role of PET is to deter mine its clinical effectiveness for cancer evaluation by its ability to accuratel y def ine the presence and extent of disease compared with conventional imaging techniques. Erroneously in our opinion, this assumes that PET is competing with these modalities.

PET AS A CANCER DIAGNOSTIC AND STAGING TECHNIQUE As a glucose analog, FDG is used for the vast majority of clinical PET studies perfor med for the evaluation of suspected or conf irmed malignancy. This tracer is discussed in more detail in the chapter on PET radiochemistr y. The schema of transport and metabolism of this tracer in normal and cancer cells is no w w ell characterized. Unlik e glucose, FDG is onl y a substrate for the initial stages of glycolysis. These are an initial cellular uptak e through facilitated transpor t b y v arious glucose transpor ters (GLUTs) and phosphor ylation b y the initial enzymatic step in the gl ycolytic pathway performed by hexokinase. Thereafter, the monophosphate form of FDG is not a substrate for subsequent enzymes in the pathw ays followed by glucose. Through upregulation of the insulin-independent GLUT , GLUT-1, and he xokinase, enhanced FDG uptake and retention relati ve to nor mal tissues is a characteristic of malignant transfor mation. Enhanced gl ycolytic metabolism was recognized to be a key feature of cancer cells b y the Ger man Nobel Laureate, Otto Warburg in the earl y 20th centur y. This phenomenon has gained renewed interest in the early 21st century through detailed studies of cancer metabolism. 8 Indeed, most of 9 are the fundamental characteristics of cancer cells accompanied b y enhanced glucose metabolism. It has been recognized for many years that GLUTs are upre gulated in cancer cells, par ticularly in the presence of hypoxia.10,11 The impor tance of h ypoxia in re gulating cancer cell beha vior has been increasingl y reco gnized

877

over recent y ears.12 Other f actors in volved in the increased uptak e of FDG in cancer cells include oncogene overexpression and increased g rowth factor signaling.13 Thus, increased glucose metabolism inte grates the effects of disparate pathways involved in malignant transformation and tumorigenesis, the comple x process b y which a malignant cell for ms a primar y cancer and spreads to form distant metastases. Impaired mitochondrial function seems to be an impor tant f actor in man y cancers, and recent studies in lung cancer ha ve demonstrated linkage betw een FDG a vidity and mitochondrial changes leading to augmented aerobic gl ycolysis.14,15 Since FDG imaging was first used for the evaluation of suspected cancer in the 1980s, it w as clear that high uptake of this tracer w as a characteristic of man y malignancies.16–25 These data w ere suppor ted b y studies performed in x enograft models of v arious human cancers 26 and has helped to estab lish PET as a v aluable tool in translational research. Because of limited a vailability of PET at that time, se veral of these studies in volved heavily collimated gamma cameras and those that did not were of a highly restricted axial field of view (FOV) since whole-body screening capability had not y et been developed. Reflecting this technical limitation, e valuation of brain tumors w as an impor tant earl y focus of research since it w as easier to localize and encompass the site of suspected tumor in the FO V.27 However, early studies of tumors outside the brain demonstrated that a normal scan had a f airly high ne gative predicti ve v alue (NPV) provided that the lesion size w as sufficiently large to not be subject to partial volume effects. This principle of partial volume undersampling is discussed in the chapter on PET physics and instr umetation. Within the brain, a relationship between the intensity of uptake and tumor grade was recognized for the f irst time, and the influence of therapeutic intervention on FDG uptake was also reported.28–31 Even though it w as reco gnized earl y in the e xperience that FDG uptak e could occur in inflammator y processes,32 the positive predictive value (PPV) was also encouraging. Thus, despite the f act that an FDG-PET scan is actuall y an in vi vo biodistribution of glucose metabolism, the simplistic assumption that focal accumulation of FDG, not related to nor mal ph ysiologic processes, reflects malignanc y has been associated with quite acceptab le diagnostic accurac y in the majority of studies. Pro gressive impro vements in instr umentation, particularly the development of PET/CT, and in the processing of emission data have further enhanced the diagnostic performance of PET in answering the fundamental questions posed by physicians who suspect the diagnosis of cancer and desire conf irmation of the e xtent of

878

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

disease. Because of this continuing evolution, the clinical effectiveness of PET cannot be necessaril y accuratel y judged from the results obtained in earlier studies. This problem is fur ther e xacerbated b y lack of robust “gold standards.” As detailed abo ve, anatomic imaging techniques have significant limitations as a comparator while pathology is subject to sampling er rors, and some diseases are not easily characterized as benign or malignant on the basis of cellular features alone. Accordingly, the choice of comparator for assessment of diagnostic accuracy is particularly problematic. Not only is PET technology improving, there is also ongoing e volution of other imaging modalities, such as the de velopment of multidetector CT and higher f ield strength MRI using new pulse sequences and impro ved image anal ysis algorithms. These innovations, combined with impro ved patholo gic techniques and the potential for PET itself to influence the choice of validation methodology, emphasize the difficulties inherent in making do gmatic statements about the diagnostic performance of PET in cancer. For both diagnosis and staging of disease e xtent, the performance of PET , lik e that of an y diagnostic test, is generally e xpressed in ter ms of sensiti vity, specif icity, NPV, PPV, and accurac y. According to Ba yes’ theorem, both the NPV and PPV are highl y influenced b y the prevalence (pretest lik elihood) of disease in the test population. Accordingly, sensiti vity and specif icity are often considered to be the most appropriate parameters to describe the intrinsic diagnostic perfor mance of a test. Nevertheless, most clinicians who are faced with a given patient are actually more interested , w hether the y are aware of it or not, in the posttest lik elihood of disease when the scan is ne gative or positi ve in their indi vidual patient. Further, they expect a diagnostic test to appropriately guide their therapeutic choices. Although standards of evidence have been proposed to evaluate technical and diagnostic perfor mance of tests, 33 these are often impractical when the test is as expensive as PET and its putati ve v alidation gold standard car ries significant cost, potential morbidity and ma y itself suf fer from imperfect accurac y. Nevertheless, it must be reco gnized that the scientif ic methodology applied to establishing the diagnostic accuracy of FDG-PET in cancer has not always been ideal. Reflecting a vailable, pragmatic, and ethical subject selection, man y preliminary evaluations of the diagnostic perfor mance of FDG-PET in volved highly biased populations. By including patients with either v ery high or v ery low pretest lik elihoods of disease, both the PPV and NPV w ere lik ely to ha ve been enhanced. Conversely, the incremental diagnostic v alue compared with less accurate staging techniques is, however, likely to

have been underestimated in such studies since the patient group most likely to benefit from superior diagnostic accuracy is that with an inter mediate pretest likelihood of disease. Furthermore, validation bias has occur red since it is usually unethical to routinely biopsy sites that are negative on imaging. Despite these ca veats, there is no w abundant evidence that whether compared to pathology or to clinical follow-up including serial imaging, FDG-PET is more accurate, due to a v ariable degree of superior sensiti vity, specificity or both, than most con ventional imaging techniques for both the diagnosis and staging of cancer .34 Furthermore, the diagnostic perfor mance of molecular imaging with FDG has been fur ther enhanced with the advent of PET/CT.1 Nevertheless, validation of the diagnostic utility of a test needs to extend beyond diagnostic accuracy alone. In particular, e ven the concept of absolute sensiti vity of a diagnostic test for cancer detection is fla wed. Comparative diagnostic sensitivity is simply a function of the most sensitive technique available for validation purposes. Did PET suddenly become less sensitive when sentinel lymph node biopsy replaced clinical examination, CT scanning, and conventional nodal sampling for nodal basin staging of malignant melanoma? The ans wer is self-e vident. It did not; the comparator just got better. It is impossible to determine the true denominator for analysis of diagnostic sensitivity of tests used for cancer detection since a single rogue cancer cell ma y be all that is required to for m the lesion that e ventually leads to death. Gi ven that various techniques ha ve sho wn that indi vidual circulating cancer cells are common in patients diagnosed with various forms of malignancy35 and a significant proportion of women with breast cancer also ha ve cancer cells in their bone mar row,36 clearly an y imaging strate gy must be deemed to be highl y insensitive against such standards. However, it is no w clear that not all circulating cancer cells, or e ven those lodged in the bone mar row, are necessarily able to for m a metastasis and lead to illness or death of an individual.37 A battle must be waged between cancer cells and the immune system.38 Thus, it is perhaps better to detect onl y those sites at w hich the battle has clearly been w on by the cancer . But at w hat size does a metastatic lesion become capab le of impacting sur vival? This is an important but imponderable question when one considers the sensiti vity of PET . Similarly, the apparent specificity of a diagnostic test in cancer e valuation depends on the rigor with w hich disease is e xcluded. Anyone in volved in follo w-up of cases with putati ve false-positive FDG-PET results will be made w ell aware of the limitations of biopsy since man y such abnor malities turn out to relate to sampling er rors (Figure 1).

PET Diagnosis and Response Monitoring in Oncolo gy

Interim Response

Post-Treatment

Correlative Biopsy

879

6/12 Follow-up

Figure 1. Diffuse large-cell lymphoma is an aggressive malignancy with only a moderate likelihood of cure. A positive FDG-PET on interim assessment during chemotherapy carries an adverse prognosis. A scan done after 3 cycles of chemotherapy (projection image, lower left panel) demonstrated a partial metabolic response (PMR) compared to baseline (not shown) but residual left axillary abnormality (fused PET/CT, upper left panel). Treatment was intensified but a posttreatment FDG-PET/CT scan showed clear disease progression in the left axilla. An excisional lymph node biopsy was organized prior to bone marrow transplantation but yielded only a sclerotic node with no lymphoma. Unfortunately for the patient, the biopsy was considered more compelling than the positive PET scan by the referring clinician. The subsequent surveillance scan 6 months following treatment (right panel) demonstrated progressive systemic relapse involving the spleen and bone marrow, begging the question as to whether biopsy or outcome should be the “gold standard” for validation of diagnostic imaging tests.

Clearly, there is a need for alter native v alidation methods. What might offer an alternative? Given that the cancer is a disease that car ries a signif icant likelihood of a decrease in the quality and duration of life, ability to provide prognostic stratification is an important first step. Progression-free sur vival and , par ticularly, o verall survival are no w recognized to be impor tant parameters in evaluating the natural histor y of cancers and therapeutic efficacy. As discussed below, there is increasing evidence that FDG-PET provides superior prognostic stratification compared with conventional imaging techniques. Unfortunately, more accurate diagnosis of cancer ma y not necessarily impro ve sur vival of an indi vidual patient, particularly if no ef fective treatment is a vailable. Nevertheless, accurate staging is particularly important in those cancers for w hich no treatment has clearl y proven efficacy once the disease becomes disseminated since it may prevent futile attempts at locore gional cure, sparing the patient the financial and physiologic costs of this therapy. Furthermore, if better pro gnostic stratif ication of patient populations is possib le, this should allo w more rigorous assessment of therapeutic response and outcome using dif ferent e xperimental therapeutic approaches.

Accordingly, another parameter b y w hich the clinical utility of FDG-PET should be judged is its ability to guide management and improve patient outcomes. In this regard, the lack of randomized-controlled trials has been seen as a limitation.39 However, the need for randomizedcontrolled trials to pro ve that impro ved outcomes deri ve from superior diagnostic tests has recentl y been questioned,40 and the need for appropriate clinical benchmarks has been emphasized. 41 Through the combination of more accurate def inition of disease e xtent and better characterization of tumor biology, FDG-PET of fers the possibility of selecting specif ic patients lik ely to benef it from specific active interventions while potentially sparing other patients from unnecessar y treatment. However, the huge variation in respect to the biolo gical nature and burden of disease renders some patients entirely inappropriate for e valuation b y this technolo gy. F or e xample, whole-body PET is not suited to staging biolo gically indolent tumors or in situ carcinomas, w here there is a very lo w lik elihood of distant spread , or for staging widely metastatic disease for w hich only palliative intervention is being considered since detection of additional metastatic sites is unlikely to alter management.

880

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

CLINICAL INDICATIONS FOR FDG-PET The clinical role of PET in oncolo gy can be summarized using a prob lem-based approach that can be applied to any cancer that generall y has good a vidity for FDG. “Indication fragmentation, ” as applied in man y recent national health technology assessments, has complicated the evaluation of the clinical ef ficacy of FDG-PET and lacks a clinical orientation. A potential range of “prob lem-based” indications is as follows: 1. The noninvasive characterization of the likelihood of malignancy of mass lesions that are not readil y amenable to biopsy or for w hich biopsy attempts have already failed (evaluation of mass lesions). 2. The detection of cancer in patients at signif icantly increased risk of malignanc y on the basis of ele vated tumor mark ers, clinical symptoms and signs, or risk factors in w hom routine tests ha ve f ailed to detect a cancer (cancer detection). 3. Staging of high-risk malignanc y amenable to potentially curative therapy for which disease extent is critical to treatment selection (staging). 4. Planning of highl y tar geted therap y w here delineation of disease is critical to ef ficient treatment delivery and thereb y therapeutic success (treatment planning). 5. Assessment of therapeutic response in diseases with a significant likelihood of treatment failure for which earlier demonstration of therapeutic failure may benefit the patient (therapeutic monitoring). 6. Surveillance of high-risk malignancies or e valuation at clinical relapse w here salvage therapies e xist and for which early intervention may be curative or prolong life (restaging). For all these clinical scenarios, there are multiple independent e xamples of FDG-PET and , more recentl y, PET/CT being effective. Indeed, when directly compared, virtually all studies ha ve sho wn the superiority of PET/CT to PET alone. These will be detailed fur ther below. In addition to these broad g roupings, the use of FDG-PET tracers as predicti ve or prognostic biomarkers is an evolving and exciting application of this technology relevant to all the categories above. There is evidence that FDG-PET uptak e and its change during treatment provides robust individual patient and patient group prognostic stratification. This role will be further enhanced by development of additional, more specif ic oncolo gical PET tracers.

In each of the clinical scenarios, the superior accurac y of FDG-PET/CT is most lik ely to prevent futile attempts at cure b y detecting otherwise occult distant metastatic disease, reducing therapeutic costs and allo wing more rational allocation of scarce or e xpensive therapies. Thus, although the unit cost of a PET scan is relatively high compared with conventional e valuation techniques, its superior accurac y and impact on management decisions ha ve the potential to both reduce global cancer costs and improve outcomes.

FDG-PET FOR EVALUATING MASS LESIONS The f irst step in management of cancer is accurate and timely diagnosis. One of the first situations where FDGPET was evaluated as a noninvasive tool in this role was for the characterization of solitar y pulmonar y nodules (SPN). This represents the prototypical e xample of the use of FDG-PET for characterizing the lik elihood of malignancy in mass lesions that are not amenab le to biopsy, or which have def ied histopathologic characterization. Using a range of imaging de vices, studies performed in various countries around the world found that PET had a relatively high accuracy with the majority of FDG-avid lesions being malignant, and the v ast majority of non-a vid lesions being benign. 22,42–46 Most of these studies relied on detection of the lesion b y conventional imaging and e xclusion of those patients in whom the diagnosis of malignanc y w as made highl y likely by the features of malignant l ymphadenopathy or features of distant metastatic disease on con ventional imaging. Presumab ly, man y patients in w hom the CT features w ere characteristic of a benign process w ere also e xcluded. Ne vertheless, these selection criteria ensured a population that w as most lik ely to benef it from the superior diagnostic accurac y of PET because of an inter mediate pretest lik elihood of malignanc y. In many series, the pre valence of malignanc y w as in the range of 40 to 60%. The high accuracy of FDG-PET for the e valuation of SPN in most pub lished series from North America, Europe, and Australia and its adoption into clinical practice has not been mir rored by experience in regions of the world with a higher prevalence of infectious diseases that are associated with enhanced glucose metabolism. Such inflammator y lesions may mimic malignant lesions on both radiologic and FDGPET studies. Accordingly, they act to decrease the relative pre valence of malignant lesions in the tar get population and therefore, the PPV and apparent specificity of FDG-PET . Ne vertheless, since most of the

PET Diagnosis and Response Monitoring in Oncolo gy

processes that cause abnormal FDG accumulation in the lungs are disease processes that w arrant acti ve treatment, a specif ic patholo gic diagnosis, or at least v ery close and earl y re view, is required for positi ve PET results. On the other hand , negative PET results can be managed more conservatively. The difference in disease prevalence between different populations mandates that difference in further investigation and management paradigms should be applied. The influence of selecting a clinicall y appropriate population in which to validate the diagnostic performance of FDG-PET is illustrated by studies evaluating its role in the primar y diagnosis of o varian cancer in w omen with adnexal masses. Early studies had a low prevalence of disease47,48 and reported relatively low specif icity, related to a low PPV. A more recent study that had selected more appropriate clinical population, as documented b ya prevalence of disease of 64%, found a high PPV and diagnostic accuracy.49 Examples of mass lesions that ma y benef it from being fur ther characterized b y FDG-PET/CT include presumed sarcomas,50 wherein demonstration of probable high-grade malignancy may influence biopsy technique, and incidentally detected adrenal lesions. 51 However, any mass that is either difficult to biopsy in a minimally invasive manner or that has failed to yield adequate tissue for histopathologic diagnosis after se veral attempts w ould potentially fit this indication.

FDG-PET FOR CANCER DETECTION Although 1 in 3 indi viduals in w estern societies will develop cancer in their lifetime, the actual number harboring an occult cancer at an y given moment in time is relatively low. The yield from most screening programs is correspondingly lo w in absolute ter ms, e ven w hen restricted to groups deemed to be at increased risk on the basis of age or other clinical characteristics. For example, the rate of breast cancer detection in most mammography programs is generally less than 1% of the screened population when limited to w omen of 50 y ears or older and even less if w omen y ounger than 50 are included. 52 As expected b y the v ery lo w incidence of disease in the screened population, the PPV of suspicious results is also relatively low with as man y of 30 to 40 cases requiring biopsy to yield each tr ue case of malignancy. The use of PET has been adv ocated as par t of screening programs and has found greatest utilization in Japan. Although possib ly in volving indi viduals at increased risk of malignanc y related to selection of high-net-worth businessmen, one such pro gram w as

881

reported to have detected previously unidentified malignancies in up to 3% of the screened population using a combination of FDG-PET , CT, MRI, and a batter y of laboratory tests. 53 However, the number of f alse positives on PET was not recorded. In patients being e valuated for kno wn or suspected cancer b y FDG-PET or PET/CT, incidental conf irmed second malignancies have also been described to occur in 1 to 3% of patients, with colonic and thyroid malignancies appearing to predominate.54–58 In a recent study e valuating o ver 1,700 consecutive patients with FDG-PET/CT, the rate of confirmed, nonth yroidal malignancies w as 0.9%. 59 This detection rate was higher than for many other screening tests and almost cer tainly relates to the adv antage of PET not being limited to detection of cancer in a single organ. Encouragingly, the PPV, in those lesions considered to be highl y suspicious for a second malignanc y and of sufficient clinical concern to warrant further validation b y the managing clinician, w as just o ver 60%. However, a lar ge number of other incidentall y detected FDG abnormalities were correctly classif ied as benign, suggesting that e xperienced readers are impor tant to achieve good diagnostic accuracy. With increasing use of screening tests for cancer , including whole-body CT and various blood markers, there are a growing number of patients suspected to ha ve cancer but in whom conventional evaluation paradigms fail to confirm a diagnosis or yield false-positive results. The significant anxiety that such results can cause patients needs to be considered. If the rationale for perfor ming such screening is to detect cancer at an earlier and hopefull y more easil y curable stage, then use of a sensiti ve imaging technique is logical. The relatively high sensitivity of FDG-PET/CT will equate to an e xcellent NPV in patients with a lo w prevalence of disease, allo wing greater reassurance for patients with a negative scan. Conversely, given the relatively high specificity of FDG-PET/CT, active investigation of positive results is less lik ely to be a fr uitless exercise than for tests with a v ery lo w PPV. In summar y, FDG-PET/CT ma y allow identif ication of the subg roup of patients in w hom the combination of a suspicious tumor mark er and a positive PET may give a sufficiently high posttest likelihood of malignancy to w arrant fur ther histopatholo gic conf irmation of malignanc y, w hereas a ne gative result on FDGPET/CT could allo w a more conser vative obser vational strategy. Ev aluation of ele vated CA-19.9, CA-15.3, CA125 or carcinoembr yonic antigen (CEA) le vels are potential situations w here FDG-PET/CT might f ind a clinical role, but strong suppor ting evidence for the use of PET in this setting is cur rently lacking. P atients suspected to be suffering from a paraneoplastic syndrome but without an

882

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

obvious primar y ha ve been successfull y e valuated b y PET.60 Similarly, some rationale ma y exist for a screening PET study in patients with a high genetic pre-disposition to cancer. It should be noted that higher sensitivity, organ specific, PET scanners might mak e a g reater contribution to lesion detection than w as feasib le with w hole-body PET alone. As an example, in selected patients, a dedicated PET breast imaging scanner sho wed high sensiti vity for lesion detection.61–63 This scanner places a detector ar ray directly around the breast to impro ve sensiti vity and resolution. Thus, as PET technology continues to improve, the role of PET may expand.

FDG-PET FOR STAGING OF CANCER Once cancer has been diagnosed, the next critical step is to determine the e xtent of disease. This process of staging often involves significant financial resources and time but is vital to cor rect management choice. The goal is for long-term sur vival, if not free of disease, at least free of progression and the symptoms that accompan y malignancy. If the cancer cannot be cured by currently available techniques, then palliative treatments that maximize quality of life should be used, avoiding the cost and morbidity of futile curative attempts. Therefore, the key performance criteria of a diagnostic test for cancer are not, w e believe, its diagnostic perfor mance alone but rather its ability to appropriately change management and to better stratify prognosis. The ability of PET to detect more lesions than CT was demonstrated soon after lar ge FOV PET scanners w ere introduced. Small pilot studies in breast cancer ,24 lymphoma,64 and lung cancer65 staging showed PET to be more sensiti ve and , in lung cancer , more accurate than staging done b y conventional diagnostic methods alone. Although these studies w ere comparisons to either conventional diagnostic imaging or to patholo gy (with its attendant limitations), they were encouraging in that PET could detect more disease than the anatomic methods available at that time. A lo gical e xpectation of superior staging accuracy would be differences in patient outcomes or changes in management. Soon after w hole-body PET became a vailable, seminal studies perfor med b y P eter Valk and colleagues 66 demonstrated that FDG-PET could have a major impact not only on the apparent stage of disease but also on the management decisions made by referring clinicians. Across a range of indications, these retrospective studies demonstrated that patient management was altered in a substantial number of cases as a consequence of stage mig ration. Where biopsy or clinical follow-up was able to ascertain the appropriateness of the

resulting stage migration, FDG-PET was also shown to be correct an overwhelming percentage of the time compared with con ventional in vestigational paradigms and to be cost-effective using medical costs of the United States. When the clinical PET facility was established at the Peter MacCallum Cancer Centre in 1996, a prospecti ve data collection process was established that required all referring clinicians to designate both the disease status based on all a vailable infor mation available up to the time that PET was being requested and what their management plan would be if PET w ere not a vailable. P articularly, in the early years of this evaluation process, conventional evaluation almost always included clinical examination, laboratory tests, and various combinations of imaging tests, with CT or MRI being almost uni versal. On the basis of this information, a pre-PET stage and management plan w ere prospectively assigned. These w ere then compared with post-PET stage and the management actuall y deli vered after PET. Using standardized criteria, the impact of the PET result on management was then assessed. The impact of PET was graded as follows: “high” when, as a result of the PET f indings, there w as a change in management intent or modality, eg, curative to palliati ve or sur gery to medical therapy; “medium” if there w as change in deli very of treatment but not intent or modality, eg, a change in radiation treatment v olume (Figure 2); “low” when management planned w as still deemed appropriate; and “no” if the management planned seemed inappropriate b ut treatment was not altered, ie, PET was ignored. The validity of discordant PET and con ventional imaging f indings that altered stage or management w as e valuated b y histopathologic e valuation, if a vailable, or b y clinical follow-up o ver an appropriatel y long inter val to assess both the accurac y of staging and the pro gnostic signif icance of pre-PET versus post-PET stage. This methodology was first used to report the impact of FDG-PET in a prospecti ve cohor t of patients with known or suspected lung cancer at v arious phases of the diagnostic process including primar y staging. 67 Similar studies w ere subsequentl y repor ted on the impact of FDG-PET for primar y staging of e xpanded cohor ts of patients with pro ven NSCLC, 68 NSCLC patients being considered for radical radiotherap y,69 rectal cancer ,70 esophageal cancer ,71 and cer vical cancer .72 In all these cohorts, FDG-PET not onl y changed the stage in a significant proportion of patients but also impacted on treatment decisions in a sizeable proportion of patients. When FDG-PET results w ere discrepant with con ventional imaging, they were validated to be correct in a large proportion of patients in w hom the e xtent of disease w as evaluable by histopathology or serial imaging. Long-term

PET Diagnosis and Response Monitoring in Oncolo gy

883

Planning CT

Fused FDG PET/CT

Pre-Treatment

Mid-Treatment

Figure 2. Accurate delineation of treatment volumes is important to the success of radiotherapy. Too small a volume risks local treatment failure while too large a volume may compromise ongoing quality of life through additional radiation damage to normal tissues that provides no greater chance of cure. Use of FDG-PET/CT in radiation treatment planning is increasing and has particular utility in defining the primary tumor extent in association with lung collapse. In this case of non-small-cell lung cancer, both the extent and location of the primary is masked by extensive lung collapse (left upper panel) leading to a very large treatment volume based on CT planning. However, fused PET/CT clearly demonstrated a smaller central primary. The extent and location of the primary was confirmed by follow-up PET/CT performed during treatment and following partial re-expansion of collapsed lung (right panels). Note that both studies were performed on a flat palette as used for radiation treatment delivery. Patient positioning using lasers allows careful setup of the patient in treatment position and allows image data to be transferred into the planning computer.

follow-up has also demonstrated that PET provides superior prognostic stratification to conventional imaging. This methodolo gy has been adopted for a national data collection process in Australia and is similar to that adopted in the United States for the National Oncolo gic PET Re gistry.73 Although it is impor tant to demonstrate that PET changes stage and management, v alidation that the management changes documented were appropriate is an impor tant aspect of such research. Although lo gistically difficult to do outcome anal ysis on a national scale, substudies with appropriate length of follo w-up to compare pro gnostic stratif ication b y FDG-PET/CT with pre-PET stage from conventional evaluation will enhance confidence in the use of molecular imaging in staging of cancer. Resource utilization studies will also be important since it is probable that detection of occult metastatic disease will decrease futile and e xpensive attempts at cure.

FDG-PET FOR TREATMENT PLANNING Recent innovations in radiotherap y have seen a mo ve to more highl y tar geted treatment methods such as three-dimensional confor mal radiation therap y

(3D-CRT) or dose escalation to subre gions within tumors (dose painting). The development of adv anced treatment deli very systems lik e intensity-modulated radiotherapy (IMR T) allo ws use of higher radiation doses to tumor w hile sparing adjacent nor mal tissues. However, such tight tumor margins mean that it is even more important that the gross tumor volume (GTV) be accurately and precisel y def ined. Without this accuracy, “geo graphic misses” ma y occur . The ability of anatomical imaging to def ine the relationship of the primary tumor to k ey str uctures can be compromised by mechanical ef fects on sur rounding tissues, such as those associated with collapse of lung parench yma beyond an obstr ucting bronchial lesion. This can lead to a GTV that is signif icantly lar ger than the actual tumor. Alternatively, similarities in the imaging characteristics of the primar y and adjacent nor mal tissue may lead to an underestimate of the tr ue GTV. An example of this is when esophageal carcinoma extends over a g reater cranio-caudal e xtent than suggested b y CT mural thick ening. Similarl y, in volvement of normal str uctures, lik e l ymph nodes, ma y not necessarily be associated with mor phologic changes w hile

884

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

reactive l ymphadenopathy ma y cause nonmalignant enlargement of lymph nodes. Accordingly, delineation of the e xtent of local nodal in volvement requiring treatment can be prob lematic but is vitall y impor tant when planning radiation treatment volumes. Due to the contrast af forded b y dif ferential FDG uptake in cancer cells, while still providing the anatomical landmarks and attenuation characteristics required for radiotherapy dose planning and deli very, PET/CT of fers the potential for impro ved dif ferentiation of malignant from benign tissues and, therefore, better def inition of the GTV.74 This is par ticularly important in patients under going radiotherapy since many such patients are not suitab le for sur gery and therefore lack comprehensi ve patholo gic staging.75 As a result, FDG-PET/CT is pla ying an increasing role in radiotherapy treatment planning. Prior to widespread a vailability of PET/CT , se veral studies using coregistration of FDG-PET with con ventional CT-based radiotherapy treatment planning were reported. These most commonly involved patients with lung cancer and sho wed that the use of PET simulation enhances the accurac y of target volume coverage,76–78 particularly when complicated by bronchial obstr uction, but also b y guiding appropriate inclusion or e xclusion of in volved and unin volved lymph nodes in the planned treatment volume.79 The adv antage of FDG-PET/CT in radiotherap y has subsequently been demonstrated in esophageal cancer ,80,81 where the axial extent and remote nodal disease pose problems to GTV planning. Preliminar y results suggest that PET/CT may be also helpful in planning IMR T f ields for para-aortic lymph node involvement in cervical cancer even when these nodes are not enlarged.82–84 Due to proximity to the kidneys, spinal cord, and gut, all of which limit conventional radiation dose delivery, IMRT is an attracti ve option for salvaging patients with para-aor tic nodal in volvement on PET since this is clearl y an adverse prognostic factor.85 The ability of FDG-PET to demonstrate nonenlarged nodes in patients with rectal cancer70,86 suggests that PET/CT will also be suitab le for planning neoadjuv ant chemoradiation, which is no w a routine component of multidisciplinar y management of locall y adv anced rectal cancer . Similarl y, FDG-PET/CT has a strong rational basis in planning radiotherapy of locally advanced head and neck cancer.87

FDG-PET FOR THERAPEUTIC MONITORING For both localized and widel y disseminated cancers, a growing array of therapeutic agents is becoming available, with an increasing focus on molecular -targeted therapies.

Many of these agents are added to conventional therapies, adding cost and potential ne w toxicities to e xisting treatment paradigms but offer the hope of impro ving survival and the quality and duration of life of those who cannot be y cured.88 A personalized use of multimodality therap requires innovative selection, planning, delivery, and, particularly, impro ved therapeutic monitoring. Earl y and robust identification of nonresponders is needed to f acilitate earlier termination of ineffective treatment allowing a change to alternative treatments that may be more efficacious, or a voidance of futile side ef fects that diminish physiological reserves, and compromise quality of life and patients’ capacity to withstand subsequent treatments. Therapeutic response assessment also plays a critical role in determining the efficacy of a cancer therapy in populations of patients, aiding establishment of treatment guidelines and guiding allocation of scarce health care resources. The latter role relies particularly on the demonstration that imaging results can act as sur rogates for measures of sur vival. There is a strong rational basis for the use of molecular imaging and par ticularly FDGPET/CT for response assessment and oncolo gic dr ug development.89,90 Although tumor mark ers are commonl y used to assess treatment response, the y are not al ways available and provide no localizing v alue. Therefore, the y cannot be used to guide salv age therapies lik e surgery or radiotherapy in the event of persisting abnor mality. Therapeutic response assessment has traditionall y been, and generally remains, primarily based on the changes in the measured dimensions of lesions identif ied on CT , and , less frequentl y, on ultrasound , x-ra y, or MRI. These changes are recorded and g raded to using def initions detailed in the Response Ev aluation Criteria in Solid Tumors (RECIST), 91 representing a modif ication of earlier World Health Or ganization (WHO) response criteria.92 Both the RECIST and WHO systems classify response based on the change in the dimensions of tar get lesions and ha ve classif ications of complete response (CR), partial response (PR), stable disease (SD) and progressive disease (PD). Both these systems ha ve their roots in a preliminar y e xperimental study that in volved simulated measurement of lesions, using wooden spheres of v arying size under a foam mattress, b y e xperienced clinicians using palpation and various measuring devices, including calipers.93 There never was a biologic basis for these response g roupings, only considerations re garding statistical reproducibility. Unfortunately, long experience has taught us the limitations of anatomic imaging in assessing response, 94 not only in adults but also in children. 95 It is well known that

PET Diagnosis and Response Monitoring in Oncolo gy

changes in lesion size are relati vely slo w to occur and may be limited b y f ibrotic healing. This ma y lead to unnecessary prolongation of treatment or even institution of more agg ressive treatment in the mistak en belief that there has been a poor response to earlier inter ventions. Conversely, structures such as lymph nodes that return to normal size may still harbor disease. Ne vertheless, these techniques are widel y accepted b y the medical community, funding bodies, and re gulators simply because the y are easy to standardize and ha ve, in general, been sho wn to correlate somewhat with survival. One of the major theoretical adv antages of PET compared with str uctural imaging techniques is that there is usuall y a more rapid decline in tumor metabolism than in tumor size for responding tumors. Preliminary studies, repor ted more than 15 years ago, demonstrated that reduced FDG uptak e in breast cancer both preceded and predicted subsequent mor phologic response to combined modality therap y.96 Since then numerous other studies have demonstrated that reduction in FDG uptak e cor relates with subsequent clinical and radiological response. This has led to recommendations for wider use of FDG-PET in therapeutic response assessment and attempts to codify response cate gories based on FDG metabolic response. One such attempt to achieve a consensus position in volved the European Organization for Research and Treatment of Cancer (EORTC), w hich subsequentl y promulgated guidelines for the methodolo gy of perfor ming serial FDG-PET evaluations and repor ting metabolic response. 97 These recommendations ha ve recentl y been augmented b y a consensus statement from the National Institutes of Health in the United States.98 A reduction in FDG uptake in lesions is usuall y seen in responding lesions. One of the most dramatic examples of this is the rapid reduction of FDG uptake in gastrointestinal stromal tumors (GIST) following treatment with an agent that b locks the c-kit oncogene product. 99 Metabolic responses can be apparent within 24 hours of commencing treatment. However, since such tumors ma y become pro gressively more hypodense on CT with treatment, the y may even appear to enlar ge with treatment. Con ventional radiolo gic response criteria ha ve been sho wn to be poorl y predictive of sur vival in this setting. 100 In some situations, a transient increase or “metabolic flare” ma y also be predictive of subsequent clinical benef it. This has been described in the setting of introduction of tamo xifen in metastatic breast cancer.101 Although there is g rowing enthusiasm that PET can provide early therapeutic response assessment, the preferred methodology for metabolic response assessment remains

885

controversial. Methods v ary in comple xity from simple visual comparison of baseline and posttreatment scans to complex computational approaches in volving dynamic imaging and serial ar terial b lood sampling. Quantitati ve, semiquantitative, and qualitative assessments each have both advantages and limitations. The usual acquisition protocol for oncolo gic imaging in volves a dela yed, e xtended FOV imaging study. Indeed, “whole-body” PET is becoming the standard of care for cancer staging for the reasons detailed above. This scan is consequentl y able to provide a baseline for therapeutic response assessment using either qualitati ve or semiquantitative analysis. Dynamic imaging, on the other hand, requires prospective determination of the appropriate imaging FOV and has a limited axial sur vey e xtent. This must, therefore, usuall y be follo wed b y a dela yed w holebody scan. Such acquisition protocols are often quite impractical for routine clinical use in busy clinics. Quantitative analysis relies on mathematical modeling and , consequently, makes assumptions that may or may not be correct. An understanding of tumor biology, the mechanisms of treatment effects, and the likelihood for cure of a given cancer with a gi ven therap y should under pin a rational discussion of the role of FDG-PET in therapeutic monitoring. The most impor tant objective of response assessment is reliab le stratif ication of pro gnosis and appropriate guidance of fur ther treatment requirements. The term “metabolic response” is now being widely used to denote the de gree of change in FDG uptak e in tar get lesions. The simplest method of e valuating metabolic response is visual anal ysis, but its subjecti vity has been seen as a limitation. To overcome this, there needs to be attention to detail with respect to achie ving a consistent display of images. This involves normalization to appropriate nor mal reference tissues and use of consistent image display scales. It is also impor tant to use a standardized nomenclature for qualitati ve reporting of serial FDG-PET scans that can be applied to all tumor types and can be consistentl y applied b y different individuals and institutions. In the schema described b y Mac Manus and colleagues,102 a complete metabolic response (CMR) is defined as a return of FDG uptake in previously documented lesions to a level equivalent to, or less than, residual radioacti vity in nor mal tissues within the or gan in question. A partial metabolic response (PMR) constitutes a signif icant visual reduction in FDG uptak e in tumor sites based on visual inspection of appropriatel y displayed comparative images. Stab le metabolic disease and pro gressive metabolic disease (PMD) are def ined respectively b y a lack of change, or b y an increase in either of metabolic abnor mality in a patter n consistent with tumor g rowth, or de velopment of ne w sites of

886

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

disease. F or those cate gories that in volve a qualitati ve change in the intensity of uptake, such as PMR, measurement of tracer uptak e could be useful to v alidate the qualitative impression. The semiquantitative parameter that is cur rently preferred to assess the change in FDG uptak e in tumors is the standardized uptake value (SUV). The SUV is a number that describes the radioactivity per unit volume of tissues as a function of the administered dose and radioactive decay. If all the administered acti vity were to be unifor mly distributed in the v olume of interest, the SUV at all time points would be 1.0. However, because of different degrees of uptak e and trapping of FDG in differing tissues, the SUV v aries considerably between various normal tissues and between most normal tissues and malignant lesions. There is not y et consensus re garding what degree of FDG signal reduction should constitute a partial or CMR. Like the RECIST and WHO systems, the EORTC schema is based primarily on the reproducibility of the measure of the SUV rather than an y biologically relevant end point. Fur ther, it must be cautioned that the reproducibility assumed in the EORTC criteria, based on papers showing 10 to 15% reproducibility,103,104 is probably optimistic in the conte xt of routine clinical practice unless strict attention to the consistenc y of scan acquisition and processing is adhered to. Although there is a strong rationale for adopting a standardized approach for PET def inition of therapeutic response cate gories, it needs to be recognized that uptake and retention of molecular tracers is a biological process. As such, it is subject to the mechanism of dr ug action and the cellular consequences of this. Although the objecti ve of cancer treatment is to kill all malignant cells, treatment ma y modulate tumor bioener getics and thereb y alter FDG uptake prior to, and sometimes independent of, cell killing. It is also apparent that technical and ph ysiological factors influence measures of tracer uptake in tissues, and therefore, it is critical that methodological variability is minimized betw een studies in a gi ven patient if proportional reduction in FDG is to be used for response assessment.98 It should also be apparent that definition of a single criterion of metabolic response is lik ely to be fraught with difficulty. Despite reser vations about the scientif ic validity of qualitative analysis, multiple studies that ha ve used this methodology ha ve demonstrated its ability to stratify prognosis based on broad cate gories of metabolic response. Indeed, most studies evaluating the use of PET in l ymphoma ha ve used visual anal ysis to dichotomize responders into complete and incomplete metabolic responses. This methodolo gy w as adopted as the most

appropriate standard for this role in a recent consensus statement on the use of FDG-PET for response e valuation,105 based on its ability to po werfully stratify patient outcome (at least when retrospectively applied to clinical data). Qualitati ve anal ysis of FDG-PET to assess response of solid tumors to treatment has been used in multiple studies performed at the Peter MacCallum Cancer Centre106–111 and has demonstrated that PET can provide statisticall y signif icant pro gnostic stratif ication, particularly w hen patients are dichotomized betw een CMR and non-CMR groups. The frequency and prognostic v alue of a CMR are lik ely to be influenced b y the responsiveness of the tumor to treatment, the biolo gical aggressiveness of the disease process, and the timing of the follow-up scan after treatment. Although patients with a CMR will not necessaril y be cured, it is lik ely that the majority of those patients w ho do achie ve a durab le remission of cancer will come from this g roup of patients. Ba yesian principles are lik ely to influence the proportion of patients with a CMR w ho are cured. F or diseases that ha ve low cure rates with cur rent therapies, such as breast cancer, a CMR is unlikely to indicate cure but may provide a useful measure of progression-free survival.110 For diseases in which the majority of patients are cured, such as Hodgkin lymphoma, a CMR is much more likely to correlate with cure. 112 While a CMR and PMD are lik ely to be f airly consistently inter preted between individual repor ting physicians and betw een institutions, the methodolo gy used to define a PMR is less clearl y def ined at this time. As opposed to l ymphoma, solid tumors rarel y respond rapidly to treatment b y depopulation of viab le cells. Therefore, PMRs ha ve predominated in “responders” within most FDG-PET therapeutic monitoring trials, particularly those involving chemotherapy. Where abnormal radiotracer uptake remains in a lesion, deter mination of the degree to which it has reduced ma y have therapeutic and prognostic implications. In such cases, measurement of lesion radiotracer uptak e may provide more objecti ve evaluation than qualitative assessment. Various investigators have generally divided patients into responders and nonresponders based on subsequent radiolo gic response or pathologic response and then used post hoc determination of the percentage reduction in FDG uptake in malignant tissues that provided the best predictive accuracy of FDG-PET for subsequent therapeutic response. Such thresholds are unlik ely to be consistent for dif ferent cancers, dif ferent therapies, or dif ferent time points of response evaluation. Nevertheless, as with the v alidation of the utility of FDG-PET for staging, the strongest evidence for the utility of FDG-PET will come from

PET Diagnosis and Response Monitoring in Oncolo gy

increasing data demonstrating that metabolic response, however def ined, better stratif ies sur vival than con ventional RECIST e valuation. There is no w increasing e vidence for semiquantitati ve measures of FDG-PET response assessment in v arious solid malignancies. 113 Importantly, PET is not limited to “measurab le” lesions in the manner that CT or MRI are. Issues that still need to be resolv ed with respect to the use of PET in therapeutic response assessment relate to the timing of the scan in relationship to the therapeutic intervention. On the basis of f irst principles, it is lik ely that if the scan is perfor med within hours or da ys of the therapy, its FDG uptak e will primaril y reflect acute effects on tissue ph ysiology and cellular metabolism, whereas if perfor med later, after these acute ef fects have subsided, it will primaril y reflect the number of residual viable cells. In most series, chemotherapeutic response, as deter mined by FDG-PET, tends to occur earlier than with radiotherap y. Accordingly, response assessment is more commonl y perfor med after 1 to 3 c ycles of chemotherapy, whereas assessment of response to radiotherapy is usually performed several weeks after completion of treatment. Trials are, ho wever, ongoing to determine the optimum time for assessing response of various tumor types to differing treatment modalities.

FDG-PET FOR RESTAGING Following def initive treatment of cancer , ongoing symptoms, residual str uctural imaging abnor malities or elevated tumor mark ers are not uncommon. There are also many patients at signif icant risk of relapse e ven in the absence of an y objecti ve e vidence of residual disease. Investigations used for the routine sur veillance of highrisk patients, or to assess symptoms that ma y present months or y ears after the completion of primar y treatment, include CT scanning and abdominal ultrasound , measurement of tumor mark ers per tinent to the type of cancer, and , w here appropriate, endoscop y. Results of these investigations may lead to other more invasive procedures including sur gical inter vention such as laparoscopy. In the absence of pathologic confirmation of disease, equi vocal f indings on an y of these tests often lead to serial imaging studies. The results of such investigations deter mine w hether fur ther acti ve treatment is required. The limitations of str uctural imaging that bedevil interpretation of staging scans are further exacerbated b y posttreatment changes, including de velopment of scar tissue and alteration of nor mal anatomy by surgical resection or reconstr uctive surgery. In the setting of posttreatment recurrence, malignant deposits may coexist

887

with scar tissue, fur ther increasing the lik elihood of sampling error on biopsy (F igure 3). Without pathologic confirmation of disease, equi vocal in vestigation results can be a considerab le source of anxiety for the patient, but acti ve treatment is clearl y unjustif ied if residual masses represent onl y scar tissue. Ho wever, waiting for clear evidence of disease progression may lessen the likelihood of successful salv age therapies b y allowing time for systemic spread to occur. Accordingly, in some cases empirical treatment is undertaken. Clinically, it is important not onl y to detect residual or recur rent cancer but also to determine whether it is suitable for salvage therapies. Even metastatic disease can be resected with curative intent if limited in extent. This has an established and expanding role in the treatment of hepatic metastases from colorectal cancer.114 Patients with colorectal cancer may also benefit from resection of limited metastatic disease at other sites including the lung 115 and o varies.116 Metastectomy of lung nodules is also perfor med in melanoma117 and sarcoma.118 In addition to major hepatic resection, hepatic metastases ma y also be treated b y ablative techniques including radiofrequenc y ab lation, ethanol injection, radioacti ve microspheres, and cryosurgery.119 Improved treatment selection and planning could be facilitated by more accurate detection of residual disease and better def inition of its e xtent. Being based on the metabolic characteristics of tissues, FDG-PET should be less susceptible to the effects of prior treatment and certainly, because mature scar tissue is metabolicall y inactive w hile most recur rent cancers are FDG-a vid, FDG-PET should ha ve a relati vely high NPV. Posttreatment FDG-PET has, for e xample, been demonstrated to have powerful prognostic value in the posttreatment setting of lymphoma.105 Since many recurrent solid malignancies are detected w ell after initial treatment, there is also an increased likelihood of additional occult metastatic disease having developed. Thus, the potential for discrepant disease status e valuation betw een con ventional imaging and FDG-PET is probab ly higher in the restaging setting than in the primar y staging domain. Man y studies throughout the w orld ha ve demonstrated that FDG-PET is more accurate than con ventional imaging for detection of residual cancer follo wing def initive treatment of v arious nonhematolo gic malignancies. Using the management impact methodolo gy described above, substantial changes in disease status were demonstrated using stand-alone FDG-PET at the P eter MacCallum Cancer Centre in patients with suspected or proven recur rence of NSCLC, 120 small-cell lung cancer,121 colon cancer,122 and head and neck squamous cell

888

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

Figure 3. Following neoadjuvant chemoradiotherapy and surgery for locally advanced rectal cancer, persisting pre-sacral masses are common and compromise the positive-predictive value of CT. The ability to guide biopsy to sites of metabolic abnormality and to identify scar tissue by virtue of its low metabolic activity potentially allows earlier confirmation of local recurrence. In this case, focal uptake in the right side of the pre-sacral mass as demonstrated by the cross-hairs on the fused transaxial, coronal, and sagittal images allowed greater confidence in the nature and location of focal uptake in the right side of the pelvis as seen on the projection image (lower right panel). Biopsy confirmed local recurrence. The ability of FDG-PET to exclude wider metastatic spread has additional implications for the choice of treatment.

carcinoma.123 These results ha ve subsequentl y been extended to FDG-PET/CT in head and neck cancer 111 and ovarian carcinoma.124 In these studies, PET findings were again overwhelmingly validated by clinical followup. In most of these studies, FDG-PET also demonstrated superior pro gnostic capability compared with conventional evaluation. The par ticular adv antage of FDG-PET/CT in the restaging setting is its ability to more def initely localize abnormal foci of uptak e, par ticularly in str uctures that may ha ve altered anatom y due to mechanical f actors related to previous surgery or radiotherapy. An advantage of PET/CT o ver stand-alone PET has been clearl y demonstrated for colorectal cancer 125,126 and is lik ely to be true of many other cancers including NSCLC. 127 Furthermore, current generation combined PET/CT scanners

allow metabolically-guided biopsy to minimize sampling errors when there is need for patholo gic conf irmation of residual disease. 128 Thus, the accurac y, management impact, and pro gnostic stratif ication v alue of FDGPET/CT are lik ely to be e ven g reater than those pre viously demonstrated for FDG-PET.

PET IMAGING BIOMARKERS As detailed above, the ability of FDG-PET and PET/CT to detect cancer deposits for diagnosis and staging is impressive and clinically important because TNM staging is currently the major pro gnostic f actor in most cancers. Similarly, the utility of FDG-PET for treatment planning, therapeutic monitoring, and restaging is superior to conventional imaging because it better characterizes the

PET Diagnosis and Response Monitoring in Oncolo gy

presence and e xtent of disease. Ho wever, to limit the discussion of PET to the definition of the volume of tumor ignores its potential role in characterizing the biolo gic nature of a cancer . There is increasing reco gnition of the capability of PET-derived parameters to act as biomarkers.129 Biomarkers may be prognostic, predictive, or both. A prognostic biomarker is one that predicts outcome ir respective of the treatment given. A predictive biomarker is one that prospectively identifies that a particular outcome is more likely when a given treatment is administered than when it is not. Some biomark ers may be both predicti ve and prognostic.130 Most current biomarkers are related to serum factors or tissue characteristics that ha ve no localizing value and many have imperfect sensitivity and specificity.131 By imaging tissue tar gets directl y or the downstream consequences of their modulation, PET potentially offers a complementar y role in selecting and monitoring the use of molecular -targeted therapeutics.13,132 Any discussion of the role of PET in molecular imaging needs to consider not onl y its diagnostic performance against conventional imaging benchmarks but also its capacity to deliver novel biomarkers. Nevertheless, the majority of studies e valuating PET ha ve focused almost entirely on the former. When w e consider the diagnostic perfor mance of PET, it is often dichotomized into cate gories of either a positive or a ne gative result. Ho wever, in reality , FDG uptake represents a continuous v ariable. In the earliest studies of PET in oncology, it was noted that the intensity of FDG uptak e w as influenced b y tumor g rade. Semiquantitative measures of FDG uptak e including the SUV have been recognized as a marker of tumor biology. Studies by Di Chiro and colleagues 27 demonstrated that highgrade gliomas had higher FDG retention than lo w-grade astrocytomas. As a corollary of this, a lower FDG uptake was associated with longer survival.28,31 Given the differing histopatholo gic characteristics of brain tumors and their v ery dif ferent known natural histories, such information could be considered of limited v alue since this prognostic infor mation w as lik ely to be a vailable more directly from histopatholo gy. Ho wever, o ver time there have been an increasing number of studies attesting to the ability of a measure as simple as the SUV to stratify prognosis above and beyond conventional prognostic factors. An e xample of this w as a study of the use of PET in esophageal cancer where the intensity of FDG uptak e on PET provided better prognostic stratification than any of the cur rently used patholo gic and imaging criteria. 133 Similar results have been obtained with many other solid malignancies, including sarcoma, 134 which are characterized by great variability in both histopatholo gic features

889

but also in natural histor y. Impor tantly, man y of the cancers for which PET has been shown to have relatively low sensiti vity, as a result of relati vely reduced FDGavidity, ha ve included cancers that also tend to ha ve a more indolent natural histor y compared with other histopathologic subtypes with w hich the y are generall y categorized. Examples include lobular carcinoma of the breast,135 bronchioloalveolar carcinoma of the lung, 136 carcinoma of the prostate, 137 and indolent l ymphoma.138 If the goal of PET is to f ind all cancer sites, then FDGPET might be considered a f ailure in some of these cancers. However, if the goal is to predict the course of the disease, the answer might be very different. Despite its ability to stratify prognosis, it is important to recognize that the SUV is a simplistic, but impor tant, downstream measure of a complex process and reflects a biologic continuum related to tissue glucose metabolism, which within tumor g roupings appears to ha ve a lo gnormal distribution. 139 Nevertheless, the SUV has been invested with almost mythical status within some sections of the nuclear medicine community with attempts to codify thresholds for dif ferentiating benign from malignant lesions or the percentage change in this parameter that is reflective of a f avorable response to treatment. Such thresholds ha ve usuall y been deter mined b y post hoc determination of the cutof f value that provided the highest accurac y in dichotomizing lesions into benign and malignant g roupings or patients into g roups based on some other measure of response. Studies ha ve f ailed to demonstrate a con vincing diagnostic adv antage of SUVbased diagnosis o ver qualitati ve inter pretation.44,140 Moreover, Israel and colleagues57 found that there was no statistical difference between the SUV of pre-malignant, malignant, benign, and ph ysiologic lesions in an e valuation of unexpected gastrointestinal foci of FDG detected by PET/CT. Nevertheless, whether assessed qualitatively or by SUV, the intensity of FDG uptak e seems to be an important biomark er of disease agg ressiveness in man y forms of malignanc y. Larger prospective trials are being conducted in lung cancer and lymphoma to further define the validity of FDG-PET and changes in SUV as a biomarker of eventual outcome in these conditions.

ALTERNATIVE TRACERS AS CANCER BIOMARKERS The diversity of f actors that contribute to tissue glucose metabolism limits the specif icity of FDG as a cancer imaging agent. Uncontrolled proliferation is one of the most important hallmarks of cancer and is also one of the major tar gets of cancer therap y. PET tracers of cellular

890

MOLECULAR IMAGING: PRINCIPLES AND PRACTICE

proliferation are likely to be more specific for malignancy than FDG and should be par ticularly helpful in characterizing the biolo gic agg ressiveness of tumors. To date, the most promising of proliferation tracers for clinical application appears to be F-18 fluorothymidine (FLT),141,142 the uptake of which is closely correlated with cellular proliferation.143 However, high bone mar row uptake, related to normal hematopoietic cell proliferation, and high hepatic uptake, due to metabolism of the tracer, lead to high background activity in major sites of metastatic in volvement by cancer. This limits the sensiti vity of FLT for detection of disease. Furthermore, reactive germinal centers in lymphoma nodes responding to antigenic challenge from inflammatory processes mak e inter pretation of l ymph nodes dif ficult. Accordingly, the major role of FL T is likely to be as an adjunct to FDG, particularly in monitoring treatments that have a cytostatic action. Extensive efforts have been also made to develop PET tracers to enable the noninvasive imaging of hypoxia as a biomarker. This is signif icant because of the impor tance of tissue hypoxia as a prognostic factor in various cancers, being associated with a poor response to therap y, particularly including radiotherapy, and also to a more aggressive metastatic phenotype. Fluorine-18 fluoromisonidazole (FMISO) has been the most widel y evaluated tracer both in pre-clinical and human studies. The ability of FMISO to stratify pro gnosis in patients recei ving con ventional radiotherapy and tar geted therap y of h ypoxia has been demonstrated.144,145 Unfortunately, FMISO is a relati vely lipophilic compound and demonstrates relati vely lo w uptake in hypoxic tissue relative to normal tissue and slow clearance from normal tissues requiring delayed scanning with consequences on contrast and image quality. This has led to the development of other hypoxic tracers with more favorable imaging properties. Fluorine-18 fluoroazomycin arabinoside (FAZA) is one such agent. 146 Due to the f avorable imaging characteristics and ready availability of fluorine-18, there has been an understandable focus on fluorinated tracers. Ho wever, a wide range of both c yclotron and generator -produced isotopes are now becoming available. These include yttrium-86 and iodine-124. These tracers lend themselv es to comple xing with biolo gic macromolecules, such as peptides 147 and antibodies. The availability of long-lived positron-emitting radionuclides particularly offers opportunities for translational research in the labeling of larger molecular species, including monoclonal antibodies, w hich can ha ve slo w accumulation in tissues.148 The generator-produced Ga-68 has a suitab ly shor t half-life to allo w e valuation of processes with a shor t te