Principles of Molecular Medicine [2 ed.] 1588292029, 9781588292025, 9781592599639

This fully updated and expanded edition of the much-acclaimed Principles of Molecular Medicine provides an integrated su

1,724 190 21MB

English Pages 1304 [1305] Year 2006

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Principles of Molecular Medicine [2 ed.]
 1588292029, 9781588292025, 9781592599639

Citation preview

Principles of

Molecular

Medicine SECOND EDITION

Edited by

Marschall S. Runge, MD, PhD Cam Patterson, MD SECTION EDITORS Richard C. Boucher, MD David A. Brenner, MD Luis A. Diaz, MD James P. Evans, MD, PhD Daniel J. Garry, MD, PhD Lowell A. Goldsmith, MD, MPH Steven M. Holland, MD Samuel Klein, MD Terry Magnuson, PhD W. Stratford May, Jr., MD, PhD

Michael J. McPhaul, MD William E. Mitch, MD Charles B. Nemeroff, MD, PhD Peadar G. Noone, MD Kerry J. Ressler, MD, PhD Anthony Rosenzweig, MD Stephen M. Strittmatter, MD, PhD Swee Lay Thein, DSc, FRCP, FRCPath, Fmedsci

PRINCIPLES OF MOLECULAR MEDICINE SECOND EDITION

SECTION EDITORS RICHARD C. BOUCHER, MD

W. STRATFORD MAY, JR., MD, PhD

CYSTIC FIBROSIS RESEARCH AND TREATMENT CENTER DIVISION OF PULMONARY DISEASES DEPARTMENT OF MEDICINE UNIVERSITY OF NORTH CAROLINA CHAPEL HILL, NC

UNIVERSITY OF FLORIDA SHANDS CANCER CENTER UNIVERSITY OF FLORIDA GAINESVILLE, FL

DAVID A. BRENNER, MD DEPARTMENT OF MEDICINE COLLEGE OF PHYSICIANS AND SURGEONS COLUMBIA UNIVERSITY MEDICAL CENTER NEW YORK, NY

LUIS A. DIAZ, MD DEPARTMENT OF DERMATOLOGY SCHOOL OF MEDICINE UNIVERSITY OF NORTH CAROLINA CHAPEL HILL, NC

JAMES P. EVANS, MD, PhD DEPARTMENT OF GENETICS UNIVERSITY OF NORTH CAROLINA CHAPEL HILL, NC

DANIEL J. GARRY, MD, PhD DEPARTMENTS OF INTERNAL MEDICINE AND MOLECULAR BIOLOGY UNIVERSITY OF TEXAS SOUTHWESTERN MEDICAL CENTER DALLAS, TX

LOWELL A. GOLDSMITH, MD, MPH DEPARTMENT OF DERMATOLOGY SCHOOL OF MEDICINE UNIVERSITY OF NORTH CAROLINA CHAPEL HILL, NC

STEVEN M. HOLLAND, MD LABORATORY OF CLINICAL INFECTIOUS DISEASES NATIONAL INSTITUTE OF ALLERGY AND INFECTIOUS DISEASE NATIONAL INSTITUTES OF HEALTH BETHESDA, MD

SAMUEL KLEIN, MD CENTER FOR HUMAN NUTRITION DIVISION OF GERIATRICS AND NUTRITIONAL SCIENCES WASHINGTON UNIVERSITY SCHOOL OF MEDICINE ST. LOUIS, MO

TERRY MAGNUSON, PhD DEPARTMENT OF GENETICS UNIVERSITY OF NORTH CAROLINA CHAPEL HILL, NC

MICHAEL J. MCPHAUL, MD DEPARTMENT OF INTERNAL MEDICINE UNIVERSITY OF TEXAS SOUTHWESTERN MEDICAL CENTER DALLAS, TX

WILLIAM E. MITCH, MD DIVISION OF NEPHROLOGY BAYLOR COLLEGE OF MEDICINE HOUSTON, TX

CHARLES B. NEMEROFF, MD, PhD DEPARTMENT OF PSYCHIATRY AND BEHAVIORAL SCIENCES EMORY UNIVERSITY SCHOOL OF MEDICINE ATLANTA, GA

PEADAR G. NOONE, MD CYSTIC FIBROSIS RESEARCH AND TREATMENT CENTER DIVISION OF PULMONARY DISEASES DEPARTMENT OF MEDICINE UNIVERSITY OF NORTH CAROLINA CHAPEL HILL, NC

KERRY J. RESSLER, MD, PhD DEPARTMENT OF PSYCHIATRY AND BEHAVIORAL SCIENCES EMORY UNIVERSITY SCHOOL OF MEDICINE ATLANTA, GA

ANTHONY ROSENZWEIG, MD CENTER FOR IMMUNOLOGY AND INFLAMMATORY DISEASES MASSACHUSETTS GENERAL HOSPITAL BOSTON, MA

STEPHEN M. STRITTMATTER, MD, PhD DEPARTMENT OF NEUROLOGY YALE UNIVERSITY SCHOOL OF MEDICINE NEW HAVEN, CT

SWEE LAY THEIN, DSc, FRCP, FRCPath, FMedSci DEPARTMENT OF HAEMATOLOGICAL MEDICINE GUY'S, KING'S AND ST. THOMAS' SCHOOL OF MEDICINE LONDON, UK

PRINCIPLES OF

MOLECULAR MEDICINE SECOND EDITION

EDITED BY

MARSCHALL S. RUNGE, MD, PhD DEPARTMENT OF MEDICINE, DIVISION OF CARDIOLOGY CAROLINA CARDIOVASCULAR BIOLOGY CENTER UNIVERSITY OF NORTH CAROLINA SCHOOL OF MEDICINE CHAPEL HILL, NC

CAM PATTERSON, MD DEPARTMENT OF MEDICINE, DIVISION OF CARDIOLOGY CAROLINA CARDIOVASCULAR BIOLOGY CENTER UNIVERSITY OF NORTH CAROLINA SCHOOL OF MEDICINE CHAPEL HILL, NC

FOREWORD BY

VICTOR A. MCKUSICK,

MD

JOHNS HOPKINS UNIVERSITY BALTIMORE, MD

© 2006 Humana Press Inc. 999 Riverview Drive, Suite 208 Totowa, New Jersey 07512 www.humanapress.com All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise without written permission from the Publisher. All papers, comments, opinions, conclusions, or recommendations are those of the author(s), and do not necessarily reflect the views of the publisher. This publication is printed on acid-free paper. ∞ ANSI Z39.48-1984 (American Standards Institute) Permanence of Paper for Printed Library Materials. Production Editor: Jennifer Hackworth Cover design by Patricia F. Cleary For additional copies, pricing for bulk purchases, and/or information about other Humana titles, contact Humana at the above address or at any of the following numbers: Tel.: 973-256-1699; Fax: 973-256-8341; E-mail: [email protected]; or visit our Website: www.humanapress.com Photocopy Authorization Policy: Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by Humana Press, provided that the base fee of US $30.00 per copy is paid directly to the Copyright Clearance Center at 222 Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy license from the CCC, a separate system of payment has been arranged and is acceptable to Humana Press Inc. The fee code for users of the Transactional Reporting Service is: [1-58829-202-9/06 $30.00]. Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1 1-59259-963-X (e-book)

Library of Congress Cataloging in Publication Data

Principles of molecular medicine / edited by Marschall S. Runge ; foreword by Victor A. Mckusick. -- 2nd ed. p. ; cm. Includes bibliographical references and index. ISBN 1-58829-202-9 (alk. paper) 1. Medical genetics. 2. Pathology, Molecular. 3. Molecular biology. I. Runge, Marschall Stevens, 1954- . [DNLM: 1. Genetics, Medical. 2. Gene Therapy. 3. Molecular Biology. QZ 50 P9573 2006] RB155.P695 2006 616'.042--dc22 2005034346

Foreword

means that there are many new opportunities and challenges for clinical medicine. One of the effects of the completion of the Human Genome Project is the increasing application of the fields of molecular biology and genetics to the understanding and management of common diseases. Assimilation of the new developments since the first edition has been ably accomplished by Drs. Runge and Patterson with the help of their many knowledgeable authors. As was evident in the first edition, molecular genetics is involved in every specialty of medicine. A recurrent theme in that edition, perhaps even more striking in the present one, is that information gleaned and research methods designed in one specialty have been highly influential on researchers and physicians in other fields—often in ways that could not have been foreseen. The editors have succeeded in considering all the disciplines while searching for connections and correlations that might otherwise be missed. The organization selected by the editors allows for the molecular bases of disease, as well as the constantly evolving areas of ethical issues and counseling that affect all disciplines, to be covered in the opening section. Specifics in the several medical disciplines are then handled splendidly in the sections that follow. Each chapter resounds with the amazing detail of what is known and simultaneously probes the many unanswered questions that provide new avenues for research in the 21st century. The state-of-the-art focus in each specialty will be much appreciated by the reader, whether practitioner, researcher or student. The authors and section editors that participated in this text are recognized leaders in their fields from around the globe. They and Drs. Runge and Patterson, who have led and coordinated this extraordinary effort, deserve commendation. The product is a text that will be useful for all interested in the molecular pathogenesis of disease.

The concept of molecular medicine dates back to Linus Pauling, who in the late 1940s and early 1950s generalized from the ideas that came from the study of the sickle cell hemoglobin molecule. With the first cloning of human genes about 1976, molecular genetics took the molecular perspective on disease to the level of DNA. The term molecular medicine achieved wide currency in the 1980s with the assignment of this designation to journals, at least one society, institutes, and academic divisions of departments of internal medicine. Undoubtedly, molecular medicine has been abetted by the Human Genome Project, which has aided greatly in the molecular characterization of disease. Mapbased gene discovery, as in positional cloning of previously unknown genes responsible for “mystery diseases,” could be now replaced by sequence-based gene discovery. What is molecular medicine? In the first edition of Principles of Molecular Medicine, Francis Collins seems to define it as “molecular genetics and medicine”—the last four words of his Foreword. He was referring to the pervasive relevance of genetics and genomics to all of medicine. In essence, molecular medicine is genetic medicine. Since the publication of the first edition Principles of Molecular Medicine in 1998, the Human Genome Project has provided a “complete” sequence of the human genome with several surprising revelations relevant to molecular medicine. As indicated in the Preface of the first edition, the total count of genes was thought to be 50,000 to 80,000. Scrutiny of the complete human sequence leads to a count only half that, perhaps fewer than 30,000. It has come to be realized that each gene can give rise to multiple protein gene products through alternative splicing of pre-messenger RNA, as well as through different posttranscriptional modification of the gene products. Each gene may on the average have as many as 10 different protein products. Mutations in different ones of these can cause quite different clinical disorders. Thus the focus has shifted to the transcriptome and to the proteins that constitute the proteome—a shift from genomics to proteomics. Compilation of the rapidly expanded topic of molecular medicine since the edition of some 8 years ago is a daunting task. The rate at which new discoveries have been made

Victor A. McKusick, MD

v

FOREWORD

Foreword to the First Edition

Until recently, medical genetics and molecular medicine were considered the exclusive province of academic specialists in tertiary-care medical centers. Queried about their familiarity with molecular genetic aspects of clinical medicine, most primary-care providers only a few years ago would have responded that such matters were irrelevant to their daily practice. Yet few could say that today. Few internists or general practitioners have not prescribed recombinant insulin, tPA, or erythropoietin; few pediatricians have not gone through the molecular evaluation of a child with dysmorphology or learning disability; few obstetricians have not performed amniocentesis or CVS for couples at increased genetic risk; and few general surgeons have not faced penetrating questions about the role of genetic testing or prophylactic surgery from women with a strong family history of breast or ovarian cancer. This level of emergence of molecular genetics into clinical medicine is still quite modest, however, compared to what is coming. As the human genome project hurtles toward completion of the sequence of a reference human genome, and the identification of all human genes, by 2005, the pace of revelations about human illness will continue to accelerate. Until recently, most disease-gene discoveries have related to single-gene disorders (cystic fibrosis, fragile X syndrome, and so on) or to Mendelian subsets of more common illnesses (BRCA1 and BRCA2, the hereditary nonpolyposis colon cancer syndromes, and so on). But with

the initiation in 1998 of an aggressive new genome project goal, cataloging all common human sequence variations, it is expected that the weaker polygenic contributors to virtually all diseases will begin to be discerned. Many consequences will result. Individualized preventive medicine strategies, rooted in the gene-based determination of future risk of illness, will become part of the regular practice of medicine. New designer drugs, based not on empiricism but on a detailed understanding of the molecular pathogenesis of disease, will appear. Pharmacogenomics, wherein the efficacy and toxicity of a particular drug regimen can be predicted based on patient genotype, will become a standard component of designing optimum therapy for the individual. And gene therapy, fed by a wealth of disease-gene discoveries, will mature into a significant part of the physician’s armamentarium against disease. As we watch this train coming down the track, this is an ideal time to collect information about molecular medicine into one authoritative text. Principles of Molecular Medicine aims to do just that, bridging the current gap between basic science and the bedside. It will thus be useful to researchers and clinicians alike. With more than 100 chapters covering a wide variety of topics, its distinguished cohort of section editors, and its abundant tables and illustrations, it provides an accessible and much needed manual to the present and the future of molecular genetics and medicine. Francis S. Collins

vii

Preface

potential to serve as cell-based therapies in chronic and degenerative diseases. New cell growth and cell death mechanisms that are dysregulated in neoplasia and that may serve as new anticancer targets have been elucidated. Advances have been made in understanding the biology of previously untreatable neurodegenerative diseases such as Huntington's disease. These and many other important advances in our understanding of human diseases are elucidated in this edition. In addition, we have been able to note the new epistemologies in the genetic basis of human disease that are rapidly emerging. For instance, characterization of candidate genes for human diseases has expanded well beyond monogenic diseases, the study of late-onset diabetes being a notable example. Molecular alterations can have far-reaching effects on many systems. The identification of genes for epithelial sodium channels has led to a deeper understanding of their role in disorders of total body Na+ homeostasis, blood volume, and blood pressure. As a result of advances like these, views that had been held for much of the 20th century are being reconsidered. For example, hereditary hemochromatosis, a familial disease characterized by excess tissue deposits of iron leading to end-organ damage, has traditionally been thought to result from mutations in a single gene. Very recently, the identification of similar phenotypes associated with mutations of at least four different iron-metabolism genes has expanded our understanding of the pathophysiology of this relatively common genetic disease. One common theme repeated in the chapters in this text is that the pathophysiology of disease is often a succession of genetic alterations, not just a single mutation. Although understanding these genetic relationships is never simple, their role in human diseases is all the more fascinating to consider. Human health often appears tenuous, but the discovery that a series of genetic missteps is often required to produce many disease states can be reassuring. The presence of several genetic steps in a disease process also suggests that multiple therapeutic targets may exist to modulate the course of these diseases. Less exhilarating is the knowledge that numerous diseases result not just from a complex succession of genetic missteps, but also from an individual’s

Since publication of the first edition of Principles of Molecular Medicine, dramatic discoveries in molecular medicine along with concomitant rapid technological advances have revolutionized the diagnosis and treatment of a broad range of human diseases. Given the pace of new discovery, genetic- and cell-based therapies may well become a common part of the physicians’ armamentarium in the near future. Direct links between genetic mutations and diseases are being mapped almost routinely. Genomic approaches to diseases such as breast cancer have led to identification of previously unrecognized malignancies and the ability to prognosticate outcomes to therapy. The delicate interplay between adipocytes and regulation of insulin sensitivity, the roles of bone morphogenetic proteins in pulmonary hypertension, and the discovery of mutations involved in an array of cardiomyopathies are but a few of the important recent advances that have direct implications for patient care. It is virtually impossible to keep track of the breadth of discovery that has led to these biomedical advances. The goal of the many authors and editors of this second edition of the Principles of Molecular Medicine has been to present the voluminous discoveries of the past decade in a format that captures the essence of scientific discovery but allows rapid assimilation in each particular area. This second edition again includes chapters describing advances in fields paralleling traditional medical texts, and will be especially useful to specialists who are updating their education, practicing physicians interested in keeping abreast of new developments, and students appropriately curious about what is known and what lies ahead. Although only 8 years have passed since the first edition was published, we have made every effort to comprehensively update chapters with recent advances and have added chapters for disease entities and areas in which discovery has accelerated during the past five years. As we have participated in the assembly of this volume, we have had the good fortune to review in depth the molecular discoveries that are transforming medical practice. For example, in the interval since the first edition of this text, stem cell populations have been discovered that regenerate muscle, heart, and neural cell populations, and that have the ix

x

PREFACE

interaction with the environment. Abundant examples of this principle are present throughout clinical medicine, and are described in detail in this edition of Principles of Molecular Medicine. Paradoxically, as new discoveries are made, new mysteries appear. The many advances described in this volume often raise as many new questions as they answer. On the one hand, this indicates that biomedical discovery and medical practice will continue to evolve. On the other hand, the changes in medical care described in the chapters of this text are an indication that the unresolved questions of today may be harbingers of new therapeutic approaches in years to come. It has been our pleasure to bring together in-depth expositions of the most recent advances in molecular medicine. We invite you to enjoy this magnificent point in biomedical history, as genetics and molecular medicine continue to merge with clinical practice. The compendium of information in Principles of Molecular Medicine: Second Edition, has been made possible by the tireless efforts of our section editors. Without their expertise and commitment to this project, this textbook would not be possible. In addition, we

thank the individual authors for sharing their expertise with all of us. In addition to the phenomenal work of the editors and contributors, we would like to extend special thanks to Ms. Katie O’Brien for her commitment to this project; to Ms. Angela Clotfelter-Rego, whose tireless efforts despite numerous obstacles made this project possible; and to Ms. Carolyn Kruse, who synthesized the work of numerous authors to create a blazingly readable text. The editors thank our families, who have tolerated yet another joint effort. Finally, we would like to dedicate this volume to the first chairmen of medicine we had the privilege to serve under, Juha P. Kokko, MD, PhD, and Victor McKusick, MD. As physicians and scientists, these gentlemen nurtured many of the contributors to this edition, and their own work as scientists is frequently cited directly and indirectly in these chapters. It is on the shoulders of men like these that the principles of molecular medicine have been determined. Marschall S. Runge, MD, PhD Cam Patterson, MD

Contents

FOREWORD FOREWORD TO THE FIRST EDITION PREFACE CONTRIBUTORS LIST OF COLOR PLATES

11

V

Carolyn Y. Ho and Christine E. Seidman

VII

12

IX XV

13

14 15

Identifying Causal Genetic Factors ................... 19 16 17 18 19 20

Genetic Counseling ............................................ 46 Animal Models in Biomedical Research: Ethics, Challenges, and Opportunities .......................................... 53

III. PULMONARY DISEASES 21

Ethical, Legal, and Social Implications ............. 61

RICHARD C. BOUCHER AND PEADAR G. NOONE Idiopathic Interstitial Pneumonias ................... 188 Talmadge E. King, Jr.

Marcia Van Riper

22

Asthma .............................................................. 198 Stephen T. Holgate, Gordon Dent, and Mark G. Buckley

II. CARDIOLOGY 10

Cardiovascular Gene Therapy .......................... 175 David A. Dichek

Robert W. Williams

9

Genomics .......................................................... 166 Calum A. MacRae and Christopher J. O’Donnell

Robin L. Bennett

8

Arrhythmias ...................................................... 157 Barry London

Hemophilia as a Model Disease for Gene Therapy of Genetic Disorders........................ 39 Jay Lozier

7

Cardiac Hypertrophy ........................................ 146 Thomas Force and Jeffrey D. Molkentin

Pharmacogenetics ............................................... 34 Renee E. Edkins and Dennis J. Cheek

6

Hypertension .................................................... 138 Khurshed A. Katki and Donald J. DiPette

Cancer Genetics and Molecular Oncology ........ 27 Sharon E. Plon

5

Lipid Metabolism and Coronary Artery Disease .............................................. 130 Mason W. Freeman

Christopher I. Amos, John S. Witte, and William G. Newman

4

Atherosclerotic Coronary Disease ................... 121 Robert E. Gerszten and Anthony Rosenzweig

Nontraditional Inheritance ................................... 9 Shawn E. McCandless and Suzanne B. Cassidy

3

Aortic Diseases ................................................. 116 Saumya Das, James L. Januzzi, Jr., and Eric M. Isselbacher

TERRY MAGNUSON AND JAMES P. EVANS Mendelian Inheritance .......................................... 3 Bruce R. Korf

2

Heart Failure: Emerging Concepts in Excitation–Contraction Coupling and β-Adrenoceptor Coupling .......................... 105 Clive J. Lewis, Federica del Monte, Sian E. Harding, and Roger J. Hajjar

XXV

I. GENETICS 1

Inherited Cardiomyopathies ............................... 98

ANTHONY ROSENZWEIG Congenital Heart Disease ................................... 69

23

Pulmonary Emphysema ................................... 214 Steven D. Shapiro

Lazaros K. Kochilas and Alvin J. Chin xi

xii

24

CONTENTS

Pulmonary Hypertension .................................. 223

43

Brian Fouty and David M. Rodman

25

Acute Lung Injury ............................................ 231

Tomonobu Hasegawa

44

David C. Christiani and Michelle Ng Gong

26

Primary Ciliary Dyskinesia .............................. 239

45

28

29 30

46

V. METABOLIC DISORDERS

Sarcoidosis ........................................................ 269 47

Disorders of Pulmonary Surfactant Homeostasis ................................................. 277

48

Cellular Regulation of Lipolysis ...................... 518

49

Fat-Induced Insulin Resistance and Atherosclerosis ...................................... 524

Patrick Tso and Min Liu Allan Green

IV. ENDOCRINOLOGY MICHAEL J. MCPHAUL Mechanisms of Hormone Action ..................... 291 Stephen R. Hammes, Carol A. Lange, and Michael J. McPhaul

32

Guenther Boden

50

Metabolic and Molecular Aspects of Sarcopenia ................................................ 529 W. Todd Cade and Kevin E. Yarasheski

51

Diabetes Mellitus .............................................. 308 William L. Lowe, Jr.

33

SAMUEL KLEIN Gastrointestinal Regulation of Food Intake ..... 513

Kelly D. Chason and Stephen L. Tilley

Jeffrey A. Whitsett, Susan E. Wert, and Bruce C. Trapnell

31

Ovarian Diseases .............................................. 495 Elizabeth A. McGee and Tammy L. Loucks

Gene Therapy for Lung Diseases ..................... 259 Jane C. Davies, Duncan M. Geddes, and Eric W.F.W. Alton

Molecular Endocrinology of the Testis ........... 473 Marco Marcelli, Glenn R. Cunningham, José M. Garcia, Kirk C. Lo, and Dolores J. Lamb

Cystic Fibrosis .................................................. 251 Scott H. Donaldson and Richard C. Boucher

Defects of Androgen Action ............................ 466 Michael J. McPhaul

Peadar G. Noone, Maimoona Zariwala, and Michael R. Knowles

27

Disorders of Pubertal Development ................. 453

Adipose Tissue Development and Metabolism ............................................ 535 Sheila Collins, Yushi Bai, and Jacques Robidoux

Pituitary Function and Neoplasia ..................... 319 Shlomo Melmed

34

Growth Hormone Deficiency Disorders .......... 327

VI. GASTROENTEROLOGY

Amy Potter and John A. Phillips, III

35

Thyroid Disorders ............................................ 337

52

Peter Kopp

36

Johnson Yiu-Nam Lau, Jane Wing-Sang Fang, Masashi Mizokami, Robert G. Gish, and Teresa L. Wright

Disorders of the Parathyroid Gland ................. 357 Andrew Arnold and Michael A. Levine

37

Congenital Adrenal Hyperplasia ...................... 365

53

Adrenal Diseases .............................................. 377

54

Multiple Endocrine Neoplasia Type 1 ............. 386

55

41

42

Pancreatic Exocrine Dysfunction ..................... 573 David C. Whitcomb and Jonathan A. Cohn

Rajesh V. Thakker

40

Hereditary Hemochromatosis ......................... 567 Francesca Ferrara, Elena Corradini, and Antonello Pietrangelo

Richard J. Auchus, William E. Rainey, and Keith L. Parker

39

Molecular Diagnostics in Hepatitis B .............. 554 Scott Bowden and Stephen Locarnini

Robert C. Wilson and Maria I. New

38

DAVID A. BRENNER Hepatitis C ........................................................ 542

Multiple Endocrine Neoplasia Type 2 ............. 393

56

Small and Large Bowel Dysfunction ............... 581

Robert F. Gagel, Sarah Shefelbine, Hironori Hayashi, and Gilbert Cote

57

Disorders of Sex Determination and Differentiation ....................................... 400

The Molecular Mechanisms of Helicobacter pylori-Associated Gastroduodenal Disease ............................... 590

Charmian A. Quigley

Peter B. Ernst

Sex Chromosome Disorders ............................. 446 Andrew R. Zinn

Deborah C. Rubin

CONTENTS

xiii

73

Colorectal Cancer ............................................. 720

VII. NEPHROLOGY 58

59

74

Christopher S. Wilcox

75

Nephrogenic Diabetes Insipidus: Water and Urea Transport .......................... 622

76

63

77

Moira R. Jackson and Stephen P. Sugrue

Interstitial Nephritis ......................................... 636 Carla Zoja and Giuseppe Remuzzi

Anthony T. Yachnis and Henry V. Baker

The Pathophysiology of Acute Renal Failure ................................................ 643

Loss of Lean Body Mass in Uremia ................ 650 S. Russ Price and William E. Mitch

65

Mechanisms of Renal Allograft Rejection ...... 656 Daniel R. Goldstein, Anirban Bose, and Fadi G. Lakkis

VIII. MUSCULOSKELETAL 66

DANIEL J. GARRY Muscle Development and Differentiation ....... 665 Eric N. Olson

67

78

HEMATOLOGICAL MALIGNANCIES 79 Acute Myeloid Leukemias ............................... 767 Iris T. Chan and D. Gary Gilliland

80 81

69

82

83

Muscular Dystrophies: Mechanisms ................ 693 Peter B. Kang and Louis M. Kunkel

71

Rhabdomyosarcomas ....................................... 700 Stephen J. Tapscott

84

X. HEMATOLOGY 85

SWEE LAY THEIN Disorders of the Red Cell Membrane .............. 830 Jean Delaunay and Gordon W. Stewart

86

Paroxysmal Nocturnal Hemoglobinuria .......... 838 Bruno Rotoli, Khedoudja Nafa, and Antonio M. Risitano

87

Iron Metabolism ............................................... 848 Nancy C. Andrews

88

Correction of Genetic Blood Defects by Gene Transfer .......................................... 854 Marina Cavazzana-Calvo, Salima HaceinBey-Abina, Adrian J. Thrasher, Philippe Leboulch, and Alain Fischer

W. STRATFORD MAY, JR.

W. Stratford May, Jr. and Xingming Deng

HIV-1, AIDS, and Related Malignancies ........ 818 Maureen M. Goodenow and James J. Kohler

IX. ONCOLOGY SOLID TUMORS 72 Apoptosis .......................................................... 709

Multiple Myeloma: A Story of Genes and the Environment .................................... 804 Kenneth H. Shain and William S. Dalton

Zhen Yan and R. Sanders Williams

70

Non-Hodgkin’s Lymphoma and Chronic Lymphocytic Leukemia ............................... 794 Ayad M. Al-Katib and Anwar N. Mohamed

Stem Cells and Muscle Regeneration .............. 682

Skeletal Muscle Hypertrophy and Response to Training ............................ 688

Chronic Myelogenous Leukemia ..................... 789 Brian J. Druker

Skeletal Muscle Structure and Function .......... 674

Cindy M. Martin, Thomas J. Hawke, and Daniel J. Garry

Acute Lymphoblastic Leukemia ...................... 776 Alexander E. Perl and Donald Small

Elizabeth M. McNally, Karen A. Lapidos, and Matthew T. Wheeler

68

Cutaneous Melanoma ....................................... 752 Applications of Gene Expression Profiling to the Study of Malignant Gliomas .................................. 760

Glomerulonephritis and Smad Signaling ......... 629

Didier Portilla, Gur P. Kaushal, Alexei G. Basnakian, and Sudhir V. Shah

64

Discoveries and Frontiers in Prostate Cancer Translational Sciences ................................. 743 Jonathan W. Simons

Hui Y. Lan and Richard J. Johnson

62

Lung Cancer ..................................................... 736 Lei Xiao

Jeff M. Sands and Daniel G. Bichet

61

Breast Cancer ................................................... 728 Yi Huang and Nancy E. Davidson

Hypertension and Sodium Channel Turnover .... 613 Douglas C. Eaton, Bela Malik, and He-Ping Ma

60

Satya Narayan

WILLIAM E. MITCH Nitric Oxide Synthase and Cyclooxygenase in the Kidneys .............................................. 606

89

Bone Marrow Failure Syndromes .................... 862 J. Eric Turner and Thomas C. Shea

xiv

90

CONTENTS

Coagulation Disorders ...................................... 871 Stephan Moll and Gilbert C. White II

91

Advances in Transfusion Safety ...................... 883 Lorna M. Williamson and Jean-Pierre Allain

XI. IMMUNOLOGY AND INFECTIOUS DISEASES 92 93

105 Epidermolysis Bullosa ................................... 1024 Jouni Uitto, Gabriele Richard, and Angela M. Christiano

107 Genetic Epidermal Diseases........................... 1043

95

Molecular Pathogenesis of Fungal Infections ..................................... 920

Marcus J. Schultz and Tom van der Poll

Brahm H. Segal

XII. DERMATOLOGY LUIS A. DIAZ AND LOWELL A. GOLDSMITH IMMUNE MEDIATED Psoriasis ............................................................ 939 James T. Elder

Atopic Dermatitis ............................................. 948 Kefei Kang, Donald Y. M. Leung, and Kevin D. Cooper

Pemphigus Foliaceus, Pemphigus Vulgaris, Paraneoplastic Pemphigus, Bullous Pemphigoid, Herpes Gestationis and Cicatricial Pemphigoid ........................ 959 Ning Li, David S. Rubenstein, Zhi Liu, and Luis A. Diaz

99

Julie V. Schaffer and Jean L. Bolognia

HIV Molecular Biology, Treatment, and Resistance .............................................. 905 Cellular and Molecular Aspects of Pneumonia ............................................... 912

98

104 Disorders of Hypopigmentation ..................... 1011

Juan C. Gea-Banacloche

94

97

Nancy E. Thomas

106 Connective Tissue Disorders: The Paradigms of Ehlers-Danlos Syndrome and Pseudoxanthoma Elasticum ................ 1035

STEVEN M. HOLLAND Immunomodulation .......................................... 893

Frank Maldarelli

96

103 Melanoma and Nevi ....................................... 1004

Systemic Lupus Erythematosus ....................... 970 Amr H. Sawalha, Luis A. Diaz, and John B. Harley

100 Systemic Sclerosis ............................................ 979 Chris T. Derk and Sergio A. Jimenez

NONIMMUNE 101 Diseases With Signaling and Transcriptional Abnormalities ............................................... 991 David S. Rubenstein, Amy Stein, and Lowell A. Goldsmith

102 Genetic Skin Diseases With Neoplasia............ 998 David S. Rubenstein

Jouni Uitto and Franziska Ringpfeil Amy Stein and Lowell A. Goldsmith

108 Genetic Hair and Nail Defects ....................... 1052 Marija Tadin-Strapps and Angela M. Christiano

109 Metabolic Genetic Disorders With Prominent Skin Findings .................. 1059 Christopher B. Mizelle and Lowell A. Goldsmith

110 Heritable Conditions Affecting Tissues of the Oral Cavity....................................... 1065 J. Tim Wright

XIII. NEUROLOGY STEPHEN M. STRITTMATTER 111 The Genetic Basis of Human Cerebral Cortical Malformations .............................. 1073 Bernard S. Chang and Christopher A. Walsh

112 Muscular Dystrophies: Molecular Diagnosis .................................. 1080 Eric P. Hoffman and Erynn S. Gordon

113 Channelopathies of the Nervous System ....... 1088 Stephen C. Cannon

114 Charcot-Marie-Tooth Disease and Related Peripheral Neuropathies ............................. 1097 James R. Lupski

115 Amyotrophic Lateral Sclerosis and Related Motor Neuron Disorders: Lessons in Pathogenesis and Therapy From Genetics ...................... 1105 Robert H. Brown, Jr.

116 Trinucleotide Repeat Disorders ..................... 1114 Huda Y. Zoghbi

117 Parkinson’s Disease ....................................... 1123 Deepak M. Sampathu and Virginia M.-Y. Lee

xv

CONTENTS

118 Genetics and Neurobiology of Alzheimer's Disease and Frontotemporal Dementias ................................................... 1130 Peter H. St. George-Hyslop

119 Prion Diseases ................................................ 1142 Adriano Aguzzi

120 Narcolepsy and Other Neurological Sleep Disorders .......................................... 1153 Thomas E. Scammell

121 Neurofibromatosis 1 and 2 ............................. 1160 Gregory J. Esper and David H. Gutmann

122 Axonal Regeneration and Recovery From Chronic Central Nervous System Injury ............................................. 1165 Stephen M. Strittmatter

XIV. PSYCHIATRY KERRY J. RESSLER AND CHARLES B. NEMEROFF 123 Molecular Mechanisms Regulating Behavior: Focus on Genetic and Environmental Influences ................................................... 1175 Kerry J. Ressler and Charles B. Nemeroff

124 The Complex Genetics of Psychiatric Disorders .................................................... 1184 Ming T. Tsuang, Stephen J. Glatt, and Stephen V. Faraone

125 Treating Depression: AntidepressantActivated Signaling Pathways as a Source for Novel Drug Targets .......... 1191 Marcelo Páez-Pereda and Florian Holsboer

126 Anxiety Disorders .......................................... 1197 Gregory M. Sullivan and Jack M. Gorman

127 Trauma Spectrum Disorders .......................... 1203 Christine Heim, J. Douglas Bremner, and Charles B. Nemeroff

128 Schizophrenia ................................................. 1211 Carol A. Tamminga, Deborah C. Medoff, and Gunvant Thaker

129 Disorders of Substance Abuse and Dependence ......................................... 1220 Eric J. Nestler and Jennifer Chao

130 Autism and Related Disorders ....................... 1228 Linmarie Sikich, Thomas H. Wassink, Kevin A. Pelphrey, and Joseph Piven

INDEX ....................................................................... 1237

Contributors

ADRIANO AGUZZI, MD, PhD, FRCP, FRCPath, Institute of Neuropathology, University Hospital, Zurich, Switzerland ERIC WFW ALTON, MD, FRCP, Department of Gene Therapy, Imperial College London, National Heart and Lung Institute, London, UK CHRISTOPHER I. AMOS, PhD, Section Head of Computational and Genetic Epidemiology, Departments of Epidemiology, and Biostatistics and Applied Mathematics, University of Texas M.D. Anderson Cancer Center, Houston, TX AYAD M. AL-KATIB, MD, Lymphoma Research Laboratory, Wayne State University School of Medicine, Detroit, Michigan and Van Elslander Cancer Center, Grosse Pointe Woods, MI JEAN-PIERRE ALLAIN, MD, PhD, Department of Haematology, University of Cambridge, Cambridge, UK NANCY C. ANDREWS, MD, PhD, Howard Hughes Medical Institute, Children’s Hospital Boston, Harvard Medical School, Boston, MA ANDREW ARNOLD, MD, Center for Molecular Medicine and Division of Endocrinology and Metabolism, University of Connecticut School of Medicine, Farmington, CT RICHARD J. AUCHUS, MD, PhD, Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, TX YUSHI BAI, MD, PhD, Endocrine Biology Program, Division of Biological Sciences, CIIT Centers for Health Research, Research Triangle Park, NC HENRY V. BAKER, PhD, Department of Molecular Genetics and Microbiology, College of Medicine, University of Florida, Gainesville, FL ALEXEI G. BASNAKIAN, MD, PhD, Department of Internal Medicine, Division of Nephrology, University of Arkansas for Medical Sciences, Central Arkansas Veterans Healthcare System, Little Rock, AR ROBIN L. BENNETT, MS, CGC, Department of Medicine, Division of Medical Genetics, University of Washington Medical Center, Seattle, WA DANIEL G. BICHET, MD, Department of Medicine, University of Montreal, Hôpital du Sacré-Coeur de Montreal, Quebec, Canada GUENTHER BODEN, MD, Department of Medicine, Section of Metabolism, Diabetes and Endocrinology, Temple University Hospital, Philadelphia, PA JEAN L. BOLOGNIA, MD, Department of Dermatology, Yale University, New Haven, CT

ANIRBAN BOSE, MD, Section of Nephrology, University of Rochester, Rochester, NY RICHARD C. BOUCHER, MD, Division of Pulmonary Diseases, Department of Medicine, Cystic Fibrosis Research and Treatment Center, University of North Carolina, Chapel Hill, NC SCOTT BOWDEN, PhD, Victorian Infectious Diseases Reference Laboratory (VIDRL), North Melbourne and Department of Microbiology, Monash University, Clayton, Victoria, Australia J. DOUGLAS BREMNER, MD, Department of Psychiatry and Behavioral Sciences, Emory University School of Medicine, Atlanta, GA DAVID A. BRENNER, MD, Department of Medicine, College of Physicians and Surgeons, Columbia University Medical Center, New York, NY ROBERT H. BROWN, JR., MD, PhD, Day Neuromuscular Research Laboratory, Department of Neurology, Harvard Medical School, Massachusetts General Hospital, Boston, MA MARK G. BUCKLEY, PhD, Respiratory Cell and Molecular Biology, University of Southampton School of Medicine, Southampton, UK W. TODD CADE, PT, PhD, Department of Internal Medicine, Washington University School of Medicine, St. Louis, MO STEPHEN C. CANNON, MD, PhD, Department of Neurology, University of Texas Southwestern Medical Center, Dallas, TX SUZANNE B. CASSIDY, MD, Department of Pediatrics, Division of Human Genetics, University of California, Irvine, Irvine, CA MARINA CAVAZZANA-CALVO, INSERM, Department of Biotherapy AP-HP, Hôpital Necker, Paris, France IRIS T. CHAN, MD, PhD, Dana-Farber Cancer Institute, Brigham and Women’s Hospital, Howard Hughes Medical Institute, Harvard Medical School, Boston, MA BERNARD S. CHANG, MD, Division of Neurogenetics, Howard Hughes Medical Institute, Beth Israel Deaconess Medical Center, Department of Neurology, Harvard Medical School, Boston, MA JENNIFER CHAO, MD, PhD, Department of Ophthalmology, University of Southern California, Los Angeles, CA KELLY D. CHASON, BS, Division of Pulmonary and Critical Care Medicine, University of North Carolina, Chapel Hill, NC DENNIS J. CHEEK, PhD, RN, FAHA, Texas Christian University, Harris College of Nursing and Health Sciences and School of Nurse Anesthesia, Fort Worth, TX

xvii

xviii

CONTRIBUTORS

ALVIN J. CHIN, MD, Department of Pediatrics, Division of Cardiology, University of Pennsylvania School of Medicine and Joseph Stokes Research Institute, The Children’s Hospital of Philadelphia, Philadelphia, PA DAVID C. CHRISTIANI, MD, MPH, MS, Pulmonary and Critical Care Unit, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Environmental Health Department, Harvard School of Public Health, Boston, MA ANGELA M. CHRISTIANO, PhD, Departments of Dermatology, and Genetics and Development, College of Physicians and Surgeons, Columbia University, New York, NY JONATHAN A. COHN, MD, Division of Gastroenterology, Duke University, Durham, NC SHEILA COLLINS, PhD, Departments of Psychiatry and Behavioral Sciences, Duke University Medical Center, Durham, NC, and Endocrine Biology Program, Division of Biological Sciences, CIIT Centers for Health Research, Research Triangle Park, NC KEVIN D. COOPER, MD, Department of Dermatology, University Hospitals of Cleveland, Cleveland, OH ELENA CORRADINI, MD, Center for Hemochromatosis and Hereditary Liver Diseases, Department of Internal Medicine, University of Modena and Reggio Emilia, Modena, Italy GILBERT COTE, PhD, Department of Endocrine Neoplasia and Hormonal Disorders, University of Texas M.D. Anderson Cancer Center, Houston, TX GLENN R. CUNNINGHAM, MD, Departments of Medicine and Molecular and Cellular Biology, Baylor College of Medicine, Houston, TX WILLIAM S. DALTON, MD, PhD, Experimental Therapeutics, H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL SAUMYA DAS, MD, Thoracic Aortic Center and Cardiology Division, Massachusetts General Hospital, Boston, MA NANCY E. DAVIDSON, MD, The Breast Cancer Program, The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD JANE C. DAVIES MRCP, MD, Department of Gene Therapy, Imperial College London, National Heart and Lung Institute, London, UK JEAN DELAUNAY, MD, PhD, Service d’Hématologie, d’Immunologie et de Cytogénétique, INSERM, Hôpital de Bicêtre, Le Kremlin-Bicêtre, France FEDERICA DEL MONTE MD, PhD, Cardiovascular Research Center, Massachusetts General Hospital, Boston, MA XINGMING DENG, MD, PhD, University of Florida Shands Cancer Center, University of Florida, Gainesville, FL GORDON DENT, PhD, Institute of Science and Technology in Medicine, Keele University, Keele, UK CHRIS T. DERK, MD, Division of Rheumatology, Department of Medicine, Thomas Jefferson University, Philadelphia, PA LUIS A. DIAZ, MD, Department of Dermatology, School of Medicine, University of North Carolina, Chapel Hill, NC DAVID A. DICHEK, MD, Department of Medicine, Division of Cardiology, University of Washington, Seattle, WA DONALD J. DIPETTE, MD, Department of Medicine, Scott and White, The Texas A&M University System Health Science Center College of Medicine, Temple, TX

SCOTT H. DONALDSON, MD, Cystic Fibrosis Research and Treatment Center, University of North Carolina, Chapel Hill, NC BRIAN J. DRUKER, MD, Oregon Health and Science University Cancer Institute is usually noted as Oregon Health and Science University Cancer Institute, Portland, OR DOUGLAS C. EATON, PhD, Department of Physiology, The Center for Cell and Molecular Signaling, Emory University School of Medicine, Atlanta, GA RENEE E. EDKINS, MA, RN, CCRN, Critical Care: Nursing Practice, Education and Research, University of North Carolina, Chapel Hill, NC JAMES T. ELDER, MD, PhD, Departments of Dermatology and Radiation Oncology, University of Michigan, Ann Arbor, MI PETER B. ERNST, DVM, PhD, Internal Medicine and Microbiology, Division of Gastroenterology and Hepatology, Digestive Health Center of Excellence, University of Virginia, Charlottesville, VA GREGORY J. ESPER, MD, Department of Neurology, Washington University School of Medicine, St. Louis, MO JAMES P. EVANS, MD, PhD, Department of Genetics, University of North Carolina, Chapel Hill, NC JANE WING-SANG FANG, MBBS, MRCP, FAAP, Kinex Pharmaceuticals, Buffalo, NY STEPHEN V. FARAONE, PhD, Medical Genetics Research Program and Department of Psychiatry and Behavioral Sciences; SUNY Upstate Medical University, Syracuse, NY FRANCESCA FERRARA, MD, Center for Hemochromatosis and Hereditary Liver Diseases, Department of Internal Medicine, University of Modena and Reggio Emilia, Modena, Italy ALAIN FISCHER, INSERM, Unité d’Immunologie et d’Hématologie Pédiatriques, Hôpital Necker, Paris, France THOMAS FORCE, MD, Center of Translational Medicine, Jefferson Medical College, Philadelphia, PA BRIAN FOUTY, MD, Division of Pulmonary Medicine, Center for Lung Biology, University of South Alabama School of Medicine, Mobile, AL MASON W. FREEMAN, MD, Lipid Metabolism Unit, Massachusetts General Hospital, Harvard Medical School, Boston, MA ROBERT F. GAGEL, MD, Division of Internal Medicine, University of Texas M.D. Anderson Cancer Center, Houston, TX JOSÉ M. GARCIA, MD, Department of Medicine, Baylor College of Medicine, Houston, TX DANIEL J. GARRY, MD, PhD, Departments of Internal Medicine and Molecular Biology, Univesity of Texas Southwestern Medical Center, Dallas, TX JUAN C. GEA-BANACLOCHE, MD, Infectious Diseases Section, Experimental Transplantation and Immunology Branch, National Cancer Institute, Bethesda, MD DUNCAN M. GEDDES, MD, FRCP, Department Thoracic Medicine, Royal Brompton Hospital, London, UK ROBERT E. GERSZTEN, MD, Program in Cardiovascular Gene Therapy, Cardiovascular Research Center, Center for Immunology and Inflammatory Diseases, Massachusetts General Hospital, Boston, MA

CONTRIBUTORS

D. GARY GILLILAND, MD, PhD, Brigham and Women’s Hospital, Dana-Farber Cancer Institute, Harvard Medical School, Howard Hughes Medical Institute, Boston, MA ROBERT G. GISH, MD, Department of Hepatology and Liver Transplantation, California Pacific Medical Center, San Francisco, CA STEPHEN J. GLATT, PhD, Center for Behavioral Genomics, Department of Psychiatry, University of California, San Diego, La Jolla, CA; and Veterans Medical Research Foundation, San Diego, CA LOWELL A. GOLDSMITH, MD, MPH, Dermatology, School of Medicine, University of North Carolina, Chapel Hill, NC DANIEL R. GOLDSTEIN, MD, Section of Cardiovascular Medicine, Department of Internal Medicine, Yale University, New Haven, CT MICHELLE NG GONG, MD, MS, Pulmonary, Critical Care and Sleep Medicine, Department of Medicine, Mount Sinai School of Medicine, New York, NY MAUREEN M. GOODENOW, PhD, Department of Pathology, Immunology, and Laboratory Medicine, Department of Pediatrics, Division of Immunology and Infectious Diseases, College of Medicine, University of Florida, Gainesville, FL ERYNN S. GORDON, MS, CGC, Research Center for Genetic Medicine, Children's National Medical Center, Washington, DC JACK M. GORMAN, MD, Partners Psychiatry and Mental Health, Harvard Medical School, McLean Hospital, Belmont, MA ALLAN GREEN, D.Phil., The Mary Imogene Bassett Hospital, Bassett Research Institute, Bassett Healthcare, Cooperstown, NY DAVID H. GUTMANN, MD, PhD, Department of Neurology, Washington University School of Medicine, St. Louis, MO SALIMA HACEIN-BEY-ABINA, INSERM, Department of Biotherapy AP-HP, Hôpital Necker, Paris, France ROGER J. HAJJAR, MD, Cardiovascular Research Center, Massachusetts General Hospital, Boston, MA STEPHEN R. HAMMES, MD, PhD, Department of Internal Medicine, University of Texas Medical Center, Dallas, TX SIAN E. HARDING, PhD, Imperial College, National Heart and Lung Institute, London, UK JOHN B. HARLEY MD, PhD, Department of Medicine, University of Oklahoma Health Sciences Center, US Department of Veterans Affairs Medical Center, Arthritis and Immunology Program, Oklahoma Medical Research Foundation, Oklahoma City, OK TOMONOBU HASEGAWA, MD, PhD, Department of Pediatrics, Keio University School of Medicine, Tokyo, Japan THOMAS J. HAWKE, PhD, Department of Pure and Applied Science, York University, Toronto, Ontario, Canada HIRONORI HAYASHI, MD, University of Texas M.D. Anderson Cancer Center, Houston, TX CHRISTINE HEIM, PhD, Department of Psychiatry and Behavioral Sciences, Emory University School of Medicine, Atlanta, GA CAROLYN Y. HO, MD, Cardiovascular Division, Brigham and Women’s Hospital, Boston, MA ERIC P. HOFFMAN, PhD, Children's National Medical Center, Washington, DC

xix

STEPHEN T. HOLGATE, MD, FRCP, DSc, Respiratory Cell and Molecular Biology, University of Southampton School of Medicine, Southampton, UK STEVEN M. HOLLAND, MD, Laboratory of Clinical Infectious Diseases, National Institute of Allergy and Infectious Disease, National Institutes of Health, Bethesda, MD FLORIAN HOLSBOER, MD, PhD, Max Planck Institute of Psychiatry, Munich, Germany YI HUANG, MD, PhD, The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD ERIC M. ISSELBACHER, MD, Thoracic Aortic Center and Cardiology Division, Massachusetts General Hospital, Boston, MA MOIRA R. JACKSON, PhD, Department of Anatomy and Cell Biology, University of Florida College of Medicine, Gainesville, FL JAMES L. JANUZZI, JR., MD, Thoracic Aortic Center and Cardiology Division, Massachusetts General Hospital, Boston, MA SERGIO A. JIMENEZ, MD, Division of Rheumatology, Department of Medicine, Thomas Jefferson University, Philadelphia, PA RICHARD J. JOHNSON, MD, Department of Medicine, Baylor College of Medicine, Houston, TX KEFEI KANG, MD, Department of Dermatology, Case Western Reserve University, Cleveland, OH PETER B. KANG, MD, Program in Genomics, Department of Neurology, Children's Hospital, Harvard Medical School, Boston, MA KHURSHED A. KATKI, PhD, Department of Medicine, Scott and White, The Texas A&M University System Health Science Center College of Medicine, Temple, TX GUR P. KAUSHAL, PhD, Departments of Internal Medicine and Biochemistry and Molecular Biology, Division of Nephrology, University of Arkansas for Medical Sciences, Central Arkansas Veterans Healthcare System, Little Rock, AR TALMADGE E. KING JR., MD, Medical Services, San Francisco General Hospital, Department of Medicine, University of California, San Francisco, CA SAMUEL KLEIN, MD, Center for Human Nutrition, Division of Geriatrics and Nutritional Sciences, Washington University School of Medicine, St. Louis, MO MICHAEL R. KNOWLES, MD, Division of Pulmonary Diseases, Department of Medicine, University of North Carolina, Chapel Hill, NC LAZAROS K. KOCHILAS, MD, Department of Pediatrics, Division of Pediatric Cardiology, Hasbro Children's Hospital/RIH, Brown University Medical School, Providence, RI JAMES J. KOHLER, PhD, Department of Pathology, College of Medicine, Emory University, Atlanta, GA PETER KOPP, MD, Division of Endocrinology, Metabolism and Molecular Medicine, Feinberg School of Medicine, Northwestern University, Chicago, IL BRUCE R. KORF, MD, PhD, Department of Genetics, University of Alabama, Birmingham, AL

xx

CONTRIBUTORS

LOUIS M. KUNKEL, PhD, Program in Genomics, Howard Hughes Medical Institute, Children's Hospital, Harvard Medical School, Boston, MA FADI G. LAKKIS, MD, Thomas E. Starzl Transplantation Institute, University of Pittsburgh, Pittsburgh, PA DOLORES J. LAMB, PhD, Scott Department of Urology and Department of Molecular and Cellular Biology, Baylor College of Medicine, Houston, TX HUI Y. LAN, MD, PhD, Department of Medicine, Baylor College of Medicine, Houston, TX CAROL A. LANGE, PhD, University of Minnesota Cancer Center, Minneapolis, MN KAREN A. LAPIDOS, PhD, Department of Molecular Genetics and Cell Biology, The University of Chicago, Chicago, IL JOHNSON YIU-NAM LAU, MBBS, MD, FRCP, Managing Director, Roth Capital Partners, Newport Beach, CA PHILIPPE LEBOULCH, Harvard Medical School and Genetics Division, Brigham and Women’s Hospital, Boston, MA VIRGINIA M.-Y. LEE, PhD, Center for Neurodegenerative Disease Research, Department of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, Philadelphia, PA DONALD Y.M. LEUNG, MD, PhD, Department of Pediatrics, University of Colorado Health Sciences Center, Pediatric Allergy/Immunology Division, National Jewish and Medical Research Center, Denver, CO MICHAEL A. LEVINE, MD, The Children's Hospital at The Cleveland Clinic, Department of Pediatrics, The Cleveland Clinic Lerner College of Medicine, Case Western Reserve University, Cleveland, OH CLIVE J. LEWIS, MD, Imperial College, National Heart and Lung Institute, London, UK NING LI, PhD, Department of Dermatology, School of Medicine, University of North Carolina, Chapel Hill, NC MIN LIU, MD, PhD, Department of Pathology, University of Cincinnati Medical Center, Cincinnati, OH ZHI LIU, PhD, Department of Dermatology, Department of Microbiology and Immunology, School of Medicine, University of North Carolina, Chapel Hill, NC KIRK C. LO, MD, CM, Male Reproductive Medicine and Surgery, Scott Department of Urology, Baylor College of Medicine, Houston, TX STEPHEN LOCARNINI, MD, PhD, Victorian Infectious Diseases Reference Laboratory (VIDRL), North Melbourne and Department of Microbiology, Monash University, Clayton, Victoria, Australia BARRY LONDON, MD, PhD, Cardiovascular Institute, University of Pittsburgh, Pittsburgh, PA TAMMY L. LOUCKS, MPH, Magee-Women's Research Institute, Pittsburgh, PA WILLIAM L. LOWE, JR., MD, Division of Endocrinology, Metabolism, and Molecular Medicine, Feinberg School of Medicine, Northwestern University, Chicago, IL JAY LOZIER, MD, PhD, Center for Biologics Evaluation and Research, Food and Drug Administration, Rockville, MD JAMES R. LUPSKI, MD, PhD, Department of Molecular and Human Genetics, Department of Pediatrics, Baylor College of Medicine, Texas Children’s Hospital, Houston, TX

HE-PING MA, MD, Department of Medicine, University of Alabama at Birmingham School of Medicine, Birmingham, AL CALUM A. MACRAE, MD, Cardiology Division and Cardiovascular Research Center, Massachusetts General Hospital, Harvard Medical School, National Heart, Lung and Blood Institute’s Framingham Heart Study, Boston, MA TERRY MAGNUSON, PhD, Department of Genetics, University of North Carolina, Chapel Hill, NC FRANK MALDARELLI, MD, PhD, HIV Drug Resistance Program, National Cancer Institute, National Institutes of Health, Bethesda, MD BELA MALIK, PhD, Department of Physiology, The Center for Cell and Molecular Signaling, Emory University School of Medicine, Atlanta, GA MARCO MARCELLI, MD, Departments of Medicine and Molecular and Cellular Biology, Baylor College of Medicine, Division of Endocrinology, Michael E. DeBakey VA Medical Center, Houston, TX CINDY M. MARTIN, MD, Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, TX W. STRATFORD MAY, JR., MD, PhD, University of Florida, Shands Cancer Center, University of Florida, Gainesville, FL SHAWN E. MCCANDLESS, MD, Department of Genetics, Case Western Reserve University, Center for Human Genetics, University Hospitals of Cleveland, Cleveland, OH ELIZABETH A. MCGEE, MD, Magee-Women's Research Institute, Pittsburgh, PA ELIZABETH M. MCNALLY, MD, PhD, Department of Medicine, Section of Cardiology, Department of Human Genetics, The University of Chicago, Chicago, IL MICHAEL J. MCPHAUL, MD, Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, TX DEBORAH C. MEDOFF, PhD, Maryland Psychiatric Research Center, Department of Psychiatry, University of Maryland School of Medicine, Baltimore, MD SHLOMO MELMED, MD, Academic Affairs, Cedars-Sinai Medical Center, David Geffen School of Medicine at UCLA, Los Angeles, CA WILLIAM E. MITCH, MD, Division of Nephrology, Baylor College of Medicine, Houston, TX CHRISTOPHER B. MIZELLE, MD, School of Medicine, University of North Carolina, Chapel Hill, NC MASASHI MIZOKAMI, MD, PhD, Department of Clinical Molecular Informative Medicine, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan ANWAR N. MOHAMED, MD, Cancer Cytogenetic Laboratory, Department of Pathology, Wayne State University School of Medicine, Detroit, MI JEFFERY D. MOLKENTIN, PhD, Department of Pediatrics, Division of Molecular Cardiovascular Biology, Children’s Hospital Medical Center, Cincinnati, OH STEPHAN MOLL, MD, Department of Medicine, Division of Hematology-Oncology, University of North Carolina, Chapel Hill, NC

CONTRIBUTORS

KHEDOUDJA NAFA, PhD, Memorial Sloan Kettering Cancer Center, New York, NY SATYA NARAYAN, PhD, Department of Anatomy and Cell Biology, UF Shands Cancer Center, College of Medicine, University of Florida, Gainesville, FL CHARLES B. NEMEROFF, MD, PhD, Department of Psychiatry and Behavioral Sciences, Emory University School of Medicine, Atlanta, GA ERIC J. NESTLER, MD, PhD, Department of Psychiatry and Center for Basic Neuroscience, The University of Texas Southwestern Medical Center, Dallas, TX MARIA I. NEW, MD, Adrenal Steroid Disorders Program, Department of Pediatrics, Mount Sinai School of Medicine, The Mount Sinai Hospital, New York, NY WILLIAM G. NEWMAN, MD, PhD, University of Manchester, UK, St. Mary's Hospital, Manchester, UK PEADAR G. NOONE, MD, Cystic Fibrosis Research and Treatment Center, Department of Medicine, Division of Pulmonary Diseases, University of North Carolina, Chapel Hill, NC CHRISTOPHER J. O’DONNELL, MD, MPH, Cardiology Division, Massachusetts General Hospital, Harvard Medical School and the National Heart, Lung and Blood Institute's Framingham Heart Study, Boston, MA ERIC N. OLSON, PhD, Department of Molecular Biology, University of Texas Southwestern Medical Center, Dallas, TX MARCELO PÁEZ-PEREDA, PhD, Affectis Pharmaceuticals and Max Planck Institute of Psychiatry, Munich, Germany KEITH L. PARKER, MD, PhD, Department of Internal Medicine, Division of Endocrinology and Metabolism, University of Texas Southwestern Medical Center, Dallas, TX KEVIN A. PELPHREY, PhD, Department of Psychological and Brain Sciences, Duke University, Durham, NC ALEXANDER E. PERL, MD, Division of Hematology-Oncology, Department of Medicine, University of Pennsylvania, Philadelphia, PA JOHN A. PHILLIPS III, MD, Division of Medical Genetics, Department of Pediatrics, Vanderbilt University School of Medicine, Nashville, TN ANTONELLO PIETRANGELO, MD, PhD, Center for Hemochromatosis and Hereditary Liver Diseases, Department of Internal Medicine, University of Modena and Reggio Emilia, Modena, Italy JOSEPH PIVEN, MD, Departments of Psychiatry and Pediatrics, Neurodevelopmental Disorders Research Center, University of North Carolina, Chapel Hill, NC SHARON E. PLON, MD, PhD, Departments of Pediatrics, and Molecular and Human Genetics, Baylor College of Medicine, Houston, TX DIDIER PORTILLA, MD, Department of Internal Medicine, Division of Nephrology, University of Arkansas for Medical Sciences, Central Arkansas Veterans Healthcare System, Little Rock, AR AMY POTTER, MD, Adult and Pediatric Endocrinology, Division of Endocrinology, Department of Pediatrics, Division of Endocrinology and Metabolism, Department of Medicine, Vanderbilt University School of Medicine, Nashville, TN

xxi

S. RUSS PRICE, PhD, Renal Division, Department of Medicine, Emory University, Atlanta, GA CHARMIAN A. QUIGLEY, MBBS, Department of Pediatrics, Indiana University School of Medicine, Lilly Research Laboratories, Indianapolis, IN WILLIAM E. RAINEY, PhD, Department of Physiology, Medical College of Georgia, Augusta, GA GIUSEPPE REMUZZI, MD, Mario Negri Institute for Pharmacological Research, and Division of Nephrology and Dialysis, Azienda Ospedaliera, Ospedali Riuniti di Bergamo, Bergamo, Italy KERRY J. RESSLER, MD, PhD, Department of Psychiatry and Behavioral Sciences, Emory University School of Medicine, Atlanta, GA GABRIELE RICHARD, MD, Department of Dermatology and Cutaneous Biology, Jefferson Medical College, Thomas Jefferson University, Philadelphia, PA FRANZISKA RINGPFEIL, MD, Department of Dermatology and Cutaneous Biology, Jefferson Medical College, Thomas Jefferson University, Philadelphia, PA ANTONIO M. RISITANO, PhD, Dipartimento di Biochimica e Biotecnologie Mediche, Federico II University of Naples, Naples, Italy JACQUES ROBIDOUX, PhD, CIIT Centers for Health Research, Endocrine Biology Program, Division of Biological Sciences, Research Triangle Park, NC DAVID M. RODMAN, MD, Division of Pulmonary Sciences and Critical Care Medicine, University of Colorado Health Sciences Center, Denver, CO ANTHONY ROSENZWEIG, MD, Program in Cardiovascular Gene Therapy, Cardiovascular Research Center, Center for Immunology and Inflammatory Diseases, Massachusetts General Hospital, Boston, MA BRUNO ROTOLI, MD, Dipartimento di Biochimica e Biotecnologie Mediche, Federico II University of Naples, Naples, Italy DAVID S. RUBENSTEIN, MD, PhD, Department of Dermatology, School of Medicine, University of North Carolina, Chapel Hill, NC DEBORAH C. RUBIN, MD, Division of Gastroenterology, Departments of Medicine and Molecular Biology and Pharmacology, Washington University School of Medicine, St. Louis, MO DEEPAK M. SAMPATHU, PhD, Center for Neurodegenerative Disease Research, Department of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, Philadelphia, PA JEFF M. SANDS, MD, Renal Division, Department of Medicine, Department of Physiology, Emory University School of Medicine, Atlanta, GA AMR H. SAWALHA, MD, Department of Medicine, University of Oklahoma Health Sciences Center, US Department of Veterans Affairs Medical Center, Arthritis and Immunology Program, Oklahoma Medical Research Foundation, Oklahoma City, OK THOMAS E. SCAMMELL, MD, Department of Neurology, Beth Israel Deaconess Medical Center, Boston, MA JULIE V. SCHAFFER, MD, Department of Dermatology, New York Univesity School of Medicine, New York, NY

xxii

CONTRIBUTORS

MARCUS J. SCHULTZ, MD, PhD, Department of Intensive Care Medicine, Laboratory of Experimental Intensive Care and Anesthesiology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands BRAHM H. SEGAL, MD, Department of Medicine, Division of Infectious Diseases, The University at Buffalo, The State University of New York (SUNY), Roswell Park Cancer Institute, Buffalo, NY CHRISTINE E. SEIDMAN, MD, Cardiovascular Division, Brigham and Women’s Hospital, Howard Hughes Medical Institute and Department of Genetics, Harvard Medical School, Boston, MA SUDHIR V. SHAH, MD, Department of Internal Medicine, Division of Nephrology, University of Arkansas for Medical Sciences, Central Arkansas Veterans Healthcare System, Little Rock, AR KENNETH H. SHAIN, MD, PhD, Experimental Therapeutics, H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL STEVEN D. SHAPIRO, MD, Pulmonary and Critical Care, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA THOMAS C. SHEA, MD, Division of Hematology and Oncology, Bone Marrow and Stem Cell Transplant Program, Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC SARAH SHEFELBINE, MD, Department of Otolaryngology, University of Florida, Gainesville, FL LINMARIE SIKICH, MD, Department of Psychiatry, University of North Carolina, Chapel Hill, NC JONATHAN W. SIMONS, MD, Winship Cancer Institute, Emory University, Atlanta, GA DONALD SMALL, MD, PhD, Departments of Pediatrics, Oncology, Cellular and Molecular Medicine, Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Baltimore, MD PETER H. ST GEORGE-HYSLOP, MD, FRCPC, FRS, Centre for Research in Neurodegenerative Diseases, University of Toronto, Department of Medicine, Division of Neurology, University Health Network, Toronto, Ontario, Canada AMY STEIN, MD, Department of Dermatology, School of Medicine, University of North Carolina, Chapel Hill, NC GORDON W. STEWART, MD, FRCP, Department of Medicine, The Rayne Institute, London, UK STEPHEN M. STRITTMATTER, MD, PhD, Department of Neurology, Yale University School of Medicine, New Haven, CT STEPHEN P. SUGRUE, PhD, Department of Anatomy and Cell Biology, University of Florida College of Medicine, Gainesville, FL GREGORY M. SULLIVAN, MD, Department of Psychiatry, Columbia University College of Physicians and Surgeons, New York, NY MARIJA TADIN-STRAPPS, PhD, Department of Dermatology, College of Physicians and Surgeons, Columbia University, New York, NY CAROL A. TAMMINGA, MD, Department of Psychiatry, Univesity of Texas Southwestern Medical Center, Dallas, TX

STEPHEN J. TAPSCOTT, MD, PhD, Fred Hutchinson Cancer Research Center, Department of Neurology, University of Washington, Seattle, WA GUNVANT THAKER, MD, Maryland Psychiatric Research Center, Department of Psychiatry, University of Maryland School of Medicine, Baltimore, MD RAJESH V. THAKKER, MD, FRCP, FRCPath, FMedSci, Nuffield Department of Clinical Medicine, University of Oxford, Academic Endocrine Unit, Oxford Centre for Diabetes, Endocrinology and Metabolism (OCDEM), Churchill Hospital, Headington, Oxford, UK SWEE LAY THEIN, DSc, FRCP, FRCPath, FMedSci, Department of Haematological Medicine, Guy's, King's and St. Thomas' School of Medicine, London, UK NANCY E. THOMAS, MD, PhD, Department of Dermatology, University of North Carolina, Chapel Hill, NC ADRIAN J. THRASHER, Molecular Immunology Unit, Institute of Child Health, London, UK STEPHEN L. TILLEY, MD, Department of Medicine, Division of Pulmonary and Critical Care Medicine, University of North Carolina, Chapel Hill, NC BRUCE C. TRAPNELL, MD, MS, Department of Pediatrics, Division of Pulmonary Biology, Cincinnati Children’s Hospital Medical Center, University of Cincinnati, Cincinnati, OH PATRICK TSO, PhD, Department of Pathology, University of Cincinnati Medical Center, Cincinnati, OH MING T. TSUANG, MD, PhD, Center for Behavioral Genomics, Department of Psychiatry, University of California, San Diego, La Jolla, CA; Veterans Affairs San Diego Healthcare System, San Diego, CA; and Harvard Institute of Psychiatric Epidemiology and Genetics, Harvard Departments of Epidemiology and Psychiatry, Boston, MA J. ERIC TURNER, MD, Division of Hematology and Oncology, University of North Carolina, Chapel Hill, NC JOUNI UITTO, MD, PhD, Department of Dermatology and Cutaneous Biology, Jefferson Medical College, Thomas Jefferson University, Philadelphia, PA TOM VAN DER POLL, MD, PhD, Laboratory of Experimental Internal Medicine, Department of Infectious Diseases, Tropical Medicine and AIDS, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands MARCIA VAN RIPER, PhD, RN, School of Nursing, University of North Carolina, Carolina Center for Genome Sciences, Chapel Hill, NC CHRISTOPHER A. WALSH, MD, PhD, Division of Neurogenetics, Howard Hughes Medical Institute, Beth Israel Deaconess Medical Center, Department of Neurology, Harvard Medical School, Boston, MA THOMAS H. WASSINK, MD, Department of Psychiatry, Carver College of Medicine, University of Iowa, Iowa City, IA SUSAN E. WERT, PhD, Department of Pediatrics, Division of Pulmonary Biology, Cincinnati Children’s Hospital Medical Center, University of Cincinnati, Cincinnati, OH MATTHEW T. WHEELER, MD, PhD, Department of Molecular Genetics and Cell Biology, The University of Chicago, Chicago, IL DAVID C. WHITCOMB, MD, PhD, Division of Gastroenterology, Hepatology and Nutrition, University of Pittsburgh, Pittsburgh, PA

CONTRIBUTORS

GILBERT C. WHITE, II, MD, Richard H. and Sarah E. Aster Chair for Medical Research, Blood Center of Wisconsin, Blood Research Institute, and Biochemistry, and Pharmacology and Toxicology, Medical College of Wisconsin JEFFREY A. WHITSETT, MD, Department of Pediatrics, Division of Pulmonary Biology, Cincinnati Children’s Hospital Medical Center, University of Cincinnati, Cincinnati, OH CHRISTOPHER S. WILCOX, MD, PhD, Division of Nephrology and Hypertension, Cardiovascular Kidney Institute, Georgetown University, Washington, DC R. SANDERS WILLIAMS, MD, Department of Medicine, Division of Cardiology, Duke University Medical Center, Durham, NC ROBERT W. WILLIAMS, PhD, Department of Anatomy and Neurobiology, Department of Pediatrics, Center for Genomics and Bioinformatics, University of Tennessee, Memphis, TN LORNA M. WILLIAMSON, BSc, MD, FRCP, FRCPath, Department of Haematology, University of Cambridge/ National Blood Service, Cambridge, UK JOHN S. WITTE, PhD, Departments of Epidemiology and Biostatistics, and Urology, University of California, San Francisco, San Francisco, CA ROBERT C. WILSON, PhD, Department of Pediatrics, Mount Sinai School of Medicine, New York, NY J. TIM WRIGHT, DDS, MS, Department of Pediatric Dentistry, School of Dentistry, University of North Carolina, Chapel Hill, NC

xxiii

TERESA L. WRIGHT, MD, Department of Medicine, University of California, San Francisco, Gastroenterology Division, Veterans Affairs Medical Center, San Francisco, CA LEI XIAO, PhD, Department of Anatomy and Cell Biology, University of Florida, Gainesville, FL ANTHONY T. YACHNIS, MD, MS, Department of Pathology and Laboratory Medicine, College of Medicine, University of Florida, Gainesville, FL ZHEN YAN, PhD, Department of Medicine, Division of Cardiology, Duke University Medical Center, Durham, NC KEVIN E. YARASHESKI, PhD, Department of Internal Medicine, Washington University School of Medicine, St. Louis, MO MAIMOONA ZARIWALA, PhD, Division of Pulmonary Diseases, Department of Medicine, University of North Carolina, Chapel Hill, NC ANDREW R. ZINN, MD, PhD, McDermott Center for Human Growth and Development, Department of Internal Medicine, University of Texas Southwestern Medical School, Dallas, TX HUDA Y. ZOGHBI, MD, Departments of Pediatrics, Neurology, Molecular and Human Genetics, Neuroscience, Baylor College of Medicine, Howard Hughes Medical Institute, Houston, TX CARLA ZOJA, PhD, Mario Negri Institute for Pharmacological Research, Bergamo, Italy

Color Plates

Plate 1 (Fig. 2, Chapter 4). Fluorescent in situ hybridization of a pediatric leukemia sample demonstrating that one chromosome 5 contains a deleted segment that includes the EGR1 gene. Plate 2 (Fig. 1, Chapter 10). The left side of the figure depicts fetal circulation in the human. The right side of the figure depicts neonatal circulation in the human. Plate 3 (Fig. 2, Chapter 10). Dosage sensitive role of Tbx1 in the etiology of cardiovascular defects in mice. Plate 4 (Fig. 1A–C, Chapter 11). Gross pathological specimens of a heart with (A) hypertrophic cardiomyopathy (HCM) and (C) dilated cardiomyopathy (DCM). Note the marked increased in left ventricular hypertrophy (HCM) and chamber dimensions (DCM) as compared with (B) the normal heart. Plate 5 (Fig. 2, Chapter 11). Histopathology of distinct human cardiomyopathies revealed by hematoxylin and eosin staining. Plate 6 (Fig. 3, Chapter 15). The palmar and tuberoeruptive xanthomas are classically seen in patients with increased concentrations of intermediate density lipoproteins whereas the eruptive xanthomas are typically found in patients with massive serum triglyceride elevations resulting from excess chylomicrons and/or very lowdensity lipoproteins. Plate 7 (Fig. 5, Chapter 22). Increased immunostaining for the caspase 3 p85 cleavage product of poly ADP-ribose polymerase in asthmatic mucosal biopsies and bronchoalveolar lavage epithelial cells. Plate 8 (Fig. 4, Chapter 24). Plexiform lesion. The end-stage of pulmonary hypertension is associated with the formation of occlusive intimal lesions. Plate 9 (Fig. 1, Chapter 29). Development of granulomatous inflammation. Antigen presenting cells interact with CD4+ T lymphocytes by presenting antigenic peptides, bound to major histocompatibility complex molecules, to T-cell receptors.

Plate 10 (Fig. 2A–D, Chapter 30). Histopathology of surfactant abnormalities found in the lungs of human patients with (A) mutations in the SFTPB gene, (B) mutations in the SFTPC gene, (C) mutations in the ABCA3 gene, (D) and antibodies to granulocyte macrophage-colony-stimulating factor. Plate 11 (Fig. 1, Chapter 66). Activation of MHC expression in fibroblasts expressing exogenous MyoD. Plate 12 (Fig. 6, Chapter 66). Diagrammatic representation of somite maturation. Plate 13 (Fig. 7, Chapter 66). Expression pattern of a myogenin-lacZ transgene in an 11.5-day mouse embryo. The myogenin promoter was linked to a β-galactosidase reporter gene and introduced into transgenic mice. Plate 14 (Fig. 1, Chapter 84). Global epidemiology of HIV-1 subtypes and estimated number of infected individuals (in millions) at the end of 2002 according to the International AIDS Vaccine Initiative and the Joint United Nations Programme on HIV/AIDS. Plate 15 (Fig. 1, Chapter 108). Features of the hairless phenotype in humans and mice. Plate 16 (Fig. 1, Chapter 119). Characteristic neuropathological features of transmissible spongiform encephalopathies. Plate 17 (Fig. 1, Chapter 129). Key neural circuits of addiction. Dotted lines indicate limbic afferents to the nucleus accumbens. Arrows represent efferents from the nucleus accumbens thought to be involved in drug reward.

xxv

GENETICS SECTION EDITORS:

TERRY MAGNUSON AND JAMES P. EVANS

I

Abbreviations I. MOLECULAR GENETICS AAV ABGC ADR AMKL AS BBS BNSF BWS CD CEERs CF cM DPD DS ELSI EM F1 FAP FDA FISH HD HER2 HGP HH htSNP IBD

IGF IND LGK LOD MCMC MDR MELAS

insulin-like growth factor investigational new drug Lander-Green-Kruglyak likelihood of the odds Monte-Carlo Markov Chain multidrug resistance mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes MEN multiple endocrine neoplasia MEN1 multiple endocrine neoplasia type 1 MEN2 multiple endocrine neoplasia type 2 MTC medullary thyroid carcinoma mtDNA mitochondrial DNA NHGRI National Human Genome Research Institute PKU phenylketonuria PM poor metabolizer PWS Prader–Willi syndrome RP retinitis pigmentosa SNP single-nucleotide polymorphism TDT transmission/disequilibrium test UC ulcerative colitis UPD uniparental disomy URM ultra rapid metabolizer UV ultraviolet WAGR Wilm’s tumor, aniridia, genital–urinary abnormalities, and mental retardation XP xeroderma pigmentosum

adeno-associated virus American Board of Genetics Counseling adverse drug reaction acute megakaryocytic leukemia Angelman syndrome Bardet-Biedl syndrome Burlington Northern Santa Fe Railroad Beckwith-Wiedemann syndrome Crohn’s disease Centers of Excellence in ELSI Research cystic fibrosis centimorgan dihydropyrimidine dehydrogenase Down syndrome ethical, legal, and social implications extensive metabolizer filial 1 familial adenomatous polyposis Food and Drug Administration fluorescent in situ hybridization Huntington disease human epidermal receptor 2 Human Genome Project hemihyperplasia haplotype tagging single-nucleotide polymorphism inflammatory bowel disease

2

1 Mendelian Inheritance BRUCE R. KORF SUMMARY

He originated the term “inborn errors of metabolism” to describe a set of disorders in which specific biochemical pathways were deranged, leading to accumulation of toxic substrates or deficiency of end products. He recognized that these conditions are familial and behave as Mendelian recessive traits. A recessive trait is only expressed in a homozygous individual who inherits a mutant allele from both parents (Fig. 1-1A). The parents are heterozygous carriers, who are asymptomatic because of the action of the dominant allele. A couple consisting of two carriers faces a 25% risk of transmission of homozygosity to any offspring. The basis of recessive inheritance of inborn errors of metabolism is that the responsible genes encode enzymes required to catalyze specific biochemical reactions. Enzymes function in a catalytic manner, so the 50% level of activity that may occur in a heterozygote is sufficient to complete the reaction and thereby avoid the phenotype. Only a homozygote will lack sufficient activity to manifest the disorder. An example is the disorder phenylketonuria, which is because of mutation in the gene that encodes the enzyme phenylalanine hydroxylase, required to convert phenylalanine to tyrosine. Homozygotes accumulate phenylalanine to toxic levels, and also have a deficiency of phenylalanine metabolites such as dopamine and melanin. Children with this disorder detected by newborn screening can be spared the severe developmental impairment of this disorder by treatment with a low-phenylalanine diet. Carrier parents are asymptomatic, but have a 25% risk of additional affected children. Dominant traits are expressed in both homozygous and heterozygous individuals (Fig. 1-1B). If the trait is rare, most affected individuals will be heterozygous. Moreover, many dominantly inherited medical conditions are lethal in the homozygous state, technically indicating that they are not “true” dominants. An individual who is heterozygous for a dominant trait has a 50% chance of passing either allele to any offspring. A prototypical autosomal-dominant disorder is Marfan syndrome, resulting from mutation in the connective tissue protein fibrillin. Affected individuals are tall, lanky, and experience complications such as joint dislocation, lens subluxation, and aortic aneurysms because of weakness of connective tissue. The disorder is compatible with survival to reproductive age, and affected individuals have a 50% risk of transmitting the mutant allele to any offspring. The molecular basis for dominance is reviewed later. SEX LINKAGE Sex determination occurs by inheritance of two X chromosomes in females or an X and a Y in males. The Y carries a limited repertoire of genes, including those involved

The basic patterns of genetic transmission in humans have been known for about a century, but are now coming to be understood at the molecular level. In addition to classical dominant, recessive, and sex-linked inheritance, more complex patterns have also been identified. These include maternal transmission of traits encoded in the mitochondrial genome, digenic traits determined by two distinct genes, and genomic imprinting. It is becoming clear that both rare and common genetic traits are determined by a complex interaction of multiple genetic and nongenetic factors. Key Words: Digenic; dominant; expressivity; imprinting; Mendelian; mitochondrial inheritance; penetrance; recessive; X-linked.

INTRODUCTION The existence of human traits that are transmitted from generation to generation in accordance with Mendel’s laws was first recognized early in the 20th century. Understanding the mechanisms that underlie Mendelian inheritance has unfolded over the ensuing decades, and forms the basis for knowledge of human genetics. With increasing sophistication it has become clear that the seemingly straightforward rules of inheritance—for example, dominance and recessiveness—are in fact complex. With more nuanced understanding, however, comes recognition of genetic principles that have a critical role in diagnosis and counseling. This chapter reviews the basics of Mendelian inheritance and explores how insights in molecular genetics are both explicating and changing views of these fundamental principles.

PATTERNS OF GENETIC TRANSMISSION The patterns of Mendelian transmission ensue from the fact that humans are diploid organisms, inheriting a complete set of genes from each parent. The two individual copies of a specific gene are referred to as alleles. The alleles on homologous chromosomes segregate at meiosis and new combinations are paired together on fertilization. The specific alleles at a locus comprise the genotype; the physical characteristic that results from action of the alleles is the phenotype. DOMINANT AND RECESSIVE INHERITANCE The first instance of Mendelian transmission in humans was recognized by Archibald Garrod, working in the early years of the 20th century. From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

3

4

SECTION I / GENETICS

Figure 1-1 Pedigrees depicting autosomal-recessive (A), autosomal-dominant (B), X-linked-recessive (C), and X-linked-dominant with male lethality (D). By convention, squares denote males, circles females, and filled-in symbols are individuals who manifest a phenotype. Half-filled symbols in the recessive pedigree are carriers, and females with dots in the X-linked recessive pedigree are heterozygotes.

in testes determination and spermatogenesis. The X carries a larger set, most of which lack counterparts on the Y. Therefore, for these genes, females have two alleles, but males only one. There are regions at the two ends of the X chromosome where homologous loci exist on the Y. These are referred to as pseudoautosomal. For the other loci found only on the X males are said to be hemizygous. Gene dosage is finely controlled, and, therefore, a mechanism exists to compensate for the dosage differences in males and females for X-linked genes. Most genes on the X chromosome are inactivated on one of the two X’s in female cells early in development. The choice of X to be inactivated is random, but once an X is “turned off,” that chromosome remains off in all descendents of a particular cell. Some genes, especially those in the pseudoautosomal regions escape inactivation. The molecular basis for X chromosome inactivation is becoming understood. It includes selection of the chromosome to be inactivated by an RNA molecule encoded by an X-linked gene called Xist and subsequent methylation of DNA on the inactive chromosome. There are few Y-linked traits of medical significance. A Y-linked gene is transmitted from a male to all his sons and none of his daughters. X-linked traits are transmitted by a heterozygous female to half her offspring; a hemizygous male passes the gene to all his daughters and none of his sons (Fig. 1-1C). The concepts of dominance and recessiveness are meaningless when applied to genes on the X chromosome, which are subject to inactivation, because only one allele is expressed in any cell in either a male or a female.

Whether or not a trait is expressed in a heterozygous female depends on whether expression of the mutant gene in approx 50% of cells is sufficient to cause the phenotype, or whether expression of the wild-type allele in 50% is insufficient. Classic “X-linkedrecessive” traits such as Duchenne muscular dystrophy and hemophilia A tends not to lead to a phenotype in heterozygous females, unless X chromosome inactivation has somehow been skewed toward inactivation of the wild-type allele. “X-linked-dominant” traits, such as hypophosphatemic rickets, are expressed in both sexes. X-linked traits such as Rett syndrome or incontinentia pigmenti are lethal in hemizygous males and, therefore, only are seen in heterozygous females (Fig. 1-1D). MATERNAL TRANSMISSION Although not a “Mendelian” pattern, maternal inheritance is another form of single gene transmission. Maternal inheritance applies to a set of genes found on the 16.5-kb circular double-stranded DNA molecules found within mitochondria. These encode 13 peptides involved in mitochondrial oxidative phosphorylation, as well as a set of transfer RNAs and ribosomal RNAs. There are thousands of DNA molecules within mitochondria in every cell. If there is a mutation in some, the cell is said to be heteroplasmic (Fig. 1-2A). Because mitochondrial DNA molecules segregate at random during cell division, the proportion of mutant and wild-type DNA molecules can vary widely between cells. Mitochondrial DNA mutations tend to interfere with cellular energy production. Because of heteroplasmy there can be a wide range of phenotypic effects, depending on the proportion of mutant mitochondria in different tissues.

CHAPTER 1 / MENDELIAN INHERITANCE

5

Figure 1-2 (A) Mutant and wild-type mitochondrial DNA may coexist in a cell, referred to as heteroplasmy. Mitochondrial DNA molecules segregate passively when a cell divides, so daughter cells may differ in their proportions of mutant and wild-type mitochondrial DNA. (B) Essentially, all the mitochondrial DNA is maternally transmitted. Hence, a female with a mitochondrial disorder will transmit it to all her offspring, whereas a male will not transmit the trait. Offspring may differ in their degree of expression of the phenotype because of heteroplasmy.

Most, if not all, mitochondria are transmitted through the oocyte. As a result, a female with a mitochondrial mutation will pass it to all of her offspring, whereas a male will not transmit it at all (Fig. 1-2B). Once again, however, heteroplasmy will account for variability, in this case among members of a sibship. This creates a challenge in recurrence risk counseling for mitochondrial disorders, because the likelihood that an offspring will inherit sufficient mutant DNA molecules to produce a phenotype cannot be predicted.

COMPLEXITIES OF MENDELIAN TRAITS Although single gene traits are transmitted in accordance with Mendel’s laws, a number of phenomena may lead to deviation from the expected segregation ratios. These include nonpenetrance, new mutation, mosaicism, anticipation, imprinting, and digenic inheritance. PENETRANCE AND EXPRESSIVITY Individuals who have the genotype associated with a particular phenotype yet do not display the phenotype are said to be nonpenetrant. For a dominant trait this may lead to a skipped generation (i.e., a trait is seen in a child and a grandparent, but the parent is not affected). Some phenotypes display age-dependent penetrance, so the probability of phenotypic expression increases with age. This is typical of disorders such as Huntington disease or adult polycystic kidney disease. Penetrance is an all-or-none phenomenon for an individual; the phenotype is either present or not at a particular time. Penetrance should not be confused with expressivity, which refers to the degree of phenotypic expression from individual to individual. The possibility of nonpenetrance or of age-dependent penetrance needs to be considered when counseling an individual at risk of a dominant trait. The lack of phenotype does not necessarily preclude one from occurring at a later age or exclude the possibility of transmission of the trait to an offspring. MUTATION AND MOSAICISM Sporadically affected individuals with a dominant or an X-linked trait may occur because of a new mutation in the sperm or egg cell. Neither parent will be found to carry the trait, and the affected child will be the first affected member of the family. That child, however, will be at risk of transmitting the trait to his or her offspring. Mutation rates vary among different loci, usually hovering in the range of 10–4 to 10–6/gamete/generation. Few distinct risk factors have been

identified, although there is a slight increase in the risk of mutation with advancing paternal age. Mutation may occur in somatic cells as well as in the germline. Somatic mutation during early development results in somatic mosaicism, in which an individual has a mixture of mutant and nonmutant cells. This may manifest as milder expression of a phenotype, or as expression of the phenotype in a limited region of the body. A dramatic example is segmental neurofibromatosis, where café-au-lait spots and neurofibromas may be restricted to part of the body. Germline mosaicism results in multiple sperm or egg cells carrying a new mutation. A parent with germline mosaicism can have multiple affected children despite not carrying the mutation in somatic cells. ANTICIPATION It has long been noted that in some families with particular dominant traits, severity increases, and age of onset decreases, from generation to generation. This phenomenon is referred to as anticipation. Although initially thought to be an artifact owing to bias of ascertainment, it is now known to be a real event that is the signature of a specific type of mutation, the triplet repeat expansion (Fig. 1-3). A number of genes include repeated triplets of bases, such as CAGCAGCAG. The exact number of repeats is a heritable polymorphism, as there is no phenotypic impact regardless of the number, up to a point. Individuals with mutations, however, have expanded numbers of repeats that lead to aberrant gene expression. Anticipation results from two characteristics of repeat expansion mutations. First, the larger the number of repeats, the more severe and earlier is the onset of the disorder. Second, the larger the repeat size, the more unstable it is, creating risk of further expansion in the next generation. The expansion, therefore, increases from generation to generation, leading to anticipation. Disorders associated with triplet repeat expansion tend to affect the nervous system, and include Huntington disease, fragile X syndrome, myotonic dystrophy, spinocerebellar ataxia, and others. IMPRINTING A subset of genes is expressed only from the maternal or paternal allele, but not both, and is referred to as imprinting. The “imprint” that identifies an allele as being of maternal or paternal origin is “erased” each generation. For example, if it is the maternally derived allele that is expressed, a maternally inherited allele will be turned on in a male, but will be turned off when he transmits it to the next generation. It appears that only

6

SECTION I / GENETICS

Figure 1-3 A prototypical gene, showing sites of triplet repeats that are prone to expansion, and examples of resultant disorders.

Figure 1-4 In this autosomal-dominant trait, an imprinted gene is only expressed when inherited from a female. Hence, only females can have affected offspring. Individuals who inherit the mutation from their fathers will not express the phenotype, but their daughters who carry the mutation can have affected children.

a relatively small number of genes are subject to imprinting, but these account for some distinct phenotypes and unusual patterns of transmission. A dominant trait because of an imprinted gene will only result in a phenotype when the transmitting parent is the one who transmits the expressed allele (Fig. 1-4). This gives rise to multiple examples of nonpenetrance. For example, this is the case in familial glomus tumors. Prader–Willi and Angelman syndromes result from deletions on chromosome 15 from a region that contains imprinted genes. The gene involved in Angelman syndrome is expressed from the maternal allele, so deletion of this allele results in the disorder. In contrast, paternal deletion of the same region produces Prader–Willi syndrome, reflecting the presence of one or more paternally expressed genes in the region. The same phenotypes can result from the inheritance of both copies of chromosome 15 from the same parent, referred to as uniparental disomy. In this case there will be absence of either maternally or paternally expressed genes, depending on whether there are two copies of mother’s or father’s chromosomes. Uniparental disomy results when a trisomic zygote produces an embryo in which disomy is restored by a second nondisjunction event. If the chromosomes that remain are from the parent in whom nondisjunction occurred, uniparental disomy will result. Aside from accounting for some cases of Prader–Willi or Angelman syndromes, uniparental disomy for several other chromosomes has been associated with a phenotype. DIGENIC INHERITANCE In rare instances, it has been found that individuals who are doubly heterozygous for two recessive alleles may manifest a phenotype (Fig. 1-5). This has been

found, for example, in some cases of the eye disorder retinitis pigmentosum. Parents who are heterozygous for two different genes can have doubly heterozygous offspring. If the genes are on different chromosomes, an affected child can transmit both mutant alleles to an offspring, producing apparent dominant transmission. Genes that are subject to digenic inheritance tend to encode proteins that interact with one another in the same pathway.

MOLECULAR BASIS OF MENDELIAN INHERITANCE The patterns of single gene transmission have been known for a long time, but only recently the phenomena have begun to be understood at the molecular level. The bases for the Mendelian patterns as well as the complexities noted in the previous section are gradually emerging. RECESSIVE VS DOMINANT ALLELES Recessive alleles, by definition, only exert a phenotypic effect in a homozygous state. In general, this implies that a mutation has resulted in loss of function of the gene product, and that partial loss of function is not sufficient to result in a phenotype. As noted, this is the case for most enzyme deficiencies. The mutations responsible for such disorders tend to be those that cause premature termination of translation, such as frame shifts or stop mutations, nonsense mutations, deletions, or splicing mutations, which significantly disrupt the coding sequence. Missense mutations may also cause a phenotype if they significantly disrupt the function of the gene product. The basis for dominance can be a diverse set of genetic changes. Some mutations lead to gain of function, for example,

CHAPTER 1 / MENDELIAN INHERITANCE

7

Figure 1-6 Tumor suppressor concept. A tumor suppressor gene is homozygously mutated in a tumor cell. Those who inherit a heterozygous mutation as a dominant trait are at increased risk of cancer if the remaining wild-type allele is mutated.

Figure 1-5 Pedigree illustrating digenic inheritance. Each parent is heterozygous for a different gene. The child who is heterozygous for both expresses the phenotype.

constitutive activation of a cell membrane receptor as occurs in the FGFR3 gene in achondroplasia. Activation of just one allele is sufficient to alter the behavior of a cell. Another mechanism is referred to as dominant negative. Here, a single mutant allele produces sufficient abnormal product to disrupt cellular function. This is the hallmark of mutations in genes that encode products of multimeric proteins. Although only 50% of the protein may be abnormal, abnormal subunits may contribute to a higher proportion of proteins, resulting in a heterozygous phenotype. This is the case for some types of collagen mutations responsible for osteogenesis imperfecta. Dominant transmission may also occur with loss-of-function mutations if 50% levels of gene product are insufficient for normal function. This occurs in mild forms of Marfan syndrome because of loss-of-function fibrillin gene mutations. Interestingly, dominant negative mutations in this gene cause more severe disease, because the presence of abnormal fibrillin molecules has a more pervasive effect on connective tissue integrity than does 50% reduction of quantity. A special case of dominant inheritance is accounted for by tumor suppressor genes. These genes account for familial susceptibility to cancers such as retinoblastoma, colon cancer, and breast or ovarian cancer. In families with an inherited susceptibility, the risk is transmitted as a dominant trait. Within a tumor cell, however, both alleles of the tumor suppressor are mutated, usually with a loss of function type mutation (Fig. 1-6). The mutation of one allele in such families is transmitted from generation to generation. An individual with heterozygous mutation faces an increased risk of cancer because somatic mutation of the remaining wild-type allele is all that is required to start a cell on the path to malignancy. GENETIC HETEROGENEITY There is a wide diversity of different mutation types, with a range of effects on gene function, including mutations that increase or decrease levels of gene product expression, or change the functional properties of the protein. Different mutations in the same gene may have differing impacts on phenotype. The consequence is that different individuals with a dominant trait may have distinct mutations. Furthermore, the two mutant alleles in an individual who is homozygous for a recessive mutation may differ. There are exceptions; in some cases only a very specific mutation will cause a specified phenotype. This is true for the FGFR3 mutation in achondroplasia. In other instances,

a specific mutation may have arisen in an isolated population and remain relatively common there. This is called the founder effect, and accounts, for example, for the high prevalence of specific mutations responsible for Tay-Sachs disease in the Ashkenazi Jewish population. Other than these instances, allelic heterogeneity is more a rule than an exception. In some cases, variable expressivity is explained by the occurrence of different mutations causing slightly different phenotypes. Study of genotype–phenotype correlations can sometimes provide information predictive of disease severity useful in genetic counseling. The difference between severe Duchenne muscular dystrophy and the milder Becker form can be predicted from whether the mutation causes complete loss of the gene product, dystrophin (Duchenne), or production of an aberrant protein (Becker). In still other cases, the different mutations result in phenotypes that would not have been regarded as the same disease. For example, different mutations in the CFTR gene can cause cystic fibrosis, chronic sinusitis, or male infertility resulting from congenital bilateral absence of the vas deferens. Genetic heterogeneity extends not only to different alleles, but also to different loci. Studies to identify genes responsible for disease often reveal multiple distinct genes that are associated with the same phenotype in different individuals. In some instances, such as congenital deafness, this locus heterogeneity reflects that a large number of genes contribute to normal hearing. Many of these genes can be disrupted by mutation, all leading to deafness. In other cases, genes that encode proteins that interact with one another in the same pathway result in an indistinguishable phenotype when mutated. This is the case in tuberous sclerosis, associated with mutation in either the TSC1 or TSC2 genes. The protein products, hamartin and tuberin, interact with one another to form a complex that is involved in the control of cell growth. Locus heterogeneity is important in clinical genetics, because testing of the incorrect gene will fail to reveal an underlying mutation and could lead to incorrect exclusion of a diagnosis. For a recessive disorder, locus heterogeneity explains why two affected parents may have unaffected children. This is a common occurrence in hereditary deafness. MODIFYING GENES Although some phenotypes are reliably associated with specific genotypes, such as sickle cell anemia with the substitution of valine for glutain acid at position 6 β-globin mutation, mutations should not be thought of as totally deterministic of any specific phenotype. All mutations act within a context that is specified by factors including genetic background and the

8

SECTION I / GENETICS

environment. Although a distinction is often made between monogenic and multilateral phenotypes, in a sense all phenotypes are multifactorial. In some instances, a single gene exerts a major effect and in others many genes exert a more modest effect, but in virtually all cases genetic background and environment play some role. Modifying loci may be intragenic or may occur as interactions between different gene loci. The poly T polymorphism with the CFTR gene exerts a modifying effect in the expression of mutations within this gene. Individuals may have an allele with 5, 7, or 9T’s with this polymorphic site, which resides within the interval between exons 7 and 8. The 5T allele is associated with skipping of exon 9, whereas the 9T allele leads to normal splicing. Genotype at the poly T site influences phenotype associated with another CFTR mutation, R117H. This mutation is associated with CBAVD when in cis with 9T and opposite another CFTR mutation, but with mild cystic fibrosis if in cis with a 5T allele. A dramatic example of gene–gene interaction is the phenomenon of epitasis, wherein the phenotype of one gene masks that of another. Individuals will only secrete A or B blood group antigens into saliva if they have at least one dominant Se alleles at the secretor locus. Someone with the se/se genotype will not have A or B antigen in saliva regardless of ABO genotype. Modifier loci with more subtle effects are likely to play a role in a large number of phenotypes. At least some degree of variable expression is probably the consequence of interactions between genes. Any single locus can be visualized as a node in a network. Changing the state of a single node will influence, and be influenced by, a large number of other nodes. In rare and common disorders, expression of a phenotype is a consequence of a complex web of interactions, and genetic analysis will increasingly require such complex systems.

CONCLUSIONS The rules of Mendelian inheritance have provided the basis for the study of human genetics for a century. Studies at the molecular level have provided an understanding of the basis for single gene inheritance and have uncovered some unexpected mechanisms. With new tools of genomics, exploration is beginning of the complexities of gene interactions as they relate to rare and

common phenotypes, and knowledge of the elements of genetics is being integrated into a broader picture of biology.

SELECTED REFERENCES Augarten A, Kerem BS, Kerem E, Gazit E, Yahav Y. Correlation between genotype and phenotype in patients with cystic fibrosis. N Engl J Med 1993;329:1308–1313. Cassidy SB, Dykens E, Williams CA. Prader-Willi and Angelman syndromes: sister imprinted disorders. Am J Med Genet 2000;97(2): 136–146. DiMauro S, Schon EA. Mitochondrial DNA mutations in human disease. Am J Med Genet 2001;106(1):18–26. Kajiwara K, Berson EL, Dryja TP. Digenic retinitis pigmentosa due to mutations at the unlinked peripherin/RDS and ROM1 loci. Science 1994;264(5165):1604–1608. Kiesewetter S, Macek M Jr, Davis C, et al. A mutation in CFTR produces different phenotypes depending on chromosomal background. Nat Genet 1993;5:274–278. Levy HL, Albers S. Genetic screening of newborns. Annu Rev Genomics Hum Genet 2000;1:139–177. Lieberman AP, Fischbeck KH. Triplet repeat expansion in neuromuscular disease. Muscle Nerve 2000;23(6):843–850. McIntosh GC, Olshan AF, Baird PA. Paternal age and the risk of birth defects in offspring. Epidemiology 1995;6:282–288. Monaco AP, Bertelson CJ, Liechti-Gallati S, Moser H, Kunkel LM. An explanation for the phenotypic differences between patients bearing partial deletions of the DMD locus. Genomics 1989;2:90–95. Plath K, Mlynarczyk-Evans S, Nusinow DA, Panning B. Xist RNA and the mechanism of X chromosome inactivation. Annu Rev Genet 2002;36:233–278. Prockop DJ, Baldwin CT, Constantinou CD. Mutations in type I procollagen genes that cause osteogenesis imperfecta. Adv Hum Genet 1990;19: 105–132. Robinson PN, Booms P, Katzke S, et al. Mutations of FBN1 and genotype-phenotype correlations in Marfan syndrome and related fibrillinopathies. Hum Mutat 2002;20(3):153–161. Ruggieri M, Huson SM. The clinical and diagnostic implications of mosaicism in the neurofibromatoses. Neurology 2001;56(11): 1433–1443. Sampson JR. TSC1 and TSC2: genes that are mutated in the human genetic disorder tuberous sclerosis. Biochem Soc Trans 2003;31(Pt 3):592–596. Shiang R, Thompson LM, Zhu Y-Z, et al. Mutations in the transmembrane domain of FGFR3 cause the most common genetic form of dwarfism, achondroplasia. Cell 1994;78:335–342. Walter J, Paulsen M. Imprinting and disease. Semin Cell Dev Biol 2003;14(1):101–110.

2 Nontraditional Inheritance SHAWN E. MCCANDLESS AND SUZANNE B. CASSIDY SUMMARY

A variety of molecular mechanisms have been identified, which explain certain phenomena that are not easily explained by traditional Mendelian patterns of inheritance. These nonMendelian mechanisms differ on a molecular basis, but can be described as a group by the term “nontraditional mechanisms of inheritance” or “nontraditional inheritance.” Stated simply, nontraditional inheritance refers to the pattern of inheritance of a trait or phenotype that occurs predictably, recurrently, and in some cases familially, but does not follow the rules of typical Mendelian autosomal or sex chromosome inheritance. Examples discussed in this chapter are the triplet repeat expansion mutations and genomic disorders including genetic imprinting, mitochondrial inheritance, and multi-allelic inheritance.

The “rules” of segregation of alleles originally defined by Gregor Mendel explained much of the phenomena associated with inheritance and have been dogmatically applied in the field of genetics. However, there are situations in which the rules of Mendelian inheritance cannot explain observed phenomena. A variety of molecular mechanisms have been identified that explain certain phenomena that are not easily explained by traditional Mendelian patterns of inheritance. These non-Mendelian mechanisms differ on a molecular basis, but can be described as a group by the term “nontraditional mechanisms of inheritance” or “nontraditional inheritance.” Stated simply, nontraditional inheritance refers to the pattern of inheritance of a trait or phenotype that occurs predictably, recurrently, and in some cases familially, but does not follow the rules of typical Mendelian autosomal or sex chromosome inheritance. Examples discussed in this chapter are the triplet repeat expansion mutations, and genomic disorders including genetic imprinting, mitochondrial inheritance, and multi-allelic inheritance.

TRIPLET REPEAT EXPANSION The first disorder identified as resulting from this form of nontraditional inheritance is fragile X syndrome (FRAXA). FRAXA is a well-recognized disorder that causes mental retardation, autisticlike behaviors, and a subtle, but characteristic, external phenotype in all males and many females possessing the mutation. Early studies confirmed that the locus of interest was on the X chromosome, and that in some cases the trait was associated with a cytogenetically visible fragile site on the X chromosome, seen only when cells were grown in a folic acid-deficient medium. The mental retardation syndrome was inherited in a classic pattern of X-linked inheritance, with carrier mothers who might have affected brothers or uncles passing the trait on to half of their male offspring. However, some unusual families caused confusion because of a pedigree pattern demonstrating what came to be known as the Sherman paradox. Specifically, there were families identified in which a male appeared to have passed the trait on to his daughters, but he himself was not affected, even though he might have affected brothers or uncles. This pattern could not be explained by typical X-linked inheritance. The solution to the Sherman paradox became apparent when the molecular basis of the FRAXA was found to be a unique type of mutation that occurs in a region of repeated nucleotides in the genetic sequence. Specifically, in the FMR1 gene (Xq27.3) there is a repeated sequence of CGG nucleotides, a “triplet repeat” (Fig. 2-1). In normal individuals this sequence is repeated 5–44 times, but in an affected individual the sequence is repeated more than 200 times. Even more interesting, the mothers of the affected individuals were found to have triplet repeats with 60–200 copies. The normal allele is stably copied during the process of meiosis,

Key Words: Angelman syndrome (AS); fragile X; Mendelian; mitochondrial inheritance; multifactorial inheritance; non-Mendelian; Prader–Willi syndrome (PWS).

INTRODUCTION The “rules” of segregation of alleles originally defined by Gregor Mendel explained much of the phenomena associated with inheritance and have been dogmatically applied in the field of genetics. However, there are situations in which the rules of Mendelian inheritance cannot explain observed phenomena. A variety of concepts have been suggested to explain such phenomena, including the idea that individual genes may function in cooperation with each other and with environmental factors to produce a given phenotype. This concept of multifactorial inheritance is well accepted; however, specific examples for which the various factors can be well defined have been difficult to identify. Other natural phenomena, such as anticipation, in which genetic traits or disorders become more severe or pronounced in successive generations, or genetically determined conditions that appear to depend on the sex of the parent of origin of the involved chromosome, have been difficult to explain, even using concepts of multifactorial inheritance. From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

9

10

SECTION I / GENETICS

Figure 2-1 Triplet repeat expansion in fragile X syndrome. The gel (A) shows Southern blot-based testing for several individuals including a normal male—lane 1, a normal female—lane 2, a female premutation carrier—lane 3, and an affected male—lane 4. DNA is double digested with EcoRI, a restriction enzyme that cuts on either side of the triplet repeat, and EagI, a methylation-sensitive enzyme that only cuts unmethylated DNA (including one site near the Fragile X triplet repeat). DNA is loaded from the top of the gel and separated by electrophoresis. A radioactively labeled probe, which binds near the triplet repeat, is used to visualize the bands of interest. Because EagI only cuts unmethylated DNA, the methylated (inactive) allele is not cut and is seen as a 5.2-kb fragment (containing the triplet repeat). The unmethylated (active) allele is cut by EagI and is seen as a 2.8-kb fragment (also containing the triplet repeat). Normal males have only the 2.8-kb fragment, representing the unmethylated allele from the active X chromosome, as seen in lane 1. Because they have two X chromosomes, normal females have both a 2.8-kb fragment and a 5.2-kb fragment, representing the methylated (inactive) and the unmethylated (active) alleles (lane 2). The female premutation carrier (lane 3) has two bands around 2.8 kb, one slightly larger because of the triplet repeat expansion of about 70 repeats (210 nucleotides). These additional 210 nucleotides represent approx 8% of the 2.8-kb fragment, so two lower bands are seen. The upper, methylated, fragment also has two bands, but because the 210 extra nucleotides only account for approx 4% of the whole fragment, the two bands do not separate enough to be visualized. The affected male in lane 4 has only one allele, seen as a fragment larger (above) than the 5.2-kb alleles in the female premutation carrier (lane 3) because of the increased size of the triplet repeat region of the fragment (estimated to be 330–530 repeats). Because males have only one X chromosome, this band represents a full-size expansion of the triplet repeat, which leads to methylation (inactivation) of the gene, resulting in Fragile X syndrome. Examples of the sequence (B) are shown for normal, a pre-expansion carrier and an affected allele, with the expansion shown in black and flanking sequence shown in gray. The normal allele in this figure has 30 CGG repeats, the premutation 74, and the full expansion 270. (Fig. 2-1A is courtesy of Stuart Schwartz and Linda Jeng.)

but alleles that are somewhat larger than normal, called “premutations,” are prone to further expansion (increase in the number of repeats), leading in some cases to the full mutation. For reasons that are not well understood, the triplet repeat expansion in FRAXA appears to be much more likely to expand during female meioses than during male meioses. Occasionally, a female carrier of a premutation may pass on to her son an allele that has not undergone further expansion. He will not be affected, but can pass on the premutation to his daughters, who then will have a high risk of having an affected child by passing on a further expanded allele. It should be noted that FRAXA is caused by loss of function of the FMR1 protein and can also be caused by other inactivating mutations or deletion of the FMR1 gene. Another interesting observation about FRAXA is that, contrary to Mendelian expectations for an X-linked recessive disorder, carrier females are often affected. In fact, as many as 50% of females who carry a full expansion will have measurable cognitive defects. This is not because of variation in the triplet repeat expansion, but is instead a result of intraindividual variation in X-chromosome inactivation. In every female cell, one or the other of the two X chromosomes is inactivated to compensate for the fact that women have

double the number of X chromosomes as men. This X-inactivation is thought to occur randomly at an early stage of development when there are only 64–128 cells in the blastocyst. On average, half of the cells would be expected to maintain one of the X chromosomes as the active one, and the other half of the cells will maintain the other X. In an individual, however, merely by chance, the ratio may be skewed toward one or the other of the X chromosomes being active. This has been well documented, so that in a population of women there is a normal distribution, with a significant minority of women having one or the other X chromosome much more often inactivated. There are also likely to be different ratios of X-inactivation in a single individual when examining different tissues, with the tissue of interest for the cognitive defects (the brain) being generally unavailable for diagnostic molecular evaluation. These and other examples of the effect of X-inactivation on expression of X-linked disorders have led some to suggest that it is inaccurate to use the distinction of X-linked-recessive or X-linked-dominant. Rather, all of these disorders should be simply called “X-linked.” A number of genetic disorders caused by triplet repeat expansions have now been described. Table 2-1 shows several examples, along with the mode of apparent Mendelian inheritance, the

11

CHAPTER 2 / NONTRADITIONAL INHERITANCE

Table 2-1 Examples of Other Triplet Repeat Expansion Disorders Disorder Myotonic dystrophy Huntington disease Spinocerebellar ataxia type I Friedreich ataxia Fragile X syndrome X-linked spinobulbar atrophy

Inheritance

Triplet sequence

Normal number of repeats

Number of repeats associated with disorder

AD AD AD AR XLR XLR

CTG CAG CAG GAA CGG CAG

5–27 9–37 19–38 7–20 6–52 19–25

>50 to >1000 >37 40 to >80 >200 >200 >40

AD, autosomal-dominant; AR, autosomal-recessive; XLR, X-linked-recessive.

repeated triplet of bases, and the number of repeats associated with the disease state. The molecular mechanism that causes disease is likely different, because some triplet repeat expansions are in coding regions of the gene (exons), some are in noncoding regions (introns), and others are completely outside of the gene, apparently affecting transcriptional regulation. In some cases, the triplet repeat expansion causes disease because it leads to the loss of function of the normal protein product. The triplet repeat expansion may also cause some new function or interaction, which has been shown to be the case in Huntington disease. Interestingly, the vast majority of disorders known to result from triplet repeat expansion are disorders of the neurological system, especially ataxias and other movement disorders. This mechanistic heterogeneity extends also to the meiotic instability of the triplet repeat expansions. Some triplet repeats are more prone to expansion during female meiosis (e.g., fragile X and myotonic dystrophy) whereas others are more likely to expand when inherited from the father (e.g., Huntington disease). The FMR1 gene also has mitotic instability, so that there may be variation in the size of expansion in different cells and different tissues in the same individual. This is not a generalized trait of triplet repeat expansions, though, as it does not occur with the Huntington disease gene, Huntingtin (4p16.3). Anticipation refers to an observed phenomenon where a genetic disorder appears to become more severe in successive generations, a condition not easily explained by simple Mendelian inheritance. For many years there was a controversy as to whether this observation was true, or was a result of ascertainment bias because mildly affected parents may only be identified if they have a more severely affected child. It is now known that anticipation does occur, at least in many triplet repeat expansion disorders, because of increasing size of the triplet repeat expansion in successive generations. In Huntington disease, the child of an affected father may present at a significantly earlier age than the father because of further expansion of the abnormal allele during male meiosis. A similar situation occurs with myotonic dystrophy, a disorder characterized by progressive weakness, especially in the distal extremities, associated with myotonia (difficulty relaxing a contracted muscle), cataracts, and frontal hair loss. A mildly affected mother, who may not know she has the disorder, can give birth to an infant with severe hypotonia and weakness causing respiratory compromise and often death in the neonatal period. The mother may have few symptoms, which can be as subtle as difficulty releasing a handshake. A similar pattern of a severely affected infant born to a mildly affected parent has been described with massive triplet repeat expansions occurring in genes associated with some forms of spinocerebellar ataxia.

GENOMIC DISORDERS AND IMPRINTING Prader–Willi syndrome (PWS) and Angelman syndrome (AS) exemplify several aspects of nontraditional inheritance. In the case of PWS and AS, the parent of origin of chromosome 15 affects the expression of some genes. The discredited hypothesis of Lamarck suggested that the parental factor of inheritance is somehow “imprinted,” and that acquired traits can be passed on to the offspring. Although Lamarck was incorrect, the concept of imprinting has survived, in this case meaning that expression of certain genes is determined by the sex of the parent who passed on that chromosome. These imprinted genes, which reside on autosomes, exist in two copies, as do all autosomal genes. The inactivation of one copy of these genes resulting from imprinting makes the genes dosage dependent. Stated another way, some autosomal genes are normally expressed only from one member of the gene pair even though both genes in the pair have normal base sequence because one allele has been inactivated. This exposes at least three different, and fascinating, causes of nontraditional inheritance not because of variation in the DNA sequence of the genes involved, but instead because of changes affecting the way the genes are transcribed and expressed. “Genomic disorders” are those disorders resulting from the loss of function of a dosage-dependant gene as a result of loss, duplication, or disruption of the region of the genome in which the genes reside. Often, these events are mediated by deletions or duplications caused by aberrant recombination resulting from closely spaced low copy number repeats flanking the critical region. Points in the DNA that break are more susceptible to breakage because of intrinsic characteristics of the DNA sequence that predispose to abnormal DNA looping at the time of meiosis. The following discussion uses PWS and concerning to demonstrate these mechanisms of nontraditional inheritance. PWS is characterized by two distinct clinical phases, both of which are seen in essentially all individuals with the disorder. The first phase, marked by profound hypotonia, can be noted prenatally with decreased fetal movement and breech position in the uterus. After birth the infant is profoundly hypotonic, sleeps excessively and has difficulty with feeding and weight gain. There is a global developmental delay, often accompanied by hypoplastic genitalia and cryptorchidism (in boys), strabismus, and evidence of growth hormone deficiency. By the end of 1 yr the feeding difficulty is generally resolved, and behavioral issues are mild. The second phase typically begins around the age of 2–4 yr. The child is noted to have an apparently insatiable appetite leading to profound obesity if not carefully monitored. There is a typical,

12

SECTION I / GENETICS

Figure 2-2 Gene map of Prader-Willi syndrome/Angelman syndrome region of chromosome 15. This represents approx 4–5 Mb of chromosome 15 just below the centromere. The common breakpoints of the recurrent deletions are shown. Open circles represent maternally imprinted (expressed only from the paternally inherited chromosome) genes. Gray squares are paternally imprinted genes, and black diamonds represent nonimprinted genes. The open ovals represent clusters of small nucleolar RNAs (SnoRNAs) that have been identified. The function of these RNAs is not known, but they are distributed in intronic regions between the 144 purported exons of SNURF/SNRPN. Arrows show the direction of transcription of genes, with the long, dashed arrow showing the direction and extent of the SNRPN exons. Any of the maternally imprinted genes potentially could contribute to the PWS phenotype, although evidence does not suggest a role for MKRN3 or IPW. The imprinting center is shown as two pieces, with the open rectangle representing the region controlling paternal imprinting (AS), and the filled portion representing the maternal imprint control (PWS) region.

but subtle, facial appearance. There is mild mental retardation or low normal intelligence and a characteristic behavioral profile with temper tantrums, obsessive behaviors, verbal perseveration, skin picking, and a variety of other traits. Management includes avoidance of obesity by careful dietary control and exercise, avoiding exposure to food except at mealtime, use of supplemental growth hormone, and an array of behavioral interventions. Without intervention, the life expectancy is significantly shortened because of complications of obesity such as obstructive sleep apnea, right-sided heart failure, and diabetes. AS is characterized by more significant mental retardation, severe speech delay or no speech at all, marked gait disturbance with ataxia, and an unusual behavioral profile with a happy demeanor, frequent bursts of laughter for no apparent reason, and rapid escalation of behaviors. There is a characteristic facial appearance, microcephaly, commonly seizures, and there may be a typical EEG pattern. PWS and AS were delineated clinically in the 1950s and 1960s. In the 1980s, the same recurrent chromosomal deletion was shown to cause both of these disparate clinical syndromes. Further study showed that both result from lack of expression of imprinted genes located near the centromere on the long arm of chromosome 15, but the parental origin of the deletion differed in the two disorders. These two disorders were the first abnormalities resulting from imprinted genes to be recognized in humans. Specifically, at least five different causes of these two disorders have now been delineated, all of which result because genes in the affected region of chromosome 15 are expressed differently when inherited from the mother than when inherited from the father. The known genes in the region are shown in Fig. 2-2, with the sex of the parent in whom the genes are expressed (active) indicated. Much has been learned about the mechanism by which differential gene expression occurs in imprinted genes. Several genes in the PWS/AS region have an excess of methyl groups attached to

cytosine nucleotides. The methylation appears to block the transcriptional machinery from attaching to or acting on these genes, so that no messenger RNA is made from the highly methylated chromosome. Therefore, the only active copies of the genes in this region are those that are unmethylated. This hypermethylation is found in a parent-of-origin-specific distribution. In Fig. 2-2, genes that are hypermethylated (inactive) when inherited from the mother are shown as white circles, whereas those that appear to be inactive when inherited from the father are shown as gray squares. Any structural change that leads to loss or disruption of the active genes will lead to an absence of the gene product. Several of the genes produce a protein product, although little is known about the function of any of these proteins. PWS is a result of the loss of genes that are only expressed from the paternally inherited chromosome 15. The most common mechanism leading to PWS is a small interstitial deletion on the chromosome 15 inherited from the father. This recurrent deletion accounts for about 70–75% of all cases of PWS, and occurs with a frequency as high as 1 in 20,000 liveborn infants (the overall incidence of PWS, resulting from all causes, is thought to be 1 in 10,000–15,000). The same deletion accounts for a similar proportion of cases of AS, when it occurs on the chromosome 15 inherited from the mother. The common deletion in both disorders has the same breakpoints, in the vast majority of cases, resulting from small duplicated sections of DNA flanking the region, spanning a distance of about 4 Mb. This type of duplication has been referred to as a “duplicon,” and a number of similar situations appear throughout the genome. Several have already been identified as causing aberrant recombination leading to recurrence of other microdeletions (e.g., deletion 22q11 syndrome). A second mechanism leading to an individual having no active copies of these imprinted genes occurs when both copies of chromosome 15 are inherited from the same parent, called “uniparental disomy” (UPD). PWS and AS were among the first

CHAPTER 2 / NONTRADITIONAL INHERITANCE

disorders caused by UPD to be described. Now, several other conditions have been shown to have a similar mechanism because UPD for chromosomes containing imprinted genes results in absent expression of the imprinted genes. Other examples include some cases of Beckwith-Wiedemann syndrome resulting from UPD for chromosome 11, transient neonatal diabetes resulting from UPD of chromosome 6, some cases of Silver-Russell syndrome resulting from UPD of chromosome 7, and a mental retardation syndrome resulting from UPD for chromosome 14. UPD most often occurs as a result of a trisomy present at fertilization. PWS occurs when there are two copies of the maternally inherited chromosome 15 present. Most often this results from malsegregation during female meiosis, leading to conception with two copies of chromosome 15 from the ovum and one copy from the sperm. This is nonviable unless there is a second postmitotic event, usually loss of one of the chromosome 15s resulting from malsegregation during mitosis (called “trisomy rescue”). This often occurs because of anaphase lag, where one chromosome fails to move along with the others as the mitotic spindles separate during cell division, and the chromosome is lost into the cytoplasm of one of the daughter cells. It is apparent that there are two possible outcomes from this event. One is the loss of one of the two chromosomes that came from the same parent, resulting in a cell that is now back to the normal and appropriate chromosomal complement, having one chromosome 15 from the father and one from the mother. Alternatively, the chromosome lost in the trisomy rescue process may be from the parent who only contributed one chromosome. This rescues the trisomy, and allows the pregnancy to continue, but if there are imprinted genes on the involved chromosome they will have abnormal expression. Specifically, as seen in PWS, although there are two copies of each gene on chromosome 15, both are from the mother so there will be no gene expression from the imprinted genes. The point in meiosis where nondisjunction occurs is another important factor concerning UPD. Meiosis I nondisjunction is a failure of separation of the homologous chromosomes, but the sister chromatids divide normally at meiosis II (Fig. 2-3). This results in two gametes that carry one copy of each original parental chromosome (heterodisomy) and two gametes with no copies of the parental chromosome involved. The alternative is that the nondisjunction occurs in meiosis II, after the homologues have successfully separated. Meiosis II defects result in one gamete with two identical copies of the same chromosome (isodisomy), one gamete with no copy of the chromosome and two normal gametes (Fig. 2-3D). Both types of meiotic errors can lead to a trisomic fertilization that is then rescued by loss of one of the chromosomes. Isodisomy can also occur when a gamete that is missing a chromosome (nullisomic) is involved in fertilization. That fertilization results in a monosomic pregnancy, most of which are not viable, unless a mitotic segregation defect occurs that leads to a duplication of the chromosome in question. This mechanism always leads to isodisomy. In AS, the majority of UPD cases have paternal isodisomy, suggesting that the mitotic mechanism is more common. In PWS, both heterodisomy and isodisomy have been seen, making it difficult to determine the mechanism. Heterodisomy and isodisomy cause a phenotype if there are imprinted genes in the region. Even when the two copies of the chromosome are different (heterodisomy), both carry the identical imprint (i.e., maternally derived or paternally derived) so that the imprinted genes are not expressed. Isodisomy, on the other hand, can also cause a phenotype because nonimprinted recessive disease-causing genes on the duplicated chromosome will be present

13

on both copies of the chromosome. The very first documented case of UPD was in a child with cystic fibrosis (CF) who was homozygous for the common ∆F508 mutation, a mutation that was only found in one of her parents. She also had short stature. Additional studies demonstrated that both copies of her chromosome 7 were inherited from the parent who carried the CF mutation. This situation has been described for other autosomal-recessive conditions. Furthermore, maternal UPD for chromosome 7 has been shown to be associated with poor growth of prenatal onset, and appears to cause some cases of the primordial dwarfing condition SilverRussell syndrome. Several possible evolutionary advantages of imprinting have been put forward. One hypothesis suggests a relative survival advantage to males of being physically large (the idea of the “strapping” boy) weighed against the survival advantage to the female, in this case the mother, of surviving the delivery to reproduce again by having a relatively smaller baby. Supportive evidence for this comes from the fact that pregnancies resulting from duplication of the male genome form a mass of trophoblastic tissue (placentation) with little or no recognizable embryonic tissue, the so-called “hydatidiform mole.” Likewise, the parental contribution of the extra set of chromosomes in triploid pregnancies correlates with the clinical findings. Triploidy with the paternal genome duplicated is usually associated with a very small fetus and large placenta, whereas maternal genomic duplication is associated with a small placenta and an early spontaneous miscarriage. Many of the earliest imprinted genes identified were found to be associated with growth, although that generalization has not held up entirely as more imprinted genes have been found. Other hypotheses purporting an evolutionary advantage for imprinted genes suggest a role either in protection from inappropriate timing of expression of certain genes, or a role in protecting mammalian females from malignant trophoblastic disease because of parthenogenic reproduction, so that paternally contributed genes are required for normal placentation. It has been suggested that there are only 100–200 imprinted genes out of the total estimated less than 30,000 genes in the human genome. The third mechanism for development of PWS and AS, accounting for less than 5% of cases of each, also results from an imprinting abnormality. During the normal process of gamete production, an individual must change the imprinting pattern of chromosomes inherited from their own opposite sex parent. For example, the chromosome 15 that a man inherits from his mother will be maternally imprinted, and the imprinted genes will not be expressed during that man’s embryonic development. When he passes on that chromosome 15 to his children he must be able to switch the imprint and turn those genes back on. If this does not happen, his offspring will inherit a normal maternally imprinted chromosome 15 from their mother, and an abnormal maternally imprinted chromosome 15 from their father. In this case, there is biparental inheritance of chromosome 15, but both copies are maternally imprinted. When molecular testing is performed there will be no deletion of chromosome 15, nor will there be molecular evidence, usually identified by microsatellite-polymorphism analysis, of UPD. Specific analysis of the methylation pattern in the PWS/AS region of chromosome 15 will be abnormal, though. This “methylation assay” relies on the use of a methylation-sensitive method of evaluating the region, either by use of a methylationsensitive restriction enzyme, or by use of a specialized methylationsensitive polymerase chain reaction protocol. In either test, the result will be production of different-sized DNA fragments from

14

SECTION I / GENETICS

Figure 2-3 (Continued)

maternally and paternally imprinted DNA, demonstrating the presence (or absence) of maternally and paternally imprinted genes. This test will identify all cases of PWS and AS resulting from deletion, UPD, or imprinting defects. The specific mechanism by which the parent of origin imprint is switched has not been elucidated. The region of chromosome 15 involved, the “imprinting center,” has been defined through examination of a series of chromosome rearrangements and progressively smaller deletions. There appear to be distinct, slightly

separated, regions responsible for initiating the imprint for the maternally silenced genes and the paternally silenced genes. Unlike deletions and UPD, which occur sporadically, some imprinting defects result from imprinting mutations (mostly very small deletions of sequence around the imprinting center) that may be familial. Imprinting mutations cause a unique situation in which the first individual in a family to acquire the defect will be normal, but half of the chromosomes that they pass on will be abnormal because the imprint will not be properly switched on

CHAPTER 2 / NONTRADITIONAL INHERITANCE

15

Figure 2-3 Uniparental disomy. This figure follows a single pair of chromosomes through a variety of meiotic and mitotic outcomes to demonstrate how UPD occurs. The chromosomes are shaded differently so that parent of origin for individual chromosomes can be followed easily. All of the meioses are shown in ova, but the process is similar in spermatogenesis. (A) Normal female meiotic gametogenesis. After duplication, the homologues separate during the first meiotic division so that each cell contains two identical chromosomes (sister chromatids). During the second meiotic division the sister chromatids separate so that each gamete contains one copy of each chromosome. (B) In meiosis I, errors the homologues fail to separate, but the sister chromatids do separate normally during the second meiotic division. Thus there are two potential types of gametes, those that contain one copy of each of the parental chromosomes, and those that contain no copy of the chromosome. The first produces a trisomic fertilization (C) that is not viable unless a second error occurs. Mitotic anaphase lag rescues the pregnancy by loss of one of the chromosomes. Depending on which chromosome is lost the result may be either normal biparental inheritance (left side) or uniparental heterodisomy (right side). (D) Meiosis II errors occur after the normal separation of the homologues, but with failure of separation of the sister chromatids during the second meiotic division. In this case, there are three potential chromosomal complements for the gametes, isodisomy, nullisomy, and normal. Fertilization of the disomic gamete (E) leads to a trisomy that can be rescued by loss of one chromosome. This results either in normal biparental inheritance (left side) or in uniparental isodisomy (right side). Note that in meiosis II errors two of the gametes are normal, whereas in meiosis I defects all of the gametes are abnormal. (F) Fertilization of the nullisomic gamete resulting from either type of meiotic error leads to monosomy, which can then be rescued by a mitotic nondisjunction event leading to duplication of the single chromosome. This always results in uniparental isodisomy.

one of their chromosomes. For example, if a woman acquires a new imprinting mutation she will be normal. If that mutation arose on the chromosome 15 that she inherited from her mother, there will be no problem in her offspring because the imprinting pattern does not need to switch. Likewise, each of her daughters who inherit this mutated chromosome will also be fine, as will all of their offspring. However, the son of a mother with a maternally inherited imprinting mutation will be unable to switch the imprint when he passes on that chromosome 15, so that half of his offspring will inherit a maternally imprinted chromosome 15 from their father and will have PWS. This fact makes it important that every child with PWS have the cause thoroughly investigated to rule out this 50% recurrence risk for offspring of the father who carries an imprinting mutation. When analyzing the pedigree of this family it will appear that the trait may skip generations, leading to what has been called a “grandmatrilineal” inheritance pattern for PWS. Similarly, if the original imprinting mutation arises in a male, the eventual result will be females with a 50% risk of having children with AS. Both PWS and AS have been caused by apparently balanced translocations that either disrupt specific genes, or, more likely,

interfere with imprinting in the region. AS can also be caused by mutations in a single gene in the region, UBE3A, possibly accounting for 10% of cases. This gene is unusual in that it appears to be expressed from both alleles in most tissues, but only from the maternal allele in certain regions of the brain. This tissuespecific imprinting pattern is not typical of the genes in the region and the mechanism is poorly understood. It does not appear to be resulting from hypermethylation of CpG islands at the 5′ end of the gene, the most common silencing mechanism for imprinted genes, but may be a result of expression of an antisense transcript of UBE3A. This antisense transcript is in the area of the imprinting region, upstream of UBE3A. It is speculated that on the paternal chromosome the imprinting center allows transcription of the paternally active genes as well as allowing transcription of the paternal copies of the UBE3A antisense region. This antisense transcript, through unknown mechanisms, may interfere with expression of the UBE3A from the same chromosome. The result is that when the imprinting center is “off,” the paternally expressed genes are transcribed, as is an antisense transcript that stops expression of UBE3A from that chromosome. Alternatively, when the imprinting center is “on,” the maternally expressed genes are

16

SECTION I / GENETICS

silenced, and the UBE3A antisense transcript is not expressed, thus allowing normal expression of UBE3A from that chromosome. No mutation in a single gene has been shown to cause PWS, and the abnormal methylation pattern is almost always seen in individuals with typical PWS. This supports the idea that PWS is a true contiguous gene syndrome, meaning that the full phenotype is the result of a combination of effects from several genes that are not properly expressed. There is also support for this idea from the various mouse models of PWS as well, none of which fully recapitulate the complete PWS phenotype. The recurrent nature of the common deletion in PWS and AS is a result of the genomic structure around the region, which is flanked by highly homologous stretches of DNA that predispose to aberrant recombination and deletion or duplication. Such areas can be found throughout the genome, and are thought to explain several recurrent deletion syndromes. It is important to note that one or more of the genes included in the deleted region must be dosage sensitive, so that the loss of a single copy can lead to disease. Developmental abnormalities resulting from this phenomenon of genomic architecture leading directly to a mechanism of disease that does not involve a traditional type of mutation, or traditional inheritance, have been called “genomic disorders.” There are other examples of genomic disorders. Smith–Magenis syndrome, a recognizable mental retardation syndrome, is caused by a recurrent microdeletion on chromosome 17p11.2. CharcotMarie-Tooth disease type 1A results from a recurrent duplication nearby on 17p11.2, involving the peripheral myelin protein 22 gene, which, when deleted instead of duplicated, causes a different disorder, a hereditary neuropathy with liability to pressure palsy. These genomic rearrangements, once they occur, segregate following Mendelian principles, as can be seen with CharcotMarie-Tooth, long known to be inherited in an autosomal-dominant fashion. There is some evidence that this particular aspect of genomic architecture, whereas in some instances predisposing to genomic disorders, is actually part of the process of primate evolution, as some of these regions appear to be associated with new genes developing as part of gene families resulting from genomic duplication.

MITOCHONDRIAL INHERITANCE The idea of nontraditional inheritance developed in response to contradictions between Mendel’s laws and observed biological facts and was initially used to describe imprinting defects. The concept, though, can be further extended to include a variety of other interesting phenomena that lead to situations in which inheritance is not easily explained by Mendel’s laws. A well-recognized example of this is the condition of mitochondrial inheritance, which appears in a matrilineal pattern. This means that the disorder can be seen in males or females, but can only be transmitted from an affected female to her children. Affected males do not transmit the disorder (although this, like most biological “rules” has not been shown to be 100% true). The cause of this inheritance pattern is now well understood, because the mitochondria contain their own small genome. Mitochondrial DNA (mtDNA) is a small, circular DNA containing only 16,569 bp, encoding 13 proteins, each of which is a part of one of the subunits of the mitochondrial electron transport chain. The mitochondrial genome also encodes a unique set of transfer RNAs (tRNAs), as well as two ribosomal RNAs. Mutations throughout the intronless genes on the mtDNA can cause disease, all of which are manifest by disturbances in

energy metabolism, as would be expected by the roles of the known proteins. During the process of gametogenesis the ovum accumulates a large number of mitochondria, each of which contains multiple copies of the mitochondrial genome. The nucleotide sequence of these mitochondrial genomes is not identical, so that in any particular ovum there may be a variety of mutations, none of which are present in every copy of the mtDNA. The sperm compartmentalizes its mitochondria to the motor unit of the tail, so that none of the mitochondria are delivered into the fertilized egg. Therefore, the mother, explaining the matrilineal inheritance pattern, supplies all of the mitochondria in the fertilized egg. Another hallmark of mtDNA diseases is that there can be tremendous clinical variation. Different mutations may predispose to different phenotypes, but even with the same mutation the phenotype may vary. One of the reasons for this becomes clear from the fact that multiple different copies of the mtDNA exist in each egg. After fertilization, mitochondria, and their mtDNA component, replicate and segregate during cell division. In this way, different developing tissues may acquire different complements of mtDNA mutations, and, depending on the effect of the mutation and the energy requirements of the tissue in question, there may be selection for one mtDNA genome over another, leading to accumulation or loss of a particular mutation in a particular tissue type. This variation of mitochondrial complement in different tissues is referred to as heteroplasmy. MELAS, or mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes, is a recurrent mtDNA phenotype most often resulting from a point mutation in the mitochondrial leucine tRNA (nucleotide 3243). There is often an accumulation of mutant mtDNA in successive generations, leading to increased severity with earlier onset in the younger generations. Affected individuals may present with poor growth, lactic acidosis, seizures and ataxia, severe headaches, recurrent strokes or stroke-like episodes, cortical blindness, or muscle weakness. The symptoms tend to progress, with death resulting from respiratory complications, infections, or bowel obstruction. Affected individuals may be identified across many generations and branches of the family, always inherited through females. Kearn-Sayre syndrome is a progressive disorder consisting of peripheral weakness, pigmentary retinopathy, progressive external ophthalmoplegia (because of weakness of the extraocular muscles), heart block or cardiomyopathy, and, occasionally, diabetes mellitus. Most cases of Kearn-Sayre are associated with large deletions of the mitochondrial genome, but some cases have been reported with point mutations in the same leucine tRNA associated with MELAS. The fact that two distinct phenotypes may result from defects in the same tRNA may be a result of tissue heteroplasmy, but it also points to the difficulty in predicting phenotype from genotype in mitochondrial diseases. Many other phenotypes have been described with mtDNA mutations, some more predictable than others, but it is important for the clinician to remember that defects of the oxidative phosphorylation process may produce almost any symptom in almost any tissue at almost any time of life. Also, although this section discusses defects resulting from mtDNA changes, many more nuclear-encoded genes are involved in the production of electron transport protein subunits and the formation and maintenance of the mitochondrial membranes, transporters and oxidative phosphorylation complexes, defects of which are most often inherited as traditional autosomal-recessive traits.

CHAPTER 2 / NONTRADITIONAL INHERITANCE

MULTI-ALLELIC INHERITANCE In the later 19th and early 20th century, as Mendel’s ideas about the independent segregation of traits were being re-evaluated, it was understood that some traits were clearly not the result of single genes. Sir Francis Galton established that height was a trait that could not be explained by Mendelian arguments, and Ronald A. Fisher, a statistician, later showed how multiple genes, each contributing more or less to the final outcome, could explain Galton’s findings on height and other quantitative traits. The concept of polygenic inheritance followed, and along with that the idea that genetic factors may interact with environmental factors to produce a trait in a multifactorial way. Multifactorial inheritance, then, can be invoked to mathematically model empirically observed incidences of a variety of traits, and the concept is now fully accepted in genetic thinking and counseling. This multifactorial model requires several assumptions, including that the genes involved all contribute something to the phenotype, without being dominant or recessive, and that they act together in an additive fashion. The number of genes and environmental factors involved in a multifactorial trait is not infinite, and may vary from just a few to a great many (as has been suggested for the development of hypertension). Findings now point to a form of nontraditional inheritance that is neither fully Mendelian nor fully multifactorial. Three specific examples are digenic inheritance, synergistic heterozygosity, and triallelic inheritance. None of these are completely unique and independent concepts, but all serve to illustrate the complexity of genetic and biological interactions. Retinitis pigmentosa (RP) is a genetically heterogeneous condition of progressive vision loss because of degeneration of the retina associated with increased retinal pigment deposition. It can be isolated or associated with a variety of genetic syndromes, and at least 26 loci have been described in the genome that cause isolated RP, some as X-linked, others as autosomal-dominant and -recessive traits. All of them, though, result from mutations in a single gene. A unique inheritance pattern was identified when a form of RP was found in several different families because of the combination of heterozygosity for mutations in two different genes, ROM1 and RDS. Homozygosity for mutations in the RDS gene can also cause RP, but heterozygosity for a mutation in either, by itself, does not. This finding, that heterozygosity at two different unlinked loci is a requirement for the development of the phenotype, represents a newly recognized form of inheritance that is neither Mendelian nor multifactorial, but is instead digenic. Specifically, it cannot be said to be multifactorial or polygenic inheritance because it is not an additive effect of the two genes, but a synergistic effect. Similar findings have now been shown for several conditions, including one form of hereditary deafness, some cases of Hirschsprung disease, and severe insulin resistance. It is interesting to note, though, that pedigree analysis of affected families might be suggestive of autosomal-recessive inheritance, because the recurrence risk for the parents of an affected child would be 25% with each pregnancy. In some cases digenic inheritance could result from mutations in two genes that interact in a developmental pathway, or that both contribute to the same developmental pathway although they may not physically interact with each other. An analogous situation has been described in several individuals presenting with metabolic myopathies. Investigation into the usual causes of disruption in fatty acid oxidation or electron transport pathways led to the observation that in some cases, heterozygosity for mutations in

17

two different genes involved in cellular energy metabolism may cause myopathy. Specific genes involved include those for carnitine palmitoyl transferase II, very long chain acyl-CoA dehydrogenase gene, and some of the nuclear-encoded subunits of the electron transport chain. Each, by itself, will be considered a recessive allele, not expected to cause disease if the other partner of the gene pair were normal. It appears that in the heterozygous state the reduction in flux through a particular pathway may be tolerated, but if there is a mild (heterozygous) defect in a different part of the same pathway, the sum of the reductions in flux through the pathway may lead to insufficiency of energy production during periods of metabolic stress. This condition has been referred to as synergistic heterozygosity, but the parallel to digenic inheritance is obvious. Although both of these forms of inheritance would lead to Mendelian (recessive) proportions of affected individuals on pedigree analysis, triallelic inheritance would not. This fascinating example of nontraditional inheritance has been described in an isolated population with a high rate of an unusual disorder called Bardet-Biedl syndrome (BBS). Individuals with BBS present with pigmentary retinal dystrophy, polydactyly, obesity, reduced cognitive function, and renal abnormalities. There is genetic heterogeneity for the disorder, with eight loci having been identified. In studying several families it was found that some affected individuals were homozygous for mutations in a previously identified BBS gene, whereas there were unaffected family members that were also homozygous for the same mutations. Further investigation revealed that the difference between the affected and the unaffected individuals was that those affected also had heterozygous mutations of another gene known to be associated with BBS. Therefore, in this population, homozygosity for mutation in the first locus was not sufficient to cause the phenotype, but required a third abnormal allele in a different gene, thus triallelic inheritance. There are different ways to interpret these findings, and debate continues whether these may really represent modifier effects; nonetheless, the complexity of inheritance is much greater than was previously imagined. Similar arguments could be made for some complex traits. A good example is in the risk of blood clotting resulting from inherited thrombophilia. This is one area in which genetic dissection of a complex, multifactorial trait has led to the recognition of a variety of more or less common genetic factors predisposing to thrombosis. The identification of certain of these factors, including the Leiden mutation in the gene for clotting factor V and the prothrombin 20210G > A mutation, along with certain environmental factors, such as cigaret use and oral contraceptive use, allows, at least partially, for the determination of broad categories of risk of abnormal thrombosis and of the relative contribution of individual factors to that risk. New technologies for exploring the human genome, the Human Genome Project, and the dedicated work of thousands of researchers are beginning to unravel the complexities of human inheritance in ways that could not have been imagined by Mendel, Galton, or other pioneers of genetics. This chapter has reviewed some of the complexities of non-Mendelian, nontraditional inheritance and discussed new ways of understanding old paradoxes. It is likely that more complexities will be discovered, giving new insight into genetic disorders both rare and common, and informing new therapeutic approaches. At the least, it seems likely that these new genetic findings will lead to more personalized medical

18

SECTION I / GENETICS

information and risk assessments. The immediate impact, unfortunately, is to make the job of the physician much more complicated. Patient demands for genetic information will make it impossible to ignore these advances, so physicians need to find tools and resources to keep up to date, and to find approaches to share this information with patients in ways that will be beneficial without raising inappropriate expectations or fears.

SELECTED REFERENCES DiMauro S, Andreu AL, Musumeci O, Bonilla E. Diseases of oxidative phosphorylation due to mtDNA mutations. Semin Neurol 2001;21: 251–260. GeneTests: Medical Genetics Information Resource. University of Washington and Children’s Health System (Seattle, OR). http://www. genetests.org 1993–2004 (updated weekly). Accessed May 21, 2004. Goldstone AP. Prader-Willi syndrome: advances in genetics, pathophysiology and treatment. Trends Endocrinol Metab 2004;15:12–20. Katsanis N, Ansley SJ, Badano JL, et al. Triallelic inheritance in Bardet-Biedl syndrome, a Mendelian recessive disorder. Science 2001;293: 2256–2259. Lupski JR. Genomic disorders: structural features of the genome can lead to DNA rearrangements and human disease traits. Trends Genet 1998;14:417–422.

McCandless SE, Cassidy SB. 15q11-13 and the Prader-Willi syndrome. In: Epstein CJ, Erickson RP, Wynshaw-Boris A, eds. Inborn Errors of Development. New York: Oxford University Press, 2004; pp. 765, 766. Ming JE, Muenke M. Multiple hits during early embryonic development: Digenic diseases and holoprosencephaly. Am J Hum Genet 2002;71: 1017–1032. Morison IM, Reeve AE. A catalogue of imprinted genes and parent-oforigin effects in humans and animals. Hum Mol Genet 1998;7: 1599–1609. Nicholls RD, Knepper JL. Genome organization, function, and imprinting in Prader-Willi and Angelman syndromes. Annu Rev Genomics Hum Genet 2001;2:153–175. Online Mendelian Inheritance in Man, OMIM. McKusick-Nathans Institute for Genetic Medicine, Johns Hopkins University (Baltimore, MD) and National Center for Biotechnology Information, National Library of Medicine (Bethesda, MD). http://www.ncbi.nlm.nih.gov/omim/2000. Accessed September 23, 2004. Preece MA, Moore GE. Genomic imprinting, uniparental disomy and foetal growth. Trends Endocrinol Metab 2000;11:270–275. Vockley J, Rinaldo P, Bennett MJ, Matern D, Vladutiu GD. Synergistic heterozygosity: disease resulting from multiple partial defects in one or more metabolic pathways. Mol Genet Metab 2000;71:10–18.

3 Identifying Causal Genetic Factors CHRISTOPHER I. AMOS, JOHN S. WITTE, AND WILLIAM G. NEWMAN SUMMARY

for which genetic studies have been particularly successful, both in identifying genetic factors as well as in providing potential new targets for therapy. We integrate examples from the study of IBD throughout the chapter and then provide a few clinical observations relevant to the findings from these studies as a summary to the chapter. A common disease has been defined as affecting 1 or more individuals per 1000 population. A number of factors may suggest that a common disease (or related clinical traits) is inherited. Clustering of affected individuals within families often provides the initial evidence of an inherited susceptibility to disease. Such findings can be supported by studies of twins; greater disease concordance between monozygotic (identical) compared with dizygotic (nonidentical) twins suggests a genetic susceptibility as it presumes similar environmental exposures among twins. Further evidence of genetic predisposition to common disease may be indicated by studies of migrant populations who retain the level of risk for a particular disease from their area of origin rather than that of the indigenous population. Higher prevalence levels of the disease in specific ethnic populations may be accounted for by a genetic variant in an ancestor (founder effect). Occasionally, common diseases can occur as manifestations of rare single gene disorders. However, further analysis often identifies minor differences between these rare, segregating forms of common disease, and the more prevalent form. For example, segregation of mutations in BRCA1 predispose to breast cancer, but the onset is usually early and there are often distinct pathological changes that can identify cancers arising in individuals with BRCA1 mutations from those with sporadic breast cancers. IBD is an excellent paradigm for a common disease with a significant heritable component. IBD encompasses two major inflammatory diseases of the bowel, Crohn’s disease (CD, MIM 266600) and ulcerative colitis (UC, MIM 191390). The heritable nature of IBD was first described over 40 yr ago by Kirsner who noted that individuals with CD and UC were more likely to have relatives affected by these diseases compared to unaffected individuals. Such familial clustering in itself does not provide conclusive evidence for the genetic basis of a disease. However, studies of disease concordance in twins revealed that both CD and UC are present more commonly in both members of monozygotic compared with dizygotic twins. Additional evidence emerged from Asian migrants to the United Kingdom who maintained a lower level of CD risk than the indigenous population. Finally, a higher prevalence of IBD has been reported in the Ashkenazi Jewish population and IBD is a

The study of a complex disease requires careful characterization of the clinical phenotypes for study. Linkage studies, which can detect relative risks of four or greater, apply stringent diagnostic criteria and restrictive rules for family selection to assure a maximally informative collection of subjects. Clinical characterizations that are adopted for association studies must be precise and should be widely accepted to facilitate large studies. The presence of linkage disequilibrium among tightly linked loci provides a basis for genome-wide association studies. A subset of “tagging” markers that maximally characterize interindividual variability can be sought to minimize genotyping costs. Association studies can detect lower relative risks than linkage methods provided there are a limited number of causal variants at each locus and linkage disequilibrium is present (or one directly studies the causal variant). For some complex diseases there may be multiple disease variants and only moderate risks from any single locus. For these complex diseases alternative strategies using comparative genomics and animal models may be required. Admixture linkage mapping may also permit the study of larger collections of patients than is feasible using traditional linkage methods. Finally, once causal loci are identified, further genotype–phenotype studies will allow the disease to be further delineated. Such studies may also identify subsets of patients with varying responsiveness to treatments. Key Words: Association; comparative genomics; disequilibrium; genetic linkage; genotype–phenotype correlations; linkage haplotypes.

INTRODUCTION Identifying genetic factors that increase an individual’s risk of disease is a major goal of genetic epidemiology. In this chapter, we provide an overview of the primary approaches that use information from families and individuals to discover disease-causing genes. For illustrative purposes, the chapter integrates examples from studies that are successfully deciphering the complex etiology of inflammatory bowel disease (IBD), identifying genetic factors, and providing potential new targets for therapy. The chapter concludes with a summary of clinical observations relevant to the findings from these studies of IBD. We have decided to use the success in understanding the complex etiology of IBD as an example From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

19

20

SECTION I / GENETICS

Table 3-1 Chromosomal Loci Containing Potential or Confirmed Susceptibility Genes for CD, UC, or Both (IBD), as Identified by Genome-Wide Scans or Subsequent Association Studies

Figure 3-1 Designs for the identification of causal genetic factors influencing risk for diseases.

Chromosome locus

Gene identified

Association with CD, UC, or IBD

IBD1 IBD2 IBD3 IBD4 IBD5

16q13 12q14 6p 14q11–q12 5q31–33

CD UC CD CD CD

IBD6 IBD7 IBD8 IBD9

19p13 1p36 16p12 3p26 3p21 10q23 Xq21.3

CARD15 – – – OCTN cluster variants – – – – – DLG5 –

IBD IBD CD IBD IBD IBD IBD

characteristic feature of rare single gene disorders including WiskottAldrich and Hermansky-Pudlak syndromes.

DEFINING THE PHENOTYPE CLINICAL CLASSIFICATION In order to identify genetic loci predisposing for disease, studies often adopt stringent criteria for designating an individual as affected. These strict criteria are used in family studies because affected individuals provide the most information about their underlying genotype, especially when studying infrequent diseases for which the penetrance, or probability to be affected given that one has inherited a riskincreasing genotype, is decreased. Often in family studies we are interested in tracking the inheritance of disease susceptibility, and the underlying genotypes can be most effectively deduced from the affected individuals. We use strict criteria in the initial family studies to avoid misclassification of genotypes. Once genetic loci influencing disease susceptibility are identified, then interest turns to evaluating the impact that disease-increasing genotypes have on the broader spectrum of disease. Figure 3-1 displays designs for identifying genetic factors influencing risk for complex diseases. As shown, for diseases in which the risk conferred by a single factor is high (i.e., exceeding a fourfold increased risk), linkage studies are highly effective for disease localization. However, if the risk associated with a mutation is lower, then alternate strategies employing association-based approaches are more likely to be successful. Precise diagnostic classification of phenotypes is vital to maximize one’s ability to identify causal genetic variants. IBD illustrates the importance of collecting complete clinical data. Individuals with CD are 30 times more likely to have a sibling affected with CD and 16 times more likely to have a sibling with UC than a healthy individual. Patients with UC have similarly raised familial risks. This suggests that there are genes that increase the risk of CD and UC specifically and IBD generally. Genetic studies to date in IBD have supported this suggestion (Table 3-1). To dissect these different genes precise classification of CD and UC is imperative. Individuals with IBD can generally be defined as having CD or UC based on endoscopic, histological, and/or radiological criteria. However, in about 10% of patients a differentiation is not apparent and the term indeterminate colitis is applied. International diagnostic guidelines such as the Vienna Classification for the diagnosis of CD can aid consistency among studies.

Phenotypic expression of common diseases as well as susceptibility may also be inherited. These clinical characteristics may be influenced by disease susceptibility genes or by disease modifier genes, which do not alter risk of the disease itself just the expression. In IBD, epidemiological studies have demonstrated that a number of clinical characteristics may be inherited, including early age of disease onset, disease involving a specific part of the bowel, and more severe disease. Therefore, clinical and demographic data including ethnicity, gender, age of disease onset, and disease severity should all be collected. Ethnicity data are particularly important to ensure that the frequency of a particular genotype in affected individuals is matched to that in an appropriate control population. This is evidenced by variants in the CARD15 gene that have been associated with CD in North American and European but not Japanese populations. In addition, some of the extraintestinal manifestations of IBD including psoriasis and ankylosing spondylitis have a significant heritable component. Genotype–phenotype correlations for mutations in CARD15 have shown associations with the site of bowel disease and earlier age at disease onset among CD patients and with psoriatic arthropathy. QUANTITATIVE PHENOTYPES An alternative to employing stringent criteria for identifying those families most likely to be segregating a simple cause for a genetic disease is to seek phenotypes that are correlated with the disease of interest but which show a more clear genetic component. For example, in the study of cardiovascular disease, identifying genetic causes of myocardial infarction is difficult, although identifying genetic factors influencing low- and high-density lipoprotein cholesterol levels has been more straightforward. Myocardial infarction as an end point is hard to study because it is caused by many different factors. Aberrant cholesterol levels can be strong risk factors for myocardial infarction, and the genetic causes influencing cholesterol levels are simpler to identify than the many factors influencing myocardial infarction as an endpoint. In addition, because cholesterol levels are measured and variable among subjects, each individual provides more statistical information than is provided by the dichotomous information provided by presence or absence of myocardial infarction. The simpler genetic causation influencing cholesterol levels is indicated by the higher heritability (i.e.,

CHAPTER 3 / IDENTIFYING CAUSAL GENETIC FACTORS

increased similarity among closely related individuals) of these traits compared with myocardial infarction as an end point. However, heritability alone does not necessarily indicate that a trait has a genetically simple architecture because even traits that are highly heritable can be influenced by multiple genetic factors. In addition to having high heritability, traits that are known from biological studies to have biochemically or physiologically simple causes would be good candidates for genetic studies. Although quantitative traits can provide more information than dichotomous traits, there are often sampling advantages in studying discrete traits. Individuals who have a disease may become “affected” because they are extreme for an underlying quantitative trait. Identifying individuals with a disease may provide a mechanism for identifying individuals who are extreme for an underlying quantitative phenotype. Families that include some individuals with extreme values may provide the most information in a linkage study because there are underlying genetic variants increasing the trait levels of these subjects. GENETIC LINKAGE ANALYSIS Genetic linkage analysis has been an extremely powerful tool for identifying specific genetic factors for diseases. Linkage analysis has typically been applied for identifying novel genetic factors, by using a genomewide analysis of the coinheritance of disease with genetic markers. Evidence in favor of linkage is typically expressed by the LOD score which is the log10 ratio of the likelihood of the data assuming linkage between a modeled disease susceptibility locus and a genetic marker to the likelihood of the data assuming no linkage of the disease susceptibility and genetic marker. To allow for the large number of tests that are indicated in a genome-wide analysis, several testing paradigms have been developed. If a Bayesian approach is adopted then a LOD score of about 3 leads to a 5% posterior probability of linkage assuming the existence of a single disease locus, even when many markers are genotyped over the entire genome. Morton developed an approach for sequentially combining data from multiple studies by adding LOD scores across studies that has been highly effective. From Bayesian and sequential analytical approaches, a LOD score of 3.0 was proposed as providing a meaningful critical value for declaring strong evidence for linkage. More recently, approaches to control the overall significance of genetic studies when studying multiple markers have been adopted. These criteria have been criticized for being excessively conservative particularly when candidate regions are of primary interest, for example, when prior studies indicated evidence for linkage to an area. The significance testing paradigm requires the slightly higher LOD score of 3.3 to declare that a significant result has been obtained while providing a genome-wide significance of 5%. If a simple genetic mechanism explains inheritance of disease, then a genetic model can be specified and tested for coinheritance of disease susceptibility with genetic markers. In order for linkage studies to be informative, the families chosen for study must be able to show inheritance of a genetic factor. For uncommon diseases for which the penetrance is reduced, the affected individuals provide the majority of information about the segregation or inheritance of genetic mutations predisposing to disease. For quantitative traits, sampling through individuals with extreme phenotypes can increase the probability of sampling a genetic variant influencing the trait of interest. Sampling through extreme individuals is an effective strategy for increasing the power of a linkage study, but may only be practical if the quantitative phenotype can be assayed inexpensively. Some studies of quantitative phenotypes study many

21

phenotypes. Sampling through extreme individuals only increases power for a single or a few correlated phenotypes. Currently available microsatellite-based mapping platforms usually study the genome at approx 10-cM intervals, and investigators usually follow-up positive signals with denser maps. Evidence for genetic linkage in a region would often be followed by finer scale mapping to (1) improve the information for detecting linkage and to identify any recombinant individuals, and (2) using much finer maps, to search for associations between the disease or trait and particular marker alleles. Standard finer mapping panels for microsatellites provide a 5-cM interval spacing, and microsatellite sequences and technologies for approx 0.2-cM interval mapping are routinely available from DeCode genetics (www.decodegenetics.com) or by custom request for even finer mapping (e.g., by request to Invitrogen genetics). Routine genotyping platforms for the purposes of genetic linkage analysis are available from Affymetrix and Illumina and provide results from genotyping of about 11,000 and 6000 genome-wide singlenucleotide polymorphisms (SNPs), respectively. These much finer mapping panels can improve the power to detect linkages and may provide narrower intervals for positional cloning. A wide range of genetic linkage methods is available. The diversity of methods reflects, in part, the considerable success that linkage methods have had in identifying genetic causes of disease, and the consequent value and interest in using the methods by the scientific community. Computing statistics over a large number of genetic markers in families for diseases that do not show simple inheritance patterns is computationally demanding and there are three basic approaches that are taken for the analysis of the genetic marker data. The Elston–Stewart algorithm resummarizes information about haplotypes (the set of alleles on a chromosome) sequentially in a pedigree and is, therefore, efficient for statistical analysis of large families, but limited in the number of markers that can be jointly modeled; usually fewer than five markers can be considered jointly. The Lander-Green-Kruglyak (LGK) method adopts a different approach that facilitates the analysis of multiple markers. The LGK model first identifies the possible inheritance patterns of genotypes within families and stores this information as inheritance vectors. Because the number of inheritance vectors increases rapidly according to the number of individuals in a family, this approach is only suitable for small- or medium-sized families, usually allowing at most 25 individuals in a family to be studied. In addition, because the method stores all possible inheritance vectors in memory, the approach requires considerable RAM to be efficient. The major advantage of the LGK approach is that computational speed increases only linearly in the number of markers so that it is highly efficient for genome-wide analyses. Analyses including many markers on large pedigrees or analyses of pedigrees that include more than a few inbred individuals may not be effectively performed using the Elston–Stewart of LGK algorithms. In this case, Monte-Carlo Markov Chain (MCMC) algorithms are used to approximate the likelihood of the data. MCMC methods provide tools for sampling the haplotype configurations in data. The MCMC procedure samples possible haplotypes according to the underlying probability distribution that generated the data and provides an accurate approximation to the likelihood. A major advantage of MCMC procedures is a decreased need for memory, because they do not require summing over all possible genotypes as in the Elston–Stewart algorithm, or over all possible inheritance vectors as in the LGK. One disadvantage may include the complexity in storing output from analysis, because

22

SECTION I / GENETICS

results from large numbers of realizations from the sampling of genotype configurations are often stored. MCMC methods infer the genotypes for all individuals that are specified as a part of the analytical file. Individuals with known genotypes have a limited number of potential haplotypes, but individuals who have not been genotyped can have a large number of potential genotypes and haplotypes. The probability distribution from which MCMC methods must sample can become quite large if many individuals who have not been genotyped are included in the analytical file. Therefore, it is often beneficial to remove the ungenotyped individuals from MCMC analyses, particularly those who are not affected, because they contribute little in most linkage analyses. An issue in performing genetic analysis is whether to use model-dependent or model-free methods for linkage analysis. Model-dependent methods have higher power for linkage analysis if an approximately valid genetic model can be specified to describe the manner in which disease susceptibility at a given locus is expressed. One approach for estimating penetrance to be used in a linkage study is to first perform a segregation analysis of families that have been ascertained according to a specified sampling scheme. The approach estimates parameters for models describing the inheritance of genetic and environmental factors that most closely fit the dependence in family data. For uncommon conditions, random sampling of families would not result in an informative family, and a sampling scheme is usually followed in which relatives of cases with a disease are preferentially sampled. When the families are not randomly sampled, an ascertainment correction for nonrandom sampling is required in order to obtain parameter estimates that reflect the more general population of families. In order to correct for the nonrandom sampling approach that is typically used, a clearly defined sampling scheme must typically be followed. Using only a binary phenotype (e.g., affection or nonaffection) one may not be able to estimate all the parameters that are necessary to describe the penetrance of the genotypes of the loci influencing disease susceptibility, unless restrictive assumptions about the interactions among the loci are made. Sampling families and collecting information for segregation analysis can be an arduous task and may not be fully informative about the parameters that describe the penetrance and disease allele frequencies. Therefore, investigators studying complex diseases may postulate genetic models from assumptions about the relative risks for disease that are observed from epidemiological studies. It has been shown that postulating an inaccurate genetic model for genetic linkage studies does not lead to false-positive results, in a model-based linkage study. However, if multiple models are tested, then there can be an inflation in the overall number of false-positive results from linkage studies because of the inherent multiple testing problem that is so introduced. A powerful approach for studying complex diseases is to evaluate the evidence for linkage assuming simple recessive and dominant models of disease and then to adjust the required critical value for the LOD score upward by about 0.3 for the small multiple testing problem so engendered. If the genetic model influencing disease susceptibility cannot be inferred with any confidence, either because the genetic model appears too complex or because there is a lack of epidemiological data from which to postulate penetrance, then model-free methods are typically adopted. One approach is to set the penetrance to an artificially low level, thus restricting analysis to include only the affected subjects: with very low penetrance, unaffected individuals provide no information about their possible genotypes and so

do not contribute in the linkage analysis, but this approach still makes some modeling assumptions about disease expression. An alternative approach is to evaluate the similarity in alleles that have been inherited by common parentage (identity by descent) and test whether or not there is evidence that affected relatives share more alleles than expected identical by descent. In some cases this approach may provide a more powerful test for linkage than a model-dependent approach, particularly when multiple independent loci additively increase disease risk. Because pedigrees are usually variable in size and contain different numbers of affected relatives, a variety of different tests have been proposed and are available for testing for linkage. These tests are optimal for varying disease penetrances (which are typically unknown). As a compromise, the pairs statistic, which includes all affected relatives in a pedigree and gives only moderately higher weight to families that include multiple affected relatives is often used. The joint analysis of covariates along with genetic markers in family studies usually has limited utility. Typically collecting covariate information in families is difficult because data cannot be directly collected from deceased or otherwise unavailable individuals. In addition, the genetic risks that are sought in linkage analyses are often large. Some nongenetic factors such as smoking and reproductive behaviors can be reliably collected through proxies (when needed) are inexpensive to collect and may have a strong effect on risk for some diseases. One of the first major investigations to identify genes/ chromosomal loci for a common disease was performed by John Todd’s group at Oxford University in 1994 in families with type 1 diabetes mellitus. Siblings have a 50% chance of inheriting the same allele from a parent. A wide range of statistical methods have been developed to identify regions showing excess sharing of inherited alleles in relatives. This theory has provided the basis for numerous further investigations in a range of common diseases where panels of evenly spaced microsatellite markers are used in genome-wide linkage studies. In IBD there has been consistency between linkage study results with replication of six loci (Table 3-1). For complex diseases, a large number of families may be needed to obtain adequate power to detect linkages. Meta-analysis combining multiple studies can assist in overcoming power limitations from a single study. However, in order for meaningful results to be obtained in meta-analysis, investigators must be studying comparable classifications of the same disease. Coordination of studies becomes necessary for the study of complex diseases. Genetic linkage analyses of IBD have identified numerous genomic regions harboring susceptibility factors. Because of the apparent genetic complexity of these diseases, simple Mendelian models have typically not been applied. Instead, model-free methods using primarily nuclear families have been the preferred method for identifying genetic factors. Genetic linkage analysis followed by positional cloning has been effective in identifying CARD15 and OCTN1/OCTN2 causal mutations for CD. The identification of CARD15 mutations as causal resulted when genetic linkage analysis indicated the chromosome 16q region as likely to contain a genetic susceptibility factor. The CARD15 locus had previously been identified by homology to a similar locus in mice and it was an excellent candidate locus for IBD because mice homozygous for CARD15 deletions showed an IBD-like phenotype. In contrast, genetic linkage analysis identified a region of chromosome 5 encompassing the cytokine cluster on 5q. Haplotype studies, discussed below showed that a single extended haplotype was often present in CD patients. However, causal variants in OCTN1

CHAPTER 3 / IDENTIFYING CAUSAL GENETIC FACTORS

and OCTN2 were only found after extensive sequencing of all the genes in an extended region. ASSOCIATION STUDIES Although parametric and nonparametric linkage analysis approaches have proved successful for mapping many disease and trait genes, in some gene mapping investigations, the limited number of meioses occurring within pedigrees limit one’s ability to detect by linkage the recombination events between closely spaced (~ 108

transducing particles/mL

No clear in vivo tropism

ssRNA virus, enveloped

approx 9.6 kb

>108 transducing particles/mL

No clear in vivo tropism

Adenovirus

dsDNA virus, nonenveloped

up to 34 kb

>1011 pfu/mL

Epithelial cells (liver, lung, intestine)

Adeno-associated virus (AAV)

ssDNA virus, nonenveloped

approx 4.5–5.0 kb

1013 vector genomes/mL

In vivo tropism varies with serotype

Herpesvirus (HSV)-1

dsDNA virus, enveloped

approx 30 kb

Not determined

Epithelial and neural cells

Molecular conjugates

“Hybrid of viral/nonviral”

Unlimited

Not applicable

Liposomes

Nonviral

Unlimited

Not applicable

Tropism may be conferred by ligands in conjugate No clear in vivo tropism

Naked DNA

Nonviral

Unlimited

Not applicable

viable. One clinical trial has been performed in which factor VIIItransfected skin fibroblasts were reimplanted in the omentum. The peak factor VIII levels achieved were only marginally therapeutic, however, and were not sustained. The need to make autologous transduced cells for each individual patient’s treatment could be approached by the use of standardized cell lines expressing factor VIII or factor IX contained in an immunoisolation device as protected allografts; however, this has not been tried in humans. Gene transfer to muscle can be performed using appropriate promoters to direct tissue-specific gene expression. Engineered cells can be implanted in the host, and factor IX has been expressed in mice using muscle cells transduced before reimplantation. Alternatively, muscle cells have been encapsulated in alginate compounds before implantation to protect the modified cells in vivo, permitting them to express human factor IX in immunocompetent mice. Immunodeficient (nude) hemophilia B mice have developed tumors of myoblast origin after implantation of encapsulated cells, which raises the possibility that muscle cell therapies may be tumorigenic in humans under some circumstances. It would be much more practical to transduce myocytes in vivo because there are no human myoblast/myotube cell lines suitable for use in humans. In vivo gene transfer to muscle by intramuscular injection would be more appealing than ex vivo approaches because the methodology is simpler; one vector could be used for all patients, and there would be no need for in vitro cultivation of patient myocytes. The most extensively studied application of this

No clear in vivo tropism

Comments Requires cell division for integration into cellular DNA Can integrate into cellular DNA of nondividing cells Adenovirus serotype 5 most commonly used as vector AAV serotype 2 most commonly used as vector; capacity varies with serotype HSV-1 most commonly used as vector Adenovirus capsids linked to DNA facilitate gene transfer Low efficiency of gene transfer in vivo; direct application to mucosal surface has been performed in vivo Typically delivered by injection (e.g., “gene gun”)

approach utilizes AAV vectors. AAV vectors can only transfer DNA of total length less than 4.5 kb, which can accommodate factor IX gene transfer, but not factor VIII gene transfer. Preclinical studies with AAV vectors in hemophilia B knockout mice and hemophilia B dogs have achieved sustained, clinically relevant plasma levels of factor IX using this approach. Poor delivery of recombinant factor IX by muscle into plasma is the key problem directly related to the use of muscle as an in vivo target for gene transfer. This has been attributed to binding of factor IX by type-IV collagen, and seems to be mediated by epidermal growth factorlike domains on factor IX. It also appears that AAV gene transfer is significantly more efficient in type-X (slow-twitch) fibers as compared with type-Y (fast-twitch) fibers, perhaps as a consequence of more AAV receptors on the cell surface for the former muscle fiber type. Clinical trials of AAV-factor IX gene transfer targeting muscle have demonstrated therapeutically significant plasma factor IX levels approaching 1% of normal in patients given 2 × 1011 vector genomes/kg. Patients reported fewer joint bleeds and administered fewer doses of factor IX after treatment as compared with the period before treatment. A placebo effect cannot be ruled out because this was an open label study, however. Hopes for greater factor IX plasma levels were not realized when higher doses of AAV-factor IX vector were given to other patients, and the factor IX levels seen in the early trials gradually disappeared. It is not clear whether the loss of factor IX was because of loss of vector, adsorption of factor IX by type-IV collagen, methylation of promoter DNA, or other epigenetic events; inhibitor

42

SECTION I / GENETICS

antibodies to human factor IX were not seen in any of these patients, however. Nonviral gene transfer by electroporation of plasmid DNA has been studied. Using in vivo electroporation of plasmid DNA into muscle of mice and dogs it has been possible to demonstrate transient factor IX expression at levels of about 1% of normal, which was followed by an antibody response to foreign (human) factor IX. This technique requires confirmation of factor IX activity in hemophilia B animals to confirm its utility. Factor VIII expression in muscle has not been shown to deliver active factor VIII to the circulation. The amount of factor VIII that can be expressed by muscle cells and secreted in active form into the circulation may be the chief limiting factor to this approach. Liver hepatocytes synthesize many plasma proteins including coagulation factors VIII and IX, and make some proteins, such as albumin, in tremendous quantities. For these reasons, in vivo gene transfer to liver is one avenue being investigated for hemophilia gene therapy. Hepatocytes cannot be easily propagated in tissue culture, but in vivo gene transfer has been demonstrated in animals using adenovirus, AAV, and retrovirus vectors. Adenovirus vectors can mediate quantitative in vivo gene transfer to hepatocytes when administered intravenously. Thus, it is not surprising that adenovirus vectors have been used for in vivo gene transfer of factor VIII or factor IX to liver cells of mice, monkeys, and hemophilic dogs. Unfortunately, adenovirus-mediated gene transfer is typically characterized by a gradual decrease in gene expression. Moreover, adenovirus vector proteins in immune-competent animals induce numerous cytokines and have powerful adjuvant effects that may cause antibodies to the gene product to be formed. Efforts to create and test vectors with most of the adenovirus genes removed have been pursued, based on the observation that deletion of adenovirus vector genes may result in longer periods of gene expression and a decreased immune response. Problems with production of large quantities of these highly deleted vectors have impeded the translation to clinical trials. At least one clinical trial is testing an extensively deleted adenovirus vector for in vivo factor VIII gene transfer. Use of adenovirus vectors may be limited by neutralizing antibodies to the common serotypes, which commonly infect humans; however, the sequential use of different serotypes of adenovirus vector might be employed to circumvent this problem. Although AAV-mediated factor VIII gene transfer to liver in vivo has been reported, gene transfer to liver hepatocytes using AAV vectors has focused largely on factor IX, because of the size constraints of vector DNA packaging. Gene transfer of factor IX DNA by intravenous, portal vein, or hepatic artery injection of AAV vectors has resulted in expression of factor IX in mice, dogs, and nonhuman primates. The results of these studies have justified clinical trials, one of which is underway in patients with hemophilia B. This trial was temporarily put on hold because of vector shedding in the semen of the first subject enrolled. After showing that the AAV vector was not incorporated in germline DNA, the trial has resumed but data on factor IX gene transfer and expression are not yet available. In the field of AAV vector development it has sometimes been difficult to predict the effectiveness of in vivo gene transfer from the results of in vitro experiments. Accordingly, many new approaches are being evaluated. It has been shown recently that different AAV vector serotypes (e.g., AAV 1, AAV 5, AAV 7, AAV 8, and so on) may have advantages for gene transfer regarding DNA capacity or efficiency of gene transfer in particular cell types or animal species, as compared with the AAV 2 vectors that have

been more commonly utilized. Some of the “new” AAV serotypes have been identified as pathogens in nonhuman primates, and their utility in human gene therapy has yet to be proven. Although retroviral vectors based on Moloney murine leukemia viruses are not suitable for in vivo gene transfer to hepatocytes owing to the low rate of cell division in this cell type, lentivirus vectors (which can transfer genes into nondividing cells) may be useful for hemophilia gene therapy. Pseudotyped lentiviral vectors have been used to transfer the factor VIII cDNA into liver more efficiently than nonpseudotyped vectors. Hematopoietic cells are logical targets for gene transfer because they could be transduced ex vivo and returned to the donor by marrow transplantation, after which progeny cells would circulate in the bloodstream. In the case of platelets, an additional advantage might be realized because they would be concentrated at the site of vascular injury where factor VIII or factor IX is needed most. Such hematopoietic cell gene transfer has been attempted as an experimental approach to both hemophilia A and B. Retroviral gene transfer of the (B domain-deleted) human factor VIII gene into hemophilia A knockout mouse bone marrow did not result in detectable plasma levels of human factor VIII after transplantation with concomitant myeloablation, but immune tolerance to human factor VIII in recipients of transduced bone marrow was conferred. A nonspecific viral (long terminal repeat) promoter was used to drive expression of factor VIII, thus the vector could in theory direct expression in any progeny cell type, including many nonhematopoietic cell lineages. It is possible that expression of factor VIII in either hematopoietic or nonhematopoietic progeny cells could mediate the immune tolerance response that was seen after gene transfer in these experiments. Similar experiments using a lentiviral vector that expressed factor VIII in hematopoietic cells resulted in antibodies to human factor VIII in mice that were transplanted with transduced bone marrow; it is suggested that the immunogenic (rather than “toleragenic”) effect might be owing to higher expression levels seen in lentivirus-transduced cells and/or more efficient transduction of antigen-presenting cells. Erythroid and megakaryocytic cell lines secrete significantly higher levels of factor VIII than do B- or T-cell lines in vitro when transduced with lentivirus vectors containing factor VIII expression cassettes. Cytokine stimulators had little effect on factor VIII expression, but differentiating agents such as phorbol myristic acid, which activates protein kinase C, induce approx two- to threefold greater factor VIII expression. Human myeloid leukemia (HL)-60, cells can express factor IX in vitro when transduced by Moloney retroviral vectors; expression from the CMV promoter is increased with phorbol myristic acid (a monocytic differentiating agent) or dimethylsulfoxide (a granulocytic differentiating agent). Human erythroleukemia cells can express factor IX in vitro when expressed under control of a megakaryocyte-specific (glycoprotein IIb) promoter, suggesting that platelets could store the expressed protein in granules that would be available for release at the site of platelet aggregation in vivo. Induction of differentiation resulted in increased levels of factor IX expression in vitro as was seen in previous work with myelomonocytic leukemia cells. However, factor IX has not been expressed at clinically significant levels in vivo by hematopoietic cells. Stromal fibroblasts are cells from bone marrow, which can be obtained by modestly invasive methods (marrow aspiration) and can be grown easily in tissue culture and transduced by retroviral vectors in vitro. Human marrow stromal cells transduced ex vivo

CHAPTER 6 / HEMOPHILIA AS A MODEL DISEASE FOR GENE THERAPY OF GENETIC DISORDERS

with a gibbon ape leukemia virus-pseudotyped Moloney retroviral vector containing a factor VIII expression cassette showed clinically significant (6% of normal) factor VIII levels after reimplantion of cells in the spleen of NOD-SCID mice. Expression of human factor VIII ceased after approx 4 mo owing to promoter inactivation, however. Reinfusion of factor IX-expressing bone marrow stromal cells in canines after irradiation of marrow (to facilitate growth of infused cells in the marrow) led to detectable levels of factor IX in plasma; it is unlikely that such a procedure would be practical in humans, however. Under appropriate culture conditions cells that have the properties of endothelial cells can be induced to grow from nucleated peripheral blood cells. These so-called blood outgrowth endothelial cells (or BOECs) have been isolated from humans, transduced with factor VIII vectors, and infused into immunodeficient mice in which they have expressed significant amounts of factor VIII. The BOECs have been shown to persist with endothelial cell properties in the spleen and bone marrow of treated mice. Use of autologous BOECs transduced with factor VIII or factor IX vectors may be another relatively noninvasive avenue for gene therapy of hemophilia in humans.

FUTURE DIRECTIONS Future issues to be addressed in the field of gene therapy include minimization of the risk for immune response and antibody formation, development of vectors that can sustain high levels of protein expression, and avoidance of germline transmission of vector nucleic acid sequences. It will be instructive to see whether inhibitor antibodies to factor VIII or factor IX in hemophilia gene therapy patients will be observed once the technology is applied to patients who have not had extensive exposure to clotting factor proteins. Novel viral vectors are rapidly being developed and their suitability for gene transfer is being assessed. Novel strategies for expression of relatively large genes such as factor VIII cDNA are required and so-called split vectors may someday permit hemophilia A gene therapy with AAV vectors. Another potential avenue for the introduction of large genes such as hemophilia A is the use of vectors that mediate splicing of normal mRNA in place of defective or missing 3′ mRNA in the message. This method has resulted in transient therapeutic factor VIII levels in hemophilia A knockout mice, and in principle would be an attractive approach to the problem in the common intron 22 inversion of the factor VIII gene in which the last 4 of 26 exons are missing. The hemophilia A dogs in which the same inversion mechanism occurs would be a good model in which to test this approach before clinical trials in patients. Concerns for the safety of retroviral gene transfer to bone marrow cells have been raised by the finding of clonal T-cell proliferation in two young patients with X-linked severe combined immunodeficiency (SCID)-X1 treated with retroviral gene transfer; in both cases the vector was shown to be integrated into genomic DNA of T cells near the LMO2 proto-oncogene promoter. It remains to be seen whether this event is specific to patients with SCID-X1, the vector construct used in the trial, or retroviral vectors in general. The lack of similar adverse events in extensive preclinical and clinical experience with retroviral vectors in other trials suggests that the problem is not generic to retroviral vectors. The problem of T-cell leukemogenesis may be related to the selective advantage that T cells gain in SCID patients after transduction with the γ-subunit of the IL-2 receptor (IL2Rγc), compared with nontransduced T cells. Further, the IL2Rγ c gene sequence may preferentially integrate in the LMO2 genomic DNA sequence. Vector constructs that mediate integration by novel methods have also been described. Novel targets, such as intestinal epithelial

43

cells that have been transduced in vitro by viral and/or nonviral vectors, need to be studied in vivo. These areas are being addressed by investigators throughout the world.

ACKNOWLEDGMENT This represents the opinion of the author and does not constitute US Government policy.

SELECTED REFERENCES Andrews JL, Shirley PS, Iverson WO, et al. Evaluation of the duration of human factor VIII expression in nonhuman primates after systemic delivery of an adenoviral vector. Hum Gene Ther 2002;13(11):1331–1336. Antonarakis SE, Rossiter JP, Young M, et al. Factor VIII gene inversions in severe hemophilia A: Results of an international consortium study. Blood 1995;86:2206–2212. Axelrod JH, Read MS, Brinkhous KM, Verma IM. Phenotypic correction of factor IX deficiency in skin fibroblasts of hemophilic dogs. Proc Natl Acad Sci USA 1990;87:5173–5177. Bi L, Lawler AM, Antonarakis SE, High KA, Gearhart JD, Kazazian HH Jr. Targeted disruption of the mouse factor VIII gene produces a model of haemophilia A. Nat Genet 1995;10:119–121. Brauker J, Frost GH, Dwarki V, et al. Sustained expression of high levels of human factor IX from human cells implanted within an immunoisolation device into athymic rodents. Hum Gene Ther 1998;9(6):879–888. Brinkhous KM, Graham JB. Hemophilia in the female dog. Science 1950;111:723. Bristol JA, Gallo-Penn A, Andrews J, Idamakanti N, Kaleko M, Connelly S. Adenovirus-mediated factor VIII gene expression results in attenuated anti-factor VIII-specific immunity in hemophilia A mice compared with factor VIII protein infusion. Hum Gene Ther 2001;12(13):1651–1661. Chao H, Liu Y, Rabinowitz J, Li C, Samulski RJ, Walsh CE. Several log increase in therapeutic transgene delivery by distinct adeno-associated viral serotype vectors. Mol Ther 2000;2(6):619–623. Chao H, Mansfield SG, Bartel RC, et al. Phenotype correction of hemophilia A mice by spliceosome-mediated RNA trans-splicing. Nat Med 2003;9(8):1015–1019. Chao H, Mao L, Bruce AT, Walsh CE. Sustained expression of human factor VIII in mice using a parvovirus-based vector. Blood 2000;95(5): 1594–1599. Chao H, Sun L, Bruce A, Xiao X, Walsh CE. Expression of human factor VIII by splicing between dimerized AAV vectors. Mol Ther 2002;5(6): 716–722. Chao H, Walsh CE. Induction of tolerance to human factor VIII in mice. Blood 2001;97(10):3311–3312. Choo KH, Gould KG, Rees DJ, Brownlee GG. Molecular cloning of the gene for human anti-haemophilic factor IX. Nature 1982;299(5879): 178–180. Chuah MK, Brems H, Vanslembrouck V, Collen D, Vandendriessche T. Bone marrow stromal cells as targets for gene therapy of hemophilia A. Hum Gene Ther 1998;9(3):353–365. Chuah MK, Schiedner G, Thorrez L, et al. Therapeutic factor VIII levels and negligible toxicity in mouse and dog models of hemophilia A following gene therapy with high-capacity adenoviral vectors. Blood 2003;101(5):1734–1743. Chuah MK, Van Damme A, Zwinnen H, et al. Long-term persistence of human bone marrow stromal cells transduced with factor VIII-retroviral vectors and transient production of therapeutic levels of human factor VIII in nonmyeloablated immunodeficient mice. Hum Gene Ther 2000;11(5):729–738. Davé UP, Jenkins NA, Copeland NG. Gene therapy insertional mutagenesis insights. Science 2004;303:333. Ehrhardt A, Kay MA. A new adenoviral helper-dependent vector results in long-term therapeutic levels of human coagulation factor IX at low doses in vivo. Blood 2002;99(11):3923–3930. Evans GL, Morgan RA. Genetic induction of immune tolerance to human clotting factor VIII in a mouse model for hemophilia A. Proc Natl Acad Sci USA 1998;95(10):5734–5739. Fang B, Eisensmith RC, Wang H, et al. Gene therapy for hemophilia B: host immunosuppression prolongs the therapeutic effect of

44

SECTION I / GENETICS

adenovirus-mediated factor IX expression. Hum Gene Ther 1995; 6: 1039–1044. Fewell JG, MacLaughlin F, Mehta V, et al. Gene therapy for the treatment of hemophilia B using PINC-formulated plasmid delivered to muscle with electroporation. Mol Ther 2001;3(4):574–583. Gerrard AJ, Austen DE, Brownlee GG. Recombinant factor IX secreted by transduced human keratinocytes is biologically active. Br J Haematol 1996;95(3):561–563. Gerrard AJ, Hudson DL, Brownlee GG, Watt FM. Towards gene therapy for haemophilia B using primary human keratinocytes. Nat Genet 1993;3:180–183. Giannelli F, Green PM, Sommer SS, et al. Haemophilia B: database of point mutations and short additions and deletions, 7th edition. Nucleic Acids Res 1997;25:133–135. Giles AR, Tinlin S, Greenwood R. A canine model of hemophilic (factor VIII: C deficiency) bleeding. Blood 1982;60:727–730. Gnatenko DV, Saenko EL, Jesty J, Cao LX, Hearing P, Bahou WF. Human factor VIII can be packaged and functionally expressed in an adenoassociated virus background: applicability to haemophilia A gene therapy. Br J Haematol 1999;104(1):27–36. Graham JB, Buckwalter JA, Hartley LJ, Brinkhous KM. Canine hemophilia: observations on the course, the clotting anomaly, and the effects of blood transfusion. J Exp Med 1949;90:97–111. Hacein-Bey-Abina S, Von Kalle C, Schmidt M, et al. LMO2-associated clonal T cell proliferation in two patients after gene therapy for SCIDX1. Science 2003;302(5644):415–419. Hao Q-L, Malik P, Salazar R, Tang H, Gordon EM, Kohn DB. Expression of biologically active human factor IX in human hematopoietic cells after retroviral vector-mediated gene transduction. Hum Gene Ther 1995;6:873–880. Herzog RW, Hagstrom JN, Kung S-H, et al. Stable gene transfer and expression of human blood coagulation factor IX after intramuscular injection of recombinant adeno-associated virus. Proc Natl Acad Sci USA 1997;94:5804–5809. Hoeben RC, van der Jagt RCM, van Tilburg NH, et al. Expression of functional factor VIII in primary human skin fibroblasts after retrovirusmediated gene transfer. J Biol Chem 1990;265(13):7318–7323. Hortelano G, Wang L, Xu N, Ofosu FA. Sustained and therapeutic delivery of factor IX in nude haemophilia B mice by encapsulated C2C12 myoblasts: Concurrent tumourigenesis. Haemophilia 2001;7(2):207–214. Hortelano G, Xu N, Vandenberg A, Solera J, Chang PL, Ofosu FA. Persistent delivery of factor IX in mice: gene therapy for hemophilia using implantable microcapsules. Hum Gene Ther 1999;10(8):1281–1288. Hough C, Kamisue S, Cameron C, et al. Aberrant splicing and premature termination of transcription of the FVIII gene as a cause of severe canine hemophilia A: similarities with the intron 22 inversion mutation in human hemophilia. Thromb Haemost 2002;87:659–665. Hurwitz DR, Kirchgesser M, Merrill W, et al. Systemic delivery of human growth hormone or human factor IX in dogs by reintroduced genetically modified autologous bone marrow stromal cells. Hum Gene Ther 1997;8:137–156. Kemball-Cook G, Barrowcliffe TW. Interaction of factor VIII with phospholipids: Role of composition and negative charge. Thromb Haemost 1992;67:57–71. Koeberl DD, Alexander IE, Halbert CL, Russell DW, Miller AD. Persistent expression of human clotting factor IX from mouse liver after intravenous injection of adeno-associated virus vectors. Proc Natl Acad Sci USA 1997;94:1426–1431. Kootstra NA, Matsumura R, Verma IM. Efficient production of human FVIII in hemophilic mice using lentiviral vectors. Mol Ther 2003;7(5): 623–631. Kundu RK, Sangiorgi F, Wu LY, et al. Targeted inactivation of the coagulation factor IX gene causes hemophilia B in mice. Blood 1998;92(1): 168–174. Lakich D, Kazazian HH, Antonarakis SE, Gitschier J. Inversions disrupting the factor VIII gene are a common cause of severe hemophilia A. Nat Genet 1993;5:236–241. Lensen R, Bertina RM, Vandenbroucke JP, Rosendaal FR. High factor VIII levels contribute to the thrombotic risk in families with factor V Leiden. Br J Haematol 2001;114(2):380–386.

Levinson B, Kenwrick S, Lakich D, Hammonds JG, Gitschier J. A transcribed gene in an intron of the human factor VIII gene. Genomics 1990;7:1–11. Lin Y, Chang L, Solovey A, Healey JF, Lollar P, Hebbel RP. Use of blood outgrowth endothelial cells for gene therapy for hemophilia A. Blood 2002;99:457–62. Lin H-F, Maeda N, Smithies O, Straight DL, Stafford DW. A coagulation factor IX-deficient mouse model for human hemophilia B. Blood 1997;90:3962–3966. Lozier JN, Csako G, Mondoro TH, et al. Toxicity of a first-generation adenoviral vector in rhesus macaques. Hum Gene Ther 2002;13(1):113–124. Lozier JN, Dutra A, Pak E, et al. The Chapel Hill hemophilia A dog colony exhibits an inversion of the factor VIII gene. Proc Natl Acad Sci USA 2002;99(20):12,991–12,996. Lozier JN, Metzger ME, Donahue RE, Morgan RA. The rhesus macaque as an animal model for hemophilia B gene therapy. Blood 1999;93(6): 1875–1881. Lozier JN, Metzger ME, Donahue RE, Morgan RA. Adenovirus-mediated expression of human coagulation factor IX in the rhesus macaque is associated with dose-limiting toxicity. Blood 1999;94(12):3968–3975. Lozier JN, Yankaskas JR, Ramsey WJ, Chen L, Berschneider H, Morgan RA. Gut epithelial cells as targets for gene therapy of hemophilia. Hum Gene Ther 1997;8(12):1481–1490. Mah C, Sarkar R, Zolotukhin I, et al. Dual vectors expressing murine factor VIII result in sustained correction of hemophilia A mice. Hum Gene Ther 2003;14(2):143–152. Manno CS, Chew AJ, Hutchison S, et al. AAV-mediated factor IX gene transfer to skeletal muscle in patients with severe hemophilia B. Blood 2003;101(8):2963–2972. Marshall E. Gene therapy. Panel reviews risks of germ line changes. Science 2001;294(5550):2268–2269. McCormack MP, Rabbitts TH. Activation of the T-cell oncogene LMO2 after gene therapy for X-linked severe combined immunodeficiency. N Engl J Med 2004;350:913–922. Nakai H, Storm TA, Kay MA. Increasing the size of rAAV-mediated expression cassettes in vivo by intermolecular joining of two complementary vectors. Nat Biotechnol 2000;18(5):527–532. Nathwani AC, Davidoff AM, Hanawa H, et al. Sustained high-level expression of human factor IX (hFIX) after liver-targeted delivery of recombinant adeno-associated virus encoding the hFIX gene in rhesus macaques. Blood 2002;100(5):1662–1669. Naylor JA, Brinke A, Hassock S, Green PM, Giannelli F. Characteristic mRNA abnormality found in half the patients with severe hemophilia A is due to large DNA inversions. Hum Mol Genet 1993;2:1773. Naylor JA, Buck D, Green P, Williamson H, Bentley D, Giannelli F. Investigation of the factor VIII intron 22 repeated region (int22h) and the associated inversion junctions. Hum Mol Genet 1995;4(7):1217–1224. Naylor JA, Nicholson P, Goodeve A, Hassock S, Peake I, Giannelli F. A novel DNA inversion causing severe hemophilia A. Blood 1996;87: 3255–3261. Nunes FA, Furth EE, Wilson JM, Raper SE. Gene transfer into the liver of nonhuman primates with E1-deleted recombinant adenoviral vectors: Safety of readministration. Hum Gene Ther 1999;10(15):2515–2526. Palmer TD, Rosman GJ, Osborne WRA, Miller AD. Genetically modified skin fibroblasts persist long after transplantation but gradually inactivate introduced genes. Proc Natl Acad Sci USA 1991;88:1330–1334. Palmer TD, Thompson AR, Miller AD. Production of human factor IX in animals by genetically modified skin fibroblasts: potential therapy for hemophilia B. Blood 1989;73(2):438–445. Plantier JL, Rodriguez MH, Enjolras N, Attali O, Negrier C. A factor VIII minigene comprising the truncated intron I of factor IX highly improves the in vitro production of factor VIII. Thromb Haemost 2001;86(2):596–603. Rodriguez MH, Enjolras N, Plantier JL, et al. Expression of coagulation factor IX in a haematopoietic cell line. Thromb Haemost 2002;87(3): 366–373. Roth DA, Tawa NE Jr, O’Brien JM, Treco DA, Selden RF. Nonviral transfer of the gene encoding coagulation factor VIII in patients with severe hemophilia A. N Engl J Med 2001;344(23):1735–1742. Sarkar R, Tetrault R, Gao G, et al. Total correction of hemophilia A mice with canine FVIII using an AAV 8 serotype. Blood 2004;103(4):1253–1260.

CHAPTER 6 / HEMOPHILIA AS A MODEL DISEASE FOR GENE THERAPY OF GENETIC DISORDERS

Sarkar R, Xiao W, Kazazian HH Jr. A single adeno-associated virus (AAV)-murine factor VIII vector partially corrects the hemophilia A phenotype. J Thromb Haemost 2003;1(2):220–226. Schnell MA, Zhang Y, Tazelaar J, et al. Activation of innate immunity in nonhuman primates following intraportal administration of adenoviral vectors. Mol Ther 2001;3(5 Pt 1):708–722. Stein CS, Kang Y, Sauter SL, et al. In vivo treatment of hemophilia A and mucopolysaccharidosis type VII using nonprimate lentiviral vectors. Mol Ther 2001;3(6):850–856. Tonn T, Herder C, Becker S, Seifried E, Grez M. Generation and characterization of human hematopoietic cell lines expressing factor VIII. J Hematother Stem Cell Res 2002;11(4):695–704. van Hylckama Vlieg A, van der Linden IK, Bertina RM, Rosendaal FR. High levels of factor IX increase the risk of venous thrombosis. Blood 2000;95(12):3678–3682. Van Raamsdonk JM, Ross CJ, Potter MA, et al. Treatment of hemophilia B in mice with nonautologous somatic gene therapeutics. J Lab Clin Med 2002;139(1):35–42. Wang L, Nichols TC, Read MS, Bellinger DA, Verma IM. Sustained expression of therapeutic level of factor IX in hemophilia B dogs by AAV-mediated gene therapy in liver. Mol Ther 2000;1(2):154–158. Wang J-M, Zheng H, Blaivas M, Kurachi K. Persistent systemic production of human factor IX in mice by skeletal myoblast-mediated gene

45

transfer: feasibility of repeat application to obtain therapeutic levels. Blood 1997;90:1075–1082. Wang L, Zoppe M, Hackeng TM, Griffin JH, Lee K-F, Verma IM. A factor IX-deficient mouse model for hemophilia B gene therapy. Proc Natl Acad Sci USA 1997;94:11,563–11,566. White SJ, Page SM, Margaritis P, Brownlee GG. Long-term expression of human clotting factor IX from retrovirally transduced primary human keratinocytes in vivo. Hum Gene Ther 1998;9(8):1187–1195. Wood WI, Capon DJ, Simonsen CC, et al. Expression of active human factor VIII from recombinant DNA clones. Nature 1984;312: 330–337. Xiao W, Chirmule N, Berta SC, McCullough B, Gao G, Wilson JM. Gene therapy vectors based on adeno-associated virus type 1. J Virol 1999;73(5):3994–4003. Yao SN, Farjo A, Roessler BJ, Davidson BL, Kurachi K. Adenovirusmediated transfer of human factor IX gene in immunodeficient and normal mice: evidence for prolonged stability and activity of the transgene in liver. Viral Immunol 1996;9(3):141–153. Yao S, Kurachi K. Expression of human factor IX in mice after injection of genetically modified myoblasts. Proc Natl Acad Sci USA 1992;89: 3357–3361. Yao S-N, Smith KJ, Kurachi K. Primary myoblast-mediated gene transfer: persistent expression of human factor IX in mice. Gene Ther 1994;1: 99–107.

7 Genetic Counseling ROBIN L. BENNETT

where scientists collected data on human traits and sometimes provided information to affected families, with the primary purpose of “…accumulating and studying records of physical and mental characteristics of human families to the end that the people may be better advised as to reproduce fit and unfit marriages, and in order to establish the potentialities of an individual.” The horrendous excesses of Nazi Germany and the eugenics movement in the first half of the 20th century led to a retreat from advising families about potential hereditary conditions to a more neutral educational model. Risk information was provided to couples and families with the main options being to “take their chances” by becoming pregnant or to refrain from pregnancy. Sheldon Reed is credited with introducing the term “genetic counseling” in 1947. In the 1940s, genetics clinics were staffed by physicians and doctoral geneticists, using a medical/prevention model that primarily involved providing “facts” (risk figures, natural history, and treatment information) so that couples and families could make informed reproductive decisions. With the advent of carrier testing and prenatal diagnosis in the 1960s and 1970s, couples and families were faced with a new range of choices. The educational model of providing genetic information did not facilitate patient decision-making. In 1969, Melissa Richter had the foresight to initiate the development of a master’s degree program in genetic counseling. The provision of genetic information shifted to a more client-centered approach with information being presented within a psychological and cultural context so clients could make informed decisions regarding genetic testing and reproductive options. A tenet of nondirectiveness, particularly regarding reproductive choices, was held firmly in this genetic counseling model, to demonstrate respect for patient’s autonomy and to clearly distance the genetic counselors of this era from the earlier eugenics movement. Genetic counseling is an expanding field in the age of genomic medicine. Genetic counselors provide services to clients across the lifespan, from preconception counseling to prenatal diagnosis, the diagnosis of newborns or pediatric genetic disorders, and the diagnosis of adults with inherited predisposition to diseases such as cancer, presenile dementia, psychiatric disorders, and heart disease. The approach to genetic counseling involves assessing family and environmental history to determine disease risk; assisting in genetic testing, diagnosis, and disease prevention and management; and offering psychosocial and ethical guidance to help patients make informed, autonomous health care and reproductive

“Genetic counseling is the process of helping people understand and adapt to the medical, psychological and familial implications of genetic contributions to disease. This process integrates: — Interpretation of family and medical histories to assess the chance of disease occurrence or recurrence. — Education about inheritance, testing, management, prevention, resources and research. — Counseling to promote informed choices and adaptation to the risk of condition.” —National Society of Genetic Counselors, 2006

SUMMARY Genetic counseling is an expanding field in the age of genomic medicine. Genetic counselors provide services to clients across the lifespan, from preconception counseling to prenatal diagnosis, the diagnosis of newborns or pediatric genetic disorders, and the diagnosis of adults with inherited predisposition to diseases such as cancer, presenile dementia, psychiatric disorders, and heart disease. The approach to genetic counseling involves assessing family, medical, and environmental history to determine disease risk; assisting in genetic testing, diagnosis, and disease prevention and management; and offering psychosocial and ethical guidance to help patients make informed, autonomous health care and reproductive decisions. Genetic counseling focuses on complex issues related to the value of genetic testing, and on medical interventions and health care practices that have varying degrees of efficacy and success. The traditional dogma that genetic counseling must be nondirective is being challenged in favor of a psychosocial approach that emphasizes shared deliberation and decision making between the counselor and the client. Key Words: Cancer genetics; genetic counseling; genetic testing; neurogenetics; prenatal genetics.

INTRODUCTION Historically, the practice of genetic “counseling” focused on providing information about recurrence risks for particular conditions within a family. The first clinics providing information on disease inheritance were established in the early 1900s in Cold Spring Harbor, New York and the University College, London, From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

46

CHAPTER 7 / GENETIC COUNSELING

decisions. Genetic counseling focuses on complex issues related to the value of genetic testing, and on medical interventions and health care practices that have varying degrees of efficacy and success. The traditional dogma that genetic counseling must be nondirective is being challenged in favor of a psychosocial approach that emphasizes shared deliberation and decision-making between the counselor and the client. As genetic counselor Beth Fine noted in 1998, “Genetic counselors function within the health care system as patient advocates who aim to empower patients and families to find solutions to problems and ways of coping that are optimal for them.”

WHAT IS UNIQUE ABOUT GENETIC INFORMATION? As more tests are developed that identify a hereditary component to common chronic disorders such as cancer, heart disease and presenile dementia, genetic disorders are no longer categorized in the exclusive realm of rare, mostly pediatric diseases. Competency in genomic medicine is important for all health professionals, and an idea of “genetics exceptionalism” may no longer apply. But there are still several features of genetic information that present unique personal, family, and social consequences that distinguish genetic disorders from other nonhereditary medical conditions. Obviously, genetic disorders are familial. A genetic diagnosis may embrace multiple generations and members of a family. Often, to provide comprehensive risk assessment to relatives, molecular testing must first be performed on an affected relative to identify the specific gene mutation. When a client seeks an opinion of a surgeon for knee surgery, no other relatives are involved; the consultation remains a private conversation between the consultant and the health provider. In contrast, a healthy woman with a strong family history of breast cancer seeking an opinion regarding her options for breast cancer risk reduction and surveillance may need to obtain medical records and death certificates on multiple affected relatives. Testing the client’s affected mother or sister to identify a gene mutation may be important to provide the consultant with critical information regarding whether she has a high lifetime risk to develop breast and possibly other cancers. In medical genetics, a whole extended family often becomes the client unit, raising unique issues of confidentiality and privacy of health and personal information. Knowledge of familial disease risks might have profound effects (both positive and negative) on interpersonal relationships among family members. Parental guilt is a common experience of the parents of affected offspring. Children might blame a parent for passing on the genetic disorder in the family, or partners may blame each other for the birth of an affected child. Survivor guilt is another common problem that is experienced by unaffected individuals in a family who have “escaped” the genetic disease in the family. The healthy individuals may wonder why they are unaffected whereas other relatives have been less fortunate. The “survivor” may feel on the outskirts of the “family team” despite knowing it is irrational to desire ill health. The familial nature of genetic disorders may alter reproductive plans of many relatives. The parental role is often threatened by learning genetic carrier status. There may be challenges to religious and ethical belief systems between couples and their extended family as couples wrestle with core values of biological parenting and views on adoption, assistive reproductive technologies (including donor gametes and preimplantation diagnosis), prenatal diagnosis, and potential abortion.

47

Although remarkable advances are continuing to be made in the general understanding of the mechanism of many genetic disorders, unfortunately, durable cures such as gene therapy have been slow to become reality. The permanent nature of genetic disease may bring a sense of fatalism or hopelessness with the diagnosis or results of genetic testing. Genetic disorders are chronic, thus there is often a continual array of new health and physical challenges over a person’s lifetime. Individuals with a genetic disorder often experience increasing medical problems as they age. The individual and the family may experience “chronic sorrow” for the “healthy person who will never be.” Genetic disorders are complex, usually affecting multiple organ systems. Frequently, a multidisciplinary team approach is needed to provide comprehensive care to persons with inherited disorders. Individuals with rare genetic disorders often face continual frustration attempting to locate health professionals with experience in their condition. They may feel extremely isolated, with a sense that no one else is quite like them. Linking these individuals through disease specific support networks (such as the Genetic Alliance) can be instrumental in providing clients with the emotional and medical support they require. Genetic disorders often demonstrate enormous variability of disease expression, with some individuals having few manifestations and others have multiple disabling features of the condition. This clinical variability further confounds the understanding of appropriate management for individuals with genetic disorders and makes prognostic predictions close to impossible. Unfortunately, genetic disorders continue to be labeling. As Francis Galton observed, “Most men and women shrink from having their hereditary worth recorded. There may be family diseases of which they hardly dare to speak, except on rare occasions, and then in whispered hints, or obscure phrases as though timidity of utterance could hush thoughts….” A genetic diagnosis still carries the potential for social stigma. A person may be considered less desirable as a mate or employee. The family may feel their heritage is “tainted.” Fear of genetic discrimination (insurance, employment, and societal) may hinder the willingness of individuals and their families to participate in genetic testing and research. The ability to test a healthy person for possible future health status provides new challenges to traditional definitions of “healthy” and “diseased.” Not long ago, the usual appointment with a health professional centered on the client’s specific medical symptoms. Now, a healthy person can be tested for a growing list of potential inherited diseases. Yet, the spectrum of disorders for which there is effective medical therapy remains relatively small compared with the burgeoning number of available genetic tests. None of these genetic tests predict the exact age a person will develop the genetic condition nor do they predict the specific manifestations and severity of the disease. Jonsen and colleagues coined the term the “unpatient” for this group of “genetically unwell” but outwardly healthy individuals who have inherited a gene mutation predisposing to disease. An individual with diagnostic results that are opposite of his or her preconceived affected status may be at higher risk for adverse psychological consequences. Results of genetic testing may alter the person’s self-concept and self-esteem, as well as his or her perceptions of wellness and genetic or social identity.

WHO ARE GENETIC COUNSELORS? Genetic counseling is a distinct medical specialty with the role of providing clinical health care, education and emotional support to individuals and families challenged by congenital and inherited diseases. The field of genetic counseling developed from a

48

SECTION I / GENETICS

Table 7-1 Resources for Locating a Genetics Professional National Society of Genetic Counselors Gene Clinics International Federation of Human Genetics Societies

need to educate, manage, and counsel individuals and families diagnosed with, or at risk for genetic diseases, with respect to how these conditions affect the psychological, medical, financial, and social aspects of life. The term “genetic counselor” is generally reserved for masterslevel health professionals with extensive training in human genetics and counseling skills. The first group of ten genetic counselors graduated from Sarah Lawrence College in 1971. There are 30 programs in the United States and Canada accredited by the American Board of Genetics Counseling (ABGC) with similar programs in place in the United Kingdom, South Africa, Australia, Cuba, Norway, the Netherlands, Taiwan, Spain, Sweden, Saudi Arabia, Japan, France, Israel, and China. As of 2004, there were more than 2300 members of the professional society of genetic counselors, the National Society of Genetic Counselors. Graduates from genetic counseling programs accredited by the ABGC demonstrate competencies in 27 areas within four critical domains: communication, critical-thinking, interpersonal counseling, and psychosocial assessment, and professional ethics and values. Didactic coursework in genetic counseling training programs involves human and medical genetics, cytogenetics, developmental biology and embryology, teratology, statistics, qualitative and quantitative research, counseling theory, communication skills, interviewing techniques, and public health. Clinical skills are obtained by a combination of role-playing and >800 h of comprehensive supervised fieldwork in a variety of practice settings. Teaching experience and completion of a thesis or other scholarly enterprise are ABGC accreditation requirements. Program lengths are approx 2 academic years. Adopted in 1993 and revised in 2006, The Code of Ethics of the National Society of Genetic Counselors is an “ethic of care,” because this approach emphasizes the interdependence of individuals and reflects the values, principles and beliefs of genetic counselors. The relationships are genetic counselors themselves; genetic counselors and their clients; genetic counselors and their colleagues; and genetic counselors and society. A primary focus is respect for client autonomy and patient advocacy. Most genetic counselors work in university medical centers, private and public hospitals, or large medical facilities. Genetic counselors work independently or as a member of a multispecialty team. An increasing number of genetic counselors work with diagnostic laboratories and pharmaceutical companies as well as in positions related to the development of government and public policy. Genetic counselors are also uniquely trained to work as research coordinators for genetic research studies. Most genetic counselors are involved in education, including providing education to a variety of health professionals and community organizations, as well as students at all levels, from elementary schools to postgraduate programs. Other health professionals that are specifically trained to provide genetic counseling services include physician (medical) geneticists and clinical nurse specialists in genetics. Medical geneticists attend a fellowship or residency program through a program accredited by

www.nsgc.org www.geneclinics.org www.ifhgs.org/genetics/ifhgs/members.htm

the American Board of Medical Genetics or the Canadian College of Medical Genetics. Clinical geneticists often have particular areas of interest such as oncology, pediatrics, dysmorphology, prenatal diagnosis, neurogenetics, or metabolic disorders. Advance Practice Genetic Nurses meet genetic competencies through a portfolio process established by the International Society of Nurses in Genetics and the Genetic Nurses Credentialing Committee. All health care professionals need to develop basic aptitude in human genetics as genomic medicine becomes incorporated into all fields of medicine. The National Coalition of Health Care Provider Education in Genetics has developed a set of core competencies that can be adapted to various health specialties. Recognizing the need for genetic counseling and referral to a genetics specialist is an important component of these competencies. Table 7-1 includes a list of resources for locating health professionals with expertise in genetic counseling and medical genetics; many of these listings include specialty areas of practice (such as cancer genetics, neurogenetics, prenatal genetics, and so on).

THE PROCESS OF GENETIC COUNSELING Genetic counseling sessions with clients and their families may involve a one-time crisis intervention dealing with a new genetic diagnosis, or may develop into a relationship over many years if the client is treated in a specialty clinic for diseases such as hemophilia, neurofibromatosis, fragile X syndrome, or Huntington disease. The genetic counseling intervention is designed to reduce the client’s anxiety, enhance the client’s sense of control, and mastery over life circumstances, increase the client’s understanding of the genetic disorder and options for testing and disease management, and provide the individual and family with the tools required to adjust to potential outcomes. A major paradigm of genetic counseling is that it is noncoercive. The information provided during genetic counseling helps the individual and family personalize often threatening information in order to clarify their values and strengthen their coping mechanisms. Whether a genetic counseling session is a one-time visit or over many years, there are three broad areas that are covered in each session, assessment, education, and counseling. ASSESSMENT The process of genetic counseling begins with “contracting”—the merging of the counselor’s and client’s expectations. Why is the client here? What are their concerns? What are the mutual expectations and goals of the session? Gathering information during the phone intake or early in the visit assists in developing patient rapport and realistic expectations of the visit as well as appropriate triage to other health professionals. Explaining what will happen and who is involved in the visit can help alleviate client anxiety. What are the client’s preconceived notions about patterns of inheritance, chances of testing positive or developing the family disorder? What is the perceived burden of the disease (financial, emotional, and social)? If the genetic counseling information is divergent from a client’s perceptions, the client may have difficulty incorporating the information or implementing disease-management recommendations. Throughout the process of genetic counseling,

CHAPTER 7 / GENETIC COUNSELING

49

Figure 7-2 Additional pedigree symbols. Reproduced with permission from the University of Chicago, from Bennett RL, Steinhaus KA, Uhrich SB, et al. Recommendations for standardized human pedigree nomenclature. Am J Hum Genet 1995;56:745–752.

Figure 7-1 Pedigree symbols. Reproduced with permission from the University of Chicago, from Bennett RL, Steinhaus KA, Uhrich SB, et al. Recommendations for standardized human pedigree nomenclature. Am J Hum Genet 1995;56:745–752.

there is continual appraisal of family beliefs about causation, and of emotional, experiential, social, educational, and cultural issues that may affect the client’s incorporation of information and coping patterns. Early in the session a medical family history is recorded using standard pedigree symbols (Figs. 7-1–7-3). Usually a pedigree includes two generations of ascent from the consultant (the person seeking information) or proband (the first affected relative who brings the family to medical attention), and two generations of descent. For example, pedigree assessment for a 60-yr-old man with a family history of colorectal cancer would include information about his parents, grandparents, aunts and uncles, and possibly his cousins, as well as information regarding his siblings, children, and likely his grandchildren. A pedigree is an important method of establishing patient rapport and serves as a visual demonstration for providing patient education on variation in disease expression in the family, as well as identifying other relatives at risk for disease. Statistical risk assessment based on pedigree analysis, epidemiology (such as Hardy-Weinberg equilibrium), various risk models (such as Bayes theorem), and the sensitivity and specificity of various genetic tests is central to genetic counseling. Genetic counseling involves explaining risks in multiple ways,

such as distinguishing absolute from relative risks (e.g., a 10% absolute risk but a threefold increased relative risk), and using percentages to frame the magnitude of risks from different perspectives. For example, if an autosomal-recessive pattern is the mode of inheritance, the chance of having an affected child would be framed in terms of a 25% or one in four chance of occurrence or recurrence, as well as a 75% chance or three in four chance that a son or daughter would be unaffected. Has a diagnosis been established? Confirmation of family medical information through review of medical records and death certificates or even obtaining family photographs can be essential to ensure that the pedigree, and thus the risk assessment provided to the client, is based on accurate information. A client’s recall of information about second- and third-degree relatives (e.g., aunts and uncles, grandparents, and cousins) is much less likely to be accurate than information about more closely related relatives (e.g., siblings, parents, and children). EDUCATION AND HEALTH PROMOTION Knowledge is more than information. Genetic counselor Ann C. M. Smith noted, “Individuals and families affected by genetic diseases face a plethora of high tech information from which they seek to gain true knowledge about their genetic circumstances. What is said and how information is communicated can have a significant impact on their ability to process the information, and to their understanding and assimilation into personal life circumstances. To place more power into the hands of individuals and their families affected by genetic disease is to be sure that they have adequate knowledge—not just information—about their genetic circumstance…the genetic counselor seeks to effectively “communicate” such highly technical genetic information in a way that is compassionate, empathic, and sensitive to the ethnocultural values of the client. In this way, the genetic encounter moves from basic patient education to the multifaceted practice of genetic counseling.” Genetic education and health promotion usually involves communication about the following areas: • Discussing options for available genetic testing or diagnostic procedures (including testing costs, sensitivity, and specificity), and arranging for testing as appropriate.

50

SECTION I / GENETICS

Figure 7-3 Pedigree relationships. Reproduced with permission from the University of Chicago, from Bennett RL, Steinhaus KA, Uhrich SB, et al. Recommendations for standardized human pedigree nomenclature. Am J Hum Genet 1995;56:745–752.

• Review of the inheritance pattern(s) and natural history of the condition, disease monitoring and management, available preventive measures, and reproductive options. • Options for prenatal diagnosis and availability of assistive reproductive technologies (e.g., gamete donation, preimplantation diagnosis). • Provision of contact information for disease specific support groups (see www.geneticalliance.org). • Referral to community resources and health specialists, as needed. • Recommendations for genetic evaluation of other relatives with information about local resources for relatives that live in other communities.

Because of the potentially profound effect that a genetic diagnosis may have on the life of the individual being evaluated, genetic test results are usually relayed in person. A support person is encouraged to attend most genetic counseling visits, especially when tests results are disclosed. A follow-up visit or phone call to discuss the client and family’s reaction to the results is advisable. Patients are likely to remember only about half of the information presented during a clinical visit. Focusing on areas of client concern that were ascertained in the initial assessment helps make the information meaningful to the client. Patients also remember best what they are told first; therefore concentrating on the most important issues in the beginning of the session aids client recall. Providing opportunity for questions and for clients to restate the

CHAPTER 7 / GENETIC COUNSELING

51

Table 7-2 Pros and Cons of Genetic Testing of an Asymptomatic Minor Potential adverse consequences of testing (focus on positive test results)

Potential benefits of testing

Damage to the minor’s self esteem

Resolution of the parent’s (and possibly the child’s) concerns about mutation status Allow child and family time to adjust to status if test is positive. No anticipation of developing disease if result is negative Affected relatives can be positive role models for child Anticipatory guidance such as choosing physical activities and occupation for possibly affected child Health status is normalized and perception of disease becomes part of the child’s sense of self Ability to make informed reproductive decisions

Distortion of the family’s perception of the child. Siblings may be treated differently depending on their genetic status Child may identify with severely affected relatives Loss of future adult autonomy and confidentiality for the tested child Adverse effects on the child’s capacity to form future relationships Fear of rejection in forming long-term relationships. Fear/guilt if person wants biological children Fear/guilt if person wants biological children. Chronic sadness for “planned” family. Discrimination (insurance, employment, education, choice of mate) Increased medical surveillance for “healthy” child. Child feels labeled

Anticipatory guidance regarding reproductive decisions (e.g., discussions of alternative forms of parenting such as adoption and donor gametes) Decreased premiums for insurance because risk factor eliminated if test result is negative Available at earliest opportunity for medical intervention (if available)

From Bennett RL, Hart KA, O’Rourke E, et al. Fabry disease in genetic counseling practice: recommendations of the National Society of Genetic Counselors. J Genet Couns 2002;11:121–146. With kind permission of Springer Science and Business Media.

information in their own terms also allows the health professional to estimate how much of the information is understood by the patient. Providing information in several ways to accommodate different learning styles (e.g., verbal, audiovisual, written, and websites) is useful. A patient follow-up letter (that can also be sent to relevant health professionals) provides a resource after the clinic visit that may be referred to even years later, and also assists with coordination of patient care and as a resource for informing at-risk relatives. COUNSELING Genetic counseling is multifaceted; it includes assessing personal, social, religious, and ethnocultural views on how a genetic diagnosis, genetic testing, and test results affect the client’s life. Assessing possible ethical concerns such as confidentiality, disparate paternity, concerns about insurability, views on discrimination, employment issues, feeling about reproduction, and prenatal diagnosis, and testing minor individuals for adult onset conditions are some of the many areas that are explored in genetic counseling. Understanding the client’s perception of risk is more important than the actual risk number. Did the client have a preconceived notion of whether the “chances” were high? What percentage risk does the client consider acceptable or too high? Is the information presented different from what the client expected? Patient decisions are supported in the context of individual values, beliefs and goals. There are several excellent books that review the many counseling issues and process of genetic counseling.

THE FUTURE OF GENETIC COUNSELING The demands for genetic counseling will increase as new genetic tests continually become available. Pharmacogenetics and susceptibility counseling pose novel challenges to the system of genetic counseling that has traditionally concentrated on rare genetic disorders with clear patterns of Mendelian inheritance. Models of genetic counseling must continue to evolve. Can the health care system afford the time intensive previsit preparation, interviewing, assessment, and counseling intervention that the present genetic counseling model embraces? It can be argued that

the subject matter at the core of genetic counseling sessions— heart-wrenching reproductive choices or decisions centering on extreme prevention measures such as prophylactic mastectomy or oophorectomy—deserve more time than what has become the usual 10-min health encounter. Genetic advances will challenge ethics of care. For example, it is generally thought that healthy minors should not be tested for adult-onset conditions in which there are no treatments, yet there may be benefits to testing children under certain circumstances (Table 7-2). With the advent of high-throughput genetic technologies such as tandem mass spectrometry, a growing number of disorders with limited treatments are being added to the diseases that are available for newborn screening—the potential is there for simultaneous screening for hundreds of diseases. How can parents truly be educated and counseled about the implications of so many disorders? Is there significant natural history data on these diseases to know the age that a medical intervention should be implemented? Will parents have a choice regarding which screening tests are given to their newborns? Finally, the atrocities that were committed in the spirit of genetic research and “ethnic cleansing” in the Eugenics era must never be forgotten. By 1926, laws mandating sterilization of the “mentally defective” were implemented by 23 of the 48 United States. Federal immigration quotas limited immigration by “inferior” ethnic groups. Euthanasia was a legal program in Nazi Germany, where the lives of over 70,000 people with hereditary disorders were extinguished, in addition to the six million Jews and others who were murdered. With the clarity of hindsight modern human geneticists shudder at these early biases and abuses. Yet, genetic counselor Robert Resta notes that genetic research as all scientific research may be “influenced by the political and social beliefs of all-toohuman geneticists. Those beliefs may be so ingrained that we mistake them for biological laws. It is, I suspect, beyond our ability to know which of our personal biases we are disguising as scientific truths. The whispers and hints of our biases may be heard only by future generations of geneticists.” The voluntary nature of genetic testing

52

SECTION I / GENETICS

and the availability of genetic counseling by skilled health professionals must continue.

SELECTED REFERENCES American Board of Genetic Counseling, Bethesda, Maryland; www.abgc. net. Accessed Feb. 22, 2006. American Board of Medical Genetics, Bethesda, Maryland; www.abmg. org. Accessed Feb. 22, 2006. Baker DL, Schuette JL, Uhlmann WR, eds. A Guide to Genetic Counseling. New York: Wiley-Liss, 1998. Bennett RL. The Practical Guide to the Genetic Family History. New York: John Wiley and Sons, Inc., 1999. Bennett RL, Hampel HL, Mandell JB, Marks JH. Genetic counselors: translating genomic science into clinical practice. J Clin Invest 2003; 112:1274–1279. Bennett RL, Hart KA, O’Rourke E, et al. Fabry disease in genetic counseling practice: recommendations of the National Society of Genetic Counselors. J Genet Couns 2002;11:121–146. Bennett RL. Steinhaus KA, Uhrich SB, et al. Recommendations for standardized human pedigree nomenclature. Am J Hum Genet 1995;56: 745–752. Biesecker BB. Back to the future of genetic counseling: commentary on “psychosocial genetic counseling in the post-nondirective era.” J Genet Couns 2003;12:213–217. Biesecker B, Peter K. Genetic counseling: ready for a new definition? J Genet Couns 2003;11:536, 537. Ciarleglio LJ, Bennett RL, Williamson J, Mandell JB, Marks JH. Genetic counseling throughout the life cycle. J Clin Invest 2003;112: 1280–1286. Claus EB, RIsch N, Thompson WD. Autosomal dominant inheritance of early-onset breast cancer: implications for risk prediction. Cancer 1994;73:643–651. Fine BA. Genetic counseling. In: Jameson JL, ed. Principles of Molecular Medicine, Totowa, New Jersey: Humana Press, 1998, pp. 89–95. Gail MH, Brinton LA, Byar DP, et al. Projecting individualized probabilities of developing breast cancer for white females who are being examined annually. J Natl Cancer Inst 1989;81:1879–1886. Genetic Alliance, Washington, DC; www.geneticalliance.org Accessed Feb. 22, 2006. Genetic Nursing Credentialing Commission, Inc., Hot Springs, Arkansas; www.geneticnurse.org. Updated April, 2005. Accessed 21-7-05. International Society of Nurses in Genetics, Pittsburgh, PA. www.isong.org. Accessed Feb. 22, 2006. Jonsen AR, Durfy SJ, Burke W, Motulsky AG. The advent of the “unpatients.” Nat Med 1996;2:622–624.

Marymee K, Dolan CR, Pagan RA, Bennett RL, Coe S, Fisher N. Development of the critical elements of genetic evaluation and genetic counseling for genetic professionals and gerontologists in Washington state. J Genet Couns 1998;6:133–165. Mazumdar PMH. Eugenics, Human Genetics and Human Failings. London:Routledge, 1992. McConkie-Rosell A, DeVellis BM. Threat to parental role: a possible mechanism of altered self-concept related to carrier knowledge. J Genet Couns 2000;9:285–302. National Coalition for Health Professional Education in Genetics, Lutherville, Maryland. www.nchpeg.org. Accessed 21-7-05. National Society of Genetic Counselors, Chicago, IL. www.nsgc.org Accessed Feb. 22, 2006. National Society of Genetic Counselors, Resta RG, Biesecker BB, Bennett RL, et al. A new definition of genetic counseling: National Society of Genetic Counselor’s Task Force Report. J Genet Couns 15 (2), in press. Nussbaum RL, McInnes RR, Willard HF, Thompson MW. Thompson & Thompson Genetics in Medicine, 6th ed, Philadelphia: W.B. Saunders, 2004. Parmigiani G, Berry DA, Aguilar O. Determining carrier probabilities for breast cancer-susceptibility genes BRCA1 and BRCA2. Am J Hum Genet 1998;62:145–158. Plumridge D, Bennett R, Dingo N, Branson C. The Student with a Genetic Disorder: Educational Implications for Special Education Teachers and for Physical Therapists, Occupational Therapists and Speech Pathologists. Springfield: Charles C Thomas, 1993. Resta RG, ed. Psyche and Helix, Psychological Aspects of Genetic Counseling. New York: Wiley Liss, 2000. Resta RG. Whispered hints. Am J Med Genet 1995;59:131–133. Schild S, Black RB. Social Work and Genetics: A Guide for Practice. New York: The Hawthorth Press, 1984. Smith ACM. Patient education. In: Baker DL, Schuette JL, Uhlmann WR, eds. A Guide to Genetic Counseling. New York: Wiley Liss, 1998; 99–121. Walker AP. Genetic counseling. In: Emery AEH, Rimoin DL, Connor JM, Pyreitz RE, Korf BR, eds. Principles and Practice of Medical Genetics, 4th edition. New York: Churchill Livingston, 2002, pp. 842–874. Weil J, Psychosocial Genetic Counseling. Oxford: Oxford University Press, 2000. Weil J. Psychosocial genetic counseling in the post-nondirective era: a point of view. J Genet Couns 2003;12:199–211. Williams JK, Schutte Dl, Evers C, Holkup PA. Redefinition: coping with normal results from predictive gene testing for neurodegenerative disorders. Res Nurs Health 2000;23:260–269. Young ID. Introduction to risk calculation in genetic counseling, 2nd ed. Oxford: Oxford University Press, 1999.

8 Animal Models in Biomedical Research Ethics, Challenges, and Opportunities

ROBERT W. WILLIAMS We are all conscious today that we are drowning in a sea of data and starving for knowledge. —Sydney Brenner (2002)

Despite this upbeat assessment and the impressive size of Principles of Molecular Medicine, biomedical research is only now beginning to move from relatively simple qualitative analysis of single factors to well-designed studies of multiple interacting factors and a multitude of genes. In his Foreword, Victor McKusik points out that we still do not have an accurate inventory of the number of genes, mRNAs, or proteins that make up a human. We know even less about the interactions among these complex molecules. We have just begun to explore the intricate molecular networks that mediate between our genomes, our environment, and our health. We know comparatively little about the linkage between gene variants and the most common and pervasive diseases. Animal models are essential for this transition to a more global and integrative approach to medical care. These models have a vital role as fast and secure routes to improving our understanding of the vocabulary and syntax of the genome and of the complex relations between gene variants and disease. Only research with animals allows us the precise experimental and genetic control required to confirm causality. Animals are also essential when cellular assays fail to provide a relevant biological context, for example, to study the regulation of blood pressure or the systemic effects of drugs or pathogens. As a result, we can expect to rely more heavily on animal models as we transition from simple approaches to more sophisticated and realistic studies that account for human genetic diversity and differences in environmental exposure. Animal models will remain a permanent part of any experimentally validated research program, and will be even more crucial to converting the sea of data alluded to by Sydney Brenner in the opening quotation into effective knowledge that will have a positive impact on human health.

SUMMARY Diagnostic power needs to be matched to prognostic accuracy. To accomplish this, we need robust animal models that incorporate the same level of genetic and genomic variation as highly diverse human populations. A key goal of biomedical research is to provide clinicians with sufficient knowledge to predict disease risk and choose effective treatment. We are on the cusp of having complete genome data for each patient. However, our current understanding of complex biological systems and individual differences does not allow us to predict effects of novel interventions and drugs. Animal models provide precise genetic and experimental control, and they also provide a route to more rapid discovery and validation of disease prevention and treatment. But the successful application of experimental results to diverse humans requires that models accurately represent both the underlying disease process and the impact of human genetic variation on treatment and outcome. Key Words: Animal model; biomedical research; humanizing mice; inbred; molecular medicine.

INTRODUCTION It has been just over 50 yr since Linus Pauling introduced the term molecular medicine in the context of groundbreaking work on sickle cell anemia, and since Roger Williams wrote Biochemical Individuality—a prescient introduction to what we now call individualized medicine. Progress in genetics and molecular biology has been relentless over these fifty years, and thanks to new technologies that range from microarrays to magnetic resonance imaging, we are acquiring information at an accelerating pace. In the next decade, patients are likely to have their individual genomes and transcriptomes stored as part of their medical records. Fine-tuning treatment on a case-by-case basis will become the norm, and we can hope that this process will involve more science than art.

USING ANIMALS IN RESEARCH: RATIONALE AND ETHICAL CONTEXT A wide variety of unusual species have been valuable in specific areas of research. Several examples highlight serendipitous discoveries and unique animal models that are making important contributions. • The common nine-banded armadillo has an unusually low average body temperature (approx 32°C), which makes it a unique nonhuman host for the leprosy bacillus, Mycobacterium

From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

53

54

SECTION I / GENETICS

leprae. In nature, prevalence of infection reaches 30% in some coastal parts of Louisiana. For three decades, armadillos have, therefore, been the pre-eminent model used to study leprosy and to develop and test vaccines. Female armadillos also give birth to litters of monozygotic quadruplets, a unique pattern of reproduction that makes armadillos an intriguing model for the developmental biology of twinning and the limits of genetic control. • Groundhogs have a high endogenous incidence of hepatitis B virus that leads to liver cancer virtually indistinguishable from that of humans infected with the same virus. This species is used to evaluate environmental and toxicological factors such as aflatoxin that affect susceptibility to liver pathology and cancer progression in humans. • Two closely related species of voles provide insight into variation in neuroendocrine modulation of reproduction and social structure. Montaine voles are monogamous, whereas meadow voles are polygamous. The activity of neuroactive peptide hormone systems—vasopressin in particular—has a strong impact on pair bonding and social structure. This work continues to lead to mechanistically sound information into how neuroactive peptides modulate aspects of reproduction and behavior. Most biomedical research, however, relies on familiar domesticated subspecies of mice, rats, hamsters, rabbits, ferrets, cats, dogs, and pigs. Although the size of these species is often confounded with their complexity and similarity to humans, mammals are not arrayed from simple to complex, from primitive to modern, or from stupid to smart. From the standpoint of evolutionary refinement or biochemical complexity, mice, rats, dogs, pigs, chimpanzees, and humans are peers—all are the latest products of roughly 200 million yr of mammalian evolution. ANTHROPOMORPHIC LEGAL THEORY Old-world monkeys and apes (macaques, baboons, chimpanzees, and so on) belong to a special category of genuinely wild species that share evolutionary history and biological affinity with humans. They are not more complex than mice or more deeply thoughtful than dogs or pigs, but they are incontrovertibly more similar biologically to humans. This in itself is a reason for using these species in research despite the cost and emotional qualms their use engenders. A vocal wing of the legal community, led by Steven Wise, Laurence Tribe, and Alan Dershowitz, is exploring the idea that “humanity quotients” can be assigned to life forms as diverse as honeybees and chimpanzees, arguing that each species should be provided with scaled legal protection. For example, according to Wise’s scale, the humanity quotient of parrots ranks them somewhere between elephants and dolphins. This fractional anthropomorphic legal theory clearly remains outside the scope of scientific discourse. Misapplying the results of cognitive neuroscience to scale species and perhaps even individuals will produce unintended consequences and a cascade of legal ambiguities. Being human is not an honorary degree that scientists or lawyers are poised to confer. Yet these misframed ideas have begun to exert a subtle but serious impact on scientists themselves, many of whom are willing to concede or even to formally endorse the inverted proposition that similarity to humans makes the use of an animal model inappropriate. This is precisely wrong. The more closely an animal is related to humans, the more potential value it has as a model. Given that animal research is ethically well grounded and is not abusive,

biomedical research is a more meaningful use of animals than their alternative uses as pets, food, transport, or articles of clothing. Their contributions become a permanent part of a knowledge base. Only biological Luddites will deny the long-term positive impact animal models are having on medical care. In this context, the intrinsic barrier to using chimpanzees in research becomes clear: their use is limited by their wild origin and endangered status, their size, and their high cost. Classification schemes based on scales of intelligence and levels of consciousness will falter because the sensory and cognitive capacity of each species is unique and of roughly equal sophistication from a neurobiological perspective. The biological quirks of human brains and the self-reported complexity of human thought are not the qualities that make humans special. Yes, humans have a slight cognitive edge, but what distinguishes humans is our highly complex cultures and unique ability to accumulate and pass knowledge from generation to generation. A sound understanding of human biology is a priceless scientific heirloom bequeathed to the next generation. Animal rights activists exert almost continuous pressure to reduce or eliminate the use of nonhuman mammals in biomedical research. It is relatively easy and effective for this advocacy group to shift the focus away from the unequivocal benefits that have already accrued from animal research onto ethical ambiguities and the difficulties of generating effective animal models. Exploiting the emotional bonds that most humans have with domestic animals and pets, opponents of animal research paint a bleak picture of the state of biomedical research and the need for animal models. The term “sacrifice” is used to refer to killing animals as part of research projects. This is not a euphemism; this word choice is an acknowledgment of their death and our intent to use them for common good. As individuals and as groups, humans constantly make active and passive decisions about the value of human lives compared with those of other species with which we have natural affinity and even sympathy. Animal research is an explicit and honest example of one of these decisions. LIMITS OF USING HUMANS IN BIOMEDICAL RESEARCH Several prominent commentators, researchers, and organizations have stressed the need for more widespread and effective use of humans to study human disease. Their point is that with appropriate safeguards, new informatic methods can be harnessed to effectively mine vast clinical resources within patient records, tissue samples, and images. Massive clinical resources can in principle be merged with data on DNA sequence variants and gene expression differences to generate deep insights into disease vulnerability and environmental modulators. The systematic efforts of deCode in Iceland and the United Kingdom BioBank in Britain are beginning to provide compelling examples of the effective integration of clinical, genetic, and molecular analyses of large populations of humans. The limitation of these massive studies is that they often cannot directly test relations between cause and effect. The critical advantage of animal models is that they can be used to directly test causal relations and mechanisms of action under well-defined experimental conditions and perturbations. The ponderous effort by the US Surgeon General to prove the causal connection between cigarette smoking and cancers in the 1950s is a compelling example of precisely why good animal models are needed. How can human association studies be expected to disprove the alternative hypotheses that a common genetic

CHAPTER 8 / ANIMAL MODELS IN BIOMEDICAL RESEARCH

factor causes both cancer and a yearning for nicotine, or perhaps a tumor suppressor gene variant is tightly linked with a nicotinergic acetylcholine receptor gene variant? Animal models can unequivocally resolve these types of ambiguity. The need for unequivocal mechanistically sound explanations cannot usually be circumvented by use of humans or cell lines. The effects of thalidomide, ethanol, or heavy metals on human embryonic development cannot be studied, and even unintended toxicological “experiments” of these types are unacceptable. Even voluntary participation in biomedical research by humans must be tightly constrained for ethical reasons. For example, for what period of time is it permissible to temporarily blind human volunteers to study brain reorganization and subsequent recovery—a day, a week, a month, or a year? —and what is the appropriate compensation for participation in such a project? THE PROBLEM OF TRANSLATION Most common diseases are inherently complex and multifactorial. Elias Zerhouni stated this point clearly: “Solving the puzzle of complex diseases, from obesity to cancer, will require a holistic understanding of the interplay between factors such as genetics, infectious agents, environment, behavior, and social structures.” Molecular medicine is not something usually considered holistic, but oddly enough, this direction seems likely with high throughput genomic tools that will soon provide many humans with their personal genome sequence. This advance in medicine will produce exceedingly well-informed patients who will expect customized care. Molecular processes and organismal responses will often not extrapolate directly across individuals, strains, and species. The supposed failure to make rapid progress in research and clinical delivery—for instance, in the war on cancer initiated in 1971— often reflects a misunderstanding of the complexity of biology (a single cell can be thought of as having greater organizational complexity than the stock exchange), the current depth of ignorance of these complex systems, and the disappointment that follows from overly optimistic media reports on the latest results and potential cures. Unjustified high expectations and incautious headlines are followed by disappointment and denunciation of the models on which they are based. Nonetheless, there are reasons to constructively critique animal models. A general problem is that the majority of models are designed to reduce genetic and environmental complexity—an unrealistic setting that fails to account for, let alone embrace, the inherent complexity of human populations. Several generations of biologists have been trained to adhere to reductionist methods in which single variables and subsystems of subsystems are studied in isolation. Although this approach is highly productive when access is limited to statistical resources and when questions and answers are focused on fundamental qualitative processes such as basic mechanisms of transcription or enzymatic pathways, it is no longer adequate. The closer that research gets to clinical treatment—the major purpose of animal models—the greater the need for more complex models that can handle multifactorial perturbations. The computational and collaborative tools are now available to retain the power of reductionist methods while encompassing the power of complex multifactorial experimental designs. A solution to this puzzle requires new types of animal models that more accurately mimic the complex structure of human populations simultaneously allowing control of genomes and environments. Research involves long-range vision, gradual progress, and delayed gratification, and it is difficult to quantify the many

55

losses that would be the consequence of failure to make discoveries through the use of animal models. Hence, these compromises are made between costs and benefits. It is usually faster, more efficient, and less costly to conduct studies using animal models rather than humans.

BUILDING MODELS AS A KEY TO PROGRESS The phrase “animal model,” first used widely in the 1960s, implicitly acknowledges the many fundamental differences between animals such as mice and our peculiar human species. Often human diseases cannot be fully replicated using other species, even chimpanzee. If a disease is hard to define in humans— autism, for example, slides across several diagnostic subcategories— then it will be hard to model in macaques or chimpanzees, let alone mice. Nonetheless, there are profound and compelling reasons to develop and refine animal models for as wide a variety of human diseases as possible, even those that might initially generate a cynical smile, such as mouse models for macular degeneration (mice have no macula or fovea) or developmental dyslexia (a subtle and uniquely human disorder affecting the ability to process and interpret text). The phrase “animal model” is also used to indicate that the production of models is often an active process that involves an intense effort in selecting or modifying the genetics of a species or in devising treatments, which consistently produce valuable information and knowledge. Building a good animal model typically involve cycles of testing, modification, and refinement. The process may take years to refine. In the 20th century, technical advances were episodic, and animal models were introduced at a leisurely pace. For example, inbred strains of mice were initially generated by Clarence Little and colleagues in the late 1920s to demonstrate that differences in cancer susceptibility had a genetic basis. Little’s first inbred strain was spontaneously prone to several types of cancer, whereas his second strain was not. The use of inbred strains diffused over a period of decades into other fields, in particular, immunology, developmental biology, neurology, behavioral genetics, and toxicology, but the underlying resources and models were not modified significantly. NEW METHODS TO MAKE NEW MODELS New methods to generate models are being introduced at a rapid rate. The successful integration of exogenous genes into the oocytes of several inbred strains of mice in the early 1980s by Palmiter and Brinster dramatically accelerated the pace with which engineered mice could be generated. Hundreds of transgenic lines of mice are now available, including lines in which human genes are overexpressed. The Huntington’s disease gene, HTT, and the β-amyloid precursor protein, APP, are two prominent examples of human genes that have been moved into mice to explore mechanisms of neurodegeneration. Transgenic models were followed in the 1990s by the first wave of knockout mice, in which single genes were inactivated in all cells throughout life—what could be called the absolute knockout. The current trend, ca. the year 2000, is to introduce subtle and specific alterations in the expression of genes and their derivative proteins—what are called conditional knockouts. The effect of the knockout may be limited to a particular organ, tissue, cell type, or the knockout may be restricted to a particular stage of development. It is now possible to imagine adding the equivalent of an expression “volume control” to each gene and to selectively control

56

SECTION I / GENETICS

the settings of those controls on a cell-by-cell and stage-by-stage basis. This possibility is only a few years behind science fiction. Every year, researchers generate not only new models but also new methods to make models. Even the most active research groups find it challenging to track, let alone exploit, the profusion of molecular and cellular techniques. Technical advances in making models sometimes outpace the basic and clinical research they are intended to produce and serve. The technologies that are used to produce models now have an impact that is equivalent to major advances in instrumentation. ALL MODELS ARE IMPERFECT Humans are a particularly difficult species to model. This fact leads to failures and misapplications, but it should spur further efforts rather than resignation. Human populations have uniquely human problems; the high prevalence of myopia, Alzheimer’s disease, obesity, diabetes, asthma, and cancer is just a few unfortunate examples. Most of these problems are produced by a combination of genetic susceptibility factors and the effects of highly variable demographics, environments, and lifestyles. Because of human diversity and environmental flexibility, some humans are not even particularly good models for other humans. Consider the well-known differences in alcohol metabolism and tolerance associated with aldehyde dehydrogenase gene variants (Han Chinese, for example, typically have low activity alleles and low ethanol tolerance), whose effects extend to social customs and population health. Dramatic individual differences in drug metabolism are associated with several cytochrome p450 gene variants. In some cases, genetic and environmentally induced variation makes the notion of the standard dose or treatment untenable. Recognition of this fact, coupled with advanced technology, drives an intense interest in individualized medicine. The difficulty of generating a good model or generalizing results from an animal to a human, or from one human to another, does not diminish the vital need for models—it simply emphasizes that multiple models are needed and that only a well-considered review of several animal models can provide sufficient grounds to begin the arduous, costly, and inherently risky process of testing new treatments in clinical trials. Models rely on biological commonality. The fundamental basis for using animal models is that virtually all human disease is partly modulated by molecular and cellular interactions that are reasonably well conserved across a broad swath of the animal kingdom. There are of course species-specific exceptions, but this biological and genetic conservation often extends through the entire chordate phylum through to model organisms in other phylae and kingdoms, including Drosophila, nematode, plants, yeast, and bacteria. Having full gene sequence data for many species has greatly enhanced the long-term ability to discover and exploit this biological commonality. New genomic data allow systematically testing the functions of genes, in isolation and in networks. With these rich comparative data, extrapolations can be made more confidently across species boundaries. The ocular disease aniridia type 2 is caused by mutations in the PAX6 gene. A feature that has been highlighted by research in Drosophila, cephalopods, rodents, and humans is the general conservation of this transcription factor in modulating key aspects of eye development. Other aspects of PAX6 are more variable, and the phenotypes naturally differ depending on the particular mutant allele of this gene and on the substantial variation between and within species. Building up and outward from the genomic level to complex and contingent cellular networks that are closer to the disease process

will remain challenging work for many years, but at least a list of the mammalian genetic vocabulary is being formed. Converting this list into a dictionary and encyclopedia of the mammalian genome is the next crucial phase that will make animal models much more powerful and useful. CONSIDERATIONS OF COST, EFFICIENCY, AND NUMBERS OF ANIMALS Investigators attempt to maximize the impact of their research within the limits imposed by budget and expertise. Maximizing impact usually means conducting and interpreting a series of experiments that often require the use of many animals. The cost of acquiring, treating, and analyzing each case therefore is a significant part of the expense. A key issue is the number of cases needed to give a compelling answer or an accurate estimate of the effect of treatment. One obvious and effective way to improve statistical power (the ability to detect a genuine biological signal) is to increase the number of cases that are studied within each group. The standard error of the sample mean is roughly proportional to the square root of the number of animals per group. A twofold reduction in experimental error requires approximately a fourfold increase in sample size. Assuming that equally precise and equally relevant measurements can be obtained from a mouse, rat, rabbit, or dog, the statistical advantages of using a larger number of mice rather than a smaller number of dogs will be substantial (the cost ratio is roughly 1/30). Both assumptions will often be wrong—more accurate measurements can often be obtained from larger species. This is precisely why dogs have been the pre-eminent model used to study heart function, cardiovascular disease, and hypertension. It is not that dogs are inherently more like humans than mice or rats are, but that instrumentation used to study heart, kidney, and cardiovascular systems in humans was readily adaptable for use in animals of comparable size. This equipment was, therefore, relatively easy to modify to yield accurate data at a modest cost. Guidelines of many scientific societies and funding agencies require that investigators use the “minimum number required to obtain valid results.” This is a key statement formalized in the influential guide for the care and use of laboratory animals. The statement presupposes that researchers know in advance the effect size that they want to detect and that the variance structure and stability of the traits being measured is understood. A minimum number, however, is a recipe for truncated experimental design. (“Appropriate number” would be a wiser term than “minimum number.”) By truncating a study to only include male rats at exactly 90 d of age, for example, numbers can effectively be minimized and investigators may be able to show a significant effect of a treatment with an α error rate of p < 0.05 with only two per group. To generate this significant result all that is needed is a lucky draw, a genuinely large effect, or an unintended confound (perhaps caging conditions varied). Should that information be published with an n of 4? Not at all. A statistically significant result is not necessarily a valid result. The actual trend in practice is to increase n to obtain better power, more precise estimates of effects, and better generality of conclusions. Instead of striving for a minimum number, researchers should use a more complex and somewhat ad hoc procedure to determine the number of animals required to obtain valid results. The intended goal is usually to obtain results that have broad utility and some applicability to humans. To claim to have discovered a potentially important therapeutic effect of an estrogen treatment on the aging process in one inbred strain of mouse, for example, it should be almost mandatory to verify that difference in several

CHAPTER 8 / ANIMAL MODELS IN BIOMEDICAL RESEARCH

other inbred and even outbred strains of mice. Differences that do not generalize across different strains of mice may interest mouse geneticists, but they are unlikely to interest a clinician treating patients. Ideally, similar results would also be demonstrated in another species before moving to clinical trials.

MOUSE MODELS IN BIOMEDICAL RESEARCH Studies that exploit mice, many of which are referenced in other chapters and on an almost daily basis in the press, are designed ultimately to improve the diagnosis and treatment of human disease. There are inevitable limitations in relying on a single species that differs so much from humans in size, ecological niche, and reproductive strategy. However, mouse models have reached a high level of molecular and genetic sophistication. Mice presage the achievements that can be expected with other species of mammals—and perhaps of the human species—over the next several decades. For example, the ability to inactivate a gene by homologous recombination (knockout technology) was introduced in mice in the early 1990s by Mario Cappechi and Oliver Smithies and successfully adapted to rats with a 12-yr lag. Many thousands of knockout lines are now in widespread use as models of human diseases and as test beds of gene function. This chapter does not review the burgeoning technical repertoire used to generate genetically engineered strains of mice, but focuses on several concepts and strategies surrounding the generation and use of mouse models. SIZE AS A FACTOR IN ANIMAL MODELS Mice are the smallest mammals commonly used in biomedical research. Adult body weight of various species and subspecies ranges from 10 to 30 g, roughly 1/3000 the size of humans. This scaling applies even to an organ such as brain that is considered unusually large in humans; the brains of both mice and humans comprise roughly 2% of total body mass, and in terms of neuron numbers mice are actually proportionally brainier than humans (approx 75 million vs 100 billion neurons). Like humans, mice have significant body size sexual dimorphism; males typically weigh 30–50% more than females. The main advantage of small size is that a set of 8–10 animals can be maintained in good health in a shoebox-sized cage. Current practice is to maintain 4–5 adults/cage (an anthropomorphic standard for which there is no scientific support). Small size is also associated with particularly rapid development and rapid aging. Gestation in mice is typically 19 d. Mice do not open their eyes until 8–10 d of age, but they develop rapidly and can be weaned after 3 wk of age. Females often produce their first litters at 60–70 d. A full generation therefore typically requires about 3 mo, and up to four generations can be produced per year. Mice are usually fertile for a year or less, and life-span typically ranges from 600 to 800 d. Thus, the biomedical research community has easy access to a mammal in which size, length of the life cycle, and breeding costs are already optimized. These factors have contributed to rapid progress in generating useful mouse models for human disease. Methods initially perfected in large species such as cats and dogs have been miniaturized to the point that it is now practical in many cases to exploit smaller species such as mice. This shift is driven not by ambiguous ethical gradations that may distinguish small from large mammals, but primarily by greatly improved prospects of generating useful results at a lower cost using instruments and methods specifically geared for studying small mammals. Almost equally accurate results will often be generated from

57

mice as from humans and other large mammals. The main inducement to use mice is not always lower animal costs and higher numbers of research subjects (the mouse/rat cost ratio is only 1/2) but rather the enormous variety of genetically engineered and selected lines of mice. In some cases there are hard limits to measurement precision that can be obtained using small animals. A case in point is the resolution limits of structural and functional MRI data sets. MRI imaging resolution in any species is not expected of much less than 10 µ because of water diffusion. The proportional in-plane resolution in the mouse brain will be roughly 1/15 that of humans (the cube root of the weight ratio), and a 50-µ resolution in mouse is equivalent in anatomical terms to a 700-µ resolution in human. This resolution limit is counterbalanced, however, by the ability to image many more mice that have been subjected to well-defined manipulations. Again, the inevitable tradeoff is evident between increased measurement precision in large animals vs increased precision and generality made possible using larger numbers of small animals. INBREEDING AND ITS CONSEQUENCES A hallmark of most mouse models is that they usually rely on a particular inbred stock or strain. Most transgenic mice, knockouts, and spontaneous mutations are bred onto inbred stocks such as C57BL/6, FVB, and 129 strains. Inbred strains make it possible to perpetuate a mouse model essentially as a permanent clone. An inbred strain can be considered a sexually reproducing “clone” of a single genome in which the only residual variation is that required to perpetuate both males and females. An important distinction needs to be made between isogenic and monozygotic human twins and two inbred mice of the same sex. Identical twins have the same allelic diversity of other humans—they will often have two different alleles (the maternal and paternal allele) at genes and markers distributed across their entire genome. In contrast, inbred mice have been generated in a way that has eliminated nearly all allelic variation. The problem with using inbred strains as models or as genetic “hosts” for specific engineered gene variants is that inbred strains are very odd types of mammals. There is almost nothing like an inbred strain in the wild (cheetahs and other species that have gone through severe population bottlenecks may come close). Human populations are not inbred to any significant extent; the inbreeding coefficients even in human populations such as Hutterites are low (0.5). Production of an inbred strain requires just over 20 generations of consecutive matings between sibling males and females. Generating an inbred strain of mouse is therefore a 7–8 yr effort that is often not successful. Inbreeding greatly diminishes reproductive fitness; the deleterious effects of recessive alleles are exposed fully, and natural selection often expresses its disapproval. But in the relatively mild laboratory environment, 50–80% of incipient inbred strains reach the finish line, each permanently archiving a unique combination of gene variants inherited from the original source population as well as a small load of spontaneous mutations acquired and fixed during inbreeding itself. The complete lack of heterozygous loci in inbred strains has the advantage of making perpetuation of a defined genome and strain feasible almost indefinitely, but the cost is that these strains are idiosyncratic and highly variable in how they respond to all kinds of exogenous factors. This fact can be viewed in a positive light—variation among strains is a great source material for genetic studies—but it also means that the effects of knockouts and mutations will typically be highly dependent on the genetic

58

SECTION I / GENETICS

background (strain of mouse) on which a mutation is placed by breeding. The effects of a mutation when present on a C57BL/6J background cannot be generalized with much confidence to be similar to those of the same mutation on a 129X1/SvJ background. There is really nothing like a “normal” mouse; wild-type means that a single gene is in its putative normal or wild-type state. In other words, the mouse happens to be wild type at a single one of approx 25,000 genes. C57BL/6J is not a wild-type mouse—it is an inbred strain that contains hundreds of recessive alleles and frank mutations that have been locked into its genome. Even nominally “wild-type” strains such as CD-1 that are bred with specific avoidance of sibling matings are clearly not wild type; these mice are albinos and are often completely blind. When a pair of different inbred strains is bred to each other, they produce a filial 1 (F1) generation of offspring. The F1 progeny are still isogenic; males and females are genetically identical, just like their inbred parents. Unlike their parents, however, these F1 mice are heterozygous at all genes that differ between the original pair of parental strains. F1 animals, therefore, retain the advantage of being isogenic but also have the advantage of being much more similar to truly wild populations of mammals, including humans. A study of a knockout on a single F1 hybrid background is analogous to a study of a disease in a single human, and there is a compelling need to evaluate gene function in a broad sense and in the context of diverse genomes and environments. STRATEGIES FOR MAKING MOUSE MODELS From a genetic perspective, human populations are inherently complex, an outcrossed population in which millions of common polymorphisms are simultaneously segregating and assorting according to Mendel’s laws. Humans are also complex in terms of the highly variable environmental factors to which we are subjected, beginning with the formation of oocytes several decades before our own conception. Mouse geneticists have intentionally constructed mice with simple genomes (fully inbred), and researchers retain tight control over most key environmental factors (diet, pathogen exposure, social structure, temperature, and so on). The reason to make mouse models with a structure diametrically opposite to the human condition is that this structure allows powerful experimental advantages. For example, a single point mutation or a single gene knockout can be introduced onto the fixed genetic “background” of an inbred strain. Similarly, the genome can be untouched and one facet of the environment be varied. Differences in phenotypes between control mice and experimental mice can therefore be confidently attributed to effects of genetic or environmental factors. The reductionist approach using such models has been effective in revealing causes and effects. The statistics of single-factor experiments are simple, and a conventional t-test will usually give nearly optimal power to reject the null hypothesis. But the downside of reductionist models is precisely that they do not attempt to model the real-world genetic and environmental complexity of human populations. This problem has led to a serious critique not just of individual animal models (e.g., those used in cancer biology) but of the whole effort to generate animal models, even to study drug toxicology. GENERAL MODELS AND INTEGRATING ACROSS SCALES Biomedical researchers are beginning to face major challenges in integrating data across scales, from single base pairs to social structure. It is also important to integrate data across systems, departments, and institutional boundaries—for example, to consider

complex interactions between the immune system, neuroendocrine responses, metabolic networks, and cardiovascular performance. Integration across scale and systems is difficult and inefficient because of the inherently fractured and specialized nature of biomedical research. Few researchers gain mastery in a single field, let alone two or more, and generalists lose in the funding and publication competition with specialists. Most models necessarily reflect the highly honed specialty of researchers and are often of use only in a narrow field of application. There are solutions to this problem, but they are unfortunately more latent than real. The foremost solution is the common use of reference panels of inbred or isogenic mice that have broad utility in many areas of research. Usually a model is considered to be a single entity—a single genetically modified mouse, such as a particular knockout line or a single inbred strain with a tendency to develop lymphomas. But a whole panel of inbred strains of mice, rather than an individual strain, can also be considered a model with which to systematically explore variation in many traits and in many environments. In this case, researchers rely on the range of variation that has been intentionally or inadvertently captured in the reference panel. Human populations in clinical research often consists of what are essentially reference panels—for example, the Framingham cohort and the very well studied panel of 60 large families that form the core of the CEPH Family Panel used throughout the 1990s in gene mapping experiments. To clarify this distinction between single strain-single gene models and panels of strains, consider two contrasting approaches to studying the genetics of obesity using single mouse models or a reference population. Approach 1 To analyze phenotypes associated with a spontaneous or induced mutation in a specific gene that leads to marked obesity in mice, a single line on a single genetic background (e.g., a new mutant allele of the leptin receptor on the background of strain DBA/2J) is generated, optimized, and studied. The homozygous knockout is compared with the heterozygote and the wild-type strain (–/–, –/+, and +/+). The phenotypic differences across these three classes may provide exquisite biochemical detail and high power. However, results may not be easily generalizable to other strains of mice such as C57BL/6 or 129X1/Sv, let alone to other species and to human populations. Because the model is unique, data generated are not easily aggregated and integrated using this novel line with data from other experiments and models. Integration will be at a comparatively high level: do the overall conclusions agree or not with those of other studies? The simple solution to the first problem of generality is to make a habit of studying monogenic manipulations on a set of different genetic backgrounds. This is quite easy in mice, because changing the background may simply involve one or two generation breeding experiment (an outcross of the mutation to a different wild-type strain). Approach 2 The interest in the genetics of obesity continues, but now an entire reference panel of 30 or more genetically and phenotypically diverse strains of mice rather than the three classes of a Mendelian mutation is used. For example, variation in fat pad mass or leptin receptor protein level in hypothalamus across this panel could be assayed. Variation in this measurement across the panel is the “experimental” signal used for two types of subsequent analyses: (1) gene mapping studies of the chromosomal regions that contribute some fraction of the variation detected across the strains, and (2) association studies of variation in fat pad mass with hundreds of other phenotypes that have already

CHAPTER 8 / ANIMAL MODELS IN BIOMEDICAL RESEARCH

been acquired in this same panel under similar and different conditions. Results of hundreds of experiments can be pooled and integrated for the simple reason that the same genetic individuals— members of the reference panel—are being used by multiple research communities. For example, data on fat pad mass could be compared with pre-existing data for the same strains on variation in blood pressure or activity levels. Networks of phenotypes can readily be explored that covary in the same way that clinical researchers search for patterns of comorbidity among diseases. The price for this integrated approach is obvious: there is more phenotyping to contend with, and the variation of interest does not have an immediate genetic cause. The mapping studies may suggest that three or more different chromosomal regions contribute to differences in phenotypes. Getting to the point of understanding the biochemical basis of a variant phenotype is still comparatively difficult. There are several types of reference panels that share the attribute that their constituent strains are a stable resource. A major project of this type, the Collaborative Cross, involves interbreeding eight mouse strains to produce a set of about 1000 derivative strains, each with a unique but fixed set of allelic variants, phenotypes, and disease susceptibilities. Several smaller reference panels of this type typically consist of 8–80 strains of mice. For example, the BXD panel of recombinant inbred strains was made 30 yr ago by combining the genomes of two strains (C57BL/6J and DBA/2J) that have both been almost fully sequenced. A multiplicity of phenotype and genotype data from these BXD strains can be exploited in genetic mapping studies and can also be used in phenotype association studies to test the consistency, strength, specificity, and coherence of associations between an amazing variety of traits (see www.genenetwork.org for an example of this integrative approach). HUMANIZING MICE Imagine a line of mice genetically engineered to biochemically resemble a human (as much as possible, given the massive scaling differences and biometric consequences). Transgenic mice in which human genes are inserted into the mouse genome represent the first step in this direction; it is now possible to extend this idea to groups of genes or even whole organs. There has been substantial progress in generating new types of mice in which the endogenous immune system has been genetically extirpated. These mice can be engrafted with human hematopoietic stem cells that mature into lymphocytes and T- and B-cell subpopulations. Such humanized mice will be especially valuable in studying host–pathogen interactions in ways that would not otherwise be possible. Similar techniques are being developed to make mice that incorporate human liver and breast tissues, and this trend is expected to continue as more control is gained over tissue differentiation in vivo and in vitro. This process will inevitably have its own set of problems, but there will be inexorable progress in generating more faithful and useful models that can be used to study basic mechanisms and test the safety and efficacy of new treatments. There is a risk (highlighted by Clifton Leaf) that researchers enamored with their sophisticated animal models may fail to promptly consider clinical applications or may be sidetracked into marginally relevant areas with little short-term payoff. Yet, strong counterbalancing incentives provided by funding agencies, pharmaceutical companies, and research institutions drive researchers to translate their results into effective clinical practice. In fact, these incentives can be so compelling that investigators may leap from research results to clinical tests without the delay imposed by an intermediate analysis of other model organisms to test generality and species specificity.

59

CONCLUSION In light of costs and benefits and the urgent need for rapid advances, animal models are an essential component of biomedical research. Computer simulations, cell-based assays, and human association studies are all critical components of research, but they cannot substitute for the rigorous mechanistic insight into cause and effects that can be gained from expanded use of more sophisticated and general animal models. A failure to exploit these models to improve quality of life and to provide effective compassionate care would be negligent. There will be a constant tension between what is regarded as uniquely human characteristics vs common denominators shared with other species. The major challenge in using animal models is to generalize results with reasonable fidelity to humans. To get to this point will often require higher numbers of cases because more variables need to be includeed—both sexes, several ages, several genotypes, and perhaps even several environmental conditions and several species. Achieving such goals may involve a balanced analysis of several model organisms, each with its own unique experimental and practical advantages.

ACKNOWLEDGMENTS My thanks to Kathryn Graehl for critical reading and editing of this work. Supported in part by grants from NIAAA (INIA program), NIDA, NIMH and NIAAA (Human Brain Project), NCRR (Mouse BIRN), and NCI (Mouse Models of Human Cancer Consortium).

SELECTED REFERENCES Biobank. University of Manchester. Manchester, England (www.biobank. ac.uk). Accessed May 11, 2005. Brenner S. Nature’s gift to science. Nobel Lecture 2002; pp. 274–282. (www.nobel.se/medicine/laureates/2002/brenner-lecture.pdf). Accessed Jan. 14, 2006. Chesler EJ, Wang J, Williams RW, Manly KF. Web QTL: Rapid exploratory analysis of gene expression and genetic networks for brain and behavior. Nat Neurosci 2004; 7:485–486. Crick F, Koch C. The unconscious homunculus. In: Metzinger T, ed. The Neuronal Correlates of Consciousness. Cambridge, MA: MIT Press, 2000; pp. 103–110. deCODE Genetics. Reykjavik, Iceland (www.decodegenetics.com). Accessed May 11, 2005. Gudrais E. Chimpanzees and the law. Harv Mag 2003;105:21, 22. Kuperwasser C, Chavarria T, Wu M, et al. Reconstruction of functionally normal and malignant human breast tissues in mice. Proc Natl Acad Sci USA 2004;101:4966–4971. Leaf C. Why we’re losing the war on cancer. Fortune 2004;149:76–97. Little CC. The role of heredity in determining the incidence and growth of cancer. Am J Cancer 1931;15:2780–2789. Lusis AJ, West D, Davis C. Animal models of complex genetic disease. In: King RA, Rotter JI, Motulsky AG, eds. The Genetic Basis of Common Disease, 2nd ed. New York: Oxford University Press, 2002; pp. 65–86. Nagy A, Gertsenstein M, Vintersten K, Behringer R. Manipulating the Mouse Embryo: A Laboratory Manual, 3rd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 2003. National Academy Press: Guide for the Care and Use of Laboratory Animals. Washington, DC: National Academy Press, 1996. Paired Box Gene 6; PAX6. OMIM. Johns Hopkins University. Baltimore, Maryland (www.ncbi.nlm.nih.gov/entrez/dispomim.cgi?id=607108). Accessed May 11, 2005. Pauling L, Itano HA. Sickle cell anemia: a molecular disease. Science 1949;110:543–548. Quiring R, Walldorf U, Kloter U, Gerhing WJ. Homology of the eyeless gene in Drosophila to the small eye gene in mice and anirdia in humans. Science 1994;2655:785–789.

60

SECTION I / GENETICS

Rader K. The mouse’s tale: standardized animals in the culture and practice of technoscience. Cabinet Mag 2001;4. www.cabinet magazine. org/issues/ 4/themousestale.php). Accessed Jan. 14, 2006. Ryan TM, Townes TM, Reilly MP, et al. Human sickle hemoglobin in transgenic mice. Science 1990;247:566–568. Silver LM. Mouse Genetics. Concepts and Applications. New York: Oxford University Press, 1995.

Threadgill DW, Hunter KW, Williams RW. Genetic dissection of complex and quantitative traits: from fantasy to reality via a community effort. Mamm Genome 2002;13:175–178. Vogel G. Scientists dream of 1001 complex mice. Science 2003;301: 456, 457. Williams RJ. Biochemical Individuality. New York: John Wiley & Sons, 1998. Zerhouni E. The NIH roadmap. Science 2003;302:63.

9 Ethical, Legal, and Social Implications MARCIA VAN RIPER SUMMARY

GENETIC INFORMATION: CHANGES IN ACCESS AND USE

This chapter provides an overview of some of the complex ethical, legal, and social issues associated with advances in genomics and molecular medicine, such as equity in access to genetic information and services, use of genetic information in nonhealth-related settings, and genetic discrimination. It is hoped that this discussion will stimulate future discussions about these and other issues that are just beginning to emerge in the exciting, ever-changing world of genomic health care. Discussions such as these are critical to the development of appropriate policies and legislation.

The term “genetic information” has been used in a variety of ways. Commonly used to indicate the information obtained from a specific genetic test, it may also be used to refer to information about a specific genetic condition, or information about a person’s entire genome. Regardless of which definition is used, the term genetic information tends to elicit a wide variety of emotions, including excitement, wonder, relief, fear, sadness, and anger. CHANGES IN ACCESS Access to Genetic Information Through Genetic Professionals Previously, the primary way to gain access to genetic information was through interactions with genetic professionals (geneticists, genetic counselors, and nurses with expertise in genetics). Typically, genetic professionals follow a genetic services paradigm that includes the construction of a family pedigree, a risk assessment, education about the genetic component of specific conditions, psychosocial assessment and counseling, a discussion of testing and management options, and support with decision making. For many conditions, no treatments exist, so management options may be limited to surveillance and riskreducing strategies. If genetic testing is available, genetic professionals usually provide both pre- and post-test counseling to give the individual or family who is considering genetic testing a chance to discuss possible ethical, legal, and social implications (ELSI) of testing. It also gives the genetic professional a chance to anticipate possible negative consequences of testing. If it appears that the individual or family might have difficulty dealing with the test results, additional counseling, and support can be provided. Access to Genetic Information Through Other Sources There are a variety of ways to gain access to genetic information. Genetic services, especially genetic testing, are becoming increasingly available in mainstream health care. In addition, advances in technology and mass communication have made it possible for many people to access genetic information in their homes. The Internet is rapidly becoming a major source of genetic information. Although many individuals are using web-based resources to expand their understanding of specific genetic conditions, others are using Internet chat rooms and discussion boards to help in their decision making about genetic testing and management options, and some are receiving ongoing information and support through websites devoted to specific conditions. Moreover, some individuals and families are becoming aware of and gaining access to genetic testing through companies that offer direct-to-consumer marketing.

Key Words: Discrimination; ethics; genetics; genetic information; genome; pharmacogenomics.

INTRODUCTION Clinical practice has been dramatically altered by advances in genomics and molecular medicine, such as the sequencing of the human genome. Although it will probably be at least 10 yr before one’s entire genome will be sequenced for $1000 or less, there have been important advances in the application of genomic information to the prevention, diagnosis, and treatment of human disease. Some of these advances include the use of preimplantation genetic diagnosis to help families affected by devastating genetic conditions have unaffected children; the use of genotyping to stratify patients according to their risk of specific diseases, such as long-QT syndrome; the use of geneexpression profiling to assess prognosis and guide treatment decisions for women with breast cancer; and the use of pharmacogenomics to tailor medications and dosages for individual genotypes. Although advances such as these are promising and may ultimately transform the model of population-based risk assessment and empirical treatment into a predictive individualized model based on the molecular classification of disease and targeted treatment, they also have the potential to complicate the web of ethical, legal, and social issues already associated with advances in genomics and molecular medicine. This chapter provides an overview of the ethical, legal, and social issues connected with advances in genomics and molecular medicine.

From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

61

62

SECTION I / GENETICS

Ethical, Legal, and Social Issues Associated With Changes in Access Changes in access to genetic information increase the possibility that individuals and families from diverse backgrounds will be able to gain access to genetic information, but these changes also increase the possibility of ethical, legal, and social issues. For example, if an African-American woman with a strong family history of breast cancer is offered BRCA1/2 testing, she can give informed consent only after (1) she receives full disclosure of pertinent information concerning the test and (2) she fully understands the risks, benefits, and limitations of the test. If she undergoes the testing in mainstream health care, this is less likely to occur because many physicians in mainstream health care lack the time and expertise necessary to adequately counsel their patients about BRCA1/2 testing. In addition, the information she receives may not be clearly understandable or culturally sensitive. Also, depending on her socioeconomic status, she may not have the financial resources or insurance coverage needed to undergo testing. Concerns about cost, confidentiality, and insurance discrimination may prevent her from being tested. In a study about BRCA1/2 testing in the community setting, 82% of the 646 women who underwent testing recalled discussing and reviewing the consent with their primary care provider. However, the time spent on this (median time spent on counseling and informed consent was 30 min, range was from 1 to 240 min) was much less than is usually devoted to this activity in specialty clinics and research programs. Almost 13% of the women did not recall discussing the consent form. Concerning how they became aware of their test results, 57% received their results during an office visit, 39% received them by telephone, 3% got their results in the mail, and 2% did not answer the question. Three of those who received their results in the mail had variants of uncertain significance. Whereas most of the women (67%) reported that the testing was paid for by their health insurance or through research funds, 25% paid for it themselves. The remaining women either had insurance claims pending or they did not report a source of payment. None of the women reported increased premiums or loss of health or life insurance because of undergoing the testing. These findings need to be viewed with caution because it had been less than 2 mo since they received their test results. As with access to genetic information through physicians in mainstream health care, a number of ethical, legal, and social issues are associated with access to genetic information through the Internet. The first concern is that the information provided on these websites may be incorrect or misleading. Next, there are concerns about privacy and confidentiality, especially if the sharing of personal information is a prerequisite to gaining access to a website. Also certain websites can be used to purchase medications, dietary supplements, and genetic tests without the supervision of a health-care provider. Findings from a study about direct-to-consumer sales of genetic services on the Internet suggest that most of the websites offered genetics services such as paternity testing, identity testing, and DNA banking. Only 14 out the 105 websites reviewed offered health-related tests. Although some of these tests were conventional genetic tests (hemochromotosis and cystic fibrosis), others related to nutrition, behavior, and aging. The authors of this study concluded that the availability of direct-to-consumer health-related genetics tests creates the potential for inadequate decision making prior to testing, a misunderstanding of test results, and access to genetic tests of questionable clinical value.

GeneWatch UK, a nonprofit group that monitors developments in genetic technologies noted a number of problems with the approach taken by one company that offers to provide products (lotions and other skin-care products) and services (guidance on training methods and nutrition requirements) that are designed to match an individual’s genetic profile. According to GeneWatch UK, (1) the company’s claims are misleading, (2) the company does not adequately inform their customers about possible risks associated with the testing, (3) the company plans to keep the DNA samples and link them to personal genetic information and lifestyle information as long as the customer continues to be a subscriber, (4) the customer’s genetic information may ultimately be patented without their knowledge, and (5) the customer may experience negative consequences, such as financial burden because of the high cost of products that may not be necessary. Their analysis would suggest that individuals who use this source to gain access to their genetic information may experience a lack of informed consent, a violation of their right not to know, loss of privacy, workplace and insurance discrimination, confidentiality issues, and unnecessary financial burden. CHANGES IN USE Use of Genetic Information to Improve Health and Well-Being Initially, the primary use of genetic information was to help individuals and families affected by genetic conditions face the associated ongoing challenges. Once genetic testing became available for select conditions, possible uses for genetic information increased dramatically. For example, expectant couples can use the genetic information that they acquire through pre- and postaminocentesis counseling to help them prepare for the birth of a child with Down syndrome or the termination of an affected pregnancy. New parents can use the genetic information they receive following their child’s positive newborn screen for phenylketonuria to start their child on a special formula known to prevent the mental retardation commonly associated with phenylketonuria. Individuals found to carry the gene for Huntington disease (HD) can use genetic information in their decision making about reproduction, as well as their decision making about how and where they want to live the rest of their life. Before the availability of genetic testing for familial adenomatous polyposis, children in families with a history of familial adenomatous polyposis were usually started on close surveillance, which comprised annual sigmoidoscopies, starting at around age 10. Because genetic testing is available, results of the test can be used to determine which children need this type of close surveillance. With the development of risk-reducing strategies and treatments for select genetic conditions, possible uses of genetic information to improve health and well-being have increased again. For example, as data came to light indicating that an individual’s short-term risk of developing a second breast cancer is substantially affected by whether she carries one of the BRCA mutations associated with hereditary breast and ovarian cancer, newly diagnosed breast cancer patients are increasingly being asked to consider undergoing BRCA1/2 testing before they make decisions about their treatment options. This way, they can use the results of their genetic test to help in decision making about possible treatment options. Women who test positive and choose to undergo a bilateral mastectomy rather than a breast-conserving procedure can avoid radiation treatment and possibly a second surgery. For some, an even more important use of their test results is to help their offspring, by making them aware of any increased risk.

CHAPTER 9 / ETHICAL, LEGAL, AND SOCIAL IMPLICATIONS

The ability to test for a specific gene mutation far exceeds the ability to treat the genetic condition associated with this mutation. In the early 1990s there was a great deal of optimism about the possibility of using genetic information to provide quick solutions to a long list of health problems. Unfortunately, the field of gene therapy, also known as gene transfer, has sustained a number of major disappointments. One of the most upsetting events was the death of an 18-yr-old volunteer, Jesse Gelsinger, in a gene therapy trial in the fall of 1999. The death of Jesse Gelsinger drew particular attention to the safety of gene therapy because unlike most other volunteers in gene therapy trials, Jesse was not seriously ill. He had a less severe form of an inborn error of metabolism called ornithine transcarbamylase deficiency. He chose to volunteer for the experiment because he wanted to try to help the children who die of the condition. Jesse died 4 d after receiving the gene therapy. Although the early optimism about gene therapy was probably never fully justified, it is likely that the development of safer and more effective methods for gene delivery will ensure a significant role for gene therapy in the treatment of some diseases. Major challenges include targeting the right gene to the right location in the right cells; expressing the transferred gene at the right time; and minimizing adverse reactions. One use of genetic information that has received a great deal of attention during the past decade is pharmacogenomics. There has been speculation that pharmacogenomics may become part of standard practice for a large number of disorders and drugs by 2020. The expectation is that by identifying common variants in genes that are associated with the likelihood of a good or bad response to a specific drug, drug prescriptions can be individualized based on the individual’s unique genetic make-up. A primary benefit of pharmacogenomics is the potential to reduce adverse drug reactions. Considering that adverse drug reactions result in significant morbidity, mortality, and excess medical costs, pharmacogenomics could prove to be important. Use of Genetic Information in Nonhealth-Care Setting Genetic information is being used or has been considered for use in a wide variety of settings other than health care, such as insurance, employment, education, adoption, and civil litigation. Insurers argue that they need access to genetic information so they can stop individuals with known mutations from “loading up” on insurance at the regular rates. Employers argue that they need access to genetic information so that they can prevent unnecessary injury or death. For example, a trucking company does not want one of their employees to be symptomatic for HD or Alzheimer’s disease. One place in which genetic information is being used extensively is in the criminal justice system. DNA profiling is routinely used to find and convict criminals. In 1995, the first national DNA database was established in the United Kingdom. This database holds about 2.1 million DNA samples from individuals and 215,000 profiles from crime scenes. In a typical month, matches are found linking suspects to 30 murders, 45 rapes, and 3200 other crimes. Genetic information is also being used to establish the innocence of individuals believed to be wrongly convicted. Two states, Illinois and Maryland, implemented a death penalty moratorium because of growing concern about the execution of wrongly convicted individuals. However, the moratorium has since been lifted in Maryland. At least eight states have enacted laws to guarantee that inmates have access to DNA testing to help establish their innocence.

63

Ethical, Legal, and Social Issues Associated With Changes in Use Use of the over 1000 available genetic tests is much less than was anticipated. When genetic testing for HD first became available, it was anticipated that demand for testing would be high (approx 60–70%). Actual use has been 5–25% of the people at risk for HD. A variety of reasons have been given for nonparticipation in genetic testing, such as lack of interest, taboos about the disease in the family, family conflicts, lack of information about the testing, and an inability to cope with the results, but the most commonly expressed reason for nonparticipation is fear of genetic discrimination. Despite limited empirical evidence demonstrating the existence of genetic discrimination, many people continue to fear that genetic information will be used in ways that could harm them. That is, they fear genetic information will be used to deny them access to insurance, employment, and education. In addition, there is concern that adoption agencies may refuse to accept a couples’ adoption request if they learn that one of the prospective parents has tested positive for certain genetic mutations. People who are concerned about genetic discrimination may avoid seeking reimbursement from their health insurer for testing, or they may seek reimbursement only if the test result is favorable. Some people chose to undergo anonymous testing. The people who are at greatest risk for health insurance discrimination are those who purchase health insurance on their own. These people have to undergo the greatest scrutiny. For people who are covered by government programs (Medicare, Medicaid, Military) or employers who have greater than 20 employees, there is actually little or no risk of genetic discrimination. The most widely publicized example of genetic discrimination in the workplace is a case involving Burlington Northern Santa Fe Railroad (BNSF). Employees of BNSF who had developed carpal tunnel syndrome and were seeking disability compensation were tested for a genetic mutation associated with hereditary neuropathy. Allegedly the testing was done based on the advice of the company’s physician, who supposedly had relied on information he received from the diagnostic company who performed the test. Employees who were tested were not aware of what they were being tested for. Although the company’s motive for doing the testing was never made explicit, it was hypothesized that BNSF planned to deny disability benefits to any employee who had the specific genetic mutation. Soon after this practice came to light, it was stopped and shortly thereafter the company settled employee claims for an undisclosed amount of money. The use of genetic information is shaped, and in many instances constrained, by factors such as social norms, where care is received, and socioeconomic status. Most pregnant women in the United States have at least one ultrasound, many undergo some type of multiple marker screening, and a growing number undergo other types of prenatal testing. The range of prenatal testing options available to a pregnant woman and her family varies significantly based on where she receives prenatal care and her socioeconomic status. Certain types of prenatal testing may not be available in smaller communities and rural settings (e.g., chorionic villus sampling; fluorescent in situ hybridization analysis). In addition, certain types of genetic testing may not be offered in conservative medical communities (e.g., preimplantation diagnosis). Concerning socioeconomic status, some types of genetics testing are expensive and typically not covered by health insurance. Because of this, these tests may only be available to a relatively small number of individuals and families, people who can

64

SECTION I / GENETICS

afford to pay for them. This issue needs to be addressed; ideally, there should be equity in access to care. Equity in access to care is also an issue in pharmacogenomics. The idea of tailor-making medications for individual genotypes goes against the model of mass-producing medications that are suitable and safe for the widest possible range of people at competitive prices. Inequalities are likely to develop between those individuals who have adequate resources and those who do not. Individuals who have adequate resources may be the only ones who can afford the tailor-made “designer drugs.” The potential for insurance discrimination is another possible ethical issue with pharmacogenomics. Insurers may try to use the results of tests for drug reactions to exclude or charge higher premiums for those individuals who need the more expensive medications. Pharmaceutical companies may need to include additional warnings on their labels if they want to avoid class action suits from individuals who have genotypes that increase their susceptibility to adverse reactions. Critics of pharmacogenomics say it represents a misallocation of scarce resources. They argue that rather than wasting a country’s scarce resources on pharmacogenomics, these resources should be devoted to solving more urgent problems facing humanity such as global famine and accessibility to clean water.

EFFORTS TO ADDRESS ETHICAL, LEGAL, AND SOCIAL ISSUES ETHICAL, LEGAL, AND SOCIAL IMPLICATIONS PROGRAMS AT THE NATIONAL HUMAN GENOME RESEARCH INSTITUTE AND THE DEPARTMENT OF ENERGY Before the beginning of the Human Genome Project (HGP), widespread concern about misuse of the genetic information gained through the HGP resulted in 3–5% of the HGP budget being designated for the study of ELSI of human genome research. Two large ELSI programs were created to identify, analyze, and address ELSI of human genome research at the same time while the basic science issues were being studied. One ELSI program was established in the National Human Genome Research Institute at the National Institutes of Health and another was established in the Office of Biological and Environmental Research at the Department of Energy. The two ELSI programs are separate, but complementary. High priority issues for these programs are those surrounding the completion of the human DNA sequence and the study of human genetic variation; issues raised by the integration of genetic technologies and information into healthcare and public health activities; issues raised by the integration of knowledge about genomics and gene environment interactions into nonclinical settings; ways in which new genetic knowledge may interact with a variety of philosophical, theological, and ethical perspectives; and how socioeconomic factors and concepts of race and ethnicity influence the use and interpretation of genetic information, the utilization of genetic services, and the development of policy. In 2004, the ELSI program at the National Human Genome Research Institute provided funding for four Centers of Excellence in ELSI Research, as well as a number of planning grants for Centers of Excellence in ELSI Research. It also called for research proposals on intellectual property rights in genetics and genomics research and development, and the effect of such laws and policies on progress in these fields and on commercialization, drug development, health-care delivery, and the public health.

LEGISLATION ON GENETIC DISCRIMINATION Although many states have enacted antidiscrimination laws, the need for federal legislation has grown as the Supreme Court has consistently narrowed the protection provided under the Americans with Disabilities Act. There have been a number of attempts to draft federal legislation on the issue of genetic discrimination in the United States, but this is likely to be an ongoing process. For example, S. 1053, “A bill to prohibit discrimination on the basis of genetic information with respect to health insurance and employment,” passed the US Senate by a vote of 95 to 0 on October 14, 2003. Then, in February 2005, the US Senate passed the Genetic Information Nondiscrimination Act of 2005 (S. 306). The two bills (S. 1053 and S. 306) are identical. In March 2005, a bipartisan group of over 100 members of Congress introduced a bill, “Genetic Information Nondiscrimination Act of 2005,” HR 1227, in Congress. Again this bill is identical to S. 1053. As of October 2005, this bill has yet to be passed in Congress.

CONCLUSION The ethical, legal, and social issues discussed in this chapter are only some of the complex issues associated with advances in genomics and molecular medicine. It is hoped that this discussion will stimulate future discussions about these and other issues that are just beginning to emerge in the exciting, ever-changing world of genomic health care. Discussions such as these are critical to the development of appropriate policies and legislation.

SELECTED REFERENCES Brower V. Genomics and health care. How genomics medicine is translated into better health care largely depends on how physicians handle this information. EMBO Rep 2004;5:131–133. Chen WY, Garber JE, Higham S, et al. BRCA1/2 genetic testing in the community setting. J Clin Oncol 2002;20:4485–4492. Christian SM, Kieffer SA, Leonard NJ. Medical genetics and patient use of the Internet. Clin Genet 2001;60:232–236. Clayton EW. Ethical, legal, and social implications of genomic medicine. N Engl J Med 2003;349:562–569. Collins FS, Green ED, Guttmacher AE, Guyer MS, US National Human Genome Research Institute. A vision for the future of genomics research. Nature 2003;422:835–847. Collins FS, Guttmacher AE. Genetics moves into the medical mainstream. JAMA 2001;286:2322–2324. Collins FS, McKusick VA. Implications of the Human Genome Project for medical science. JAMA 2001;285:540–544. Forensic Science Service. Fact sheet: The National DNA Database (NDNAD). http://www.forensic.gov.uk/forensic_t/inside/news/docs/ NDNAD.doc. Accessed July 20, 2004. Freedman AN, Wideroff L, Olson L, et al. US physicians’ attitudes toward genetic testing for cancer susceptibility. Am J Med Genet 2003;120A: 63–71. GeneWatch UK. Genetic testing and the body shop. http://www.genewatch. org/HumanGen/Tests/BodyShop.htm. Accessed July 20, 2004. Gollust SE, Wilfond BS, Hull SC. Direct-to-consumer sales of genetic services on the internet. Genet Med 2003;5:332–337. Guttmacher AE, Collins FS. Welcome to the genomic era. N Engl J Med 2003;349:996–998. Hampton T. Breast cancer gene chip study under way: can new technology help predict treatment success? JAMA 2004;291:2927–2930. Hartmann LC, Schaid DJ, Woods JE, et al. Efficacy of bilateral prophylactic mastectomy in women with a family history of breast cancer. N Engl J Med 1999;340:77–84. Hartmann LC, Sellers TA, Schaid DJ, et al. Efficacy of bilateral prophylactic mastectomy in BRCA1 and BRCA2 gene mutation carriers. J Natl Cancer Inst 2001;93:1633–1637. Hayden MR. Predictive testing for Huntington’s disease: the calm after the storm. Lancet 2000;356:1944, 1945.

CHAPTER 9 / ETHICAL, LEGAL, AND SOCIAL IMPLICATIONS

Khoury MJ. Genetics and genomics in practice: the continuum from genetic disease to genetic information in health and disease. Genet Med 2003;5:261–268. Meijers-Heijboer H, van Geel B, van Putten WL, et al. Breast cancer after prophylactic bilateral mastectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med 2001;345:159–164. Miller PS. Genetic discrimination in the workplace. J Law Med Ethics 1998;26:189–197. O’Neill O. Informed consent and genetic information. Stud Hist Philos Biol Biomed Sci 2001;32:689–704. Peterson EA, Milliron KJ, Lewis KE, Goold SD, Merajver SD. Health insurance and discrimination concerns and BRCA1/2 testing in a clinic population. Cancer Epidemiol Biomarkers Prev 2002;11:79–87. Priori SG, Schwartz PJ, Napolitano C, et al. Risk stratification in the longQT syndrome. N Engl J Med 2003:348:1866–1874. Ramaswamy S. Translating cancer genomics into clinical oncology. N Engl J Med 2004;350:1814–1816. Robertson JA. The $1000 genome: ethical and legal issues in whole genome sequencing of individuals. Am J Bioeth 2003;3:W35–W42. Roses AD. Pharmacogenetics and the practice of medicine. Nature 2000;405:857–865.

65

Schwartz MD, Lerman C, Brogan B, et al. Impact of BRCA 1/BRCA 2 counseling and testing on newly diagnosed breast cancer patients. J Clin Oncol 2004;22:1823–1829. Sermon K, Van Steirteghem A, Liebaers I. Preimplantation genetic diagnosis. Lancet 2004;363:1633–1641. Singer E, Antonucci T, Van Hoewyk J. Racial and ethnic variations in knowledge and attitudes about genetic testing. Genet test 2004;8:31–43. Thompson HS, Valdimarsdottir HB, Jandorf L, Redd W. Perceived disadvantages and concerns about abuses of genetic testing for cancer risk: differences across African American, Latina, and Caucasian women. Patient Educ Couns 2003;51:217–227. van de Vijver MJ, He YD, van’t Veer LJ, et al. A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med 2002;347: 1999–2009. Weitzel JN, McCaffrey SM, Nedelcu R, MacDonald DJ, Blazer KR, Cullinane CA. Effect of genetic risk assessment on surgical decisions at breast cancer diagnosis. Arch Surg 2003;138:1323–1328. Wertz DC. Ethical, social and legal issues in pharmacogenomics. Pharmacogenomics J 2003;3:194–196. Williams-Jones B. Where there’s a web, there’s a way: commercial genetic testing and the Internet. Community Genet 2003;6:46–57.

CARDIOLOGY SECTION EDITOR:

ANTHONY ROSENZWEIG

II

Abbreviations II. CARDIOLOGY 4E-BP1 AAA ABCA1 AC ACE ACS ACTH ADP AGT Ang-I Ang-II APA Apo A-I Apo B ARH ARVD ATPase β-MHC CABG CAD cAMP cbEGF CETP cGMP CGRP CHD CK cMyBPC CRP CsA cTnT CVD DCM DGS DMD EAD E-C ECM eEF2K EGF eIF ENaC eNOS EP EPCs ERK

ET-1 FBN1 FDB FH FISH Gi GPCR GRA GSK-3 HCM HDL HMG CoA ICAM-1 ICD IDL IGF-1 ISA JNKs KO LCAT LDL LDLR-/LMWH LV LVAD LVH MAP MAPK MAS MCIP MCP-1 MEK

endothelin-1 fibrillin-1 gene familial defective apolipoprotein B familial hypercholesterolemia fluorescence in situ hybridization G inhibitory guanine nucleotide protein coupled receptor glucocorticoid-remediable hyperaldosteronism glycogen synthase kinase-3 hypertrophic cardiomyopathy high-density lipoprotein 3-Hydroxy-3-methyl-glutaryl coenzyme A intercellular adhesion molecule-1 implantable cardiac defibrillator intermediate density lipoproteins insulin-like growth factor-1 intrinsic sympathomimetic activity c-Jun N-terminal kinases knockout lecithin cholesterol acyl transferase low-density lipoprotein low-density lipoprotein receptor deficient low-molecular-weight heparin left ventricular left-ventricular assist device left-ventricular hypertrophy mitogen-activated protein mitogen-activated protein kinase marker-assisted selection muscle-enriched calcineurin inhibitory protein monocyte chemotactic protein-1 mitogen-activated protein and extracellular signal-regulated kinases MEN-2 multiple endocrine neoplasia 2 MFS Marfan syndrome MHC myosin heavy chain MI myocardial infarction MLP muscle LIM protein MMP matrix metalloproteases mTOR mammalian target of rapamycin NBD nucleotide binding domains NCEP National Cholesterol Education Program NFAT nuclear factor of activated T cells NHANES III Third National Health and Nutrition Examination Survey NO nitric oxide NOS nitric oxide synthase PAI-1 plasminogen-activator inhibitor

4E-binding protein 1 abdominal aortic aneurysms ATP binding cassette transporter A1 adenylate cyclase angiotensin-converting enzyme acute coronary syndrome adrenocorticotropic hormone adenosine diphosphate angiotensinogen angiotensin I angiotensin II aldosterone-producing adrenocortical adenoma apolipoprotein A-I apolipoprotein B autosomal-recessive hypercholesterolemia arrhythmogenic right ventricular dysplasia adenosine triphosphatase β-myosin heavy chain coronary artery bypass surgery coronary artery disease w/e (cyclic adenosine monophosphate) calcium-binding epidermal growth factor-like cholesterol ester transfer protein cyclic guanosine monophosphate calcitonin gene-related peptide coronary heart disease creatine kinase cardiac myosin binding protein C C-reactive protein cyclosporine A cardiac troponin T cardiovascular disease dilated cardiomyopathy DiGeorge syndrome Duchenne muscular dystrophy early after depolarization excitation-contraction extracellular matrix eukaryotic elongation factor-2 kinase epidermal growth factor eukaryotic initiation factor epithelial sodium channels endothelial nitric oxide synthase electrophysiology endothelial progenitor cells extracellular signal-regulated kinase

67

68 PCI PDK1 PI3-K PKA PKB PKC PRKAG2 PTEN PTX QTL RyR SCD SDH SHR SIDS SMC

SECTION II / CARDIOLOGY

percutaneous coronary intervention phosphoinositide-dependent protein kinase-1 phosphoinositide 3-kinase protein kinase A protein kinase B protein kinase C γ-2 regulatory subunit of AMP-activated protein kinase phosphatase and tensin homolog pertussis toxin quantitative trait loci ryanodine receptor sudden cardiac death succinate dehydrogenase spontaneously hypertensive rat sudden infant death syndrome smooth muscle cells

SNP SNPs SP SR SSRE STAT STE SVAS TIMP TOF t-PA u-PA VCAM-1 VCFS VEGF VHL VLDL

single-nucleotide polymorphism single-nucleotide polymorphisms substance P sarcoplasmic reticulum shear stress response element signal transducer and activator of transcription ST elevation supravalvar aortic stenosis tissue inhibitors of MMP Tetralogy of Fallot tissue plasminogen activator urokinase-type plasminogen activator vascular cell adhesion molecule-1 Velo-cardio-facial syndrome vascular endothelial growth factor von Hippel-Lindau syndrome very low-density lipoproteins

10 Congenital Heart Disease LAZAROS K. KOCHILAS AND ALVIN J. CHIN

85% of these infants can expect to survive to adulthood. However, congenital heart defects are the leading cause of death during the first 5 yr of life and are associated with significant morbidity in neonatal life and beyond. Because only the most severe structural malformations thwart management schemes, prevention and early in utero detection are the remaining strategies to reduce the impact of heart malformation on the health of infants and children; however, any preventive approach would depend largely on knowing the percentage of congenital heart defects that are genetic. For example, if most congenital heart anomalies result from intrauterine insults (infectious agents, toxins, nutritional factors, mechanical stress, and so on) to the embryo during the first 30 d of gestation, then a prevention strategy would be ineffective, and improving fetal imaging skills should be a focus of effort; because most pregnancies are already monitored by ultrasound at least once for accurate dating, perhaps adding a “cardiac surveillance” portion to the imaging protocol would suffice to increase the chance of prenatal identification. However, evidence suggests that most congenital heart malformations result from gene alterations. Therefore, the scientific challenge is to unravel the sequence of molecular decisions that result in the construction of the heart and blood vessels from the first embryonic tissue layers. Understanding the development of the cardiovascular system in more detail will allow exploration of novel interventional strategies for treatment or prevention of disease. In vertebrates studied, the heterogeneity of phenotypes associated with even single-gene null mutations points to the importance of modifier genes, environmental factors, and genetic polymorphisms in determining the severity and type of congenital heart disease. It will be important to identify these “modifier genes” because they may partially explain the variability of defects in human conditions such as the DiGeorge syndrome (DGS), usually associated with a multigene deletion. Much of the variability in diseases caused by point mutations occurs from the generation of hypomorphic alleles of different strengths. It is also important to study the effects of environmental factors like toxins, nutrients, and alcohol that are similarly implicated in cardiac development. More than 150 genes are involved in cardiovascular morphogenesis (Table 10-1), but how they function to form the heart and great vessels remains obscure. Both forward and reverse genetic approaches are being investigated, and a variety of organisms are being scrutinized because the underlying mechanisms of patterning appear to be widely shared among vertebrates. Among these the mouse has been the classic model for cardiac genetic studies because

SUMMARY Cardiogenesis is a complex process involving different cell types, such as muscle, endothelial, neural crest, and matrix cells. These cells follow a “protocol” that emerges through changes in gene expression induced by developmental and mechanical cues. Data from human genetics and animal mutants suggest that most congenital heart malformations are arise from gene alterations. The next challenge will be to unravel the sequence of molecular decisions that result in the formation of heart and blood vessels from the first embryonic tissue layers. This knowledge is expected to result in novel strategies for diagnosis, treatment or prevention of heart diseases. Key Words: Cardiac development; congenital heart defects; endocardium; genes; mutations; myocardial cells; neural crest cells; transcription factors.

INTRODUCTION The heart is the first organ to form and function during mammalian organogenesis. The formation of the heart and vasculature is an extremely complex process because of its multicomponent constituency, with the heart at its center and the vascular network at its periphery. A diversity of cell types including muscle, endothelial, neural crest, and matrix secretory cells have to follow a strictly coordinated “protocol” to build the normal system. Underlying the complexity at the macroscopic level is corresponding complexity at the molecular level. In the human cardiovascular system, thousands of genes are expressed in any given single cell. Although a significant proportion of these genes are devoted to maintenance of basic cellular function, a subset of them has more restricted expression (spatial and temporal) and contributes to the cardiovascular system’s diversity of cell types. The development of the cardiovascular system as a whole is the end product of the changes in gene expression in response to developmental and mechanical cues that drive growth, pattern formation and differentiation. Cardiac malformations occur in 5–8 of 1000 live births, and worldwide approx 1 million infants are born annually with heart defects (20,000/yr in the United States). Sufficient progress has occurred in identifying, characterizing, and surgically repairing physiologically important congenital cardiac defects so that From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

69

Table 10-1 Genes Involved in Cardiovascular Development Protein family function

Gene name

1. Gene/protein expression Transcription factors NK Nkx2.5

Tinman

70

GATA

GATA-4

Pannier (pnr)

GATA-5 FOG

Fog2

Organism

Human: Heterozygous mutations: ostium secundum type ASD, TOF, Ebstein’s anomaly, AV conduction defects and “left ventricular noncompaction” Mouse: Nkx2.5+/– : ostium secundum type ASD, AV conduction defects and vulnerability to arrhythmia Nkx2.5–/–: embryonic lethal at E10-E11 from cardiac insufficiency; cardiac developmental arrest at the stage of looping; poor chamber differentiation and trabeculation pattern Drosophila: tinman mutant embryos lack heart or visceral muscles Human: Heterozygous GATA-4 missense mutations are associated with familial VSDs Mouse: GATA-4–/–: embryonic lethal at E8.5-E10.5 with cardiac bifida and absent pericardial cavity Drosophila: Pnr mutant embryos have excess pericardial cells and deficient development of cardioblasts Zebrafish: Fau (GATA-5 mutants): cardiac bifida Human: Heterozygous FOG2 missense mutations were reported to be associated with congenital diapharagmatic hernia and lung hypoplasia, but no cardiac defects Mouse: Fog2–/–: embryonic lethal at midgestation with various cardiac defects characterized by thin ventricular myocardium, endocardial cushion defects (common atrioventricular canal), conotruncal abnormalities and hypoplastic coronary arteries

Expression sites and stages of involvement

Other family genes with potential role in cardiac development

Nkx2.5 is expressed in myocardial precursors (mouse, Xenopus, zebrafish) and is required for completion of cardiac looping morphogenesis.

Nkx2.3, Nkx2.6, Nkx2.7, and Nkx2.8 (various species) Bagpipe (Drosophila) Bax: Nkx3.1 (mouse)

GATA-4 is expressed in endocardium, endocardial cushions and myocardium except distal outflow tract (mouse) and is required for ventral migration and fusion of cardiac primordia.

GATA-5, GATA-6 (mouse): display overlapping but distinct spatio-temporal expression patterns during cardiogenesis. GATA6 is dispensable for early cardiac development but may play a role in later cardiomyocyte differentiation.

pannier is expressed in the dorsal mesoderm and required for cardial cell formation, while repressing a pericardial cell fate.

Expressed throughout the myocardium and is important for the normal looping and septation of the heart (mouse).

U-shaped (ush) zfh-1

T-BOX

T/Brachyury/ntl/ Xbra

Dorsocross (Doc 1/2/3) Tbx1

71 Tbx5

Drosophila: Ush mutants lack heart formation. Zfh-1 mutants form a heart tube that is specifically missing the even-skipped (eve) –expressing subset of pericardial cells Mouse: T–/–: randomization of cardiac looping and orientation Zebrafish: no tail–/– mutants have randomization of cardiac orientation Xenopus: Xbra overexpression or inactivation causes randomization of heart looping Drosophila: Combined mutants for Doc1-3 lack cardioblasts Human: TBX1 is included in the DGCR, whose haploinsufficiency is associated with DiGeorge syndrome (DGS) Mouse: Tbx1+/–: truncus arteriosus, IAA, RAA, aberrant RSCA and other aortic arch defects Tbx1–/–: perinatally lethal and occasionally embryonic lethal at E13.5-E14.5; DGS-like features with a range of conotruncal and aortic arch defects including truncus arteriosus, IAA, aberrant RSCA, VSD, TOF, TOF/Pulmonary atresia and branch pulmonary stenosis Tbx1 transgenics with Tbx1 overexpression display also conotruncal and arch anomalies Zebrafish: vgo/tbx1–/–: bilateral arch anomalies Human: HOS-1 (Tbx5): Heterozygous mutations are associated with Holt–Oram syndrome (limb defects and range of cardiac abnormalities: AV block, failure of septa formation and HLHS, aortic stenosis, mitral valve prolapse, and TOF) Mouse: Tbx5+/–: 50–80% perinatal lethality; ASD, VSD, limb defects

U-shaped promotes heart development, by maintaining tinman expression in the cardiogenic region. zfh-1 is expressed in the early mesoderm and later in the forming heart. Brachyury is not expressed in the vertebrate heart.

Tbx2/3 (human, mouse, chicken), Tbx4 (mouse) xEomes (Xenopus) Tbx18 (mouse, zebrafish)

Doc1-3 are expressed in distinct areas of the cardiogenic mesoderm and the dorsal vessel in Drosophila. Tbx1 is expressed in cardiac crescent, cardiac outflow tract, mesodermal core of pharyngeal arches (mouse). In zebrafish tbx1 expression is found in sinus venosus, atrium, ventricle and outflow tract.

Tbx5 is expressed in embryonic epicardium, myocardium, and atrioventricular tissue; endocardial expression is restricted only in left ventricle in mouse but bilaterally in human; it can act as cellular arrest signal during cardiogenesis and thereby modulates cardiac growth and development.

(Continued)

Table 10-1 (Continued) Protein family function

Gene name

Organism

spt/tbx16/ XvegT

Tbx5–/–: embryonic lethal at E10.5; severely hypoplastic posterior cardiac segments, delayed looping, absent atrial septum, and atrioventricular cushions Zebrafish: Heartstrings (tbx5) mutants have hearts that initially form and function normally but later fail to loop and display progressive hypoplasia of ventricle and atrium Zebrafish: hr-T morphants: have dysmorphic hearts Zebrafish/Xenopus: spt–/–: randomization of left–right development and heart looping situs

NFAT

NFATc1/CNB1, NFATc2, NFATc3, NFATc4

Mouse: NFATc1–/–: Embryologically lethal at E13.5-E17.5; cardiac failure secondary to abnormal semilunar valve formation.

FOX

Foxc1, Foxc2

Mouse: Foxc1–/–, Foxc2–/–, and Foxc1+/–; Foxc2+/–: perinatally lethal, various aortic arch defects (mostly fourth arch derivatives), and VSD. FoxH1–/–: embryonic lethal between E8.5-E10.5. Primitive heart tube is formed but fails to develop outflow tract and right ventricle, development arrests at looping stage, and cardiac failure occurs Zebrafish: schmalspur (sur) mutants have severe defects in all axial structures and randomization of heart looping Mouse: Foxp1–/–: embryonically lethal at E14.5; thin myocardium and severe cardiac outflow tract, septation and cushion defects Foxp4–/–: embryonically lethal at E12.5; cardiac bifida Hfh4–/–: perinatally lethal, failure to grow, situs inversus or heterotaxy with randomization of heart position

hr-T/tbx20

72

FoxH1

Foxp1

Foxp4 Foxj1(Hfh4)

Expression sites and stages of involvement

tbx20 is expressed in cardiac precursors at the anterior lateral plate and later throughout the myocardium (zebrafish). tbx16 is expressed in mesoderm (zebrafish).

Calcineurin/NFAT signaling functions sequentially from myocardium to endocardium and is required for repression of VEGF in the endocardium (mouse). This repression allows endocardial cells to transform into mesenchymal cells and initiate valve formation in vertebrates. Foxc1 is expressed in the endothelial cells and mesenchyme of the branchial arches (mouse). Foxc2 is also expressed in the valves and cardiac outflow tract (mouse). FoxH1 is expressed initially in the lateral plate mesoderm and then is restricted to the heart (mouse). FoxH1 functionally interacts with Nkx2-5 and is essential for development of the anterior heart field and its derivatives.

Foxp1 is expressed in myocardium and endocardium (early stages up to E14.5). Foxp4 is expressed in ciliated epithelial cells.

Other family genes with potential role in cardiac development

The Tbx20 Drosophila orthologs, mid, and H-15, are required for the functional diversification of cardioblasts and the expression of tinman-dependent terminal differentiation genes within the dorsal vessel.

Homeotic genes

Hoxa-3–/–

Gax

SOX

Sox4

Myc

c-myc N-myc

Jun proto-oncogenes c-Jun (components of AP-1 transcription factors)

73

Iroquois

Irx4

Prh Homeobox (Hex)

Prh/Hhex (Hex)

ras-GTPaseNf1 activating proteins (ras-GAPs).

Pax (Paired domain homeoboxes)

Pax3

Mouse: Hoxa-3–/–: die around the time of birth with various conotruncal anomalies, thymic and parathyroid agenesis reminiscent of the human DGS Mouse: Forced expression of Gax inhibits cardiomyocyte proliferation and causes thin myocardium phenotype Mouse: Sox4–/–: embryonic lethal at E14; die of heart failure; abnormal formation of semilunar valves and persistent truncus arteriosus Mouse: c-myc–/–: embryonic lethal between E9.5-E10.5; heart failure. N-myc–/–: embryonic lethal at E10.5-12.5; poorly developed ventricular myocardium, VSD Mouse: c-jun–/–: embryonic lethal at E13 with TA, RAA, aortic arch anomalies, and thin right ventricle Mouse: Irx4–/–: Hypertrophic cardiomyopathy in the adult mouse Mouse: Hhex–/–: right ventricular hypoplasia, thin myocardium Zebrafish: Hhex knock-down causes randomization of gastrointestinal chirality Human: Patients with neurofibromatosis 1 have increased incidence of certain congenital heart defects like pulmonic stenosis and coarctation. Also a case with multiple coronary aneurysms has been reported in a child with neurofibromatosis Mouse: Nf1–/–: embryonic lethal at E14; poor development of ventricular myocardium, VSD, PTA, abnormal atrioventricular valves Human: Heterozygous mutations are associated with Waardenburg’s syndromes 1 and 3 (no cardiac defects); homozygous mutations are presumed to be early embryonic lethal.

Expressed in the mesenchyme and endothelial cells of the third and fourth pharyngeal arches

hoxd-3, hoxa-4, and hoxd-4: expressed at the early stages of heart formation in the chicken.

Gax has a biphasic pattern of expression in cardiomyocytes (an early one from begin of cardiogenesis up to E12.5 and a late one within compact layer at E15.5) Expressed in endocardial ridges

Mtsh 1, Mtsh2 (mouse) are expressed in mesenchyme of branchial arches. tsh (Drosophila) Sox7/8/9/10/17/18 (Human, mouse, chick, Drosophila)

c-myc is expressed in myocardial cells. N-myc is expressed only in compact myocardium.

The different AP-1 components including Jun proteins are expressed in a development- and tissue-specific manner in multiple tissues including the heart.

Expressed in ventricular myocardium and plays a critical role in establishing chamber-specific gene expression in the developing heart Expressed in the anterior tip of the primitive streak in the mouse embryo, as well as also in the endothelium throughout the developing vascular network and in the third pharyngeal pouch

c-jun and junD are genetically redundant. Mice lacking both jun genes display cardiovascular and angiogenic defects during embryonic development. Irx1, Irx2, and Irx5 may partially compensate for loss of Irx4 function in the heart

Neurofibromin (Nf)-1 modulates epithelialMutations in the related genes TSC1, mesenchymal transformation and proliferation in TSC2 are associated with tuberous the developing heart by down-regulating ras activity. sclerosis and cardiac It is expressed in human atrial and ventricular tissue. rhabdomyomas in humans In the mouse, Nf1 is expressed ubiquitously but is developmentally regulated in the heart and neural crest derivatives.

Expressed in migratory neural crest cells

(Continued)

Table 10-1 (Continued) Protein family function

bHLH

Gene name

dHAND (Hand2)

eHAND (Hand1)

Hey1/Hesr1 Hesr2 (hairy and enhancer of split2)/Hey2/ HRT2/grl

74 MesP1 MesP2

Bicoid-related homeobox genes

Pitx2a,b,c,d

Organism Mouse: Pax3+/–(Splotch+/–): no cardiac defects, only skin pigmentation defects Pax3–/–: embryonic lethal at E13.5, persistent truncus arteriosus, thin ventricular myocardium, VSD Mouse: dHand–/–: embryonic lethal at E10.5; absence of heart looping, absent right ventricle eHAND–/–: embryonic lethal at E8.5-9.5; heart development arrests at late cardiac crescent stage showing incomplete heart tube fusion and no observable contractions Mouse: Hey1–/–;Hey2–/–: embryonic lethal at E11.5 with hypoplastic trabeculae, single ventricle, impaired EMT in atrioventricular cushions and defects in blood vessel development Hey2–/–: die perinatally from cardiac failure and display various cardiac malformations including ventricular septal defects, tetralogy of Fallot, dysplastic atrio-ventricular valves and cardiac hypertrophy Zebrafish: grl–/–: abnormal arterial development with variable disruption of aorta segments that resembles coarctation of the aorta Mouse: MesP1–/–: embryonic lethal at E10.5 with cardiac bifida MesP1–/–;MesP2–/–: embryonic lethal at E9.5 with absent mesodermal layer and heart formation MesP2–/–: perinatally lethal with skeletal anomalies but no heart defects Human: Rieger syndrome (RIEG/Pit2a heterozygous mutants): occular, dental, umbilical cord abnormalities Mouse: Pitx2–/–: deletion of all Pitx2 isoforms is embryologically lethal at E14.5 with multiple developmental abnormalities (ASD, complete AV canal, abnormal arterio-ventricular

Expression sites and stages of involvement

Other family genes with potential role in cardiac development

dHAND is predominantly expressed in the morphologically right ventricle and eHAND in the left ventricle (mouse).

Xenopus dHAND and eHAND are also expressed bilaterally in the lateral mesoderm but without any left-right asymmetry. No eHAND homologue was found in zebrafish.

The Hesr genes are expressed in the cardiac crescent. At later stages Hesr1 is expressed in the atria, while Hesr2 is also expressed in the atria, but predominantly in the ventricles, aortic and pharyngeal arches.

Hesr1/Hey-1/HRT-1/CHF-1 and hesr2 are redundantly required to mediate Notch signaling in the developing cardiovascular system (mouse).

MesP1 is expressed in cardiac precursors in the primitive streak (mouse). MesP2 is expressed in early mesoderm but its expression is down-regulated after E7.0 (mouse).

Pitx2 isoforms are expressed initially in the left lateral plate mesoderm at E8-9; later are expressed bilaterally in the derivatives of the lateral plate mesoderm and only Pitx2c is expressed asymmetrically in the left side of most organ primordia (heart, lungs and gut). Pitx2c is also expressed asymmetrically in the ventral part of branchial arches and at the junction with the aortic sac

The bHLH transcriptional repressor mSharp-1/mDEC2 is expressed in the mouse heart at E8.5 and remains active until at least E16.5

nieuwkoid/ dharma homeobox

nieuwkoid/ dharma

SAP domain nuclear proteins

Myocardin Myocardin-related ttranscription factor B (MRTF)-B

Hop

Cited nuclear proteins (CBP/p300 -interacting transactivator with ED-rich tail) ZFHX

Cited2

ZFHX1B/SIP1

SMAD

SMAD1-8

75

Hop homeobox

connections, abnormal cardiac positioning [but normal cardiac d-looping], hypoplastic RV, right lung isomerism, open ventral wall, eye abnormalities) Pitx2c–/–: lethal in neonatal age with abnormalities of the aortic arch patterning and the outflow tract Zebrafish: bozozok/dharma mutants display independent randomization of heart and gut chirality Mouse: Myocardin–/–: embryonically lethal at 10.5 with severely impaired embryonic vasculogenesis but no evidence of cardiac defects MRTF-B–/–: homozygous mutant mice die between embryonic day (E)17.5 and postnatal day 1 from cardiac failure and display cardiac outflow tract defects (TA, Double outlet right ventricle, IAA and large VSDs) Mouse: Hop–/–: two phenotypic classes characterized by an excess (hypertrophic myocardium in adult age) or deficiency of cardiac myocytes (thin myocardium and embryonic death during midgestation from heart failure). Hop–/– mutants display also conduction defects below the level of the AV node Zebrafish: Inhibition of Hop activity in zebrafish embryos likewise disrupts cardiac development and results in severely impaired cardiac function Mouse: Cited2–/–: embryonic lethal at E13.5-E16.5 with cardiac outflow tract defects (TOF, DORV, truncus arteriosus, right aortic arch), adrenal agenesis, abnormal cranial ganglia and exencephaly Human: SIP1 mutations are associated with Hirshprung’s disease and PDA Mouse: Smad4–/–: display cardiac hypertrophy Smad5–/–: embryonic lethality between E9.5 and E11.5 with defects in heart looping and embryonic turning

dharma is expressed in the dorsal blastoderm and is required at the blastula stages for formation of the embryonic shield and expression of multiple organizer-specific genes. Master gene regulator of many cardiac specific genes, which is diffusely expressed in myocardial tissue.

MRTF-B is expressed in the premigratory neural crest, in rhombomeres 3 and 5, and in the neural crest-derived mesenchyme surrounding the aortic arch arteries.

Myocardin and MRTFs physically associate with SRF and activate SMC-restricted transcription. MRTF-A is expressed by multiple cell lineages and is a potent transcriptional coactivator of some SRF-dependent genes.

Expressed in cardiac precursors and cardiac myocytes (mouse). Balances growth vs differentiation of ventricular myocytes.

Expressed in primitive streak in areas fated to become heart tissue and in the presomitic and lateral plate mesoderm.

Expressed ubiquitously

Smad4 is expressed specifically in cardiomyocytes. Smad5 is ubiquitously expressed in the mouse and zebrafish embryo.

Smads act as BMP receptors by transducing their signaling effects.

(Continued)

Table 10-1 (Continued) Protein family function

Gene name

MADS

Mef2c

ETS

Ets1, Ets2

pointed

GLI superfamily

ZIC3

76 Gli1/2/3

TEA domain (TEAD) family AP-2

Transcriptional enhancer factor (TEF-1/TEAD1) TFAP-2B

LIM

Isl1

ladybird (lb)

lbe (early) lbl (late)

Protein synthesis/modification Subtilisin-like SPC1/Furin

Organism Mouse: Mef2c–/–: Embryonic lethal at E9.5-10.5; absence of heart looping, lack of OFT and right ventricle, single LV fused directly to atrial chamber, lacking trabeculae and dissociated cardiomyocytes. Drosophila: D-mef-2 mutants fail to express contractile protein genes in all muscle lineages (heart, visceral and somatic muscles) Chicken: Targeting Ets1/2 mRNA by antisense oligonucleotides causes abnormal coronary vessels Drosophila: Pnt mutants have excess cardioblasts

Human: ZIC3 mutations are associated with X-linked heterotaxy (HTX1) Mouse: Zic3–/–: embryonic lethal, heterotaxy Human: Gli3 mutations are associated with skeletal malformations syndromes (mostly polydactuly) and cancer Mouse: Gli2–/–;Gli3+/–: VACTER-like phenotype, PTA and heart looping defects Mouse: TEF-1–/–: embryonic lethal at E12; noncompaction, VSD Human: Missense mutations cause Char syndrome with persistence of ductus arteriosus Mouse: Isl1–/–: display cardiac malformations with absent outflow tract and most of the atria Drosophila: lb–/–: display hypoplasia of cardiac precursors. Overexpression of both lbe and lbl reveals hyperplasia of heart progenitors Mouse:

Expression sites and stages of involvement

Other family genes with potential role in cardiac development

Essential for differentiation of muscle cell lineages (mouse, Drosophila)

Mef2α, Mef2b, Mef2d (mouse). SRF is expressed in cardiac and skeletal muscle cells and is implicated in the control of cardiac muscle gene expression (mouse)

ETS are important for the epithelial-mesenchymal transformation in the primary and secondary heart fields (chicken)

XI-fli, Xsap 1 (Xenopus): Expressed in heart and branchial arches

Pnt is a key regulator of cardioblast and pericardial cell fates in the posterior segments of the heart, by promoting pericardial cell development and opposing cardioblast development (Drosophila). Expressed in mesoderm (mouse)

Expression of Gli1 is more restricted and transcriptionally regulated by Shh, Gli2/3 expression is broader and less dependent from Shh signaling (mouse).

Expressed in myocardium Expressed in third, fourth, and sixth pharyngeal arches and later in ductus arteriosus

Expressed in cardiomyocytes in the secondary (anterior) cardiac field. lb genes are expressed in clusters of about four cells per hemisegment in the developing heart region.

Spcs induce endoproteolytic maturation of TGF-β-related molecules. Spc1 is expressed in the visceral endoderm

proprotein Convertases (SPCs)

SPC4/ PACE4

SPT translation elongation factors

Spt5

Spt6

Chromatin modification HDACs Hdac1

77 Hdac5, Hdac9

CREB-binding protein

CREBBP

p300/CBP transcriptional activators

p300

Polycomb genes

rae28

Fur–/–: embryonic lethal at E10.5 with failure and the cardiogenic mesoderm. of cardiac primordial to fuse in the ventral Spc4 is expressed in the definite endoderm midline or with formation of a short and the left lateral plate. unlooped heart with reduced number of cardiomyocytes Spc4–/–: embryological lethal at E13.5-E15.5 with situs ambiguous, congenital heart defects (PTA, VSD, common atrium, DORV), combined with left pulmonary isomerism or complex craniofacial malformations including cyclopia Zebrafish: Spt5 and 6 are both transcriptional elongation sk8/s30 (Spt5) mutants have similar factors that interact with each other. phenotype with pandora Pandora (Spt6) mutants have dysmorphic hearts, decreased number of ventricular myocytes and pericardial effusion Mouse: Hdac1–/–: embryonic lethal at E10.5. The mutant embryos are severely dysmorphic and growth restricted, so that the cardiac phenotype could not be assessed. Zebrafish: hdac1 mutants have dysmorphic and hypoplastic hearts with absent atrio-ventricular valve Mouse: Hdac5–/– and Hdac9–/–: sensitize animals to hypertrophic stimuli Human: Haploinsufficiency of the CREBBP gene (16p13.3 deletion) has been associated with a severe form of Rubinstein-Taybi syndrome that includes mental retardation, skeletal abnormalities, accessory spleens, lung laterality defects and congenital cardiac defects (hypoplastic left heart, VSD). Mouse: p300–/–: mutant mice die between E9.5-E11.5 from cardiac failure associated with defects of cardiac muscle differentiation Mouse: rae28–/–: die around the time of birth with conotruncal anomalies (tetralogy of Fallot and aortic stenosis), asplenia and abnormalities in the eyes, hard palate, and parathyroid glands

Several class II HDAC members (in particular HDAC5 and 9) are very strongly expressed in cardiomyocytes and postulated to inhibit post-mitotic cardiac growth and repress cardiac hypertrophy (human, mouse).

Several class I HDAC members (1, 2, 3, 8) are also expressed in cardiac tissue (mouse, zebrafish) but their particular role in cardiac development is still unknown.

CREBBP and its paralog EP300 are ubiquitously expressed transcriptional coactivators and act as histone acetyl transferases (HATs).

EP 300

Expressed ubiquitously from early developmental stages.

p300 and CBP play a critical role in cardiac hypertrophy that is dependent on their histone acetyltransferase activity.

The gene contains DNA binding domains that belong in the chromatin-associated group of proteins.

(Continued)

Table 10-1 (Continued) Protein family function

Gene name

Organism

78

2. Intercellular signaling and signal transduction (signals, ligands and receptors) a. Secreted factors Hedgehog Shh/syu Mouse: Shh–/–: alterations in cardiac looping, tetralogy of Fallot with pulmonary atresia Zebrafish: syu–/–: absent dorsal aorta and axial vein Ihh Mouse: Shh–/– ;Ihh–/–: developmental arrest at somitogenesis with small, linear heart tube, open gut and cyclopia (identical phenotype with Smo–/–) TGF-β TGF-β2 Mouse: TGF-β2–/–: die in perinatal period with malformations of the cardiac outflow tract, the AV canal region, the aortic arches and the semilunar valves BMP-1/Tolloid like Mouse: proteases (Tld): Tll-1–/–: embryonic lethal during midegestation Tolloid-like 1 with AV-canal type of VSD and abnormal (Tll-1) semilunar valves Bmp2 Mouse: Bmp2–/–: Early embryonic lethal with amnion and chorion malformations and abnormal development of the heart in the exocoelomic cavity. No cardiac defects in heterozygotes Zebrafish: Swirl/bmp2b mutants display dorsoventral patterning defects but no cardiac abnormalities Bmp4 Mouse: Bmp4+/–: No cardiac defects in heterozygotes. Bmp4–/–: Early embryonic lethal. Cardiac specific ablation of Bmp4 is embryonic lethal at E13.5 with outflow tract defects including truncus arteriosus, and in combination with a hypomorphic allele causes common atrioventricular canal Drosophila: Dpp mutant embryos: heart and visceral mesoderm formation is abolished (as in tinman)

Expression sites and stages of involvement

Other family genes with potential role in cardiac development

Expressed in endoderm of pharyngeal arches

TGF-β-related signals (like Nodal, lefty, southpaw, Gdf1) specify in vertebrates the anteroposterior and left-right axes. TGF-β2 is expressed in the endocardium and the myocardium adjacent to the endocardial cushion tissue.

TGF-β1 is expressed in endothelial cells. TGF-β3 is expressed in endocardial cushions

Tll-1 is expressed specifically in precardiac tissue and endocardium at E7.5-8.5 mouse embryos.

Bmp2 is expressed in the AV junctional myocardium and valve tissue and interacts with TGF-α3 for the formation of endocardial cushions.

Bmp4 is expressed in the cardiac outflow tract (incl. truncus arteriosus) and the endocardial cushion ridges (mouse). It has also a developmentally regulated expression in zebrafish.

Dpp (decapentaplegic) is required along with the secreted signaling molecule, wingless (wg) to specify the heart in Drosophila.

Bmp5/6/7

Bmp10

Nodal

southpaw nodal related gene

79 LEFTA / lefty 1(Ebaf)

Gdf-1/Vg-1

TGF-β antagonists

Chordin(Chrd)

Mouse: Bmp5–/–;Bmp7–/–: embryos die at E10.5 with absent endocardial cushions Bmp6–/–;Bmp7–/–: embryos die between E10.5 and E15.5 due to cardiac failure with delayed formation of the outflow tract endocardial cushions, defects in valve morphogenesis and chamber septation Bmp10–/–: embryologically lethal at E10.5 with hypoplastic ventricular wall

Mouse: Nodal–/–: die in gastrulation but compound heterozygotes between hypomorphic and null alleles (Nodalfl1/) die at E14.5 and display abnormal cardiac looping, dextrocardia or mesocardia, TGA and VSDs. HNF3-β+/–; nodallacZ/+ double-heterozygous embryos display randomization of heart looping and left pulmonary isomerism. Zebrafish: southpaw morphants display severe disruption of early (cardiac jogging) and late (cardiac looping) aspects of cardiac left-right asymmetry. Human: Mutations identified in patients with laterality defects similar to mouse lefty 1–/–. Mouse: lefty 1–/–: laterality defects described as left thoracic isomerism. Mouse: Gdf1–/–: spectrum of laterality defects, including visceral situs inversus, right pulmonary isomerism, and a range of cardiac anomalies incl. abnormalities of cardiac looping, transposition of great arteries, VSD, and atrio-ventricular canal. Mouse Chrd–/–: die during mid-gestation or at time of birth with DGS phenotype incl. Truncus arteriosus, right aortic arch and aortic arch branching abnormalities. Zebrafish/Xenopus. Chordino(dino) mutants display randomization of heart looping

Bmp5 is expressed in the developing myocardium; Bmp6 is expressed in the outflow tract and the atrioventricular cushions; Bmp7 is expressed in overlapping and adjacent sites with Bmp6, including the endocardial cushions. It is also expressed diffusely in axial and lateral mesoderm, gut endoderm and allantois. BMP-10 expression is restricted initially in the trabeculated part of the common ventricular chamber and the bulbus cordis, but after E12.5, additional expression is seen in the atrial wall (mouse)

BMP-10 is important for the trabeculation and ventricular growth of the embryonic heart.

Nodal is expressed in left lateral plate mesoderm and is required for formation of the primitive streak. It is indirectly implicated in laterality defects in human, mouse, chicken, and frog.

Mutations of the zebrafish nodal related gene cyclops (ndr2) does not cause laterality defects.

Lefty1 is expressed asymmetrically in the left lateral plate in the mouse embryo.

Lefty2 (LEFTB) is also asymmetrically expressed in the left lateral plate in the mouse embryo and seems to be controlled by lefty1.

Gdf-1 is expressed bilaterally in intermediate and lateral plate mesoderm. Vg1 cell-signaling pathway plays a central role in left–right coordinator function in Xenopus. Chrd is a secreted Bmp-binding protein that is expressed in the mouse node and its derivatives, notochord and pharyngeal endoderm (mouse).

The role of other bmp inhibitors like noggin, follistatin, and Sog on cardiac development is unknown

Chordin gene is first expressed on the dorsal side of the embryo (starting before gastrulation) (Zebrafish, Xenopus). (Continued)

Table 10-1 (Continued) Protein family function Early response gene growth factor

Gene name SIL

80

Neurotrophin growth factors

Nt3

Vegf

Vegf

Erythropoietin

Epo

Wnt/Wingless

Dvl2(Dsh2)

EGF-CFC

CFC1/Cryptic

Adapter proteins

CRKL

Organism Mouse: SIL–/–: embryonic lethal around E10.5 with axial midline defects and randomized cardiac looping Mouse: Nt3–/–: perinatally lethal with severe cardiovascular abnormalities including atrial and ventricular septal defects, and tetralogy of Fallot Human: Certain VEGF haplotypes are associated with increased risk for cardiovascular defects in del22q11 individuals Mouse: VEGF isoforms associated with decreased VEGF expression display various aortic arch abnormalities. Transgenics overexpressing VEGF die at E12-E14 and display conotruncal anomalies (TOF, pulmonary hypoplasia), VSD and thin myocardium Mouse: Epo–/–: embryonic lethal during midgestation from failure of erythropoiesis. The mutants display also ventricular hypoplasia, detachment of epicardium and abnormalities of the vascular network Mouse: Dvl2–/–: 50% perinatally lethal (predominantly the males) DORV, TGA, PTA

Human: Loss-of-function mutations are associated with heterotaxic phenotypes with a variety of congenital heart defects (d-TGA, DORV, Pulmonary atresia, complete AV canal, HLHS, ASD, VSD, PDA) Mouse: cryptic–/–: right pulmonary isomerism, dextrocardia, malposition or transposition of the great arteries and DORV Human: Included within the DGCR Mouse:

Expression sites and stages of involvement Ubiquitously expressed

Other family genes with potential role in cardiac development

Essential for the survival and/or migration of cardiac neural crest

Expressed in the endoderm of the pharyngeal arches, the aortic sac, cardiac outflow tract, and ventricular septum

Erythropoietin affects cardiomyocyte proliferation in a cell non-autonomous manner.

Similar results were found also in the erythropoietin receptor knockout mice.

Widely expressed in embryonic and adult tissues. Important for cardiac neural crest development

Wnt1/Wingless: Wnt antagonism initiates cardiogenesis in Drosophila, Xenopus laevis and chick. Wnt1 is expressed in subset of cardiac neural crest cells Cripto and oep act as co-receptors or presenting molecules in nodal signaling pathway.

Expressed at E7.5 in the precardiac and lateral plate mesoderm in the mouse; later is expressed in the cardiac inflow and outflow tract.

Expressed in dorsal neural tube, dorsal root ganglia, some clearly defined some of the branchial arches and the front nasal mass.

Semaphorins

Sema3A

Sema3C

Ephrins

ephrinB2

EGF Neuregulin and EGFR signaling

NRG1

HRG

81

HB-EGF

Betacellulin (BTC) TACE FGF

Fgf8

Endothelins

ET-1

Nonmembranous protein tyrosine phosphatases

PTPN11

Crkol–/–: abnormalities of the cardiac outflow tract and the aortic arch arteries Mouse: Sema3A–/–: perinatally lethal with selective hypertrophy of the right ventricle and right atrial dilation Sema3C–/–: perinatally lethal, interrupted aortic arch, and PTA.

Sema3A is expressed in cardiac tissue (rat)

Sema3C is expressed in migratory and postmigratory neural crest cells in the myocardium surrounding the outflow tract. Important for the migration of cardiac neural crest cells to the outflow tract EphrinB2 is expressed in arterial endothelial cells.

Mouse: ephrinB2–/–: embryonic lethal before E11.5 with poorly organized vessels, deficient myocardial trabeculation, pericardial effusion and no cardiac looping Endocardiac-specific ephrinB2 inactivation leads to similarly defective cardiac morphogenesis Mouse: NRG1–/–: embryonic lethal during midgestation due to the aborted development of myocardial trabeculae in ventricular muscle (Long 2003) HRG–/–: embryonic lethal at E10.5 with thin myocardium, irregular heart beat HB-EGF–/–: die in neonatal age from heart failure secondary malformed semilunar and atrioventricular heart valves, and hypoplastic, poorly differentiated lungs (Meno 1998) HB-EGF–/–; BTC–/–: similar defects as described above with more accelerated occurence of cardiac failure and death TACE–/–: similar phenotype with defective valvulogenesis as above Zebrafish: acerebellar: homozygous mutants for fgf8 have dysmorphic hearts with reduced ventricular mass Mouse: Fgf8 homozygous hypomorphic alleles display looping abnormalities as well as various aortic arch defects, including outflow tract, intracardiac and aortic arch anomalies Mouse: ET-1–/–: conotruncal and aortic arch defects

Expressed in the neural crest-derived pharyngeal mesenchyme, as well as endoderm and ectoderm

Human: Defects in PTPN11 account for 50% of patients with Noonan’s syndrome (which includes cardiovascular anomalies such as pulmonary

PTPN11 is widely expressed in both embryonic and adult tissues and encodes the nonmembranous protein tyrosine phosphatases SHP-2, which mediates signal transduction from numerous

NRG1 is expressed by pluripotent neural crest cells and endocardium.

Heparin-binding epidermal growth factor (HB-EGF) Although BTC and TACE do not belong and betacellulin (BTC) are activating ligands for to the same family of genes they EGF receptor (EGFR/ErbB1) and ErbB4. all act through activation of the HB-EGF is expressed by endocardial cells lining EGFR signaling pathway and are the margins of cardiac valves. Tumor necrosis bundled together in this context. factor-alpha converting enzyme (TACE) activates EGFR by deriving soluble HB-EGF. Proper cardiac valvulogenesis is dependent on EGFR activation that is required to regulate BMP signaling in endocardial cushions.

Fgf8 is expressed in cardiac precursors and later in the ventricular tissue, ectoderm of first pharyngeal arch and core mesenchyme of more caudal pharyngeal arches (mouse) Fgf8 is required for induction and patterning of myocardial precursors Fgf8 is a left determinant in the mouse but in the chicken appears to be a right determinant.

Fgf1, 2, 4, 7, 12, 13, 16

Fgf10 is expressed in pharyngeal mesodermal cells that contribute to the arterial pole of the heart.

(Continued)

Table 10-1 (Continued) Protein family function

MicroRNAs

Gene name

microRNA-1-1 (miR-1-1)

82 b. Receptors and Ligands Notch Notch1

Notch2

Notch4

Dsl proteins

Dll1

Organism stenosis and hypertrophic cardiomyopathy) and some cases with Leopard syndrome (multiple lentigines, congenital cardiac abnormalities mostly pulmonic stenosis, ocular hypertelorism, and retardation of growth) Mouse: Ptpn11D61G/D61G: Homozygous mutants for the D61G mutation in the Ptpn11 gene are embryonic lethal at E13.5 (grossly edematous and hemorrhagic, have diffuse liver necrosis and severe cardiac defects). Heterozygous embryos exhibit cardiac defects (DORV, thin myocardium, thickened atrioventricular and outflow tract valves, growth failure, perturbed craniofacial development and a mild myeloproliferative disease Mouse: Excess miR-1-1 in the developing heart leads to a decreased pool of proliferating ventricular cardiomyocytes

Expression sites and stages of involvement

Other family genes with potential role in cardiac development

receptor tyrosine kinases. It is positively associated with pulmonic stenosis but negatively associated with hypertrophic cardiomyopathy, suggestive of a more critical role for SHP-2 in valvulogenesis than cardiomyocyte proliferation

miR-1-1 and miR-1-2 are specifically expressed in cardiac and skeletal muscle precursor cells

Human: Notch1 and Notch2 are expressed in the Notch 1 mutations cause a spectrum of outflow tracts and the epicardium, in developmental aortic valve anomalies and specific cell populations that express severe valve calcification in non-syndromic JAG1. Notch 1 transcripts are particularly autosomal-dominant human pedigrees enriched in the aortic valve. Mouse: Notch1–/–: embryonic lethal with severe neuronal and somitic defects. In addition, Notch1–/– mutants show defects in angiogenic vascular remodeling in the embryo, yolk sac, and placenta. Endothelial-specific Notch1 inactivation recapitulates the Notch1–/–phenotype Notch2del1/del1: mutants die perinatally and display myocardial hypoplasia, edema, kidney and ocular vascular defects Notch4 is restricted to the vascular Notch4–/–: No vascular phenotype endothelium Notch1–/–; Notch4–/–: More severe vascular phenotype than Notch1–/– Mouse: Dll1 encodes one of the four known mammalian DSL Dll1–/–: situs ambiguous phenotype, including

During cardiogenesis miR-1 genes titrate the effects of cardiac regulatory proteins to control the balance between differentiation and proliferation.

Notch1 (Xotch) and Serrate1are expressed in overlapping patterns in the early heart field in Xenopus. Notch signaling suppresses cardiomyogenesis through suppressor hairless, Su(H) and this is essential for the correct specification of myocardial and nonmyocardial cell fates

Jagged-1 (JAG-1)

Receptors of TGF-β-like ligands

TBRII

Endoglin

83 ALK1 (acvrl1)

ALK2 (acvrl2)

ALK3 (BMPR1A)

randomization of the direction of heart looping and embryonic turning. Human: Mutations in JAG1 are associated with Alagille syndrome (AGS) and cardiac defects (TOF, VSD, ASD, PDA, PTA, pulmonary atresia, hypoplasia of the entire pulmonary vascular tree) Mouse: Jag1–/–: embryonic lethal at E10.5 from hemorrhage and exhibit defects in remodeling of the embryonic and yolk sac vasculature (Hamblet 2002) Jag1+/–; Notch2+/–: Phenotype reminiscent of human Alagille syndrome Mouse: TBRIInull/flox;Wnt1-Cre+/–: Neural crest specific TBRII inactivation is perinatally lethal with DGS phenotype including truncus arteriosus and abnormal aortic arch branching Endoglin+/–: Hereditary hemorrhagic telangiectasia type 1 (HHT1) Endoglin–/–: embryonic lethal at E10.5 with diffuse vascular abnormalities, absence of cardiac cushions formation and pericardial effusion Human: Heterozygous mutations are associated with primary pulmonary hypertension and hereditary hemorrhagic telangiectasia type 2 (HHT2) Mouse: ALK1+/–: Hereditary hemorrhagic telangiectasia type 2 (HHT2) Zebrafish: Vbg(acvrl1)–/–: dilated cranial vessels Mouse: ALK2 null/flox;Tie2-Cre+/–: Endothelial-specific ALK2 inactivation is embryonically lethal and leads to endocardial cushion defects ALK2;Wnt1-Cre+/–: Neural crest-specific ALK2 inactivation causes craniofacial anomalies and aortic arch abnormalities reminiscent of human DGS phenotype ALK3 null/flox;αMHC-Cre+/–: Cardiac-specific deletion of ALK3 is embryologically lethal around E15.5 with endocardial cushion defects and hypoplastic trabeculae

proteins and acts as a ligand for the receptor Notch. Dll1 is expressed in the human embryonic heart (atria, ventricular endocardium, coronary arteries and epicardium). In the mouse embryo. Jag-1 has more global expression but comparable in the heart (atria, ventricles, outflow tracts, valve precursors) and the great arteries (pulmonary arteries, descending aorta and all pharyngeal and their vascular derivatives)

All receptors are expressed diffusely in several issues during embryonic mouse development

TBRI,TBRII, TBRIII and ALK2 mediate TGF-β signaling and induce epithelial-mesenchymal transformation in the AV cushions

Endoglin is expressed at high levels on endothelial cells of capillaries, veins, and arteries (human) as well as on mesenchymal stromal cells in several tissues (mouse)

ALK1 is expressed in endocardial cells (mouse, chicken)

ALK2 is expressed abundantly in endocardial cells of the outflow tract, ventricle, and AV cushion (mouse, chicken)

ALK3 is expressed ubiquitously in mouse embryonic tissues (with the exception of liver)

ALK3(BMPR1A) is required for mesodermal formation in vertebrates. (Continued)

Table 10-1 (Continued) Protein family function

Gene name

ALK5

ACVR2B (ActRIIB)

BmpR2

84 EGFR family/

Neuregulin receptors

EGFR

Organism

Expression sites and stages of involvement

ALK3 null/flox cGATA6-Cre+/–: Targeted Alk3 deletion in cardiomyocytes of the AV canal results in atrio-ventricular valve defects reminiscent of Ebstein’s disease and WPW αMHC-ALK5: Transgenics with cardiac-specific ALK5 is expressed throughout the heart. ALK5 activation display arrest of looping morphogenesis and a linear, dilated and hypoplastic heart tube Human: ACVR2B mutations were only rarely found among cases of LR axis malformation Mouse: ActRIIB–/–: Lethal at birth; randomized heart position, malposition of the great arteries, atrial and ventricular septal defects; right pulmonary isomerism and splenic abnormalities, recapitulating the clinical features of the human asplenia syndrome. Human: Mutations in BMPR2 were found in patients with primary pulmonary hypertension (PPH1) Mouse: BmpR2+/–: susceptible to pulmonary vascular obstructive lung disease BmpR2–/–: die in utero at gastrulation but homozygous mice for a BmpR2 hypomorphic allele die at midgestation with cardiovascular defects like truncus arteriosus with interrupted aortic arch and dysmorphic semilunar valves Mouse: EGFR is ubiquitously expressed during gastrulation, somitogenesis and later stages (zebrafish). Egfr–/–: EGFR mutant and null mice have abnormal semilunar valves with thickened leaflets Zebrafish: Inhibition of zebrafish EGFR activity in vivo impeded blood flow via the outflow tract into the aorta and impeded circulation in the axial and intersegmental vessels by 80 hpf. Analysis of the heart showed that the heart chambers and pericardial sacs were dilated and the outflow tracts were narrowed Mouse: ErbB2 expressed in cardiomyocytes

Other family genes with potential role in cardiac development

ALK5 is not involved in the endothelial–mesenchymal transformation of the endocardial cushions.

The ActRIIB-mediated signaling pathway interacts with retinoic acid and plays a critical role in patterning both anteroposterior and left–right axes.

BmpR2 is required for primitive streak formation.

ErbB4 expressed in cardiomyocytes

ErbB2

Erb3

ErbB2–/–: similar with HRG–/– ; die at midgestation because of heart malformation. Conditional mutation of the ErbB2 in cardiomyocytes leads to dilated cardiomyopathy ErbB3–/–: embryonic lethal at E13.5 with underdeveloped AV valves and mildly reduced myocardial thickness Human: CRELD1 is deleted in the cytogenetic disorder 3p– syndrome that is associated with common atrioventricular canal ASD and VSD Mouse: trkC–/–: mice lacking all trkC receptor isoforms die in the early postnatal period with severe cardiac defects such as atrial and ventricular septal defects, and valvular defects including pulmonic stenosis Mouse: ephB2–/–;ephB3–/–: similar phenotype with ephrinB2 mutants but even more severely affected vessel formation ephB4–/–: similar phenotype as ephB2–/–.

85

Cysteine-Rich with EGF-Like Domains

CRELD1

Neurotrophin receptors

trkC

Eph receptors tyrosine kinases

EphB2/B3

Semaphorin A receptors: neuropilins

Np1

Mouse: Np1–/–: embryonic lethal at E12.5 with congenital heart defects like transposed great arteries, truncus arteriosus and arch anomalies most notable absent fourth and sixth aortic arches

Integrins

α4β1 integrin

FGFR

DFGF-R2 (DFGRI)/ Heartless (Htl)

Mouse: α4β1 integrin–/–: embryonic lethal at E12.5 with absent epicardium and hemorrhagic cardiac disease. Drosophila: Htl mutants lack visceral mesoderm, heart, and the dorsal somatic muscles.

PDGFRα

PDGFRα

Mouse: PDGFRα–/–: conotruncal defects, thin myocardium, pericardial effusion

ErbB3 expressed in endocardial cushion mesenchyme CRELD1 is ubiquitously expressed in early development but later becomes more markedly expressed in the developing heart and branchial arches (chicken). Expression persists in adulthood in most tissues (human). Expressed in the outflow tract, the posterior wall of the aorta and the cardiac ganglia

ephB2 is expressed in endocardiac cells and mesenchyme. ephB3 is expressed in all endothelial cells and aortic arches. ephB4 is specifically expressed in endothelial cells of all major veins. Np1 is a receptor for class 3 semaphorins and for VEGF and is preferentially expressed in the endothelial cells of arteries.

Np2 is similarly a receptor for class 3 semaphorins but is preferentially expressed in the endothelial cells of veins. Plexin A2 belongs to the plexins family of receptors for transmembrane semaphorins and is expressed in migratory and postmigratory cardiac neural crest cells.

α4β1 integrin is expressed both in epicardial cells and cardiac outflow tract cells and appears to be essential for epicardium formation and maintenance Htl is expressed in the embryonic mesoderm and is essential for cell migration and establishment of several mesodermal lineages into primordia for the heart, visceral and somatic muscles. Expressed in cardiac neural crest cells (Continued)

Table 10-1 Protein family function

Gene name

Ig superfamily

VCAM-1

Retinoid acid receptors

RXR

RAR(α1/β2, α/β2, αγ)

Agpt1

Erythropoietin receptor

EpoR

Endothelin receptors ECE

ETA

86

Angiopoietin

ECE-1

ECE-2 Orphan nuclear receptors

COUP-TFII (Nr2f2)

Phospatidyl-serine recptor

Ptdsr

Organism Mouse: VCAM–/–: most of them embryonic lethal between E10.5 and E12.5 with severe placental defects, absent epicardium, bloody pericardial effusion and reduced compact layer of ventricular myocardium and intraventricular septum. Mouse: RXRα–/–: embryonic lethal at E14.5 from heart failure as result of a thin walled myocardium RARα–/– and β–/–: thin walled myocardium RAR inhibition of all isoforms (α, β, γ) impairs the formation of third, fourth, and sixth pharyngeal arches and their derivatives and is associated with conotruncal defects (interrupted aortic arch, truncus arteriosus, VSD, DORV). Mouse: Agpt1–/–: embryonic lethal at E12.5 with severe vascular anomalies reminiscent of those seen in embryos lacking TIE2. Mutant mice have abnormal ventricular endocardium with decreased trabeculations and collapsed atrial endothelial lining. Mouse: EpoR–/–: embryonic lethal during midgestation with similar phenotype as the Epo–/– mice Mouse: ETA–/–: conotruncal and aortic arch defects Mouse: ECE-1–/–: conotruncal and aortic arch defects (Rebagliati 1998) ECE-1–/–;ECE-2–/–: as above plus atrioventricular valve abnormalities Mouse: COUP-TFII–/–: embryonic lethal with defects in angiogenesis and heart development; the atria and sinus venosus fail to develop past the primitive tube stage. Mouse: Ptdsr–/–: Ablation of Ptdsr function in knockout mice causes perinatal lethality from cardiac failure associated with cardiac defects such as VSDs, double-outlet right ventricle, and

(Continued) Expression sites and stages of involvement

Other family genes with potential role in cardiac development

Expressed in the epicardium and the underlying myocardium, as well as in the intraventricular septum

Pecam-1 is expressed in the endothelium of the dorsal aorta and the ventricular and outflow tract endocardium (mouse).

Retinoic acid and its metabolites are signals that trigger and modulate cardiac morphogenesis during vertebrate development. The mediators of response to retinoids are transcription factors and proteins of the nuclear hormone receptor superfamily (RARs, RXRs), that bind these ligands. RXRs are expressed in neural crest cells (mouse, chicken). RARs are expressed in neural crest and its derivatives, in endocardial cells and sites of ectomesodermal interaction.

There is considerable overlap and redundancy among the various RAR receptors, so the exact role of each one is difficult to be determined. Conoventricular and conoseptal hypoplasia type VSDs have been identified in compound heterozygotes for various RAR receptor isoforms in association with excessive cell death in the conoseptum.

Expressed in myocardium wall surrounding the TIE2 expressing endocardium

EPOR is expressed in the developing heart in a subset of cardiac tissues (endocardium, epicardium, pericardium and endocardial cushions) but not myoocardium. Similar expression pattern as ET-1 Expressed in the pharyngeal ectoderm and the aortic arch endothelium. Expressed in the endocardiac cushions COUP-TFII is expressed in mesoderm but only in sinoatrial region of the forming heart.

svp (Drosophila)

hypoplasia of the pulmonary artery. They also display thymus hypoplasia, ocular lesions, hematopoietic defects, growth retardation as well as delay in terminal differentiation of the kidney, intestine, liver and lungs during embryogenesis c. Channels Connexins

Cx40 Cx43 Cx45

3. Cell structure and motility KIF Kif3A

Kif3B DNAH11/ Lrdynein (lrd)/iv

Smoothened

Smo

Laminin

Laminin

Hyaluronidase synthase

Has2

87

Axonemal dynein

Mouse: Cx40–/–: cardiac conduction abnormalities but not complete cardiac block Cx43–/–: lethal in neonatal period, conotruncal defects in the RVOT, coronary artery abnormalities Cx45–/–: Embryologically lethal at E10; heart failure, endocardial cushion defect, conduction block Mouse: Kif3A–/–: embryonic lethal at E10.5; cardiovascular insufficiency, pericardial effusion, randomization of laterality in heart looping Kif3B–/–: left-right randomization Human: Mutation in DNAH11 has been found in a patient with Kartagener syndrome. Mouse: lrd–/–: Homozygous mutants display randomized left-right development but no congenital heart defects (identical with iv/iv mutant mice) Mouse: Smo–/–: developmental arrest at somitogenesis with small, linear heart tube, open gut, and cyclopia Drosophila: laminin A deficient embryos:display twists and breaks of cardioblasts at late embryonic stages as a result of dissociation of the pericardial cells Mouse: laminin–α2–/–: display phenotype equivalent with muscular dystrophy; mutants die prematurely at 5 mo from unidentified cause Mouse: Has2–/–: embryonic lethal at E9.5 with abnormal formation of AV cushions and outflow tract, thin myocardium, pericardial edema

Both Cx40 and 45 are expressed in the His bundles in mice. Cx43 is expressed in proepicardial and cardiac neural crest cells and is involved in the proper epithelial–mesenchymal transformation of the proepicardial cells and the development of cardiac neural crest cells.

Expressed in nodal cilia and ubiquitously. Important for leftward nodal flow

Initially is expressed symmetrically in the node at embryonic day 7.5 but at embryonic day 8, a striking asymmetric expression pattern is observed in all three germ layers, suggesting roles in both the establishment and maintenance of left–right asymmetry.

Smo acts epistatic to Ptc1 to mediate Shh and Ihh signaling in the early mouse embryo.

Expressed in extracellular mattrix

Important for AV cushion transformation

(Continued)

Table 10-1 (Continued) Protein family function

Gene name

88

UDP-glucose dehydrogenase

UDP-GD

Perlecan (heparin–sulfate proteoglycan)

Perlecan

Versican

Cspg2 (versican)

Fibulin

Fibulin-2

Fibrillin

Fbn-1 Fbn-2

Elastin

Elastin

Collagen

COL3A1

Collagen6 α1/α2 Vinculin

Vinculin

4. Metabolism Folbp

Folbp1

RALDH

RALDH1,2

Organism Zebrafish: jekyll: zebrafish mutants for UDP-GD have thin myocardium and hypoplastic endocardial cushions Mouse: Perlecan–/–: embryonic lethal at E10 or at birth; TGA, hyperplastic semilunar valves

Mouse: Hdf (Cspg2 –/–): Embryonic lethal at E10.5; the future right ventricle and conotruncus of the single heart tube fail to form and the endocardial cushions are absent Mouse: Fibulin-2 –/–: No cardiac developmental abnormalities Human: Heterozygous mutations of Fbn1 and Fbn2 are associated with Marfan’s phenotype Human: Haploinsufficiency of Ch7q11.23 that includes elastin is associated with Williams-Bueren syndrome (supravalvular aortic stenosis, elfin facies, hypercalcemia and mental retardation) Human: Ehlers-Danlos syndrome type IV results from mutations in the COL3A1 gene, which encodes the polypeptides in type III collagen Mouse: Specific alleles of the collagen genes associated with atrioventricular canal defects Mouse: Vinculin–/–: early embryonic lethal between E8-E10 with severely reduced and akinetic myocardial and endocardial structures Mouse: Folbp1–/–: embryonic lethal at E10, growth retardation, neural tube defects and aortic arch defects Mouse: RALDH2–/–: embryonic lethal at E10.5 with formation of an unlooped medial distended cavity and severe impairment of posterior

Expression sites and stages of involvement

Other family genes with potential role in cardiac development

Perlecan is present in the basal surface of myocardium and endocardium, as well as surrounding neural crest cells. It controls production of heparin–sulfate proteoglycan and is important for AV cushion transformation. Cspg2 gene controls expression of the chondroitin sulfate proteoglycan versican and is required for the successful development of the endocardial cushion swellings and the embryonic heart segments that give rise to the right ventricle and conotruncus. Expressed in AV cushions

Expressed in endocardial cells

Collagen VI genes are expressed in the AV canal tissue of embryonic mouse hearts and are included in the region of the human chromosome 21 that is considered critical for congenital heart defects. Ubiquitously expressed (mouse)

Folbp1 is expressed in the mesenchyme of the pharyngeal arches. Phenotype was rescued by folate rich diet. RALDH2 is expressed in the mesenchyme and the early heart epicardium

RALDH1 is coexpressed with Raldh2 in the early heart epicardium, and is later specifically expressed in developing heart valves.

chamber (atria and sinus venosus) development. The developing ventricular myocardium consists of a thick layer of loosely attached and prematurely differentiated cardiomyocytes. Mutant mice harboring a hypomorphic allele of RALDH2 die perinatally and exhibit the features of the human DiGeorge syndrome (DGS) with heart outflow tract septation defects and selective defects of the posterior (third to sixth) branchial arches ASD, atrial septal defect; VSD, ventricular septal defect; TOF; Tetralogy of Fallot; IAA, interrupted aortic arch; RAA, right aortic arch; TA, truncus arteriosus; AV canal, atrioventricular canal; DORV, Double outlet right ventricle; HLHS, hypoplastic left heart syndrome; DGS, DiGeorge syndrome; DGCR, DiGeorge critical region; RSCA, retroesophageal subclavian artery; bHLH, basic helix-loop-helix; Prh, proline-rich homeobox is the new name for Hex, hematopoietically expressed homeobox; TGF, transforming growth factor; FGF, fibroblast growth factor; EGF, epidermal growth factor; PDGF, platelet-derived growth factor; Bmp, Bone morphogenetic protein; VEGF, vascular endothelial growth factor; RALDH2, retinaldehyde dehydrogenase.

89

90

SECTION II / CARDIOLOGY

of its close phylogenetic relationship to humans and the similarities between the mouse and the human heart at the anatomic and physiological levels. The completion of sequencing of the mouse genome adds an additional advantage over other species for dissecting genetic pathways associated with congenital malformations. However, murine cardiac development is difficult to observe in a serial fashion in vivo without new imaging techniques because of the relatively inaccessible embryo, and because most mouse embryos with severe cardiac malfunction die early in utero, complicating functional analysis. Systematic screens of the mouse genome for recessive mutations affecting morphogenesis of several organs have been initiated but are decidedly expensive. Experimentation with the chick Gallus gallus, whose embryonic material is easily accessible for physical manipulation from the earliest stages, has proven beneficial for some areas like the study of neural crest. The African clawed frog Xenopus laevis has been used mostly to study the inductive interactions in the establishment of polarity and tissue differentiation. The drawback of these vertebrates is the long generation time and the inapplicability of classic genetics. The teleost fish Danio (formerly Brachydanio) rerio (zebrafish) has emerged as a model system because of its prolific egg production, rapid (90 d) generation time, optically transparent embryo, and extremely rapid heart development (48 h). In addition the sequencing of the zebrafish genome is close to completion. Loss-of-function gene models have been successfully created in zebrafish with the use of morpholino-modified antisense oligonucleotides. In zebrafish embryos the cardiovascular system is functional at 24 h but not essential for survival of the early embryo, which obtains adequate oxygen by simple diffusion, and even mutant embryos with complete lack of circulation can survive for 3 d. This is an important advantage over the mammalian embryos, whose survival is dependent on an intact circulation, and allows the study of zebrafish mutants with cardiovascular defects, whose mammalian embryonic counterparts undergo rapid degeneration and absorption. Despite the advantages of the zebrafish system for mutational analysis, there are some aspects of cardiovascular construction, namely septation and pulmonary artery formation that must be studied in higher vertebrates. Finally, the fruit fly, Drosophila melanogaster, completes the panel of experimental models used in cardiogenomics. One of the most remarkable discoveries was of the gene needed in Drosophila to produce a heart. Finding this gene, named “tinman” after the Wizard of Oz character in need of a heart, led to the identification of the evolutionarily conserved, and previously unknown, class of homeodomain-containing “NK” genes, which are critical for cardiac muscle development in all animals. Orthologous relationships between NK and other cardiac gene regulatory networks in Drosophila and mouse have been identified, showing that control of cardiac formation is likely to have been conserved over 600 × 106 yr. Information on the genetic and molecular bases of cardiovascular development, function, and disease that has accumulated from the combination of data from animal models and human genetic studies is leading to a redefinition of the pathophysiology of congenital heart diseases, and innovative diagnostic and prognostic tools are emerging. This chapter discusses the anatomic malformations, which comprise the majority of cases of human congenital heart disease, but specifically excludes heritable cardiomyopathies, long QT syndrome and Marfan syndrome, discussed in other chapters.

CLINICAL FEATURES The majority of human newborns with hemodynamically significant heart defects present in the first 2 wk of life with one out of four physiological arrangements: obstruction to pulmonary arterial circulation, obstruction to systemic arterial circulation, inadequate mixing between pulmonary and systemic circulations, or pulmonary venous obstruction. Because of the two-connectedpumps-working-in-parallel construction of fetal cardiac circulation (Fig. 10-1, left panel), these four types of hemodynamic aberrations are well tolerated for a 40-wk gestation. An obstruction to the pulmonary or aortic outflow is reliably compensated by blood flow shifting to the contralateral side of the heart. Transposition of the great arteries with intact ventricular septum does not have hemodynamic consequences because it merely constitutes a different variety of two-connected-pumps-working-inparallel configuration. Anomalous pulmonary venous connection with pulmonary venous obstruction does not substantially alter prenatal hemodynamics because minimal blood flows into the lungs in utero. The four physiological derangements described earlier are unmasked when the circulation is acutely changed at birth to a two-unconnected-pumps-working-in-series arrangement, with lungs but without placenta (Fig. 10-1, right panel). The rapidity of presentation is critically dependent on the time-course of foramen closure or ductal closure or both. Impaired pulmonary flow manifests as cyanosis, whereas impaired systemic circulation manifests as “low output syndrome.” Inadequate mixing between the pulmonary and systemic circulations also appears as cyanosis. Because of the interstitial edema, pulmonary venous obstruction manifests as tachypnea in addition to cyanosis. Although these four hemodynamic subsets account for virtually all cases of heart defects appearing in the first 2 wk of life, severe cases of the rare malformation “absent pulmonary valve syndrome,” whose cardinal feature is ventilatory failure because of airway impingement by adjacent markedly dilated pulmonary arteries, may also present on the first day of life. A few types of malformation are evident after 2 wk of age. Isolated large septation defects resulting in left-to-right shunt physiology typically do not present in the first 2 wk of age because the magnitude of the left-to-right shunt depends mostly on the pulmonary vascular resistance, the latter parameter falling to its adult value gradually over the first 6 mo. The uncommon condition of anomalous origin of the left coronary artery from the pulmonary artery, which results in gradual-onset myocardial ischemia, also manifests after the neonatal period because the reduction in perfusion of the left ventricle is closely linked to the fall in pulmonary vascular resistance. Finally, in some rare forms of congenital heart defects, a physiologically corrected but inherently abnormal organization of the cardiac segments has no immediate hemodynamic effects (L-TGA). However, at a much later point, the longevity of this arrangement is challenged by the long-term mechanical cues (e.g., pressure or volume overload) normally not destined for these anatomic structures, and heart failure frequently occurs.

DIAGNOSIS Although the screening protocol includes physical examination, arterial blood gas, electrocardiogram, and chest roentgenogram, the principal tool for detailed characterization of morphology and hemodynamics in the human newborn is ultrasound imaging. With

CHAPTER 10 / CONGENITAL HEART DISEASE

91

Figure 10-1 (Left) Fetal circulation in the human. Oxygenated blood travels from the placenta to the fetus through the umbilical vein and then the ductus venosus. After combining with deoxygenated inferior vena caval blood, it preferentially streams across the foramen ovale into the left heart and eventually supplies the ascending aorta. Deoxygenated superior vena caval blood preferentially fills the right heart and supplies the descending aorta through the ductus arteriosus. Because they do not function as the gas exchange site, the lungs are not inflated in utero and receive very little blood flow. The two sides of the heart function as parallel pumps, communicating at the level of the atria (foramen ovale) and the great arteries (ductus arteriosus). (Right) Neonatal circulation in the human. The lungs inflate, assume their gas exchange function and begin to receive more blood flow. Ductus venosus and ductus arteriosus constrict in the first few days. The two sides of the heart function as pumps-inseries now; any blood passing through the right heart must eventually also pass through the left heart. (Reproduced with permission from Fray H Clemente CD: Gray’s Anatomy of the Human Body, 1986, Lea and Febiger, Philadelphia.) (Please see color insert.)

progressive refinement over the last 20 yr, its accuracy in assessing anatomy and circulatory physiology in the newborn and young infant now approaches that of cardiac catheterization. Taking a detailed family history and surveying the patient for subtle extracardiac anomalies will become an increasingly important part of the workup. For example, a prolonged PR interval in patients with ostium secundum atrial septal defect in healthy firstdegree relatives can be useful in genetic counseling because it indicates the possibility of mutation in the CSX/NKX2.5 gene. For the detection of deletions, translocations, or duplications, fluorescence in situ hybridization will become standard as probes become widely available. With the advances in detecting underlying genetic defects, it is expected that classification of the patients by phenotype will give way to classification by genotype.

GENETIC BASIS OF DISEASE AND MOLECULAR PATHOPHYSIOLOGY The prevailing view of the etiology of congenital heart disease had been that 8% of defects were single-gene defects, 2% were teratogens, and 90% were “multifactorial.” Three classes of observation argue against this view. First, as the number of adult survivors

of congenital heart disease in the United States approaches 1 million, surveys of the F1 generation reveal a much higher incidence of congenital heart malformation than previously reported. Second, an increasing number of genes responsible for syndromes involving the heart are being identified by positional cloning, a technique based on identifying a DNA marker with which a phenotype cosegregates. Third, mutations producing phenotypes similar to human congenital heart disease are being identified in several lower vertebrates (Table 10-2). In mouse, the technique of targeted gene disruption has uncovered models for DGS (Tbx1, Hoxa-3, RALDH2, CRKL knockout mice) and subpulmonary stenosis (connexin 43 knockout). Spontaneous mutations in the iv and pax-3 genes produce models for heterotaxy and truncus arteriosus, respectively. In zebrafish, initial mutagenesis screens have revealed mutants lacking endocardium or valves, manifesting tachyarrhythmias, and demonstrating myocardial hypoplasia or hyperplasia. Besides the growing list of molecules found to affect heart development, epigenetic factors are likely to be also involved in this process. For example, intracardiac hemodynamics and early myocardial function seem to affect the endothelial-mesenchymal transformation that precedes the development of endocardial

92

SECTION II / CARDIOLOGY

Table 10-2 Mutant Phenotypes Classified by Affected Heart Segment Left-right asymmetry

Great arteries

Outflow tract

Ventricle

Atrioventricular canal

Atrium

Global

Interruption/obstruction/septation

Septation

Septation

Septation

Septation

Cryptic ntl/T boz/dharma flh/znot ActRIIB inv kif3B kif3A Zic3 FoxJ1 Foxh1 Gdf1 Lrdyneina DNAH5a spt/tbx16a din/chordina

grd/hey2 Foxc2 Foxc1 Tbx1 Fgf8 Sema3C c-Jun ET-1 ETA ECE-1 JAGGED1 ELASTIN Bmp4 Np1 RXR; RAR Plexin D1

Pax3 PDGFRa Tbx1 Fgf8 Bmpr2 Dvl2 Nf1 Hey2 Sema3C c-Jun Cited2 GATA4 Bmp4 TGFb2

Bmp4 Hey2

Fog2 RXRα GATA4 Cx40; Cx43 Collagen6α1/α2

TBX5 Bmp4 Nkx2.5

Hypoplasia, valve abnormality

Hypoplasia, abnormal trabeculation

Hypoplasia, valve abnormality

Hypoplasia

NFATc1 JAGGED1 Sox4 Bmp6; Bmp7 Egfr;shp2 RXRa PTPN11 Cx45 Cx43 Isl1 Mef2c ALK2 ALK3 Foxp1

dHand foggy/Spt5 pandora/Spt6 Notch2 Pax3 5-HT(2B) ErbB2 EphrinB2 EphB4 N-Myc Hey2 VCAM1 RXRα Cx45 GATA4 Bmp10 Alk3 Epo, EpoR Agpt1

jekyll/udpgd FIBRILLIN1 Cx45 Bmp4 Alk3 ErbB3

RALDH2 Isl1

Regional oep Lefty1

Dilation COLL3A1 FIBRILLIN1

aLrdynein, spt, din and DNAH5 model primary ciliary dyskinesia (Kartagener-like conditions with randomization to situs inversus totalis and situs solitus). The other mutants in first column all model heterotaxy syndrome.

cushions and formation of heart valves. In addition, clinical and experimental data suggest that abnormal cardiac function because of alterations of the cardiac contractile and cytoskeletal properties can impair ventricular growth. Reverse genetics in humans has been successful as exemplified by the mapping of several familial disorders to specific chromosomal loci. Familial supravalvar aortic stenosis (SVAS), an autosomal-dominant disorder affecting many large arteries (particularly the pulmonary arteries and the ascending aorta) was found to be coinherited with a DNA polymorphism at 7q11.23. Analysis of a SVAS family in which a balanced translocation, involving chromosome 7 cosegregated with the phenotype showed that the elastin gene was disrupted. Another SVAS family manifested a large deletion in the elastin gene. In addition, SVAS also occurs as part of Williams syndrome (supravalvar aortic or pul-

monary stenosis, stenoses of other systemic arteries, elfin facies, precocious verbal ability, hypersensitivity to sounds, poor visualmotor integration); 95% of Williams syndrome patients have a 1.5 Mb deletion encompassing 28 genes including elastin. Of patients who have a complete atrioventricular canal, approx 85% exhibit Down syndrome. In the complete form of common atrioventricular canal, the atria and ventricles are incompletely septated; in addition, the atria drain into the ventricles via a single (common) atrioventricular valve with typically five leaflets. Through studies of families with various partial 21 trisomies, a critical region of 9 Mb for atrioventricular canal defects has been identified (21q22.2-21q22.3). However, a specific gene has not yet been identified as responsible for the common atrioventricular canal. Holt Oram syndrome (an autosomal-dominant condition with forelimb anomalies, ostium secundum atrial septal defects, and

CHAPTER 10 / CONGENITAL HEART DISEASE

atrioventricular block) is caused by mutations in the gene TBX5, which was cloned from the disease locus on human chromosome 12q24. Finally, progress has been made in the search for the etiology of interrupted aortic arch, truncus arteriosus, and Tetralogy of Fallot (TOF) (Tables 10-1 and 10-2). These defects with Velocardio-facial syndrome (VCFS)/DGS are associated with the microdeletion in the region 22q11 (DiGeorge Critical Region). The anomalies in the VCFS/DGS phenotype are in structures to which migrating neural crest cells contribute, so it is understandable that much attention is being focused on genes known to be important in cell–cell signaling and cell movement. Other work centering on transcription factors, which may function as master controls, is also important. TBX1, a transcription factor within the deleted region, is emerging as the critical gene for cardiac outflow tract development with a dose dependent function (Fig. 10-2). Intriguingly, human patients with an isolated deletion or mutation of TBX1 had not been reported until a few patients with DGS in Japan were identified, who harbor nonfunctional TBX1 mutations. The interaction of TBX1 with modifier genes is a possible explanation for this paradox, although only a few potential TBX1 partners have been identified. However, it is likely that TBX1 has several other partners that contribute to the variable expressivity of the mutant phenotype, and this model could be operational in other congenital heart diseases. Although the physiologically based schema mentioned in the clinical features section works well for the clinician, a developmentally oriented classification of defects may be more appropriate for the molecular biologist. The important steps of cardiac development as known from different species should be taken into account, starting with the genesis and migration of precardiac cells from the lateral plate mesoderm followed by their ventral migration to form a single beating heart tube and the activation of genes responsible for the subsequent morphogenetic steps of looping, wedging, atrio-ventricular septation, formation of the outflow tracts largely from pharyngeal mesoderm and endocardial cushions, and aortic arch formation and remodeling.

MANAGEMENT/TREATMENT Today, surgical or catheter-based reconstruction of most congenital heart malformations is virtually always feasible. Even the varieties with hypoplasia of one or the other ventricle can be palliated for approx 20 yr with a modified Fontan operation, in which the systemic venous circulation is rerouted directly into the pulmonary arteries. Assessment of the quality of the surgical repair is no longer done 6 or 12 mo postoperatively but rather in the operating room or shortly before hospital discharge. Long-term follow-up typically uses noninvasive techniques such as ultrasound, MRI, and exercise testing. However, in most cases the treatment is palliative, not curative, and the incidence of residual disease remains relatively high among the exponentially expanding population of adult survivors with “corrected” congenital heart disease. For example, patients with Fontan physiology need a series of “staged” operations and are at risk for developing recurrent effusions, arrhythmias, hepatic dysfunction and protein losing enteropathy. Moreover, the single functional ventricle in these patients gradually decompensates, and they eventually need heart transplantation. Patients with severe forms of Tetralogy of Fallot may need successive surgeries for replacement of their conduits because of recurrent stenosis or for inability to tolerate severe residual pulmonary insufficiency. Lesions repaired through ventriculotomy (such as

93

double-outlet right ventricle and truncus arteriosus) pose increased risk for fatal arrhythmias later in life. Further medical and surgical advances may address some of these remaining questions; in particular the advent of fetal cardiac surgery may open new perspectives in the management of congenital heart diseases. Alternatively, other ways of further lowering the morbidity of congenital heart defects are prevention (by genotyping, counseling, or some form of early “genetic” intervention), termination in utero on phenotype recognition, and better prenatal as well as postnatal screening. Although the second may be practiced in some countries, for religious and ethical reasons it is doubtful whether it will ever be widely implemented in the United States. The third way (improved screening methods) has not proved cost-effective. Hence, efforts will likely be concentrated on the first approach. As a prerequisite, more work is needed to understand the construction of the heart and vessels at the molecular level. Once the specific genetic basis for cardiac defects is identified, it should become possible to establish the basis for genetic counseling and treatment. In particular, genotypic analysis could determine the risk status of the patients’ relatives or identify individuals predisposed for specific defects at an early stage; bioengineering technology could allow the delivery of a corrected gene copy or correction of a structural defect. Stem cell biology and tissue engineering could generate parts (valves, vessels, or tissue for patches) to replace or repair defective heart tissue, thereby avoiding the complications associated with the use of artificial prostheses and heart transplantation. It would necessitate technological development in the fields of gene therapy, biopharmaceuticals, organ culture, and fetal surgery. The knowledge that secondary factors affect the ultimate phenotype of genetic mutations predicts that at least some types of congenital heart defects might be prevented by modulating these factors. Hence, it may become possible to choose drugs that would regulate gene expression to overcome a problem. For example, virtually all morphological and molecular abnormalities elicited by dietary or genetic retinoid deficiency can be rescued by systemic (maternal) administration, reminiscent of preventing neural tube defects with folic acid supplementation in women of reproductive age.

FUTURE DIRECTIONS The following are some of the questions that biologists hope to answer: How do the myocardial, endocardial, endothelial, and “cardiac neural crest” lineages become specified? How is the heart tube correctly positioned in the embryo and then “patterned?” Do the same families of molecules responsible for body axis patterning get reused in the patterning of organs? What is the role of the endocardium in the development of the heart? How are the cardiac valves formed? How is the “specialized” conduction system formed? (Humans with abnormal lateralization or looping have anomalous conduction systems.) Given that in mammalian hearts two atria and two ventricles function in parallel (and develop in tandem in utero), how is the growth of the left heart exquisitely matched to that of the right heart throughout gestation? How are the pulmonary vasculature and coronary vasculature formed? How are myocardial cells maintained in a state of terminal differentiation and how can nonheart or stem cells be reprogrammed toward a cardiogenic fate? A variety of model organisms will yield answers to these questions. Although laboratories studying the mouse system will continue to create knockouts (including double and triple knockouts,

94

SECTION II / CARDIOLOGY

Figure 10-2 Dosage sensitive role of Tbx1 in the etiology of cardiovascular defects in mice. Aortic arch variations in E18.5 mouse embryos with different Tbx1 dosage as seen by corrosion casts (A, F) or tissue dissection (B–E). Higher than normal Tbx1 dosage in transgenics overexpressing Tbx1 on wild type background (TG;+/+) causes several outflow tract and aortic arch abnormalities (A, B). (A) Pulmonary atresia with absent main pulmonary artery: the ductus arteriosus supplies the branch pulmonary arteries bilaterally. (B) Interrupted aortic arch type B (*indicates the site of interruption). As the Tbx1 dosage normalizes in mice harboring the overexpressing transgene on Tbx1–/– background (TG;–/–) or in wild mice (+/+) the cardiovascular abnormalities are rescued (C, D). Both (C) and (D) display normal aortic arch anatomy; the arrow in D indicates the bifurcation of the innominate artery into right subclavian and right carotid artery. Lower than normal Tbx1 dosage as in mice happloinsufficient (+/–) or null for Tbx1 (–/–) causes similar conotruncal and aortic arch defects as higher than normal Tbx1 dosage (E, F). (E) Aberrant right subclavian artery: the arrow points to the right carotid artery originating as a single vessel from the aortic arch because of the aberrant origin of the right subclavian from the descending aorta (the aberrant retroesophageal subclavian artery cannot be seen in this view). (F) Truncus arteriosus with the pulmonary arteries originating directly from the common arterial trunk. (Please see color insert.)

CHAPTER 10 / CONGENITAL HEART DISEASE

and so on), mice carrying specific mutations of nonnull, i.e., hypomorphic alleles may be more illuminating, because they may mimic the human conditions more closely. “Conditional,” such as tissue- or stage-specific, mutant lines will also become more common. Zebrafish labs will continue to analyze the zygotic recessive mutations affecting cardiovascular development. Cloning the genes affected in these mutants will be facilitated by the completion of sequencing of the organism’s genome. However, the construction of additional genetic tools such as transgenic fluorescent-reporter zebrafish lines whose embryos fluoresce green when a structure or tissue of interest is first specified could serve as the starting reagent for large-scale “targeted” mutagenesis screens in this organism. In the current postgenomic era, with the complete genome sequences of several species (including human, mouse, and zebrafish) at hand, the challenge will be to understand the function of genes within the cardiovascular system. One of the most powerful methods available to assign function to a gene is to inactivate the gene by reverse genetic approaches. In this regard loss-of-function gene models have been successfully created in zebrafish with the use of antisense morpholino oligonucleotides that can specifically inhibit the translation of a specific gene. However, this method is suited only for early developmental stages with the additional drawback of nonspecific side effects. Recently, a new reverse genetic approach, hitherto utilized in plants, target induced local lesions in genomes is being applied to zebrafish to generate allelic series in individual genes of interest. In humans, positional cloning will continue to be employed for large families with single-gene disorders. Attempts to correlate particular amino acid substitutions with specific human cardiac malformations will afford further insight into the structure-function relationships of the relevant proteins, whether they are transcription factors, signals, or receptors.

SELECTED REFERENCES Abu-Issa R, Smyth G, Smoak I, Yamamura K, Meyers EN. Fgf8 is required for pharyngeal arch and cardiovascular development in the mouse. Development 2002;129:4613–4625. Bao ZZ, Bruneau BG, Seidman JG, Seidman CE, Cepko CL. Regulation of chamber-specific gene expression in the developing heart by Irx4. Science 1999;283:1161–1164. Bartman T, Walsh EC, Wen KK, et al. Early myocardial function affects endocardial cushion development in zebrafish. PLoS Biol 2004;2:E129. Basson CT, Bachinsky DR, Lin RC, et al. Mutations in human TBX5 cause limb and cardiac malformation in Holt-Oram syndrome. Nat Genet 1997;15:30–35. Biben C, Weber R, Kesteven S, et al. Cardiac septal and valvular dysmorphogenesis in mice heterozygous for mutations in the homeobox gene Nkx2-5. Circ Res 2001;87(10):888–895. Bisgrove BW, Essner JJ, Yost HJ. Multiple pathways in the midline regulate concordant brain, heart and gut left-right asymmetry. Development 2000;127:3567–3579. Brown CB, Boyer AS, Runyan RB, Barnett JV. Requirement of type III TGF-beta receptor for endocardial cell transformation in the heart. Science 1999;283:2080–2082. Brown CB, Feiner L, Lu MM, et al. PlexinA2 and semaphorin signaling during cardiac neural crest development. Development 2001;128: 3071–3080. Bruneau BG, Nemer G, Schmitt JP, et al. A murine model of Holt-Oram syndrome defines roles of the T-box transcription factor Tbx5 in cardiogenesis and disease. Cell 2001;106:709–721. Cai CL, Liang X, Shi Y, et al. Isl1 identifies a cardiac progenitor population that proliferates prior to differentiation and contributes a majority of cells to the heart. Dev Cell 2003;5:877–889. Camenisch TD, Spicer AP, Brehm-Gibson T, et al. Disruption of hyaluronan synthase-2 abrogates normal cardiac morphogenesis and

95

hyaluronan-mediated transformation of epithelium to mesenchyme. J Clin Invest 2000;106:349–360. Carraway KL III. Involvement of the neuregulins and their receptors in cardiac and neural development. Bioessays 1996;18:263–266. Chang CP, Neilson JR, Bayle JH, et al. A field of myocardial-endocardial NFAT signaling underlies heart valve morphogenesis. Cell 2004;118: 649–663. Chang H, Zwijsen A, Vogel H, Huylebroeck D, Matzuk MM. Smad5 is essential for left-right asymmetry in mice. Dev Biol 2000;219:71–78. Charron J, Malynn BA, Fisher P, et al. Embryonic lethality in mice homozygous for a targeted disruption of the N-myc gene. Genes Dev 1992;6: 2248–2257. Chen F, Kook H, Milewski R, et al. Hop is an unusual homeobox gene that modulates cardiac development. Cell 2002;110:713–723. Conway SJ, Henderson DJ, Kirby ML, Anderson RH, Copp AJ. Development of a lethal congenital heart defect in the splotch (Pax3) mutant mouse. Cardiovasc Res 1997;36:163–173. de la Pompa JL, Timmerman LA, Takimoto H, et al. Role of the NF-ATc transcription factor in morphogenesis of cardiac valves and septum. Nature 1998;392:182–186. Delot EC, Bahamonde ME, Zhao M, Lyons KM. BMP signaling is required for septation of the outflow tract of the mammalian heart. Development 2003;130:209–220. Deng Z, Morse JH, Slager SL, et al. Familial primary pulmonary hypertension (gene PPH1) is caused by mutations in the bone morphogenetic protein receptor-II gene. Am J Hum Genet 2000;67:737–744. Donovan J, Kordylewska A, Jan YN, Utset MF. Tetralogy of Fallot and other congenital heart defects in Hey2 mutant mice. Curr Biol 2002; 12:1605–1610. Elliott DA, Kirk EP, Yeoh T, et al. Cardiac homeobox gene NKX2-5 mutations and congenital heart disease: associations with atrial septal defect and hypoplastic left heart syndrome. J Am Coll Cardiol 2003;41(11): 2072–2076. Funke B, Epstein JA, Kochilas LK, et al. Mice overexpressing genes from the 22q11 region deleted in velo-cardio- facial syndrome/DiGeorge syndrome have middle and inner ear defects. Hum Mol Genet 2001; 10:2549–2556. Gaio U, Schweickert A, Fischer A, et al. A role of the cryptic gene in the correct establishment of the left- right axis. Curr Biol 1999;9:1339–1342. Gajewski K, Fossett N, Molkentin JD, Schulz RA. The zinc finger proteins Pannier and GATA4 function as cardiogenic factors in Drosophila. Development 1999;126:5679–5688. Garg V, Kathirya IS, Barnes R, et al. GATA4 mutations cause human congenital heart defects and reveal an interaction with TBX5. Nature 2003;24:443–447 Gerety SS, Anderson DJ. Cardiovascular ephrinB2 function is essential for embryonic angiogenesis. Development 2002;129:1397–1410. Gessler M, Knobeloch KP, Helisch A, et al. Mouse gridlock: no aortic coarctation or deficiency, but fatal cardiac defects in Hey2 –/– mice. Curr Biol 2002;12:1601–1604. Ghyselinck NB, Wendling O, Messaddeq N, et al. Contribution of retinoic acid receptor beta isoforms to the formation of the conotruncal septum of the embryonic heart. Dev Biol 1998;198:303–318. Goldmuntz E. Recent advances in understanding the genetic etiology of congenital heart disease. Curr Opin Pediatr 1999;11:437–443. Goldmuntz E, Bamford R, Karkera JD, dela Cruz J, Roessler E, Muenke M. CFC1 mutations in patients with transposition of the great arteries and double-outlet right ventricle. Am J Hum Genet 2002;70: 776–780. Hamblet NS, Lijam N, Ruiz-Lozano P, et al. Dishevelled 2 is essential for cardiac outflow tract development, somite segmentation and neural tube closure. Development 2002;129:5827–5838. Hatcher CJ, Kim MS, Mah CS, et al. TBX5 transcription factor regulates cell proliferation during cardiogenesis. Dev Biol 2001;230:177–188. Hornberger LK, Sanders SP, Rein AJ, Spevak PJ, Parness IA, Colan SD. Left heart obstructive lesions and left ventricular growth in the midtrimester fetus. A longitudinal study. Circulation 1995;92:1531–1538. Hove JR, Koster RW, Forouhar AS, Acevedo-Bolton G, Fraser SE, Gharib M. Intracardiac fluid forces are an essential epigenetic factor for embryonic cardiogenesis. Nature 2003;421:172–177.

96

SECTION II / CARDIOLOGY

Iwamoto R, Yamazaki S, Asakura M, et al. Heparin-binding EGF-like growth factor and ErbB signaling is essential for heart function. Proc Natl Acad Sci USA 2003;100:3221–3226. Izraeli S, Lowe LA, Bertness VL, et al. The SIL gene is required for mouse embryonic axial development and left- right specification. Nature 1999;399:691–694. Jackson LF, Qiu TH, Sunnarborg SW, et al. Defective valvulogenesis in HB-EGF and TACE-null mice is associated with aberrant BMP signaling. EMBO J 2003;22:2704–2716. Jerome LA, Papaioannou VE. DiGeorge syndrome phenotype in mice mutant for the T-box gene, Tbx1. Nat Genet 2001;27:286–291. Kasahara H, Lee B, Schott JJ, et al. Loss of function and inhibitory effects of human CSX/NKX2.5 homeoprotein mutations associated with congenital heart disease. J Clin Invest 2000;106:299–308. Keating M. Elastin and vascular disease. Trends Cardiovasc Med 1994; 4:202–206. Kelly RG, Brown NA, Buckingham ME. The arterial pole of the mouse heart forms from Fgf10-expressing cells in pharyngeal mesoderm. Dev Cell 2001;1:435–440. Kim RY, Robertson EJ, Solloway MJ. Bmp6 and Bmp7 are required for cushion formation and septation in the developing mouse heart. Dev Biol 2001;235:449–466. King T, Beddington RS, Brown NA. The role of the brachyury gene in heart development and left-right specification in the mouse. Mech Dev 1998;79:29–37. Kirby ML, Gale TF, Stewart DE. Neural crest cells contribute to normal aorticopulmonary septation. Science 1983;220:1059–1061. Kitajima S, Takagi A, Inoue T, Saga Y. MesP1 and MesP2 are essential for the development of cardiac mesoderm. Development 2000;127: 3215–3226. Kitamura K, Miura H, Miyagawa-Tomita S, et al. Mouse Pitx2 deficiency leads to anomalies of the ventral body wall, heart, extra- and periocular mesoderm and right pulmonary isomerism. Development 1999; 126:5749–5758. Klinedinst SL, Bodmer R. Gata factor Pannier is required to establish competence for heart progenitor formation. Development 2003; 130:3027–3038. Kuo CT, Morrisey EE, Anandappa R, et al. GATA4 transcription factor is required for ventral morphogenesis and heart tube formation. Genes Dev 1997;11:1048–1060. Lai YT, Beason KB, Brames GP, et al. Activin receptor-like kinase 2 can mediate atrioventricular cushion transformation. Dev Biol 2000; 222:1–11. Laugwitz KL, Moretti A, Lam J, et al. Postnatal isl1 + cardioblasts enter fully differentiated cardiomyocyte lineages. Nature 2005;433:647–653. Li S, Zhou D, Lu MM, Morrisey EE. Advanced cardiac morphogenesis does not require heart tube fusion. Science 2004;305:1619–1622. Liao J, Kochilas L, Nowotschin S, et al. Full spectrum of malformations in velo-cardio-facial syndrome/DiGeorge syndrome mouse models by altering Tbx1 dosage. Hum Mol Gen 2004;13(15):1–9. Lin Q, Lu J, Yanagisawa H, et al. Requirement of the MADS-box transcription factor MEF2C for vascular development. Development 1998;125:4565–4574. Lindsay EA, Botta A, Jurecic V, et al. Congenital heart disease in mice deficient for the DiGeorge syndrome region. Nature 1999;401:379–383. Lindsay EA, Vitelli F, Su H, et al. Tbx1 haploinsufficieny in the DiGeorge syndrome region causes aortic arch defects in mice. Nature 2001; 410:97–101. Liu C, Liu W, Lu MF, Brown NA, Martin JF. Regulation of left-right asymmetry by thresholds of Pitx2c activity. Development 2001;128: 2039–2048. Meno C, Shimono A, Saijoh Y, et al. lefty-1 is required for left-right determination as a regulator of lefty-2 and nodal. Cell 1998;94:287–297. Merscher S, Funke B, Epstein JA, et al. TBX1 is responsible for cardiovascular defects in velo-cardio-facial/DiGeorge syndrome. Cell 2001;104:619–629. Meyers EN, Martin GR. Differences in left-right axis pathways in mouse and chick: functions of FGF8 and SHH. Science 1999;285:403–406. Michelson AM, Gisselbrecht S, Zhou Y, Baek KH, Buff EM. Dual functions of the heartless fibroblast growth factor receptor in development of the Drosophila embryonic mesoderm. Dev Genet 1998;22:212–229.

Milewicz DM. Molecular genetics of Marfan syndrome and EhlersDanlos type IV. Curr Opin Cardiol 1998;13:198–204. Molkentin JD, Lin Q, Duncan SA, Olson EN. Requirement of the transcription factor GATA4 for heart tube formation and ventral morphogenesis. Genes Dev 1997;11:1061–1072. Nasevicius A, Ekker SC. Effective targeted gene “knockdown” in zebrafish. Nat Genet 2000;26:216–220. Niederreither K, Vermot J, Messaddeq N, Schuhbaur B, Chambon P, Dolle P. Embryonic retinoic acid synthesis is essential for heart morphogenesis in the mouse. Development 2001;128:1019–1031. Ranger AM, Grusby MJ, Hodge MR, et al. The transcription factor NF-ATc is essential for cardiac valve formation. Nature 1998;392:186–190. Reaume AG, de Sousa PA, Kulkarni S, et al. Cardiac malformation in neonatal mice lacking connexin43. Science 1995;267:1831–1834. Reifers F, Walsh EC, Leger S, Stainier DY, Brand M. Induction and differentiation of the zebrafish heart requires fibroblast growth factor 8 (fgf8/acerebellar). Development 2000;127:225–235. Saga Y, Kitajima S, Miyagawa-Tomita S. Mesp1 expression is the earliest sign of cardiovascular development. Trends Cardiovasc Med 2000; 10:345–352. Sakata Y, Kamei CN, Nakagami H, Bronson R, Liao JK, Chin MT. Ventricular septal defect and cardiomyopathy in mice lacking the transcription factor CHF1/Hey2. Proc Natl Acad Sci USA 2002;99: 16197–16202. Satoda M, Zhao F, Diaz GA, et al. Mutations in TFAP2B cause Char syndrome, a familial form of patent ductus arteriosus. Nat Genet 2000; 25:42–46. Schneider VA, Mercola M. Wnt antagonism initiates cardiogenesis in Xenopus laevis. Genes Dev 2001;15:304–315. Sissman NJ, Lefkowitz RJ, Willerson JT. Incidence of congenital heart disease. JAMA 2001;285:2579–2580. Srinivasan S, Hanes MA, Dickens T, et al. A mouse model for hereditary hemorrhagic telangiectasia (HHT) type 2. Hum Mol Genet 2003;12: 473–482. Srivastava D. HAND proteins: molecular mediators of cardiac development and congenital heart disease. Trends Cardiovasc Med 1999;9:11–18. Srivastava D, Olson EN. A genetic blueprint for cardiac development. Nature 2000;407:221–226. Stainier DY, Fouquet B, Chen JN, et al. Mutations affecting the formation and function of the cardiovascular system in the zebrafish embryo. Development 1996;123:285–292. Stalmans I, Lambrechts D, De Smet F, et al. VEGF: a modifier of the del22q11 (DiGeorge) syndrome? Nat Med 2003;9:173–182. Sucov HM, Dyson E, Gumeringer CL, Price J, Chien KR, Evans RM. RXR alpha mutant mice establish a genetic basis for vitamin A signaling in heart morphogenesis. Genes Dev 1994;8:1007–1018. Supp DM, Brueckner M, Kuehn MR, et al. Targeted deletion of the ATP binding domain of left-right dynein confirms its role in specifying development of left-right asymmetries. Development 1999;126: 5495–5504. Tartaglia M, Kalidas K, Shaw A, et al. PTPN11 mutations in Noonan syndrome: molecular spectrum, genotype-phenotype correlation, and phenotypic heterogeneity. Am J Hum Genet 2002;70:1555–1563. Tevosian SG, Deconinck AE, Tanaka M, et al. FOG-2, a cofactor for GATA transcription factors, is essential for heart morphogenesis and development of coronary vessels from epicardium. Cell 2000;101: 729–739. Thomas T, Yamagishi H, Overbeek PA, Olson EN, Srivastava D. The bHLH factors, dHAND and eHAND, specify pulmonary and systemic cardiac ventricles independent of left-right sidedness. Dev Biol 1998;196:228–236. Trembath RC. Mutations in the TGF-beta type 1 receptor, ALK1, in combined primary pulmonary hypertension and hereditary haemorrhagic telangiectasia, implies pathway specificity. J Heart Lung Transplant 2001;20:175. Tsukui T, Capdevila J, Tamura K, et al. Multiple left-right asymmetry defects in Shh(–/–) mutant mice unveil a convergence of the shh and retinoic acid pathways in the control of Lefty-1. Proc Natl Acad Sci USA 1999;96:11,376–11,381. Vermot J, Niederreither K, Garnier JM, Chambon P, Dolle P. Decreased embryonic retinoic acid synthesis results in a DiGeorge syndrome

CHAPTER 10 / CONGENITAL HEART DISEASE

phenotype in newborn mice. Proc Natl Acad Sci USA 2003;100: 1763–1768. von Both I, Silvestri C, Erdemir T, et al. Foxh1 is essential for development of the anterior heart field. Developmental Cell 2004;7(3): 331–345. Wallace KN, Yusuff S, Sonntag JM, Chin AJ, Pack M. Zebrafish hhex regulates liver development and digestive organ chirality. Genesis 2001; 30:141–143. Wang B, Weidenfeld J, Lu MM, et al. Foxp1 regulates cardiac outflow tract, endocardial cushion morphogenesis and myocyte proliferation and maturation. Development 2004;131(18):4477–4487. Wendling O, Dennefeld C, Chambon P, Mark M. Retinoid signaling is essential for patterning the endoderm of the third and fourth pharyngeal arches. Development 2000;127:1553–1562. Wienholds E, van Eeden F, Kosters M, et al. Efficient target-selected mutagenesis in zebrafish. Genome Res 2003;13:2700–2707. Winnier GE, Kume T, Deng K, et al. Roles for the winged helix transcription factors MF1 and MFH1 in cardiovascular development revealed

97

by nonallelic noncomplementation of null alleles. Dev Biol 1999; 213:418–431. Xue Y, Gao X, Lindsell CE, et al. Embryonic lethality and vascular defects in mice lacking the Notch ligand Jagged1. Hum Mol Genet 1999;8: 723–730. Ya J, Schilham MW, de Boer PA, Moorman AF, Clevers H, Lamers WH. Sox4-deficiency syndrome in mice is an animal model for common trunk. Circ Res 1998;83:986–994. Yanagisawa H, Hammer RE, Richardson JA, et al. Disruption of ECE-1 and ECE-2 reveals a role for endothelin-converting enzyme-2 in murine cardiac development. J Clin Invest 2000;105:1373–1382. Yanagisawa H, Hammer RE, Richardson JA, Williams SC, Clouthier DE, Yanagisawa M. Role of Endothelin-1/Endothelin-A receptor-mediated signaling pathway in the aortic arch patterning in mice. J Clin Invest 1998;102:22–33. Yin Z, Haynie J, Yang X, et al. The essential role of Cited2, a negative regulator for HIF-1alpha, in heart development and neurulation. Proc Natl Acad Sci USA 2002;99:10,488–10,493.

11 Inherited Cardiomyopathies CAROLYN Y. HO AND CHRISTINE E. SEIDMAN SUMMARY

traditionally typified by unexplained cardiac hypertrophy in a nondilated ventricle (Fig. 11-1). By definition, this refers to myocardial hypertrophy that occurs in the absence of any systemic or cardiac condition (such as hypertension or aortic stenosis) which may account for an increased load on the heart. The most common pattern of hypertrophy is asymmetric septal hypertrophy, but there is a significant variation in both the location and extent of left-ventricular hypertrophy (LVH) in affected individuals. The histopathological hallmarks of this condition are myocyte hypertrophy with myocardial disarray and fibrosis (Fig. 11-2). Genetic studies have defined HCM to be a disease of the sarcomere, caused by mutations in genes encoding different components of the contractile apparatus. Clinical manifestations of HCM are diverse and its description has been largely shaped by available diagnostic tools. Traditionally the most obvious and easily characterized features of this condition have been emphasized: outflow tract obstruction and LVH. However, resting outflow tract obstruction occurs in only approx 25% of patients. A provocable gradient may be detected in other patients by administering medications or performing maneuvers that reduce afterload or preload or increase cardiac contractility. EPIDEMIOLOGY The prevalence of unexplained LVH in the general population is estimated to be 1 in 500. Most of these cases are likely attributable to mutations in genes that encode sarcomere proteins, making HCM the most common genetic cardiovascular disorder. HCM is also the leading cause of sudden death among competitive athletes in the United States. SYMPTOMS The clinical spectrum of HCM is diverse. Although some individuals experience no or minor symptoms and are diagnosed only in the course of family screening, others develop refractory symptoms of pulmonary congestion, progressing to end-stage heart failure that may require cardiac transplantation. In a small subset, sudden cardiac death (SCD) may be the presenting event. Shortness of breath, particularly on exertion, is the most common symptom of HCM, occurring in about 90% of patients. Other manifestations include chest pain (approx 30%, often exertional), palpitations, orthostatic lightheadedness, presyncope and syncope (15–25%), orthopnea/paroxysmal nocturnal dyspnea, and fatigue. There is no close correlation between the degree of LVH or outflow tract obstruction and the severity of symptoms. Although often multifactorial in etiology, the occurrence of syncope may be a marker for increased risk of SCD, especially in younger individuals.

Genetics is an emerging field in cardiovascular medicine. Remarkably, less than 50 yr ago genetics was a nascent field of basic research with little apparent relevance to cardiovascular science or any other medical subspecialty. Yet today cardiovascular genetics is a discipline that fully integrates high technology laboratory investigation and clinical medicine. From this unusual hybrid have emerged discoveries that precisely identify cause in here-to-for “idiopathic” disorders, that provide fundamental insights into disease processes, and that delineate subtypes in well-defined pathologies. Insights from these discoveries uproot traditional anatomic classifications of disease and integrate cell physiology and molecular biochemistry into the study of pathology. For researchers, practitioners, and patients alike, cardiovascular genetics has a growing impact on the definition and diagnosis of disease, on explaining prognosis, and expanding treatments. Key Words: Dilated cardiomyopathy; DNA; gene; genetic risk; hypertrophic cardiomyopathy; mutation; modes of inheritance; preclinical disease; sequence.

INTRODUCTION Cardiomyopathies are disorders of the myocardium, which arise from a variety of etiologies and trigger pathways that may culminate in hypertrophic or dilated remodeling of the heart. The details of the cellular and molecular events that lead to compensatory forms of cardiac remodeling in response to a superimposed load remain largely unknown. In contrast, there have been greater advances in the study of primary cardiomyopathies, disorders of cardiac myocytes, which remodel the myocardium in the absence of other underlying pathology. Inherited gene defects are increasingly recognized as the most common cause of hypertrophic cardiomyopathy (HCM) and a frequent cause of dilated cardiomyopathy (DCM). Elucidating the molecular mechanisms leading from genetic mutation to the clinical expression of disease will have profound effects on the understanding of broader issues of basic myocyte structure and function, and influence the approach to disease management.

HYPERTROPHIC CARDIOMYOPATHY CLINICAL ASPECTS HCM was first described over 100-yr-ago but the modern characterization dates to 1959. It has been From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

98

CHAPTER 11 / INHERITED CARDIOMYOPATHIES

99

Figure 11-1 Gross pathological specimens of a heart with (A) hypertrophic cardiomyopathy (HCM) and (C) dilated cardiomyopathy (DCM). Note the marked increased in left-ventricular hypertrophy (HCM) and chamber dimensions (DCM) as compared with (B) the normal heart. (Please see color insert.)

Figure 11-2 Histopathology of distinct human cardiomyopathies revealed by hematoxylin and eosin staining. (A) Hypertrophic cardiomyopathy specimen shows characteristic features of myocyte hypertrophy with myocardial disarray and fibrosis. (B) In contrast, histological specimens from patients with PRKAG2 mutations show myocytes with nonmembrane bound vacuoles (arrows) that stain for glycogen and amylopectin. Note mild amounts of fibrosis and absence of significant myocyte disarray. (From Arad M, Benson DW, Perez-Atayde AR, et al., 2002. Constitutively active AMP kinase mutations cause glycogen storage disease mimicking hypertrophic cardiomyopathy. J Clin Invest 2002;109:357–362. [Reproduced with permission of J Clin Invest.]) (C) Dilated cardiomyopathy caused by a phospholamban missense mutation shows myocyte enlargement without disarray and marked interstitial fibrosis. (From Science 2003;299:1410–1413.) (Please see color insert.)

PHYSICAL EXAMINATION Left ventricular (LV) systolic function is typically preserved; however, abnormal diastolic function has been well documented and may largely account for symptoms of pulmonary congestion and exercise intolerance. Diastolic dysfunction may be a fundamental manifestation of altered sarcomere

function. It precedes the development of LVH in preliminary studies of individuals who have inherited a causal sarcomere gene mutation, but have not yet developed other typical features of disease. In addition to impaired contractile function, some individuals with HCM have altered blood pressure response to exercise, possibly related to

100

SECTION II / CARDIOLOGY

abnormal vasomotor tone. This finding may indicate a worse prognosis with a higher risk for sudden death. Typical findings on physical examination include a prominent LV apical impulse or lift, a fourth heart sound (S4), and a brisk, occasionally bifid carotid upstroke. If obstruction is present, there may be a harsh crescendo-decrescendo systolic murmur typically best heard at the apex and lower left sternal border, radiating to the axilla and base, but usually not to the neck. This murmur may reflect outflow tract obstruction as well as mitral regurgitation. NATURAL HISTORY The natural history of HCM is highly variable, even among family members who have inherited the same causal mutation. HCM rarely presents in infancy or early childhood; development of LVH typically occurs in adolescence in conjunction with the pubertal growth spurt. The age of onset of hypertrophy may be determined to some extent by the specific nature of the underlying gene defect. Disease caused by mutations in the β-myosin heavy chain (β-MHC) gene is generally associated with phenotypically obvious disease with near-universal development of LVH by the second decade. In contrast, disease caused by mutations in the cardiac myosin binding protein C (cMyBPC) gene may not display clinically evident hypertrophy until the fourth or fifth decade of life. cMyBPC mutations have been associated with elderly onset HCM with initial clinical manifestations developing late in life. Estimates of the annual mortality of HCM vary. Evaluation of populations drawn from specialized referral centers suggests a significant annual mortality rate of 4–6%. In contrast, communitybased studies, which may be less susceptible to selection bias, suggest a more benign outcome with a projected annual mortality rate of 1–2%. HCM-specific causes of morbidity and mortality include SCD, progressive heart failure, atrial fibrillation, and heart failure, and stroke. Sudden death is the most feared complication of HCM and accurate estimation of an individual’s risk for sudden death is a challenge. The risk for sudden death in the overall HCM population varies from 1 to 5% with about 10–20% of patients at highest risk. Less than 10–20% of patients progress to the “burntout” phase of HCM, marked by worsening symptomatic heart failure, LV systolic dysfunction, regression of LVH, and chamber dilatation. These patients may ultimately require cardiac transplantation for end-stage heart failure. MANAGEMENT There are two major aspects to HCM management: alleviation of symptoms and assessment of the risk for SCD. Medical therapy is the cornerstone of treatment for symptomatic HCM and typically incorporates agents such as βadrenergic and calcium channel antagonists to increase diastolic filling time, slow the heart rate, decrease contractility, and help normalize intracardiac filling pressures. For medically refractory patients with obstructive physiology, mechanical intervention to relieve outflow tract obstruction may be considered, including surgical septal myomectomy or catheterization-based alcohol septal ablation. Clinical indicators of increased risk for sudden death include: a history of cardiac arrest or sustained ventricular tachycardia; significant nonsustained ventricular tachycardia on ambulatory monitoring; recurrent syncope; a hypotensive response to exercise in patients younger than 50 yr; massive LVH (>30–35 mm); and a family history of SCD or identification of a malignant genotype. However, estimation of an individual’s risk is imprecise. If two or more risk factors are present, increased risk of SCD is present with an estimated annual mortality of 4–6%. Sudden death may be the

initial presenting symptom of HCM or occur in only mildly symptomatic patients. Although sudden death has been associated with vigorous physical activity and HCM is the leading cause of SCD in US competitive athletes, many episodes have occurred in the setting of only mild or no exercise. Implantation of a cardiovertordefibrillator is effective in decreasing the risk of sudden death from lethal arrhythmias in appropriate individuals. GENETIC ASPECTS The observation that HCM occurred in families defined it as a genetic cardiovascular disorder with Mendelian autosomal-dominant inheritance. Linkage analysis of large kindreds with HCM identified several disease loci on chromosomes 1, 11, 14, and 15. Positional cloning and candidate gene analysis identified discrete mutations in genes that encode for different elements of the contractile apparatus, including cardiac β-MHC, cardiac troponins T and I, cMyBPC, α-tropomyosin, actin, the essential and regulatory myosin light chains, and titin (Table 11-1). Thus, genetic studies established the paradigm of HCM as a disease of the sarcomere (Fig. 11-3). The sarcomere is the functional unit of myocyte contraction. Proteins are organized into a latticework of thick (myosin heavy and light chains) and thin (actin, the troponin complex, and α-tropomyosin) filaments that interdigitate during muscle fiber shortening and lengthening. The detachment and attachment of actin and the myosin head serve as the molecular motor of contraction and relaxation. The hydrolysis of ATP provides fuel, and changes in intracellular Ca2+ concentration coordinate thick and thin filament interaction. More than 300 individual mutations have been identified in 11 different components of the contractile apparatus. Types of mutations include missense mutations (resulting in amino acid substitution), nonsense mutations (resulting in premature termination of translation), short insertions and deletions, and alteration of splice donor or acceptor sites (resulting in altered transcripts). There is no significant founder effect in HCM as mutations tend to be “private”—unique from family to family with rare reappearances in unrelated kindreds. De novo or sporadic mutations are also well described. Mutations in cardiac β-MHC, cMyBPC, and cardiac troponins T and I account for approx 80–90% of described HCM cases. Significant diversity exists in the clinical expression of these inherited gene defects. Although a handful of mutations have been described as characteristically “benign” or “malignant,” heterogeneity of phenotype is the rule. Further identification of causal mutations and accurate description of genotype–phenotype correlations remain a work in progress, but the wide genetic and clinical spectrum of HCM make this challenging. As the ability to more precisely and sensitively assess phenotype improves, more accurate associations will emerge. Broad, recurring themes are outlined in Table 11-1. It remains unclear why some sarcomere mutations cause more severe disease than others and why individuals with the same mutation have a wide range of clinical features. Description of wider genetic and environmental factors that shape the expression of the underlying mutation is an active area of investigation. Cardiac β-MHC Mutations Mutations in the cardiac β-MHC gene on chromosome 14 are thought to account for approx 40% of cases of HCM. This protein is organized into two functional domains: an amino terminal globular head that interacts with actin and a carboxyl terminal rod. The force of the power stroke is transduced through a hinge region connecting the rod and head domains. Most HCM-causing mutations are of the missense

101

CHAPTER 11 / INHERITED CARDIOMYOPATHIES

Table 11-1 Mutations That Cause Hypertrophic Cardiomyopathy and General Phenotypic Correlations Gene HCM-sarcomere proteins

Inherited left ventricular hypertrophy

Designation

Chromosome

Frequency

Number of mutations

Phenotypic correlation

β-myosin heavy chain

β-MHC

14q1

~30–40%

>80

Cardiac myosin binding protein C

cMyBPC

11q1

~30–40%

>50

Cardiac troponin T

cTnT

1q3

approx 5%

>20

Cardiac troponin I α-tropomyosin Myosin essential light chain Myosin regulatory light chain Actin Titin γ-subunit AMP kinase Lysosome associated membrane protein 2 Muscle LIM protein

cTnI α-TM MLC-1

19p1 15q2 3p

5% 10 8 2

Skeletal myopathy

MLC-2

12q

Rare

8

Skeletal myopathy

PRKAG2

11q 2q3 7q3

Rare Rare ?

5 1 3

LAMP2

X

?

11p

?

CRP3

Typically obvious disease with significant LVH; several severe phenotypes (end-stage heart failure and sudden death) Typically more mild disease, but severe phenotypes have been described; elderly onset HCM Typically mild LVH but increased association with sudden death

Glucose metabolism; preexcitation and conduction disease; catecholeminergic polymorphic ventricular tachycardia

3

Adapted from J Mol Cell Cardiol 2001;33:655–670; Circulation 2001;104:2113–2116; Seidman JG, Seidman C. The genetic basis for cardiomyopathy: from mutation identification to mechanistic paradigms. Cell 2001;104;557–567; http://cardiogenomics.med.harvard.edu/ project-detail?project_id=230, with permission from Elsevier.

Figure 11-3 Mutations in genes that encode sarcomere proteins form the genetic basis of hypertrophic cardiomyopathy as well as some forms of dilated cardiomyopathy. (Reprinted with permission from N Engl J Med 2000; 343:1688–1696.)

variety and clustered within the globular head. The genetic basis of HCM was first described as missense mutation in this gene in which a single base pair is changed, resulting in the substitution of glutamine for arginine at residue 403. Over 80 different β-MHC missense mutations have been reported in familial and sporadic disease. The phenotypic expression of these mutations tends to be fairly obvious with significant degrees of LVH apparent by late adolescence. Although heterogeneous, the clinical course of certain MHC mutations has been severe, associated with

a striking risk of sudden death or development of end-stage heart failure. The precise determinants of the relationship between prognosis and underlying gene mutation are likely multifactorial and poorly understood. Mutations that result in a change in the charge of the substituted amino acid appear to result in more severe disease, presumably owing to more dramatic effects on protein structure and function. Cardiac Myosin Binding Protein C Missense, splice site, and deletion/insertion mutations in cMyBPC likely account for approx 30–40% of HCM cases. This large protein (1274 amino acids) is encoded on chromosome 11 and may function to provide structural support to the sarcomere and to modulate myosin ATPase activity in response to sympathetic stimulation. In a significant subset of individuals with cMyBPC mutations, the development of clinically apparent LVH is delayed until age 50 yr or older. Elderly onset HCM is associated with mutations in cMyBPC. Disease course is generally mild and not typically associated with attenuated survival, but there are reports of increased sudden death risk related to cMyBPC mutations. Cardiac Troponin T Troponin T links the troponin complex to α-tropomyosin and, therefore, plays a central role in regulating contraction. The cardiac-specific isoform is encoded on chromosome 1. Approximately 5% of HCM is thought to be attributable to cardiac troponin T (cTnT) mutations. Traditionally, the clinical phenotype has been characterized by modest degrees of LVH but an increased risk of sudden death, although “benign” cTnT mutations are also described. EXPERIMENTAL MODELS OF HCM Mutations in sarcomere proteins may alter actin–myosin crossbridge formation, intracellular calcium cycling, the energetics of force generation, or the

102

SECTION II / CARDIOLOGY

transmission of force. However, the heterogeneity of HCM in the human population has challenged the dissection of the precise molecular mechanisms that lead from inherited gene defect to clinical phenotype. To evaluate the consequences of specific mutations in a more precise and controlled manner, a variety of in vitro and in vivo experimental models of HCM have been developed. In vivo models are based on genetically modified animals in which β-MHC, cTnT, and cMyBPC mutations that cause human disease are introduced. The resultant animals develop a phenotype that recapitulates HCM phenotype, developing myocardial fibrosis, hypertrophy, and disarray in an age-dependent manner. Hemodynamic changes and abnormalities of diastolic function typically precede the development of LVH. Interrogation of these models remains a work in progress, but will allow further investigation into basic questions such as whether LVH is primary or secondary in response to either enhanced or impaired contractility; the influence of genetic background, modifier genes, the environment, and pharmacologic manipulation; and clarification of the early molecular events that lead to myocyte hypertrophy. NEW PARADIGMS OF INHERITED CARDIAC HYPERTROPHY HCM is caused by mutations in genes that encode elements of the contractile apparatus. Mutations have been described in nonsarcomere proteins that mimic the phenotype of HCM. Genetic studies of families and sporadic cases of unexplained LVH with conduction abnormalities (progressive atrioventricular block, atrial fibrillation, ventricular pre-excitation/ Wolff–Parkinson–White syndrome) have identified a novel disease entity caused by mutations either in the γ-2 regulatory subunit (PRKAG2) of adenosine monophosphate-activated protein kinase, or in the lyosome-associated membrane protein 2 (LAMP2) gene both enzymes involved with glucose metabolism. Ventricular pre-excitation typically occurs early in life and is often symptomatic. Progressive conduction disease occurs with increasing age such that permanent pacemaker implantation was necessitated in 30% of affected individuals and helped serve as a discriminating feature from HCM caused by sarcomere mutations. Severe clinical outcomes were noted in a subset of patients with PRKAG2 and LAMP2 mutations, including progression to endstage heart failure or transplantation and SCD. Inherited LVH caused by PRKAG2 and LAMP2 mutations is a disease entity distinct from HCM caused by sarcomere protein mutations. Despite superficial similarities, this distinction is illustrated by the different morphologic appearances of these glycogen storage cardiomyopathies as compared with HCM. LVH in mutations caused by LAMP2 mutations is typically striking and concentric. Histopathologically, PRKAG2 and LAMP2 mutations do not display the myocardial disarray characteristic of HCM, but rather a relatively minor amount of fibrosis and prominent nonmembrane bound vacuoles that, if appropriately handled and stained, demonstrate glycogen and amylopectin accumulation. These histologic features are more consistent with other disorders of glycogen storage, such as Danon or Pompe disease. Different molecular signaling pathways are involved and therefore the clinical approach to individuals glycogen storage cardiomyopathies should not be predicated on management tenets of HCM. Mutations in the muscle LIM protein (MLP) may also cause inherited LVH. This protein plays an essential role as a promoter of myogenesis and may also act as a cofactor in the regulation of muscle-specific gene expression in skeletal and cardiac muscle. Mice with a MLP gene knockout have been described to develop both a dilated and hypertrophic cardiac phenotype. Studies to analyze

for mutations in the CRP3 gene (encoding MLP) in humans with hypertrophic heart disease revealed three novel missense mutations in three unrelated families.

DILATED CARDIOMYOPATHY CLINICAL ASPECTS Heart failure is an important public health problem affecting 5 million patients in the United States. It is responsible for 1 million hospitalizations and 300,000 deaths annually. Pathologic remodeling of the heart resulting in increased chamber volume is termed DCM. Cardiac mass is increased because of enlargement of the chambers with only modest increase in ventricular wall thickness. Histopathologic changes may be relatively subtle with only minor myocyte hypertrophy, degeneration, and interstitial fibrosis. DCM is a common cause of heart failure with a worldwide prevalence of about 36.5/100,000. Clinical manifestations may be somewhat protean, particularly in the early stages of disease. Characteristic symptoms include exertional dyspnea, fatigue, orthopnea, and lower extremity edema. Diagnosis is generally based on identifying increased cardiac dimensions and decreased contractile function. Exclusion of disorders that may cause cardiac dilation and dysfunction (coronary artery disease, alcohol abuse, thyroid disease, viral myocarditis, and infiltrative disorders such as hemochromatosis) generally leads to the diagnosis of “idiopathic” DCM; approx 25–30% are thought to have a genetic cause. The genetic basis of DCM is less specifically described than that of HCM. The clinical syndrome of DCM is also less distinctive than HCM, further challenging precise characterization. There is significant variability in the clinical features of DCM because of single gene mutations. Age of onset ranges from early childhood to late adulthood, although most patients present during the fourth or fifth decades of life. Families may also manifest additional phenotypes that cosegregate with disease, including additional cardiac involvement (mitral valve prolapse and conduction system disease) and extracardiac conditions (sensorineuronal hearing loss and muscular dystrophies). These additional manifestations may be expressed in advance of DCM and therefore serve as supplemental measures to assign phenotypic status. GENETIC ASPECTS The most common mode of inheritance of familial DCM is autosomal-dominant, but infrequent autosomalrecessive, X-linked, and mitochondrial transmission have been described (Table 11-2). Autosomal-dominant inheritance has been defined for a number of chromosomal loci and several specific genetic mutations have been characterized. Dominant mutations causing DCM may be associated with additional phenotypes as described above. Mutations that result in altered force generation and force transmission have been reported as a cause of autosomaldominant DCM. Missense mutations in β-MHC, α-tropomyosin, and an in-frame deletion of a lysine residue in cTnT have been identified as causes of autosomal-dominant DCM. Surprisingly, these mutations occur in sarcomere protein genes traditionally associated with causing HCM. The mechanisms by which mutations in the same gene lead to either a hypertrophic or dilated phenotype remain unclear. One speculation is that mutations associated with HCM may involve areas of the protein that are directly involved with generating energy or initiating the power stroke, resulting in impaired force generation, whereas mutations associated with DCM may result in impaired force transmission with subsequent cardiac enlargement and dysfunction. Proper cardiac function also requires effective transmission of the force generated by the contractile apparatus to the extracellular

103

CHAPTER 11 / INHERITED CARDIOMYOPATHIES

Table 11-2 Mutations That Cause Dilated Cardiomyopathy

Sarcomere proteins

Intermediate filament proteins/structural proteins

Miscellaneous genes

Gene

Inheritance

Chromosomal location

Associated phenotypes

β-myosin heavy chain Cardiac troponin T α-tropomyosin Actin δ-sarcoglycan Desmin Desmoplakin Plakoglobin

Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-recessive Autosomal-recessive

14q1 1q3 15q22 11q 5q33-34 2q35 6p24 17q21

Dystrophin Lamin A/C Tafazzin

X-linked Autosomal-dominant X-linked

Xp21 1p1-q21 Xq28

None None None None None Skeletal myopathy Wooly hair and keratoderma Arrhythmogenic right ventricular dysplasia, wooly hair, and keratoderma (Naxos syndrome) Muscular dystrophy Conduction disease Short stature and neutropenia (Barth syndrome)

Adapted from Schonberger J, Seidman CE. Many roads lead to a broken heart: the genetics of dilated cardiomyopathy. Am J Hum Genet 2001:69:249–260, with permission of The University of Chicago Press.

matrix. Filamentous proteins linking the sarcomere to the sarcolemma may assist in the propagation of force. Mutations in the portion of actin that participates in actin–cytoskeletal (rather than actin–myosin) interactions and mutations in the giant structural molecule, titin, have been implicated in causing isolated DCM. Intermediate-filament proteins, including desmin, connect actin to the dystrophin–sarcoglycan complex beneath the plasma membrane of all muscle cells. Mutations in desmin, dystrophin, and δ-sarcoglycan cause skeletal myopathies, some associated with myocardial involvement, as well as DCM with subclinical or no skeletal muscle involvement. The basis for the cardiac specificity of these mutations is unclear. Rare syndromes with autosomal-recessive inheritance have been attributed to mutations in cellular adhesion molecules plakoglobin (associated with arrhythmogenic right ventricular dysplasia with palmoplantar keratosis and wooly hair—Naxos syndrome) and desmoplakin (similar phenotype with more prominent LV involvement). These molecules play critical roles at the junctions of desmosomes and adherens and therefore may participate in cell-to-cell force propagation. Sporadic or maternally inherited mutations in mitochondrial DNA can result in complex, multisystem phenotypes, including cardiac involvement with DCM. Apart from disruption of the production or transmission of force, a separate paradigm of familial DCM has been suggested by the identification of a missense mutation in phospholamban as a cause of autosomal-dominant DCM. Phospholamban is a transmembrane sarcoplasmic phosphoprotein that regulates the Ca2+ ATPase pump, SERCA2a. This mutation may result in constitutive inhibition of SERCA2a via trapping of the phospholamban regulatory protein, protein kinase A. Therefore, alterations in myocyte Ca2+ homeostasis may be a primary trigger for the development of DCM. The mechanisms by which other genetic mutations give rise to a DCM phenotype remain elusive. Mutations in tafazzin, an X chromosome-encoded acetyltransferase protein of unknown function, give rise to Barth syndrome, a triad of DCM, neutropenia, and 3-methylglutaconicaciduria. Dominant mutations in the nuclear envelope proteins, lamin, cause progressive conduction system disease with eventual development of DCM. Of note, distinct lamin A/C mutations can give rise to autosomal-dominant Emery–Dreyfus mus-

cular dystrophy, limb-girdle muscular dystrophy, and to familial partial lipodystrophy. Each of these clinical phenotypes may be associated with cardiac disease, predominantly DCM.

CONCLUSION Unraveling the molecular and genetic basis of inherited cardiomyopathies will provide further insight into their fundamental pathophysiology, and to the mechanisms underlying more common forms of acquired dilated and hypertrophic heart disease. Screening of family members in DCM kindreds may identify individuals with asymptomatic LV dysfunction or dilatation. Early initiation of medical therapy at this stage improves symptoms, morbidity, and prognosis. More definitive treatment to interrupt the natural history of dilated and hypertrophic cardiomyopathies will be drawn from better mechanistic understanding of how inherited gene defects remodel the heart.

SELECTED REFERENCES Arad M, Benson DW, Perez-Atayde AR, et al. Constitutively active AMP kinase mutations cause glycogen storage disease mimicking hypertrophic cardiomyopathy. J Clin Invest 2002;109:357–362. Blanchard E, Seidman C, Seidman JG, LeWinter M, Maughan D. Altered crossbridge kinetics in the alphaMHC403/+ mouse model of familial hypertrophic cardiomyopathy. Circ Res 1999;84:475–483. Bonne G, Carrier L, Richard P, Hainque B, Schwartz K. Familial hypertrophic cardiomyopathy: from mutations to functional defects. Circ Res 1998;83:580–593. Braunwald E, Morrow AG, Cornell WP, Aygen MM, Hilbish TF. Idiopathic hypertrophic subaortic stenosis—clinical, hemodynamic and angiographic manifestations. Am J Med 1960;29:924–945. Carrier L, Hengstenberg C, Beckmann JS, et al. Mapping of a novel gene for familial hypertrophic cardiomyopathy to chromosome 11. Nat Genet 1993;4:311–313. Charron P, Carrier L, Dubourg O, et al. Penetrance of familial hypertrophic cardiomyopathy. Genet Couns 1997;8:107–114. Charron P, Dubourg O, Desnos M, et al. Clinical features and prognostic implications of familial hypertrophic cardiomyopathy related to the cardiac myosin-binding protein C gene. Circulation 1998;97: 2230–2236. Epstein ND, Cohn GM, Cyran F, Fananapazir L. Differences in clinical expression of hypertrophic cardiomyopathy associated with two distinct mutations in the beta-myosin heavy chain gene. A 908Leu–Val mutation and a 403Arg–Gln mutation. Circulation 1992;86:345–352.

104

SECTION II / CARDIOLOGY

Geier C, Perrot A, Ozcelik C, et al. Mutations in the human muscle LIM protein gene in families with hypertrophic cardiomyopathy. Circulation 2003;107:1390–1395. Geisterfer-Lowrance AA, Christe M, Conner DA, et al. A mouse model of familial hypertrophic cardiomyopathy. Science 1996;272:731–734. Geisterfer-Lowrance AA, Kass S, Tanigawa G, et al. A molecular basis for familial hypertrophic cardiomyopathy: A beta cardiac myosin heavy chain gene missense mutation. Cell 1990;62:999–1006. Ho CY, Sweitzer NK, McDonough B, et al. Assessment of diastolic function with Doppler tissue imaging to predict genotype in preclinical hypertrophic cardiomyopathy. Circulation 2002;105:2992–2997. Kamisago M, Sharma SD, DePalma SR, et al. Mutations in sarcomere protein genes as a cause of dilated cardiomyopathy. N Engl J Med 2000;343:1688–1696. Klues HG, Schiffers A, Maron BJ. Phenotypic spectrum and patterns of left ventricular hypertrophy in hypertrophic cardimoypathy: morphologic observations and significance as assessed by two-dimensional echocardiography in 600 patients. J Am Coll Cardiol 1995; 26:1699–1708. Marian AJ. Pathogenesis of diverse clinical and pathological phenotypes in hypertrophic cardiomyopathy. Lancet 2000;355:58–60. Marian AJ, Roberts R. The molecular genetic basis for hypertrophic cardiomyopathy. J Mol Cell Cardiol 2001;33:655–670. Maron BJ. Hypertrophic cardiomyopathy. Lancet 1997;350:127–133. Maron BJ, Gardin JM, Flack JM, Gidding SS, Kurosaki TT, Bild DE. Prevalence of hypertrophic cardiomyopathy in a general population of young adults—echocardiographic analysis of 4111 subjects in the CARDIA study. Circulation 1995;92:785–789. Moolman JC, Corfield VA, Posen B, et al. Sudden death due to troponin T mutations. J Am Coll Cardiol 1997;29:549–555. Nagueh SF, Bachinski LL, Meyer D, et al. Tissue Doppler imaging consistently detects myocardial abnormalities in patients with hypertrophic cardiomyopathy and provides a novel means for an early diagnosis before and independently of hypertrophy. Circulation 2001;104:128–130. Niimura H, Bachinski LL, Sangwatanaroj S, et al. Mutations in the gene for cardiac myosin-binding protein C and late-onset familial hypertrophic cardiomyopathy (see comments). N Engl J Med 1998;338: 1248–1257. Niimura H, Patton KK, McKenna WJ, et al. Sarcomere protein gene mutations in hypertrophic cardiomyopathy of the elderly. Circulation 2002;105:446–451.

Schmitt JP, Kamisago M, Asahi M, et al. Dilated cardiomyopathy and heart failure caused by a mutation in phospholamban. Science 2003; 299:1410–1413. Schonberger J, Seidman CE. Many roads lead to a broken heart: the genetics of dilated cardiomyopathy. Am J Hum Genet 2001;69: 249–260. Seidman CE, Seidman JG. Hypertrophic cardiomyopathy. In: Scriver CR, Beaudet AL, Valle D, et al, eds. The Metabolic and Molecular Bases of Inherited Disease, 8th ed., vol. 4. New York: McGraw-Hill, 2000; pp. 5433–5452. Seidman JG, Seidman C. The genetic basis for cardiomyopathy: from mutation identification to mechanistic paradigms. Cell 2001;104: 557–567. Spirito P, Seidman CE, McKenna WJ, Maron BJ. The management of hypertrophic cardiomyopathy. N Engl J Med 1997;336:775–785. Thierfelder L, MacRae C, Watkins H, et al. A familial hypertrophic cardiomyopathy locus maps to chromosome 15q2. Proc Natl Acad Sci USA 1993;90:6270–6274. Tyska MJ, Hayes E, Giewat M, Seidman CE, Seidman JG, Warshaw DM. Single-molecule mechanics of R403Q cardiac myosin isolated from the mouse model of familial hypertrophic cardiomyopathy. Circ Res 2000;86:737–744. Watkins H. Genetic clues to disease pathways in hypertrophic and dilated cardiomyopathies. Circulation 2003;107:1344–1346. Watkins H, Anan R, Coviello DA, Spirito P, Seidman JG, Seidman CE. A de novo mutation in alpha-tropomyosin that causes hypertrophic cardiomyopathy. Circulation 1995;91:2302–2305. Watkins H, Conner D, Thierfelder L, et al. Mutations in the cardiac myosin binding protein-C gene on chromosome 11 cause familial hypertrophic cardiomyopathy. Nat Genet 1995;11:434–437. Watkins H, MacRae C, Thierfelder L, et al. A disease locus for familial hypertrophic cardiomyopathy maps to chromosome 1q3. Nat Genet 1993;3:333–337. Watkins H, McKenna WJ, Thierfelder L, et al. Mutations in the genes for cardiac troponin T and alpha-tropomyosin in hypertrophic cardiomyopathy. N Engl J Med 1995;332:1058–1064. Watkins H, Rosenzweig A, Hwang DS, et al. Characteristics and prognostic implications of myosin missense mutations in familial hypertrophic cardiomyopathy. N Engl J Med 1992;326:1108–1114. Wigle ED, Sasson Z, Henderson MA, et al. Hypertrophic cardiomyopathy: The importance of the site and the extent of hypertrophy, a review. Prog Cardiovasc Dis 1985;28:1–83.

12 Heart Failure Emerging Concepts in Excitation–Contraction Coupling and β-Adrenoceptor Coupling CLIVE J. LEWIS, FEDERICA AND ROGER J. HAJJAR

MONTE, SIAN E. HARDING,

DEL

SUMMARY

EXCITATION–CONTRACTION COUPLING

The process that begins contraction in the heart is known as excitation–contraction (E–C) coupling because it couples electrical signals on the membrane of the cardiac cell to activation of the myofilament and cross-bridge cycling. The cardiac action potential is produced by the coordinated interaction of many ion channels, which transduce physiological signals within and between cardiomyocytes. These cardiomyocytes are further regulated by a number of receptors that control the strength of the contraction on a beat-to-beat basis and their morphology in a chronic fashion. In heart failure, a number of steps in E–C coupling become abnormal. In this chapter, we will examine the role of these abnormalities in the development of heart failure.

On a beat-to-beat basis, depolarization of the cell membrane in cardiac myocytes leads to the opening of voltage gated L-type Ca2+ channels resulting in the influx of transsarcolemmal influx of Ca2+ into the cell (Fig. 12-1). These channels are in close proximity to the ryanodine receptors (RyRs), which are Ca2+ release channels located on the sarcoplasmic reticulum (SR). Ca2+ entering the cells through a single L-type Ca2+ channel induces the opening of one or a cluster of RyRs resulting in the local release of Ca2+ from the SR. During membrane depolarization, a large number of L-type Ca2+ channels are opened, resulting in a large release of Ca2+ from the RyRs, raising cytosolic Ca2+ from 0.1–0.2 to 2–10 µM. A functional Ca2+ signaling and releasing unit has been characterized, namely the Ca2+ spark. Ca2+ sparks were first identified as the “elementary events” of spontaneous increases in intracellular [Ca2+], which were detected by laser scanning confocal microscopy and the fluorescent Ca2+ indicator fluo-3. Ca2+ sparks also can be produced by the triggering Ca2+ entered through the L-type Ca2+ channel during the E–C coupling process. Functionally, the Ca2+ sparks represent Ca2+ releases from the SR through the opening of the SR Ca2+ release channels/RyRs. Ca2+ sparks are produced by activation of 10–100 RyRs based on the ratio of the Ca2+ sparks’ current and the single RyR channel current and because the morphology of Ca2+ sparks varies even within a single cell. Ca2+ sparks are depicted/measured by their morphology and the frequency of occurrence. The morphology information includes the sizes of Ca2+ sparks (amplitude, width, and duration), and the kinetics is described by the spark rising and decaying dynamics. The activity of individual RyRs and the number of the RyRs recruited during a spark play an important role in the spark morphology. Thus, the more active the Ca2+ release channels, the more Ca2+ would be released and consequently, increased size and more frequent occurrence of Ca2+ sparks would be observed. Therefore, direct modifications that change the RyRs activity or the modulators that regulate RyR activity will affect the size and frequency of Ca2+ sparks. SR Ca2+ content also regulates Ca2+ sparks because luminal (Ca2+) plays a critical role in regulating RyR by increasing

Key Words: β-receptor kinase; β-receptors; calcium; contractility; excitation contraction coupling; G proteins; L-type calcium channel; myofilaments; phospholamban; sarcoplasmic reticulum; sodium calcium exchanger.

INTRODUCTION The understanding of cardiac excitation contraction coupling and β-adrenoreceptor signaling continues to evolve. Defects in the steps of excitation contraction coupling and β-adrenergic signaling have been identified in human and experimental models of heart failure. Abnormalities in ionic channels, transporters, kinases, and various signaling pathways collectively contribute to the “failing phenotype.” β-adrenoceptors are widely expressed in human tissues and activated by neuronally released and circulating catecholamines. They are important mediators of the sympatho-adrenal axis regulating numerous physiological events, including relaxation of vascular smooth muscle, cardiac inotropy, chronotropy, and lusitropy. The traditional view of signal transduction is changing because of the emergence of concepts in the complexity of guanine nucleotide protein coupled receptor (GPCR)-effector coupling, an intracellular coupling.

From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

105

106

SECTION II / CARDIOLOGY

Figure 12-1 Depolarization of the membrane by the action potential leads to the opening of voltage gated L-type Ca2+ channels allowing the entry of a small amount of Ca2+ into the cell. Through a coupling mechanism between the L-type Ca2+ channel and the SR release channels (RyRs), a larger amount of Ca2+ is released activating the myofilaments leading to contraction. During relaxation, Ca2+ is reaccumulated back into the SR by the SERCA2a and extruded extracellularly by the sarcolemmal Na+/Ca2+ exchanger.

the activity of the RyRs and by sensitizing the threshold of RyR for activation to the stimulus. The signaling between L-type Ca2+ channel and RyR Ca2+ sparks can be triggered by the Ca2+ influx through the L-type Ca2+ sparks during voltage pulses. Thus, the kinetics, fidelity, and stoichiometry of coupling between L-type Ca2+ channels and RyR play a critical role in determination of the signal transduction during the E–C coupling process. The distance of the cleft between the cell plasma membrane and the SR membrane, roughly 12 nm in normal cardiomyocytes, is critical for such coupling and signal transduction. Extension of the distance between them by pulling the plasma membrane under tight-seal condition decreases/abolishes the signaling reliability. Relaxation occurs when calcium detaches from troponin C and is reaccumulated back into the SR by the cardiac isoform of the Ca2+ adenosine triphosphatase (ATPase) pump (SERCA2a) and extruded extracellularly by the sarcolemmal Na+/Ca2+ exchanger. The contribution of each of these mechanisms for lowering cytosolic Ca2+ varies among species. In humans approx 75% of the Ca2+ is removed by SERCA2a and approx 25% by the Na+/Ca2+ exchanger. SERCA2a transports Ca2+ back to the luminal space of the SR against a Ca2+ gradient by an energy dependent mechanism (one molecule of ATP is hydrolyzed for the transport of two molecules of Ca2+), where it binds to a calcium buffering protein, calsequestrin. The Ca2+ pumping activity of SERCA2a is regulated by phospholamban. In its unphosphorylated state, phospholamban inhibits the Ca2+ ATPase, whereas phosphorylation of phospholamban by cAMP-dependent protein kinase and by Ca2+-calmodulin dependent protein kinase reverses this inhibition.

CALCIUM HANDLING IN HEART FAILURE The myopathic heart exhibits abnormalities in the systolic and diastolic phases. Changes in diastolic function often appear earlier than systolic dysfunction. Compensated hypertrophy phenotypically demonstrates impaired relaxation parameters in the presence of normal or increased systolic function. The first report of abnormalities in calcium handling was made when calcium transients, recorded with the calcium indicator aequorin from trabeculae from myopathic human hearts removed at the time of cardiac transplantation, revealed a significantly prolonged calcium transient with an elevated end-diastolic intracellular calcium. This was corroborated by recordings in single isolated cardiomyocytes loaded with the fluorescent indicator Fura-2 from myopathic hearts. The calcium transients were characterized as having elevated diastolic calcium levels, decreased systolic Ca2+, and a prolonged relaxation phase. Studies both in muscle strips and isolated cardiomyoctes found that systolic calcium concentration was decreased in the failing state whereas diastolic calcium concentrations were elevated. These differences were more accentuated at higher stimulation rates. Because intracellular calcium transients and calcium concentrations are regulated by the voltage-dependent Ca2+ channel, the SR ryanodine Ca2+ release channels, the SERCA2a/phospholamban complex, and the Na/Ca exchanger, several studies have focused on the relative changes in these proteins between myopathic and normal hearts. Conflicting findings have been reported when calcium channels were measured with radioligand binding assays. One group reported an approx 20–30% reduction in mRNA and Bmax for dihydropyridine binding. However, two other groups found no difference in the

CHAPTER 12 / HEART FAILURE

107

Figure 12-2 There are four β-adrenoceptor subtypes that modulate human myocardial function. β3-adrenoceptor mediates cardiodepression mediated by the Gi protein. “Putative” β4-adrenoceptor causes cardiostimulant effects at concentrations 1000 times greater than those required to antagonize the cardiostimulant effects of catecholamines at β1- and β2-adreonceptors with high affinity.

number of calcium channels as measured with radioligand binding. Despite these conflicting reports it is appreciated that radioligandbinding assays can detect nascent channels and are only a surrogate marker of the number of active calcium channels. Altered myocyte calcium handling might play an important role in the development of heart failure in humans. This hypothesis is consistent with alterations found in some of the key proteins involved in the uptake and release of calcium in myopathic hearts. Calcium is released from the SR through a calcium sensitive release channel, the RyR. RyR mRNA is reduced in both myopathic human hearts and animal models. Inconsistent reports on RyR protein levels in myopathic human myocardium have appeared in the literature. Some research has shown altered gating mechanisms and altered responses to ryanodine in myopathic human myocardium, whereas other studies have found normal basic properties. Thus, although gene expression of the RyR receptor might be unchanged, its gating behavior, i.e., activity, might be impaired in failing myocardium. Even though the number of receptors might not be changed in heart failure and hypertrophy, there is a defective coupling between the L-type calcium channel and the RyR. The molecular correlates of this uncoupling seem to be a hyperphosphorylation of the RyR.

HUMAN MYOCARDIAL FUNCTION IS MODULATED THROUGH FOUR β-ADRENOCEPTORS β1- and β2-adrenoceptors coexist in human and rat myocytes with the β1-adrenoceptor the predominant subtype, even in heart failure when the relative proportion of β2-adrenoceptors increases (Fig. 12-2). The proportion of β2-adrenoceptors is greatest in humans of all species studied. β1- and β2-adrenoceptors are pharmacologically and genetically distinct entities with only 71% amino acid identity in the transmembrane-spanning domains and 54% identity overall; however, they appear to have the same mechanism of signal transduction through Gs and cAMP. Furthermore, epinephrine and norepinephrine are able to enhance contraction, relaxation, arrhythmia, and protein kinase A (PKA)dependent phosphorylation of downstream regulatory proteins to

a similar extent through human β1- and β2-adrenoceptors, respectively, despite the difference in receptor number. To explain this, tighter coupling of the β2-adrenoceptor to the Gs/adenylate cyclase (AC) pathway was proposed. Evidence suggests that β1- and β2-adrenoceptors have distinct physiological actions, coupled to distinct signal transduction pathways, and might occupy different spatial localization within the heart or within single cells (compartmentalization). In addition, β2-adrenoceptors might exist in two active forms that couple in parallel to Gs and G inhibitory (Gi), respectively. In addition to β1- and β2-adrenoceptors, evidence is consistent with the involvement of four β-adrenoceptor subtypes in the modulation of human myocardial function. A third β-adrenoceptor has been cloned and found to mediate vasodilatation, adipocyte metabolism, and cardiodepression mediated by the Gi protein. Pharmacological data also support existence of an atypical or novel β-adrenoceptor subtype, which mediates cardiostimulant effects including inotropy, chronotropy,and lusitropy and led to the proposal of the fourth cardiac β-adrenoceptor. “Putative” β4adrenoceptor pharmacology has been defined using nonconventional partial agonist β-adrenoceptor antagonists that cause cardiostimulant effects at concentrations 1000 times greater than those required to antagonize the cardiostimulant effects of catecholamines at β1- and β2-adreonceptors with high affinity. These agents include clinically used β-adrenoceptor antagonists such as pindolol, alprenolol, oxprenolol, and bucindolol as well as the useful experimental agent CGP 12177A. This receptor phenotype differs from the “classic” described function of known GPCRs and might be explained by novel receptor states or conformations or receptor dimerization. β1- AND β2-ADRENOCEPTOR COUPLING The coupling of β1- and β2-adrenoceptors to G proteins mediates the beneficial effects of catecholamines and is also involved in the pathophysiology of the heart, thus consideration of the emerging molecular mechanisms that can alter coupling of these subtypes is important. There has been much interest in single-nucleotide polymorphisms (SNPs) of GPCRs because they are thought to affect the responses

108

SECTION II / CARDIOLOGY

to endogenous agonists and/or drugs and therefore alter the pathophysiology of disease states by altering G protein coupling and signaling. The β1-adrenoceptor gene is polymorphic with 18 SNPs of which seven lead to amino acid substitutions within the coding exon. Two loci have been studied closely: Glycine49Serine (A for G at nucleotide 145) and Glycine389Arginine (C for G at nucleotide position 1165). Expression studies have demonstrated that despite similar affinities for both agonists and antagonists, the 49Serine receptor variant was relatively resistant to agonist-promoted downregulation compared to the 49Glycine variant. The Glycine389Arginine polymorphism, however, in the cytoplasmic tail Gs coupling domain, affects coupling to AC. In expression studies, the 389Arginine variant demonstrated and approx 30-fold greater ability to couple to AC when activated by isoproterenol and approx twofold for the β4-adreonceptor agonist CGP 12177A. Several studies have investigated whether any clinical correlate of this pharmacogenetic effect can be demonstrated, but the lack of agreement means that no conclusions can be drawn. The 49Glycine variant does not appear more frequently in patients with heart failure but is associated with a greater mortality. In dilated cardiomyopathy patients, the 389Arginine variant was more frequent in those without ventricular tachycardia. Neither polymorphism appears associated with long QT syndrome or acute coronary syndrome but higher blood pressures were observed in subjects with at least one 389Arginine allele and the 389Arginine genotype in combination with a deletion mutant of the α2C-adrenoceptor significantly increases the risk of heart failure in African Americans. In addition to the β1-adrenoceptor, the β2-adrenoceptor has several SNPs, the most interesting of which is Isoleucine164. This amino acid change results in decreased β2-adrenoceptor activity via decreased G protein/AC coupling. For the Isoleucine164 variant there is a fivefold increased relative risk of mortality in patients with heart failure. Thus SNPs that affect G protein coupling and signaling might have important implications in understanding cardiac physiology, disease, and pharmacological therapy. NON-CAMP DEPENDENT β1-ADRENORECEPTOR COUPLING (DIRECT L-TYPE CA2+ CHANNEL COUPLING) E–C coupling of the β1-adrenoceptor via Gs/cAMP is well established, but the possibility of a membrane delimited, non-cAMP dependent mechanism exists involving activated Gαs directly interacting with the sarcolemmal L-type Ca2+ channel. Evidence for a noncAMP dependent activation of Ca2+ channel comes from five lines of experiments: 1. Gαs reactivates Ca2+ channels that have run down in excised guinea-pig ventricular patches. 2. Gαs stimulates L-type Ca2+ channel activity after reconstitution in lipid bilayers. 3. Dihydropyridine and radio-labeled Gαs binding suggest that G proteins bind and activate purified Ca2+ channels. 4. Isoproterenol stimulated atrial myocytes demonstrate biphasic Ca2+ currents with the fast response (~5% total) resistant to phosphodiesterase inhibition by isobutyl methyl xanthine and Gαs activation by forskolin. 5. Incomplete blockade of β-AR responses by inhibitors of PKA. However, there are some inconsistencies and the overall relevance of these findings under physiological conditions has been questioned.

Supportive evidence exists for a non-cAMP dependent activation of Ca2+ channel mechanism under physiological conditions in isolated, paced single rat ventricular cardiomyocytes overexpressing human sequence β1-adrenoceptors. Increased basal, unstimulated myocyte contraction resulting from constitutive activity of overexpressed β1-adrenoceptors (ability of a receptor to signal through a second messenger pathway and activation of downstream effectors in the absence of agonist) has been observed. The raised basal contraction could be returned to control levels by CGP 20712A acting as an inverse agonist. The constitutive activity appeared non-cAMP dependent because cAMP levels were not increased by β1-overexpression and neither RpcAMPS (cAMP antagonist) nor carbachol (inhibits cAMP-mediated contraction) had a significant effect on contraction. However, the L-type Ca2+ channel current was significantly enhanced by β1-overexpression supporting a direct membrane-delimited activation by Gαs of the constitutively active β1-adrenoceptor. Controversy remains over constitutive activity of the β1-adrenoceptor. cAMP-dependent constitutive activity was observed in atria from transgenic mice and cell lines overexpressing β1-adrenoceptors. However, no change was found in basal contraction amplitude or basal cAMP in myocytes from β1/β2 double knockout (KO) mice overexpressing β1-adrenoceptors. Constitutive activity of the β2-adrenoceptor has also been observed. In transgenic mice with approx 200-fold overexpression of the β2-adrenoceptor, cardiac contractility and AC activity are raised to levels observed with maximal agonist stimulation in control. The constitutive activity could be reduced by the inverse agonist activity of ICI 188,551. GPCRs are proposed to exist in equilibrium between two conformational states, an inactive form (R) and an active form (R*) that can interact with G proteins. Agonist binds to and stabilizes the R form resulting in a shift to the R* form that involves a conformational change allowing G protein activation by guanosine triphosphate binding. In the β-adrenoceptor overexpressing myocytes, the levels of R + R* are increased and sufficient to produce increased basal contraction (constitutive activity). Inverse agonism is thought to result from the antagonist binding to the inactive R form and shifting the equilibrium away from R* with a corresponding decrease in contraction and AC activity. DIFFERENCES BETWEEN β1- AND β2-ADRENOCEPTOR COUPLING Differences between β1- and β2-adrenoceptors emerged by studying chronic heart-specific overexpression of each subtype in transgenic mice. β1-overexpressing mice have early mortality owing to the development of myocyte hypertrophy and myocardial fibrosis leading to heart failure despite unaltered cardiac function and basal AC activity. Proapoptotic effects mediated by activation of β1-adrenoceptor signaling pathways might mediate this observation. Transgenic mice overexpressing β2adrenoceptors at a similar level did not show increased mortality or heart failure, although the explanation for this was unclear. As opposed to the selective coupling of the β2-adrenoceptor to Gs/AC in humans, a non-cAMP E–C mechanism has been suggested for mouse and rat heart. It was based on the observations that isoproterenol produced increases in contraction through β2-adrenoceptors without changes in the speed of relaxation or phospholamban phosphorylation and that agents reducing cAMP-mediated effects did not affect contraction. The coupling of β2-adrenoceptors to Gi was observed by the use of pertussis toxin (PTX) when cAMPdependent E–C coupling was revealed. Interestingly, Gi activation

CHAPTER 12 / HEART FAILURE

Figure 12-3 Description of stimulus-trafficking in which different effects can be mediated by the same receptor acting though different G proteins, previously described for several GPCRs including β2-adrenoceptors.

through β2-adrenoceptors could oppose the proapoptotic effect of β1-adrenoceptors through Gs. Mitogen-activated protein and phosphatidylinositol 3′-kinase pathways were identified as downstream mediators of β2-/Gi signaling. The protection from early mortality owing to heart failure in β2-overexpressing mice as opposed to β1-overexpressing mice was linked to spontaneous upregulation of Gi (decreasing constitutive activity and hyperresponsiveness to catecholamines) and opposition of proapoptotic pathways. Subsequently β2-adrenoceptors were observed to couple to both Gs and Gi resulting in activation of opposing signaling pathways. Coupling of human β2-adrenoceptors to Gi in failing ventricle has been confirmed. ICI 118,551 (selective β2-adrenoceptor antagonist), at inverse agonist concentrations, produced a profound reduction of contraction in isolated ventricular myocytes from human heart, which could be prevented by treatment with PTX, indicating the involvement of Gi. The use of specific agonists/antagonists demonstrated Gi coupling to β2- rather than β1or β3-adrenoceptors. PTX-sensitive inverse agonist effects of ICI 118,551 have been observed in rat and rabbit by overexpressing human β2-adrenoceptors or Giα2. In failing human heart, β2-adrenoceptors do not demonstrate constitutive activity and there is no “spare” β-adrenoceptor capacity, yet Gi is upregulated and there are dampened β2-adrenoceptor responses. Thus a “stimulus-trafficking” modification has been proposed to the original theory of inverse agonism (Fig. 12-3). Stimulus trafficking describes the mechanism in which different effects can be mediated by the same receptor acting through different G-proteins; it has previously been described for several GPCRs including β2-adrenoceptors. Binding of the inverse agonist is not to the inactive R form but to another active Gi-coupled form (R#), which occurs in parallel with the usual R*-Gs coupling. In nonfailing heart of mouse, rat or humans, total levels of β2-adrenoceptor (R* + R#) and Gi levels are moderate with no effect on basal contraction. In failing heart, when Gi is upregulated, the Gi-R# form will shift the equilibrium away from R* resulting in the dampened contractile response to catecholamines observed in failing human heart. Inverse agonists bind to R# but because of the increased amount of Gi-R# in heart failure, a direct negative

109

inotropic effect of β-adrenoceptor antagonists is observed mediated by the β2-adrenoceptor. In explaining the findings above, ICI 118,551 is an agonist through the Gi-R# but not the Gs-R* pathway. However, the downstream Gi linked targets for mediating the negative inotropic effect of ICI 118,551 have not been identified. This negatively inotropic effect of ICI 118,551 has been observed for other β-adrenoceptor antagonists and is clearly important in human heart failure where they are used therapeutically to reduce mortality. The initial decrease in cardiac output during titration of β-adrenoceptor antagonists may, in addition to the removal of sympathetic tone, be owing to their direct negative inotropic effects. COMPARTMENTALIZATION OF GPCR SIGNALING A large body of functional data supports the existence of subcellular compartmentalization of the cAMP/AC/PKA signaling pathway in heart. There appears to be a fixed and specific spatial relationship between components of cAMP generation, response mechanisms and degradation. Thus within a cardiac myocyte, not all cAMP gains access to all PKA and its downstream regulators, and only a limited amount of phosphodiesterases might degrade cAMP. Molecular mechanisms contributing to compartmentalization include localization of receptors, G proteins, and AC in caveolae within the sarcolemma; localization of PKA by A-kinase anchoring proteins resulting in scaffolding and inhibition of activity; localization of activated PKC isoforms, their downstream target substrates, and phosphodiesterase isoforms in caveolae; and compartmentalization of cAMP generation and phosphodiesterase degradation in sarcolemmal microdomains. Evidence has highlighted compartmentalization of cAMP signaling in cardiac myocytes. In neonatal rat cardiac myocytes, specificity of signaling is guaranteed by tight localization of signaling events to the surface membrane. β-adrenoceptor stimulation generates multiple microdomains with increased concentrations of cAMP, but free diffusion of the second messenger is limited by the activity of phosphodiesterases and cannot transmit to nonsarcolemmal proteins. The possibility of compartmentalized cAMP signaling has also been investigated for the β2-adrenoceptor in adult rat ventricular myocytes where contraction appeared dissociated from cAMP. L-type Ca2+ current and contraction to β2- as well as β1-adrenoceptor agonists was entirely abolished by the use of PKA inhibitors, suggesting functional compartmentalization of cAMPdependent signaling. Direct G protein activation of the L-type Ca2+ channel has also been suggested for β2-adrenoceptor. However, a compartmentalized cAMP/PKA pathway mediating β2-adrenoceptor signaling to the L-type Ca2+ channel seems more likely. In hippocampal neurons, the β2-adrenoceptor appears to form a macromolecular signaling complex with the L-type Ca2+ channel, a G protein, AC, cAMP-dependent PKA and the counterbalancing phosphatase PP2A. This signaling complex ensures specific and rapid signaling. Moreover, β2-adrenoceptors appear compartmentalized spatially, residing exclusively in caveolae isolated from quiescent rat cardiomyocytes allowing regulation of cAMP signaling in microdomains. Further specificity of signaling might result from the selective coupling to different isoforms of AC. Nine mammalian isoforms of AC have been identified, of which types 5 and 6 are the predominant isoforms expressed in heart and are inhibited by multiple intracellular signaling pathways including Gi, PKA, PKC, and Ca2+. In myocytes overexpressing AC6, a selective enhancement

110

SECTION II / CARDIOLOGY

of β-adrenoceptor stimulated cAMP formation was observed with colocalization of β-adrenoceptor and AC in caveolar microdomains allowing rapid and specific signal transduction in cardiac myocytes. Whether selective AC isoform coupling occurs with different β-adrenoceptor subtypes is under investigation.

HOMO- AND HETERODIMERIZATION OF β1AND β1-ADRENOCEPTORS The classic GPCR model predicts that the stoichiometry of receptor, G protein and effector is 1/1/1; however, several studies have demonstrated that GPCRs exist as dimeric or oligomeric complexes. In addition to homodimers, heterodimers between members of the GPCR family exist. Homodimerization of both β1adrenoceptors and β2-adrenoceptors has been observed in vitro and in vivo, and intermolecular interactions between receptors might have functional and structural implications for G proteinmediated signaling. The sixth transmembrane domain of the β2adrenoceptor appears to be involved in homodimerization because a peptide derived from this domain inhibits both dimerization and β-adrenoceptor Gs/AC coupling, indicating a functional role for dimerization. Agonist stimulation was also found to stabilize the homodimer, whereas inverse agonists favored the monomeric state, further suggesting that conversion between monomer and dimer forms might play a physiological role. The coexistence of β1- and β2-adrenoceptor subtypes within a single myocyte raises the possibility of heterodimerization between these two receptors giving rise to potential novel pharmacology. Heterodimeriation of β1- and β2-adrenoceptors has been observed in cell lines and mouse cardiomyocytes cooverexpressing β1- and β2-adrenoceptors. β1- and β2-adrenoceptor colocalization has been demonstrated on the cell membrane surface together with novel functional properties. There was approx 1.5 log unit leftward shift in the concentration-response curve to isoproterenol of myocyte contraction when β1- and β2-adrenoceptors were coexpressed in myocytes from double β1-/β2-adrenoceptor KO mice. In addition, decreased affinity of known ligands was observed (Kd of CGP 20712A decreased approx 10-fold and the Kd of ICI 118,551 decreased approx 70-fold compared with each subtype expressed alone), suggesting that β1- and β2-adrenoceptors can form highaffinity heterodimers with a novel binding site and altered ligand binding properties. However, similar efficacy and potency of isoproterenol-stimulated cAMP production has been demonstrated when β1- and β2-adrenoceptors were expressed in cell lines alone or together when they were observed to form heterodimers. However, a unique functional property of β1-/β2-heterodimers was observed, with complete loss of the isoproterenol stimulation of ERK 1/2 mitogen-activated protein kinase activation, which is usually seen when β2- (but not β1-) adrenoceptors are expressed alone. Thus homo- and heterodimerization of β1- and β2-adrenoceptors alters G protein coupling and downstream signaling, although the functional basis of these interactions remains to be explored. β3-ADRENOCEPTOR COUPLING IN HUMAN HEART The functional role and coupling mechanism for cardiac β3adrenoceptors remains unclear. β3-adrenoceptors are pharmacologically characterized by four criteria: 1. Low affinity of classic β-adrenoceptor antagonists such as propranolol and nadolol. 2. Activation by selective β3-adrenoceptor agonists BRL 37344, SR 58611A, and CL 316243.

3. Activation by nonconventional partial agonists CGP 12177A, cyanopindolol, and pindolol. 4. Blockade by selective β3-adrenoceptors antagonists SR 59230A. Activation of recombinant β3-adrenoceptors increases adenylyl cyclase activity by signaling via Gs. β3-adrenoceptor mRNA has been reported in human atrium and ventricle in addition to gastrointestinal and adipose tissue, and protein expression has been demonstrated in human heart. Nonetheless, little evidence suggests a cardiostimulant role for the β3-adrenoceptor because cardiostimulant effects of selective β3adrenoceptor agonists have not been observed in human ventricle or atrium, or rat or ferret heart. β3-adrenoceptor selective agonists BRL 37344 and SR 58611 increase L-type Ca2+ current 1.7- and 2.2-fold, respectively in human atrial myocytes by a nadololresistant mechanism and increased atrial inotropic and chronotropic response, but these effects are most likely mediated by activation of classic β1- and β2-adrenoceptors. However, β3-adrenoceptors also appear to couple to Gi and produce cardiodepression and action potential shortening in human ventricle with nanomolar potency. In addition, high concentrations of endogenous catecholamines were shown to elicit a nadolol-resistant cardiodepression. The cardiodepressant effect of β3-adrenoceptor agonists was attenuated by pretreatment by PTX, suggesting mediation by coupling to Gi. The observation that the cardiodepressant effects were reduced by nitric oxide (NO)-antagonists led to the suggestion of a Gi/NO-dependent cyclic guanosine monophosphate signaling pathway mediating β3-adrenoceptor negative inotropic effects in human ventricle. However, although NOmediated β3-adrenoceptor stimulation is present in human atrium, at least in nonfailing heart, it does not directly affect atrial contraction. β3-adrenoceptors are more resistant to desensitization than classic β1- and β2-adrenoceptors, and β3-adrenoceptor expression appears increased in failing human ventricle. This suggests that in heart failure, the relative importance of β3-adrenoceptors might be increased with impairment of contractility mediated by activation of β3-adrenoceptors by high circulating concentrations of norepinephrine and coupling to a Gi/cyclic guanosine monophosphate/ NO signaling pathway. However, the observation of cardiodepression by β3-adrenoceptor selective agonists has not been confirmed in isolated human cardiac preparations by others. β3-adrenoceptor selective agonists failed to cause cardiodepression in human atrium, and in failing ventricle in both whole trabeculae and isolated myocytes. Furthermore, the cardiodepressant effect of BRL 37344 was observed with nanomolar potency, whereas this compound has been observed to have micromolar potency at recombinant and adipocyte β3-adrenoceptors. As described, negative inotropic effects of β-adrenoceptor antagonists have been demonstrated in failing human ventricle mediated primarily by a Gi-dependent interaction with the β2-adrenoceptor. BRL 37344 has a similar affinity for the β2- and β3-adrenoceptor and produces positive inotropic effects in atrium mediated by β1- and β2-adrenoceptors. Thus in failing ventricle, the cardiodepressant effects of BRL 37344 could be mediated by β2-adrenoceptor coupling to Gi. Clearly, the implication of β3-adrenoceptor mediated cardiodepression is important, particularly in the context of heart failure, and further elucidation is required to address this controversy. THE “PUTATIVE” β4-ADRENOCEPTOR The β4-adrenoceptor is of particular interest, because of its potential importance

CHAPTER 12 / HEART FAILURE

in cardiac physiology, its implication in arrhythmogenesis and as a target for clinically used β-adrenoceptor antagonists. This apparent novel receptor subtype is an alternate state or conformation of the β1-adrenoceptor in addition to the “classic” catecholamine activation site. Species homologues of the “putative” β4-adrenoceptor are present in all mammals studied and are expressed in sinoatrial node, atrium, and ventricle. In common with the activation of “classic” β1-adrenoceptors, β4-adrenoceptor agonists produce positive inotropy, chronotropy, and lusitropy and effect signal transduction in human and rat heart through a Gs/AC pathway increasing cAMP and PKA. β4-adrenoceptor’s effects are relatively resistant to those of propranolol yet are antagonized with moderate potency by bupranolol, carvedilol, and a range of β-adrenoceptor antagonists but with approx 100 times lower affinity than at the “classic” β1-adrenoceptor. Functional human β4-adrenoceptor responses have also been found in vivo. Pindolol, in the presence of β1- and β2-adrenoceptor blockade by propranolol, induces a 10% increase in heart rate in healthy volunteers, which can be prevented by predosing with carvedilol. CGP 12177A is more potent than norepinephrine at shortening the ventricular monophasic action potential but does not alter the refractory period, suggesting differential coupling to individual ion channels. CGP 12177A also increases both intracellular Ca2+ transient and current through the L-type Ca2+ channel, although less than via “classic” β1-adrenoceptors. In part, the continuing interest in the β4-adreonceptor’s pharmacology is owing to its potential involvement in arrhythmogenesis. CGP 12177A produces ventricular extrasystoles in whole ferret ventricle, arrhythmic Ca2+ transients in atrial and ventricular rat myocytes and arrhythmic Ca2+ transients in mouse ventricular myocytes with a potency 40 times greater than through the “classic” β1-adrenoceptor. The pharmacology of the β4-adrenoceptor partly resembles that of the β3-adrenoceptor. However, in rat, mouse and human heart, selective β3-adrenoceptor agonists do not produce cardiostimulant effects and neither selective β3-adrenoceptor antagonists nor mice with targeted disruption of the β3-adrenoceptor gene (β3 KO mice) alter the effect of CGP 12177A. A novel low-affinity state or conformation of the β1-adrenoceptor is responsible for the “putative” β4-adrenoceptor pharmacology. Thus, the β1-adrenoceptor might exist in at least two different states, a “high-affinity” state (“classic” β1-adrenoceptor catecholamine site with high affinity to propranolol) and another “low-affinity” state stabilized by CGP 12177A, responsible for β4adrenoceptor pharmacology and with low affinity to propranolol. CGP 12177A stimulates adenylyl cyclase in cell lines expressing recombinant rat and human β1-adrenoceptors and its cardiostimulant effects were significantly more resistant to β-adrenoceptor antagonism than catecholamines were. This, therefore, suggests that the β1-adrenoceptor fulfilled some of the pharmacological criteria for the β4-adrenoceptor. Adenoviral overexpression of β1-adrenoceptors in isolated, cultured adult rat ventricular myocytes has been accomplished and a similar increase in the cardiostimulant potency of both isoproterenol and CGP 12177A has been demonstrated, indicating involvement of a propranololresistant state of the β1-adrenoceptor. This concept is supported by the demonstration of agonist specific differences in antagonist affinity revealing two different affinity states of the β1-adrenoceptor in recombinant β1-adrenoceptors in cell lines and in ferret heart

111

preparations. Moreover, an obligatory role for β1-adrenoceptors in the cardiostimulant effect of CGP 12177A was demonstrated in the β1-/β2-adrenoceptor double KO mice. CGP 12177A increased sinoatrial rate and left atrial contractile force in heart from wild-type and β2-adrenoceptor KO mice but were absent in β1-/β2-adrenoceptor double KO mouse heart. β4-adrenoceptor’s involvement in heart failure has been suggested by the parallel desensitization and resensitization of β1- and β4-adrenoceptors in a rat model of heart failure. The resensitization induced by PTX suggests that in cardiac failure, both β1- and β4-adrenoceptors can couple to Gi as well as Gs because Giα2 mRNA is increased in this model of heart failure. A similar desensitization in cardiostimulant responses (maximal effect and potency of CGP 12177A) through both states of the β1-adrenoceptor was seen in right atrium of failing human heart. In addition, a parallel decrease in both the high (catecholamine site) and lowaffinity (β4-adrenoceptor site) [3H]-CGP 12177A binding sites was observed in right atrium from failing compared with nonfailing human myocardium. The ability of the low-affinity site of β1adrenoceptors to couple to Gi appears to be restricted to heart failure because PTX had no effect on the CGP 12177A response in nonfailing heart in mouse and rat. In addition to a low-affinity state of the β1-adrenoceptors mediating the cardiostimulant effects of CGP 12177A, reports suggest that part of the stimulant effect is through β2-adrenoceptors. If novel states of receptors are common throughout the GPCR super family, then CGP 12177A might be activating a novel site on the β2-adrenoceptor with low affinity for propranolol and ICI 118,551, similar to the novel state of the β1-adrenoceptor. The concept that the β2-adrenoceptor can exist in more than one active conformation has been suggested previously. A propranolol and ICI 118.551 (selective β2-adrenoceptor antagonist) resistant enhancement of the positive inotropic effect of CGP 12177A has been demonstrated in isolated rat ventricular myocytes adenovirally overexpressing β2-adrenoceptors. This is time-dependent, with an early (24 h) enhancement of the positive effect lost by 48 h, paralleling changes in β2-adrenoceptor stimulation of contraction. The attenuation of the effect with time can be restored by pretreatment with PTX suggesting that the CGP 12177A cardiostimulant site, like the β2-adrenoceptor, can couple through Gi as well as Gs. Mediation of the cardiostimulant effect of CGP 12177A via β2-adrenoceptors is consistent with the weak partial agonist activity of CGP 12177A (intrinsic activity only 10–20% of that at expressed β1-adrenoceptors) on cAMP accumulation in cell lines overexpressing recombinant β2-adrenoceptors. Alternatively, the possibility that the cardiostimulant effects of CGP 12177A are mediated by an interaction with β1- and β2adrenoceptor homo- or heterodimers, as shown for the effect of isoproterenol, has also been suggested. In support of this possibility, a further approx 10-fold enhanced cardiostimulant effect of CGP 12177A has been observed in isolated rat ventricular myocytes adenovirally coexpressing β1- and β2-adrenoceptors compared with myocytes expressing each subtype alone. This suggests that heterodimerization of β1- and β2-adrenoceptors creates a novel pharmacological profile to CGP 12177A in addition to isoproterenol. Because native β1-adrenoceptors are present in rat myocytes, the effect of β2-adrenoceptor overexpression to enhance effects of CGP12177A might also be mediated by heterodimers. Under normal circumstances, different membrane localization of β1-adrenoceptors might prevent receptor dimerization by spatial

112

SECTION II / CARDIOLOGY

separation from the caveolar domains where β2-adrenoceptors are found, but these spatial arrangements could be disrupted after adenovirally mediated overexpression.

CLINICAL IMPLICATIONS In a meta-analysis of β-adrenoceptor antagonists given after myocardial infarction, those with high intrinsic sympathomimetic activity (ISA) appeared to have worse effects on mortality and a lower level of risk-reduction than those with low ISA. Xamoterol (β1-selective antagonist) with ISA increased mortality in heart failure, and bucindolol (nonselective β-adrenoceptor antagonist) had no overall benefit on mortality and a worse outcome in severe heart failure. Interestingly, bucindolol has ISA at the low-affinity state of the β1-adrenoceptor in humans, suggesting that the study outcome was owing to interaction with this novel site. Carvedilol improves mortality in heart failure and has antagonist effects at both the classic and low-affinity site of the β1-adrenoceptor, suggesting that part of its beneficial effect in heart failure and its superiority over metoprolol may be resulting from an interaction at the low-affinity state of the β1-adrenoceptor. Alternatively, activation of protective Gi-mediated pathways by an action of carvedilol through β2/Gi coupling could contribute to recovery.

CONCLUSION Cardiac β1- and β1-adrenoceptors differ in G protein coupling, cAMP signaling, target protein phosphorylation, and alteration of cardiac E–C coupling. The dual coupling of β2-adrenoceptors to both Gi and Gs explains many of these differences. Additionally, functional localization and spatial compartmentalization of β2adrenoceptor cAMP/PKA signaling to membrane microdomains further contributes to specificity of signal. The stimulus-trafficking model of parallel pathway β2-adrenoceptor Gi and Gs coupling might explain the dampened β-adrenoceptor response in failing human heart and might explain the negative inotropic effects of clinically used β-adrenoceptor antagonists. The concept of novel “low-affinity” sites or state of GPCRs, elucidated for the “putative” β4-adrenoceptor, requires further investigation, especially because drugs in clinical use interact at the novel site of the β1adrenoceptor. In addition, the observation of functional homo- and heterodimers of β- and β2-adrenoceptors indicates the complexity of β-adrenoceptor signaling and the need to study the impact of β1- and β2-adrenoceptors’ differences and interactions both in their physiological and pathophysiological roles. Studies are needed to define the molecular, structural and spatial mechanisms underlying differential G protein coupling, downstream signaling and regulation of apoptosis of β2- vs β1-adrenoceptors.

SELECTED REFERENCES Angers S, Salahpour A, Bouvier M. Dimerization: an emerging concept for G protein-coupled receptor ontogeny and function. Annu Rev Pharmacol Toxicol 2002;42:409–435. Aprigliano O, Rybin V, Pak E, Robinson R, Steinberg S. Beta 1-and beta 2-adrenergic receptors exhibit differing susceptibility to muscarinic accentuated antagonism. Am J Physiol 1997;272:H2726–H2735. Arch S, Ainsworth A, Cawthorne M, et al. Atypical β-adrenoceptor on brown adipocytes as target for anti-obesity drugs. Nature 1984;309:163–165. Australia/New Zealand Heart Failure Collaborative Group. Randomised, placebo controlled trial of carvedilol in patents with congestive heat failure due to ischaemic heart disease. Lancet 1997;349:375–380. Baker J, Hill S. Pharmacological characterization of CGP 12177 at the human beta(2)-adrenoceptor. Br J Pharmacol 2002;137:400–408.

Bayewitch M, Avidor-Reiss T, Levy R, et al. Inhibition of adenylyl cyclase isoforms V and VI by various Gβγ subunits. FASEB J 1998;12:1019–1025. Begin-Heick. β3-adrenoceptor activation of adenylyl cyclase in mouse white adipocyes: modulation by GTP and effect of obesity. J Cell Biochem 1995;58:464–473. Bengtsson K, Melander O, Orho-melander M, et al. Polymorphism in the β1-adrenergic receptor gene and hypertension. Circulation 2001;104:187–190. Berkowitz D, Nardone N, Smiley R, et al. Distribution of β3-adrenoceptor mRNA in human tissues. Eur J Pharmacol 1995;289:223–228. Beta-Blocker Evaluation of Survival Trial Investigators. A trial of the beta-blocker bucindolol in patients with advanced chronic heart failure. N Engl J Med 2001;344:1659–1667. Bianchetti A, Manara L. In vitro inhibition of intestinal motility by phenylethanolaminotetralines: evidence of atypical β-adrenoceptors in rat colon. Br J Pharmacol 1990;100:831–839. Bloom J, Dutia M, Johnson B, et al. Disodium (R,R) -5-[2-[2-(3Chlorophenyl)-2-hydroxymethyl]-amino]pro-pyl]-1,3-benzodioxole2,2-dicarboxylate (CL 316,243). A potent β-adrenergic agonist virtually specific for β3-adrenoceptors: a promising antidiabetic and antiobesity agent. J Med Chem 1992;35:3081–3084. Bond R, Leff P, Johnson T, et al. Physiological effects of inverse agonists in transgenic mice with myocardial overexpression of the β2-adrenoceptor. Nature 1995;374:272–276. Borjesson M, Magnusson Y, Hjalmarson A, Andersson B. A novel polymorphism in the gene coding for the β1-adrenergic receptor associated with survival in patients with heart failure. Eur Heart J 2000;21: 1853–1858. Bundkirchen A, Brixius K, Bolck B, Schwinger R. Bucindolol exerts agonistic activity on the propranolol-insensitive state of β1-adrenoceptros in human myocardium J Pharmacol Exp Ther 2002;300:794–801. Chaudry A, MacKenzie R, Georgic L, Granneman J. Differential interaction of the beta (1)- and beta (3)-adrenergic receptors with G (i) in rat adipocytes. Cell Signal 1994;6:457–465. Chen-Izu Y, Xiao R, Izu L, et al. G(i)-dependent localization of beta(2)adrenergic receptor signaling to L-type Ca(2+) channels. Biophys J 2000;79:2547–2556. Chesley A, Lundberg M, Asai T, et al. The beta(2)-adrenergic receptor delivers an antiapoptotic signal to cardiac myocytes through G(i)dependent coupling to phosphatidylinositol 3-kinase. Circ Res 2000;87:1172–1179. Communal C, Singh K, Sawyer D, Colucci W. Opposing effects of beta(1)- and beta(2)-adrenergic receptors on cardiac mycoyte apoptosis: role of a pertussis toxin-sensitive G protein. Circulation 1999;100:2210–2212. Davare M, Avdonin V, Hall D, et al. A beta2 adrenergic receptor signaling complex assembled with the Ca2+ channel Cav1.2. Science 2001;293:98–101. Del Monte F, Kaumann A, Poole-Wilson P, Wynne D, Pepper J, Harding S. Coexistence of functioning beta 1- and beta 2-adrenoceptors in single myocytes from human ventricle. Circulation 1993;88:854–863. Devi L. Heterodimerization of G-protein-coupled receptors: pharmacology, signaling and trafficking. Trends Pharmacol Sci 2001;22: 532–537. Dixon R, Kobilka B, Strader D, et al. Cloning of the gene and cDNA for mammalian β-adrenergic receptor and homology with rhodopsin. Nature 1986;321:75–79. Emorine L, Marullo S, Briend-Sutren M, et al. Molecular characterization of the human β3-adrenergic receptor. Science 1989;245:1118–1121. Engel G, Hoyer D, Berthold R, Wagner H. (+)[125Iodo]cyanopindolol, a new ligand for β-adrenoceptors: identification and quantitation of subclasses of β-adrenoceptors in guinea pig. Naunyn Schmiedebergs Arch Pharmacol 1981;317:277–285. Engelhardt S, Grimmer Y, Fan G, Lohse M. Constitutive activity of the human β1-adrenergic receptor in β1-receptor transgenic mice. Mol Pharmacol 2001;60:712–717. Engelhardt S, Hein L, Wiesmann F, Lohse M. Progressive hypertrophy and heart failure in beta1-adrenergic receptor transgenic mice. Proc Natl Acad Sci USA 1999;96:7059–7064.

CHAPTER 12 / HEART FAILURE

Freestone N, Heubach J, Wettwer E, Ravens U, Brown D, Kaumann A. β4-Adrenoceptors are more effective than β1-adrenoceptors in mediating arrhythmic Ca2+ transients in mouse ventricular myocytes. Naunyn Schmiedebergs Arch Pharmacol 1999;360:445–456. Freestone N, Kaumann A. Activation of an atypical β-adrenoceptor increases calcium and can elicit arrhythmias in rat atrial cardiomyocytes. Pharmacologist 1997;39:74. Gao T, Ouri T, Gerhardstein B, et al. Identification and subcellular localization of the subunits of L-type calcium channels and adenylyl cyclase in cardiac myocytes. J Biol Chem 1997;272:19,401–19,407. Gauthier C, Langin D, Balligand J. β3-Adrenoceptors in the cardiovascular system. Trends Pharmacol Sci 2000;1:426–431. Gauthier C, Leblais V, Kobzig L, et al. The negative inotropic effect of β3adrenoceptor stimulation is mediated by activation of anitric oxide synthase in human ventricle. J Clin Invest 1998;102:1377–1384. Gauthier C, Tavernier G, Charpentier F, Langin D, Le Marec H. Functional β3-adrenoceptor in the human heart. J Clin Invest 1996;98: 556–562. Gilbert E, Hershberger R, Wiechmann R, Movsesian M, Bristow M. Pharmacologic and hemodynamic effects of combined β-agonist stimulation and phosphodiesterase inhibition in the failing human heart. Chest 1995;108:1524–1532. Gille R, Lemoine H, Ehle B, Kaumann A. The affinity of (–)-propranolol for β1- and β2-adrenoceptors of human heart. Differential antagonism of the positive inotropic effects and adenylate cyclase stimulation by (–)-noradrenaline and (–)-adrenaline. Naunyn Schmiedebergs Arch Pharmacol 1985;331:60–70. Gong H, Adamson D, Ranu H, et al. The effect of Gi-protein inactivation on basal, and beta(1)- and beta(2)AR-stimulated contraction of myocytes from transgenic mice overexpressing the beta(2)-adrenoceptor. Br J Pharmacol 2000;131:594–600. Granneman J. The putative β4-adrenergic receptor is a novel state of the β1-adrenergic receptor. Am J Physiol Endocrinol Metab 2001;43: E199–E202. Hall J, Kaumann A, Brown M. Selective β1-adrenoceptor blockade enhances positive inotropic effects of endogenous catecholamines through β2-adrenoceptors in human atrium. Circ Res 1990;66: 1610–1623. Hamilton S, Codina J, Hawkes M, et al. Evidence for direct interaction of Gs alpha with the Ca2+ channel of skeletal muscle. J Biol Chem 1991;266:19,528–19,535. Hanoune J, Pouille Y, Tzavara E, et al. Adenylyl cyclases: structure, regulation and function in an enzyme superfamily. Mol Cell Endocrinol 1997;128:179–194. Harding S. Lack of evidence for β3-adrenoceptor modulation of contractile force in human ventricular myocytes. Circulation 1997;95:1–53. Hartzell HC, Fischmeister R. Direct regulation of cardiac Ca2+ channels by G proteins: neither proven nor necessary? Trends Pharmacol Sci 1992;13:380–385. He J, Xu J, Castleberry A, Lau A, Hall R. Glycosylation of beta(1)-adrenergic receptors regulates receptor surface expression and dimerization. Biochem Biophys Res Commun 2002;297:565–572. Hebert T, Moffett S, Morello J, et al. A peptide derived from a beta2-adrenergic receptor transmembrane domain inhibits both receptor dimerization and activation. J Biol Chem 1996;271:16,384–16,392. Heubach J, Blaschke M, Harding S, Ravens U, Kaumann A. Cardiostimulant and cardiodepressant effects through overexpressed human β2-adrenoceptors in murine heart: regional differences and functional role of β1-adrenoceptors. Naunyn Schmiedebergs Arch Pharmacol 2003;367:380–390. Iwai C, Akita H, Shiga N, et al. Suppressive effect of the Gly389 allele of the β1-adrenergic receptor gene on the occurrence of ventricular tachycardia in dilated cardiomyopathy. Circ J 2002;66:723–728. Joseph S, Lynham J, Grace A, Colledge W, Kaumann A. The effects of (–)-isoprenaline but not of (–)-CGP12177 are markedly reduced at Gly-389 β1–adrenoceptors compared to Arg-389 β1–adrenoceptors. Br J Pharmacol 2002;135:336P. Kanki H, Yang P, Xie H, Kim R, George A, Roden D. Polymorphisms in β-adrenergic receptor genes in the acquired long QT syndrome. J Cardiovasc Electrophysiol 2002;13:252–256.

113

Kaumann A, Bartel S, Molenaar P, et al. Activation of β2-adrenergic receptors hastens relaxation and mediates phosphorylation of phospholamban, troponin I and C protein in ventricular myocardium form patients with terminal heart failure. Circulation 1999;99:65–72. Kaumann A, Blinks J. β-Adrenoceptor blocking agents as partial agonists in isolated heart muscle: dissociation of stimulation and blockade. Naunyn Schmiedebergs Arch Pharmacol 1980;311:205–218. Kaumann A, Engelhardt S, Molenaar P, Hein L, Molenaar P, Lohse M. (–)-CGP 12177-evoked cardiostimulation in double β1-/β2-adrenoceptor knockout mice. Obligatory role of β1-adrenoceptors for putative β4-adrenoceptor pharmacology. Naunyn Schmiedebergs Arch Pharmacol 2001;363:87–93. Kaumann A, Lynham J, Sarsero D, Molenaar P. The atypical cardiac cardiostimulant β-adrenoceptor is distinct from β3-adrenoceptor and is coupled to a cyclic AMP-dependent pathway in rat and human myocardium. Br J Pharmacol 1997;120:102P. Kaumann A, Lynham J. Stimulation of cyclic AMP-dependent protein kinase in rat atria by (–)-CGP 12177 through an atypical β-adrenoceptor. Br J Pharmacol 1997;120:1187–1189. Kaumann A, Molenaar P. Differences between the third cardiac β-adrenoceptor and the colonic β3-adrenoceptor in the rat. Br J Pharmacol 1996;118:2085–2098. Kaumann A, Molenaar P. Modulation of human cardiac function through 4 β-adrenoceptor populations. Naunyn Schmiedebergs Arch Pharmacol 1997;355:667–681. Kaumann A, Preitner F, Sarsero D, Molenaar P, Revelli J, Giacobino J. (–)-CGP 12177 causes cardiostimulation and binds to cardiac putative β4-adrenoceptors in both wild-type and β3-adrenoceptor knockout mice. Mol Pharmacol 1998;53:670–675. Kaumann A, Saunders L. Both β1- and β2-adrenoceptors mediate catecholamine-evoked arrhythmias in isolated human atrium. Naunyn Schmiedebergs Arch Pharmacol 1993;348:536–540. Kaumann A. Is there a third heart β-adrenoceptor? Trends Pharmacol Sci 1989;10:316–319. Kaumann A. Four β-adrenoceptor subtypes in mammalian heart. Trends Pharmacol Sci 1997;18:70–76. Kenakin T. Inverse, protean, and ligand-selective agonism: matters of receptor conformation. FASEB J 2001;15:598–611. Kenakin T. Recombinant roulette versus the apparent virtue of ‘natural’ cell receptor systems: receptor genotypes versus phenotypes. Trends Pharmacol Sci 2002;23(9):403, 404. Kompa A, Gu X, Evans B, Summers R. Desensitization of cardiac betaadrenoceptor signaling with heart failure produced by myocardial infarction in the rat. Evidence for the role of Gi but not Gs or phosphorylating proteins. J Mol Cell Cardiol 1999;31:1185–1201. Kompa A, Summers R. Desensitizataion and resensitization of β1- and putative β4-adrenoceptor mediated responses occur in parallel in a rat model of cardiac failure. Br J Pharmacol 1999;128:1399–1406. Konkar A, Zhai Y, Granneman J. β1-Adrenergic receptors mediate β3adrenergic-independent effects of CGP 12177 in brown adipose tissue. Mol Pharmacol 2000;57:252–258. Konkar A, Zhu Z, Granneman J. Aryloxypropanolamine and catecholamine ligand interactions with the β1-adrenergic receptor: evidence for interaction with distinct conformations of β1-adrenergic receptors. J Pharmacol Exp Ther 2000;294:923–932. Krief S, Lonnqvist F, Raimbault S, et al. Tissue distribution of β3-adrenoceptor mRNA in man. J Clin Invest 1993;91:344–349. Lavoie C, Mercier J, Salahpour A, et al. β1/β2-adrenergic receptor heterodimerization regulates β2-adrenergic receptor internalization and ERK signaling efficacy. J Biol Chem 2002;277:35,402–35,410. Lewis C, Gong H, Koch W, Brown M, Harding S. The beta-four adrenoceptor is a novel state of the beta-one adrenoceptor. J Am Coll Cardiol 2002;39(5):165A. Liggett S, Tepe N, Lorenz J, et al. Early and delayed consequences of beta(2)-adrenergic receptor overexpression in mouse hearts: critical role for expression level. Circulation 2001;101:1707–1714. Liggett S, Wagoner L, Craft L, Hornung R, Hoit B, McIntosh T. The Ile164 β2-adrenergic receptor polymorphism adversely affects the outcome of congestive heart failure. J Clin Invest 1998;102: 1534–1539.

114

SECTION II / CARDIOLOGY

Lowe M, Grace A, Vandenberg J, Kaumann A. Action potential shortening through the putative β4-adrenoceptor in ferret ventricle: comparison with β1- and β2-adrenoceptor-mediated effects. Br J Pharmacol 1998;124:1341–1344. Lowe M, Lynham J, Grace A, Kaumann A. Comparison of the affinity of beta-blockers for two states of the beta 1-adrenoceptor in ferret ventricular myocardium. Br J Pharmacol 2002;135:451–461. Lowe M, Lynham J, Grace A, Kaumann A. Comparison of the affinity of beta-blockers for two states of the beta 1-adrenoceptor in ferret ventricular myocardium. Br J Pharmacol 2002;135:451–461. Maack C, Cremers B, Flesch M, Hoper A, Sudkamp M, Bohm M. Different intrinsic activities of bucindolol, carvedilol and metoprolol in human failing myocardium. Br J Pharmacol 2000;130:1131–1139. Malinowska B, Schlicker E. Mediation of the positive chronotropic effect of CGP 12177 and cyanopindolol in the pithed rat by atypical betaadrenoceptors, different from beta 3-adrenoceptors. Br J Pharmacol 1996;117:943–949. Manara L, Badone D, Baroni M, et al. Functional identification of rat atypical β-adrenoerocetor by the first β3-selective antagonists, arlyoxypropanolaminetetratine. Br J Pharmacol 1996;117:1374–1376. Maqbool A, Hall A, Ball S, Balmforth A. Common polymorphisms of β1-adrenoceptor: identification and rapid screening assay. Lancet 1999;353:897. Mason D, Moore J, Green S, Liggett S. A gain-of-function polymorphism in a G-protein coupling domain of the human beta 1-adrenergic receptor. J Biol Chem 1999;274(19):12,670–12,674. Michel M, Insel P. Receptor gene polymorphisms: lessons on functional relevance from the β1–adrenoceptor. Br J Pharmacol 2003;138: 279–282. Milano C, Allen L, Rockman H, et al. Enhanced myocardial function in transgenic mice overexpressing the beta 2-adrenergic receptor. Science 1994;264:582–586. Mohell N, Dicker A. The β-adrenergic radioligand [3H] CGP-12177, generally classified as an antagonist, is a thermogenic agonist in brown adipose tissue. Biochem J 1989;261:401–405. Molenaar P, Sarsero D, Kaumann A. Proposal for the interaction of nonconventional partial agonists and catecholamines with the ‘putative β4adrenoceptor’ in mammalian heart. Clin Exp Pharmacol Physiol 1997; 24:647–656. Moniotte S, Kobzik L, Feron O, Trochu J, Guthier C, Balligand J. Upregulation of b3-adrenoceptors and altered contractile response to inotropic amines in human failing myocardium. Circulation 2000;103:1649–1655. Nantel F, Bonin H, Emorine L, et al. The human β3-adrenoceptor is resistant to short-term agonist-promoted desensitization. Mol Pharmacol 1993;43:548–555. Oostendorp J, Kaumann A. Pertussis toxin suppresses carbachol-evoked cardiodepression but does not modify cardiostimulation mediated through beta1- and putative beta4-adrenoceptors in mouse left atria: no evidence for beta2- and beta3-adrenoreceptor function. Naunyn Schmiedebergs Arch Pharmacol 2000;361:134–145. Ostrom R, Violin J, Coleman S, Insel P. Selective enhancement of β-adrenergic receptor signaling by overexpression of adenylyl cyclase type 6: colocalization of receptor and adenylyl cyclase in caveolae of cardiac myocytes. Mol Pharmacol 2000;57:1075–1079. Packer M, Bristow M, Cohn J, et al. The effect of carvedilol on morbidity and mortality in patients with chronic heart failure. U.S. Carvedilol Heart Failure Study Group. N Engl J Med 1996;334:1349–1355. Pak M, Fishman P. Anomalous behaviour of CGP 12177A on beta 1adrenergic receptors. J Recept Signal Transduct Res 1996; 16(1,2): 1–23. Pavoine C, Magne S, Sauvadet A, Pecker F. Evidence for a beta2-adrenergic/arachidonic acid pathway in ventricular cardiomyocytes. Regulation by the beta1- adrenergic/camp pathway. J Biol Chem 1999; 274:628–637. Podlowski S, Wenzel K, Luther H, et al. Beta1-adrenoceptor gene variations: a role in idiopathic dilated cardiomyopathy? J Mol Med 2000; 78:87–93. Poole-Wilson PA, Swedberg K, Cleland JGF, et al. Comparison of carvedilol and metoprolol on clinical outcomes in patients with

chronic heart failure in the carvedilol or metoprolol European trial (COMET): randomised controlled trial. Lancet 2003;362:9377–9384. Pott C, Brixius K, Bundkirchen A, et al. The preferential β3-adrenoceptor agonist BRL 37344 increases force via β1-/β2-adrenoceptors and induces endothelial nitric oxide synthase via β3-adrenoceptors in human atrial myocardium. Br J Pharmacol 2003;138:521–529. Rana B, Shina T, Insel P. Genetic variations and polymorphisms of G protein-coupled receptors: functional and therapeutic implications. Annu Rev Pharmacol Toxicol 2001;41:593–624. Rathz D, Brown K, Kramer L, Liggett S. Amino acid 49 polymorphisms of the human β1–adrenergic receptor affect agonist-promoted trafficking. J Cardiovasc Pharmacol 2002;39:155–160. Rybin V, Xu X, Lisanti M, Steinberg S. Differential targeting of beta adrenergic receptor subtypes and adenylyl cyclase to cardiomyocyte caveolae. A mechanism to functionally regulate the cAMP signaling pathway. J Biol Chem 2000;275:41,447–41,457. Rybin V, Xu X, Steinberg S. Activated protein kinase C isoforms target to cardiomyocyte caveolae: stimulation of local protein phosphorylation. Circ Res 1999;84:980–988. Samama P, Cotecchia S, Costa T, Lefkowitz R. A mutation-induced activated state of the b2-adrenergic receptor: extending the ternary complex model. J Biol Chem 1993;268:4625–4636. Samama P, Pei G, Costa T, Cotecchia S, Lefkowitz R. Negative antagonists promote an inactive conformation of the β2-adrenergic receptor. Mol Pharmacol 1994;45:390–394. Sarsero D, Molenaar P, Kaumann A, Freestone N. Putative β4Adrenoceptors in rat ventricle mediate increases in contractile force and cell Ca2+: comparison with atrial receptors and relationship to (–)[3H]-CGP 12177 binding. Br J Pharmacol 1999;128:1445–1460. Sarsero D, Molenaar P, Kaumann A. The “putative β4-adrenoceptor” mediates positive inotropic responses and hastens relaxation through a cAMP pathway in human heart. Aust N Z J Med 1998;28:147. Sarsero D, Russell F, Lynham J, et al. (–)-CGP 12177 increases contractile force and hastens relaxation of human myocardial preparations through a propranolol-resistant state of the β1-adrenoceptor. Naunyn Schmiedebergs Arch Pharmacol 2003;367:10–21. Sennitt M, Kaumann A, Molenaar P, et al. The contribution of classical (β1/2-) and atypical β-adrenoceptors to the stimulation of human white adipocyte lipolysis and right atrial appendage contraction by novel β3adrenoceptor agonists of differing selectivities. J Pharmacol Exp Ther 1998;285:1084–1095. Sheares K, Haydock S, Brown M. Evidence in vivo of a functional β4adrenoceptor in the human heart? Br J Clin Pharmacol 2000;49:496P. Shuba Y, Hesslinger B, Trautwein W, McDonald T, Pelzer D. Whole-cell calcium current in guinea-pig ventricular myocytes dialysed with guanine nucleotides. J Physiol 1990;424:205–228. Skeberdis V, Jurevicius J, Fischmeister R. β3-Adrenergic regulation of Ltype Ca2+ current in human atrial myocytes. Abstract to the 43rd Annual Meeting of the Biophysical Society February 13–17, 1999, Baltimore, MD. Small K, Wagoner L, Levin A, Kardia S, Liggett S. Synergistic polymorphsims of β1 and α2C –adrenergic receptors and the risk of congestive heart failure. N Engl J Med 2002;347:1135–1142. Steinberg S, Brunton L. Compartmentation of G protein-coupled signaling pathways in cardiac myocyte. Annu Rev Pharmacol Toxicol 2001;41:751–773. Summers R, Molenaar P. Autoradiography of β1 and β2-adrenoceptors. In: Kendall D, Hill S, eds. Methods in Molecular Biology 41: Signal Transduction Protocols. Totowa, NJ: Humana Press Inc., 1995; pp. 25–39. Sunahara R, Dessauer C, Gilman A. Complexity and diversity of mammalian adenylyl cyclases. Annu Rev Pharmacol Toxicol 1996;36: 461–480. Waagstein F, Caidahl K, Wallentin I, Bergh C, Hjalmarson A. Long-term beta-blockade in dilated cardiomyopathy. Effects of short and longterm metoprolol treatment followed by withdrawal and readministration of metoprolol. Circulation 1989;80:551–563. Walter M, Lemoine H, Kaumann A. Stimulant and blocking effects of optical isomers of pindolol on the sinoatrial node and trachea of guinea pig. Role of β-adrenoceptor subtypes in the dissociation between

CHAPTER 12 / HEART FAILURE

blockade and stimulation. Naunyn Schmiedebergs Arch Pharmacol 1984;327:159–175. Wenzel K, Felix S, Bauer D, et al. Novel variants in 3 kb of 5′UTR of the b1-adrenergic receptor gene (-93C4T, -210C4T, and -2146T4C): -21a6C homozygotes present in patients with idiopathic dilated cardiomyopathy and coronary heart disease. Hum Mutat 2000;16:534. Wheeldon N, Mcdevitt D, Lipworth B. Cardiac Effects of the β3-adrenoceptor agonist BRL 35135 in man. Br J Clin Pharmacol 1994;37:363–369. White H, Maqbool A, McMahon A, et al. An evaluation of the beta-1 adrenergic receptor Arg389Gly polymorphism in individuals at risk of coronary events. A WOSCOPS substudy. Eur Heart J 2002;23: 1087–1092. Wickman K, Clapham D. Ion channel regulation by G proteins. Physiol Rev. 1995;75:864–885. Xamoterol in Severe Heart Failure Study Group. Xamoterol in severe heart failure. Lancet 1990;336:1–6. Xiao R, Avdonin P, Zhou Y, et al. Coupling of beta2-adrenoceptor to Gi proteins and its physiological relevance in murine cardiac myocytes. Circ Res 1999;84:43–52. Xiao R, Cheng H, Zhou Y, Kuschel M, Lakatta E. Recent advances in cardiac β2-adrenergic signal transduction. Circ Res 1999;85:1092–1100. Xiao R, Hohl C, Altschuld R, et al. β2-Adrenergic receptor-stimulated increase in cAMP in rat heart cells is not coupled to change in Ca2+ dynamics, contractility or phospholamban phosphorylation. J Biol Chem 1994;269:19,151–19,156.

115

Xiao R, Lakatta E. β1-Adrenoceptor stimulation and β2-adrenoceptor stimulation differ in their effects on contraction, cyotsolic Ca2+ and Ca2+ current in single rat ventricular cells. Circ Res 1993;73:286–300. Xiao R. β-Adrenergic signalling in the heart: dual coupling of the β2adrenergic receptor to Gs and Gi proteins. Sci STKE 2001;104:RE15. Yatani A, Brown A. Rapid beta-adrenergic modulation of cardiac calcium channel currents by a fast G protein pathway. Science 1989; 245:71–74. Yatani A, Codina J, Imoto Y, Reeves J, Birnbaumer L, Brown A. A G protein directly regulates mammalian cardiac calcium channels. Science 1987;238:1288–1292. Yatani A, Imoto Y, Codina J, Hamilton S, Brown A, Birnbaumer L. The stimulatory G protein of adenylyl cyclase, Gs, also stimulated dihydropyridine-sensititive Ca2+ channels. J Biol Chem 1988;263:9887–9895. Zaccolo M, Pozzan T. Discrete microdomains with high concentration of cAMP in stimulated rat neonatal cardiac myocytes. Science 2002; 295(1):1711–1715. Zhou Y, Cheng H, Bogdanov K, et al. Localized cAMP-dependent signalling mediates beta 2-adrenergic modulation of cardiac excitationcontraction coupling. Am J Physiol 1997;273:H1611–H1618. Zhou Y, Yang D, Zhu W, et al. Spontaneous activation of beta(2)- but not beta(1)-adrenoceptors expressed in cardiac myocytes from beta(1)beta(2) double knockout mice. Mol Pharmacol 2000;58:887–894. Zhu W, Yang D, Zhang S, et al. Heterodimerization of β1- and β2-adrenergic receptors in cardiac myocytes. Circulation 2002;106(Supp II):1124.

13 Aortic Diseases SAUMYA DAS, JAMES L. JANUZZI, JR., AND ERIC M. ISSELBACHER SUMMARY Aneurysms are the most important disorder that affects the aorta. Aneurysms involving the abdominal aorta are typically associated with atherosclerosis, whereas those involving the thoracic aorta have many causes, including congenital abnormalities in the structure of the aortic wall. This chapter will discuss thoracic aortic aneurysms, such as Marfan syndrome, bicuspid aortic valve, and familial thoracic aortic aneurysm syndrome, and also abdominal aortic aneurysms. Key Words: Aneurysm; aorta; bicuspid aortic valve; familial thoracic aortic aneurysm syndrome; Marfan syndrome.

INTRODUCTION Although diseases of the aorta are significantly less common than those of the heart, because aortic disease can be life threatening, its diagnosis and treatment are clinically relevant. Aneurysms are the most important disorder that affect the aorta. Aneurysms involving the abdominal aorta are typically associated with atherosclerosis, whereas those involving the thoracic aorta have many causes, including congenital abnormalities in the structure of the aortic wall. Because aortic aneurysms typically do not produce symptoms until they are large, the diagnosis is often not made until a rupture or dissection occurs. However, mass population screening is not likely to be practical or cost effective. Therefore, the ability to identify genetic factors that could predict risk for aortic aneurysms would be a valuable guide screening and therapy in clinical practice.

THORACIC AORTIC ANEURYSMS MARFAN SYNDROME Marfan syndrome (MFS) is the best-studied aortic disease regarding genetic basis and molecular mechanisms. The search for the causes of MFS exemplifies a systematic approach used to identify candidate genes and gain insight into the disorder’s biology and pathophysiology. MFS is a systemic disorder of connective tissue with protean manifestations, with the skeletal, ocular, and cardiovascular systems most often affected. The hallmark of MFS is abnormality of the medial layer of the aortic wall, characterized by fragmentation and disorganization of the elastic fibers, a generalized loss of elastin content, and deposition of amorphous matrix components. The mechanical properties of the aorta are primarily a function of the From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

elastic fibers within the media, so loss or abnormalities of elastin could weaken the tensile strength of the aortic wall. Indeed, the most threatening consequence of MFS is dilatation of the aortic root and the ascending aorta, which, if untreated, can result in potentially fatal aortic dissection or rupture. The syndrome is relatively uncommon, with a prevalence of one in 5000–10,000. Traditionally, the diagnosis has relied on a series of clinical criteria as outlined in the Ghent nosology, reflecting the pleiotropic manifestation of the disease: the phenotypes of the affected individuals form a continuum from the severe neonatal MFS to milder forms known by the acronym MASS (myopia, mitral valve prolapse, aortic dilatation without dissection, and skin and skeletal muscle abnormalities). The Ghent nosology was designed to account for varying presentations by defining highly specific major criteria with less specific minor criteria. A positive diagnosis requires a combination of major criteria in at least two organ systems and minor criteria in a third organ system. The clustering of MFS in families is well recognized and classic genetic analysis has shown it to be autosomal-dominant with complete penetrance (but with pleiotropic manifestations). In exploring for the genetic cause of MFS, researchers first excluded certain candidate genes such as elastin. Then, in one affected family, positional cloning indicated that the mutation was localized to chromosome 15q21. Cloning and sequencing of the DNA in that region led to the identification of the fibrillin (FBN)-1 gene. Subsequently, FBN1 mutations were linked to the majority of the cases of MFS. In addition, a distinct gene on chromosome 3p24 was linked to another family with MFS. The identification of the FBN1 prompted investigation into the role of FBN1, the protein product of the FBN1, in the development and homeostasis of the aortic wall. Studies undertaken to elucidate the basic structure of the fibrillin gene progressed on two fronts. Some investigators mapped the various mutations associated with MFS, attempting to correlate genotype and phenotype. Others examined the biology of fibrillin (and they are made up of microfibrils), trying to understand how specific abnormalities in protein structure may alter biology at the cellular, organ, and systemic level. Of the 337 mutations in the FBN1 reported in patients with MFS, 69 have been found in more than one unrelated individual, eight in three or more unrelated individuals, and only three mutations have been found in four of more unrelated individuals. Unfortunately, the difficulty expressing fibrillin and constituting microfibrils in vitro has hampered the correlation of structure and function. Consequently, only a few of these mutations have been tested experimentally.

116

CHAPTER 13 / AORTIC DISEASES

The majority of FBN1 mutations in MFS are missense mutations that lead to substitution of one of six highly conserved cysteines. Cysteines are important in the formation of intramolecular disulfide bonds critical to protein folding, so substitution of any of these amino acids could disrupt fibrillin protein structure and, consequently, have deleterious effects on the global structure of microfibrils. Some of the other missense mutations are in the conserved calcium-binding motif. The binding of calcium by this motif may serve to rigidify the protein and perhaps even stabilize FBN1 against proteolytic degradation. Nevertheless, it remains unclear whether the primary consequence of the alteration in the structure of the fibrillin protein is an increased sensitivity to proteolysis, an altered protein–protein interaction, individual microfibril disarray, or overall disruption of the architecture of polymeric microfibrillin arrays owing to distortion in the individual monomeric fibrillin modules. Most of the identified mutations do not seem to segregate with particular phenotypes. Indeed, even in classic MFS, there is a striking intrafamily variability. For example, family members sharing the same mutation can have clinical syndromes that vary widely in terms of the age of onset, the organ systems involved, and the severity of disease. Nevertheless, a minority of mutations do produce more consistent phenotypes. For example, 20% of reported mutations are either nonsense mutations or have a frameshift resulting in a premature stop codon. Messenger RNAs containing these transcripts display a reduced concentration owing to a phenomenon called nonsense-mediated decay, and most of these manifest as a mild clinical phenotype. Indeed, none of these mutations has been reported to cause neonatal MFS, the most serious manifestation of the disease. Conversely, patients with the phenotype of neonatal MFS—who are diagnosed at birth and die primarily of heart failure from severe tricuspid and mitral regurgitation rather than aortic dissection—have mutations that cluster in exons 24–32. Another clinical phenotype, atypically severe MFS, which presents with early-onset cardiovascular complications with aortic dissection or need for aortic surgery before age 16, is also associated with mutations in exons 24–32, although these mutations are typically distinct from mutations found in neonatal and classic MFS. Nonetheless meaningful conclusions about genotype–phenotype correlations have been difficult to draw. Without further detailed knowledge of the three-dimensional structure of fibrillin, its protein partners, and the biology of microfibrils, it would be hard to correlate mutations with clinical phenotypes. Furthermore, given that most mutations in FBN1 (apart from those seen in classic MFS) have been identified in only one affected individual or family, it would be hard to generalize genotype–phenotype correlations. Finally, MFS is likely only a small percentage of fibrillinopathies, because FBN1 mutations have been reported in Marfan-related aortic syndromes and disorders of skeletal system such as WeillMarchesani syndrome. Further progress awaits characterization of these disorders and more knowledge of the role of microfibrils. The histological hallmark of MFS is the disorganization and fragmentation of elastic fibers and a decrease in elastin content in the medial layer of the aorta. No comprehensive theory has emerged that explains how mutations in the FBN1 lead to this pathological hallmark. However, several major hypotheses have emerged. The first proposed mechanism is the dominant-negative model of pathogenesis. Microfibrils consist of polymers of the FBN1 protein. According to the dominant-negative model, mutant FBN1 proteins interfere with the assembly of wild-type FBN1—produced by the

117

normal allele in heterozygotes—into microfibrils. This model is attractive from a theoretical standpoint because large polymers are especially sensitive to defects in monomeric components; in the case of fibrillin, mutations in the cysteine residues might affect not only intramolecular disulfide bonds but also intermolecular bonds that might be essential to the assembly of microfibrils. Premature truncation codon mutations are associated with lower levels of mutant transcripts leading to decreased numbers of mutant fibrillin monomers. If the dominant-negative model were true, fewer mutant monomers should produce less interference with assembly of the remaining wild-type monomers, and hence a milder phenotype. Indeed, in one study the patient with the lowest proportion of mutant transcript (6%) had a clinically mild MASS phenotype, whereas another patient with higher levels of mutant transcript (16%) had the more severe classic MFS phenotype. However, this model may not alone be sufficient, as in one report patients with 2 and 7% mutant transcript levels both had severe manifestation of MFS. Studies using cultured fibroblasts from MFS patients also seem to support the dominant-negative hypothesis. Using pulse-chase experiments with radioactive sulfur to label the cysteine-rich fibrillin proteins, the synthesis, secretion and aggregation of fibrillin molecules were studied. Fibroblast cell lines derived from patients with MFS (with a variety of mutations) showed only 35% expected fibrillin deposition (based on normal controls) in the extracellular matrix (ECM). Interestingly, three nonsense mutations with mutant transcripts proportions of 15–25% demonstrated ECM deposition rates of 7–25% of normal, whereas one nonsense mutation with a lower mutant transcript level of 6% showed ECM deposition of 54% normal, supporting a dominant-negative effect. A “threshold effect” may exist in which low levels of mutant transcript may be insufficient to disrupt the assembly of microfibrils from the wild-type fibrillin protein expressed from the wild-type allele, whereas higher levels of mutant proteins are able to exert their dominant-negative effect. Finally, in a classic experiment, a mutant allele was expressed from an MFS patient in normal human and murine fibroblasts. There was substantially reduced ECM deposition of fibrillin in these cultures. Thus, even in the background of two normal alleles, the expression of the mutant monomer led to disruption of fibrillin deposition in the ECM. The second proposed mechanism of the pathogenesis of MFS is an altered homeostasis of the microfibril. The elastic fibers within the aortic media consist of an amorphous core comprised primarily of cross-linked tropoelastin monomers surrounded by microfibrils. During development, microfibrils are found at the margins of maturing elastic fibers, leading to the hypothesis that microfibrils and FBN1 may play a crucial role in the deposition of elastic fibers during embryogenesis, and that the distortion in the architecture of elastic fibers seen in MFS thus reflects abnormal elastogenesis. Mice with targeted mutations in the FBN1 were created to test this hypothesis, leading to a 10-fold reduction in the level of the mutant transcript. Heterozygous mice expressed very low levels of the mutant transcript and were morphologically and histologically indistinguishable from the wild-type mice. Homozygous mutant mice appeared normal at birth, but died shortly thereafter of vascular complications, with histologic analysis revealing thinning of the proximal aortic wall. Immunostaining showed substantial reduction in the amount of extracellular fibrillin, but a normal amount of elastin staining, suggesting that elastogenesis may

118

SECTION II / CARDIOLOGY

proceed in a normal fashion even in the absence of normal development of microfibrils. The investigators hypothesized that microfibrils may be important in the homeostasis of the aortic wall, and in the absence of proper microfibril development, the aortic media is unable to sustain the hemodynamic stress to which it is subjected, eventually leading to aortic dilatation. The disruption of the elastic network may then be a secondary event. Studies involving another mutant mouse model producing a fivefold reduction in gene expression support the concept. The homozygous mutant mice gradually develop severe kyphoscoliosis and die prematurely of Marfan-like vascular complications. Histological examination of these mice at birth shows normal vascular anatomy with a seemingly normal appearing elastin network in the aortic media. However, by 6 wk of age a sequence of events commences with focal calcification in the aortic elastic lamellae, progressing to intimal hyperplasia, monocytic infiltration, fragmentation of elastic lamellae, and eventually dilatation of the aortic wall. An increase in fibrillin proteolysis is the third mechanism proposed of the pathogenesis of MFS. It is recognized that calcium binding to the calcium-binding epidermal growth factor-like (cbEGF) domains of fibrillin is essential for the proper assembly and integrity of the microfibrillar aggregates, and that calcium may protect wild-type fibrillin from proteolysis. Because FBN1 mutations reduce the calcium affinity of cbEGF motifs in vitro, investigators have hypothesized that mutations in the cbEGF domains might render the FBN1 aggregates more prone to proteolysis. Studies have been hampered by the difficulty of expressing fulllength fibrillin in vitro, so most experiments have used fibrillin fragments. Nonetheless, these experiments have shown that certain FBN1 mutations increase protease susceptibility by exposing enzyme-specific cryptic cleavage sites for particular proteases. The clinical sequelae remain unclear. In one study, however, surgical thoracic aortic aneurysm specimens from patients with MFS did show increased immunofluorescence for certain matrix metalloproteinases (proteolytic enzymes), especially at the border of areas of cystic medial necrosis. Nonetheless, in contrast to the pathogenesis of abdominal aortic aneurysms (AAAs, discussed later), direct evidence linking proteases to the pathogenesis of MFS is lacking. The pathogenesis of MFS thus remains unclear. It appears that all three mechanisms described may contribute to the ultimate disruption of the cellular and molecular architecture of the aortic wall’s tunica media. A unified theory will require a better understanding of the dynamic biology of the microfibril, its interaction with other proteins, and its interaction with adjacent vascular smooth cells. Also unclear is the potential role of modifying genetic elements, which may be responsible for modulating the phenotypic expression of a particular MFS genotype. BICUSPID AORTIC VALVE Bicuspid aortic valve is the most common congenital cardiac malformation, affecting 1–2% of the population. Bicuspid aortic valve is highly associated with other congenital aorta abnormalities, such as coarctation and patent ductus arteriosis. Bicuspid aortic valve is often associated with dilatation of the aortic root or ascending aorta and may progress to frank thoracic aortic aneurysms and aortic dissection. Consequently, despite its name, bicuspid aortic valve should be considered an abnormality of both the valve and aorta, rather than just the valve. Indeed, over time most individuals with a bicuspid aortic valve will develop complications—either valve dysfunction or aneurysm of the thoracic aorta—requiring surgical treatment. The pathogenesis of the bicuspid aortic valve is not well understood. However, it seems likely that the abnormal aortic

cusp formation during valvulogenesis is the result of a complex developmental process involving the ECM. During development the ECM provides scaffolding and cues that are critical in cellular migration and pattern formation. In the aorta, differentiation of the cushion mesenchymal cells into mature valve cells correlates with the expression of microfibrillar proteins fibrillin and fibulin. Genetic mutations that result in inadequate amounts of such proteins, or that disrupt the timing or function of such proteins, lead to abnormal valvulogenesis early in life, and perhaps a weakened aortic wall later in life, in turn resulting in aortic aneurysms and dissection. Although there does appear to be a familial risk, no single mutation resulting in the bicuspid aortic valve phenotype has been identified. Research is focusing on transcriptional elements and signaling molecules that may alter the expression of the important ECM proteins. Mice deficient in endothelial nitric oxide synthase (eNOS) have a high incidence of congenital bicuspid aortic valves. In wild-type mice, the expression of endothelial nitric oxide synthase is noted in the myocardium as well as the endothelium during development, but is localized to the endothelium lining the valve at maturity, suggesting a role in valvulogenesis. Conversely, there was no difference in the diameter of the aortic lumen between the two groups. Research has also concentrated on the etiology of the vascular complications associated with bicuspid aortic valve. The pathology of the aortic wall is likely the result of intrinsic defects in the ECM rather than a secondary consequence of valvular dysfunction. This is supported by the observation that dilatation of the ascending thoracic aorta can be found in young adults with a bicuspid aortic valve but without significant valvular stenosis or regurgitation. Similarly, even those with a bicuspid aortic valve who have undergone aortic valve replacement can develop ascending thoracic aortic aneurysms at a later date. The thoracic aortic aneurysms associated with bicuspid aortic valve typically demonstrate accelerated degeneration of the aortic media. The histopathology in such cases is similar to that of patients with MFS who also have abnormal FBN1 content in their aortic wall. FAMILIAL THORACIC AORTIC ANEURYSM SYNDROME In addition to the association of ascending thoracic aortic aneurysms with MFS and bicuspid aortic valve, some patients with such aneurysms and proven cystic medial degeneration have no other identifiable congenital abnormalities. Indeed, evidence suggests that many of these patients also have a genetic mutation that may account for cystic medial degeneration. Moreover, it is recognized that cases of thoracic aortic aneurysms in the absence of overt connective tissue disorders are often familial in nature and reflect a familial thoracic aortic aneurysm syndrome. In an analysis of a large database of patients with thoracic aortic aneurysms, at least 19% of those without MFS had a family history of a thoracic aortic aneurysm. Most pedigrees suggested an autosomal-dominant mode of inheritance, although some suggested a recessive mode and possibly X-linked inheritance as well. Another series examined the families of 158 patients referred for surgical repair of thoracic aortic aneurysms or dissections and found that first-degree relatives of probands had a higher risk of thoracic aortic aneurysms compared with controls. Investigation of the genetic basis for this syndrome is still in its infancy. Several genes have been implicated. A mutation on 3p24.2-25 has been identified that can cause both isolated and familial thoracic aortic aneurysms owing to cystic medial

CHAPTER 13 / AORTIC DISEASES

degeneration. However, despite a dominant pattern of inheritance, there is a marked variability in the expression and penetrance of the disorder, with some inheriting the gene but showing no manifestation. Two other studies of familial thoracic aortic aneurysm syndromes have mapped the mutations to at least two different chromosomal loci, but other families mapped to neither of these, suggesting genetic heterogeneity in addition to variable expression and penetrance.

ABDOMINAL AORTIC ANEURYSMS Unlike the ascending thoracic aortic aneurysms discussed, abdominal aortic aneurysms (AAAs) are typically associated with aging and atherosclerosis. Gender plays a role; men are 10 times more likely than women to have an AAA of 4 cm or greater. However, women with an AAA have a significantly greater risk of rupture than men. It is also recognized that genes influence aneurysm formation. Those having a first-degree relative with an AAA have an increased risk of 13–32% compared with the 2–5% risk in the general population. In addition, those with familial aneurysms tend to be younger and have higher rates of rupture than those with sporadic aneurysms. However, no single gene defects for AAA have been identified, and pedigree analysis suggests that the increased risk is probably polygenic. To clarify the genetic underpinnings and foster improvements in prevention and therapy, there has been intense investigation to elucidate the pathophysiology of AAA. Examination of human AAA tissue resected at surgery and experiments from explanted AAA tissue maintained in tissue culture have provided important clues. First, the pathologic hallmark seems to be destructive remodeling of the elastic media of the aortic wall, including progressive degradation of the fibrillar matrix proteins. Second, evidence of an infiltration of inflammatory cells exists, including B lymphocytes, T lymphocytes, and macrophages. Third, there is increased immunoreactivity for elastolytic matrix metalloproteases (MMP), particularly MMP-2, MMP-9, and MMP-12. Of these, MMP-9 is the most abundant elastase secreted by AAA human tissue explants in vitro and seems to be actively expressed by macrophages at the site of tissue damage in resected AAA tissue. Of the animal models developed to study the specific roles of the various MMPs, one of the most durable is the elastase-induced rat or mouse model of AAA. In it, the infrarenal aorta is temporarily ligated proximally (just below the level of the renal artery) and distally at the level of the bifurcation. A catheter is introduced via an aortotomy at the distal site of ligation and the portion of the aorta between the two ligations is perfused with elastase for 5 min. The sutures are then removed and the aortic dimensions can be measured at various time-points. A moderate amount of aortic dilatation appears immediately after the perfusion. There is then no further increase in aortic diameter for the next 7 d. However, at 14 d, the aortas in the elastase infusion group were significantly dilated (compared with a heat-inactivated elastase control group) and reached aneurysmal proportions. Light microscopy revealed minimal damage to the medial elastic lamellae for up to 7 d in both groups. By 14 d, however, the elastase-perfused aortas exhibited a dense inflammatory infiltrate along with extensive degradation of the elastin framework. In contrast, the aortas infused with inactive elastase appeared normal at 14 d. This supports a model in which the initial elastase infusion serves as an insult that evokes an inflammatory response leading to tissue destruction and elastolysis, and ultimately, aneurysmal dilatation of the aortic wall.

119

Work has focused on the molecular players immediately upstream of elastolysis, and hence may form reasonable targets for therapeutic intervention. As mentioned, the most obvious candidates were MMP-9, MMP-12, and MMP-2. Using the mouse model of elastase-induced aneurysmal dilatation, investigators found that the development of aneurysms in this model was temporally and spatially correlated with “chronic inflammation.” Staining with antibodies seemed to indicate that these macrophages were the likely source of MMP-9. Furthermore, levels of MMP-9 and MMP-12 were both increased during development of elastase-induced AAA, and treatment of these animals with doxycycline, a nonselective MMP-inhibitor, led to a significant reduction in the size of the aneurysms. In a final set of elegant experiments, the authors attempted to uncover the role of various specific MMPs using mice with targeted disruption of the MMP9 and MMP-12 genes. Elastase infusion resulted in a significantly smaller AAA size in MMP-9 knockout mice and in mice that had been bred to have combined deficiencies in MMP-9 and MMP-12. However, there was no reduction in AAA size in the MMP-12 knockout mice. Examination of aortic tissue from these mice models revealed no suppression of the elastase-induced inflammatory response. Indeed, the MMP-9 knockout mice continued to display infiltration by macrophages and polymorphonuclear leukocytes, yet at 14 d the elastic lamellae remained well preserved. However, when the bone marrow of the MMP-9 knockout mice was reconstituted with wild-type bone marrow (i.e., not MMP-9 deficient), a significant increase in aortic diameter resulted. These experiments suggest that the signals to recruit the inflammatory response following exposure to elastase are not dependent on the presence of the MMPs examined. The expression of MMP-9 by the infiltrating macrophages appears to be crucial to elastolysis and the subsequent development of aortic aneurysms. It is unclear what other agents may be involved in the MMP cascades that seem to underlie the development of aneurysms. For example, plasmin and plasminogen activators seem to be involved in the development of aneurysms in some mice models, and the overexpression of plasminogen-activator inhibitor (PAI-1) can suppress aneurysm formation in a rat model. Proteases such as plasmin/PA and elastases may be part of a cascade that leads to progressive activation of other proteases (such as MMP-9 from pro-MMP-9) much like the coagulation and complement cascade. Furthermore, protease inhibitors such as plasminogen-activator inhibitor and tissue inhibitors of MMP might act as checks to prevent rampant activation of the proteases. Disruptions or changes in the “balance of power” could alter the process in favor of proteases and hence elastolysis. Despite the extensive progress in the understanding of the protease cascades that may be responsible for elastolysis, much less is known about the signals that initiate the inflammatory process in the wall of aortic aneurysms. There is a clear association between atherosclerosis and aneurysm formation, so atherosclerotic plaques (known to contain an inflammatory infiltrate) might be an “insult” to the arterial wall, akin to the exposure of mice aorta to elastase. A model of atherosclerosis and underlying endothelial cell dysfunction has emerged in the form of mice deficient in apo-E, which have accelerated atherosclerotic lesions. However, despite their extensive atherosclerotic lesions these mice do not develop AAA. Hence just the presence of atherosclerosis and the associated inflammation may not be sufficient to induce the formation of aneurysms. One possible explanation is that

120

SECTION II / CARDIOLOGY

compensatory mechanisms in the vascular bed—such as the release of nitric oxide by endothelial cells—counteract some effects of inflammation. Although endothelial cell-dependent relaxation of the vascular bed is mediated by nitric oxide, it appears that nitric oxide may have multiple other roles, including inhibition of smooth muscle cell proliferation, platelet aggregation/adhesion, and leukocyte activation. Perhaps removing this versatile signaling molecule in the presence of atherosclerotic plaque induces the formation of an aneurysm. This was accomplished experimentally by crossing e-NOS knockout and apo-E knockout mice to create mice deficient in both proteins. The “double knockout” mice were more hypertensive that control wild-type mice, and a significant proportion of these mice developed an AAA. The development of the AAA was not simply owing to hypertension, because lowering the blood pressure to levels seen in the control mice had no effect in reducing aneurysm development, suggesting that there may be a “multihit” mechanism for the development of AAA. Atherosclerotic lesions, such as those seen in the apo-E knockout mice, may serve as the initial insult leading to a chronic inflammatory state; meanwhile, the presence of preserved endothelial function mediated by nitric oxide may protect from the development of AAA. However, the addition of significant endothelial dysfunction (in this case the lack of the critical endothelial regulator nitric oxide) may be the second factor that accelerates the process of AAA formation.

SELECTED REFERENCES Allaire E, Hasenstab D, Kenagy RD, et al. Prevention of aneurysm development and rupture by local overexpression of plasminogen activator inhibitor-1. Circulation 1998;98:249–255. Aoyoma T, Francke U, Dietz HC, Furthmayr H. Quantitative differences in biosynthesis and extracellular deposition of fibrillin in cultured fibroblasts distinguish five groups of Marfan syndrome patients and suggest distinct pathogenetic mechanisms. J Clin Invest 1994;94:130–137. Biddinger A, Rocklin M, Coselli J, Milewicz DM. Familial thoracic aortic dilatations and dissections: a case control study. J Vasc Surg 1997;25:506–511. Booms T, Tiecke F, Rosenberg T, Hagemeier C, Robinson PN. Differential effects of FBN1 mutations on in vitro proteolysis of recombinant fibrillin-1 fragments. Hum Genet 2000;107:216–224. Carmeliet P, Moons L, Lijnen R, et al. Urokinase-generated plasmin activates matrix metalloproteinases during aneurysm formation. Nat Genet 1997;17:439–444. Chen J, Kuhlencordt PJ, Astern J, Gyurko R, Huang P. Hypertension does not account for the accelerated atherosclerosis and development of aneurysms in male apolipoprotein E/endothelial nitric oxide synthase double knockout mice. Circulation 2001;104:2391–2394. Coady MA, Davis RR, Roberts M, et al. Familial patterns of thoracic aortic aneurysms. Arch Surg 1999;134:361–367. Collod G, Babron M, Jondeau G, et al. A second locus for Marfan syndrome maps to chromosome 3p24.2-p25. Nat Genet 1994;8:264–268. Curci JA, Liao S, Huffman MD, Shapiro SD, Thompson RW. Expression and localization of macrophage elastase in abdominal aortic aneurysms. J Clin Invest 1998;102:1900–1910. Daugherty A, Cassis LA. Mechanisms of abdominal aortic aneurysm formation. Curr Atheroscler Rep 2002;4(3):222–227. De Paepe A, Devereux RB, Dietz HC, Hennekam RC, Peyritz RE. Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet 1996;62:417–426. Dietz HC, Cutting GR, Pyeritz RE, et al. Marfan syndrome caused by a recurrent de novo missense mutation in the fibrillin gene. Nature 1991;352:337–339. Dietz HC, McIntosh I, Sakai LY, et al. Four novel FBN 1 mutations: Significance for mutant transcript level and EGF-like domain calcium binding in the pathogenesis of Marfan syndrome. Genomics 1993;17: 468–475.

Eisenberg LM, Markwald RR. Molecular regulation of atrioventricular alvulospetal morphogenesis. Circ Res 1995;77:1–6. Eldadah ZA, Brenn T, Furthmayr H, Dietz HC. Expression of a mutant human fibrillin allele upon a normal human or murine genetic background recapitulates a Marfan cellular phenotype. J Clin Invest 1995;95:874–880. Fedak PWM, Verma S, David TE, Leask RL, Weisel RD, Butany J. Clinical and pathophysiological implications of a bicuspid aortic valve. Circulation 2002;106:900–904. Freestone T, Turner RG, Coady A, et al. Inflammation and matrix metalloproteinases in the enlarging abdominal aortic aneurysm. Arterioscler Thromb Vasc Biol 1995;15:1145–1151. Fukui D, Miyagawa S, Soeda J, Tanaka K, Urayama H, Kawasaki S. Overexpression of transforming growth factor beta1 in smooth muscle cells of human abdominal aortic aneurysm. Eur J Vasc Endovasc Surg 2003;25(6):540–545. Kuhlencordt PJ, Gyurko R, Han F, et al. Accelerated atherosclerosis, aortic aneurysm formation, and ischemic heart disease in apolipoprotein E/endothelial nitric oxide synthase double-knockout mice. Circulation 2001;104(4):448–454. Lee TC, Zhao YD, Courtman DW, et al. Abnormal aortic valve development in mice lacking endothelial nitric oxide synthase. Circulation 2000;101:2345–2348. Loeys B, Nuytinck L, Delvaux I, De Bie S, De Paepe A. Genotype and phenotype analysis of 171 patients referred for molecular study of the fibrillin-1 gene FBN1 because of suspected Marfan syndrome. Arch Intern Med 2001;161:2447–2454. Milewicz DM, Chen H, Park E-S, et al. Reduced penetrance and variable expressivity of familial thoracic aneurysms/dissections. Am J Cardiol 1998;82:474–479. Miralles M, Wester W, Sicard GA, Thompson R, Reilly J. Indomethacin inhibits expansion of experimental aortic aneurysms via inhibition of the cox2 isoform of cyclooxygenase. J Vasc Surg 1999;29(5):884–893. Nkomo VT, Enriquez-Sarano M, Ammash NM, et al. Bicuspid aortic valve associated with aortic dilatation. Arterioscler Thromb Vasc Biol 2002;23:351–356. Pepe G, Giusti B, Atlanasio M, et al. A major involvement in the cardiovascular system in patients affected by Marfan syndrome: novel mutations in fibrillin 1 gene. J Mol Cell Cardiol 1997;29:1877–1884. Pereira L, Andrikopoulds K, Tian J, et al. Targetting of the gene encoding fibrillin-1 recapitulates the vascular aspect of Marfan syndrome. Nat Genet 1997;17:218–222. Pereira L, Lee SY, Gayraud B, et al. Pathogenetic sequence for aneurysm revealed in mice underexpressing fibrillin-1. Proc Natl Acad Sci USA 1999;96:3819–3823. Pyeritz RE. The Marfan syndrome. Annu Rev Med 2000;51:481–510. Pyo R, Lee JK, Shipley M, et al. Targeted gene disruption of metalloproteinase-9 suppresses development of experimental abdominal aortic aneurysms. J Clin Invest 2000;105:1641–1649. Reinhardt DP, Ono RN, Notbohm H, Muller PK, Bachinger HP, Sakai LY. Mutations in calcium-binding epidermal growth factor modules render fibrillin-1 susceptible to proteolysis: a potential disease causing mechanism in Marfan syndrome. J Biol Chem 2000;275: 12,339–12,345. Robinson PN, Booms P. The molecular pathogenesis of the Marfan syndrome. Cell Mol Life Sci 2001;58:1698–1707. Robinson PN, Booms P, Katzke S, et al. Mutations of FBN1 and genotype-phenotype correlations in Marfan syndrome and related fibrillinopathies. Hum Mutat 2002;20:153–161. Saito S, Zempo N, Yamashita A, Takenaka H, Fujioka K, Esato K. Matrix metalloproteinase expressions in arteriosclerotic aneurysmal disease. Vasc Endovascular Surg 2002;36(1):1–7. Shah PK. Inflammation, metalloproteinases, and increased proteolysis: an emerging pathophysiological paradigm in aortic aneurysm. Circulation 1997;96:2115–2117. Thompson RW, Holmes DR, Mertens RA, et al. Production and localization of 92-kilodalton gelatinase in abdominal aortic aneurysms: an elastolytic metalloproteinase expressed by aneurysm-infiltrating macrophages. J Clin Invest 1995;96:318–326. Vaughan CJ, Casey M, He J, et al. Identification of a chromosome 11q23.2-q24 locus for familial aortic aneurysm disease, a genetically heterogeneous disorder. Circulation 2001;103:2469–2475.

14 Atherosclerotic Coronary Disease ROBERT E. GERSZTEN AND ANTHONY ROSENZWEIG SUMMARY Atherosclerotic coronary artery disease is a complex biological process resulting in narrowing of the arterial vessels that supply the heart muscle with oxygen and nutrients. Despite major advances in the understanding of this process and its clinical management, it remains a major cause of morbidity and mortality throughout the world. In the United States, coronary artery disease accounts for about 500,000 deaths each year, of which about half are sudden. This chapter reviews the clinical presentation and treatment of this condition, focusing on advances and the biological insights they provide. Some of the leading hypotheses regarding the molecular mechanisms underlying this incompletely understood condition are considered. Key Words: Angina; atherosclerosis; coronary artery disease; endothelial cells; genetic models; infarction; inflammation; ischemia; lipids; platelets; reperfusion; smooth muscle cells.

DEFINITION OF TERMS The major complication of coronary artery disease (CAD) is myocardial injury and dysfunction from inadequate delivery of essential nutrients. Coronary obstruction can be severe before it limits blood flow, particularly under resting conditions, and thus this process can develop over years while remaining clinically silent. Inadequate blood flow that does not produce permanent myocardial cell damage is termed myocardial ischemia. Reduction or obstruction in flow sufficient in severity and duration to cause irreversible, clinically detectable myocardial damage is termed myocardial infarction (MI). Improved biochemical markers of myocyte injury with greater sensitivity and specificity (such as cardiac troponin isoforms) have made the detection of small amounts of myocardial injury feasible in a clinical setting. Detection of these markers identifies a subset of patients at higher risk for subsequent complications and increases the number of patients with diagnosable infarction who would previously have been considered to have suffered “only” an episode of ischemia. The clinical and biological spectrum from mild ischemia to substantial infarction appears to be a continuum rather than distinct categories. The clinical symptoms associated with myocardial ischemia are termed angina pectoris. Many patients with coronary disease experience exertional symptoms in a consistent and predictable pattern termed chronic stable angina. A marked acceleration in the frequency, severity, or duration of angina, or a significant decrease in From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

the level of exertion inducing angina, constitutes unstable angina and carries a substantial risk of progression to MI. Together unstable angina and MI comprise the acute or unstable coronary syndromes. The biological basis for the transition from no or chronic stable angina to an unstable coronary syndrome is the subject of intense investigation and has important clinical implications.

CLINICAL FEATURES Elements contributing to the development of atherosclerotic lesions include genetic predisposition, elevated lipid levels, diabetes, tobacco smoking, and hypertension. Peripheral markers of inflammation such as C-reactive protein may provide independently informative risk stratification. Many of these clinical risk factors generally fit well with the prevailing inflammatory hypothesis of atherogenesis discussed next. However, a precise and comprehensive understanding of the pathophysiological mechanisms responsible remains elusive. Moreover, known factors do not fully account for individual or heritable risk of developing CAD. Nevertheless, efforts to mitigate these now-established risk factors in combination with improvement in medical and surgical management have led to a significant decline in age-adjusted mortality rates for patients with cardiovascular disease over the past several decades. One of the major advances in management of MI has been the recognition of the importance of rapidly restoring reperfusion to the affected myocardium. The rapidity of reperfusion appears critical whether achieved through thrombolysis or catheter-based intervention. Though studies suggest a clinical advantage to the latter approach, logistical considerations and the availability of experienced interventionalists may sway the decision toward the former. Yet the increase in aggressive management of coronary disease has produced valuable clinical information, and in part through this experience, theories about the pathogenesis of the atherosclerotic plaque have also evolved. For example, serial angiographic studies show that often it is not the most angiographically severe lesions that become unstable clinically. Rather, nonobstructive lesions may progress rapidly to obstruction through plaque rupture and/or thrombosis. Thus the angiographic appearance of a lesion is a poor predictor of its clinical behavior. Lessons learned from clinical trials with lipid lowering agents are consistent with these observations. Although lipid reduction only modestly improves the angiographic severity of high-grade atherosclerotic lesions, it substantially reduces the risk of acute coronary events such as unstable angina or MI. In this sense, CAD is not simply a mechanical problem of luminal obstruction but also

121

122

SECTION II / CARDIOLOGY

a biological process, which increases the challenges and opportunities of managing such patients.

DIAGNOSIS AND MANAGEMENT STABLE CORONARY DISEASE The chest discomfort associated with ischemia is termed angina and is classically described as a substernal chest pressure brought on by exertion and relieved by rest. Radiation of this discomfort (to the arm or jaw) and associated symptoms such as dyspnea or diaphoresis are common. Other precipitants include emotional stress, cold weather, or even large meals. Comorbid diseases such as anemia, thyrotoxicosis, or infection should be considered as possible contributors. Angina most frequently results from epicardial coronary stenoses that impair coronary flow reserve. As it is decreased, stress-induced myocardial ischemia typically results in anginal chest discomfort. Angina often correlates with coronary stenoses of 70% of the luminal cross-section, whereas rest angina usually does not develop until stenoses are greater than 90%. Many patients with CAD have documented episodes of ischemia in the absence of anginal symptoms. The full implications and management of silent ischemia remain controversial. Although the physical examination provides important insights into the status of the patient’s cardiovascular system and general health, it can neither establish nor refute a diagnosis of CAD or angina. Further evaluation is useful for diagnosis in patients with suspected CAD and for prognosis in patients with known CAD. As with all clinical testing, a positive or negative result must be interpreted in the context of the prior probability that a diagnosis such as ischemic heart disease exists in the patient or population under study. Further testing is most useful when clinical evaluation places the prior probability of CAD in an intermediate level. If the prior probability of disease is extremely low, a positive test is most likely to be a false-positive and may precipitate a cascade of expensive, otherwise unnecessary and potentially harmful invasive testing. Conversely, if the prior probability is extremely high, a negative test is likely falsely negative and a positive test contributes little to the diagnosis. However, testing can be useful in patients with an intermediate probability of CAD, because a positive test may appropriately tip the balance toward invasive evaluation whereas a negative test may lower the likelihood of CAD sufficiently to obviate the associated risks. In patients with known CAD, exercise testing can provide a useful functional assessment and prognostic information. Evocative testing is necessary because ischemia connotes a relative deficiency of blood flow that is usually not apparent at rest until stenoses become severe. In general, provocative approaches include exercise and pharmacological manipulations such as IV adenosine, dipyridamole, or dobutamine. The pharmacological approaches do not provide as much functional information as exercise-testing and generally should be reserved for patients who cannot exercise adequately. Clinical status (symptoms, blood pressure, and ECG) is monitored but is generally a more reliable indicator of ischemia with exercise than pharmacologic testing. Noninvasive assessment of myocardial perfusion is most commonly accomplished with radionuclide scintigraphy following the intravenous administration of a radioisotope such as thallium-201 or Technetium-99m. Such radionuclide scintigraphy improves the sensitivity and specificity of evocative testing. In addition, these techniques often allow differentiation between ischemic and infarcted zones. In the former, isotope uptake is reduced during initial stress but normalizes over time with isotope redistribution

(thallium) or reinjection (technetium). In contrast, in infarcted tissues, uptake is persistently diminished. Information about coronary flow can also be inferred from dynamic changes in systolic wall movement and thickening most commonly assessed by echocardiography. The sensitivity and specificity of stress echocardiography are similar to that of thallium stress testing. These imaging modalities increase the sensitivity, specificity and expense of evocative testing. UNSTABLE CORONARY SYNDROMES The hallmark of unstable coronary syndromes is rapidly progressive substernal chest pressure, often radiating to the jaw or left arm. The initial presentation, however, may be symptoms at rest. The discomfort of acute MI is qualitatively similar, though often more severe and prolonged. It is acute in onset and often associated with diaphoresis, dyspnea, and an impending sense of doom. Elderly and diabetic patients often have less typical symptoms. Underlying illnesses such as anemia or infection may precipitate the presentation of CAD. Acute infarction occurs in a minority of patients presenting with chest pain of presumed cardiac origin. Chest discomfort described as “pressure” or “burning” is most associated with acute infarction. Positional or pleuritic pain should prompt consideration of other etiologies. It should be noted that approx 25% of MIs are silent and associated with no or only mild and atypical symptoms. The prognosis of such silent infarctions appears similar to that of more classic presentations. PHYSICAL EXAMINATION Physical examination neither establishes nor excludes the diagnosis of acute MI or unstable angina. It does, however, provide critical clinical information of practical importance in patient management. Evaluation of the vital signs is particularly important in patients with chest pain. The heart rate is usually normal or slightly elevated, though infarctions involving the diaphragmatic portion of the heart are often accompanied by bradycardia and nausea. An irregular rhythm should raise the suspicion of ventricular ectopy or atrial fibrillation. Hypotension in the setting of chest pain may be secondary to pump failure and is a particularly ominous finding. Mild hypertension is more commonly found with acute onset of symptoms. Tachypnea may be secondary to evolving congestive heart failure. The jugular venous pulsation is usually normal. An elevated measurement may denote chronically elevated right-sided filling pressures but should raise the possibility of right ventricular involvement. The diagnosis of right ventricular infarction may also be suggested by a paradoxical rise in venous pressure with inspiration, known as Kussmaul’s sign. Evaluation of the arterial pulse amplitude and duration provides important clues to possible valvular disease or associated noncoronary arterial disease. The pulse contour also reflects the patient’s cardiac output. This evaluation may be confounded in elderly patients by noncompliant arteries in which the pulse amplitude may appear normal despite a significantly reduced cardiac output. The status of peripheral arteries and the quality of the lower extremity veins should be carefully assessed before consideration of interventions such as cardiac catheterization, angioplasty, intraaortic balloon counterpulsation or bypass surgery. The possible auscultatory findings in atherosclerotic heart disease are protean. An S4 heart sound is common. A third heart sound, however, reflects significant systolic impairment. Cardiac murmurs may be fixed or transient, emphasizing the importance of serial examination. Both stable and unstable symptoms may be precipitated or exacerbated by underlying valvular heart disease such as aortic stenosis. The murmur of papillary muscle dysfunction is often transient whereas mechanical complications such as

CHAPTER 14 / ATHEROSCLEROTIC CORONARY DISEASE

ventricular septal defect may not occur until days into the hospital stay. A pericardial friction rub often occurs several days after infarction. Evaluation of the lungs centers on the presence of rales consistent with congestive heart failure. In the acute coronary syndromes (ACSs), pulmonary rales coupled with an S3 gallop is a poor prognostic sign. LABORATORY EVALUATION The classic ECG pattern of acute MI includes ST segment elevation in multiple leads reflecting a coronary distribution, followed by evolution of T-wave inversion and significant Q waves in these same leads, reflecting a Q-wave MI. However, patients with infarction can also present with ST segment depression or T-wave inversion alone without evolution of Q waves, termed non-Q wave-MI. The correlation of these ECG findings with pathological findings of transmural or subendocardial infarction is sufficiently imprecise to favor exclusive use of the ECG descriptors in clinical practice. Serological markers of myocardial necrosis remain the cornerstone in establishing the diagnosis of MI. The serum creatine kinase (CK) and its myocardial-specific isozyme, CK-MB, are widely used. However, as previously noted, cardiac troponin isoforms have improved sensitivity and specificity compared with CK, and generally become abnormal by 6 h after injury facilitating more rapid risk stratification. The degree of elevation in all these markers correlates approximately with the amount of myocardium damaged but is influenced by other factors. Elevated levels of serum transaminase and lactate dehydrogenase with isoform reversal occur later after infarction. Measurement of these levels generally adds little to clinical management and should not be routinely employed in evaluation of acute chest pain. Panels of markers reflecting myocyte necrosis (Troponin), neurohormonal activation (brain natriuretic peptide), and inflammation (CRP) are being evaluated to see if together they provide more precise prognostic information. MANAGEMENT In approaching an individual patient, several goals must be considered simultaneously: prevention of disease progression, control of symptoms, and optimizing prognosis. Available tools include patient education and risk factor modification, pharmacological therapy, and mechanical interventions including surgical and catheter-based approaches. All these considerations must be integrated into a coherent clinical approach, which in practice is often influenced dramatically by the acuity of the patient’s clinical presentation. Patient education and behavior modification form the cornerstone of primary and secondary prevention. Aggressive risk factor modification, particularly reduction in serum lipids, can have a dramatic clinical impact even within a few years in populations with documented CAD. Randomized trials have documented angiographic regression of plaques, establishing that regression is possible in principle. However, even more intriguing is the observation that although the degree of regression has been modest, the improvement in clinical end points has been more dramatic. Reduction in serum lipids appears to have clinical benefits, which are only imperfectly reflected in the angiographic appearance of lesions. These clinical benefits can be realized during a relatively brief (2–3 yr) period of treatment, which underscores the importance of secondary prevention efforts. Although lifestyle modifications are emphasized as the first step in this process, it is often necessary to move to pharmacological lipid lowering therapies. The particular agent(s) employed is determined by the specifics of the patient’s lipid profile. Other therapies of proven clinical benefit in patients who have a history of MI include aspirin, β-blockade, and angiotensinconverting enzyme inhibitors, particularly in patients with a reduced

123

left ventricular ejection fraction. These agents reduce the incidence of clinical cardiovascular events and mortality. For this reason, they should be considered in all patients after MI, even if they are asymptomatic. Medications that may be useful in the long-term treatment of coronary patients but have not been demonstrated to improve prognosis include calcium channel blockers and nitrates. In fact, in the setting of non-Q-wave infarction, the calcium channel blocker diltiazem had a deleterious effect on 1-yr survival in patients manifesting congestive heart failure during the initial hospitalization. These medications may help control anginal symptoms, hypertension, or arrhythmias, but should not be routinely prescribed independent of a specific indication because of the lack of proven outcome benefit. Therapy for chronic stable angina includes pharmacological and mechanical interventions. The three classes of pharmacological agents primarily employed are β-blockers, calcium channel blockers, and nitrates. As noted, only the first has documented survival benefit in postinfarction patients but all three have been demonstrated to improve exercise tolerance and can reduce ischemic symptoms, both valid goals. Mechanical interventions encompass a variety of catheter-based approaches such as percutaneous transluminal angioplasty as well as coronary artery bypass surgery graft (CABG). CABG provides a survival benefit over medical therapy alone in patients with significant left main coronary disease or severe three-vessel coronary disease (particularly in the setting of compromised left ventricular function). In patients with one- or two-vessel coronary disease and mild symptoms, CABG does not confer a survival benefit. However, surgery eliminates or reduces anginal symptoms more effectively than medical therapy alone. Catheter-based interventions are also highly effective at reducing ischemic symptoms but have generally not been demonstrated to improve survival. Percutaneous coronary interventions (PCIs) have historically been plagued by a significant rate of restenosis. The advent of mechanical support (stents) has significantly reduced this rate. The development of stents that simultaneously mediate local release of pharmacological agents has had an even more dramatic reduction in restenosis. Although the indications for such interventions are undergoing evaluation, it is likely the number of such procedures will continue to grow. Basic science findings have best been translated into therapy in relationship to platelet biology and evolving therapies for patients presenting with ACSs. In particular, efforts have centered on the abrogation of signaling triggered by the two major platelet agonists, adenosine diphosphate (ADP) and thrombin. Heparin, an indirect thrombin antagonist, works via activation of antithrombin 3, and is a mainstay of the medical regimen of all patients with ACSs. Comparative trials of the more specific low-molecular-weight heparin have demonstrated its superiority over unfractionated heparin in reducing cardiac events in ACS. Two direct thrombin inhibitors, hirudin, and bivalirudin, show trends toward benefit over heparin therapy alone in patients with ACS. The direct thrombin antagonist bivalirudin has also shown promise as an adjunctive therapy in patients undergoing percutaneous revascularization. However, owing to tremendous cost differences between lowmolecular-weight heparin and direct thrombin inhibitors vs unfractionated heparin, these newer therapies have not gained widespread acceptance. Direct thrombin inhibitors are commonly used, though for anticoagulation of patients with heparin induced thrombocytopenia, a growing problem worldwide. ADP receptor antagonists have become a mainstay of therapies for ACS and PCI. Clopidogrel in combination with aspirin

124

SECTION II / CARDIOLOGY

confers a 20% reduction in cardiovascular death, MI or stroke compared with aspirin alone in both low- and high-risk patients with ACS. The benefit of treatment before PCI has also been seen with greater than 30% reduction in events even at 1 yr. Ongoing studies are also evaluating ADP receptor antagonism in primary prevention. New strategies for profound inhibition of platelet activity at the injured coronary plaque have also focused on blockade of the platelet surface membrane glycoprotein IIbIIIa receptor, which binds circulating fibrinogen or von Willebrand factor and crosslinks platelets as the final common pathway to aggregation. Four agents directed against this receptor include chimeric monoclonal antibody fragments, peptide inhibitors, and nonpeptide mimetics. Multiple large scale placebo-controlled trials have evaluated approx 50,000 patients. Patients undergoing percutaneous interventions and high-risk patients with ACS appear to derive the greatest benefits. Benefits have been relatively uniform between the various classes of inhibition. Monoclonal antibody fragments may be most efficacious, potentially via their dual effects on platelet integins and the vitronectin receptor on the disrupted endothelium. Efforts to generate oral IIbIIIa antagonists for chronic therapeutic intervention have been unsuccessful likely because of partial agonist effects. In acute MI, the immediate goal is restoration of adequate coronary flow as quickly as possible to minimize or avoid irreversible myocardial damage. Two general approaches are widely employed to achieve this goal: primary angioplasty and thrombolytic therapy. Additional clinical benefit exists from primary angioplasty if it can be accomplished quickly by experienced interventionalists. However, in many communities this is not practical and immediate administration of thrombolytic agents may be preferable. One study suggested that the benefits of angioplasty outweighed the potential disadvantages of the delays inherent in transferring patients to specialized centers, at least in some settings. Whether this approach is further validated and becomes generally accepted remains to be seen. Thrombolytic agents have not been beneficial in the treatment of unstable angina. However, therapy with heparin and aspirin—alone and in combination— reduces the rate of progression to frank infarction. Patients with unstable symptoms that cannot be controlled by heparin/aspirin and maximal anti-ischemic therapy should undergo early catheterization to define possible options for immediate mechanical intervention. The platelet glycoprotein IIb/IIIa receptor is a pivotal mediator of platelet aggregation as noted earlier. Platelet adhesion, the first step in the process of hemostasis, can be triggered by endothelial dysfunction or injury, resulting in interaction of platelets with the subendothelial matrix. Adhesion molecules of the vessel wall, along with clotting proteins such as fibrinogen, interact with platelet-membrane glycoproteins—of which integrins such as IIb/IIIa play a key role. Although some combination of these approaches appears likely to optimize early reperfusion, additional efforts are directed at determining whether adjunctive therapies can maximize myocardial salvage after reperfusion and lead to additional therapeutic benefit. Animal studies suggest this is, in principle, possible but these concepts have not been validated clinically.

GENETIC BASIS OF DISEASE Atherosclerosis is a complex phenotype, which most commonly appears modulated by interactions between environmental factors and multiple genetic loci, only some of which have been

Figure 14-1 The central role of vascular endothelium. Many different stimuli may induce a similar repertoire of dysfunctional endothelial responses that ultimately contribute to clinical atherosclerosis. The endothelium provides a potential pathophysiologic link between wellestablished clinical risks factors such as hypercholesterolemia, cigarette smoking, or hypertension, and atherogenesis. The role of the endothelial effects of other agents such as homocysteine or viral infection as well as inflammatory cascades, remains more controversial. ET, endothelin-1; NO, nitric oxide; PGI2, prostaglandin I2.

identified. Genetic dyslipoproteinemias can substantially accelerate disease progression. Assessment of serum lipid levels should be included in the evaluation of all patients with CAD, thereby providing an initial screen for genetic dyslipoproteinemias as well. Similarly, understanding the genetic bases for other recognized clinical risk factors such as diabetes mellitus or hypertension has an important impact on the development and progression of CAD. Some studies have suggested an association between the deletional allele of the angiotensin converting enzyme gene and the risk of MI. However, this was not confirmed in a prospective study of a large cohort of US physicians. Elevated serum levels of homocysteine can result from genetic defects in specific metabolic enzymes and is independently associated with ischemic heart disease. The epidemiological relationship of CAD and markers of thrombosis such as fibrinogen levels are also being investigated. However, the genetic basis of most ischemic heart disease remains elusive. In part, this reflects the nature of most CAD seen clinically as a polygenic, quantitative trait. Experimental and statistical approaches to analysis of such traits have been developed and are being applied to a variety of CAD phenotypes. Ultimately, it is hoped that this approach can identify patients at risk, the likelihood of favorable responses to specific therapies, and unanticipated genes involved in disease pathogenesis. Although this represents an exciting approach to an important clinical problem, the impact of such approaches on clinical management remains to be validated. The conceptual and practical issues associated with such studies are addressed elsewhere in this text.

PATHOPHYSIOLOGY MOLECULAR PATHOPHYSIOLOGY Hypotheses concerning the molecular pathogenesis of ischemic heart disease seek to explain two related but distinct phenomena: the gradual development of obstructive atherosclerotic lesions responsible for stable angina and the rapid progression of these lesions to the ACSs.

CHAPTER 14 / ATHEROSCLEROTIC CORONARY DISEASE

125

Figure 14-2 Leukocyte interactions with the vessel wall. An activated or dysfunctional endothelium may initiate a cascade of events that augment atherogenesis, including the recruitment of leukocytes through endothelial expression of cytokines and adhesion molecules, as well as alterations in smooth muscle cell function mediated by growth factors, cytokines, and vasoactive substances (indicated by arrows). Mononuclear cells and the foam cells they give rise to, as well as vascular smooth muscle cells, can also release growth factors and cytokines that further perpetuate this cycle.

It was first proposed that atherosclerotic lesions develop as a “response to injury” of the vascular endothelium, and much of the view of atherogenesis arises from these concepts. Subsequently this model has been modified to suggest that atherogenesis may begin with endothelial activation or dysfunction rather than actual injury or loss of the endothelium. Indeed, endothelial dysfunction may represent a common response to a wide variety of clinically relevant factors (Fig. 14-1), and the endothelium may represent a critical interface integrating these stimuli and modulating the behavior of circulating leukocytes and subintimal constituents of the vessel wall, such as smooth muscle cells (SMCs, Fig. 14-2). Activation of vascular endothelium may initiate a cascade of events leading to mononuclear leukocyte recruitment into the vessel wall with subsequent release of cytokines and growth factors contributing to SMC migration and proliferation, as well as abnormalities of extracellular matrix formation. Although this model is intuitively appealing and supported by a wealth of correlative studies, the precise molecular mechanisms involved in vivo are only incompletely understood. Even less well delineated are the mechanisms underlying the abrupt clinical change from stable angina (or no symptoms at all) to the unstable coronary syndromes. The anatomic correlate of this clinical transformation appears to be acute plaque rupture. Exposure of the atherosclerotic fatty core to the blood likely serves as a substrate for the propagation of clot and inflammation. Consistent with this model, pathological studies have emphasized the importance of a central lipid core in the process of plaque fissuring and confirmed the role of thrombosis in unstable angina and MI. MURINE MODELS OF ATHEROSCLEROSIS The understanding of atherogenesis and the ability to deduce causal relationships has been substantially advanced by the use of inbred murine

models. Although most wild-type mouse strains appear resistant to atherosclerosis, certain strains—such as C57BL/6—develop lesions when placed on a high-cholesterol diet (often in combination with cholic acid). Studies in these mice have explored the role of specific mediators in atherogenesis and the genetic differences between atherogenesis-prone and resistant strains. However, lesions in this model do not progress beyond “fatty streaks.” The advent of genetically manipulated murine models that recapitulate more of the morphological features of seen in advanced human atheroma has considerably enhanced such investigation. The two most widely studied murine models of atherosclerosis are ApoE deficient (ApoE–/–) and low-density lipoprotein receptor deficient (LDLR–/–) mice. ApoE–/– mice fed a “Western” diet (21% fat by weight/40% by calories, 1.25% cholesterol) develop marked hyperlipidemia most consistent with Type III hyperlipidemia (rare in humans). On a standard chow diet, ApoE–/– mice develop fatty streaks lesions after approx 10 wk and these lesions progress, although the process is substantially accelerated on a Western diet. The complex plaques seen in this model are reminiscent of the human disease, although thrombosis is not generally described. LDLR–/– were initially developed as a model of familial hypercholesterolemia. In contrast to ApoE–/– mice, LDLR–/– mice do not develop significant atherosclerotic lesions on a normal chow diet, thus facilitating kinetic studies of atherogenesis. However, complex lesions are less common in LDLR–/– mice fed a Western diet and some investigators use supplemental, dietary cholic acid, which may have confounding pro-inflammatory effects. Both ApoE–/– and LDLR–/– models of atherosclerosis have been extensively characterized and although each has its potential advantages and advocates, results in both models have generally been concordant. Other models (e.g., transgenic overexpression of apoB100) provide important

126

SECTION II / CARDIOLOGY

insights into the role of specific lipoproteins but have been less extensively characterized. In virtually all models of atherosclerosis, atheroma develops, preferentially in specific regions marked by disturbed flow and increased expression of pro-inflammatory endothelial cell effectors. In mice, these regions exhibit enhanced expression and increased activation of the family of transcription factors, nuclear factor-κB (NF-κB), which act as key regulators of the expression of genes modulating inflammation and survival. The development of these models has enabled investigators to exploit the power of murine germline genetics to analyze the role of specific modifier loci as well as putative effectors in atherogenesis. By breeding atherogenic murine lines (e.g., ApoE–/– and LDLR–/–) to different in-bred genetic backgrounds or mice engineered to lack or overexpress specific candidates, many useful insights into the role of these pathways in atherogenesis have been generated. However, these models do not generally recapitulate the biology of plaque rupture associated with unstable human lesions. A tacit hypothesis underlying much work in this field is that interventions that mitigate the overall burden of atherosclerosis also favorably modify the tendency to plaque rupture. This has been true clinically with some therapies such as statin treatment, which improve serum lipid profiles, inhibit lesion progression, and even more dramatically reduce the rate of unstable coronary syndromes. Nevertheless, the development of animal models that accurately reflect these processes—particularly in species amenable to genetic analysis—would further facilitate investigation into the biology of the unstable plaque and the events leading to ACSs. ROLE OF LIPIDS Autopsy studies have shown that fatty streaks frequently exist in the coronary arteries and aortae of teenagers. These lesions may constitute the earliest recognizable precursor of atherosclerotic plaques, although it is formally possible that they either do not progress or actually regress with time. The major lipid component of these lesions is oxidized LDL. In vitro experiments have shown that oxidized LDL stimulates the adherence of monocytes to vascular endothelium most likely through increased expression of endothelial adhesion molecules. Oxidized LDL also stimulates transcription and secretion of monocyte chemotactic protein (MCP)-1 by human aortic and SMCs in vitro. MCP-1 is a powerful chemoattractant for monocytes and memory T cells in vitro, which are the predominant leukocyte populations in atherosclerotic lesions. Lysophosphatidylcholine, which constitutes a significant fraction of oxidized LDL, is a chemoattractant for monocytes and also induces expression of the endothelial adhesion molecules vascular cell adhesion molecule (VCAM)-1 and intercellular adhesion molecule (ICAM)-1. In addition, oxidized LDL can stimulate platelet aggregation and promote procoagulant activity on the surface of macrophages by an increase in tissue thromboplastin activity and by stimulating the expression and secretion of tissue factor by monocytes or aortic endothelial cells. Finally, oxidized LDL may also contribute to the vasomotor dysfunction that can promote or exacerbate the atherosclerotic lesion. Therefore, in many patients with atherosclerosis, lipids (particularly oxidized lipids) likely constitute an early and persistent precipitant of endothelial activation and dysfunction. SMOOTH MUSCLE CELLS Abnormal growth of vascular SMCs is prominent in atherosclerosis. At least two phenotypes of the SMCs make up the vascular wall, based on examination of myosin filaments and details of the secretory protein apparatus. When cells are in a contractile phenotype, they respond to elements that promote vasoconstriction or vasodilation such as

Table 14-1 Cellular and Molecular Mediators of Atherosclerosis Factor Growth factors PDGF bFGF M-CSF VEGF Cytokines IL-1 TNF-α IFN-γ Chemokines (MCP-1) Vasoactive substances Nitric oxide Endothelin Prostaglandin

Source

Target

EC, WBC EC, SMC, WBC EC SMC

SMC EC, SMC WBC

EC, SMC, WBC EC, SMC, WBC WBC EC, SMC, WBC

EC, WBC, SMC EC, WBC, SMC EC, WBC, SMC WBC

EC EC EC

SMC, WBC SMC SMC

bFGF, basic fibroblast growth factor; EC, endothelial cell; IFN, interferon; IL, interleukin; MCP, monocyte chemotactic protein; M-CSF, monocyte-colony stimulating factor; PDGF, platelet-derived growth factor; SMC, smooth muscle cells; TNF, tumor necrosis factor; VEGF, vascular and endothelial growth factor; WBC, white blood cells.

endothelin, angiotensin II, prostaglandin I2, or nitric oxide. A second synthetic phenotype has also been identified. In this state, SMC synthesize numerous growth factors and their receptors, cytokines, as well as matrix proteins. Multiple factors may be able to stimulate SMC proliferation and migration from the vessel media, through the internal elastic membrane. These activated SMCs, in turn, release substances that act in complex paracrine and autocrine manners, involving vascular endothelial and SMCs, as well as blood-borne elements such as white cells and platelets (Table 14-1 and Fig. 14-2). A particularly dramatic example of this process occurs after mechanical injury of the vessel during angioplasty, but a similar process may contribute to atherogenesis. A central issue in the investigation of atherogenesis is to define the source(s) of the critical factors that initiate and perpetuate the shift in SMC phenotype. Of note, a substantial advance in the management of CAD has been the development of coronary artery stents that release pharmacological agents, providing mechanical support and a local biological intervention, termed drug-eluting stents. Drug-eluting stents appear to dramatically reduce the rate of restenosis after coronary interventions. One of the preliminary successes of this approach has utilized stents that release rapamycin, an inhibitor of the signaling molecule mammalian target of rapamycin (mTOR). mTOR plays an important role transducing signals to growth and proliferation pathways in a wide variety of cell types, including SMCs. However, whether inhibition of mTOR and/or modulation of these signaling pathways underlies the clinical success seen with rapamycin eluting stents has not been demonstrated. LEUKOCYTES Because of their early and consistent association with atherosclerotic lesions, mononuclear cells appear to be central to atherogenesis. The mononuclear cells in atherosclerotic lesions are predominantly monocytes and lymphocytes of the memory T-cell phenotype (CD45 RO+). Monocyte-derived foam cells are a major component of atheroma, comprising up to 60% of cells found in the necrotic lipid core, and 10–20% of cells in the fibrous cap. Recruitment of monocytes by activated endothelium marks an early event in atherosclerotic lesion formation. Mononuclear cells serve as the progenitors of foam cells and a

CHAPTER 14 / ATHEROSCLEROTIC CORONARY DISEASE

potential source of growth factors and cytokines (Fig. 14-2). These chemical signals in turn may mediate intimal and smooth muscle hyperplasia and amplify recruitment of more leukocytes to the vessel wall. Thus, monocytes also appear to be important participants in the progression of atherosclerosis through effects on inflammation, extracellular matrix remodeling, and coagulation. Monocytes may be important in the plaque instability thought to underlie ACSs. More importantly, correlative studies are supported by evidence from multiple studies that mice genetically engineered to lack specific signals in monocyte recruitment and function are resistant to atherosclerosis. ENDOTHELIAL PROGENITOR CELLS Work in many fields has prompted a growing realization that adult cells and tissues have substantially greater plasticity and potential for regeneration than previously realized. In the context of vascular biology, circulating endothelial progenitor cells (EPCs) in animal models and humans have been identified. EPCs have the potential to incorporate into vascular endothelium in adult tissues, although little is understood about the molecular signals controlling this process. Intriguingly, the number of circulating EPCs appears to correlate inversely with established risk factors for atherosclerotic vascular disease, and is enhanced by treatment with statins. Moreover, the number of EPCs correlates better with vascular reactivity than did a cumulative index of established risk factors, leading to the hypothesis that reduced EPCs may be mechanistically connected to abnormal vasoreactivity because of a role in ongoing vascular renewal. Although intriguing, this hypothesis remains largely speculative and such correlative findings could simply reflect common underlying pathophysiology. Thus, although much is unknown about the biology of EPCs, it is possible that they play an important role in vascular homeostasis and could influence atherogenesis. Further characterizing this role and exploring the therapeutic potential of these observations is the focus of intense investigation. MOLECULAR SIGNALS Adhesion Molecules Based on the discussion earlier, it is understandable that many investigators have explored the mechanisms responsible for recruitment of leukocytes into the vessel wall. Findings have delineated a multistep sequence of events. Leukocyteendothelial interaction often begins with a relatively weak adhesive interaction manifested as leukocyte rolling. This rolling is largely mediated by the selectin family of adhesion molecules, which appear capable of rapid and reversible, though relatively weak, interactions with their counter-ligands. Conversion of leukocyte rolling to arrest and firm adhesion is likely a critical step in recruitment. The chemokine family of cytokines (discussed earlier) plays an important role in this conversion. Firm adhesion appears mediated predominantly by leukocyte integrins binding to endothelial adherence receptors of the immunoglobulin family, such as ICAM-1 and VCAM-1. Firm adhesion can then be followed by diapedesis and the cascade of pro-atherogenic events previously described. Studies combining atherosclerosis-prone mouse strains (e.g., ApoE–/–) and genetically engineered deficiency or mutation of specific adhesion molecules suggest important roles for VCAM-1 in atherosclerosis. Moreover, expression of VCAM-1 correlates well with recognized local and systemic risk factors for atherogenesis (Fig. 14-3). An effect of ICAM-1 ablation has also been noted, though appears less dramatic and consistent. Some studies suggest that P-selectin may individually play a larger role than E-selectin in atherosclerosis but work with doubly deficient mice suggests the two molecules likely act in a coordinate manner contributing to disease progression.

127

Figure 14-3 Expression of VCAM-1 at sites of atherogenesis. In the left panel, the aorta of a hypercholesterolemic rabbit has been stained with Oil-Red-O to reveal characteristic sites of fat deposition and atherogenesis (darker) involving areas of disturbed flow including the aortic arch and intercostal branch points. In the right panel, an autoradiograph of the same aorta using radiolabeled antibody to rabbit vascular cell adhesion molecule-1 (VCAM-1) demonstrates that these areas of fat deposition and atherogenesis correlate well with surface VCAM-1 expression. (Courtesy of Dr. Myron Cybulsky, Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto General Research Institute, Toronto, Canada.)

Chemoattractant Cytokines In addition to cell surface expressed adhesion molecules, smaller secreted proteins also appear to play an important role in atherogenesis. The chemoattractant cytokines, called “chemokines,” are proinflammatory cytokines postulated to contribute to the recruitment of specific mononuclear leukocyte subpopulations into atherosclerotic plaques. As noted, in addition to acting as chemoattractants, an important additional property of these molecules is their ability to convert leukocyte rolling under laminar flow conditions to arrest, a necessary prelude to diapedesis. Monocyte chemoattractant protein (MCP)-1 is a chemokine that acts specifically on monocytes and memory T cells, the two leukocyte populations most prevalent in atheroma. Expression of MCP-1 by vascular SMCs is induced in vitro by oxidized lipids. Furthermore, in situ studies show increased MCP-1 mRNA in vascular SMCs, mesenchymal intimal cells and macrophages in atherosclerotic lesions, as compared with normal vessels. These observational studies were substantially advanced by work in murine models demonstrating that ablation of MCP-1 or its receptor mitigated the development of atherosclerosis. Another compelling candidate

128

SECTION II / CARDIOLOGY

for a role in atherosclerosis signaling is the chemokine receptor CX3CR1, which is expressed on monocytes. Deletion of CXC3CR1 in atherosclerotic murine models severely retards atherogenesis, similar to deletion of MCP-1/CCR2. Equally compelling are data from three human studies, including the Framingham Heart Study, where there is a clear association of a common receptor variant with lower prevalence of atheroscerosis. Importantly, this mutation confers diminished monocyte adhesion in functional studies thus bolstering the epidemiological results. Nuclear Factor-κB Most of the molecules previously discussed, including chemokines and adhesion molecules, are transcriptionally regulated under the control of the nuclear factor (NF)-κB family of transcription factors (p65 or RelA, p50, p52, BCl-3, c-Rel, and RelB). Interestingly, NF-κB is activated in the early stages of atheroma formation, as well as in advanced lesions. In addition, NF-κB also activates pro-coagulant (e.g., tissue factor) and proliferative targets (Cyclin D1) that could participate in atherogenesis. A striking correlation exists between modulation of local NF-κBactivation and atherosclerotic plaque formation. NF-κB activation in leukocytes from patients with unstable coronary syndromes has also been demonstrated. An intuitively appealing hypothesis is that NF-κB provides a critical master control point linking a variety of biological and mechanical stimuli to the expression of molecules demonstrated to play an important role in atherogenesis by regulating the transcription of these genes. This hypothesis raises the possibility that modulation of NF-κB could provide a therapeutic target in atherosclerosis. However, NF-κB-independent signaling pathways, such as p38 and JNK/SAPK, also appear important in endothelial activation. Moreover, NF-κB promotes a variety of potentially beneficial effects including endothelial survival, as well as expression of specific antioxidants. Finally, NF-κB signaling appears important in many different disease processes from fighting infection to cancer. Thus, the full ramifications of NF-κB signaling are complex, and given that a direct requirement for NF-κB activation in atherogenesis has not been demonstrated, consideration of its therapeutic potential should proceed with caution. Hemodynamic Forces Certain sites within the vasculature are more prone to develop atherosclerosis. They are often sites of disturbed flow such as vessel branch points. Hemodynamic forces induce various functional changes in vascular endothelium, including morphological changes and alterations in the production of important product such as coagulation factors, fibrinolytic factors, growth factors, and cytokines. One mechanism linking hemodynamic forces and endothelial gene expression likely involves a cisacting DNA element called the shear stress response element (SSRE) found in the promoters of several pathophysiologically relevant genes, including PDGF-A and ICAM-1. The SSRE is a functional NF-κB binding site in vascular endothelial cells under these conditions. This SSRE may represent a common pathway by which mechanical forces influence gene expression. Other adhesion molecules such as VCAM-1 or E-selectin, whose promoters lack this motif, are not upregulated under these conditions. Identification of the molecular basis of VCAM-1 upregulation under more complex hemodynamic conditions may provide insights into the differential regulation of these genes and their role in atherogenesis. Nitric Oxide and Atherosclerosis Nitric oxide (NO) is produced by three major isoforms of nitric oxide synthase (NOS) and plays a critical role in many physiological processes. In arteries, NO produced by endothelial NOS (eNOS) is a critical determinant of vascular tone. In addition, NO may inhibit atherogenesis through other effects including inhibition of SMC proliferation, platelet

aggregation, and leukocyte activation, adhesion, and emigration. NO donors inhibit NF-κB activation, although potentiation of NFκB-dependent genes has not been seen in mice lacking eNOS. In a variety of disease states including human and experimental atherosclerosis, effective eNOS activity is reduced, a condition often termed “endothelial dysfunction,” which typically develops before advanced anatomic lesions. Multiple mechanisms likely contribute to reduced eNOS activity in atherosclerosis, including relative deficiency of substrate (L-arginine) or cofactor (tetrahydrobiopterin or BH4), decreased eNOS activity and/or protein, enhanced degradation of NO, and increased endogenous inhibitor levels. In human atheroma, there is markedly decreased luminal endothelial eNOS expression, but increased expression in endothelial cells of the vasa vasorum. Endothelial dysfunction (reduced NO) occurs in ApoE–/– and LDLR–/– mice. Interestingly, NOS can produce not only NO but superoxide (O2–), which may enhance atherogenesis through increased oxidative stress and inactivation of NO. O2– production by eNOS appears primarily regulated by tetrahydrobiopterin levels and signaling through the Lox-1 receptor. Perhaps for these reasons, O2– production is increased in hypercholesterolemia and may contribute to endothelial dysfunction in atherosclerotic ApoE–/– mice. However, mice that are doubly deficient for eNOS and ApoE develop substantially more aggressive atherosclerosis, as well as aneurysms. Overall, these studies suggest NO production plays an important modulatory role in atherogenesis. It seems plausible that genetic variations in components in this pathway may well contribute to predisposition (or resistance) to atherosclerosis in humans and may represent therapeutic targets, though these concepts remain to be validated.

FUTURE DIRECTIONS Further genetic analysis of atherosclerosis as a complex trait will be imperative. Such studies should provide substantial insight into the significant proportion of clinical coronary risk that remains unattributable. Improved assessment of genetic risk for atherosclerosis would focus and facilitate efforts at primary prevention and intervention. Ultimately, the identification of the relative contribution of specific genes will also improve the understanding of the fundamental biology of atherogenesis. Although germline manipulation of mice has already provided important information, future studies employing high-throughput, unbiased genomic, or proteomic screens for important modulatory pathways could greatly increase the yield of this approach. The development of models recapitulating not only atherogenesis but plaque instability would also provide an important foundation for addressing this clinically important issue. The anatomic degree of stenosis is a poor predictor of biological behavior. Identification of markers of plaque instability and the ability to measure or image these in vivo would greatly improve the management of patients with CAD. Thus, although substantial advances in the understanding of atherogenesis have occurred, it remains to translate most of these to application in the clinical arena.

SELECTED REFERENCES Andersen HR, Nielsen TT, Rasmussen K, et al. A comparison of coronary angioplasty with fibrinolytic therapy in acute myocardial infarction. N Engl J Med 2003;349:733–742. Boring L, Gosling J, Cleary M, Charo I. Decreased lesion formation in CCR2–/– mice reveals a role for chemokines in the initiation of atherosclerosis. Nature 1998;394:894–897.

CHAPTER 14 / ATHEROSCLEROTIC CORONARY DISEASE

Braunwald E, Zipes DP, Libby P, (eds.) Heart Disease: A Textbook of Cardiovascular Medicine. 6th ed. Philadelphia: W. B. Saunders Company, 2001. Breslow JL. Mouse models of atherosclerosis. Science 1996;272:685–688. Collins T, Cybulsky MI. NF-kappaB: Pivotal mediator or innocent bystander in atherogenesis? J Clin Invest 2001;107:255–264. Colombo A, Orlic D, Stankovic G, et al. Preliminary observations regarding angiographic pattern of restenosis after rapamycin-eluting stent implantation. Circulation 2003;107:2178–2180. Cybulsky MI, Gimbrone MA Jr. Endothelial expression of a mononuclear leukocyte adhesion molecule during atherogenesis. Science 1991; 251:788–791. Davies MJ, Richardson PD, Woolf N, Katz DR, Mann J. Risk of thrombosis in human atherosclerotic plaques: Role of extracellular lipid, macrophage, and smooth muscle cell content. Br Heart J 1993;69: 377–381. Fazio S, Linton MF. Mouse models of hyperlipidemia and atherosclerosis. Front Biosci 2001;6:D515–525. Fuster V, Lewis A. Conner Memorial Lecture. Mechanisms leading to myocardial infarction: insights from studies of vascular biology. Circulation 1994;90:2126–2146. Gerszten RE, Garcia-Zepeda EA, Lim YC, et al. MCP-1 and IL-8 trigger firm adhesion of monocytes to vascular endothelium under flow conditions. Nature 1999;398:718–723. Gu L, Okada Y, Clinton SK, et al. Absence of monocyte chemoattractant protein-1 reduces atherosclerosis in low density lipoprotein receptordeficient mice. Mol Cell 1998;2:275–281. Hajra L, Evans AI, Chen M, Hyduk SJ, Collins T, Cybulsky MI. The NFkappa B signal transduction pathway in aortic endothelial cells is primed for activation in regions predisposed to atherosclerotic lesion formation. Proc Natl Acad Sci USA 2000;97:9052–9057. Haley KJ, Lilly CM, Yang JH, et al. Overexpression of eotaxin and the CCR3 receptor in human atherosclerosis: Using genomic technology to identify a potential novel pathway of vascular inflammation. Circulation 2000;102:2185–2189. Heeschen C, Dimmeler S, Hamm CW, et al. Soluble CD40 ligand in acute coronary syndromes. N Engl J Med 2003;348:1104–1111. Hill JM, Zalos G, Halcox JP, et al. Circulating endothelial progenitor cells, vascular function, and cardiovascular risk. N Engl J Med 2003;348: 593–600. Hirschhorn JN, Lohmueller K, Byrne E, Hirschhorn K. A comprehensive review of genetic association studies. Genet Med 2002;4:45–61. Iiyama K, Hajra L, Iiyama M, et al. Patterns of vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 expression in rabbit and mouse atherosclerotic lesions and at sites predisposed to lesion formation. Circ Res 1999;85:199–207. Kuhlencordt PJ, Gyurko R, Han F, et al. Accelerated atherosclerosis, aortic aneurysm formation, and ischemic heart disease in apolipoprotein E/endothelial nitric oxide synthase double-knockout mice. Circulation 2001;104:448–454. Lander E, Kruglyak L. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results . Nat Genet 1995;11: 241–247.

129

Lindahl B, Toss H, Siegbahn A, Venge P, Wallentin L. Markers of myocardial damage and inflammation in relation to long-term mortality in unstable coronary artery disease. FRISC Study Group. Fragmin during Instability in Coronary Artery Disease. N Engl J Med 2000;343:1139–1147. Linton MF, Farese RV Jr, Chiesa G, et al. Transgenic mice expressing high plasma concentrations of human apolipoprotein B100 and lipoprotein(a). J Clin Invest 1993;92:3029–3037. Lohmueller KE, Pearce CL, Pike M, Lander ES, Hirschhorn JN. Metaanalysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nat Genet 2003; 33: 177–182. Nakashima Y, Plump AS, Raines EW, Breslow JL, Ross R. ApoE-deficient mice develop lesions of all phases of atherosclerosis throughout the arterial tree. Arterioscler Thromb 1994;14:133–140. Park SJ, Shim WH, Ho DS, et al. A paclitaxel-eluting stent for the prevention of coronary restenosis. N Engl J Med 2003;348:1537–1545. Plump AS, Smith JD, Hayek T, et al. Severe hypercholesterolemia and atherosclerosis in apolipoprotein E-deficient mice created by homologous recombination in ES cells. Cell 1992;71:343–353. Quaini F, Urbanek K, Beltrami AP, et al. Chimerism of the transplanted heart. N Engl J Med 2002;346:5–15. Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med 2002;347:1557–1565. Ritchie ME. Nuclear factor-kappaB is selectively and markedly activated in humans with unstable angina pectoris. Circulation 1998;98:1707–1713. Rosenzweig A. Endothelial progenitor cells. N Engl J Med 2003;348: 581, 582. Ross R. Atherosclerosis—an inflammatory disease. N Engl J Med 1999;340: 115–126. Scandinavian Simvastatin Study Group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: The Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383–1389. Serruys PW, Degertekin M, Tanabe K, et al. Intravascular ultrasound findings in the multicenter, randomized, double-blind RAVEL (Randomized study with the sirolimus-eluting Velocity balloonexpandable stent in the treatment of patients with de novo native coronary artery Lesions) trial. Circulation 2002;106:798–803. Shi W, Haberland ME, Jien ML, Shih DM, Lusis AJ. Endothelial responses to oxidized lipoproteins determine genetic susceptibility to atherosclerosis in mice. Circulation 2000;102:75–81. Smith JD, Trogan E, Ginsberg M, Grigaux C, Tian J, Miyata M. Decreased atherosclerosis in mice deficient in both macrophage colony-stimulating factor (op) and apolipoprotein E. Proc Natl Acad Sci USA 1995;92:8264–8268. Vasa M, Fichtlscherer S, Adler K, et al. Increase in circulating endothelial progenitor cells by statin therapy in patients with stable coronary artery disease. Circulation 2001;103:2885–2890. Zhang SH, Reddick RL, Piedrahita JA, Maeda N. Spontaneous hypercholesterolemia and arterial lesions in mice lacking apolipoprotein E. Science 1992;258:468–471.

15 Lipid Metabolism and Coronary Artery Disease MASON W. FREEMAN SUMMARY The treatment of patients with lipid disorders constitutes one of the most common interventions in adult medical practice and remains the centerpiece of strategies aimed at reducing the incidence of coronary heart disease. This chapter examines lipoproteins, the genetic basis of lipid disorders, such as monogenic low-density lipoproteins disorders and monogenic high-density lipoproteins disorders, and the management of coronary heart disease with sections about risk assessment as a guide to selection of therapy and the dietary and drug treatment of lipid disorders. Key Words: Chylomicrons; coronary heart disease (CHD); high-density lipoproteins (HDL); intermediate-density lipoproteins (IDL); lipid disorders; low-density lipoproteins (LDL); very low-density lipoproteins (VLDL).

INTRODUCTION The relationship between alterations in serum lipoprotein levels and heart disease risk was recognized more than 50 yr ago. Since then, remarkable expansion has occurred in the knowledge of the genetics, molecular, and cellular biology, and physiology of the lipoprotein metabolism system, leading to large-scale clinical trials of lipid-lowering drugs to test their efficacy in preventing acute coronary syndromes. The value of this approach has been repeatedly affirmed. As a result, the treatment of patients with lipid disorders constitutes one of the most common interventions in adult medical practice and remains the centerpiece of strategies aimed at reducing the incidence of coronary heart disease (CHD).

LIPOPROTEINS To circulate in the aqueous environment of the blood, highly hydrophobic lipids, such as cholesterol ester and triglyceride, must be packaged so they can be transported from sites of synthesis to sites of utilization. The lipoprotein structure solves this packaging problem by complexing the neutral lipids with both proteins and more polar lipids, such as unesterified cholesterol and phospholipids. The more hydrophobic lipid is sequestered on the inside of the spherical lipoprotein, whereas more polar lipids and proteins decorate the outer surface (Fig. 15-1). The protein components of From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

the lipoproteins, known as apoproteins, play structural and functional roles in the metabolism of lipoprotein particles. They determine the specificity of receptor interactions by lipoproteins and also affect their regulation by enzymes involved in lipid transfer or degradation processes. Mutations in the genes encoding apoproteins are rare, but they can have a profound influence on the circulating levels of lipoproteins and, consequently, on the risk of developing cardiovascular disease. Lipoprotein particles were originally classified using density gradient separation by ultracentrifugation. In the 1960s and 1970s, attention was focused on the electrophoretic properties of lipoproteins, giving rise to alternative names such as α- and β-lipoproteins, referring to the migration of those particles in an electrophoretic field. Since then, however, the fields have reunified and the older density nomenclature has been retained. Although lipoproteins form a continuous spectrum of density within the plasma, they can be divided into five major classes: chylomicrons, very low-density lipoproteins (VLDL), intermediate-density lipoproteins (IDL), low-density lipoproteins (LDL), and high-density lipoproteins (HDL) (see Fig. 15-1). Further subdivisions of these major lipoprotein classes have been proposed, especially of HDL, but, these additional complexities have not been widely assimilated in clinical practice. The least dense of the lipoproteins are also the largest in size and the most triglyceride-rich, the chylomicrons and VLDL. IDL contain substantial amounts of triglyceride and cholesterol, whereas the LDL and HDL are predominantly cholesterol-carrying particles. CHYLOMICRONS Chylomicrons derive from dietary fat and carry triglycerides throughout the body. They have the lowest density of all lipoproteins and will float to the top of a plasma specimen refrigerated overnight. The chylomicron itself is probably not atherogenic, but the atherosclerotic role of the triglyceridedepleted chylomicron remnant remains controversial. Triglyceride makes up most of the chylomicron and is removed by the action of an enzyme that is bound to the surface of endothelial cells, lipoprotein lipase. Patients deficient in this enzyme or its cofactors (insulin and apo C-II) have very high serum triglyceride levels and are at increased risk of acute pancreatitis. VERY LOW-DENSITY LIPOPROTEINS VLDL are also triglyceride-rich and are acted on by lipoprotein lipase. Their function is to carry triglycerides synthesized in the liver and intestines to capillary beds in adipose tissue and muscle, where they are

130

CHAPTER 15 / LIPID METABOLISM AND CORONARY ARTERY DISEASE

131

Figure 15-1 Lipoprotein classification and structure. Plasma lipoproteins can be divided into five density classes. The lightest density (chylmicrons and VLDL) are the most triglyceride-rich. LDL and HDL are cholesterol-rich. The general structure of a lipoprotein is shown in the lower half of the figure. More hydrophobic lipids are packaged on the inside of the lipoprotein whereas more polar lipids and the protein constituents are on the outside, where they can interact with the aqueous environments of blood and interstitial fluid.

hydrolyzed. After removal of their triglyceride, VLDL remnants (called IDL) can be further metabolized to LDL. VLDL serve as acceptors of cholesterol transferred from HDL, accounting in part for the inverse relation between HDL cholesterol and VLDL triglyceride. This transfer process is mediated by an enzyme called cholesterol ester transfer protein. LOW-DENSITY LIPOPROTEINS LDL are the major carriers of cholesterol in humans, responsible for supplying cholesterol to tissues with the highest sterol demands. LDL are also the lipoproteins most clearly implicated in causing atherogenic plaque formation. Circulating LDL levels can be increased in persons who consume large amounts of saturated fat and/or cholesterol. LDL levels are also elevated in those who have genetic defects that affect LDL receptor function (familial hypercholesterolemia [FH] and autosomal-recessive hypercholesterolemia [ARH]), the structure of LDL’s apoprotein, apolipoprotein B-100 or who have polygenic disorders affecting LDL metabolism. When serum LDL exceed a threshold concentration, they traverse the endothelial wall and can become trapped in the arterial intima, where they may undergo oxidation or other biochemical modification, be taken up by macrophages, and stimulate atherogenesis. The association of total serum cholesterol with CHD is predominantly a reflection of the role of LDL. HIGH-DENSITY LIPOPROTEINS HDL are believed to function to protect tissues from the unwanted accumulation of cholesterol. They participate in a reverse cholesterol transport pathway in which peripheral tissues efflux cholesterol back to lipid-poor forms of HDL for return to the liver (Fig. 15-2). There, cholesterol can be secreted into the bile. The unesterified cholesterol from tissues that is transferred to HDL is esterified by the action of lecithin cholesterol acyl transferase (LCAT) and stored in the central core of HDL. This esterified cholesterol can be transferred back to lower density lipoproteins by the action of

cholesterol ester transfer protein or it may be removed at the liver by the action of a plasma membrane receptor called scavenger receptor B-1. A particularly effective reverse transport system is thought to explain, at least in part, why elevated HDL levels are associated with a reduced risk of developing CHD. Apolipoprotein (apo)-A-I is the major apoprotein of HDL, and its serum concentration also correlates inversely with the risk of CHD. Women have higher levels of HDL cholesterol than men and this may partly explain the lower incidence of CHD in premenopausal women. Exercise increases HDL levels, whereas obesity, hypertriglyceridemia, and smoking lower them.

CLINICAL FEATURES In most individuals, alterations in serum lipoproteins produce no clinical manifestations until an atherosclerotic vascular event supervenes (e.g., a myocardial infarct, stroke, or peripheral vascular occlusion). Individuals with extremely elevated levels of serum triglycerides, because of VLDL, or chylomicron accumulation, or both, can develop pancreatitis. Such patients may first come to clinical attention because of an episode of pancreatitis or because their plasma is noted to be milky in appearance when routine blood tests are performed. Individuals with such marked triglyceride elevations (typically plasma triglyceride levels >1000 mg/dL) can also develop eruptive xanthomas, accumulations of lipid in the skin (Fig. 15-3). Individuals whose LDL levels are significantly elevated can also develop xanthomas, though these are typically located in either the Achilles tendon or extensor tendons of the hands. Individuals with accumulations of IDL can also develop xanthomas, which are classically of the tubero-eruptive or palmar type. Xanthelasmas, seen in individuals with elevated serum cholesterol levels, can often be detected in older individuals whose lipid levels are not dramatically increased. Individuals with some of the rare HDL disorders can

132

SECTION II / CARDIOLOGY

Figure 15-2 LDL and HDL metabolism. A schematized model of HDL and LDL metabolism. Cholesterol in a peripheral tissue cell, such as a macrophage, is initially stored in the cell as cholesterol ester. When stimulated by a lipid-poor HDL apoprotein A-I, the cell hydrolyzes the fatty acid from the cholesterol ester and the unesterified cholesterol is transported to the cell membrane. The action of the ABCA1 transporter results in the transfer of the free cholesterol to the apoprotein. Once bound to the apoprotein, the cholesterol is re-esterified by the action of LCAT creating the spherical HDL lipoprotein via storage of cholesterol ester in the core of the lipoprotein. HDL can then either donate this lipid back to LDL particles it encounters in the blood or travel to the liver in which its cholesterol ester content can be selectively removed by the action of scavenger receptor B-1. The cholesterol can be stored in the liver, reused for new lipoprotein synthesis, or excreted in the bile. Cholesterol that is transferred to LDL can be sent back to the periphery or the liver, in which LDL receptors bind and endocytose the particle. The action of the ARH gene product is important for LDL receptor activity in the liver. LDL originally derived from VLDL is secreted by the liver in a process not depicted in this slide.

Figure 15-3 Xanthomas. Several different types of xanthomas are shown. The palmar and tubero-eruptive xanthomas are classically seen in patients with increased concentrations of IDL whereas the eruptive xanthomas are typically found in patients with massive serum triglyceride elevations resulting from excess chylomicrons and/or VLDL. (Please see color insert.)

133

CHAPTER 15 / LIPID METABOLISM AND CORONARY ARTERY DISEASE

present with corneal opacities, tendon xanthomas (deposits typically embedded in extensor tendons), or a peripheral neuropathy. Rarely, a lipid abnormality may be detected because of abnormal liver function tests caused by steatosis of the liver or by lipid accumulation in the eye (corneal arcus). With regular measurement of lipids in adults, standard practice in the United States, lipid disorders are most commonly detected by routine laboratory screening tests.

DIAGNOSIS The diagnosis of a lipid disorder should be based on more than one measurement of serum lipids, because combined analytic and biologic variations in serum lipids range from 10 to 20%. The technology for measuring LDL levels directly has improved steadily, but in most laboratories, it remains a calculated value. To perform this calculation, the total and HDL cholesterol levels as well as the triglyceride value are measured. The LDL cholesterol concentration is then estimated using the following formula: LDL cholesterol = total cholesterol – [HDL cholesterol + triglyceride/5] The triglyceride/5 factor represents an estimate of VLDL cholesterol. The validity of this formula for estimating LDL cholesterol has been confirmed by ultracentrifugal measurement of lipoprotein levels and remains reasonably accurate as long as the total triglyceride is less than 400 mg/dL. To obtain an accurate calculation, patients must fast for at least 12 h to clear their blood of any chylomicrons, as these lipoproteins distort the triglyceride ratio on which the formula relies. If the triglyceride level is greater than 400 mg/dL, the LDL cholesterol must be determined by alternative methods. Apo B-100, present in both LDL and VLDL, can be measured directly to get an assessment of lower density lipoprotein particle numbers, but this is not routine in the United States. With increasing evidence for a greater atherogenicity of smaller, denser LDL particles, more sophisticated assessments of LDL number, and composition are being introduced into clinical practice in some regions of the world. Their place in the routine diagnostic evaluation of most patients remains controversial. Before embarking on a treatment plan, conditions must be excluded that could cause hyperlipidemia secondarily. The most common clinical conditions that influence lipids levels are obesity, diabetes, and hypothyroidism. The latter two are best screened using a serum glucose (or Hgb A1-c) measurement and a thyrotropin stimulating hormone level, respectively. Many medications commonly cause a secondary hyperlipidemia, with antiretroviral therapy, estrogens, and glucocorticoids heading the list. In the mid-1960s, a classifications scheme for lipid disorders was devised based on the phenotype of lipoprotein abnormalities. Because a better understanding of the genetics of these disorders has emerged that classification is rarely used. However, no unified classification of comparable simplicity has replaced it. Instead, the National Cholesterol Education Program (NCEP), Adult Treatment Panel III has promulgated a set of guidelines that are primarily aimed at targeting therapy in patients with elevated LDL levels or decreased HDL values. Table 15-1 summarizes the LDL-based treatment guidelines. Although the NCEP treatment targets for HDL and serum triglycerides are not as detailed as those for LDL values, HDL levels >40 mg/dL in men and 50 mg/dL in women are considered normal as is a serum triglyceride value 2 risk factors >160 10-yr risk 130a 10-yr risk >20%

150 different isoforms identified as products of >30 different P450 genes. Two notable variants are the CYP2D2 and N-acetyltransferase 2 variants. CYP2D2 mutations lead to excessive β-blockade of alprenolol, bufarolol, carvedilol, metoprolol, propranolol, and timolol presumably resulting from reduced metabolism of these blockers. Carriers of the “rapid accelerator” N-acetyltransferase 2 polymorphism require a higher dose of the vasodilator hydralazine to control blood pressure. Thus, genotyping of participants might allow tailoring of antihypertensive drug therapy. Such measurements are a necessary step toward studying the interaction between environmental factors such as age, diet and other medications that influence antihypertensive drug response. This ability to accurately identify pathophysiological subgroups and reliably predict the blood pressure response in individually tailored therapies will lead to fewer adverse reactions and ultimately lowered therapeutic costs. Human hypertension is a complex disease. With the exception of monogenic forms of hypertension, knowledge of essential hypertension and its therapy is severely restricted because of its polygenic nature and the compounding interaction between genes and the environment. With the use of the latest molecular technology of generating animal models, the sequencing of the human, mouse, and rat genomes, as well as the onset of the era of proteomics and genomics, this challenge might be overcome. Successful forays into the genetics of human hypertension should result in early intervention and more clinically and cost effective therapies that will limit complications of this common disease. The challenge for research is to develop “molecular profiles” of essential hypertension and drug response genes in order to develop individualized drug therapy. Drugs that are more specific for the molecular characteristics of individual patients should contribute to greater efficacy and reduced toxicity. With genetic screening of hypertensive patients as a possibility, and the availability of newer molecular techniques, there is the potential to revolutionize the way hypertension and its associated target organ diseases are diagnosed and treated.

SELECTED REFERENCES Bader M, Peters J, Baltatu O, Muller DN, Luft FC. Tissue renin-angiotensin systems: new insights from experimental animal models in hypertension research. J Mol Med 2001;79(2–3):76–102. Beckers A, Abs R, Willems PJ, et al. Aldosterone-secreting adrenal adenoma as part of multiple endocrine neoplasia type 1 (MEN1): loss of heterozygosity for polymorphic chromosome 11 deoxyribonucleic acid markers, including the MEN1 locus. J Clin Endocrinol Metab 1992;75(2):564–570. Burt VL, Whelton P, Roccella EJ, et al. Prevalence of hypertension in the US adult population results from the Third National Health and Nutrition Examination Survey, 1988–1991. Hypertension 1995;25(3): 305–313. Cardia SLR. Gene–environment interactions. In: Izzo JL, Black HR, eds. Hypertension Primer: Essentials of High Blood Pressure, 3rd ed. Baltimore: Lippincott Williams and Wilkins, 2002; pp. 221–224. Conn JW. Primary aldosteronism, a new clinical syndrome. J Lab Clin Med 1995;45:3–17. Conn JW, Knopf RF, Nesbit RM. Clinical characteristics of primary aldosteronism from an analysis of 145 cases. Am J Surg 1964;107: 159–172. Cooper RS. Geographic patterns of hypertension: a global perspective. In: Izzo JL, Black HR, eds. Hypertension Primer: Essentials of High Blood

CHAPTER 16 / HYPERTENSION

Pressure, 3rd ed. Baltimore: Lippincott Williams and Wilkins, 2002; pp. 231–233. Cusi D, Barlassina C, Azzani T, et al. Polymorphisms of alpha-adducin and salt sensitivity in patients with essential hypertension. Lancet 1997;349(9062):1353–1357. Doris PA. Hypertension genetics, single nucleotide polymorphisms, and the common disease: common variant hypothesis. Hypertension 2002; 39(2 Pt 2):323–331. Ferdinand KC. Hypertension in blacks. In: Izzo JL, Black HR, eds. Hypertension Primer: Essentials of High Blood Pressure, 3rd ed. Baltimore: Lippincott Williams and Wilkins, 2002; pp. 264–266. Hoffman MA, Ohh M, Yang H, Klco JM, Ivan M, Kaelin WG Jr. von HippelLindau protein mutants linked to type 2C VHL disease preserve the ability to downregulate HIF. Hum Mol Genet 2001;10(10): 1019–1027. Hrabe de Angelis MH, Flaswinkel H, Fuchs H, et al. Genome-wide, largescale production of mutant mice by ENU mutagenesis. Nat Genet 2000;25(4):444–447. Jackson RV, Lafferty A, Torpy DJ, Stratakis C. New genetic insights in familial hyperaldosteronism. Ann N Y Acad Sci 2002;970:77–88. Jeunemaitre X, Soubrier F, Kotelevtsev YV, et al. Molecular basis of human hypertension: role of angiotensinogen. Cell 1992;71(1): 169–180. Katki KA, Supowit SC, Li J, Wang DH, DiPette DJ. Subtotal nephrectomy-salt hypertension in a-CGRP knockout mice-role of substance P. 56th Annual Fall Conference and Scientific Sessions of the Council for High Blood Pressure Research of the American Heart Association September 25–28, 2002. Kato N. Genetic analysis in human hypertension. Hypertens Res 2002; 25(3):319–327. Kim HS, Krege JH, Kluckman KD, et al. Genetic control of blood pressure and the angiotensinogen locus. Proc Natl Acad Sci USA 1995; 92(7):2735–2739. Kreutz R, Hubner N. Congenic rat strains are important tools for the genetic dissection of essential hypertension. Semin Nephrol 2002; 22(2):135–47. Lifton RP, Dluhy RG, Powers M, et al. A chimaeric 11 betahydroxylase/aldosterone synthase gene causes glucocorticoid-remediable aldosteronism and human hypertension. Nature 1992;355(6357): 262–265. Maher ER, Eng C. The pressure rises: update on the genetics of phaeochromocytoma. Hum Mol Genet 2002;11(20):2347–2354. Manger WM, Gifford RW, Eisenhofer GF. Pathophysiology of pheochromocytoma. In: Izzo JL, Black HR, eds. Hypertension Primer: Essentials of High Blood Pressure, 3rd ed. Baltimore: Lippincott Williams and Wilkins, 2002, pp. 148–150. McNamara DM, MacGowan GA, London B. Clinical importance of beta-adrenoceptor polymorphisms in cardiovascular disease. Am J Pharmacogenomics 2002;2(2):73–78. Miller ER III. Geographic patterns of hypertension in the United States. In: Izzo JL, Black HR, eds. Hypertension Primer: Essentials of High Blood Pressure, 3rd ed. Baltimore: Lippincott Williams and Wilkins, 2002, pp. 233–235.

145

Nabel E. Genomic medicine: cardiovascular disease. N Engl J Med 2003; 349:60–72. Nolan PM, Peters J, Strivens M, et al. A systematic, genome-wide, phenotype-driven mutagenesis programme for gene function studies in the mouse. Nat Genet 2000;25(4):440–443. Pradervand S, Wang Q, Burnier M, et al. A mouse model for Liddle’s syndrome. J Am Soc Nephrol 1999;10(12):2527–2533. Rapp JP. Genetic analysis of inherited hypertension in the rat. Physiol Rev 2000;80(1):135–172. Rossier BC, Pradervand S, Schild L, Hummler E. Epithelial sodium channel and the control of sodium balance: interaction between genetic and environmental factors. Annu Rev Physiol 2002;64:877–897. Rubattu S, Struk B, Kreutz R, Volpe M, Lindpaintner K. Animal models of genetic hypertension: what can we learn for human hypertension? Clin Exp Pharm Physiol 1995;22(12):S386–S393. Shimkets RA, Warnock DG, Bositis CM, et al. Liddle’s syndrome: heritable human hypertension caused by mutations in the beta subunit of the epithelial sodium channel. Cell 1994;79(3):407–414. Simon DB, Karet FE, Hamdan JM, DiPietro A, Sanjad SA, Lifton RP. Bartter’s syndrome, hypokalaemic alkalosis with hypercalciuria, is caused by mutations in the Na-K-2Cl cotransporter NKCC2. Nat Genet 1996;13(2):183–188. Stoll M, Jacob HJ. Genetic rat models of hypertension: relationship to human hypertension. Curr Hypertens Rep 2001;3(2):157–164. Sugiyama F, Yagami K, Paigen B. Mouse models of blood pressure regulation and hypertension. Curr Hypertens Rep 2001;3(1):41–48. Sutherland DJ, Ruse JL, Laidlaw JC. Hypertension, increased aldosterone secretion and low plasma renin activity relieved by dexamethasone. CMAJ 1966;95(22):1109–1119. Tiret L, Rigat B, Visvikis S, et al. Evidence, from combined segregation and linkage analysis, that a variant of the angiotensin I-converting enzyme (ACE) gene controls plasma ACE levels. Am J Hum Genet 1992;51(1):197–205. Turner ST, Schwartz GL, Chapman AB, Boerwinkle E. Use of gene markers to guide antihypertensive therapy. Curr Hypertens Rep 2001; 3(5): 410–415. Vincent M, Samani NJ, Gauguier D, Thompson JR, Lathrop GM, Sassard J. A pharmacogenetic approach to blood pressure in Lyon hypertensive rats. A chromosome 2 locus influences the response to a calcium antagonist. J Clin Invest 1997;100(8):2000–2006. Ward R. Familial aggregation and genetic epidemiology of blood pressure. In: Laragh JH, Brenner BM, eds. Hypertension: Pathophysiology, Diagnosis, and Management. New York, NY: Raven Press Limited, 1990, pp. 81–100. Watson R, DiPette DJ. Experimental models of hypertension. In: Izzo JL, Black HR, eds. Hypertension Primer: Essentials of High Blood Pressure, 3rd ed. Baltimore: Lippincott Williams and Wilkins, 2002, pp. 120–122. Wright FA, O’Connor DT, Roberts E, et al. Genome scan for blood pressure loci in mice. Hypertension 1999;34(4 Pt 1):625–630. Yagil Y, Yagil C. Genetic models of hypertension in experimental animals. Exp Nephrol 2001;9(1):1–9.

17 Cardiac Hypertrophy THOMAS FORCE AND JEFFERY D. MOLKENTIN

SUMMARY Pathological cardiac hypertrophy develops in response to stresses, and can be concentric, eccentric, or both. An excess pressure load placed on the heart, for example, resulting from uncorrected hypertension or valvular disease, results in concentric hypertrophy. This hypertrophy is initially believed to be adaptive, normalizing systolic wall stress, though it is not clear that hypertrophy is necessary to maintain systolic function in the face of moderately elevated pressure loads. Eccentric hypertrophy results most often from volume loads such as those in valvular insufficiency. Finally, the hypertrophy that occurs in the remote noninfarcted myocardium, as part of the remodeling process following a myocardial infarction, may be both concentric and eccentric. Key Words: Cardiac hypertrophy; concentric; eccentric; mechanical deformation; mTOR; NFAT; protein synthesis; reprogramming.

INTRODUCTION Cells that are terminally differentiated, by definition, cannot undergo hyperplastic growth. Rather, normal growth of these cells, which include cardiomyocytes, is hypertrophic. This form of hypertrophy, termed physiological hypertrophy, is both concentric (characterized by addition of sarcomeres in parallel, leading to increased width of the myocyte) and eccentric (characterized by the addition of sarcomeres in series, leading to increased length of the myocyte). Pathological cardiac hypertrophy develops in response to stresses, and can be concentric, eccentric, or both. An excess pressure load placed on the heart, for example, resulting from uncorrected hypertension or valvular disease, results in concentric hypertrophy. This hypertrophy is initially believed to be adaptive, normalizing systolic wall stress, though it is not clear that hypertrophy is necessary to maintain systolic function in the face of moderately elevated pressure loads. Eccentric hypertrophy results most often from volume loads such as those in valvular insufficiency. Finally, the hypertrophy that occurs in the remote noninfarcted myocardium, as part of the remodeling process following a myocardial infarction (MI), may be both concentric and eccentric. If the load placed on the heart is not normalized (such as via effective antihypertensive therapy), the heart may continue to From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

hypertrophy, eventually leading to elevated filling pressures and the so-called diastolic heart failure. The hypertrophied heart may also begin to decompensate, leading to progressive dilatation, systolic dysfunction, and heart failure on that basis. Not surprisingly, left-ventricular hypertrophy is a significant risk factor for the development of heart failure, increasing the risk of this end point by 6- to 17-fold. Furthermore, within 5 yr of the first detection of left-ventricular hypertrophy, one-third of men and one-fourth of women are dead, usually from cardiac disease. Given the importance of hypertrophy as a cardiac risk factor, investigators have begun to identify the molecular pathways that regulate the cardiac hypertrophic response in an attempt to identify novel pharmacological targets of potential clinical relevance. The focus of these investigations has been on the cell surface receptors for agonists that are believed to trigger the hypertrophic response, such as receptors for angiotensin (Ang)-II, endothelin (ET)-1, and α- and β-adrenergic agents, and therapies directed at the Ang-II receptor have been effective in regressing hypertrophy and in reducing cardiovascular end points even in patients with diabetes. However, given the vast number of agents that have been reported to induce hypertrophy (see below), and the increasing evidence that hypertrophy is multifactorial in origin, focus has shifted to intracellular signaling pathways that may function as final common pathways necessary for the hypertrophic response, irrespective of the inciting stimuli. With the rapid advances in the development of small molecule inhibitors of components of these pathways that can be used in vivo, it is essential to understand the signaling networks that regulate hypertrophic growth. The hypertrophic response requires a dramatic reprograming of gene expression to upregulate gene products necessary for the growth of the cardiomyocyte. These include genes encoding contractile elements and proteins of the basic transcriptional and translational machinery that allow new protein production), and genes that encode proteins that remodel the extracellular matrix (e.g., matrix metalloproteases), allowing growth to proceed. Reprograming of gene expression occurs in response to specific growth signals that are generated from a multitude of sources and include soluble factors (growth factors and neurohormonal mediators), or biomechanical forces (stretch of the myocyte induced, for example, by an acute MI or an acute pressure load). To reprogram gene expression, these signals must be sensed at the cardiomyocyte membrane, and transmitted into the interior of the

146

CHAPTER 17 / CARDIAC HYPERTROPHY

cell, and eventually into the nucleus. This process is called signal transduction. This chapter introduces the field of signal transduction, specifically as it applies to hypertrophy of the heart, and describes how gene expression becomes reprogramed. Many growth factors trigger growth of cardiomyocytes, and an even larger number of signaling molecules mediate the growth response. The complexity of the field makes it impossible to cover all pathways involved in the hypertrophic response; two pathways are reviewed for which strong evidence exists implicating components of the pathway in both physiological and pathological hypertrophic responses: the calcineurin pathway and the phosphoinositide 3-kinase (PI3-K)/Akt pathway and its interactions with the mammalian target of rapamycin (mTOR) pathway, which has been implicated in cellular growth of a wide variety of species. In each case, the role of these pathways is evaluated in the rapidly growing field of mouse models of human disease. Finally, dysregulation of these pathways in the hearts of patients with advanced heart failure is reviewed, and potential targets for the treatment of heart failure are discussed.

STIMULI TRIGGERING THE HYPERTROPHIC RESPONSE Two types of stimuli are thought to trigger the hypertrophic response, mechanical deformation of the membrane (cell stretch) and growth-promoting ligands binding to their cognate receptors in the myocyte cell membrane, though mechanical deformation activates ligand/receptor interactions as well. MECHANICAL DEFORMATION Stretch of cardiomyocytes in culture leads to protein synthesis and a pattern of gene transcription that resembles the load-induced hypertrophic response in vivo. Stretch triggers these responses both via the direct activation of signaling molecules and by inducing the release of prohypertrophic agonists that appear to be of paramount importance in the maintenance of the response. The identity of the stretch “sensor” is not known. However, stretch may directly activate integrin signaling and stretch-activated ion channels, and either or both of these could be the sensor of membrane deformation. Heterotrimeric G proteins (Gq and Gi), which transduce signals from prohypertrophic factors released following stretch (see Humoral Factors), and possibly small G proteins, are activated so early after stretch that it is possible that these are also directly activated. One critical consequence of activation of these proximal mediators of the stretch response is an increase in cytosolic calcium concentration, necessary for activating calcineurin (discussed later). Only when the sensors are identified will the mechanisms be understood by which stretch activates cytosolic signaling pathways. In most biological response pathways, however, a stretch signal gradually accommodates so that the response is lost unless a new or greater stimulus is applied. Thus, additional factors are needed to maintain the response and to induce long-term hypertrophic growth, and these factors may be humoral. HUMORAL FACTORS One of the central tenets of hypertrophic signaling is that membrane deformation induces the release of growth factors that act in an autocrine or paracrine fashion to amplify hypertrophic responses. Ang-II was reported to be released by stretched cardiomyocytes in culture, and autocrine/paracrine effects of Ang-II were reported to be, at least in part, responsible for the hypertrophic response of cardiomyocytes to cell stretch. Ang-II activates prohypertrophic signaling pathways via two mechanisms, one triggered directly by the receptor and its associated heterotrimeric G protein, Gq (see below), and the other triggered by transactivation of growth factor receptors with intrinsic tyrosine

147

kinase activity, most notably the epidermal growth factor receptor. Ang-II induces a cytosolic calcium transient (and production of reactive oxygen species) that activates a metalloprotease that releases heparin-binding epidermal growth factor-like growth factor. This then binds to its receptor, activating additional signaling pathways not directly activated by the Ang-II receptor. A second group of prohypertrophic factors released by mechanical stretch are the IL-6 family of cytokines, including cardiotrophin-1. The insulin-like growth factor (IGF)-1 axis, including growth hormone, which acts in large part via inducing production of IGF-1, is the dominant regulator of normal postnatal growth of the mammalian heart, but IGF-1 may also play a role in pathological hypertrophy. In pathological states, IGF-1 is released in response to a variety of stimuli that induce remodeling including pressure overload and following MI. IGF-1 signaling is also activated in the remodeled hearts of patients with advanced heart failure. Many of the prohypertrophic peptides, including Ang-II, ET-1, and α-adrenergic agents bind to receptors that are linked to heterotrimeric G proteins (i.e., consists of three subunits, α, β, and γ) of the Gq family. These G proteins convert receptor activation into mobilization of intracellular signaling pathways, and are the initial trigger for downstream events. Overexpression specifically in the heart of the α-subunit of Gq (the subunit that activates most of the signaling pathways downstream of receptors coupled to Gq) or of an activated mutant of αq led to cardiac hypertrophy. Expressing a peptide that inhibited Gq-dependent signaling significantly limits the hypertrophic response to pressure overload in vivo. Conditional inactivation of the gene encoding the α-subunit of Gq (and the related G11) blunts the hypertrophic response. These studies confirm a critical role for Gq in hypertrophic signaling and in the hypertrophic response to pressure overload, and also support a central role for Gq-coupled receptors and their ligands in this process.

INTRACELLULAR SIGNALING PATHWAYS There are two essential features of hypertrophic growth: reprograming of gene expression and protein synthesis. Each is regulated by a series of intracellular pathways activated by events at the membrane as described above. Each component is considered separately, though they are inextricably intertwined. REPROGRAMMING OF GENE EXPRESSION To hypertrophy, the cardiomyocyte must upregulate the expression of a number of genes including those encoding components of the sarcomere and more specific growth-related and stress-induced genes. Other genes are downregulated as part of the response. Characteristic of the response is a re-establishment of a gene program that is often described as “embryonic” or “fetal” because several of the reexpressed genes are normally expressed in utero, but expression declines rapidly after birth. The genes induced by the hypertrophic response are often divided into three groups based on their time of expression: immediate early, intermediate, and late. Immediate early genes include the neurohormonal mediator, brain natriuretic peptide, and several stress-induced genes or genes involved in growth control. These include c-fos, c-jun, c-myc, egr-1, and heat shock protein 70. Intermediate response genes include atrial natriuretic peptide, angiotensinogen, and several sarcomeric components, β-myosin heavy chain (β-MHC) (and corresponding downregulation of α-MHC), myosin light chain-2, and skeletal α-actin (replacing cardiac α-actin). Late response genes include angiotensin converting enzyme and the Na/Ca exchanger.

148

SECTION II / CARDIOLOGY

Figure 17-1 Schematic representation of a signal transduction pathway linking the cell surface receptor to nuclear events. Hypertrophic stimuli sensed at the membrane lead via a variety of signal transducers (e.g., Gq) to the activation of a protein kinase at the top of a multitiered cascade of protein kinases. This kinase phosphorylates kinase 2 in the cascade that in turn phosphorylates kinase 3. Kinase 3 then phosphorylates one or more transcription factors that bind to promoter elements in the regulatory regions of various “hypertrophic response genes” and activate gene expression. This multitiered cascade serves both to amplify the signal (one kinase molecule at each level activates many molecules at the next level) and to prevent the need for kinase 1 to translocate to the nucleus each time it is activated.

Reprograming of gene expression requires signal transmission from the cell membrane, in which the initiating stimulus is generated, to the nucleus. This is generally accomplished by linear cascades of proteins, most commonly protein kinases (but also the protein phosphatase calcineurin [see below]), phosphorylating and activating one another in sequence, culminating in the phosphorylation, and activation of one or more transcription factors (Fig. 17-1). The transcription factors then bind to promoter elements, specific DNA sequences, usually of approx 6–12 basepairs in length, within the promoters of genes. Thus each transcription factor usually targets several genes. The net result of the activation of the entire set of genes is the hypertrophic response. To illustrate how gene expression is reprogramed, the calcineurin-nuclear factor of activated T-cells (NFAT) signaling pathway, implicated as a key regulator of stress-induced hypertrophic growth, is used (Fig. 17-2). However, first the mouse models used to evaluate the role of signaling pathways in various disease states are reviewed and their advantages and disadvantages are examined. Mouse Models of Cardiac Disease Two basic models are employed: transgenic models and so-called “knockouts.” Transgenics generally express a gene of interest specifically in the heart by expressing it under the control of a promoter that is only active in cardiac myocytes. The most commonly employed promoter is the α-MHC promoter. Expression from this promoter is constitutive. The gene of interest can encode the normal gene, a constitutively active mutant of the gene (e.g., a protein kinase that does not require activation by an upstream regulator), or a dominant inhibitory mutant of the gene (such a protein kinase that cannot be activated, interferes with the normal functioning of the pathway by binding to and sequestering upstream activators and downstream targets of the normal kinase). A number of different pathways have been implicated in regulating the hypertrophic response. However, many of these studies

relied exclusively on transgenic overexpression of a signaling molecule in the heart. Because the level of expression of the transgene is often many-fold higher than the level of the endogenous protein, producing nonphysiological levels of activation of normal downstream targets, caution in interpreting these studies is needed. In addition, gross overexpression of a transgene can lead to “cross talk” with other signaling pathways that are not normally regulated by that transgene. That said, many signaling pathways have fairly striking fidelity so that cross talk is often minimal and valuable information can be obtained using an overexpression approach. Transgenes have been expressed conditionally. That is, expression can be regulated by the investigator to occur at specific times. This approach may reduce the long-term adaptations that can occur with constitutive expression of a transgene. The optimal proof of a role for a specific signaling factor in hypertrophic growth is that a specific molecule is sufficient to induce the hypertrophic response using a transgenic approach and that the molecule must also be demonstrated as necessary for the response. On rare occasions this can be accomplished with pharmacological inhibitors but usually it is done with strategies leading to deletion of a gene of interest (e.g., the complementary studies of the role of Gq in hypertrophic growth previously discussed). This is a much more difficult task because deleting a gene is technically much more demanding than creating a transgenic animal. In addition, the approach is often compounded by embryonic or neonatal lethality if the gene of interest serves essential functions in the organism. This problem has been addressed by an ingenious method (e.g., in the studies on Gq) employing transgenic (or adenovirus-mediated) expression of a viral protein, the Cre recombinase, expression of which can be induced at desired times in the life of the mouse. Cre then excises the target gene producing a so-called conditional knockout.

CHAPTER 17 / CARDIAC HYPERTROPHY

149

Figure 17-2 Schematic representation of the calcineurin/nuclear factor of activated T cells (NFAT) pathway. Diagram shows the possible sources of Ca2+ necessary for the activation of calcineurin and negative regulators of NFAT nuclear localization (GSK-3 and JNKs). The possible contribution of Ca2+-independent inputs into calcineurin activation is also shown. See The Calcineurin-NFAT Signaling Pathway Reprograms Gene Expression for details.

Another problem with gene deletion strategies to study function of a gene is that many genes involved in hypertrophic signaling are members of multigene families and deletion of one is compensated for by the other family members. Crossing a mouse with one deleted gene with a second mouse with the other gene deleted can produce the desired knockout of all alleles within the family, but increases the likelihood of embryonic or neonatal lethality. Proving that a gene is necessary for a specific response, such as hypertrophy, remains an essential element in the evaluation of the role of a specific protein in that response. The Calcineurin-NFAT Signaling Pathway Reprograms Gene Expression The calcineurin-NFAT pathway can serve to illustrate how gene expression is reprogramed. This pathway (Fig. 17-2) in a number of different models regulates hypertrophic growth. Calcineurin (also known as protein phosphatase 2B) is a calcium-calmodulin-activated protein phosphatase. It is uniquely activated by sustained elevations in intracellular calcium. Calcineurin specifically dephosphorylates proteins previously phosphorylated on serine or threonine residues. Calcineurin consists of a 59–63 kDa catalytic subunit referred to as calcineurin A, a 19 kDa calcium binding protein called calcineurin B, and the calcium binding protein calmodulin. Three mammalian calcineurin A catalytic genes have been identified (α, β, γ) that are highly homologous to one another. The calcineurin Aα and Aβ gene products are expressed in a ubiquitous pattern throughout the body, whereas calcineurin Aγ expression is more restricted. Calcineurin catalytic activity is inhibited by the immunosuppressive drugs cyclosporine (Cs) A and FK506 through complexes with immunophilin protein. Thus, calcineurin plays a critical role in the regulation of T-cell reactivity and cytokine gene expression. Once activated, calcineurin directly dephosphorylates members of the NFAT family of transcription factors. When phosphorylated, NFATs are retained in the cytosol. Dephosphorylation by calcineurin exposes nuclear localization signals, promoting their translocation into the nucleus. Once in the nucleus, NFAT family

members participate in the transcriptional induction of various immune response genes. Calcineurin was identified as a hypertrophic signaling factor in the heart suggesting conservation in its function as a reactive signaling factor in multiple cell types. For example, overexpression of an activated form of calcineurin in the hearts of transgenic mice induced a profound hypertrophic response (two- to threefold increase in heart size) that rapidly progressed to dilated heart failure within 2–3 mo. These data demonstrated that calcineurin was sufficient to induce hypertrophy, and identified calcineurin as a potential causative factor associated with the transition to decompensation and heart failure. Investigation has focused on the evaluation of whether calcineurin is also necessary for the hypertrophic response to physiologically relevant stimuli. Hypertrophic agonists (e.g., phenylephrine, Ang-II, and ET-1) led to increased calcineurin enzymatic activity in cultured cardiomyocytes, and treatment of cultured neonatal cardiomyocytes with the calcineurin inhibitory agent CsA attenuated agonist-induced hypertrophy in vitro. Furthermore, calcineurin enzymatic activity and protein levels were upregulated in hearts from juvenile tropomodulin transgenic mice, a model of dilated heart failure. Similarly, increased cardiac calcineurin activity has been reported in hearts in which hypertrophy was induced by aortic banding, exercise, or salt feeding of salt-sensitive hypertensive rats. Other studies have examined the activity of calcineurin in samples from failing or hypertrophied human hearts. Analysis of hypertrophied human hearts or failing hearts because of ischemic or idiopathic dilated cardiomyopathy revealed a significant increase in calcineurin activity. Studies in patients with hypertrophic obstructive cardiomyopathy and aortic stenosis demonstrated a significant increase in cardiac calcineurin activity associated with a differentially processed form of the calcineurin catalytic subunit in the heart, presumably because of partial proteolysis. Collectively, these observations have demonstrated a link between cardiac hypertrophy and failure and the activation of a pivotal reactive signaling molecule in the heart, calcineurin.

150

SECTION II / CARDIOLOGY

Two approaches have been employed to determine whether calcineurin is necessary for the hypertrophic response in vivo, pharmacological inhibition with CsA or FK506 and gene targeting. Although CsA can attenuate agonist-induced cardiomyocyte hypertrophy in vitro, its effectiveness in vivo is somewhat more controversial, particularly when the hypertrophy is induced by pressure overload. Approximately 20 individual reports have shown that inhibition of calcineurin with either CsA or FK506 can antagonize cardiac hypertrophy and/or disease progression in a variety of rodent models. Thus, the majority of pharmacological animal studies support a role for calcineurin in the hypertrophic response, and the few negative accounts may reflect factors such as drug dosage, differences in the surgical preparations, sex, age, or type of animal model. As with most pharmacological inhibitors, however, CsA and FK506 have targets other than calcineurin and inhibition of these targets, rather than calcineurin, could in theory account for the observed effects of the drugs on hypertrophy. To address the issue of specificity, targeted inhibition of calcineurin was achieved by creating a transgenic mouse overexpressing endogenous inhibitors of calcineurin. In one set of studies, the noncompetitive calcineurin inhibitory domains from the calcineurin interacting proteins, Cain/Cabin-1 and AKAP79, were expressed and in the other, muscle-enriched calcineurin inhibitory proteins MCIP1 and MCIP2 (DSCR1 and ZAKI-4) were expressed. Both models had reduced hypertrophy in response to pressure overload. In addition, transgenic mice expressing a dominant inhibitory mutant of calcineurin within the heart also demonstrated reduced cardiac hypertrophy to stress stimuli (aortic banding). Mice deleted for calcineurinAβ were generated and had reduced cardiac calcineurin activity that was associated with an impaired hypertrophic response to Ang-II infusion, isoproterenol infusion, or abdominal aortic constriction. These data extend the transgenic approaches discussed, and more specifically implicate the calcineurinAβ gene in regulating the hypertrophic response. Calcineurin Targets Regulating the Hypertrophic Response The downstream transcriptional mechanisms whereby calcineurin might function in vivo remain largely uncharacterized. However, both NFAT and myocyte enhancer factor 2 transcriptional regulators are regulated by calcineurin (see Fig. 17-2), suggesting obvious candidates for genetic analysis in the heart. The NFAT family consists of five members, four of which (NFATc1-c4) are regulated by calcineurin. Calcineurin activation leads to dephosphorylation and nuclear translocation of NFATs where they activate gene expression. To identify the candidate downstream effectors of calcineurin that might mediate the hypertrophic response, NFATc3 and NFATc4 null mice were evaluated. Remarkably, NFATc3 null mice, but not NFATc4, were determined to have impaired hypertrophy induced by activated calcineurin, abdominal aortic constriction, or Ang-II infusion. These data suggest that NFATc3 functions as a necessary transducer of calcineurin signaling in mediating the cardiac hypertrophic response. Collectively, analysis of multiple geneticallymodified mouse models with altered calcineurin-NFAT signaling supports that calcineurin is an important regulator of the cardiac hypertrophic response. The calcineurin/NFAT pathway illustrates the paradigm of how a signal (increased cytosolic [Ca2+]) generated in response to deformation of the membrane or to hypertrophic agonist binding to its receptor, activates one signaling

factor (calcineurin), which then dephosphorylates and activates a transcription factor (NFATc3) that, in turn, translocates to the nucleus, reprograms gene expression and, in so doing, regulates hypertrophic responses. REGULATION OF PROTEIN SYNTHESIS The second key component of hypertrophic growth is the ability to dramatically upregulate protein synthetic capabilities. This is regulated by two complex, interacting pathways: the mTOR pathway, and the PI3-K pathway that, in addition to its role in regulating protein synthesis, also functions in reprograming gene expression. These pathways are essential in determination of cell, organ, and body size (i.e., normal growth) in species as diverse as Drosophila and humans. The pathways are also recruited in, and regulate the response to, pathological stress-induced hypertrophic growth (Fig. 17-3). Both the mTOR and PI3-K pathways regulate protein synthesis by modulating the activity of various translation factors, either initiating factors (that initiate the translation of mRNAs into proteins) or elongation factors (responsible for elongation of the polypeptide chain). The mTOR Pathway mTOR is a protein kinase whose importance is suggested by its conservation in structure and function, throughout evolution, from yeast to human. Rapamycin, a specific inhibitor of mTOR, attenuates the hypertrophic response to pressure overload in mice, demonstrating the importance of mTOR in regulating hypertrophic growth in vivo. Although mTOR is absolutely critical in the regulation of protein synthesis, and studies employing rapamycin have identified specific targets of mTOR that regulate protein synthesis, the mechanisms regulating mTOR activity remain ill-defined. In brief, mTOR is activated when amino acids and energy supplies are plentiful (see Fig. 17-3). Teleologically this makes sense because protein translation is an enormous consumer of cell energy, so translation proceeding at times of amino acid or energy deprivation is undesirable. In addition, growth factors, including those leading to cardiac growth such as insulin and IGF-1, also activate mTOR probably via activation of PI3-K signaling. One of mTOR’s major targets is a protein, 4E-binding protein 1, which binds to and inactivates the eukaryotic initiation factor (eIF) 4E, preventing the initiation of translation. When activated, mTOR leads to the phosphorylation of 4E-binding protein 1, causing it to dissociate from eIF4E, thus allowing translation to proceed (see Fig. 17-3). A second target activated by mTOR, acting in cooperation with the PI3-K pathway, is the p70S6 kinase that phosphorylates the S6 protein of the small ribosomal subunit. p70S6K may regulate the translation of a specific set of mRNAs, the so-called 5′-TOP (tract of pyrimidines) mRNAs that encode ribosomal proteins. In addition, it may regulate one of the elongation factors, eEF2. The importance of p70S6K in cell and organ growth is illustrated by the marked reduction in cell, organ, and body size in mice deleted for even one of the two p70S6K genes. Furthermore, p70S6K is activated by pressure overload, and it is likely that the ability of rapamycin to block pressure overload hypertrophy is, in part, because of the inhibition of p70S6K activation by rapamycin. The PI3-K Pathway The PI3-K pathway is also highly conserved throughout evolution (see Fig. 17-3). This pathway is remarkable because virtually every component has been shown in animal models in vivo to regulate cell and organ growth, including growth of the heart. The pathway is activated by most (if not all) of the agonists implicated in inducing cardiac hypertrophy including pressure overload. When activated, the PI3-K phosphorylates

CHAPTER 17 / CARDIAC HYPERTROPHY

151

Figure 17-3 Schematic representation of the PI3-K and mTOR pathways. See the P13-K Pathway for details. rpS6, S6 ribosomal protein; eEF2K, the eukaryotic elongation factor 2 kinase that phosphorylates and inactivates eEF2 (eEF2K is inhibited by p70S6K).

the membrane phosholipid, phosphatidylinositol, at the 3′ position of the inositol ring, leading to the recruitment of the protein kinase, Akt (also known as protein kinase B [PKB]) to the cell membrane via interactions of a specific domain of PKB/Akt (the plekstrin homology domain) with the 3′-phosphorylated lipid. This brings PKB/Akt into proximity to its activator, the 3′-phosphoinositidedependent protein kinase-1, which phosphorylates and activates PKB/Akt (in addition to the p70S6K discussed). PKB/Akt then plays a role in activating mTOR and, consequently, p70S6K and the protein translation machinery. Hence, it is probably not surprising that PI3-K, phosphoinositide-dependent protein kinase-1, and PKB/Akt all regulate cell and organ size, including size of the heart. One of the more striking cardiac-specific transgenic models is the mouse overexpressing PKB/Akt which has a markedly enlarged heart. Although activation of the translational machinery is probably an important mechanism driving the increased heart size in these mice, PKB/Akt also regulates transcription factors, either directly (e.g., members of the Forkhead family) or indirectly (via regulation of glycogen synthase kinase-3 [GSK-3]) that likely play a role in reprogramming gene expression in response to hypertrophic stress. PKB/Akt regulates GSK-3, which also plays a role in regulating normal and pathological stress-induced growth of the heart. GSK-3β (and likely GSK-3α as well) is a negative regulator of cardiac growth. Transgenic animals expressing GSK-3β have dramatic reductions in normal cardiac growth and also have a markedly reduced hypertrophic response to pressure overload. Inhibition of GSK-3β, mediated by PKB/Akt, is necessary for the hypertrophic response to a number of agonists. GSK-3β-mediated inhibition of growth occurs via several mechanisms. For example, GSK-3β negatively regulates activity of the initiation factor, eIF2B. Thus inhibition of GSK-3β may be important for enhancing protein translation. However, GSK-3β also inhibits the activity of a number of transcription factors. Because several factors have been implicated in growth, including growth of the heart, GSK-3β may be particularly important in the reprograming of gene expression and relatively less important in regulating protein translation. This might make GSK-3β an attractive target for therapeutic intervention because

therapies targeting the translation machinery can be expected to have significant toxicity with long-term use. The GSK-3β targets include the NFATs, and thus GSK-3β acts in opposition to the calcineurin pathway (Fig. 17-2). Phosphorylation of NFATs by GSK-3β leads to exclusion of NFATs from the nucleus, preventing access to target genes. When the GSK-3β transgenic was bred with the calcineurin transgenic, hypertrophy was markedly reduced. Other known growth regulators negatively regulated by GSK-3β include c-Myc, GATA-4, β-catenin, and c-Jun. The PI3-K pathway is negatively regulated by a phosphatase that dephosphorylates 3-phosphorylated phosphoinositides at the 3′ position called phosphatase and tensin homolog (PTEN) (see Fig. 17-3). Overexpression of a catalytically inactive mutant of PTEN in cultured cardiomyocytes led to cardiomyocyte hypertrophy. Furthermore, conditional inactivation of the PTEN gene in the heart also led to hypertrophy, further supporting a critical role for the PI3-K pathway in regulating cardiomyocyte growth. In summary, the consistency of the message from studies of multiple components of the PI3-K pathway has confirmed a role for this pathway in normal growth and in pathological stress-induced hypertrophic growth. This pathway, acting in concert with the mTOR pathway, is a dominant determinant of cell and organ size in mammals. OTHER PATHWAYS INVOLVED IN GROWTH REGULATION As suggested, a large number of signaling pathways have been implicated in the regulation of cardiomyocyte growth. However, either there are insufficient data to support a claim, or the data are too conflicting to make definitive statements concerning their role. This is probably most apparent from the literature on the role of stress-activated mitogen-activated protein (MAP) kinases in the hypertrophic response (Fig. 17-4). These kinases, the c-Jun N-terminal kinases (JNKs) and the p38-MAP kinases, have been exhaustively studied but there remains no consensus on their role, in large part owing to the lack of adequate pharmacological inhibitors that can be used in vivo. As a result, investigators have turned to gene-targeting approaches, but because both of these kinases are members of multigene families, studies have been complicated by functional redundancy, and crosses of knockouts have led to embryonic lethality. It is likely

152

SECTION II / CARDIOLOGY

Figure 17-4 Schematic representation of the major MAPK pathways. The three-tiered kinase cascade consists of a MAP kinase kinase kinase (MAP3K) that phosphorylates and activates a MAP kinase kinase (also called MAPK and ERK kinase, MEK), that, in turn, phosphorylates the MAP kinases (MAPK), either the ERKs, JNKs, or p38-MAPKs.

that a definitive conclusion on the role of these kinases in hypertrophic growth awaits the availability of truly specific inhibitors of these kinases, or better yet, inhibitors of specific isoforms of these kinases, because there is evidence that different isoforms (and even different splice variants) may serve different functions within the cell. Assessment of the data suggests that the p38-MAP kinases may be more involved in the progression of heart failure, including remodeling of the matrix, and may play less of a role in hypertrophic growth. If confirmed, they may be attractive drug targets in the failing heart. Other work employing mice deleted for three of the four JNK1/JNK2 alleles suggests that the JNKs may reduce the hypertrophic response to pressure overload, in part by targeting NFATs much as GSK-3β does. The extracellular signal-regulated kinases family (ERK) of MAP kinases (see Fig. 17-4) are also activated by hypertrophic stimuli. For studies of the role of the ERKs in hypertrophy, relatively specific inhibitors of the mitogen-activated protein kinase (MAPK) and ERK (MEKs) immediately upstream of the ERKs are available. Generally studies have found them to inhibit at least part of the hypertrophic response of cardiomyocytes in culture. Cardiacspecific overexpression of MEK1 (one of the immediate upstream activators of the ERKs) produced concentric cardiac hypertrophy, but unlike most other models, such as the calcineurin transgenic, the MEK1 transgenic animals did not progress to contractile failure. These data raised the concept of “beneficial” hypertrophy vs “detrimental” hypertrophy by clearly demonstrating that hypertrophy need not inexorably progress to heart failure, possibly because the ERKs (in contrast to many other prohypertrophic signaling molecules including calcineurin, the JNKs, and the p38-MAP kinases) are in many circumstances, cytoprotective. Of note, hypertrophy induced by overexpression of PI3-K also leads to hypertrophy without heart failure, and the PI3-K pathway, like the ERK pathway, is cytoprotective. These data suggest a potential approach to the treatment of patients with heart failure wherein pathways promoting progression of heart failure are inhibited whereas those prohypertrophic pathways that are also cytoprotective are stimulated. The PI3-K and mTOR pathways are critical regulators of cell and organ growth via effects on the activity of several transcription factors regulating expression of hypertrophic response and on the general protein synthesis machinery. Calcineurin is another

important regulator that appears to act largely via effects on the NFAT family of transcription factors. Although other pathways have been implicated, data supporting an important role for these other pathways in physiological or pathological hypertrophy are not nearly as convincing. Thus it seems likely as drugs become available for the treatment of hypertrophic disorders, the initial focus will be on the components of the calcineurin and PI3K/mTOR pathways. Whether long-term inhibition of these pathways, which regulate basic cellular responses of all cells in the body, will be tolerated is unknown. Alternative approaches, such as gene therapy, which can be delivered in an organ-specific (and even cell type-specific) manner, or organ-specific drug delivery may be necessary to target these pathways safely.

ALTERATIONS IN SIGNALING IN THE DISEASED HUMAN HEART Discussion of the signaling profile of the hearts of experimental animals exposed to pressure overload raises two questions: How do the signaling alterations seen in the hearts of these animals compare with signaling alterations in the hearts of patients with hypertrophy or heart failure? Is there any evidence that abnormalities in these clinical scenarios are a cause of heart failure (as opposed to a consequence of the heart failure) and, therefore, will manipulating their activity alter the progression of disease? One study has examined the signaling profile of hypertrophied hearts. Patients were scheduled to be transplant donors but for various reasons were not considered appropriate. Several had significant cardiac hypertrophy, allowing a comparison of the signaling profiles in those hearts vs normal hearts that were also rejected as donors. Of the signaling factors examined (calcineurin, ERK1/2, JNK, p38-MAP kinase, Akt, and GSK-3), the only factor consistently found to be activated in the hypertrophied hearts was calcineurin. Of note, patients with hypertrophy clearly had hypertension, but they were not suspected of having any cardiac disorder and in all cases, systolic function was not depressed. Therefore, clinically, they would be described as “compensated” hypertrophy. These data raised calcineurin as one potential therapeutic target in this phase of the disease. In contrast to the limited data available on hearts with compensated hypertrophy, several studies have examined the signaling

CHAPTER 17 / CARDIAC HYPERTROPHY

profile of hearts explanted from patients with advanced failure either going to transplant or undergoing left-ventricular assist device placement before transplant. These signaling factors were examined in at least one study: calcineurin, ERK1/2, JNKs, p38-MAP kinases, ERK5, the MAP kinase phosphatases (which inactivate the MAP kinases), Akt, GSK-3, and a signaling pathway activated by Ang-II and cytokines, the Janus kinase/signal transducer, and activator of transcription pathway. Where examined, calcineurin expression and activity were increased, though not to the same degree as in the hypertrophied hearts. Unfortunately, no clear consensus has emerged from the studies examining the MAP kinase pathways. The most consistent results appear to be for p38-MAP kinase activity, with three studies reporting activation of p38-MAP kinases in ischemic cardiomyopathy, and only one reporting inhibition. In contrast, in idiopathic cardiomyopathy, p38-MAPK activity has generally been reported to be decreased or unchanged; only one reported activation, although the magnitude of activation was very low. No consistent results have been reported for the JNKs and ERKs. Single reports examined the other factors. In these, ERK5 was inhibited and the MAP kinase phosphatases were activated. Akt was also found to be activated, irrespective of the etiology of the heart failure, and accordingly, its downstream target, GSK-3β, was inhibited. This profile suggested that the heart may be attempting to mount a hypertrophic response in the face of severe contractile dysfunction. Thus, a signaling profile of the failing heart cannot be defined and it is unclear whether any signaling alterations cause or are consequential to the heart failure. It is unclear what if any effect manipulating these pathways would have on the progression of disease. Hearts that have been mechanically unloaded with left ventricular assist device show changes in activity of MAP kinases, with ERK1/2 and JNK activity decreasing and p38-MAP kinase activity increasing, concomitant with a decrease in cardiomyocyte size (i.e., regression of hypertrophy) and a decline in the rate of myocyte apoptosis. However, it is unclear if the regression of hypertrophy and reduction in apoptosis is because of the changes in activity of these MAP kinases. Finally, the complexity of the heart failure signaling abnormalities and the changing activities of various pathways at various times in the progression of disease (as evidenced by the differences in signaling in the hypertrophied vs failing hearts) create uncertainty and lead to significant challenges for translational research. Patients with compensated hypertrophy vs advanced heart failure were at different ends of the pathophysiological spectrum of heart failure, and between these points, including the transition to and early progression of heart failure, very little data exists as to which pathways might be reasonable targets. It is likely, however, that interventions at different points in the progression of heart failure will have to focus on different targets, and it is not clear that these interventions will be equally or uniformly successful. Furthermore, the heterogeneity of signaling abnormalities, even among patients with the same clinical diagnosis, suggests that as has been proposed for cancer therapy directed at protein kinase pathways, therapy may need to be defined by the kinase “profile” of the individual patient rather than the clinical diagnosis. Otherwise, promising therapies may be discarded as ineffective because they were used in the wrong patients. Given the aggressive pursuit of inhibitors of these pathways by the pharmaceutical/ biotechnology industry, the tools to be able to address these questions should be forthcoming.

153

SELECTED REFERENCES Adams JW, Sakata Y, Davis MG, et al. Enhanced Gαq signaling: a common pathway mediates cardiac hypertrophy and apoptotic heart failure. Proc Natl Acad Sci USA 1998;95:10,140–10,145. Ahkter SA, Luttrell LM, Rockman HA, Iaccarino G, Lefkowitz RJ, Koch WJ. Targeting the receptor-Gq interface to inhibit in vivo pressure overload myocardial hypertrophy. Science 1998;280:574–577. Antos CL, McKinsey TA, Frey N, et al. Activated glycogen synthase kinase-3β suppresses cardiac hypertrophy in vivo. Proc Natl Acad Sci USA 2002;99:907–912. Beals CR, Sheridan CM, Turck CW, Gardner P, Crabtree GR. Nuclear export of NF-ATc enhanced by glycogen synthase kinase-3. Science 1997;275:1930–1934. Browne GJ, Proud CG. Regulation of peptide-chain elongation in mammalian cells. Eur J Biochem 2002;269:5360–5368. Bueno OF, Wilkins BJ, De Windt LJ, Molkentin JD. Impairment of cardiac hypertrophy in CnAβ-deficient mice. Proc Natl Acad Sci USA 2002;99:9398–9403. Bueno OF, De Windt LJ, Tymitz KM, et al. The MEK1-ERK1/2 signaling pathway promotes compensated cardiac hypertrophy in transgenic mice. EMBO J 2000;19:6341–6350. Buttini M, Limonta S, Luyten M, Boddeke H. Distribution of calcineurin A isoenzyme mRNAs in rat thymus and kidney. Histochem J 1995; 27:291–299. Cantley LC, Neel BG. New insights into tumor suppression: PTEN suppresses tumor formation by restraining the phosphoinositide 3-kinase/Akt pathway. Proc Natl Acad Sci USA 1999;96:4240–4245. Chen WS, Xu PZ, Gottlob K, et al. Growth retardation and increased apoptosis in mice with homozygous deletion of the Akt1 gene. Genes Dev 2001;15:2203–2208. Choukroun G, Hajjar R, Fry S, et al. Regulation of cardiac hypertrophy in vivo by the stress-activated protein kinases/c-Jun NH2-terminal kinases. J Clin Invest 1999;104:391–398. Coghlan VM, Perrino BA, Howard M, et al. Association of protein kinase A and protein phosphatase 2B with a common anchoring protein. Science 1995;267:108–111. Cohen P, Frame S. The renaissance of GSK-3. Nat Rev Mol Cell Biol 2001;10:769–776. Communal C, Colucci WS, Remondino A, et al. Reciprocal modulation of mitogen-activated protein kinases and mitogen-activated protein kinase phosphatase 1 and 2 in failing human myocardium. J Card Fail 2002;8:86–91. Cook SA, Sugden PH, Clerk A. Activation of c-Jun N-terminal kinases and p38-mitogen-activated protein kinases in human heart failure secondary to ischaemic heart disease. J Mol Cell Cardiol 1999; 31:1429–1431. Cook SA, Matsui T, Li L, Rosenzweig A. Transcriptional effects of chronic Akt activation in the heart. J Biol Chem 2002;277: 22,528–22,533. Crabtree GR. Generic signals and specific outcomes: Signaling through Ca2+, calcineurin, and NF-AT. Cell 1999;96:611–614. Crackower MA, Oudit GY, Kozieradzki I, et al. Regulation of myocardial contractility and cell size by distinct PI3K-PTEN signaling pathways. Cell 2002;110:737–749. Dahlof B, Devereux R, de Faire U, et al. The losartan intervention for endpoint reduction (LIFE) in hypertension study. Am J Hypertens 1997; 10:705–713. Dahlof B, Devereux RB, Kjeldsen SE, et al. Cardiovascular morbidity and mortality in the losartan intervention for endpoint reduction in hypertension study (LIFE): a randomized trial against atenolol. Lancet 2002;359:995–1003. Dancey J, Sausville EA. Issues and progress with protein kinase inhibitors for cancer treatment. Nat Rev Drug Discov 2003;2:296–313. D’Angelo DD, Sakata Y, Lorenz JH, et al. Transgenic Gαq overexpression induced cardiac contractile failure in mice. Proc Natl Acad Sci USA 1997;94:8121–8126. Daub H, Weiss FU, Wallasch C, Ullrich A. Role of transactivation of the EGF receptor in signaling by G-protein-coupled receptors. Nature 1996;379:557–560. De Windt LJ, Lim HW, Bueno OF, et al. Targeted inhibition of calcineurin attenuates cardiac hypertrophy in vivo. Proc Natl Acad Sci USA 2001;98:3322–3327.

154

SECTION II / CARDIOLOGY

Deng L, Huang B, Qin D, Ganguly K, El-Sherif N. Calcineurin inhibition ameliorates structural, contractile, and electrophysiologic consequences of postinfarction remodeling. J Cardiovasc Electrophysiol 2001;12:1055–1061. Dennis PB, Jaeschke A, Saitoh M, Fowler B, Kozma SC, Thomas G. Mammalian TOR: a homeostatic ATP sensor. Science 2001;294: 1102–1105. Ding B, Price RL, Borg TK, Weinberg EO, Halloran PF, Lorell BH. Pressure overload induces severe hypertrophy in mice treated with cyclosporine, an inhibitor of calcineurin. Circ Res 1999;84:729–734. Dolmetsch RE, Lewis RS, Goodnow CC, Healy JI. Differential activation of transcription factors induced by Ca2+ response amplitude and duration. Nature 1997;386:855–858. Eguchi S, Dempsey PJ, Frank GD, Motley ED, Inagami T. Activation of MAPKs by angiotensin II in vascular smooth muscle cells. J Biol Chem 2001;276:7957–7962. Eto Y, Yonekura K, Sonoda M, et al. Calcineurin is activated in rat hearts with physiological left ventricular hypertrophy induced by voluntary exercise training. Circulation 2000;101:2134–2137. Fingar DC, Salama S, Tsou C, Harlow E, Blenis J. Mammalian cell size is controlled by mTOR and its downstream targets S6K1 and 4EBP1/eIF4E. Genes Dev 2002;16:1472–1487. Flesch M, Margulies KB, Mochmann HC, Engel D, Sivasubramanian N, Mann DL. Differential regulation of mitogen-activated protein kinases in the failing human heart in response to mechanical unloading, Circulation 2001;104:2273–2276. Force T, Michael A, Kilter H, Haq S. Stretch-activated pathways and left ventricular remodeling. J Card Fail 2002;8:S351–S358. Frame S, Cohen P. GSK3 takes centre stage more than 20 years after its discovery. Biochem J 2001;359:1–16. Fuentes JJ, Genesca L, Kingsbury TJ, et al. DSCR1, overexpressed in Down syndrome, is an inhibitor of calcineurin-mediated signaling pathways. Hum Mol Genet 2000;9:1681–1690. Goldspink PH, McKinney RD, Kimball VA, Geenen DL, Buttrick PM. Angiotensin II induced cardiac hypertrophy in vivo is inhibited by cyclosporin A in adult rats. Mol Cell Biochem 2001;226:83–88. Graef IA, Mermelstein PG, Stankunas K, et al. L-type calcium channels and GSK-3 regulate the activity of NF-ATc4 in hippocampal neurons. Nature 1999;401:703–708. Gudi SRP, Lee AA, Clark CB, Frangos JA. Equibiaxial strain and strain rate stimulate early activation of G proteins in cardiac fibroblasts. Am J Physiol 1998;274:C1424–C1428. Haq S, Choukroun G, Kang ZB, et al. Glycogen synthase kinase-3β is a negative regulator of cardiomyocyte hypertrophy. J Cell Biol 2000; 151:117–129. Haq S, Choukroun G, Lim HW, et al. Differential activation of signal transduction pathways in human hearts with hypertrophy versus advanced heart failure. Circulation 2001;103:670–677. Haq S, Michael A, Andreucci M, et al. Stabilization of β-catenin by a Wnt-independent mechanism regulates cardiomyocyte growth. Proc Nat Acad Sci USA 2003;100:4610–4515. Hill JA, Karimi M, Kutschke W, et al. Cardiac hypertrophy is not a required compensatory response to short-term pressure overload. Circulation 2000;101:2863–2869. Hill JA, Rothermel B, Yoo KD, et al. Targeted inhibition of calcineurin in pressure-overload cardiac hypertrophy. Preservation of systolic function. J Biol Chem 2002; 277:10,251–10,255. Ho KK, Levy D, Kannel WB, Pinsky JL. The epidemiology of heart failure: The Framingham study. J Am Coll Cardiol 1993;22:6–13. Hu H, Sachs F. Stretch-activated ion channels in the heart. J Mol Cell Cardiol 1997;29:1511–1523. Jiang H, Xiong F, Kong S, Ogawa T, Kobayashi M, Liu JO. Distinct tissue and cellular distribution of two major isoforms of calcineurin. Mol Immunol 1997;34:663–669. Kannel WB. Prevalence and natural history of electrocardiographic left ventricular hypertrophy. Am J Med 1983;75(Suppl 3A):4–11. Keller RS, Shai S-Y, Babbitt CJ, et al. Disruption of integrin function in the murine myocardium leads to perinatal lethality, fibrosis, and abnormal cardiac performance. Am J Pathol 2001;158:1079–1090. Kjeldsen SE, Dahlof B, Devereux RB, et al. Effects of losartan on cardiovascular morbidity and mortality in patients with isolated systolic

hypertension and left ventricular hypertrophy: a losartan intervention for endpoint reduction (LIFE) substudy. JAMA 2002;288:1491–1498. Klee CB, Ren H, Wang X. Regulation of the calmodulin-stimulated protein phosphatase, calcineurin. J Biol Chem 1998;273:13,367–13,370. Kunisada K, Tone E, Fujio E, Matsui H, Yamauchi-Takihara K, Kishimoto T. Activation of gp130 transduces hypertrophic signals via STAT3 in cardiac myocytes. Circulation 1998;98:346–352. Kyriakis JM, Avruch J. Sounding the alarm: protein kinase cascades activated by stress and inflammation. J Biol Chem 1996;271: 24,313–24,316. Lai MM, Burnett PE, Wolosker H, Blackshaw S, Snyder SH. Cain, a novel physiologic protein inhibitor of calcineurin. J Biol Chem 1998;273: 18,325–18,331. Lawlor MA, Mora A, Ashby PR, et al. Essential role of PDK1 in regulating cell size and development in mice. EMBO J 2002;21:3728–3738. Lemke LE, Bloem LJ, Fouts R, Esterman M, Sandusky G, Vlahos CJ. Decreased p38 MAPK activity in end-stage failing human myocardium: p38 MAPK alpha is the predominant isoform expressed in human heart. J Mol Cell Cardiol 2001;33:1527–1534. Levy D, Garrison RJ, Savage DD, Kannel WB, Castelli WP. Prognostic implications of echocardiographically determined left ventricular mass in the Framingham heart study. N Engl J Med 1990;322: 1561–1566. Liao P, Georgakopoulos D, Kovacs A, et al. The in vivo role of p38 MAP kinases in cardiac remodeling and restrictive cardiomyopathy. Proc Nat Acad Sci USA 2001;98:12,283–12,288. Lim HW, De Windt LJ, Mante J, et al. Reversal of cardiac hypertrophy in transgenic disease models by calcineurin inhibition. J Mol Cell Cardiol 2000;32:697–709. Lim HW, De Windt LJ, Steinberg L, et al. Calcineurin expression, activation, and function in cardiac pressure-overload hypertrophy. Circulation 2000;101:2431–2437. Lim HW, Molkentin JD. Calcineurin and human heart failure. Nat Med 1999;5:246, 247. Lindholm LH, Ibsen H, Dahlof B, Devereux RB, Beevers G, de Faire U. Cardiovascular morbidity and mortality in patients with diabetes in the losartan intervention for endpoint reduction in hypertension study (LIFE): a randomized trial against atenolol. Lancet 2002;359: 1004–1010. Lopéz-Rodríguez C, Aramburu J, Rakeman AS, Rao A. NFAT5, a constitutively nuclear NFAT protein that does not cooperate with Fos and Jun. Proc Natl Acad Sci USA 1999;96:7214–7219. Luo Z, Shyu KG, Gualberto A, Walsh K. Calcineurin and cardiac hypertrophy. Nat Med 1998;10:1092, 1093. Lupu F, Terwilliger JD, Lee K, Segre GV, Efstratiadis A. Roles of growth hormone and IGF-1 in mouse postnatal growth. Dev Biol 2001;229: 141–162. MacKenna DA, Dolfi F, Vuori K, Ruoslahti E. ERK and JNK activation by mechanical stretch is integrin-dependent and matrix-specific in rat cardiac fibroblasts. J Clin Invest 1998;101:301–310. Matsui T, Li L, Wu JC, et al. Phenotypic spectrum caused by transgenic overexpression of activated Akt in the heart. J Biol Chem 2002; 277:22,896–22,901. McManus EJ, Alessi DR. TSC2: a complex tale of PKB-mediated S6K regulation. Nat Cell Biol 2002;4:E214–E216. Meguro T, Hong C, Asai K, et al. Cyclosporine attenuates pressure-overload hypertrophy in mice while enhancing susceptibility to decompensation and heart failure. Circ Res 1999;84:735–740. Mende U, Kagen A, Cohen A, Aramburu J, Schoen FJ, Neer EJ. Transient cardiac expression of constitutively active Gαq leads to hypertrophy and dilated cardiomyopathy by calcineurin-dependent and independent pathways. Proc Natl Acad Sci USA 1998;95:13,893–13,898. Mervaala E, Muller DN, Park JK, et al. Cyclosporin A protects against angiotensin II-induced end-organ damage in double transgenic rats harboring human renin and angiotensinogen genes. Hypertension 2000;35:360–366. Michael A, Haq S, Chen X, et al. Glycogen synthase kinase-3β regulates growth, calcium homeostasis, and diastolic function in the heart. J Biol Chem 2004;279:21,383–21,393. Minamino T, Yujiri T, Terada N, et al. MEKK1 is essential for cardiac hypertrophy and dysfunction induced by Gq, Proc Nat Acad Sci USA 2002;99:3866–3871.

CHAPTER 17 / CARDIAC HYPERTROPHY

Molkentin JD, Dorn GW. Cytoplasmic signaling pathways that regulate cardiac hypertrophy. Annu Rev Physiol 2001;63:391–426. Molkentin JD, Lu JR, Antos CL, et al. A calcineurin-dependent transcriptional pathway for cardiac hypertrophy. Cell 1998;93:215–228. Montagne J, Stewart MJ, Stocker H, Hafen E, Kozma SC, Thomas G. Drosophila SK kinase: a regulator of cell size. Science 1999; 285: 2126–2129. Morisco C, Seta K, Hardt SE, Lee Y, Vatner SF, Sadoshima J. Glycogen synthase kinase 3β regulates GATA4 in cardiac myocytes. J Biol Chem 2001;276:28,586–28,597. Morisco C, Zebrowski D, Condorelli G, Tsichlis P, Vatner SF, Sadoshima J. The Akt-glycogen synthase kinase 3β pathway regulates transcription of atrial natriuretic factor induced by β-adrenergic receptor stimulation in cardiac myocytes. J Biol Chem 2000;275:14,466–14,475. Müller JG, Nemoto S, Laser M, Carabello BA, Menick DR. Calcineurin inhibition and cardiac hypertrophy. Science 1998;282:1007. Murat A, Pellieux C, Brunner HR, Pedrazzini T. Calcineurin blockade prevents cardiac mitogen-activated protein kinase activation and hypertrophy in renovascular hypertension. J Biol Chem 2000;275: 40, 867–40,873. Murata M, Fukuda K, Ishida H, et al. Leukemia inhibitory factor, a potent cardiac hypertrophic cytokine. enhances L-type Ca2+ current and [Ca2+]i transient in cardiomyocytes. J Mol Cell Cardiol 1999;31: 237–245. Muramatsu T, Giri PR, Higuchi S, Kincaid RL. Molecular cloning of a calmodulin-dependent phosphatase from murine testis: identification of a developmentally expressed nonneural isoenzyme. Proc Natl Acad Sci USA 1992;89:529–533. Ng DC, Court NW, dos Remedios CG, Bogoyevitch MA. Activation of signal transducer and activator of transcription (STAT) pathways in failing human hearts. Cardiovasc Res 2003;57:333–338. Oh H, Fujio Y, Kunisada K, et al. Activation of phosphatidylinositol 3-kinase through gp130 induces protein kinase B and p70 S6 kinase phosphorylation in cardiac myocytes. J Biol Chem 1998;273: 9703–9710. ∅ie EB, Reidar OPF, Clausen H. Attramadal. Cyclosporin A inhibits cardiac hypertrophy and enhances cardiac dysfunction during postinfarction failure in rats. Am J Physiol Heart Circ Physiol 2000;278: 2115–2123. Pap M, Cooper GM. Role of translation initiation factor 2B in control of cell survival by the phosphatidylinositol 3-kinase/Akt/glycogen synthase kinase-3beta signaling pathway. Mol Cell Biol 2002;22: 578–586. Pete G, Fuller CR, Oldham JM, et al. Postnatal growth responses to insulin-like growth factor I in insulin receptor substrate-1-deficient mice. Endocrinology 1999;140:5478–5487. Petrich BG, Molkentin JD, Wang Y. Temporal activation of c-Jun N-terminal kinase in adult transgenic heart via cre-loxP-mediated DNA recombination. FASEB J 2003;17:749–751. Pham CG, Harpf AE, Keller RS, et al. Striated muscle-specific β1D-integrin and FAK are involved in cardiac myocyte hypertrophic response pathway. Am J Physiol 2000;279:H2916–H2926. Prenzel N, Zwick E, Daub H, et al. EGF receptor transactivation by G-protein-coupled receptors requires metalloproteinase cleavage of proHB-EGF. Nature 1999;402:884–888. Proud CG. Regulation of mammalian translation factors by nutrients. Eur J Biochem 2002;269:5338–5349. Rao A, Luo C, Hogan PG. Transcription factors of the NFAT family: Regulation and function. Annu Rev Immunol 1997;15:707–747. Ren J, Samson WK, Sowers JR. IGF-1 as a cardiac hormone: Physiological and pathophysiological implications in heart disease. J Mol Cell Cardiol 1999;31:2049–2061. Ritter O, Hack S, Schuh K, et al. Calcineurin in human heart hypertrophy. Circulation 2002;105:2265–2269. Rohde J, Heitman J, Cardenas ME. The TOR kinases link nutrient sensing to cell growth. J Biol Chem 2001;276:9583–9586. Ross RS, Pham CG, Shai S-Y, et al. β1 integrins participate in the hypertrophic response of rat ventricular myocytes. Circ Res 1998;82: 1160–1172. Rothermel B, Vega RB, Yang J, Wu H, Bassel-Duby R, Williams RS. A protein encoded within the Down syndrome critical region is enriched

155

in striated muscles and inhibits calcineurin signaling. J Biol Chem 2000;275:8719–8725. Rothermel BA, McKinsey TA, Vega RB, et al. Myocyte-enriched calcineurininteracting protein, MCIP1, inhibits cardiac hypertrophy in vivo. Proc Natl Acad Sci USA 2001;98:3328–3333. Ruwhof C, van der Laarse A. Mechanical stress-induced cardiac hypertrophy: mechanisms and signal transduction pathways. Cardiovasc Res 2000;47:23–37. Sadoshima J, Izumo S. Mechanical stretch rapidly activates multiple signal transduction pathways in cardiac myocytes: potential involvement of an autocrine/paracrine mechanism. EMBO J 1993;12: 1681–1692. Sadoshima J, Montagne O, Wang Q, et al. The MEKK1-JNK pathway plays a protective role in pressure overload but does not mediate cardiac hypertrophy. J Clin Invest 2002;110:271–279. Sadoshima J, Xu Y, Slayter HS, Izumo S. Autocrine release of angiotensin II mediates stretch-induced hypertrophy of cardiac myocytes in vitro. Cell 1993;75:977–984. Sakata Y, Hoit BD, Liggett SB, Walsh RA, Dorn, GW. Decompensation of pressure-overload hypertrophy in Gαq-overexpressing mice. Circulation 1998;97:1488–1495. Sakata Y, Masuyama T, Yamamoto K, et al. Calcineurin inhibitor attenuates left ventricular hypertrophy, leading to prevention of heart failure in hypertensive rats. Circulation 2000;102:2269–2275. Scanga SE, Ruel L, Binari RC, et al. The conserved PI3’K/PTEN/Akt signaling pathway regulates both cells size and survival in Drosophila. Oncogene 2000;19:3971–3977. Schmelzle T, Hall MN. TOR, a central controller of cell growth. Cell 2000;103:253–262. Schwartzbauer G, Robbins J. The tumor suppressor gene PTEN can regulate cardiac hypertrophy and survival. J Biol Chem 2001;276: 35,786–35,793. Shima H, Pende M, Chen Y, Fumagalli S, Thomas G, Kozma SC. Disruption of the p70(s6k)/p85(s6k) gene reveals a small mouse phenotype and a new functional S6 kinase. EMBO J 1998;17:6649–6659. Shimoyama M, Hayashi D, Takimoto E, et al. Calcineurin plays a critical role in pressure overload-induced cardiac hypertrophy. Circulation 1999;100:2449–2454. Shimoyama M, Hayashi D, Zou Y, et al. Calcineurin inhibitor attenuates the development and induces the regression of cardiac hypertrophy in rats with salt-sensitive hypertension. Circulation 2000;102: 1996–2004. Shioi T, Kang PM, Douglas PS, et al. The conserved phosphoinositide 3-kinase pathway determines heart size in mice. EMBO J 2000; 19:2537–2548. Shioi T, McMullen JR, Kang PM, et al. Akt/Protein kinase B promotes organ growth in transgenic mice. Mol Cell Biol 2002;22:2799–2809. Shioi T, McMullen JR, Tarnavski O, et al. Rapamycin attenuates loadinduced hypertrophy in mice. Circulation 2003;107:1664–1670. Shiojima I, Yefremashvili M, Luo Z, et al. Akt signaling mediates postnatal heart growth in response to insulin and nutritional status. J Biol Chem 2002;277:37,670–37,677. Stocker H, Hafen E. Genetic control of cell size. Curr Opin Genet Dev 2000;10:529–535. Subramanian A, Jones W, Gulick J, Wert S, Robbins J. Tissue-specific regulation of the alpha-myosin heavy chain gene promoter in transgenic mice. J Biol Chem 1991;266:24,613–24,620. Sun L, Youn HD, Loh C, Stolow M, He W, Liu JO. Cabin 1, a negative regulator for calcineurin signaling in T lymphocytes. Immunity 1998; 8:703–711. Sussman MA, Lim HW, Gude N, et al. Prevention of cardiac hypertrophy in mice by calcineurin inhibition. Science 1998;281:1690–1693. Taigen T, De Windt LJ, Lim HW, Molkentin JD. Targeted inhibition of calcineurin prevents agonist-induced cardiomyocyte hypertrophy. Proc Natl Acad Sci USA 2000;97:1196–1201. Takaishi T, Saito N, Kuno T, Tanaka C. Differential distribution of the mRNA encoding two isoforms of the catalytic subunit of calcineurin in the rat brain. Biochem Biophys Res Commun 1991;174:393–398. Takeda Y, Yoneda T, Demura M, Usukura M, Mabuchi H. Calcineurin inhibition attenuates mineralocorticoid-induced cardiac hypertrophy. Circulation 2002;105:677–679.

156

SECTION II / CARDIOLOGY

Takeishi Y, Huang Q, Abe J, et al. Activation of mitogen-activated protein kinases and p90 ribosomal S6 kinase in failing human hearts with dilated cardiomyopathy. Cardiovasc Res 2002;53:131–136. Tamemoto H, Kadowaki T, Tobe K, et al. Insulin resistance and growth retardation in mice lacking IRS-1. Nature 1994;372:182–186. Ushio-Fukai M, Griendling KK, Becker PL, Hilenski L, Halleran S, Alexander RW. Epidermal growth factor receptor transactivation by angiotensin II requires reactive oxygen species in vascular smooth muscle cells. Arterioscler Thromb Vasc Biol 2001;21:489–495. Vlahos CJ, McDowell SA, Clerk A. Kinases as therapeutic targets for heart failure. Nat Rev Drug Discov 2003;2:99–113. Wang Z, Kutschke W, Richardson KE, Karimi M, Hill JA. Electrical remodeling in pressure-overload cardiac hypertrophy: Role of calcineurin. Circulation 2001;104:1657–1663. Wang Z, Nolan B, Kutschke W, Hill JA. Na+-Ca2+ exchanger remodeling in pressure overload cardiac hypertrophy. J Biol Chem 2001;276: 17,706–17,711. Weinkove D, Leevers SJ. The genetic control of organ growth: Insights from Drosophila. Curr Opin Genet Dev 2000;10:75–80. Wettschureck N, Rutten H, Zywietz A, et al. Absence of pressure overload induced myocardial hypertrophy after conditional inactivation of Gαq/Gα11 in cardiomyocytes. Nat Med 2001;7:1236–1240. Wilkins BJ, De Windt LJ, Bueno OF, et al. Targeted disruption of NFATc3, but not NFATc4, reveals an intrinsic defect in calcineurin-mediated cardiac hypertrophic growth. Mol Cell Biol 2002;22: 7603–7613.

Woodgett JR. Judging a protein by more than its name: GSK-3. Sci STKE 2001;100:RE12. Xia Y, McMillin JB, Lewis A, et al. Electrical stimulation of neonatal cardiac myocytes activates the NFAT3 and GATA4 pathways and upregulates the adenylosuccinate synthetase 1 gene. J Biol Chem 2000; 275:1855–1863. Yang G, Meguro T, Hong C, et al. Cyclosporine reduces left ventricular mass with chronic aortic banding in mice, which could be due to apoptosis and fibrosis. J Mol Cell Cardiol 2001;33:1505–1514. Yang J, Cron P, Thompson V, et al. Molecular mechanism for the regulation of protein kinase B/Akt by hydrophobic motif phosphorylation. Mol Cells 2002;9:1227–1240. Youn TJ, Piao H, Kwon JS, et al. Effects of the calcineurin dependent signaling pathway inhibition by cyclosporin A on early and late cardiac remodeling following myocardial infarction. Eur J Heart Fail 2002; 4:713–718. Zhang W, Kowal RC, Rusnak F, Sikkink RA, Olson EN, Victor RG. Failure of calcineurin inhibitors to prevent pressure-overload left ventricular hypertrophy in rats. Circ Res 1999;84:722–728. Zhu W, Zou Y, Shiojima I, et al. Ca2+/calmodulin-dependent kinase II and calcineurin play critical roles in endothelin-1-induced cardiomyocyte hypertrophy. J Biol Chem 2000;275:15,239–15,245. Zou Y, Hiroi Y, Uozumi H, et al. Calcineurin plays a critical role in the development of pressure overload-induced cardiac hypertrophy. Circulation 2001;104:97–101.

18 Arrhythmias BARRY LONDON SUMMARY Ion channel mutations cause long QT syndrome, Brugada syndrome, conduction disorders, catecholinergic ventricular tachycardia, and some forms of familial atrial fibrillation and pre-excitation. Transgenic and gene-targeted mouse models of these disorders have further increased the understanding of links between ion channel mutations and these rare arrhythmia syndromes. Molecular genetics, pathophysiology, and implications of these findings are discussed later. It is important to realize, however, that the genetic basis of other inherited arrhythmic syndromes remains unclear, as does the role of genes that are not ion channels. In addition, the relationship of common genetic variants (polymorphisms) to arrhythmic risk is only beginning to be studied. This chapter highlights the avenues of future research that seem most likely to yield results. Key Words: Arrhythmias; atrial fibrillation; Brugada syndrome; conduction disease; ion channel; long QT syndrome: WolffParkinson-White syndrome (WPW).

INTRODUCTION Modern experimental electrophysiology began with the explanation of the basis of the action potential in the squid giant axon, and clinical cardiac electrophysiology began with the development of electrocardiography. The fields progressed in two rather independent directions. The development of intracellular electrodes, the voltage clamp, and the single channel patch clamp revealed the subcellular events that underlie cardiac excitability and automaticity. Meanwhile, arrhythmia mapping, devices, and clinical trials defined the mechanisms that underlie human arrhythmias and assessed the efficacy of therapeutic interventions. The cloning and molecular biological analysis of ion channels has linked structure to function, and human genetic analysis has identified the genes and mutations responsible for rare inherited arrhythmia syndromes. Although the relation of single channel properties to the more common arrhythmias associated with ischemic and cardiomyopathic processes has lagged behind, molecular electrophysiology holds great promise as a unifying discipline in the age following the completion of the human genome project. Ion channel mutations cause long QT syndrome, Brugada syndrome, conduction disorders, catecholinergic ventricular tachycardia, and some forms of familial atrial fibrillation and pre-excitation. From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

Transgenic and gene-targeted mouse models of these disorders have further increased the understanding of links between ion channel mutations and these rare arrhythmia syndromes. Molecular genetics, pathophysiology, and implications of these findings are discussed later. It is important to realize, however, that the genetic basis of other inherited arrhythmic syndromes remains unclear, as does the role of genes that are not ion channels. In addition, the relationship of common genetic variants (polymorphisms) to arrhythmic risk is only beginning to be studied. The avenues of future research that seem most likely to yield results are highlighted next.

INHERITED ARRHYTHMIA SYNDROMES LONG QT SYNDROME Clinical Presentation The autosomal-dominant form of the rare long QT syndrome was first described in the early 1960s. Although the exact gene frequency is not known, it is one of the more common causes of sudden cardiac death among otherwise healthy adolescents and young adults (along with hypertrophic cardiomyopathy and myocarditis). It is characterized by syncope, unexplained seizures, and sudden death. Symptoms often begin near adolescence and are preceded by emotional events such as surprise and fear. The electrocardiographic manifestation of the disease is a prolonged QT interval and an abnormal T-wave morphology on the surface ECG, although the QT interval has considerable variability and some symptomatic affected individuals have QT intervals within the normal range (Fig. 18-1A). The most common arrhythmia is torsade de pointes, or polymorphic ventricular tachycardia with a rotating axis (Fig. 18-1B). The symptoms often respond to treatment with β-blockers, although some patients are refractory. The role of sympathectomy and timing of implantable cardiac defibrillator (ICD) therapy remain uncertain. The autosomal-recessive long QT syndrome is associated with congenital deafness, and is extremely rare. The acquired long QT syndromes are caused by medications (particularly type-IA and type-III antiarrhythmics, certain antibiotics, and nonsedating antihistamines), ischemia, metabolic disorders, and neurological disorders. Acquired long QT syndrome is considerably more common than the autosomal-dominant and -recessive forms, and plays a major role in the design of novel pharmaceutical agents. Molecular Basis of the Disease QT prolongation results from action potential prolongation, and ion channel defects were therefore likely candidates. Positional cloning identified a number of K+ channel and Na+ channel mutations that cause the autosomaldominant forms of this syndrome (Table 18-1). Of note, KvLQT1

157

158

SECTION II / CARDIOLOGY

Figure 18-1 ECG manifestations of the long QT syndrome. (A) 12 Lead ECG of a child with the Romano-Ward autosomal-dominant long QT syndrome. (B) Lead II rhythm strip demonstrating torsade de pointes. Table 18-1 Long QT Syndrome Loci Locus

Protein

Gene

Pathophysiology

Chromosome

LQT1 LQT2 LQT3 LQT4 LQT5 LQT6 LQT7

KvLQT1 HERG Nav1.5 Ankyrin-B MinK/IsK MiRP1 Kir2.1

KCNQ1 KCNH2 SCN5A Ank-B KCNE1 KCNE2 KCNJ2

↓ IKs ↓ IKr ↑ INa ? ↓ IKs ↓ IKr ↓ IK1

11p15.5 7q35-36 3p21 4q25-27 21q21-22 21q21-22 17q23

was first identified by the positional cloning at the LQT1 locus. Mutations of the K+ channels KvLQT1 and HERG account for most cases, whereas Na+ channel mutations are present in 6 mo) blood pressure reduction in spontaneously hypertensive rats

In vivo gene transfer in animal models of limb ischemia: increased vascularity and limb salvage In vivo gene transfer into patients with peripheral artery disease (in progress) In vivo gene transfer into injured animal arteries: 35–70% reduction in neointimal formation

Decreased thrombus formation, increased fibrinolysis Normalized sarcomere function

In vivo gene transfer in animal models of intravascular thrombosis: reduction in platelet deposition, fibrin accumulation, thrombus mass None; efficient in vivo gene correction currently technically unfeasible

Angiogenesis

In vivo gene transfer in animal models of cardiac ischemia: increased vascularity, myocardial perfusion, and cardiac systolic function. Human trials in progress In vivo gene transfer in animal models of heart failure: increased systolic function, decreased mortality

Improved calcium handling, increased contractility

aPartial

lists only. anticipated mechanism has not always been experimentally verified in individual studies. ATPase, adenosine triphophatase; FGF, fibroblast growth factor; HDL, high-density lipoprotein; LDL, low-density lipoprotein; VEGF, vascular endothelial growth factor; VLDL, very lowdensity lipoprotein. bThe

transfer of the gene that encodes the very low-density lipoprotein (VLDL) receptor that binds both LDL and VLDL significantly decreased plasma cholesterol and atherosclerosis in LDL receptordeficient mice and gene transfer of apo A-I (the principal protein component of antiatherogenic high-density lipoprotein or HDL cholesterol), decreased atherosclerosis and promoted lesion regression in atherosclerosis-prone mice. Transfer of genes that encode either tissue-type plasminogen activator (t-PA) or u-PA into baboon endothelial cells decreased thrombus deposition in ex vivo arteriovenous shunts. Finally, myocardial transfer of the genes expressing either sarcoplasmic reticulum Ca2+-adenosine triphosphatase (ATPase) or the β2-adrenergic receptor improved cardiac function in animal models of heart failure. In each of these studies, overexpression of a single normal allele produced a therapeutic effect in an animal not known to have a specific genetic defect.

METHODS OF CARDIOVASCULAR GENE TRANSFER To achieve delivery of a therapeutic gene product in vivo, three steps must occur: (1) recombinant DNA sequences must be

introduced into the nucleus, (2) the DNA must be transcribed into RNA, and (3) the RNA must be translated into a functional protein. Steps 2 and 3 must occur in vivo; however, step 1, gene delivery, may be accomplished either “in vivo” or “ex vivo.” In vivo gene transfer involves delivery of genetic material to cells within an intact animal (or human) tissue. Previously, ex vivo gene delivery involved removal of differentiated cells from the targeted tissue, gene transfer outside the organism, and reimplantation of the genetically modified cells into the tissue from which they were derived. However, the discovery that adult tissues may be colonized and possibly even regenerated by circulating or bone marrow-derived stem cells has opened the additional approach of stem cell removal, ex vivo gene transfer, and infusion of the genetically modified stem cells with targeting of the modified cells to a new location in a diseased tissue. Both in vivo and ex vivo gene therapy strategies have been used for cardiovascular gene therapy studies (Fig. 20-2). EX VIVO GENE THERAPY In ex vivo gene therapy, cells are removed, maintained in culture, targeted with a therapeutic gene, and then reintroduced into the donor animal/patient. The role of the

CHAPTER 20 / CARDIOVASCULAR GENE THERAPY

177

Figure 20-1 Gene therapy approaches to polygenic cardiovascular diseases. Environmental and hereditary factors are involved in the development of polygenic diseases such as hypertension, hypercholesterolemia, and thrombosis. Transfer of one therapeutic gene can potentially correct these polygenic diseases.

Figure 20-2 Methods for cardiovascular gene therapy. (A) Ex vivo gene transfer. This approach has been used to deliver the gene that encodes the low-density lipoprotein receptor to the liver in patients with FH. After partial hepatectomy, hepatocytes are transduced in vitro using a retrovirus vector, and recombinant DNA integrates into the hepatocyte genome. Finally, the transduced hepatocytes are reinfused via the portal vein. They exit the circulation and re-establish themselves within the liver. (B) In vivo gene transfer. A catheter is inserted into a peripheral artery and directed to the heart. This approach has been used to deliver genes that encode angiogenic growth factors, with the expectation that gene therapy will stimulate growth of collateral arteries and provide relief of myocardial ischemia.

reintroduced cells is primarily to deliver a recombinant gene product. The cells are usually not required to reconstitute a particular organ or tissue; therefore, the excised cells may be reimplanted in a location different from their site of excision. For example, hepatocytes harvested from one liver lobe and genetically modified ex vivo may be reinfused throughout the donor liver. Endothelial cells harvested from a donor vein or derived in vitro from progenitor cells harvested from the peripheral blood may be reimplanted as a luminal lining of an artery. Skeletal muscle cells or bone marrow stem cells might be harvested, genetically engineered to acquire biochemical features that are characteristic of cardiac myocytes, then reimplanted within the myocardium. The primary advantages of ex vivo gene therapy are: 1. Gene delivery is performed under controlled, optimized conditions in which the efficiency of gene transfer into the targeted cells may be very high.

2. Gene delivery may be restricted to a specific cell type in which expression is optimized by careful design of a celltype-specific transgene construct. 3. The potential for an immune response to the gene transfer vector is minimized by performing gene delivery in a setting remote from the host immune system (i.e., in a tissue culture dish). The disadvantages of ex vivo gene transfer derive primarily from technical considerations: 1. Except in rare cases of monozygotic siblings, the requirements of histocompatibility mandate that the excised cells are reintroduced only into the donor individual. Thus, every animal (or human) receiving ex vivo gene therapy must undergo two invasive procedures: cell harvest and cell reintroduction.

178

SECTION II / CARDIOLOGY

2. Cell culture and ex vivo gene transfer must be performed under strict aseptic conditions to avoid the introduction of pathogenic microorganisms at the time of cell reimplantation. 3. The requirement that cells are removed and reinfused in ex vivo gene therapy imposes limitations on the type and number of cells that can be used. For cells to be removed for ex vivo genetic modification, they must be nonessential, and their removal and reimplantation must be practical. Thus, ex vivo cardiovascular gene therapy may be attempted with hepatocytes derived from partial hepatectomy and with endothelial and smooth muscle cells derived from excised, nonessential vessels. However, for cardiac gene therapy, the entire myocardium (or large sections of it) cannot be removed, engineered ex vivo, and reimplanted except as a brief, draconian procedure performed during cardiopulmonary bypass. Moreover, there is a limit to the number of cells that can be implanted en masse and survive after injection into the wall of a functioning heart. IN VIVO GENE THERAPY The advantages and disadvantages of in vivo gene therapy are essentially the reverse of those of ex vivo gene therapy. Advantages of in vivo gene therapy include: 1. Only one invasive procedure is required (i.e., injection of the gene vector). 2. Laborious and technically demanding steps of cell harvesting and reimplantation are eliminated. 3. Any cell within an intact tissue or organ is theoretically a target for in vivo gene therapy. The disadvantages of in vivo gene therapy are: 1. Gene delivery may be technically difficult. Gene transfer may need to be performed within the myocardium or in the presence of flowing blood in a narrow end-artery supplying an ischemic territory. In clinical settings such as these, the gene transfer vector may be washed away or undesirable clinical consequences (ischemia, infarction) may intervene before successful gene delivery. 2. The target cells for in vivo gene delivery are likely to be a heterogeneous population. Moreover, these cells are exposed to vector only transiently. Optimization of gene transfer and expression is more difficult under these conditions than with ex vivo, isolated cells. 3. There is an obligatory exposure of the gene transfer vector to the immune system. This exposure may produce an immune response that causes rejection of the transduced cells or blocks gene delivery entirely if the recipient organism is already immune to the gene transfer vector. In its most severe form, immune responses to the vector can precipitate a systemic inflammatory syndrome leading to multiorgan failure and death. Despite these disadvantages, the practical advantages of in vivo vs ex vivo gene transfer are overwhelming. For this reason, virtually all ongoing cardiovascular gene therapy trials employ an in vivo gene transfer approach.

VECTORS In general, the uptake of foreign genetic material (unmodified DNA or RNA) by mammalian cells is an inefficient process. This

inefficiency makes sense teleologically, as there is little advantage for a cell to allow its highly regulated and evolved genetic program to be altered at random by fragments of DNA and RNA that land on its surface. Although there are notable exceptions, to introduce genetic material into mammalian cells with reasonable efficiency, it has been necessary for investigators to associate the genetic material with a “vector” that can mediate entry, nuclear transport, and, in some cases, chromosomal integration. Gene therapy vectors belong to two general categories: nonviral vectors, which mediate gene transfer largely by physical means; and viral vectors, which make use of viral proteins and nucleic acid sequences to mediate efficient gene transfer and occasional chromosomal integration (Table 20-2). NAKED DNA AND NONVIRAL VECTORS Nonviral means of gene transfer include unmodified (naked) DNA, liposomes, and microparticle bombardment. The principal theoretical advantage of nonviral gene transfer is that the components can be prepared as standardized pharmaceutical reagents, which can maximize reproducibility and minimize toxicity. In some settings, nonviral gene transfer is quite effective. For example, muscle cells are particularly amenable to direct in vivo gene transfer with naked DNA. Promising preliminary clinical data has been reported from human trials using naked DNA injection into skeletal or cardiac muscle for cardiovascular gene therapy. However, the efficiency of gene transfer with nonviral vectors is low. In no case have naked DNA or other nonviral gene transfer approaches achieved long term, stable expression in the cardiovascular system nor have these approaches demonstrated unequivocal benefit in a clinical trial. VIRAL VECTORS Retroviral, adenoviral, adeno-associated viral, and lentiviral vectors are in use for cardiovascular gene therapy in animal models. These vectors differ regarding their ability to integrate into chromosomal DNA, their efficiency of gene transfer, their capacity for incorporating foreign DNA, and their safety profiles. Retroviral and lentiviral vectors can mediate integration of therapeutic DNA into the target cell chromosome, permitting long-term expression of transgenes and ensuring transmission of inserted DNA to the progeny of transduced cells. Adeno-associated viral vectors integrate into chromosomal DNA in some cases, but are also able to persist as functional episomal (i.e., extrachromosomal) genetic elements. Adenoviral vectors integrate at a very low frequency but may also persist for years as functional episomal elements. In general, retroviral and lentiviral vectors are inefficient mediators of gene transfer to the cells of the cardiovascular system. Adeno-associated virus is able to transfer genes to hepatocytes and muscle cells with reasonable efficiency but is an inefficient vector for gene transfer to endothelium. In contrast, adenoviral vectors achieve efficient gene transfer to most cell types (including skeletal and cardiac muscle cells, hepatocytes, endothelial, and smooth muscle cells) and are, therefore, widely used in both preclinical studies and early cardiovascular gene therapy trials. Nevertheless, the clinical use of adenovirus has been limited by toxicity, immunogenicity, and brevity of expression. Engineering of the adenoviral genome and capsid proteins to decrease toxicity and immunogenicity has shown great promise in improving the safety profile of adenoviral vectors and in extending the duration of adenovirus-mediated transgene expression.

CURRENT STATUS AND FUTURE DIRECTIONS ATHEROSCLEROSIS RESULTING FROM DYSLIPIDEMIA Because plasma LDL cholesterol is a genetically determined, major modifiable risk factor for the development and progression

179

CHAPTER 20 / CARDIOVASCULAR GENE THERAPY

Table 20-2 Major Advantages and Disadvantages of Available In Vivo Gene-Transfer Vectors Method

Advantages

Disadvantages

Naked DNA

Favorable safety profile No integration into host genomea Permits large DNA inserts Easy to manipulate

Low efficiency

Liposomes

Favorable safety profile Wide range of target cells Permits large DNA inserts Commercially available

Low efficiency

Adenovirus

High efficiency Transduces nonreplicating cells No integration into host genomea

Retrovirus

Stable, long-term gene expression Probably safe in most clinical settings

Adeno-associated virus

Long-term expression possible Parent virus is not a pathogen Stable, long-term expression Can transduce nondividing cells

Derived from potential pathogen DNA insert size limited ( women

Subacute to chronic (months to years); women > men

Subacute (25 mmHg at rest and 30 mmHg with exercise. Pulmonary hypertension is a pathophysiological state, however, not a specific disease, and understanding its underlying cause is required to make progress at clinical and scientific levels. Because the pulmonary circulation is made up of an arterial bed and a venous bed, separated by the pulmonary capillaries, the most important causes of pulmonary hypertension clinically are increases in left atrial pressure (pulmonary venous hypertension) or increases in pulmonary vascular resistance (pulmonary arterial hypertension). Pulmonary venous hypertension is usually resulting from congenital or acquired heart disease (Table 24-1). Anything that increases left atrial pressure such as mitral stenosis or induces left ventricular failure (valvular incompetence, ischemia, hypertension, or nonischemic cardiomyopathy) leads to an increase in the mean pulmonary pressure and the development of pulmonary hypertension. Noncardiac causes of postcapillary pulmonary hypertension include veno-occlusive disease, fibrosing mediastinitis, or pulmonary vein stenosis, all conditions that directly affect the venules or veins. The topic of this chapter, pulmonary arterial hypertension, refers to an increase in resistance in the pulmonary arterial circulation owing to pathological increases in vascular tone and vessel remodeling. Pulmonary arteries and arterioles, the major site of resistance to blood flow in the pulmonary vascular bed, consist of endothelial cells, smooth muscle cells (SMCs), and fibroblasts. Understanding how these three cell types, responding to genetic and environmental influences, contribute to the increased vascular tone and vessel remodeling characteristic of pulmonary arterial hypertension is critical in designing targeted therapy to prevent and reverse this disease.

VASOCONSTRICTION In vascular smooth muscle an increase in cytosolic calcium is an initiating event for contraction. As intracellular calcium increases, it binds to calmodulin and this complex activates myosin light-chain kinase, which then phosphorylates the light chain of myosin. In an adenosine triphosphate-dependent reaction myosin then binds to actin filaments causing contraction. The increase in intracellular calcium occurs owing to the release of intracellular stores from the sarcoplasmic reticulum (SR) and to an influx of calcium from the extracellular space through a variety of selective and nonselective ion channels. Although sustained contraction within SMCs is initiated by calcium–calmodulin mediated phosphorylation of myosin light

223

224

SECTION III / PULMONARY DISEASES

Table 24-1 Causes of Pulmonary Hypertension Pulmonary arterial hypertension (mean PA pressure >25 mmHg with left atrial [wedge] pressure 25 mmHg with left atrial [wedge] pressure >18 mmHg)

Pulmonary vascular disease Cardiac causes PPH Mitral stenosis Familial Atrial myxoma Sporadic Left ventricular dysfunction Collagen vascular disease-related Valvular disease Scleroderma Systemic hypertension Lupus Cardiomyopathy Rheumatoid arthritis Ischemic Mixed connective disease Nonischemic Congenital heart disease Hypertrophic Atrial or ventricular septal defects Patent ductus arteriosus Drug-induced Embolic disease Thromboembolic In situ thrombosis Tumor emboli Chronic hypoxia Chronic alveolar hypoventilation Chronic high altitude exposure Associated with pleural-pulmonary Noncardiac causes disease Veno-occlusive disease Emphysema Pulmonary vein stenosis Interstitial lung disease Fibrosing mediastinitis Chest wall deformities Sickle cell disease Cystic fibrosis PA, pulmonary arterial; PPH, primary pulmonary hypertension.

that influx of calcium is critical. However, L-type channel blockers do not completely inhibit hypoxia-induced vasoconstriction suggesting that calcium enters through other pathways as well. Another voltage-gated calcium channel, the T-type calcium channel that has lower conductance than the L-type channel and is activated at more negative membrane potentials (around –70 mV), may also allow calcium into the cell in response to hypoxia. However, evidence suggests that this channel has a greater effect on cell proliferation than on cell contraction. Finally, nonselective cation channels and ligand-gated channels that are not specific for calcium, but also permit sodium entry, may contribute to the increase in intracellular calcium. Vasoconstriction of the pulmonary circulation can also be initiated by circulating mediators such as ET-1, serotonin (5-hydroxytryptamine [HT]), and angiotensin II (AGII). Similar to the effects of hypoxia, none of these agents can induce vasoconstriction in the absence of calcium, but unlike hypoxia, they have a similar vasoconstricting effect on the pulmonary and systemic circulations. They increase intracellular calcium through multiple mechanisms directed by binding to G protein coupled receptors. These receptors stimulate phospholipase C activity, leading to the formation of two important intracellular signaling messengers, inositol triphosphate and diacylglycerol. Binding of inositol triphosphate to the SR triggers calcium release into the cytoplasm; diacylglycerol

CHAPTER 24 / PULMONARY HYPERTENSION

225

Figure 24-2 Calcium entry and signaling. Increases in intracellular calcium are the initiating event in vascular smooth muscle contraction. Calcium can enter the cell from the extracellular space through voltage-activated L-type and T-type channels or through voltage-independent nonselective or store-operated cation channels. Intracellular calcium can also be increased by the release of stores within the SR. SR release occurs in response to activation of receptor operated channels and subsequent generation of the intracellular messengers inositol triphosphate and diacylglycerol. The increase in cytosolic calcium activates calcium-calmodulin-dependent pathways including the phosphorylation of myosin light chain kinase. Activated myosin light chain kinase phosphorylates the light chain of myosin and allows it to interact with actin to cause contraction. Myosin phosphatase dephosphorylates myosin light chain and allows relaxation. AGII, angiotensin II; ET-1, endothelin-1; 5-HT, 5-hydroxy-tryptamine.

along with calcium activates protein kinase C, which phosphorylates specific target proteins important in contraction and proliferation. In addition to the release of calcium stores, receptor activation allows calcium to enter the cell through ion channels. These include the L- and T-type voltage-activated channels and voltage-independent channels such as nonselective cation channels and store-operated channels. Store-operated channels allow calcium to enter the cell in response to depletion of calcium in the SR. Although vasoconstriction requires active contraction of PA SMCs, endothelial cells play an important role in regulating the degree and duration of this contraction. This is achieved through the release of two important vasodilators, prostacyclin, and nitric oxide (NO). Prostacyclin is produced by the enzyme prostacyclin synthase and causes vasodilation when given intravenously. Increases in intracellular cAMP, with consequent activation of protein kinase A, is thought to mediate a number of the vasodilatory functions of prostacyclin. cAMP lowers intracellular calcium concentrations and decreases calcium sensitization so that for equal concentrations of cytosolic calcium there is less myosin light chain phosphorylation. Consistent with prostacyclin’s vasodilatory effect on the pulmonary circulation, mice genetically engineered to overexpress prostacyclin synthase (and thus produce more prostacyclin) have a blunted vasoconstrictor response to acute hypoxia; they also develop less pulmonary hypertension when exposed to chronic hypoxia. NO is also produced by endothelial cells through the actions of the enzyme endothelial nitric oxide synthase (eNOS) on L-arginine. NO produced by eNOS is used locally to promote SMC relaxation. NO produced in the endothelium diffuses into PA SMCs and binds to soluble guanylyl cyclase. Guanylyl cyclase generates cGMP, an intracellular messenger that promotes vasodilation principally through the actions of its main substrate, protein

kinase G. Similar to the mode of action for cAMP, it is thought that cGMP induces vasodilation either by reducing cytosolic calcium or by decreasing the calcium sensitivity of the contractile proteins. cGMP is degraded by phosphodiesterases (PDE) to inactive 3′5′cGMP terminating the vasodilating effects of NO. Mice lacking the eNOS gene develop increased pulmonary pressures even at mild altitude; the elevated pressures can be reversed by the administration of exogenous NO. Supporting the putative role of prostacyclin and NO in maintaining the low resistance of the normal pulmonary vascular bed, patients with severe pulmonary hypertension demonstrate decreased expression of both prostacyclin synthase and eNOS in PA endothelial cells. Whether the loss of these vasodilators is a cause of, or a result of, pulmonary hypertension is unclear.

PULMONARY VASCULAR REMODELING The mechanistic relationship between vasoconstriction and vascular remodeling has begun to be elucidated. Most agents that induce pulmonary vasoconstriction, such as hypoxia, ET-1, 5-HT, and AGII, also stimulate vessel wall remodeling. For remodeling to occur, quiescent cells must be stimulated to proliferate (leading to hyperplasia and hypertrophy), activated cells must be able to migrate (leading to neointimal formation and neomuscularization of small arteries and arterioles), and cells (primarily fibroblasts) must be stimulated to increase the extracellular matrix through deposition of collagen and elastin. This combination of cell proliferation with an increase in extracellular matrix makes the pulmonary vessels less compliant increasing pulmonary vascular resistance. Chronic increases in blood flow (congenital heart disease), chronic hypoxia (long-term exposure to high altitude or in patients with primary alveolar hypoventilation), direct toxins (diet supplements), autoimmune disease (scleroderma), thromboembolic

226

SECTION III / PULMONARY DISEASES

Figure 24-3 Cellular contributions to PA remodeling. All cell types within the pulmonary vessel are involved in pulmonary vascular remodeling. The response of each cell type depends on the type, severity, and duration of injury. Therapies to control and reverse pulmonary vascular remodeling need to address these pathological cell responses. ET-1, endothelin-1; NO, nitric oxide; PGI2, prostacyclin.

disease, or genetic causes (primary pulmonary hypertension [PPH]) can all initiate remodeling of the pulmonary vasculature. All three cell types within the arterial wall, endothelial cells, SMCs, and fibroblasts, are involved in the vascular remodeling associated with pulmonary hypertension (Fig. 24-3). Endothelial cells line the vessel lumen and are directly exposed to circulating mitogens, shear stress, and changes in flow. Injury to the endothelium can lead to the loss of lumen integrity, thus exposing SMCs directly to the effects of circulating mediators and shear stress that can promote medial proliferation. Smooth muscle hyperplasia and hypertrophy occur in most forms of pulmonary hypertension. Extension of smooth muscle into previously nonmuscularized arteries and arterioles, termed neomuscularization, also occurs in most forms of pulmonary hypertension. This occurs resulting from the differentiation of precursor cells (pericytes) within the vessel wall into contractile smooth muscle, although some studies suggest that migration of SMCs into these arterioles also plays a role. Neomuscularization not only increases pulmonary vascular resistance by reducing lumen size, it plays an important role in increasing pulmonary vascular tone. Fibroblasts compose the outer vessel layer and some evidence suggests they play a key role in pulmonary vascular remodeling. Fibroblasts possess a number of attributes that likely contribute to vascular remodeling. In response to stimuli such as flow or hypoxia they can proliferate, migrate, and release components of the extracellular matrix via the secretion of matrix metalloproteinases (MMPs). In addition they can secrete cytokines and growth factors such as transforming growth factor (TGF)-β, epithelial growth factor, and insulin growth factor that operate in a paracrine fashion to stimulate SMC proliferation and extracellular matrix deposition. Subsets of fibroblasts have been identified in bovine pulmonary arteries that are highly proliferative in response to hypoxia and growth factors. These subsets express different

isoforms of protein kinase C that may explain their distinct proliferative phenotypes.

INITIATORS OF PULMONARY REMODELING INCREASED PULMONARY BLOOD FLOW Congenital heart defects with significant left-to-right shunts can cause pulmonary hypertension that progresses over time. The initial response of the pulmonary circulation to chronically increased flow is the extension of muscle into previously nonmuscularized small arteries and arterioles. With time, the previously muscularized arteries respond to the increased flow and stretch by undergoing medial hypertrophy and hyperplasia. Combined with the deposition of extracellular matrix the medial thickness of the arteries and arterioles is increased and the compliance drops. If the shunt persists arterial concentration is reduced and neointimal formation develops. In advanced disease there is occlusive neointimal formation (plexiform lesions) and pulmonary pressures can exceed systemic ones initiating reversal of flow through the shunt. This is termed Eisenmenger’s syndrome and is thought to represent an irreversible form of pulmonary hypertension. Figure 24-4 shows an example of a plexiform lesion with the small, residual lumen indicated by the arrows. The normal appearance of alveolar septae can be readily appreciated. Infants with large ventricular septal defects or a patent ductus arteriosus are at especially high risk for developing this complication. Infants whose shunts are repaired within the first 8 mo of life usually normalize their pulmonary pressures over time; those repaired after 2 yr of age invariably have elevated pressures throughout their life. Loss of endothelial vasodilator function has been identified as an early finding in animal models of left-to-right shunts. Endothelial injury also exposes the SMC directly to the circulating mitogens and shear stress, an unnatural environment that promotes proliferation. Changes within the adventitial layer are also seen in

CHAPTER 24 / PULMONARY HYPERTENSION

227

Figure 24-4 Plexiform lesion. The end-stage of pulmonary hypertension is associated with the formation of occlusive intimal lesions. Despite the often bizarre appearance of the lesions, adjacent lung tissue is generally normal, and distal blood vessels may also appear surprisingly unaffected. Clonal expansion of genetically abnormal endothelial cells may contribute to abnormal proliferation within the lesions, and chronic viral infection by human herpesvirus type VII may also be a predisposing factor to development of plexiform lesions. PAH, pulmonary arterial hypertension. (Please see color insert.)

individuals with chronically increased pulmonary blood flow. Immunostaining from lung biopsies demonstrates increased deposition of the glycoproteins tenascin C and fibronectin in the matrix. Association of tenascin C with proliferating SMCs has been demonstrated in animal models of pulmonary hypertension and in patients with congenital heart disease suggesting a facilitating role in promoting hyperplasia and hypertrophy. Also identified in humans and animal models of left-to-right shunt was an increase in elastase activity. Increased elastase activity can break down the extracellular matrix and release mitogens such as fibroblast growth factor 2 and basic growth factor. This is likely important in promoting the proliferation and migration of medial and adventitial cells. HYPOXIA Acute hypoxia causes pulmonary vasoconstriction whereas chronic hypoxia additionally leads to vascular remodeling. The remodeling associated with chronic hypoxia is most pronounced in the medial layer of larger arteries in which hyperplasia and hypertrophy of SMCs occur and in small arteries and arterioles in which new smooth muscle is formed. Hypoxia increases shear stress, tension, and pressure all of which can initiate vascular remodeling. Whether hypoxia is a direct inducer of vessel remodeling or whether the stimulus to proliferate is related to the changes in shear stress resulting from vasoconstriction is unclear. In an animal model using unilateral banding of the main PA, hypoxia without flow did not lead to vascular remodeling or medial hypertrophy. This suggests that the effect of hypoxia on flow and shear stress is central to its induction of remodeling. Hypoxia has also been demonstrated to reduce the production and release of prostacyclin and NO while stimulating ET-1 release effectively tipping the balance in favor of vascular proliferation. The extracellular matrix is a biologically active composition of collagen, elastin, and proteoglycans. Cells within the vessel and matrix can detect small changes, and disruption of the matrix in response to atherosclerosis and hypertension can stimulate vessel

remodeling. Specific proteolytic enzymes, MMPs, regulate the composition and integrity of the extracellular matrix. Their expression is regulated by growth factors and cytokines; they are commonly upregulated in malignancies, closely correlating with a tumor’s ability to metastasize. They are released by cells in the connective tissue particularly fibroblasts and myofibroblasts. MMP-2 and MMP-9 are two subtypes important in vascular smooth muscle cell (VSMC) activation and neointimal formation after balloon injury. Inhibitors of MMP-2 and MMP-9 decrease VSMC migration in primate arterial explants and overexpression of MMP-9 enhances VSMC migration in vitro. Secretion of MMPs by endothelial cells is an important step in angiogenesis. As endothelial cells migrate they secrete bursts of MMP-9 to degrade the basement membrane and allow cell transit. This ability to break down extracellular matrix to allow cells to migrate is critical for neovascularization. Activated MMPs are not usually present in the quiescent adult pulmonary vessel. Hypoxia increases the activity of collagenolytic metalloproteinases, which results in the appearances of collagen breaks and fragments. The breaks in collagen allow migration of SMCs or fibroblasts from the media or adventitia and can lead to neointimal proliferation. Cleaved collagen fragments can also induce SMC and fibroblast proliferation. Extracts from the small peripheral pulmonary arteries of chronically hypoxic rats have an increase in collagenolytic activity relative to normoxic controls. This activity was most pronounced at 4 d and then again during recovery from hypoxia when resolution of vessel remodeling occurs. Inhibitors of MMPs can prevent hypoxia-induced pulmonary hypertension in rats, providing support for an important role of MMPs in pulmonary vascular remodeling. ENDOTHELIN-1 ET-1 is a vasoactive peptide synthesized by the endothelium. It can bind receptors on both endothelial and SMCs although with different actions on each. ET-1 is a ligand for both the ETA and ETB receptors on the SMC and promotes

228

SECTION III / PULMONARY DISEASES

vasoconstriction and proliferation. ET-1 also binds ETB receptors on endothelial cells that promote the release of prostacyclin and NO. Elevated circulating levels of ET-1 are present in most forms of pulmonary hypertension suggesting a pathogenic role in its development or perpetuation. Elevated circulating levels of ET-1 are found in almost every type of vascular disease including atherosclerosis and vasculitis, however, suggesting that it is a nonspecific marker of vascular injury. SEROTONIN 5-HT is a known mitogen for SMCs isolated from bovine and rat pulmonary arteries. The ability of 5-HT to remodel the pulmonary circulation has received great interest because of the development of pulmonary hypertension in individuals taking appetite suppressants that are known to interfere with the metabolism of 5-HT. Similar to ET-1, circulating 5-HT levels are elevated in most individuals with pulmonary hypertension. Treatments that increase 5-HT levels in rats exposed to long-term hypoxia worsen pulmonary hypertension, an effect that can be blocked by inhibitors of 5-HT uptake. In addition, mice that lack the 5-HT transporter required for the internalization of 5-HT develop less severe pulmonary remodeling when exposed to chronic hypoxia. ANGIOTENSIN II Evidence that AGII plays a role in the development of pulmonary hypertension comes from experiments demonstrating induction of hypertrophy and hyperplasia in rat PA SMC in vitro. Expression of the angiotensin-converting enzyme (ACE) in the endothelial layer of small arteries and evidence of increased AGII binding to its receptor in the arterial walls of rats with hypoxia-induced pulmonary hypertension all support a role for AGII as a causative agent for the development of pulmonary hypertension. Additional evidence is provided by the ability of ACE inhibitors to attenuate pulmonary vascular remodeling. CHRONIC LUNG DISEASE Pulmonary hypertension also develops in many forms of chronic lung disease including emphysema, interstitial lung disease, cystic fibrosis, and sickle cell disease. Pulmonary hypertension resulting from chronic lung disease is likely a different process than pulmonary hypertension arising from isolated involvement of the pulmonary vessels (like PPH, collagen vascular-associated pulmonary hypertension, and chronic increases in blood flow from congenital heart disease). Hypoxia, inflammation, and increases in flow/shear stress appear to be the initiating events. Inflammation in particular may play a role in the pulmonary hypertension associated with emphysema. Cigarette smoke impairs endothelial vasodilator function in vitro. Vascular remodeling similar to that seen in patients with hypoxia has been demonstrated in smokers with relatively preserved oxygenation. Whether the similarity is because of the like effects of cigaret smoking on shear stress or flow is not clear. It does suggest that direct injury to the pulmonary vessels from inflammation can occur independent of its impact on airways and lung parenchyma.

PULMONARY ARTERIAL HYPERTENSION AS A GENETIC DISEASE The effect of mutations within the TGF-β superfamily of genes on the development and progression of pulmonary hypertension has received a great deal of attention. Of patients with PPH, approx 6% demonstrate an autosomal-dominant pattern of inheritance. In 2000, two separate groups identified germline mutations in the bone morphogenic protein receptor type 2 (BMPR2) that strongly correlated with the development of familial PPH. Additionally, mutations in the Alk1 gene, a TGF-β type-I receptor, were also identified in families with hereditary hemorrhagic

telangectasia and PA hypertension. BMPR2 is a member of the TGF-β superfamily that transduces signals by binding to heterotrimeric complexes of type-I and -II receptors to activate serine/threonine kinases. This leads to the activation of intracellular messengers known as SMADs that act as transcriptional regulators. In PA SMC culled from the conduit PA of patients with PPH, it was demonstrated that TGF-β stimulated DNA incorporation and proliferation whereas it had a growth inhibitory effect on cells from normal controls. In addition, the normal antiproliferative effect of the secreted cytokine bone morphogenic protein 7, which is a ligand for BMPR2, was lost in cells from patients with PPH. However this loss of effectiveness was not seen in cells obtained from individuals with secondary pulmonary hypertension. In a mouse model, expression of a dominant negative BMPRII mutation in smooth muscle led to the development of pulmonary hypertension, providing strong support for the theory that mutations in this receptor can lead to PPH. Which intracellular signaling mechanisms are disrupted because of these mutations, the role of mutations in endothelium and other cell types, and how these mutations lead to pulmonary hypertension is not understood. Epidemiological data also indicate that mutations in BMPR2 are present in the majority of individuals with sporadic (nonfamilial) forms of PPH. Mutations of the BMPR2 have also been identified in patients with veno-occlusive disease, a form of postcapillary pulmonary hypertension resulting from the obliteration of small pulmonary venules and veins. No clear defects in systemic vascular remodeling have been identified in these individuals suggesting some specificity of this pathway in regulating pulmonary vascular remodeling.

THERAPIES FOR PULMONARY HYPERTENSION Attempts to treat pulmonary hypertension have adopted two general strategies: the use of specific blockers against mitogens (including calcium), which are known mediators of pulmonary hypertension, and attempts to increase the concentration and duration of action of the endothelial derived vasodilators, prostacyclin, and NO. Calcium channel blockers such as nifedipine prevent the influx of calcium through the voltage-gated calcium (L-type channels) and are the first-line therapy in patients with pulmonary hypertension (Fig. 24-5). They can reverse vasoconstriction in some patients and blocking calcium entry may also have an antimitogenic function on SMCs. Unfortunately, in humans they are effective in only a minority of individuals with pulmonary hypertension. Elevated circulating levels of ET-1 are present in all forms of pulmonary hypertension. Bosantan, a nonspecific ETA/ETB blocker, has been approved for the treatment of PA hypertension. Although experimentally ACE inhibitors can attenuate the vascular remodeling associated with hypoxia-induced pulmonary hypertension, they are not used clinically because of their effects on systemic blood pressure. NO is a potent vasodilator whose effect is mediated through the cyclic nucleotide cGMP. Intense scientific and pharmaceutical effort has investigated ways to augment NO’s and cGMP’s effects on the pulmonary circulation. Experimentally, gene therapy using adenovirus to overexpress eNOS has had some success in limiting pulmonary vascular remodeling. Technical barriers make it unsuitable for clinical use at this time, however. Nitroprusside and nitroglycerin are NO donors that can induce vasodilation even in the presence of an injured endothelium. Unfortunately, both can cause

CHAPTER 24 / PULMONARY HYPERTENSION

Figure 24-5 Flow diagram for initial therapeutic decision making in pulmonary arterial hypertension. Because of the complexity of caring for patients with pulmonary hypertension, initiation and maintenance of therapy should only be undertaken by a care team with extensive experience and adequate access to tertiary care services, including lung transplantation.

systemic hypotension making them less than ideal drugs to treat pulmonary hypertension. Inhaled NO has the advantage of acting locally in the pulmonary circulation because it is inactivated by hemoglobin long before reaching the systemic circulation. Toxicity of the gas as well as difficulties with delivery make this useful only in hospitalized patients particularly those awaiting lung transplant. Another strategy to prolong the vasodilator and antiproliferative effects of NO is to prevent the degradation of cGMP. cGMP is generated when NO binds soluble guanylyl cyclase, which is central to the vasodilating and antimitogenic effects of NO. cGMP is hydrolyzed by PDE, a family of enzymes (numbering 20 and counting) that bind and inactivate cAMP and cGMP thus terminating their intracellular effects. Each PDE varies in location and specificity of substrate. PDE5 is primarily located in vascular smooth muscle and is specific for cGMP making it an attractive pharmacological target. Inhibitors of PDE5 such as zapronist and sildenafil prolong the half-life of cGMP and promote vascular relaxation. In cooperation with agents that increase NO availability such as inhaled NO or nitroglycerin, PDE5 inhibitors have a potent (and potentially dangerous) vasodilator effect. Clinical trials looking at the effect of PDE5 inhibitors on persistent pulmonary hypertension of the newborn and PPH are ongoing. Intravenous prostacyclin is the drug of choice for individuals with PPH or scleroderma-associated pulmonary hypertension. Although prostacyclin is an effective vasodilator acutely, its ability to act as an anti-proliferative agent is central to its beneficial effect in pulmonary hypertension. Even patients who have no acute vasodilatory response to prostacyclin demonstrate improvement in pulmonary hemodynamics over 18 mo. Unfortunately, intravenous prostacyclin is expensive and the delivery system, which requires a central venous catheter, has many technical limitations. Inhaled and oral forms of prostacyclin are being studied as an alternative.

POTENTIAL NEW THERAPIES The only approved therapies for PA hypertension are calcium channel blockers, intravenous prostacyclin, and the nonselective

229

ET-1 blocker bosantan. All act at least in part to reduce pulmonary vascular tone. New therapies specifically targeted at reversing the vascular remodeling associated with pulmonary hypertension are being investigated. Serine elastase inhibitors are an example of this strategy. Serine elastase activity is increased in animal models of pulmonary hypertension and is also elevated in humans with congenital heart disease. The increased activity leads to proteinasedependent deposition of extracellular matrix proteins such as elastin, fibronectin, and tenascin C, which contributes to SMC proliferation. When given orally serine elastase inhibitors completely reversed the vascular lesions and normalized pulmonary pressures in an animal model of monocrotaline-induced pulmonary hypertension. This was associated with PA SMC apoptosis and loss of extracellular matrix, specifically elastin and tenascin C. Inhibitors of the RhoA/Rho kinase pathway are also being considered as a treatment for pulmonary hypertension. RhoA GTPase mediates a number of cellular responses central to the development of pulmonary hypertension including activation of the contractile proteins, cell proliferation, and induction of gene expression. Animal models indicate a role for activation of the RhoA/Rho kinase pathway in both acute and chronic hypoxia. Acute administration of Rho kinase inhibitors attenuate acute hypoxic vasoconstriction in isolated mouse lungs whereas chronic administration decreases the degree of right ventricular hypertrophy and neomuscularization following 3 wk of hypoxia. The HMG Co-A reductase inhibitors, (the statins) have replaced aspirin as the new wonder drug and look promising as a potential therapy for pulmonary hypertension. Statins have a potent antiproliferative effect on VSMCs, independent of cholesterol metabolism. They inhibit isoprenylation, which is required for the membrane localization and subsequent activation of small G proteins such as Ras and Rho, which are important intracellular messengers. In vitro, HMG Co-A reductase inhibitors not only block VSMC proliferation in response to serum but can also induce apoptosis. Already approved for the treatment of coronary artery disease they are being considered for clinical trials to treat PA hypertension.

CONCLUSION PA hypertension results from the pathological elevation in pulmonary vascular resistance because of a combination of increased vascular tone and vessel wall remodeling. It is a heterogeneous disorder with the vascular phenotype dependent on the mechanism and severity of injury. With increasing duration and severity, pulmonary hypertension is associated with progressive vessel wall remodeling ultimately resulting in fixed, irreversible lesions. All three cell types within the vessel wall contribute to these vascular lesions through the combined effect of cell proliferation, cell migration, and matrix deposition. Therapies aim to restore vasodilator function while limiting mitogen-induced proliferation. New therapies need to reverse these vascular lesions by breaking down extracellular matrix and inducing controlled regression (apoptosis) within the vessel wall to provide the best chance to restore the pulmonary circulation to its natural low resistive state and reduce the morbidity and mortality associated with this disease.

SELECTED REFERENCES Albert AP, Large WA. Store-operated Ca2+-permeable non-selective cation channels in smooth muscle cells. Cell Calcium 2003;33: 345–356.

230

SECTION III / PULMONARY DISEASES

Cowan KN, Heilbut A, Humpl T, Lam C, Ito S, Rabinovitch M. Complete reversal of fatal pulmonary hypertension in rats by a serine elastase inhibitor. Nat Med 2000;6:698–702. Eddahibi S, Raffestin B, Hamon M, Adnot S. Is the serotonin transporter involved in the pathogenesis of pulmonary hypertension? J Lab Clin Med 2002;139:194–201. Geraci MW, Gao B, Shepherd DC, et al. Pulmonary prostacyclin synthase overexpression in transgenic mice protects against development of hypoxic pulmonary hypertension. J Clin Invest 1999;103:1509–1515. Gianetti J, Bevilacqua S, De Caterina R. Inhaled nitric oxide: more than a selective pulmonary vasodilator. Eur J Clin Invest 2002;32:628–635. Hopkins N, McLoughlin P. The structural basis of pulmonary hypertension in chronic lung disease: remodelling, rarefaction or angiogenesis? J Anat 2002;201:335–348. Jeffery TK, Wanstall JC. Pulmonary vascular remodeling: a target for therapeutic intervention in pulmonary hypertension. Pharmacol Ther 2001;92:1–20. Lane KB, Machado RD, Pauciulo MW, et al. Heterozygous germline mutations in BMPR2, encoding a TGF-beta receptor, cause familial primary pulmonary hypertension. The International PPH Consortium. Nat Genet 2000;26:81–84. Li S, Westwick J, Poll C. Transient receptor potential (TRP) channels as potential drug targets in respiratory disease. Cell Calcium 2003;33: 551–558. Loyd JE. Genetics and pulmonary hypertension. Chest 2002;122(6 Suppl): 284S–286S. Loyd JE, Parker B. Francis Lecture. Genetics and gene expression in pulmonary hypertension. Chest 2002;121(3 Suppl):46S–50S. Mandegar M, Remillard CV, Yuan JX. Ion channels in pulmonary arterial hypertension. Prog Cardiovasc Dis 2002;45:81–114. Mandegar M, Yuan JX. Role of K+ channels in pulmonary hypertension. Vascul Pharmacol 2002;38:25–33. Nagaoka T, Morio Y, Casanova N, et al. Rho/Rho-kinase signaling mediates increased basal pulmonary vascular tone in chronically hypoxic rats. Am J Physiol Lung Cell Mol Physiol 2004;287(4):L665–L672.

Novotna J, Herget J. Possible role of matrix metalloproteinases in reconstruction of peripheral pulmonary arteries induced by hypoxia. Physiol Res 2002;51:323–334. Pellicelli AM, Palmieri F, Cicalini S, Petrosillo N. Pathogenesis of HIVrelated pulmonary hypertension. Ann NY Acad Sci 2001;946:82–94. Rabinovitch M. Pathobiology of pulmonary hypertension. Extracellular matrix. Clin Chest Med 2001;22:433–449, viii. Review. Rabinovitch M. Pulmonary hypertension: pathophysiology as a basis for clinical decision making. J Heart Lung Transplant 1999;18:1041–1053. Rybalkin SD, Yan C, Bornfeldt KE, Beavo JA. Cyclic GMP phosphodiesterases and regulation of smooth muscle function. Circ Res 2003;93: 280–291. Stenmark KR, Bouchey D, Nemenoff R, Dempsey EC, Das M. Hypoxiainduced pulmonary vascular remodeling: contribution of the adventitial fibroblasts. Physiol Res 2000;49:503–517. Stenmark KR, Gerasimovskaya E, Nemenoff RA, Das M. Hypoxic activation of adventitial fibroblasts: role in vascular remodeling. Chest 2002;122(6 Suppl):326S–334S. Strange JW, Wharton J, Phillips PG, Wilkins MR. Recent insights into the pathogenesis and therapeutics of pulmonary hypertension. Clin Sci (Lond) 2002;102:253–268. Sweeney M, Yuan JX. Hypoxic pulmonary vasoconstriction: role of voltage-gated potassium channels. Respir Res 2000;1:40–48. Thomas AQ, Carneal J, Markin C, et al. Specific bone morphogenic protein receptor II mutations found in primary pulmonary hypertension cause different biochemical phenotypes in vitro. Chest 2002;121(3 Suppl):83S. Thomas AQ, Gaddipati R, Newman JH, Loyd JE. Genetics of primary pulmonary hypertension. Clin Chest Med 2001;22:477–491, ix. van den Driesche S, Mummery CL, Westermann CJ. Hereditary hemorrhagic telangiectasia: an update on transforming growth factor beta signaling in vasculogenesis and angiogenesis. Cardiovasc Res 2003;58: 20–31. Webb RC. Smooth muscle contraction and relaxation. Adv Physiol Educ 2003;27:201–206.

25 Acute Lung Injury DAVID C. CHRISTIANI AND MICHELLE NG GONG SUMMARY The acute respiratory distress syndrome (ARDS)/acute lung injury is a major cause of morbidity and mortality throughout the world. ARDS is an acute syndrome of lung inflammation and increased permeability associated with severe hypoxia and bilateral infiltrates on chest radiographs with no evidence of left heart failure. This chapter reviews the evidence for genetic determinants in acute lung injury and ARDS with a focus on potential candidate gene polymorphism that may be important in the development and outcome of ARDS. Key Words: Acute lung injury (ALI); acute respiratory distress syndrome (ARDS); angiotensin-converting enzyme (ACE); glutathnione-S-transferase (GST).

INTRODUCTION The acute respiratory distress syndrome (ARDS)/acute lung injury (ALI) is a major cause of morbidity and mortality throughout the world. Annually approx 150,000 cases are reported in the United States with a reported mortality of 40–60%. The American-European Consensus Committee on ARDS defines ARDS as an acute syndrome of lung inflammation and increased permeability associated with severe hypoxia and bilateral infiltrates on chest radiographs with no evidence of left heart failure. Major risk factors for the development of ARDS have been described and include sepsis, trauma, pneumonia, burns, multiple transfusions, cardiopulmonary bypass, and pancreatitis. Other factors such as older age, chronic alcoholism, tobacco abuse, absence of diabetes, and greater severity of illness have also been found to contribute to the risk of developing ARDS. Despite the common occurrence of these risk factors, only a minority of patients with the acute injuries listed above develops ARDS. It is likely that given the same type and degree of insult, there are individual differences in susceptibility to developing ARDS. Since the initial description of ARDS in 1967, research has focused on defining the pathogenesis, clinical presentation, course, and outcome of the syndrome. Initially, studies investigated the role of complement and endotoxin in lung injury. Then research concentrated more on the role of inflammation in the pathogenesis and course of ALI/ARDS. Additionally, clinical studies have evaluated variables that may influence the development and outcome of

From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

ALI. Although many animal studies have been consistent, many human studies have reported conflicting results. Consequently, efforts to find clinical characteristics or biomarkers to predict, diagnose, or prognosticate outcomes in ARDS have been often disappointing. The understanding of why some patients develop and die from ARDS whereas others do not is incomplete. For example, although major risk factors for ARDS have been identified, most patients with these risk factors do not develop ARDS. Only approx 4% of patients with documented bacteremia and 41% of patients with sepsis syndrome develop ARDS. Likewise, the search for molecular biomarkers in ARDS has been disappointing. Some studies have found increased plasma levels of tumor necrosis factor (TNF)-α to correlate with the development of ARDS and the severity and mortality in ALI. Other studies from different institutions did not detect the same association. Variable timing, sample type, and method of measurement may explain some of the conflicting results on cytokines in ARDS. The role of genetic variability in the development and course of ALI had previously not been considered. However, discoveries about the genetic control and regulation of the innate immune defense and inflammatory response have raised the question of whether the multiple polymorphic alleles of genes that encode for cytokines and other mediators of inflammation may result in phenotypic differences in host inflammatory response. These differences may account for some of the heterogeneity in individual susceptibility to, and prognosis in, ARDS. This chapter reviews the evidence for genetic determinants in ALI and ARDS with a focus on potential candidate gene polymorphism that may be important in the development and outcome of ARDS.

EVIDENCE FOR GENETIC DETERMINANTS OF ALI/ARDS Almost all cases of ARDS/ALI occur as a complication of an initial injury such as sepsis, pneumonia, trauma, or aspiration. Whether a patient develops ARDS as a result of this initial injury depends partly on environmental factors such as comorbid diseases and treatment adequacy. Genetic variability is likely to affect development of ARDS in multiple aspects of this pathway (Fig. 25-1). There may be heritable determinants to the likelihood of developing ARDS after the initial injury. Alternatively, there may be variable genetic susceptibility to developing and manifesting the initial injury of sepsis or pneumonia. The likelihood by which a person with infection presents with shock may be genetically influenced.

231

232

SECTION III / PULMONARY DISEASES

relative risk (RR) of dying prematurely from infections in an adopted individual is 5.81 (95% confidence interval [CI] 2.47–13.7) if one biological parent died before the age of 50 from infection. In contrast, if the adopted parent were to die prematurely of infection, the RR of dying from infection was 0.73 (95% CI 0.1–5.36) in the adoptee. These data suggest that premature death from infection has a stronger genetic than environmental component.

POTENTIAL CANDIDATE GENES IN ALI/ARDS

Figure 25-1 Genetic determinants in the development of ARDS. Genetic variability is likely to play a role in multiple aspects in the pathogenesis of ARDS from the susceptibility to developing the initial injury, to variable presentation of the injury (i.e., with or without shock) and ultimately to the likelihood of the progression from initial injury to ARDS.

Finally, genetic variability may influence outcome once ALI/ ARDS develops. Classic approaches to determining the genetic vs environmental contribution to diseases using twin, adoptees, or families to determine heritability are not feasible in ARDS. Workable criteria for a consistent diagnosis of ARDS were not available until 1988. The current definition of ARDS and ALI was proposed in 1994. Thus, it is likely that ARDS was underdiagnosed in the past. Additionally, the high mortality and older age of onset of ALI/ARDS limits the availability of affected family members for analysis. Thus, it is not surprising that there are no known family aggregates of ARDS. Nevertheless, intriguing studies indicate likely genetic determinants in ARDS. There have been multiple reports of recurrent ARDS in some individuals. One report describes six patients with multiple episodes of biopsy-proven diffuse alveolar damage. The discovery of a group of patients prone to repeated lung injury raises the possibility of an inherent, possibly heritable, predisposition to ARDS. Because mortality in ARDS is high, it is not surprising that these cases of recurrent ARDS are rare. However, with mortality in ARDS improving, genetically susceptible patients may be more likely to survive their first bout of lung injury to develop another episode. Although there are no family studies on ARDS, there are intriguing twin and adoption studies on serious infections and premature deaths. Genetic susceptibility to severe infection has important implications for ALI. Sepsis is the leading cause of ALI/ARDS and is associated with worse mortality than ALI secondary to other etiologies. Bacterial infection is found frequently in ARDS patients, with one autopsy study indicating a prevalence of 98%. Most ARDS fatalities result from refractory infection and sepsis, not from respiratory failure. Thus, it is likely that genetic polymorphisms that are important in developing severe infections may also be important in ALI. In an epidemiological study of 218 pairs of Danish twins, a monozygotic twin whose cotwin died prematurely before the age of 60 had a significantly increased risk of dying prematurely. In another study involving 960 families with adopted children, the

As a result of these studies, there have been some preliminary investigations into genetic susceptibility in sepsis and ALI. Although promising, these preliminary studies also demonstrate the difficulties involved in designing genetic studies on complex diseases like ALI/ARDS. A number of common genetic variants may have a role in ALI/ARDS because of their role in altering function (e.g., protein production or post-translational expression) in various pathophysiological steps that are important in the development of ALI/ARDS (Table 25-1). Some of these polymorphisms have been associated with variable protein levels or expression, but caution is advised in interpreting such associations. Plasma levels of protein are a crude assessment of the functional consequence of the polymorphism and a positive or negative association does not necessarily mean the polymorphism is the functional causative loci. Below is a brief evidence-based review of a few genes that can serve as potential candidates for investigation into the genetic susceptibility of ALI. INFLAMMATORY/ANTI-INFLAMMATORY PATHWAY It is generally accepted that the development and evolution of ALI involves the activation of the inflammatory cascade. Instillation of inflammatory cytokines such as TNF-α and interleukin (IL)-6 in animals can result in histological changes identical to ARDS. In human studies, the results of these inflammatory cytokines in ALI have been inconsistent. Proinflammatory cytokines, such as TNFα, IL-1, IL-6, IL-8, and anti-inflammatory modulators such as IL10 and IL-1-receptor antagonists (IL-1ra) have featured prominently in studies of ALI, but there have been conflicting results on their role in predicting the development of or mortality in ALI. Potential reasons abound for the heterogeneous results from these studies of individual biomarkers in ALI. Given the complexity of the condition, it is unlikely that any particular mediator will dominate in the pathogenesis of ALI. Also, the cytokine profile may differ by the type of injury that predisposes an individual to ALI. Soluble intercellular adhesion molecule-1 and Eselectin concentrations are significantly higher in ARDS patients with sepsis than in trauma-related ARDS patients. Interest in the role of genetic variability in the development and course of ALI has grown. It is possible that multiple alleles of genes that encode for cytokines and other mediators of inflammation can result in phenotypic differences in host inflammatory response. A number of candidate genes in the inflammatory/anti-inflammatory pathway may be important in the pathogenesis of ALI. PreB-cell colony enhancing factor (PBCEF) is a cytokine that have been previously found in sepsis to inhibit neutrophil apoptosis (Jia et al., 2004). Recently, the PBCEF gene was recently implicated in ALI (Ye, SQ et al., 2005). Increased expression of PBCEF was found in both animal models of ALI and in humans with ALI and increased PBCEF protein was found in the blood and serum of ALI patients. Two single-nucleotide polymorphisms (SNPs) were identified on the PBCEF gene, T-1001G and C-1543T. The G allele of

233

CHAPTER 25 / ACUTE LUNG INJURY

Table 25-1 Potential Candidate Genes for Risk or Clinical Course of ALI/ARDS Potential role in ALI/ARDS Inflammatory cytokines

Anti-inflammtory modulators Coagulation/Fibrinolysis pathway Antioxidant Cell repair Innate immunity

Lung function Fibrosis and inflammatory Unclear

Candidate gene

Association with variable protein levels or function

Association with sepsis or ARDS

PBCEF TNF-β TNF-B IL-6 IL-1ra IL-10 PAI-1 GST HSP TLR-2 TLR-4 MBL SP-B TGF-β1 ACE

Yes Variable Variable Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes

Yes Variable Variable Yes Yes Yes No No No No Yes No Yes No Yes

the T-1001G polymorphism and the GC haplotype was found to be associated with increased odds of developing ALI although no association with mortality in ALI was found. TNF-α has often been studied in ALI. The –308 G to A transition polymorphism in the promoter region of TNF-α gene and the NcoI restriction fragment length polymorphism in intron 1 of the TNF-α gene are in linkage disequilibrium and have found to be associated with variable plasma levels of TNF-α and variable risk for sepsis and cerebral malaria in some but not all studies. We have found that the –308A allele of the TNF-α gene but not the TNFB2 allele of the TNF-α gene was associated with the development of ARDS but only in the presence of a gene–environment interaction with whether the etiology for ARDS was directly pulmonary vs extrapulmonary. The –308A allele was significantly associated with increased mortality in ARDS (Gong et al., 2005). Another potentially important candidate gene in ALI is IL-6. A SNP in position –174 in the promoter region of the IL-6 gene on chromosome 7 has been identified. The GG genotype was found to be associated with increased plasma IL-6 levels although others have found higher IL-6 levels with the C allele. The GG genotype is associated with a twofold increased risk of culture positive sepsis in neonates, whereas in adults, there was no association to sepsis but the GG genotype was associated with survival in sepsis. The variant C allele was not associated with ARDS but it was found in lower frequency in nonsurviving critically ill individuals with and without ARDS. Anti-inflammatory genes that are potentially important in ALI include IL-10 and IL-1ra. Increased plasma levels of IL-10 correlate with severity of illness in pneumonia and mortality in sepsis and menningococcemia. A SNP in position –1082 is associated with increased IL-10 and increased severity of illness and risk of septic shock in patients with pneumonia. A functional pentaallelic polymorphism of IL-1ra consists of 86 basepair variable number of tandem repeats in intron 2. In vitro and in vivo studies have shown allele 2 of this polymorphism to be associated with increased IL-1ra. Allele 2 is associated with sepsis development and, along with polymorphisms of IL-1, with mortality in sepsis. Other gene polymorphisms may be relevant in ALI but have not been associated with sepsis or lung injury. One example is NF-κβ, a regulatory protein for many of the proinflammatory

cytokines such as TNF-α, IL-1, and IL-6. Because it serves as a gatekeeper to many inflammatory mediators and function, variability in the NF-κβ gene could be important in variable susceptibility to ALI. Although polymorphisms of the NF-κβ gene have been identified, none has been associated with functional variability, sepsis, or lung injury. COAGULATION/FIBRINOLYSIS PATHWAY Interaction exists between the activation of the coagulation and fibrinolysis system and the inflammatory/anti-inflammatory pathway in sepsis and ARDS. Disseminated intravascular coagulation is a strong predictor of ARDS and mortality in sepsis. Commonly, microvascular thrombi are found in the lungs of patients with ARDS on autopsy and ARDS associated with disseminated intravascular coagulation has a high mortality rate. Several polymorphisms of coagulation/fibrinolysis factors such as protein C and fibrinogen have been described in thrombotic diseases. Only one, the plasminogen activator inhibitor (PAI)-1 gene, has been implicated in severe infectious disease. PAI-1 is an inhibitor of the fibrinolysis system in that it can complex with plasminogen activator compromising their ability to convert plasminogen to the active form, plasmin. Increased PAI-1 level is a good predictor for ARDS and mortality in ALI. A 4G/5G deletion/insertion polymorphism at position –675 in the promoter region of the PAI-1 gene has been described, and is associated with increased serum PAI-1 levels and worse outcome in mennigococcal disease. It is possible that PAI-1 polymorphism may be associated with risk of development and/or outcome in ARDS. ANTIOXIDANT/CELL REPAIR A number of studies have found a possible imbalance between reactive oxygen species and antioxidants in the pathogenesis of ALI. The glutathione S-transferase (GST) are a family of enzymes that help regulate glutathione, a major cellular antioxidant. Several subtypes of GSTs include GST-α, GST-µ, GST-π, and GST-θ. Multiple polymorphisms of the GST enzymes have been found, some with demonstrated functional consequences. The genes for these subtypes have been implicated in the genetic susceptibility to lung cancer, chest abnormalities in rheumatoid arthritis, severe bronchial hyper-responsiveness in asthma, and chronic obstructive pulmonary disease (COPD). As yet, no clinical studies link the GST genes to sepsis or ALI.

234

SECTION III / PULMONARY DISEASES

Heat shock protein (HSP)-70 is the best-studied member of a group of intracellular stress proteins that protect against cell death from physiological stress like hyperthermia, environmental toxicants, or Gram-negative bacteria. HSP-70 limits lung injury in animal models of ALI. Peripheral blood monocytes from patients with ARDS had significantly lower expression of HSP-70 after hyperthermic stress compared with healthy controls with a positive correlation between HSP-70 inducibility and duration of mechanical ventilation. Three genes encode for HSP-70 on chromosome 6. One, HSP70-2, exhibits a biallelic HSP-70-2 G/A Pst I polymorphism that is associated with a lower expression of mRNA for HSP-70 after stimulation. However, human studies have not demonstrated an association with sepsis or worse outcome in trauma. Another NcoI polymorphism on the HSP70-Hom gene was associated (but not significantly) with increased risk of ARDS (odds ratio 2.1, 95% CI 0.8–5.9) in one small (N = 80) study. Case–control studies of sepsis and ARDS/ALI are needed to evaluate the potential role of HSP-70 gene polymorphisms, disease risk, and outcome. INNATE IMMUNITY As noted, sepsis and ALI are closely linked. Sepsis is a leading cause of ARDS and the leading cause of mortality in ARDS. Thus, it is possible that interindividual variation in host defense may also influence susceptibility to, and mortality in, ALI. To defend against infection, the innate immune system needs to be able to detect foreign pathogens, eliminate them rapidly before they can propagate or if needed, trigger a cascade of specific inflammatory response to contain and eradicate the organism. Thus, genes involved in innate immunity may be important in the pathogenesis of ALI. A family of toll-like receptors (TLR) binds to bacterial constituents such as lipopolysaccharide and institutes the appropriate innate response. A SNP at A-896G in exon 4 of the TLR-4 gene is associated with a replacement of an aspartic acid residue with glycine at amino acid 299 (Asp299Gly) and results in an alteration of the extracellular domain of the TLR-4. The variant allele of this polymorphism is associated with hyporesponsiveness to inhaled lipopolysaccharide. Although one study did not find any association with meningococcal disease, others found that the variant allele was associated with septic shock and Gram-negative infections. A SNP in TLR-2, G2258A, was associated with decreased response to bacterial peptides derived from Borrelia burgdorferi and Treponema pallidum. In one study, the variant allele was found in 2/22 (9%) of patients with Gram-positive septic shock compared with 0/69 (0%) of patients with other types of septic shock, although the sample size was too small for analysis. It is possible that polymorphism in TLR-2 or TLR-4 genes may play a role in susceptibility to ARDS/ALI. Another potential candidate gene in ALI is the mannosebinding lectin (MBL) gene. MBL is a serum protein in the innate immune system that binds to polysaccharides on the surface of various bacteria and viruses, and facilitates opsonization and phagocytosis. SNPs have been identified on exons 1, 3 (known collectively as allele O) and in the promoter region of the MBL gene. The variant alleles of these polymorphisms are associated with lower serum levels of MBL. These variant alleles are associated with an increased risk of meningococcal disease, increased infections in children and patients with COPD and lupus, and greater severity of illness and worse outcomes in patients with cystic fibrosis. No studies of MBL variants in ARDS have been published.

OTHER POTENTIAL CANDIDATE GENES A deletion polymorphism in the angiotensin-converting enzyme (ACE) gene, which is associated with higher ACE levels and activity, has been implicated in a number of diseases including myocardial infarction and hypertension. The biological role of ACE in sepsis and ALI is unclear. Most studies report ACE levels or activity to be low in ARDS. However, the high ACE producing DD genotype was found to be associated with severe meningococcal disease, ARDS, and increased mortality in ARDS. Further study is needed to confirm this finding. After an acute injury, the lung often responds with evidence of fibrogenesis. Evidence of early pulmonary fibrosis in biopsy of ARDS patients has also been associated with a poorer outcome. Transforming growth factor (TGF)-β is likely to play a key role in the fibrinogenesis of the lung in ARDS. TGF-β is a cytokine produced in response to tissue injury. Its actions include the regulation and inhibition of inflammatory cytokines like TNF-β, IL-1, platelet-derived growth factors, and fibroblast growth factor; the induction of extracellular matrix deposition; the inhibition of protease degradation of the matrix; and the modulation of the expression of integrins to increase cellular adhesion to the matrix. Increased TGF-β mRNA has been found in the rat model of shock-induced ARDS. A SNP at position –509 in the promoter region of TGF-β1 gene has been described that is associated with variable TGF-β levels. Polymorphisms in TGF-β1 may be useful in predicting susceptibility to developing fibrosis, making them potential candidate genes for ALI. Pulmonary surfactant is synthesized primarily by type-II alveolar cells and it has multiple important functions in the lung including lowering the surface tension on the alveolar surface and enhancing bacterial phagocytosis and chemotaxis of alveolar macrophages. One surfactant protein (SP), SP-B is predictive of ARDS development. Polymorphisms in the SP-B gene have been associated with ARDS in a few studies. In one small study, the frequency of an insertion/deletion variant polymorphism in intron 4 was 46.6% among 15 ARDS patients in contrast to 4.3% in the control group of normal blood donors (p < 0.05). An increased association has been found between the variant SP-B polymorphism and ARDS in women at risk for ARDS secondary to sepsis, aspiration, pneumonia, trauma, or massive transfusion after adjusting for age, race, and severity of illness. No association was found for the men, indicating possible gender modification of the risk conferred by the variant polymorphism of the SP-B gene. Another study found an association between the (1580C/T missense mutation in exon 4 of the SP-B gene and the development of ARDS especially in those patients with lung injury from predominantly direct pulmonary insults such as pneumonia. Another study on a different population of patients with community acquired pneumonia confirmed the association between the –1580 C allele and ARDS (Quasney et al., 2004).

ALI AS A COMPLEX DISORDER Although there are likely to be genetic determinants in the development and evolution of ALI, elucidating these determinants will not be straightforward, because ALI is a complex disorder with complex genetic and exogenous determinants. The innate immune and inflammatory response, like many other mechanisms in physiology, involves a number of integrated biochemical and physiological systems that respond to and are modulated by environmental stimuli. To ensure stability and adaptability, the inflammatory

CHAPTER 25 / ACUTE LUNG INJURY

response has both pro- and anti-inflammatory pathways with builtin redundancies, biofeedback loops for modulation, and counterregulatory mechanisms. Although this system is under genetic regulation, it is likely to be controlled by multiple genes (genetic heterogeneity) with interaction between genes (gene–gene interaction) and between gene and exogenous stimuli such as infection, trauma, or other lung injuries (gene–environment interaction). The redundancies built into the system could result in a threshold effect in which the function of several genes needs to be affected before ALI will be manifested. Because of these multiple interactions, any single susceptibility gene in ALI will exhibit incomplete penetrance (i.e., not all individuals with the gene polymorphism will develop ALI). Epidemiologically, penetrance translates to RR, in which a highly penetrant gene corresponds to high RR of disease for the individual, whereas a low-penetrant gene confers a low to moderate RR of the disease. The role of the environment is particularly critical in influencing the genetic determinants in a complex disorder like ALI. A predisposing injury is essential to its development. In two prospective studies on the development of ARDS, 78% of the ARDS cases developed after known conditions such as trauma, aspiration, sepsis, massive transfusion, drug overdose, near drowning, or pneumonia. In one of these studies, only 9% of the patients did not have a clearly recognized acute condition leading to ARDS. In addition, the type of injury affects the incidence of ARDS, the cytokine profile, and mortality. Thus, it is likely that any genetic determinants of ALI/ARDS will be modified by the type of injury that led to lung injury. Just as it is important to account for tobacco use in genetic studies of lung cancer and COPD, it is important to account for the type of injury that predisposes to lung injury in studies of populations at risk for ARDS/ALI. Because of incomplete penetrance and interactions with other factors such as age, other genes and the environment, susceptibility genes for complex diseases like ALI may not determine the development and outcome of the disease. Rather, these gene polymorphisms, singly or in combination, will affect the probability of developing the syndrome after an injury. Traditionally, epidemiology has been concerned with discovering causal associations for diseases and disorders. Molecular epidemiology combines the use of molecular biological techniques such as modern genetics with epidemiology to identify and characterize disease in populations in the context of environmental exposures. In sorting out the genetic basis for lung injury, the discipline and tools of classic epidemiology will be important in study design and in accounting for multiple interactions and environmental influences.

CONCLUSION The use of molecular epidemiology to better understand the genetic basis for ALI on a population level has important implications for research. This approach is an example of translational research in which significant findings in the laboratory are examined on a population level to determine their relative contribution to actual disease occurrences. There is also the potential for “backward” translation. Given the large number of molecules implicated in the pathogenesis of lung injury and the complex interactions between different molecular pathways, it is not always easy to discern in the laboratory which mechanism may be most important. The identification of a particular gene or genetic polymorphism that is important in a large population and the identification of environmental conditions in which genetic contribution is strongest can

235

help guide the molecular biologist to focus on the protein or molecular pathway with the greatest potential impact for the largest number of patients. In addition, the identification of a group of patients who may be genetically at a higher risk for the development of ARDS under certain conditions has important implications for prevention and treatment. A goal of epidemiology is the identification of risk factors by which a population can be identified for future intervention. Although an individual’s genetic susceptibility cannot be changed, the physician and the patient can alter other risk factors to decrease the overall risk of disease. In addition, the identification of individuals who are genetically susceptible to ARDS can pinpoint subgroups of patients who may most benefit from certain therapeutic modalities. Surfactant replacement therapy was not found to be beneficial in ARDS. However, it is possible that certain individuals with genetic susceptibility to ALI/ARDS because of a polymorphism in the SP genes may be more responsive to surfactant replacement because the lack of surfactant is more pivotal in the pathogenesis of their lung injury than other individuals with high or normal surfactant production. A better knowledge of the genetic predisposition to ALI could lead to individually tailored therapy. Molecular epidemiological studies represent an exciting and novel approach to the study of ALI. It is not likely that a single susceptibility gene produces disease. Rather, susceptibility genes will be additional but important risk factors for determining the ultimate probability of disease. Although it will be undoubtedly challenging, the quest for genetic determinants in ALI holds great promise in helping clarify the risks and outcomes in this condition. Unlike other risk factors for ARDS, the nature of genetics allows for the prospective determination of individuals at high risk for the development of or mortality from ALI. This knowledge will be important in the design of future preventive and therapeutic trials.

ACKNOWLEDGMENTS The above work is supported by Research Grants R01 HL60710 (D. Christiani), ES00002 (D. Christiani), and K23 HL67197 (M.N. Gong) from the National Institutes of Health.

SELECTED REFERENCES Agnese DM, Calvano JE, Hahm SJ, et al. Human toll-like receptor 4 mutations but not CD14 polymorphisms are associated with an increased risk of gram-negative infections. J Infect Dis 2002;186(10): 1522–1525. Arbour NC, Lorenz E, Schutte BC, et al. TLR-4 mutations are associated with endotoxin hyporesponsiveness in humans. Nat Genet 2000;25:187–191. Ashbaugh DG, Bigelow DB, Petty TL, Levine BE. Acute respiratory distress in adults. Lancet 1967;2:319–323. Bell RC, Coalson JJ, Smith JD, et al. Multiple organ system failure and infection in adult respiratory distress syndrome. Ann Intern Med 1983;99:293–298. Bernard G, Artigas A, Brigham KL, et al., and the Consensus Committee. The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes and clinical trial coordinations. Am J Respir Crit Care Med 1994;149(3 Pt 1):818–824. Parsons PE. Mediators and mechanisms of acute lung injury. Clin Chest Med 2000;21(3):467–476. Bersten AD, Hunt T, Nicholas TE, Doyle IR. Elevated plasma surfactant protein-B predicts development of acute respiratory distress syndrome in patients with acute respiratory failure. Am J Respir Crit Care Med 2001;164(4):648–652. Bone RC, Francis PB, Pierce AK. Intravascular coagulation associated with the adult respiratory distress syndrome. Am J Med 1976;61(5):585–589.

236

SECTION III / PULMONARY DISEASES

Border WA, Noble NA. Transforming growth factor β in tissue fibrosis. N Engl J Med 1994;331(19):1286–1292. Brull DJ, Montgomery HE, Sanders J, et al. Interleukin-6 gene –174G>C and –572G>C promoter polymorphisms are strong predictors of plasma interleukin-6 levels after coronary artery bypass surgery. Arterioscler Thromb Vasc Biol 2001;21(9):1458–1463. Brun-Buisson CJ, Bonnet F, Bergeret S, Lemaire F, Rapin M. Recurrent high-permeability pulmonary edema associated with diabetic ketoacidosis. Crit Care Med 1985;13(1):55, 56. Cambien F, Poirier O, Lecerf L, et al. Deletion polymorphism in the gene for angiotensin-converting enzyme is a potent risk factor for myocardial infarction. Nature 1992;359:641–644. Chevrolet JC, Guelpa G, Schifferli JA. Recurrent adult respiratory distress-like syndrome associated with propylthiouracil therapy. Eur Respir J 1991;4:899–901. Clark JG, Milberg JA, Steinberg KP, Hudson LD. Type III procollagen peptide in the adult respiratory distress syndrome. Association of increased peptide levels in bronchoalveolar lavage fluid with increased risk for death. Ann Intern Med 1995;122(1):17–23. Danis VA, Millington M, Hyland V, Grennan D. Cytokine production by normal human monocytes: Inter-subject variation and relationship to an IL-1 receptor antagonist gene polymorphism. Clin Exp Immunol 1995;99:303–310. Dawson SJ, Wiman B, Hamsten A, Green F, Humphries S, Henney AM. The two allele sequences of a common polymorphism in the promoter of the plasminogen-activator inhibitor 1 gene respond differently to interleukin-1 in HepG2 cells. J Biol Chem 1993;268:10,739–10,745. Deitch EA, Beck SC, Cruz NC, De Maio A. Induction of heat shock gene expression in colonic epithelial cells after incubation with Escherichia coli or endotoxin. Crit Care Med 1995;23:1371–1376. Donnelly TJ, Meade P, Jagels M, et al. Cytokine, complement and endotoxin profiles associated with the development of adult respiratory distress syndrome after severe injury. Crit Care Med 1994;22:768–776. Doyle IR, Bersten AD, Nicholas TE. Surfactant proteins-A and -B are elevated in plasma of patients with acute respiratory failure. Am J Respir Crit Care Med 1997;156(4 Pt 1):1217–1229. Durand P, Bachelet M, Brunet F, et al. Inducibility of the 70 kD heat shock protein in peripheral blood monocytes is decreased in human acute respiratory distress syndrome and recovers over time. Am J Respir Crit Care Med 2000;161(1):286–292. Entzian P, Huckstadt A, Kreipe H, Barth J. Determinations of serum concentrations of type III procollagen peptide in mechanically ventilated patients. Am Rev Respir Dis 1992;142:1079–1082. Farjanel J, Hartmann DJ, Guidet B, Luquel L, Offenstadt G. Four markers of collagen metabolism as possible indicators of disease in the adult respiratory distress syndrome. Am Rev Respir Dis 1993;147:1091– 1099. Ferring M, Vincent JL. Is outcome from ARDS related to the severity of respiratory failure? Eur Respir J 1997;10:1297–1300. Fishman D, Faulds G, Jeffery R, et al. The effect of novel polymorphisms in the interleukin-6 gene on IL-6 transcription and plasma IL-6 levels, and an association with systemic-onset juvenile chronic arthritis. J Clin Invest 1998;102:1369–1376. Fourrier F, Chopin C, Goudemand J, et al. Septic shock, multiple organ failure and disseminated intravascular coagulation. Compared patterns of anti-thrombin III, protein C, and protein S deficiencies. Chest 1992;101:816–823. Fourrier F, Chopin C, Wallaert B, Mazurier C, Mangalaboyi J, Durocher A. Compared evolution of plasma fibronectin and angiotensin-converting enzyme levels in septic ARDS. Chest 1985;87:191–195. Fowler AA, Hamman RF, Good JT, et al. Adult respiratory distress syndrome: risk with common predispositions. Ann Intern Med 1983;98:593–597. Fryer AA, Bianco A, Hepple M, Jones PW, Strange RC, Spiteri MA. Polymorphism at the glutathione S-transferase GSTP1 locus: a new marker for bronchial hyperresponsiveness and asthma. Am J Respir Crit Care Med 2000;161:1437–1442. Gallagher PM, Lowe G, Fitzgerald T, et al. Association of IL-10 polymorphism with severity of illness in community acquired pneumonia. Thorax 2003;58:154–156.

Gando S, Nakanishi Y, Tedo I. Cytokines and plasminogen activator inhibitor-1 in posttrauma disseminated intravascular coagulation: Relationship to multiple organ dysfunction syndrome. Crit Care Med 1995;23(11):1835–1842. Garred P, Madsen HO, Halberg P, et al. Mannose-binding lectin polymorphisms and susceptibility to infection in systemic lupus erythematosus. Arthritis Rheum 1999;42:2145–2152. Garred P, Pressler T, Madsen HO, et al. Association of mannose-binding lectin gene heterogeneity with severity of lung disease and survival in cystic fibrosis. J Clin Invest 1999;104:431–437. Glynn P, Coakley R, Kilgallen I, Murphy N, O’Neill S. Circulating interleukin 6 and interleukin 10 in community acquired pneumonia. Thorax 1999;54:51–55. Gong MN, Zhou W, Williams PL, et al. –308GA and TNFB polymorphisms in acute respiratory distress syndrome. Eur Respir J 2005;26:382–389. Gong MN, Zhou W, Xu L, Miller D, Thompson BT, Christiani DC. Polymorphism in the surfactant protein-B gene, gender and the risk of direct pulmonary injury and ARDS. Chest 2004;125:203–211. Gonzolez ER, Cole T, Grimes MM, Fink RA, Fowler AAI. Recurrent ARDS in a 39-year-old woman with migraine headaches. Chest 1998; 114:919–922. Grainger DJ, Heathcote K, Chiano M, et al. Genetic control of the circulating concentration of transforming growth factor type β1. Hum Mol Genet 1999;8(1):93–97. Harding D, Baines PB, Brull D, et al. Severity of meningococcal disease in children and the angiotensin-converting enzyme insertion/deletion polymorphism. Am J Respir Crit Care Med 2002;165:1103–1106. Harding D, Dhamrait S, Millar A, et al. Is interleukin-6 –174 genotype associated with development of septicemia in preterm infants? Pediatrics 2003;112:800–803. Headley AS, Tolley E, Meduri GU. Infections and the inflammatory response in acute respiratory distress syndrome. Chest 1997;111:1306–1321. Hermans PW, Hibberd ML, Booy R, et al. 4G/5G promoter polymorphism in the plasminogen-activator-inhibitor-1 gene and outcome of meningococcal disease. Lancet 1999;354:556–560. Hibberd ML, Sumiya M, Summerfield JA, Booy R, Levin M. Association of variants of the gene for mannose-binding lectin with susceptibility to meningococcal disease. Meningococcal Research Group. Lancet 1999;353:1049–1053. Hierholzer C, Kalff JC, Omert L, et al. Interleukin-6 production in hemorrhagic shock is accompanied by neutrophil recruitment and lung injury. Am J Physiol 1998;275:L611–L621. Hoffman R, Claypool W, Katyal S, Singh G, Rogers R, Dauber J. Augmentation of rat alveolar macrophage migration by surfactant protein. Am Rev Respir Dis 1987;135:1358–1362. Hudson LD, Milberg JA, Anardi D, Maunder RJ. Clinical risks for development of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1995;151:293–301. Hulkkonen J, Pertovaara M, Antonen J, Pasternack A, Hurme M. Elevated interleukin-6 plasma levels are regulated by the promoter region polymorphism of the IL6 gene in primary Sjogren’s syndrome and correlate with clinical manifestation of the disease. Rheumatology 2001;40:656–661. Hurme M, Santtila S. Il-1 receptor antagonist plasma levels are co-ordinately regulated by both IL-1ra and IL-1β genes. Eur J Immunol 1998;28:2598–2602. Idell S. Coagulation, fibrinolysis and fibrin deposition in acute lung injury. Crit Care Med 2003;31(Suppl. 4):S213–S220. Iribarren C, Jacobs DR, Jr, Sidney S, Gross MD, Eisner MD. Cigarette smoking, alcohol consumption, and risk of ARDS: a 15-year cohort study in a managed care setting. Chest 2000;117:163–168. Ishii T, Matsuse T, Teramoto S, et al. Glutathione S-transferase P1 (GSTP1) polymorphism in patients with chronic obstructive pulmonary disease. Thorax 1999;54:693–696. Jardin F, Fellahi JL, Beauchet A, Vieillard-Baron A, Loubieres Y, Page B. Improved prognosis of acute respiratory distress syndrome 15 years on. Intensive Care Med 1999;25:936–941. Jia SH, Li Y, Parodo J, Kapus A, Fan L, Rotsein OD, Marshall JC. Pre-B cell colony-enhancing factor inhibits neutrophil apoptosis in experimental inflammation and clinical sepsis. J Clin Invest 2004;113:1318–1327.

CHAPTER 25 / ACUTE LUNG INJURY

Johnson J, Brigham KL, Jesmok G, Meyrick B. Morphologic changes in lungs of anesthetized sheep following intravenous infusion of recombinant tumor necrosis factor α. Am Rev Respir Dis 1991;144(1): 179–186. Johnson AR, Coalson JJ, Ashton J, Larumbide M, Erdos EG. Neutral endopeptidase in serum samples from patients with adult respiratory distress syndrome. Comparison with angiotensin-converting enzyme. Am Rev Respir Dis 1985;132:1262–1267. Kwiatkowski D, Hill A, Sambou I, et al. TNF concentration in fatal cerebral, non-fatal cerebral and uncomplicated Plasmodium falciparum malaria. Lancet 1990;336:1201–1204. Lane DA, Grant PJ. Role of hemostatic gene polymorphisms in venous and arterial thrombotic disease. Blood 2000;95(5):1517–1532. Lang JD, McArdle PJ, O’Reilly PJ, Matalon S. Oxidant-antioxidant balance in acute lung injury. Chest 2002;122(Suppl. 6):314S–320S. Lin Z, Pearson C, Chinchilli V, et al. Polymorphisms of the human SP-A, SP-B, and SP-D genes: Association of SP-B Thr131Ile with ARDS. Clin Genet 2000;58:181–191. Lindquist S. The heat shock response. Annu Rev Biochem 1986;55: 1151–1191. Lipscombe RJ, Sumiya M, Hill AV, et al. High frequencies in African and non-African populations of independent mutations in the mannosebinding protein gene. Hum Mol Genet 1992;1:709–715 (Erratum in: Hum Mol Genet 1993 2(3):342). Ma P, Chen D, Pan J, Du B. Genomic polymorphism within interleukin-1 family cytokines influences the outcome of septic patients. Crit Care Med 2002;30(5):1046–1050. Madsen HO, Garred P, Kurtzhals JAL, et al. A new frequent allele is the missing link in the structural polymorphism of the human mannanbinding protein. Immunogenetics 1994;40:37–44. Madsen HO, Garred P, Thiel S, et al. Interplay between promoter and structural gene variants control basal serum levels of mannan-binding protein. J Immunol 1995;155:3013–3020. Majetschak M, Flohe S, Obertacke U, et al. Relation of a TNF gene polymorphism to severe sepsis in trauma patients. Ann Surg 1999;230(2): 207–214. Marshall RP, Webb S, Hill MR, Humphries SE, Laurent GJ. Genetic polymorphisms associated with susceptibility and outcome in ARDS. Chest 2002;121(Suppl 3):68S–69S. Marshall RP, Webb S, Bellingan GJ, et al. Angiotensin converting enzyme insertion/deletion polymorphism is associated with susceptibility and outcome in acute respiratory distress syndrome. Am J Respir Crit Care Med 2002;166:646–650. Martin C, Papazian L, Payan MJ, et al. Pulmonary fibrosis correlates with outcome in adult respiratory distress syndrome: A study in mechanically ventilated patients. Chest 1995;107(1):196–200. Mattey DL, Hassell AB, Dawes PT, et al. Influence of polymorphism in the manganese superoxide dismutase locus on disease outcome in rheumatoid arthritis: Evidence for interaction with glutathione S-transferase genes. Arthritis Rheum 2000;43(4):859–864. Max M, Pison U, Floros J. Frequency of SP-B and SP-A1 gene polymorphism in the acute respiratory distress syndrome (ARDS). Appl Cardiopulm Pathophysiol 1996;6:111–118. McGue M, Vaupel JW, Holm N, Harvald B. Longevity is moderately heritable in a sample of Danish twins born 1870–1880. J Gerontol 1993;48(6):B237–B244. McKay CJ, Gallagher G, Brooks B, Imrie CW, Baxter JN. Increased monocyte cytokine production in association with systemic complications in acute pancreatitis. Br J Surg 1996;83:919–923. McWilliams JE, Sanderson BJ, Harris EL, Richert-Boe KE, Henner WD. Glutathione S-transferase M1 (GSTM1) deficiency and lung cancer risk. Cancer Epidemiol Biomarkers Prev 1995;4(6):589–594. Meduri GU, Tolley EA, Chrousos GP, Stentz F. Prolonged methylprednisolone treatment suppresses systemic inflammation in patients with unresolving acute respiratory distress syndrome: Evidence for inadequate endogenous glucocorticoid secretion and inflammation-induced immune cell resistance to glucocorticoids. Am J Respir Crit Care Med 2002;165:983–991. Meduri GU, Headley S, Kohler G, et al. Persistent elevation of inflammatory cytokines predicts a poor outcome in ARDS. Chest 1995;107: 1062–1073.

237

Metnitz PG, Bartens C, Fischer M, et al. Antioxidant status in patients with acute respiratory distress syndrome. Intensive Care Med 1999; 25(2):180–185. Milberg JA, Davis DR, Steinberg KP, Hudson LD. Improved survival of patients with acute respiratory distress syndrome (ARDS): 1983–1993. JAMA 1995;273(4):306–309. Minambres E, Cemborain A, Sanchez-Velasco P, et al. Correlation between transcranial interleukin-6 gradient and outcome in patients with acute brain injury. Crit Care Med 2003;31(3):933–938. Mira JP, Cariou A, Grall F, et al. Association of TNF2, a TNF-α promoter polymorphism, with septic shock susceptibility and mortality. JAMA 1999;282(6):561–568. Montgomery AB, Stager MA, Carrico CH, Hudson LD. Causes of mortality in patients with the adult respiratory distress syndrome. Am Rev Respir Dis 1985;132:485–489. Moss M, Bucher B, Moore FA, Moore EE, Parsons PE. The role of chronic alcohol abuse in the development of acute respiratory distress syndrome in adults. JAMA 1996;275(1):50–54. Moss M, Gillespie MK, Ackerson L, Moore FA, Moore EE, Parsons PE. Endothelial cell activity varies in patients at risk for the adult respiratory distress syndrome. Crit Care Med 1996;24(11):1782–1786. Moss M, Guidot DM, Steinberg KP, et al. Diabetic patients have a decreased incidence of acute respiratory distress syndrome. Crit Care Med 2000;28(7):2187–2192. LaForce FM. Effect of alveolar lining material on phagocytotic and bactericidal activity of lung macrophages against Staphylococcus aureus. J Lab Clin Med 1976;88:691–699. Lorenz E, Mira JP, Frees KL, Schwartz DA. Relevance of mutations in the TLR-4 receptor in patients with gram-negative septic shock. Arch Intern Med 2002;162:1028–1032. Murray JF, Matthay MA, Luce JM, Flick MR. An expanded definition of the adult respiratory distress syndrome. Am Rev Respir Dis 1989; 138(3);720–723. O’Neill SJ, Lesperance E, Klass DJ. Human lung lavage surfactant enhances staphylococcal phagocytosis by alveolar macrophages. Am Rev Respir Dis 1987;135:1358–1362. Orfanos SE, Armaganidis A, Glynos C, et al. Pulmonary capillary endothelium-bound angiotensin-converting enzyme activity in acute lung injury. Circulation 2000;102(16):2011–2018. Park WY, Goodman RB, Steinberg KP, et al. Cytokine balance in the lungs of patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2001;164(10 Part 1):1896–1903. Parsons PE, Moore FA, Moore EE, Ikle DN, Henson PM, Worthen GS. Studies on the role of tumor necrosis factor in adult respiratory distress syndrome. Am Rev Respir Dis 1992;146:694–700. Pittet JF, Mackersie RC, Martin TR, Matthay MA. Biological markers of acute lung injury: Prognostic and pathogenetic significance. Am J Respir Crit Care Med 1997;155(4):1187–1205. Pociot F, Ronningen KS, Nerup J. Polymorphic analysis of the human MHC-linked heat shock protein 70 and HSP 70-HOM genes in insulin dependent diabetes mellitus. Scand J Immunol 1993;38(5):491–495. Prabhakaran P, Ware LB, White KE, Cross MT, Matthay MA, Olman MA. Elevated levels of plasminogen activator inhibitor-1 in pulmonary edema fluid are associated with mortality in acute lung injury. Am J Physiol Lung Cell Mol Physiol 2003;285:L20–L28. Quasney MW, Waterer GW, Dahmer MK, et al. Association between surfactant protein B + 1580 polymorphism and the risk of respiratory failure in adults with community-acquired pneumonia. Crit Care Med 2004;32:1115–1119. Quinlan GJ, Lamb NJ, Tilley R, Evans TW, Gutteridge JM. Plasma hypoxanthine levels in ARDS: implications for oxidative stress, morbidity and mortality. Am J Respir Crit Care Med 1997;155(2): 479–484. Read RC, Pullin J, Gregory S, et al. A functional polymorphism of tolllike receptor 4 is not associated with likelihood or severity of meningococcal disease. J Infect Dis 2001;184:640–642. Rinaldo J, Christman JW. Mechanisms and mediators of the adult respiratory distress syndrome. Clin Chest Med 1990;11(4):621–632. Rocco TR Jr, Reinert SE, Cioffi W, Harrington D, Buczko G, Simms HH. A 9-year, single-institution, retrospective review of death rate and

238

SECTION III / PULMONARY DISEASES

prognostic factors in adult respiratory distress syndrome. Ann Surg 2001;233:414–422. Russell JA. Genetics of coagulation factors in acute lung injury. Crit Care Med 2003;21(Suppl 4):S243–S247. Savici D, Katzenstein AL. Diffuse alveolar damage and recurrent respiratory failure: report of 6 cases. Hum Pathol 2001;32(12):1398–1402. Schaaf BM, Boehmke F, Esnaashari H, et al. Pneumococcal septic shock is associated with interleukin-10 –1082 gene promoter polymorphism. Am J Respir Crit Care Med 2003;168(4):476–480. Schluter B, Raufhake C, Erren M, et al. Effect of the interleukin-6 promoter polymorphism (–174 G/C) on the incidence and outcome of sepsis. Crit Care Med 2002;30:32–37. Schroder O, Schulte KM, Osterman P, Roher HD, Ekkernkamp A, Schulte KM. Heat shock protein 70 genotypes HSPA1B and HSPA1L influence cytokine concentrations and interfere with outcome after major injury. Crit Care Med 2003;31(1):73–79 (Erratum in: Crit Care Med 2003;31[4]1296). Schroeder S, Borger N, Wrigge H, et al. A tumor necrosis factor gene polymorphism influences the inflammatory response after cardiac operation. Ann Thorac Surg 2003;75:534–537. Schroeder S, Reck M, Hoeft A, Stuber E. Analysis of two human leukocyte antigen-linked polymorphic heat shock protein 70 genes in patients with severe sepsis. Crit Care Med 1999;27(7):1265–1270. Shenkar R, Coulson WF, Abraham E. Anti-transforming growth factor-β monoclonal antibodies prevent lung injury in hemorrhaged mice. Am J Respir Cell Mol Biol 1994;11(3):351–357. Sorensen TI, Nielsen GG, Andersen PK, Teasdale TW. Genetic and environmental influences on premature death in adult adoptees. N Engl J Med 1988;318(12):727–732. Spragg RG, Lewis JF, Wurst W, et al. Treatment of acute respiratory distress syndrome with recombinant surfactant protein C surfactant. Am J Respir Crit Care Med 2003;167(11):1562–1566. Steinberg KP, McHugh LG, Hudson LD. Causes of mortality in patients with adult respiratory distress syndrome. Am Rev Respir Dis 1993;147:A347. Strange RC, Spiteri MA, Ramachandran S, Fryer AA. Glutathione-Stransferase family of enyzmes. Mutat Res 2001;482:21–26. Stuber F, Petersen M, Bokelmann F, Schade U. A genomic polymorphism within the tumor necrosis factor locus influences plasma tumor necrosis factor-α concentrations and outcome of patients with severe sepsis. Crit Care Med 1996;24(3):381–384. Stuber F, Udalova IA, Book M, et al. –308 Tumor necrosis factor (TNF) polymorphism is not associated with survival in severe sepsis and is unrelated to lipopolysaccharide inducibility of the human TNF promoter. J Inflamm 1996;46:42–50. Suarez M, Krieger BP. Bronchoalveolar lavage in recurrent aspirininduced adult respiratory distress syndrome. Chest 1986;90:452, 453. Summerfield JA, Sumiya M, Levin M, Turner MW. Association of mutations in mannose binding protein gene with childhood infection in consecutive infection in consecutive hospital series. BMJ 1997;314: 1229–1232. Suzuki H, Matsui Y, Kashiwagi H. Interleukin-1 receptor antagonist gene polymorphism in Japanese patients with systemic lupus erythematosus. Arthritis Rheum 1997;40:389, 390.

Tang GJ, Huang SL, Yien HW, et al. Tumor necrosis factor gene polymorphism and septic shock in surgical infection. Crit Care Med 2003;28:1733–1736. Turner DM, Williams DM, Sankaran D, Lazarus M, Sinnott PJ, Hutchinson IV. An investigation of polymorphism in interleukin-10 gene promoter. Eur J Immunogenet 1997;24(1):1–8. Van der Poll Y, de Waal Malefyt R, Coyle SM, Lowry SF. Anti-inflammatory cytokine response during clinical sepsis and experimental endotoxemia: sequential measurements of plasma soluble interleukin-1 receptor type II, IL-10, and IL-13. J Infect Dis 1997;175:118–122. Vervloet MG, Thijs LG, Hack CE. Derangements of coagulation and fibrinolysis in critically ill patients with sepsis and septic shock. Semin Thromb Hemost 1998;24(1):33–43. Villar J, Edelson JD, Post M, Mullen JB, Slutsky AS. Induction of heat stress proteins is associated with decreased mortality in an animal model of acute lung injury. Am Rev Respir Dis 1993;147:177–181. Villar J, Ribeiro SP, Mullen JB, Kuliszewski M, Post M, Slutsky AS. Induction of the heat shock response reduces mortality rate and organ damage in a sepsis-induced acute lung injury model. Crit Care Med 1994;22:914–921. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000;342(18):1334–1349. Waterer GW, Quasney MW, Cantor RM, Wunderink RG. Septic shock and respiratory failure in community-acquired pneumonia have different TNF polymorphism associations. Am J Respir Crit Care Med 2001;163(7):1599–1604. Westendorp RGJ, Hottenga JJ, Slagboom PE. Variation in plasminogenactivator–inhibitor-1 gene and risk of meningococcal septic shock. Lancet 1999;354:561–563. Westendorp RG, Langermans JA, Huizinga TW, Verweij CL, Sturk A. Genetic influence on cytokine production and fatal meningococcal disease. Lancet 1997;329:170–173. Westendorp R, Langermans J, Huizinga T, et al. Genetic influence on cytokine production and fatal meningococcal disease. Lancet 1997;349:170–173. Wheeler AP, Bernard GR. Treating patients with severe sepsis. N Engl J Med 1999;340(3):207–214. Yang IA, Seeney SL, Wolter JM, et al. Mannose-binding lectin gene polymorphism predicts hospital admissions for COPD infections. Genes Immun 2003;4(4):269–274. Ye SQ, Simon BA, Maloney JP, et al. Pre-B-cell colony-enhancing factor as a potential novel biomarker in acute lung injury. Am J Respir Crit Care Med 2005;171:361–370. Yost DA, Michalowski E. Adult respiratory distress syndrome complicating recurrent antepartum pyelonephritis. J Fam Pract 1990;31:81, 82. Yoshida K, Ishigami T, Nakazawa I, et al. Association of essential hypertension in elderly Japanese with I/D polymorphism of the angiotensinconverting enzyme (ACE) gene. J Hum Genet 2000;45(5):294–298. Zhang H, Slutsky AS, Vincent JL. Oxygen free radicals in ARDS, septic shock and organ dysfunction. Intensive Care Med 2000;26:474–476. Zilberberg MD, Epstein SK. Acute lung injury in the medical ICU: Comorbid conditions, age, etiology and hospital outcome. Am J Respir Crit Care Med 1998;157(4 Pt 1):1159–1164.

26 Primary Ciliary Dyskinesia PEADAR G. NOONE, MAIMOONA ZARIWALA, AND MICHAEL R. KNOWLES SUMMARY Mucociliary clearance is an important part of airway host defense. Abnormalities in this system may result in disease, for example, cystic fibrosis and primary ciliary dyskinesia (PCD). PCD reflects genetic-based abnormalities of airway ciliary structure, resulting in disease predominantly in the sinuses, middle ear, and lung. Although the biology of PCD has been known for decades, it is only recently that the genetics of the disease have begun to be elucidated. This chapter primarily focuses on the molecular basis of PCD, with a brief review of biology of ciliary structure/function, and the clinical aspects of the disease. Key Words: Axonemes; bronchiectasis; cilia; dynein; mucus; mucociliary clearance; neonatal respiratory distress; nitric oxide; nontuberculous mycobacteria; obstructive pulmonary disease; otitis media; primary ciliary dyskinesia; Pseudomonas aeruginosa; sinusitis; situs inversus.

INTRODUCTION Efficient airway defense is critical for protection of all lung surfaces, including the airways and distal alveolar surfaces, given the exposure of the lung to the potentially harmful contents of the outside atmosphere. Mucociliary clearance (MCC), which includes effective ciliary function, the integrated actions of airway epithelia to regulate surface liquid properties, and mucus secretion, is a key component of this defense system in the lung. Abnormalities in this defense system are reflected in the clinical expression of lung diseases. Cystic fibrosis (CF) is a prototype genetic disease of airway host defense that has been intensively studied, and the molecular basis of classic and nonclassic disease has largely been elucidated, though not completely. Abnormalities in the CF transmembrane regulator result in airway epithelial ion transport defects, and abnormal mucociliary/cough clearance, leading to chronic sinopulmonary disease and bronchiectasis as primary phenotypic manifestations. There is some phenotypic overlap of CF with another genetic disease of airway host defense, primary ciliary dyskinesia (PCD), although the latter has a completely different molecular and cell biological basis. PCD is a disease that reflects genetic abnormalities of airway ciliary structure and function, with a phenotype that reflects the anatomic distribution of ciliary organelles—i.e., otosinopulmonary disease and From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

female and male fertility problems. Although the ciliary ultrastructural and functional defects in PCD were described several decades ago, this disease’s genetics have only begun to emerge. This chapter primarily focuses on the molecular basis of PCD, but first reviews ciliary structure and function and the clinical aspects of PCD.

PATHOGENESIS AND CLINICAL ASPECTS The clinical presentation of PCD, a disease associated with defective ciliary structure and function with associated abnormalities in MCC, is predictable given the anatomic location and function of cilia in the human body. Ciliated epithelia are located in many tissues, including that of the embryo (nodal “monocilia,” important for left–right [L–R] asymmetry), the ependyma of the brain, the middle ear and Eustachian tube, the conducting airways (including the sinuses), and the Fallopian tubes in females. The sperm tail in the male also has a structure and function analogous to that of the cilium. Thus, defective ciliary function results in clinical disease expressed predominantly in those tissues; situs inversus (resulting from abnormal monociliary function in early embryogenesis), otosinopulmonary disease, reduced female fertility, and male infertility. CILIARY ULTRASTRUCTURE AND FUNCTION Because the clinical phenotype is related directly to the location and function of cilia, it is worth considering normal ciliary ultrastructure and function before addressing the spectrum of disease associated with PCD. The molecular aspects of dyneins and related dynein genes are discussed later. Electron microscopic studies have led to significant insights into the structure of human cilia and similar structures. In addition, parallel studies in the flagellate protozoa, Chlamydomonas, have led to an increased understanding of the molecular and biochemical aspects of both normal and abnormal cilia structure and function, given the homology between the organelles of these primitive organisms and those of the human. Cilia are projections of the cell membrane, with a “root system” made up of the ciliary necklace at the base of the cilium, and a ciliary rootlet and basal body embedded in the cell itself (Fig. 26-1). Adjacent cilia are functionally oriented in the same direction, as measured by the direction that the rootlets are facing, which has implications for overall mucociliary function in a synchronous, organized fashion—i.e., with all cilia beating in the same direction. The long axis of the ciliary shaft contains several microtubule-associated proteins (MAP, also termed axonemal proteins), the most obvious of which are dynein arms (DAs), arranged

239

240

SECTION III / PULMONARY DISEASES

Figure 26-1 Top panels: Diagram of a longitudinal and cross-section of a cilium showing complex arrangement of axonemes and dyneins. The central complex and radial spokes are not labeled, but are clearly visible in both this diagram and the images in the bottom panel. (Image courtesy of Johnny Carson PhD, UNC Chapel Hill.) Bottom panel: (A) Cross-sectional electron micrograph images of human cilia from healthy control subject. (B and C) Two subjects with PCD. (A), Normal ODA and normal IDA visible. (B), Normal ODA, missing IDA. (C), Missing/stubby ODA, normal IDA. (Reproduced with permission from Johnny L. Carson.)

in pairs and in a configuration of nine outer microtubule pairs and a central pair (9 + 2 configuration) (see Fig. 26-1). Spokes radiate from the two central microtubules (the central apparatus [CA]) toward the peripheral microtubule doublets. The DAs (inner dynein arm [IDA] and outer dynein arm [ODA]) are spaced at 24-nm intervals along the length of one of the microtubule doublets (microtubule A), and are the “motor” proteins of cilium. Each cilium (5 µm long) contains approx 4000 DAs in pairs. Dynein utilizes ATP to effect conformational changes in the ciliary microtubules relative to each other, such that ciliary bending results. The molecular aspects of DAs are discussed in detail later; however, a brief outline to illustrate the relationship between DA structure/function, and overall ciliary function is valuable at this stage. ODA structure appears less complicated than IDA structure. Electron microscopic techniques to visualize DAs in ciliary crosssections from human epithelial specimens show that the ODA is easier to visualize in the normal and abnormal (missing, shortened) state than the IDA, which can be difficult to distinguish (see Fig. 26-1, bottom panel). The composition and arrangement of the IDA is complex, and varies along the flagellar axoneme (see also below, dynein structure/function). The IDA includes several subunits, termed light, intermediate, and heavy chains. From a functional standpoint, this structural difference between IDA and ODA appears to relate to the different role these structures play in

ciliary function. From a genetic standpoint, different genetic mutations lead to different “ciliary” phenotypes (abnormal ODA vs IDA vs “other”), although the clinical phenotype may be similar. Overall, ciliary beat consists of a power stroke, followed by a relaxation stroke, which makes biological sense when thinking about the swimming actions of protozoa, or the actions of cilia to propel mucus and inhaled “debris” proximally out of the respiratory tract in humans. Evidence from both mutant Chlamydomonas and humans with clinical disease, suggests that the ODA relates primarily to ciliary beat frequency and force, whereas IDA appears to relate more to ciliary bending patterns (“stiff”/“less supple”). For example, several mutant loci in the ODA in Chlamydomonas result in slow swimmers, and humans with abnormal ODA have reduced ciliary activity as compared with humans with abnormal IDA. However, Chlamydomonas with mutant inner DA systems exhibit reduced shear amplitude. Highspeed video microscopy analyses of ciliary activity may further elucidate abnormal ODA and IDA, in terms of specific defects and their relationship with abnormal ciliary frequency/bending/“stiffness.” Regulation of shear force is also required for effective ciliary beat, which involves rotational mechanisms between the CA, the radial spokes, and the outer axonemes. Mutant Chlamydomonas with defective CA show paralysis of flagellar beat, suggesting that this structure acts as a distributor for the regulation of dynein activity. Patients with PCD generally have visible

CHAPTER 26 / PRIMARY CILIARY DYSKINESIA

ultrastructural and functional abnormalities in cilia that lead in vivo to abnormal MCC, loss of the protection afforded by this mechanical clearance system, and, thus, organ level expression of clinical disease manifested mainly in the ear, sinuses, and lung. However, occasionally patients may have structurally normal cilia; yet have a strong clinical “PCD phenotype.” This suggests that there may be a spectrum of PCD; from that of “classic” disease with overt, defined ciliary abnormalities, to variant or nonclassic disease, with less easily defined ciliary ultrastructural defects. Finally, one clinical observation in PCD relates to an unexpected role for another ciliary structure. “Monocilia” are ciliary structures similar to respiratory cilia but lacking a central doublet, which have an important role in early organ development in the embryo. When abnormal in PCD, defective monocilia are associated with another significant phenotypic marker of PCD, situs inversus. ASSOCIATED CLINICAL DISEASE The clinical spectrum of PCD has been well described for decades, although reviews continue to refine the phenotype. A neonatal respiratory syndrome, similar to transient tachypnea of the newborn, is a usual presenting feature of PCD, suggesting a link between effective ciliary function and the clearance of lung liquid in the neonatal period. Another striking phenotypic feature that may be detected at birth is situs inversus totalis (SI, known as Kartagener’s syndrome when accompanied by PCD). Interestingly, SI is not genetically predetermined in patients with PCD, but occurs as a random phenomenon. Studies in a mouse model with abnormal situs suggest a role for nodal primary cilia (monocilia) in cell signaling in the formation of normal L–R asymmetry, such that defective monocilia structure and function lead to randomization of L–R asymmetry in affected animals. Monocilia are structurally different to epithelial cilia, for example, lacking a central doublet, and functionally beat in a rotatory manner, rather than the usual ciliary beat patterns. Thus, the normal rotatory action of monocilia initiates an asymmetric calcium signal at the left side of the node, and thus normal L–R asymmetry. In contrast, abnormalities of monociliary structure and rotatory beat (as occurs in PCD) lead to randomization of L–R cell signaling, and SI. Note that abnormalities in axonemal dynein also result in malfunction of monocilia. The combination of SI with any neonatal respiratory symptoms, or subsequent airway, sinus, or ear infections should prompt a diagnostic workup for PCD. In the vast majority of patients with situs inversus and PCD, the organs are a mirror image of normal, with no other structural or functional defects. As patients with PCD grow older, several phenotypic markers feature prominently in the disease. A series from a large North American cohort of patients showed a phenotypic pattern consistent with earlier reports from Europe and Australia. Symptoms and signs of chronic airways infection, sinusitis, and otitis media are the cardinal features of the disease, and are responsible for the morbidity and mortality associated with PCD. Ear infections and “glue-ear” may be a prominent symptom in childhood, such that patients may be referred first to an otolaryngologist, who thus must retain a high degree of suspicion for the disease in the appropriate setting. Most patients and family members report a chronic cough as a prominent symptom. It appears that cough compensates for the lack of effective MCC in the disease and allows a degree of lung protection. Physiological data have shown that “effective” coughs may result in lung clearance almost equal to normal MCC over short time periods, even in PCD subjects with

241

Figure 26-2 Lung function (forced expiratory volume in the first second [FEV1]; % predicted) vs age in subjects with PCD (n = 70), plotted by the main organisms isolated in airway secretions (some subjects cultured more than one organism). Closed circles represent H. influenza (n = 14), open circles S. aureus (n = 3). Closed squares represent smooth P. aeruginosa (n = 10), open squares represent mucoid P. aeruginosa (n = 12, approx 35% of subjects, most over the age of 30 yr). Closed half circles represent nontuberculous mycobacteria (n = 8 [five of whom also cultured smooth P. aeruginosa]). Open stars represent subjects in whom sputum culture was negative for bacterial pathogens (n = 8), closed stars represent subjects in whom no sample was obtained (n = 15). Thirteen subjects had respiratory failure and an FEV1 ≤ 40% predicted. (Adapted with permission from Noone PG, Leigh MW, Sannuti A, et al., 2004.)

no clearance at baseline. Clinical and radiographic evidence of bronchiectasis usually develops as the disease progresses. Comparisons with CF as a similar disease phenotype are useful, as PCD appears to run a milder course, presumably because of the different pathogenesis of the two diseases, CF being more complex. For example, there are overlaps in the microbiology of PCD airway secretions with CF. One study showed a variety of organisms in the sputum of patients with PCD (Fig. 26-2): Haemophilus influenza, Staphylococcus aureus, Pseudomonas aeruginosa (including mucoid P. aeruginosa in individuals over 30 yr of age) and nontuberculous mycobacteria, for example Mycobacterium avium complex and M. abscessus. The isolation of mucoid P. aeruginosa is interesting, as this organism has traditionally been associated with CF, even at a young age. Its isolation in a significant number of patients with PCD suggests that chronic failure of the mucociliary apparatus allows colonization/infection in the airways with this alginate (hence mucoid) secreting organism. Most patients with these organisms (especially mucoid P. aeruginosa and nontuberculous mycobacteria) in their airway secretions will be screened for CF, but with an appropriate phenotype, consideration should be given to screening for PCD also (see below, diagnostic workup). Generally, PCD is a milder disease than CF, and cross-sectional studies of lung function across a wide age spectrum suggest that the loss of lung function over time is not as rapid as that of CF: approx 0.8%/yr loss of forced expiratory volume in the first second (FEV1) in PCD vs approx 3.6%/yr loss of FEV1 in CF. This likely relates to the complex pathogenesis of CF lung disease (loss

242

SECTION III / PULMONARY DISEASES

of both mucociliary and cough clearance), as compared with the more discrete ciliary defect in MCC in PCD, in addition to the gastrointestinal, endocrine, and bone complications that occur in CF. However, data do suggest a negative impact of the disease on quality of life as PCD patients grow older. There are no generally devised and accepted “state-of-the-art” treatment protocols for PCD as a disease, other than to follow the general rules for that of non-PCD bronchiectasis. Thus, a therapeutic plan for PCD that is probably appropriate involves judicious use of pharmacological methods (e.g., bronchodilators, hypertonic saline, and “directed” antibiotics), and nonpharmacological methods (airway clearance, exercise). Not all drugs for use in CF have usage in other diseases such as PCD; for example, use of DNA-ase, a developed drug for use in CF may be less useful in non-CF bronchiectasis because of a different anatomic distribution of disease in non-CF bronchiectasis. As patients with PCD grow older, many develop more severe symptoms, and most develop digital clubbing. A significant percentage (up to 25% in one series) may develop respiratory failure as defined by hypoxemia, or an FEV1 less than 40% predicted (see Fig. 26-2); a small proportion may eventually require lung transplantation. Male infertility results from abnormal sperm tail function, whereas female fertility is more variable (some women being infertile, with others reporting a delay in conception following unprotected intercourse), presumably resulting from abnormal ciliary function in the Fallopian tube. DIAGNOSIS The diagnosis of PCD requires a compatible clinical phenotype, in combination with laboratory studies demonstrating abnormal ciliary structure and/or function. The traditional approach is to search for ciliary ultrastructural defects (see Fig. 26-1). Samples of nasal ciliated epithelium can easily be obtained from the inferior turbinate using a plastic rhinoprobe and a nasoscope, without the need for local anesthesia, or invasive techniques. Analyses of ciliary cross-sections obtained from these nasal samples show structural abnormalities consisting of absent (or shortened) DAs in most patients with PCD. Defective ODA or IDA, or both, occur in more than 90% of patients with PCD, whereas a few patients may have other miscellaneous abnormalities, such as radial spoke defects, or “ciliary disorientation.” Significant expertise is required in the technical aspects of producing high quality transmission electron micrographs, and in the interpretation of ciliary ultrastructure. For example, IDA defects may be more difficult to interpret than defects of the ODA, for reasons previously outlined. In addition, the IDA may look abnormal in non-PCD patients, presumably resulting from the nonspecific effects of inflammation and infection. Occasionally, patients with a strong phenotype (e.g., otosinopulmonary disease with situs inversus) have apparently normal looking cilia, which, given the complex structure of cilia, probably reflects abnormalities in structures not easily visible on electron microscopy, and may also relate to different genetic mutations encoding for structures other than dyneins, or for functional rather than structural abnormalities. A firm diagnosis is difficult to establish in such patients, and they may be labeled as having “nonclassic” disease, pending further advances in the molecular etiology of PCD in general. From a functional standpoint, in vivo assessments may be determined using measures of nasal MCC (the saccharin test), or measures of lung MCC employing radio isotopic techniques. The significant limitation with the saccharin test is “noise” in the data

generated. Valid results of the test, for example, depend on voluntary suppression of swallow, head tilting, and other positional changes, which may be difficult for children and those with intractable cough. Measures of isotopic clearance from airway surfaces are more quantitative, although less available than other tests. Patients with PCD have no (or little) clearance of isotope from the lung when cough is suppressed. More rigorous techniques using highspeed video microscopy may provide better quantitation of ciliary beat frequency and patterns. Qualitative assessments of ciliary beat can define a dyskinetic beat pattern or immotility, and may provide an approximate, semi-quantitative assessment of “ciliary activity.” Such methodology may yield false negatives, however, as some patients with PCD may have normal looking ciliary activity by this method. Thus, structural assessments of cilia obtained from patients suspected of having PCD are the diagnostic “gold standards” for PCD. These studies of structure and adjunctive functional tests require the availability of significant expertise in terms of equipment and personnel, which may be prohibitive outside academic and research institutions. In addition, there is room for subjective error in diagnosing abnormalities in cilia structure, particularly that of the inner DA. As seen in the next section, nasal nitric oxide (NO) measures provide an additional noninvasive test that reliably helps diagnose PCD. REGULATION OF CILIARY BEAT FREQUENCY AND THE ROLE OF NITRIC OXIDE: NASAL NO AS AN ADJUNCTIVE DIAGNOSTIC TEST Regulation of ciliary beat is complex, and several factors are involved, including intracellular calcium concentrations, cAMP, extracellular nucleotides such as ATP and UTP, and NO. Despite ciliary immotility in most patients with PCD, certain pharmacological agents that modulate ion transport as well as ciliary beat frequency (such as UTP), increase cough clearance in PCD, presumably acting via changes in airway surface liquid rheology rather than any direct actions on MCC. Ion transport appears normal in PCD, in contrast to the deranged sodium and chloride ion transport observed in CF airway epithelia. Biological observations may lead to increased insights into the regulation of ciliary activity in vivo. Early observations of very low levels of nasal NO as measured in air from the upper respiratory tract in patients with PCD have been verified in larger datasets across a wide age and lung function spectrum, including pediatric and adult patients. There is a striking difference in the levels of NO produced in the upper airway of patients with PCD (very low levels) as compared with normal and disease controls (Fig. 26-3). There was no relationship between the levels of NO in PCD and the magnitude of lung disease as measured by FEV1, or ciliary activity, or ciliary structural defect. Further, there was little overlap in nasal NO levels between PCD and disease controls such as CF, so that the test may prove to be a useful diagnostic tool. The mechanism of the low NO in PCD remains elusive. Because NO has been linked to the upregulation of ciliary beat frequency, it is intriguing to speculate that low NO in PCD is related to the primary ciliary defect. NO is formed by the nitric oxide synthase (NOS) enzyme system, consisting of three NOS isoforms (NOS1, NOS2, and NOS3), in the presence of appropriate substrate (L-arginine and oxygen) and cofactors tetrahydrobiopterin. Of these, NOS3 appears to be the best candidate for involvement in ciliary beat regulation, because it is localized close to the base of cilia. NOS3 is activated in endothelium by stress; by analogy,

CHAPTER 26 / PRIMARY CILIARY DYSKINESIA

Figure 26-3 Nasal nitric oxide levels in subjects with PCD (n = 61 in total); n = 16 with both DA defective, n = 32 with ODA defective, and n = 13 with IDA defective. The data are compared with healthy controls (n = 27), and disease subjects with CF (n = 11). No differences were seen between any of the PCD subgroups, but significant differences were seen between PCD and healthy controls (p = 0.0001). Levels of nasal nitric oxide in CF were also different (higher) than PCD (p = 0.0001). (Adapted with permission from Noone PG, Leigh MW, Sannuti A, et al., 2004.)

the “stress” of ciliary bending may activate NOS3 via signaling mechanisms, and stimulate the production of NO, which in turn plays a role in regulation of ciliary beat in an autocrine or paracrine fashion. NOS3 is bound to caveolin-1 at the apical membrane of ciliated airway cells, in which it can be dissociated from this inhibitory binding and also activated by calmodulin and Ca2+. In PCD, one hypothesis to explain low NO levels relates to reduced or absent ciliary “force” leading to reduced NOS3 activity, with a decrease in airway epithelial NO production. To more fully elucidate the role of NO in ciliary regulation, studies addressing NOS expression in PCD vs controls and physiological studies of ciliary beat and NO production in various normal and disease states are needed. Thus, although a mechanistic explanation is lacking, reduced production of NO in the upper airway is a useful phenotypic marker of PCD that may serve as an adjunct “clinical” test in the diagnostic workup in PCD. From a research standpoint, low nasal NO levels in PCD may serve as a useful physiological marker of the deranged ciliary beat/NO axis.

MOLECULAR ASPECTS As discussed, ciliary structure is complex, with many different structural components required for normal function. Thus, when addressing the genetics of PCD, research challenges include the appropriate selection of candidate genes for further study and prioritization of work. Because DA defects comprise the majority of ultrastructural defects in patients with PCD (or stated another way, PCD associated with DA defects is within the limits of diagnostic capability), it seems logical to start with a study of DA structure and function: that is, the ODA, the IDA, and then other “nondynein” ciliary structures (e.g., radial spokes, central doublets).

243

Increased understanding of the structure and function of the various dyneins will augment knowledge of the relations between the various ciliary components and how mutations in different genes result in different “ciliary” phenotypes. DYNEIN ARM STRUCTURE AND RELATED CANDIDATE GENES The axoneme is conserved phylogenetically, and the unicellular organism Chlamydomonas provides the opportunity to study the components of ciliary ultrastructure, and more specifically, DAs, radial spokes, and the CA (Fig. 26-4). Thus, there is a high degree of homology between structural components of Chlamydomonas and human cilia, and genes that encode for structural proteins. The 9 + 2 axoneme contains about 250 well-conserved polypeptides. Dyneins are molecular “motors,” with a mass of approx 1–2 MDa. Up to 15 different heavy chains have been identified. The ODA in Chlamydomonas possesses three dynein heavy chains, two intermediate chains, and at least eight light chains (see Fig. 26-4, top panel). Evidence suggests that mammalian ODA contain only two heavy chains. Several human homologs of Chlamydomonas ODA components are candidates for study in patients with PCD. Dynein heavy chain candidate genes are DNAH5 (γ-heavy chain) and DNAH9 (β-heavy chain), intermediate chain candidate genes are DNAI1 (IC78) and DNAI2 (IC69) and light chain candidate genes include LC1 (which is directly attached to heavy chain motor domain), TCTEX2 (LC2) and LC4. Heavy and intermediate chains are good candidates with which to start, as these are likely to be defective in humans with missing or shortened (“stubby”) ODA, evaluable using electron microscopic techniques, as shown by studies of patients with genetic mutations encoding for heavy or intermediate-chain dyneins (ICs) that show missing or stubby ODA. The IDA appears to be different from the ODA, and is both structurally and functionally diverse (see Fig. 26-4, bottom panel). There are ICs, light-chain dyneins (LC), and heavy-chain dyneins, with different isoforms, with a very complex arrangement. Of several (approx 7–11) dynein heavy chain genes in the IDA, three have been identified (DNAH1, DNAH3, and DNAH7). To generate synchronous ciliary beat, the various IDA motors must coordinate with each other, with the ODA, the central pair and the radial spokes. Thus, elucidation of the many potential IDA defects that might relate to the expression of disease in humans presents significant challenges, particularly with reference to selection of candidate genes for testing in human disease models. Candidate genes for patients with PCD and IDA defects include DNAH1, DNAH3, and DNAH7 (heavy chains), hp28 and TCTEL1 (light chain), and DPCD. Candidate genes for patients with both ODA and IDA defects include DNAI1 and DNAI2 (intermediate chains), and also DNAH3 and LC8, which are expressed by both IDA and ODA. As discussed, other structural components of cilia also require attention because they may be structurally and functionally defective, leading to PCD. A pair of axonemes lies in the center of each cilium—the central pair—and appears to serve an anchoring function for the radial spokes, which themselves connect outward from the central pair to the peripheral doublets (see Fig. 26-4). Twenty-two polypeptides comprise the radial spoke and at least seven Chlamydomonas genes specifically affect the function of the flagellar radial spoke, with similar radial spoke defects observed in humans. The CA is made up of two single microtubules designated as C1 and C2. CA-associated structures

244

SECTION III / PULMONARY DISEASES

Figure 26-4 Top panel: ODA of Chlamydomonas. Human homologs are shown (DNAH5, DNAI1, DNAI2, DNAH9). (Adapted with permission from S.M. King http://www2.uchc.edu/˜king/ .) Bottom panel: diagrammatic representation of the complexity of the arrangement of structures along one axoneme of the outer pair. Visible are the repeating nature of the structures (96 nm), the ODA on top, the radial spokes on the bottom, and the IDA represented by a complex “islands” of structures—the dynein heavy chains (DHC) 1α and 1β in a trilobed structure known as I1 IDA isoform, made up of dynein-heavy chains, three intermediate chains, and three light chains. DRC, dynein regulatory complex; located above the second radial spoke, S2. This illustration shows the complexity of the IDA particularly, and how different mutations in different structures result in IDA function, and thus ciliary abnormalities. (Adapted and reproduced with permission from Porter ME, Sale WS, 2000.)

include the central pair projections, the central pair bridges linking the two tubules and the central pair cap, which are attached to the distal or plus end of the microtubule. In Chlamydomonas, the CA includes 23 polypeptides ranging in molecular weight from 14 to 360 kDa. Ten of these polypeptides are unique to C1 and seven are unique to C2. CA ultimately regulates dynein-driven microtubule sliding in a pathway involving the radial spoke. CA agenesis was reported as causing clinical disease in three related patients with a strong phenotype for PCD. Thus, for patients who have no detectable DA defect, candidate genes include not only genes that encode ODA and IDA function, but also include genes that encode radial spoke proteins (RSHL1, RSP3, and RSHL2), and the CA (PF15, SPAG6, and hPF20). Thus, the opportunity arises for the selection and testing of candidate genes derived from studies of Chlamydomonas, and studies in human disease, in which a well-defined PCD phenotype has been correlated to specific visible electron microscopic defects in ciliary structure. GENETICS PCD is inherited as an autosomal-recessive trait in most cases. Occasional reports suggest an autosomal-dominant inheritance. Digenic mode of inheritance (i.e., a genetic disease caused by interplay of a recessive mutation in two different genes that are functionally related) for PCD has not been seen, but it is

possible, as it occurs in other diseases. As might be predicted from the complexity of ciliary structure and function discussed, the disease is genetically heterogeneous at a ciliary structural level. Hence patients with PCD may resemble each other at a clinical phenotype level, whereas having a diverse array of ciliary abnormalities, and, thus, a completely different genetic basis for the same disease. This presents difficulties for categorization of patients, and identification of mutated genes. However, the emerging definition of ciliary ultrastructure in lower organisms and humans offers an excellent opportunity for the selection of candidate genes. Although the disease appears to be relatively uncommon (at least in terms of a clear diagnosis), the prevalence is estimated to be approx 12,000–17,000 individuals in the United States, as extrapolated from radiographic surveys in Norway and Japan, based on the presence of dextrocardia, in association with clinical evidence of bronchiectasis. It might be hypothesized that PCD, or alternately, lung disease associated with genetic-based abnormalities in ciliary structure and function, is more common than generally appreciated. As the diagnosis may not be obvious or straightforward, especially in the absence of situs inversus, accurate numbers are not available. An added complication is the lack of a central registry for patients with PCD, though efforts are underway with a patient support group to establish such a registry.

CHAPTER 26 / PRIMARY CILIARY DYSKINESIA

Another dimension is the possibility of variant (“nonclassic”) forms of PCD, manifested by a milder or intermediate phenotype, intermediate (or no) changes in ciliary ultrastructure, and with less severe functional derangements in ciliary beat. These patients may have “milder” mutations in genes associated with subtle defects in ciliary structure/function, but still be able to mount a partially effective MCC defense mechanism, in a manner analogous to the scenario that has emerged over the past decade with CF. Identification of the genetic mutations associated with “classic” PCD will enable further study of individuals suspected of having disease related to abnormal ciliary structure and function. For example, splice mutations on one allele might lead to partial protein function conferring a milder (than “classic”) phenotype on that individual. Studies of the level of expression of mRNA from known mutated genes (e.g., splice mutations), performed using methodologies already tested in atypical CF (using e.g., nasal epithelial scrape samples), will likely offer the opportunity to expand such hypotheses and studies to other genes. STRATEGIES FOR IDENTIFYING ASSOCIATED MUTATIONS Genome-wide linkage analysis is one approach to identify candidate loci putatively involved in PCD. This approach requires access to large numbers of families to yield meaningful data, and involves collection of sufficient numbers of DNA samples from affected patients and family members for significant results. More importantly, genetic heterogeneity presents a major challenge, because multiple genes are likely to be involved. For example, in a large study published in 2000, 61 families from Europe and North America were reviewed, and despite sophisticated analyses and biostatistics using 31 multiplex families (169 individuals, including 70 affected with PCD), no major locus for disease was found, although several potential loci on different chromosomes were described. However, the two genes identified as disease causing in PCD (DNAI1 and DNAH5; see below) were not identified using this approach. An alternate way to address the problem of genetic heterogeneity is to subgroup the patients into categories based on the ciliary ultrastructural defect. The categories would include clear-cut DA defects: specifically, in individual patients with an ODA defect alone, an IDA defect alone, either DA defective, or finally those with less common central pair/radial spoke defects. As indicated, this approach relies on high-quality electron microscopic images and rigorous analysis of adequate ciliary samples (ideally with more than one observer, at least one of whom is blinded to the identity of the subject), followed by careful segregation of multiple patients into categories depending on the ciliary ultrastructural defect. Once patients have been categorized as such, genetic testing can be targeted using a stepwise approach: those with ODA defects being tested (using genomic DNA) for known mutations in ODA genes (e.g., DNAI1 and DNAH5), those with IDA defects for IDA mutations (e.g., DNAH7, DPCD), and other genes as the field advances and other candidate genes become known. If a genetic mutation is detected on one allele, full gene sequencing can be conducted to search for the novel mutation on the opposite allele. This strategy was successfully used to isolate mutations in DNAI1 and DNAH5, which have now been established as disease-causing in PCD (see below). To “include” or “exclude” families of interest, testing for patterns of inheritability using intragenic polymorphisms may also be useful if parental/sibling DNA is available. If the mutation is not discovered in these initial tests, other techniques could be used, such as reverse transcription (RT)-PCR to

245

check for the possible splice defects, or Southern blot analysis to check for the possible large rearrangements. GENES TESTED TO BE DISEASE-CAUSING Two genes have been identified as causing PCD—DNAI1 (an IC) and DNAH5 (a heavy-chain dynein), both associated with defective ODA (http://www.gentest.org). Other genes have been tested in variable numbers of patients, and disease-causing mutations have not been found in patients with PCD, for example, DNAH7, DNAH9, DNA12, LC8, hp28, LC4, SPAG6, hPF20, TCTEX2, HFH4/FoxJ1, DPCD, and POLL (Table 26-1). A defined mutation as the molecular basis for PCD was first reported in 1999. DNAI1 (homologous to the Chlamydomonas IC IC78) gene was cloned as a candidate gene because it was found to be disrupted in an ODA-deficient mutant of Chlamydomonas. DNAI1 (a 20 exon gene located on 9p13-p2) was tested in patients with defective ODA, and loss-of-function mutations were reported in one individual with PCD. One mutation was a 4-bp insertion in exon 5 (285′286insAATA), the other was a splice mutation in intron 1 (219 + 3insT now known as IVS1 + 2_3insT); both mutations result in a premature termination signal. Further reports of mutations in DNAI1 in patients with PCD and ODA defects followed. Twelve mutant alleles of DNAI1 have been found from six families tested, of which six mutant alleles harbored the “original” 219 + insT mutations. Two novel mutations were found in the same codon (568) in two other families. One was a missense mutation (W568S) and the other was a nonsense mutation (W568X). In the remaining families, all three carried the 219 + insT mutation on one allele, with novel mutations (G515S and a 12-bp deletion from codons 553 to 556) found on the opposite allele. Further genetic studies of DNAI1 and PCD are ongoing at many centers. Other candidate genes have also been cloned by the group who cloned DNAI1, DNAI2 (orthologous to IC69 in Chlamydomonas), and Hpf20 (orthologous to pf20 in Chlamydomonas). DNAH5 is another candidate gene located on chromosome 5p14p15, which was cloned by a homozgygosity mapping strategy. This gene encodes a protein highly similar to Chlamydomonas γ-dynein heavy chain. Mutations in this gene have been identified in 8 patients from 25 tested families with PCD and ODA defects. One consanguineous family from the United States was homozygous for a splice site mutation in DNAH5 (IVS74→–1G > C), which presumably leads to an inframe deletion of exon 75. This family with the splice mutation is of particular interest because it raises the possibility of measuring levels of gene expression in affected and unaffected (carrier) individuals, as compared with noncarriers. Homozygous disruption of the mouse DNAH5 gene led to a PCDlike phenotype, and this mouse model may be useful to decipher the biochemical interaction of the DNAH5 gene with other ciliary components. OTHER CANDIDATE GENES Although the following genes have not been specifically associated with clinical expression of PCD, these candidates remain of interest for further investigation. For example, a deletion of DNA polymerase (Poll) was reported to cause a ciliary IDA defect in mice associated with hydrocephalus, situs inversus, chronic sinusitis, and male infertility. The mechanism(s) proposed might reflect the absence of the polymerase BRCT motif, which might be necessary for protein–protein interactions and ciliary formation, or transcription of genes necessary for cilia formation. Unexpectedly, work suggests that an alternative candidate gene (Dpcd) in the genomic region around the Poll might be the real gene of interest in the

Table 26-1 Candidate Genes for PCD Chain type/ structure

Gene (human)

ODA defect

Heavy

DNAH5

Chlamy.HC

5p14-p15

S. urchin β-HC

17p12

Intermediate

DNAH9 (DNEL1) DNAI1

Chlamy. IC78

9p21-p13

DNAI2

Chlamy. IC69

17q25

Seq. not available DNLC2A DNAL4

Chlamy. LC1

Not available

Chlamy. LC2? Chlamy. LC4?

20q11.22 22q13.1

TCTE3 (TCTEX2) DNAH3

Murine Tcte3 (Tctex2) Chlamy. HC

6q25-q27 16p12

DNAH7

Chlamy. HC

2q31-q33.2

Seq. not available

Chlamy. IC140

Not available

Seq. not available hp28

Chlamy. IC138

Not available

Chlamy.LC p28

1p35.1

TCTEL1

Chlamy. Tctex1

6p25.2-p25.3

DPCD

Murine Dpcd

10q23

246

Ciliary ultrastructure

Light

IDA defect

Heavy

Intermediate

Light

Others

Homolog

Chromosome (human)

Comments Mutations detected in PCD (eight families tested) Mouse KO yield hydrocephalus, situs inversus, sinusitis, immotile cilia with ODA defect No mutations detected in PCD (two families tested) Mutations detected in PCD (six families tested) No mutations detected in PCD (16 families tested) The only light chain associated with motor domain of γ-HC. Not tested in PCD patients Not tested in PCD patients. No mutation detected in 54 unrelated PCD patients No mutation detected in 36 PCD families Sequence variant of unknown significance detected in PCD (six patients tested) No mutation detected in PCD (one patient tested) IC140 is required for the assembly and docking of I1 complex to the doublet microtubule cargo Critical phosphoprotein required for the regulation of flagellar motility No mutation detected in 65 PCD families (four from UNC) Flagellar inner arm I1 dynein; attaches cargoes to dynein motor This gene resides in a close proximity of DNA POLL. No mutation detected in 51 PCD families

Reference Olbrich H, Haffner K, Kispert A, et al., 2002; Ibanez-Tallon I, Gorokhave S, Heintz N, 2002.

Bartoloni L, Blouin JL, Chung E, et al., 2001. Pennarun G, Escudier E, Chapelin C, et al., 1999; Zariwala M, Noone PG, Sannuti A, et al., 2001; Guichard C, Harricane MC, Lafitte JJ, et al., 2001. Pennarun G, Chapelin C, Escudier E, et al., 2000; Bartoloni L, Mitchison H, Pazour G, et al., 2000. Benashski SE, Patel-King RS, King SM., 1999. Benashski SE, Patel-King RS, King SM., 1999. Benashski SE, Patel-King RS, King SM., 1999. Gehrig C, Albrecht C, Duriaus-Sail G, et al., 2002. Neesen J, Dreckhahn JD, Tiede S, et al., 2002. Blouin JL, Gehrig C, Jeganathan D, et al., 2001. Zhang YJ, O’Neal WK, Randell SH, et al., 2002. Perrone CA, Tritschler D, Taulman P, Bower R, Yoder BK, Porter ME., 2003. Habermacher G, Sale WS., 1996. Gehrig C, Albrecht C, Duriaus-Sail G, et al., 2002. Pennarun G, Bridoux AM, Escudier E, Anselem S, Duriez B., 2001. DiBella LM, Benashski SE, Tedford HW, Harrison A, Patel-King RS, King SM., 2001. Zariwala M, O’Neal WK, Noone PG, Leigh MW, Knowles MR, Ostrowski LE., 2004.

Both DAs defects

Other defects (normal DAs)

Heavy

DNAH3

Chlamy.HC

16p12

Light

LC8

Chlamy.LC8

17q

Heavy

DNAH1

3p21.3

RSP3 (AKAP)

Murine MDHC7 Murine lrd S. urchin β-HC Chlamy. RSP4 and RSP6, S. urchin p63 Chlamy.RSP30

hPF20

Chlamy.PF20

2q34

SPAG6 Seq. Not available

PF16 Chlamy.PF15, KLP1

10p12.2 Not available

HFH4/FoxJ1

Mouse

17q22-q25

DNAH11 Radial spoke

Central apparatus genes

Cilia absent

RSHL1

Hfh4/FoxJ1

7p 19q13.3

6q25.3

IDA protein, locus linked in PCD patients with ODA defect Associates with ICs Chlamydomonas ODA, and also associates with IDA. No mutation detected in 58 PCD families KO of this IDA dynein in mice yield reduced CBF, but no abnormal ultrastructure One homozygous mutation in motor domain in PCD patient with 7 UPD RSP 4 and RSP 6 make up a radial spoke head and important for regulation of DA activity and flagellar beating Located near the inner arm dynein, regulates flagellar motility No mutation detected in five PCD families

Jeganathan D, Meeks M, Gehrig C, et al., 2001.

No mutation detected in 54 PCD patients Chlamydomonas mutants lacking central apparatus have paralyzed flagella; central apparatus important for motility KO mice yield situs inversus and absence of cilia No mutation detected in eight PCD families

Blouin JL, Albrecht C, Gehrig C, et al., 2003. Smith EF, Lefebvre PA., 1997.

Bartoloni L, Mitchison H, Pazour G, et al., 2000. Benashski SE, Patel-King RS, King SM., 1999. Neesen J, Kirschner R, Ochs M, et al., 2001. Bartoloni L, Blouin JL, Pan Y, et al., 2002. Eriksson M, Ansved T, Anvret M, Carey N., 2001.

Gaillard AR, Diener DR, Rosenbaum JL, Sale WS., 2001 Pennarun G, Bridoux AM, Escudier E, et al., 2002.

Brody SL, Yan XH, Wuerffel MK, Song S, Shapiro S., 2000. Maiti AK, Bartoloni L, Mitchison HM, et al., 2000.

247

248

SECTION III / PULMONARY DISEASES

clinical expression of a PCD phenotype. Moreover, subsequent studies have shown that another Poll knockout mouse (which did not knockout Dpcd) did not exhibit a PCD phenotype. Supporting its relationship with ciliary function is the observation of increasing expression of DPCD during ciliogenesis. Although no mutations have been discovered in 51 unrelated PCD patients, DPCD remains a good candidate gene for patients with an IDA defect, especially in relation to splicing mutants, which have not been tested. A human inner arm dynein-heavy chain (DNAH7; 12 kb in coding sequence) has also been cloned, using a proteomics approach. The entire coding region of DNAH7 has been sequenced from overlapping segments of RT-PCR products from airway epithelial RNA from a PCD patient with missing IDA, and no disease-causing mutations have been identified. However, intragenic polymorphisms have been detected and are being used for further analysis. Finally, a splice mutation in the gene encoding p28, a light chain of the IDA of Chlamydomonas axonemes, results in absent IDAs. No mutations in the human homologue (hp28) have been found to be associated with clinical PCD. Table 26-1 lists other candidate genes of interest with the associated references.

CONCLUSION PCD is associated with significant morbidity from birth to adulthood. An increased understanding of the molecular pathophysiology of the structural and functional abnormalities in PCD will undoubtedly lead to an increased understanding of normal and abnormal physiology in the sinopulmonary and reproductive tracts. It will also lead to easier diagnosis through genetic testing, as well as open possibilities for targeted, novel treatments for PCD and other similar diseases. Identification of the genetic basis of “classic” PCD will also make it possible to identify less severe molecular abnormalities leading to milder disease phenotypes (“nonclassic” disease), and exploration of the hypothesis that less severe defects in cilia structure and function contribute to other airway and lung diseases. Recent exciting developments, to increase awareness of PCD from both a clinical and a research standpoint, include the formation of a patient advocacy and support group (www.pcdfoundation.org), and an NIH-funded consortium (http://rarediseasenetwork. epi.usf.edu/gdmcc/index.htm) of clinical research centers. This will allow large-scale efforts to increase the numbers of patients to receive an expert diagnostic workup (including research and genetic studies), as well as receive continuing care for their disease. Finally, therapeutic approaches will likely evolve to increase MCC in these patients with potentially conserved or “partially conserved” ciliary function.

ACKNOWLEDGMENTS Elizabeth Godwin for editorial assistance, and the technical assistance of Susan Minnix RN and Rhonda Pace BS, and grants from the NIH (HL04225, HL34322, and RR00046).

SELECTED REFERENCES Afzelius BA. A human syndrome caused by immotile cilia. Science 1976;193:317–319. Afzelius BA. Electron microscopy of the sperm tail. J Biophys Biochem Cytol 1959;5:805–834. Afzelius BA. Cilia-related diseases. J Pathol 2004;204:470–477. Bartoloni L, Blouin JL, Pan Y, et al. Mutations in the DNAH11 (axonemal heavy chain dynein type 11) gene cause one form of situs inversus

totalis and most likely primary ciliary dyskinesia. Proc Natl Acad Sci USA 2002;99(16):10,282–10,286. Bartoloni L, Blouin JL, Chung E, et al. Axonemal beta heavy chain dynein DNAH9: cDNA sequence, genomic structure and investigation in its role in primary ciliary dyskinesia. Genomics 2001;72:21–33. Bartoloni L, Mitchison H, Pazour G, et al. No deleterious mutations were found in three genes (HF4, LC8, IC2) on human chromosome 17q in patients with primary ciliary dyskinesia (abstract). Eur J Hum Genet 2000;8:484. Benashski SE, Patel-King RS, King SM. Light chain 1 from the Chlamydomonas outer dynein arm is leucine rich repeat protein associated with the motor domain of the gamma heavy chain. Biochemistry 1999;38:7253–7264. Bertocci B, De Smet A, Flatter E, et al. DNA polymerases mu and lambda are dispensable for IgG gene hypermutation. J Immunol 2002;168: 3702–3706. Blouin JL, Albrecht C, Gehrig C, et al. Primary ciliary dyskinesia (Kartagener’s syndrome): searching for genes in a highly heterogeneous disorder (abstract). Am J Hum Genet 2003;73(Suppl):2440. Blouin JL, Gehrig C, Jeganathan D, et al. DNAH3: characterization of the full length gene and mutation search in patients with primary ciliary dyskinesia (abstract). Am J Hum Genet 2001;69:1081. Blouin JL, Meeks M, Radhakrishna U, et al. Primary ciliary dyskinesia: a genome wide linkage analysis reveals extensive locus heterogeneity. Eur J Hum Genet 2003;8:109–118. Boucher RC. Human airway ion transport (Pt 1). Am J Respir Crit Care Med 1994;150:271–281. Boucher RC. Human airway ion transport (Pt 2). Am J Respir Crit Care Med 1994;150:581–593. Bowman AB, Patel-King RS, Benashski SE, McCaffery JMl, Goldstein JS, King SM. Drosophila roadblock and Chlamydomonas LC7: a conserved family of dynein associated proteins involved in axonal transport, flagellar motility, and mitosis. J Cell Biol 1999;146: 165–180. Braiman A, Ulzaner N, Priel Z. Enhancement of ciliary beat frequency is dominantly controlled by PKG and or PKA. In: Salathe M, Adler KB, Boucher RC, Satir P, eds. Cilia and Mucus: From Development to Respiratory Defense. New York: Marcell Dekker, 2001; pp. 67–80. Brody SL, Yan XH, Wuerffel MK, Song S, Shapiro S. Ciliogenesis and left right axis defects in forkhead factor HFH-4 null mice. Am J Respir Cell Mol Biol 2000;23:45–51. Brueckner M. Cilia propel the embryo in the right direction. Am J Med Genet 2001;101(4):339–344. Bush A, Cole P, Hariri M, et al. Primary ciliary dyskinesia: diagnosis and standards of care. Eur Respir J 1998;12(4):982–988. Camner P, Mossberg B, Afzelius BA. Evidence for congenitally nonfunctioning cilia in the tracheobronchial tract in two subjects. Am Rev Respir Dis 1975;112:807–809. Camner P, Mossberg B, Afzelius BA. Measurements of tracheobronchial clearance in patients with immotile cilia syndrome and its value in differential diagnosis. Eur J Respir Dis 1983;64(Suppl 127):57–63. Carson JL, Collier AM. Ciliary defects: cell biology and clinical perspectives. Adv Pediatr 1988;35:139–166. Chilvers MA, Rutman A, O’Callaghan C. Ciliary beat pattern is associated with specific ultrastructural defects in primary ciliary dyskinesia. J Allergy Clin Immunol 2003;112(3):518–524. Coren ME, Meeks M, Morrison I, Buchdahl RM, Bush A. Primary ciliary dyskinesia: age at diagnosis and symptom history. Acta Paediatr 2002;91(6):667–669. Corey M, Edwards L, Levison H, Knowles M. Longitudinal analysis of pulmonary function decline in patients with cystic fibrosis. J Pediatr 1997;131:809–814. Davis ME, Cai H, Drummond GR, Harrison DG. Shear stress regulates endothelial nitric oxide synthase expression through c-Src by divergent signaling pathways. Circ Res 2001;89(11):1073–1080. DiBella LM, Benashski SE, Tedford HW, Harrison A, Patel-King RS, King SM. The Tctex1/Tctex2 class of dynein light chains: dimerization, differential expression, and interaction with the LC8 protein family. J Biol Chem 2001;276(17):14,366–14,373.

CHAPTER 26 / PRIMARY CILIARY DYSKINESIA

DiBella LM, King SM. Dynein motors of the Chlamydomonas flagellum. Int Rev Cytol 2001;210:227–268. de Iongh RU, Rutland J. Ciliary defects in healthy subjects, bronchiectasis, and primary ciliary dyskinesia. Am J Respir Crit Care Med 1995;151(5):1559–1567. Eliasson R, Mossberg B, Camner P, Afzelius BA. The immotile cilia syndrome: a congenital ciliary abnormality as an etiologic factor in chronic airways infection and male sterility. N Engl J Med 1977;297:1–6. Ellerman A, Bisgaard H. Longitudinal study of lung function in a cohort of primary ciliary dyskinesia. Eur Respir J 1997;10(10):2376–2379. El Zein L, Omran H, Bouvagnet P. Lateralization defects and ciliary dyskinesia: lessons from algae. Trends Genet 2003;19(3):162–167. Eriksson M, Ansved T, Anvret M, Carey N. A mammalian radial spokehead-like gene, RSHL1, at the myotonic dystrophy-1 locus. Biochem Biophys Res Commun 2001;281(4):835–841. Escudier E, Couprie M, Duriez B, et al. Computer-assisted analysis helps detect inner dynein arm abnormalities. Am J Respir Crit Care Med 2002;166(9):1257–1262. Fisher AB, Chien S, Barakat AI, Nerem RM. Endothelial cellular response to altered shear stress. Am J Physiol Lung Cell Mol Physiol 2001;281(3):L529–L533. Gaillard AR, Diener DR, Rosenbaum JL, Sale WS. Flagellar radial spoke protein 3 is an A-kinase anchoring protein (AKAP). J Cell Biol 2001;153(2):443–448. Gaston B, Drazen JM, Loscalzo J, Stamler JS. The biology of nitrogen oxides in the airways. Am J Respir Crit Care Med 1994;149(2 Pt 1):538–551. Gehrig C, Albrecht C, Duriaus-Sail G, et al. Primary ciliary dyskinesia: Mutation analysis in dynein light chain genes mapping to chromosome 1 (HP28) and 22 (DNAL4). Paper presented at: European Human Genetics Conference 2002; May 25–28, 2002; Strasbourg, France. Geremek M, Witt M. Primary ciliary dyskinesia: genes, candidate genes and chromosomal regions. J Appl Genet 2004;45:347–361. Greenstone M, Rutman A, Dewar A, Mackay I, Cole PJ. Primary ciliary dyskinesia: cytological and clinical features. Q J Med 1988;67: 405–430. Guichard C, Harricane MC, Lafitte JJ, et al. Axonemal dynein intermediate-chain gene (DNAI1) mutations result in situs inversus and primary ciliary dyskinesia (Kartagener syndrome). Am J Hum Genet 2001;68(4):1030–1035. Habermacher G, Sale WS. Regulation of flagellar dynein by an axonemal type-1 phosphatase in Chlamydomonas. J Cell Sci 1996; 109(Pt 7):1899–1907. Hadfield PJ, Rowe-Jones JM, Bush A, Mackay IS. Treatment of otitis media with effusion in children with primary ciliary dyskinesia. Clin Otolaryngol 1997;22:302–306. Halbert SA, Patton DL, Zarutskie PW, Soules MR. Function and structure of cilia in the fallopian tube of an infertile woman with Kartagener’s syndrome. Hum Reprod 1997;12(1):55–58. Holzmann D, Felix H. Neonatal respiratory distress syndrome—a sign of primary ciliary dyskinesia? Eur J Pediatr 2000;159(11): 857–860. Ibanez-Tallon I, Gorokhave S, Heintz N. Loss of function of axonemal dynein Mdnah5 causes primary ciliary dyskinesia and hydrocephalus. Hum Mol Genet 2002;11:715–721. Jeganathan D, Meeks M, Gehrig C, et al. A genome wide scan reveals a putative novel locus for primary ciliary dyskinesia (abstract). Am J Hum Genet 2001;69:2015. Kartagener M, Stucki P. Bronchiectasis with situs inversus. Arch Pediatr 1962;79:193–207. Katsuhara K, Kawamoto S, Wakabayashi T, Belsky JL. Situs inversus totalis and Kartagener’s syndrome in a Japanese population. Chest 1972;61:56–61. Kispert A, Petry M, Olbrich H, et al. Genotype-phenotype correlations in PCD patients carrying DNAH5 mutations. Thorax 2003;58: 552–554. Knowles MR, Boucher RC. Mucus clearance as a primary innate defense mechanism for mammalian airways. J Clin Invest 2002;109: 571–577.

249

Knowles MR, Durie PR. What is cystic fibrosis? N Engl J Med 2002;347(6):439–442. Knowles MR, Friedman KJ, Silverman LM. Genetics, diagnosis and clinical phenotype. In: Yankaskas JR, Knowles MR, eds. Cystic Fibrosis in Adults. Philadelphia, PA: Lippincott-Raven, 1999:27–42. Kobayashi Y, Watanabe M, Okada Y, et al. Hydrocephalus, situs inversus, chronic sinusitis, and male infertility in DNA polymerase lambdadeficient mice: possible implication for the pathogenesis of immotile cilia syndrome. Mol Cell Biol 2002;22(8):2769–2776. LeDizet M, Piperno G. ida4-1, ida4-2, and ida4-3 are intron splicing mutations affecting the locus encoding p28, a light chain of Chlamydomonas axonemal inner dynein arms. Mol Biol Cell 1995;6(6):713–723. Olbrich H, Haffner K, Kispert A, et al. Mutations in DNAH5 cause primary ciliary dyskinesia and randomization of left/right asymmetry. Nat Genet 2002;30(2):143, 144. Maiti AK, Bartoloni L, Mitchison HM, et al. No deleterious mutations in the FOXJ1 (alias HFH-4) gene in patients with primary ciliary dyskinesia (PCD). Cytogenet Cell Genet 2000;90(1–2):119–122. LeDizet M, Piperno G. The light chain p28 associates with a subset of inner dynein arm heavy chains in Chlamydomonas axonemes. Mol Biol Cell 1995;6(6):697–711. McGrath J, Somlo S, Makova S, Tian X, Brueckner M. Two populations of node monocilia initiate left-right asymmetry in the mouse. Cell 2003;11:61–73. McManus I, Mitchison H, Chung E, Stubbings G, Martin N. Primary ciliary dyskinesia (Siewert’s/Kartagener’s syndrome): respiratory symptoms and psycho-social impact. BioMed Central Pulmonary Medicine (serial online). November 27, 2003;3. Meeks M, Bush A. Primary ciliary dyskinesia (PCD). Pediatr Pulmonol 2000;29(4):307–316. Meeks M, Walne A, Spiden S, et al. A locus for primary ciliary dyskinesia maps to chromosome 19q. J Med Genet 2000;37(4):241–244. Mickle JE, Cutting GR. Genotype-phenotype relationships in cystic fibrosis. Med Clin North Am 2000;84(3):597–607. Munro NC, Currie DC, Lindsay KS, et al. Fertility in men with primary ciliary dyskinesia presenting with respiratory infection. Thorax 1994;49:684–687. Narang I, Ersu R, Wilson NM, Bush A. Nitric oxide in chronic airway inflammation in children: Diagnostic use and pathophysiologic significance. Thorax 2002;57:586–589. Narayan D, Krishnan SN, Upender M, et al. Unusual inheritance of primary ciliary dyskinesia (Kartagener’s syndrome). J Med Genet 1994;31(6):493–496. Neesen J, Dreckhahn JD, Tiede S, et al. Identification of the human ortholog of the t-complex-encoded protein TCTE3 and evaluation as a candidate gene for primary ciliary dyskinesia. Cytogenet Genome Res 2002;98:38–44. Neesen J, Kirschner R, Ochs M, et al. Disruption of an inner arm dynein heavy chain gene results in asthenozoospermia and reduced ciliary beat frequency. Hum Mol Genet 2001;10(11):1117–1128. Noone PG, Bali D, Carson JL, et al. Discordant organ laterality in monozygotic twins with primary ciliary dyskinesia. Am J Med Genet 1999;82(2):155–160. Noone PG, Bennett WD, Regnis JA, et al. Effect of aerosolized uridine-5′triphosphate on airway clearance with cough in patients with primary ciliary dyskinesia. Am J Respir Crit Care Med 1999;160(1):144–149. Noone PG, Leigh MW, Sannuti A, et al. Primary ciliary dyskinesia: diagnostic and phenotypic features. Am J Respir Crit Care Med 2004;169(4): 459–467. Noone PG, Pue CA, Zhou Z, et al. Lung disease associated with the IVS8 5T allele of the CFTR gene. Am J Respir Crit Care Med 2000; 162(5):1919–1924. O’Donnell AE, Barker AE, Ilowite JS, Fick RB. Treatment of idiopathic bronchiectasis with aerosolized recombinant human DNase. Chest 1998;113:1329–1334. Omoto CK, Gibbons IR, Kamiya R, Shingyoji C, Takahashi K, Witman GB. Rotation of the central pair microtubules in eukaryotic flagella. Mol Biol Cell 1999;10(1):1–4. Omran H, Haffner K, Volkel A, et al. Homozygosity mapping of a gene locus for primary ciliary dyskinesia on chromosome 5p and identifi-

250

SECTION III / PULMONARY DISEASES

cation of the heavy dynein chain DNAH5 as a candidate gene. Am J Respir Cell Mol Biol 2000;23(5):696–702. Ostrowski LE, Blackburn K, Radde KM, et al. A proteomic analysis of human cilia: Identification of novel components. Mol Cell Proteomics 2002;1(6):451–465. Pazour GJ,Agrin N, Walker BL, Witman GB. Identification of predicted human outer dynein arm genes: candidates for primary ciliary dyskinesia genes. J Med Genet 2006;43:62–73. Pennarun G, Bridoux AM, Escudier E, Anselem S, Duriez B. The human HP28 and HFH4 genes: evaluation as candidate genes for primary ciliary dyskinesia. Am J Respir Crit Care Med 2001;163:A538. Pennarun G, Bridoux AM, Escudier E, et al. Isolation and expression of human hPF20 gene orthologous to Chlamydomonas pf20. Am J Respir Cell Mol Biol 2002;26:362–370. Pennarun G, Chapelin C, Escudier E, et al. The human dynein intermediate chain 2 gene (DNAI2): cloning mapping expression pattern and evaluation as a candidate for primary ciliary dyskinesia. Hum Genet 2000;107:642–649. Pennarun G, Escudier E, Chapelin C, et al. Loss-of-function mutations in a human gene related to Chlamydomonas reinhardtii dynein IC78 result in primary ciliary dyskinesia. Am J Hum Genet 1999; 65(6): 1508–1519. Perrone CA, Tritschler D, Taulman P, Bower R,Yoder BK, Porter ME. A novel dynein light intermediate chain colocalizes with the retrograde motor for intraflagellar transport at sites of axoneme assembly in chlamydomonas and mammalian cells. Mol Biol Cell 2003; 14(5): 2041–2056. Piperno G, Mead K. Transport of a novel complex in the cytoplasmic matrix of Chlamydomonas flagella. Proc Natl Acad Sci USA 1997; 94(9):4457–4462. Piperno G, Mead K, Henderson S. Inner dynein arms but not outer dynein arms require the activity of kinesin homologue protein KHP1(FLA10) to reach the distal part of flagella in Chlamydomonas. J Cell Biol 1996;133(2):371–379. Razani B, Engelman JA, Wang XB, et al. Caveolin-1 null mice are viable but show evidence of hyperproliferative and vascular abnormalities. J Biol Chem 2001;276(41):38,121–38,138. Regnis JA, Zeman KL, Noone PG, Knowles MR, Bennett WD. Prolonged airway retention of insoluble particles in cystic fibrosis versus primary ciliary dyskinesia. Exp Lung Res 2000;26(3):149–162. Rossman CM, Forrest JB, Ruffin RE, Newhouse MT. Immotile cilia syndrome in persons with and without Kartagener’s syndrome. Am Rev Respir Dis 1980;121:1011–1016. Rossman CM, Lee RM, Forrest JB, Newhouse MT. Nasal ciliary ultrastructure and function in patients with primary ciliary dyskinesia compared with that in normal subjects and in subjects with various respiratory diseases. Am Rev Respir Dis 1984;129(1):161–167. Rutland J, De Iong RU. Random ciliary orientation. N Engl J Med 1990;323:1681–1684. Ruusa J, Svartengren M, Philipson K, Camner P. Tracheobronchial particle deposition and clearance in immotile cilia syndrome patients. J Aerosol Med 1993;6:89–98. Sanderson MJ, Lansley AB, Evans JH. The regulation of airway ciliary beat frequency by intracellular calcium. In: Salathe M, Adler KB, Boucher RC, Satir P, eds. Cilia and Mucus: From Development to Respiratory Defense. New York: Marcel Dekker, 2001: pp. 39–58. Sannuti A, Noone PG, Daines C, et al. Quantification of inner and outer dynein arms in primary ciliary dyskinesia (PCD). Am J Respir Crit Care Med 1999;159:A39.

Satir P. The cilium as a biological nanomachine. FASEB J 1999;13(Suppl 2):S235–S237. Sisson JH, Carson JL. Bench to bedside: new findings in primary ciliary dyskinesia. Pulmonary and Critical Care Update [serial online] 2003;17:lesson 5. Available at www.chestnet.org/education/online/ pccu/vol17/lessons5_6/index.php. Smith EF, Lefebvre PA. The role of central apparatus components in flagellar motility and microtubule assembly. Cell Motil Cytoskeleton 1997;38(1):1–8. Smith EF, Sale WS. Structural and functional reconstitution of inner dynein arms in Chlamydomonas flagellar axonemes. J Cell Biol 1992;117:573–581. Stannard W, Rutman A, Walne A, O’Callaghan C. Central microtubular agenesis causing primary ciliary dyskinesia. Am J Respir Crit Care Med 2004;169:634–637. Sturgess JM, Chao J, Turner JAP. Transposition of ciliary microtubules: Another cause of impaired motility. N Engl J Med 1980;303:318–322. Sturgess JM, Chao J, Wong J, Aspin N, Turner JAP. Cilia with defective radial spokes: A cause of human respiratory disease. N Engl J Med 1979;300:53–56. Supp DM, Witte DP, Potter SS, Brueckner M. Mutation of an axonemal dynein affects left-right asymmetry in inversus viscerum mice. Nature 1997;389:963–966. Thai CH, Gambling TM, Carson JL. Freeze fracture study of airway epithelium from patients with primary ciliary dsykinesia. Thorax 2002;57:363–365. Torgersen J. Situs inversus, asymmetry, and twinning. Am J Hum Genet 1950;2:361–370. van der Baan S, Veerman AJ, Wulffraat N, Bezemer PD, Feenstra L. Primary ciliary dyskinesia: ciliary activity. Acta Otolaryngol 1986; 102(3–4):274–281. Wanner A, Salathe M, O’Riordan TG. Mucociliary clearance in the airways: state of the art. Am J Respir Crit Care Med 1996;154:1868–1902. Wills PJ, Wodehouse T, Corkery K, Mallon K, Wilson R, Cole PJ. Short term recombinant human DNase in bronchiectasis: effect on clinical state and in vitro sputum transportability. Am J Respir Crit Care Med 1996;154:413–417. Witt M, Wang Y-F, Wang S, et al. Exclusion of chromosome 7 for Kartagener Syndrome but suggestion of linkage in families with other forms of primary ciliary dyskinesia. Am J Med Genet 1999; 64:313–318. Wu H, Maciejewski MW, Marintchev A, Benashski SE, Mullen GP, King SM. Solution structure of a dynein motor domain associated light chain. Nat Struct Biol 2000;7:575–579. Xue C, Botkin SJ, Johns RA. Localization of endothelial NOS at the basal microtubule membrane in ciliated epithelium of rat lung. J Histochem Cytochem 1996;44(5):463–471. Zariwala M, O’Neal WK, Noone PG, Leigh MW, Knowles MR, Ostrowski LE. Investigation of the possible role of a novel gene, DPCD, in primary ciliary dykinesia. Am J Respir Cell Mol Biol 2004;30:1–7. Zariwala M, Noone PG, Sannuti A, et al. Germline mutations in an intermediate chain dynein cause primary ciliary dyskinesia. Am J Respir Cell Mol Biol 2001;25(5):577–583. Zhang YJ, O’Neal WK, Randell SH, et al. Identification of dynein heavy chain 7 as an inner arm component of human cilia that is synthesized but not assembled in a case of primary ciliary dyskinesia. J Biol Chem 2002;277(20):17,906–17,915.

27 Cystic Fibrosis SCOTT H. DONALDSON AND RICHARD C. BOUCHER SUMMARY Cystic fibrosis is a genetic disease that affects multiple organ systems, yet leads to respiratory failure and premature death in most afflicted patients. Rapid advancements in our understanding of disease pathogenesis have occurred because of the cloning of the cystic fibrosis transmembrane conductance regulator gene. Recent work has shown that volume depletion of the periciliary fluid layer and mucus dehydration reduces mucus clearance and promotes the initiation and progression of CF lung disease. Therapies that correct dysregulated salt and water transport in the lung, either by correcting or bypassing the basic cystic fibrosis transmembrane conductance regulator defect, are now being developed and hold great promise. Key Words: Airway surface liquid; bronchiectasis; Cftr; cystic; ENaC; fibrosis; ion transport; mucus clearance; mucociliary; Pseudomonas.

INTRODUCTION In 1938, Anderson et al. first described “cystic fibrosis of the pancreas” as a distinct clinical entity. At that time, malnutrition resulting from insufficiency of the exocrine pancreas and overwhelming respiratory infection typically led to death within the first year of life. Subsequent observations began to lay the framework on which the understanding of the cystic fibrosis (CF) disease process has been built. Initially, it was recognized that multiple epithelial lined organs contained abnormally viscous mucus, whereas sweat from CF patients contained excessive amounts of salt. In the 1980s, multiple investigators described the altered epithelial ion transport processes that are characteristic of CF airways epithelia, raising the possibility that these abnormalities were fundamental to CF disease pathogenesis. Finally, the cystic fibrosis transmembrane conductance regulator (CFTR) gene was cloned in 1989 and was found to encode a cAMP-regulated plasma membrane polypeptide. From these seminal observations, our understanding of the CF disease process has evolved, although considerable controversy and competing hypotheses about aspects of the disease persist. As a result of this ongoing progress, clinical outcomes have continued to improve and novel therapies are being developed at an unprecedented rate. The predicted survival of patients with CF has increased to 34 yr of age, although further improvement in survival may require the development of new therapeutics. This chapter outlines the clinical From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

CF phenotype, describes the pathogenesis of CF lung disease, and suggests how new therapies may utilize the growing knowledge of CF pathogenesis to delay or prevent the development of this devastating disease.

CLINICAL DESCRIPTION OF A MULTISYSTEM DISEASE Pulmonary disease accounts for considerable morbidity and greater than 90% of the mortality from CF. However, a broad spectrum of clinical manifestations is encountered in CF, stemming from epithelial dysfunction in multiple organ systems. The spectrum of disease manifestations greatly increases the complexity of care required by this population, mandating the establishment of specialized CF care centers in which accumulated expertise and experience has improved patient outcomes. RESPIRATORY MANIFESTATIONS Newborns with CF have normal lung function and sterile, uninflamed airways. Early in life, however, persistent airway infection and inflammation develop as the result of impaired airway host defenses. The nature of this host defense deficit has been intensively studied and debated and is discussed later. CF lung disease progresses from subclinical infection and inflammation to overt symptoms (episodic or persistent cough, sputum) and airway obstruction (demonstrated by spirometry) in a highly variable fashion, but the disease process is usually evident during childhood. Overtime, periods of clinical worsening termed “exacerbations” can be identified, which require intensification of secretion clearance and antibiotic therapy aimed at the patient’s airway microbial flora. Progressive lung destruction results from the relentless infectious/inflammatory process that is characteristic of CF airways, culminating in the pathological finding of bronchiectasis, obstructive lung disease, and ultimately respiratory failure and death in the majority of patients. CF lung infections are caused by a typical group of pathogens. Early in life, viral infections are normal in frequency but may cause significantly more morbidity and may trigger the onset of persistent airway inflammation. Haemophilus influenzae is frequently encountered during childhood but rarely persists. Staphylococcus aureus and Pseudomonas aeruginosa, however, often occur early in life and generally persist, especially P. aeruginosa. Although these pathogens are also found in other clinical settings, and certainly in other causes of bronchiectasis, the frequency of Pseudomonas infection (80% of adults) and tendency to convert to the mucoid phenotype is particularly striking and suggests that the CF airway provides a unique environmental niche that favors this organism.

251

252

SECTION III / PULMONARY DISEASES

Similarly, the inability to clear Pseudomonas from the CF airway, despite the aggressive use of antibiotics and an intense host inflammatory response, defines the problem confronted by clinicians who treat CF lung disease. The persistence of P. aeruginosa likely reflects adaptation of this organism to the CF airway milieu and the development of bacterial mechanisms that allow evasion of secondary host defenses and therapeutic interventions. A number of other organisms are less commonly encountered in the CF lung, including Stenotrophomonas maltophilia and Alcaligenes xylosoxidans. Perhaps most notable, however, is the emergence of the Burkholderia cepacia complex as a group of pathogens with the capacity to lead to rapid clinical deterioration and frank sepsis (“cepacia syndrome”). Unfortunately, organisms in the B. cepacia complex are often resistant to most, if not all, known antibiotics and certain strains have proven to be highly transmissible between patients. Other respiratory manifestations of CF are also encountered frequently. With increasing age, the incidence of pneumothorax and massive hemoptysis rises. Pneumothoraces (16–20% incidence in adult CF patients) result from the rupture of subpleural cysts. Hemoptysis is common and may occasionally be lifethreatening (approx 5% incidence) in CF because the arterial circulation of bronchiectatic airways is often massively hypertrophied and vulnerable to erosion from mucosal ulcerations consequent to persistent intralumenal infection/inflammation. Less lifethreatening but common respiratory tract manifestations include allergic bronchopulmonary aspergillosis (1–4% incidence), chronic sinusitis, and nasal polyposis. EXTRAPULMONARY MANIFESTATIONS The function of the exocrine pancreas is insufficient at birth or becomes so early in life in the majority of patients. Pancreatic insufficiency in CF results from the obstruction of pancreatic ducts with inspissated secretions and the subsequent autodigestion of pancreatic acinar regions. The inability to secrete pancreatic enzymes results in protein and fat malabsorption, causing steatorrhea. If unaddressed, persistent malabsorption leads to malnutrition, impaired growth, and fat-soluble vitamin deficiencies (vitamins A, D, E, and K). A small number of CFTR mutations are associated with retention of pancreatic sufficiency, however, and these patients typically are better nourished and typically have less severe respiratory phenotypes. Pancreatic sufficient patients are at risk for pancreatitis, however, which likely reflects the presence of residual exocrine tissue and opportunity for acute duct obstruction. Endocrine pancreatic function is preserved in CF patients at birth, but may also be progressively lost overtime. The development of “cystic fibrosis-related diabetes” is age dependent, with the incidence peaking at 25% in 35–44 yr olds. Many more patients will have episodic hyperglycemia, especially during times of respiratory exacerbation. Of note, unrecognized or poorly controlled diabetes is thought to negatively impact nutrition and lung health and, therefore, must be screened for and managed carefully in these patients. Calorie restriction must be avoided to prevent worsening the patient’s nutritional status, and insulin therapy is usually required to treat cystic fibrosis-related diabetes. The CF intestinal tract epithelium has a limited ability to secrete chloride and water resulting from the absence of CFTR chloride channel function in the apical membrane of enterocytes lining the gut. As a result, patients may manifest both diarrhea from malabsorption and/or severe constipation and intestinal obstruction as the result of insufficient hydration of bulky, poorly digested intestinal

contents. When terminal ileum obstruction occurs at birth (17% incidence), the resulting “meconium ileus” is nearly diagnostic of CF. A similar phenomenon occurring later in life is termed the “distal intestinal obstruction syndrome” (20% incidence). Rectal prolapse and gastroesophageal reflux are common in children with CF, and fibrosing colonopathy may occur in any patient treated with very high doses of pancreatic enzymes. Focal biliary cirrhosis is an extremely common pathological finding in CF at autopsy, but produces clinically apparent disease in less than 5% of CF patients and is the cause of death in approx 2% of CF patients. Unlike many complications of CF, hepatic disease has a peak incidence during adolescence and a decreased prevalence in patients over age 20. Hepatic abnormalities can present as hepatosplenomegaly or as a persistent elevation of hepatic enzymes. Rarely, patients may present with esophageal varices and hemorrhage resulting from portal hypertension. Fatty liver is also common, and may improve with adequate nutrition. Finally, 10–30% of patients have dysfunctional gallbladders or gallstones. More than 98% of men with CF are infertile because of azoospermia from vas deferens obstruction. The vas deferens may be absent at the time of birth, and its absence serves as a diagnostic clue pointing toward CF in situations of previously undiagnosed bronchiectasis. It appears that the male reproductive tract is the most sensitive organ to CFTR mutations, as a sizeable percentage of infertile men with congenital, bilateral absence of the vas deferens, who have no other demonstrable CF manifestations, have mild CFTR mutations. Interestingly, these mutations may be in gene regions that cause abnormal RNA splicing but produce normally functioning, though reduced amounts of CFTR protein. This syndrome explains that although 5% of normal CFTR message levels may be insufficient for male reproductive tract function, it is often sufficient to maintain normal pancreatic and respiratory tract function. This observation has important implications when considering therapies aimed at either delivering a normal CFTR gene (“gene therapy”) or improving the processing/function of the existing, mutant CFTR protein. Importantly, female patients with CF maintain near normal fertility when factored for nutritional and respiratory status.

PATHOGENESIS OF CF LUNG DISEASE The requirement to explain how CF lung disease results from CFTR gene mutations has spurred a great deal of research and a number of competing hypotheses. Fortunately, a more clear vision of CF disease pathogenesis is emerging as discussed in the following sections. THE CFTR GENE AND PROTEIN The CFTR gene was localized to a region on the long arm of chromosome 7 via positional cloning and consists of 250 kB of genomic DNA containing 27 exons. The CFTR gene encodes a 1480 amino acid protein, belonging to the adenosine triphosphate-binding casset family of transport proteins, which is typically targeted to the apical membrane of epithelial cells. The protein can be viewed as having two halves, each made up of six transmembrane domains and a nucleotide-binding domain, which are separated by a highly charged “R domain” containing multiple protein kinase A and protein kinase C phosphorylation sites. Nearly 1300 mutations have been reported, although the ∆F508 mutation accounts for 66% of mutant CFTR alleles. Because CF is an autosomal-recessive disorder, patients carry two mutant CFTR gene alleles, whereas parents are “carriers” with a normal CFTR gene and a single mutant CFTR allele and exhibit no clinical stigmata of the CF disease.

CHAPTER 27 / CYSTIC FIBROSIS

253

Figure 27-1 CFTR mutation classifications. Class I mutations result in the complete absence of CFTR protein production. Class II mutations cause protein misfolding, impaired cellular processing and retention in the endoplasmic reticulum. Class III mutations yield CFTR molecules that reach the plasma membrane but are resistant to activation by usual (e.g., cAMP) signaling pathways. Class IV mutations result in CFTR channels with reduced ionic conductivity. Class V mutations result in dramatically reduced quantity of normal CFTR protein.

CFTR mutations may be grouped into classes that reflect the mechanism by which loss of CFTR function occurs (Fig. 27-1). Class I mutations, including nonsense (i.e., stop mutation), frameshift and splice junction mutations generally result in the loss of protein production and complete absence of full length CFTR. Class II mutations reflect abnormal protein folding, and hence processing and targeting to the plasma membrane. This class of mutations includes the common ∆F508 mutation, in which misfolding prevents normal maturation and transport to the apical cell membrane. Class III mutations in the CFTR gene affect the regulation/activation of a mutant CFTR chloride channel that folds correctly and reaches the plasma membrane. Class IV mutations also reach the plasma membrane, but the mutant CFTR exhibits poor chloride conductance through the channel pore. Finally, class V CFTR mutations decrease the abundance of full-length CFTR mRNA and, hence, protein levels, and include mutations in the CFTR promoters and regions that influence mRNA splicing. These mutations may, however, permit the production of enough CFTR protein and function to yield a less severe disease phenotype. The concept of mutation classes carries therapeutic implications in that novel therapies are being developed to: (1) produce “read-through” of premature termination mutations (class I mutation; e.g., topical gentamicin); (2) promote movement of ∆F508 CFTR to the plasma membrane in which it may function adequately (class II mutation; e.g., 4-phenylbutyrate); and (3) activate mutant channels that reach the plasma membrane (class III/IV mutations; e.g., genistein). CFTR AND EPITHELIAL TRANSPORT Even prior to the cloning of the CFTR gene, it was recognized that a cAMPstimulated chloride conductance was deficient in CF epithelia. Expression of the cloned CFTR gene in various heterologous systems demonstrated that CFTR itself encoded this missing chloride channel. Further, as the name implies, CFTR regulates other epithelial functions/ion channels, most notably the epithelial sodium channel (ENaC). Through a poorly defined mechanism, normal CFTR in airway epithelia provides a tonic inhibition of ENaC activity. In CF, therefore, ENaC activity is dysregulated and sodium hyperabsorption represents the dominant basal ion transport activity measured in many experimental systems, including the human nasal epithelium in vivo. CFTR may also regulate the

activity of calcium-activated chloride channels (CACC), as the activities of CACC and CFTR are reciprocally related in many epithelia. A number of hypotheses have attempted to explain the development of CF lung disease via mechanisms that do not require CFTR channel function, including: (1) increased binding of Pseudomonas to CF airway surfaces resulting from observed alterations in glycosylation of membrane and/or secreted proteins; (2) persistent airway infection resulting from loss of CFTR-mediated epithelial phagocytosis of pseudomonas; and (3) intrinsic hyperinflammation of CF airways via an unexplained mechanism. Each of these hypotheses is not widely accepted resulting from conflicting experimental observations and the possibility that they may represent a secondary phenomena resulting from chronic airway infection/inflammation. More generally accepted are hypotheses that directly link CFTR malfunction to altered ion transport processes. AIRWAY SURFACE LIQUID REGULATION IN NORMAL AND CF AIRWAYS The liquid bathing airway surface is composed of two components, a periciliary gel (PGL) layer and a more viscous mucus layer that is positioned between the PGL and the airway lumen. This configuration allows the mucus layer to efficiently trap inhaled pathogens and particulates. The underlying PGL layer provides a physical environment in which cilia can beat freely and, thus, propel the mucus layer toward the mouth. The PGL also acts as a lubricant layer that prevents adhesion of the mucus layer to cell surfaces. Proper regulation of airway surface liquid (ASL) volume and the hydration of its component layers, therefore, are critical to the maintenance of mucus clearance. Whereas maintenance of PGL height is necessary to facilitate cilia motion, adequate hydration of the mucus layer is a key determinant of viscoelastic properties and transportability. As ASL moves from distal airways toward the trachea, an enormous reduction in airway surface area occurs, necessitating active absorption of liquid across the airway to prevent obstruction of airway lumens. Conversely, there may be situations (e.g., exercise) or lung regions in which addition of liquid to ASL is necessary. The superficial epithelium lining airway surfaces regulates PGL height and mucus hydration through the coordinate regulation of Na+-mediated liquid absorption and Cl–-mediated liquid secretion.

254

SECTION III / PULMONARY DISEASES

Figure 27-2 Pathogenesis of CF lung disease. (Upper left) Normal regulation of ASL volume requires the reciprocal regulation of CFTR and ENaC activities, which in turn supports normal MCC. (Upper right) In CF, absent CFTR and excessive ENaC activities result in depletion of ASL volume and concentration of airway mucus. (Middle right) Mucostasis ensues and is accompanied by the development of a hypoxic zone within mucus plaques. (Lower right) Finally, infection with typical CF pathogens, conversion to a biofilm mode of growth, and exuberant but ineffective inflammation results in clinically apparent CF lung disease. CaCC, calcium-activated chloride channels.

The absorption of airway liquid is accomplished via active sodium transport, with ENaC mediating the rate-limiting step in this process. Liquid secretion across the superficial epithelium is mediated, at least in part, by CFTR. Liquid may also be added to the PGL layer via bulk flow from converging distal lung units and by chloride/water secretion from submucosal glands (in part via CFTR). The relative contribution of each process is unclear but must vary between proximal airways, in which glands are present, and distal airway regions in which glands are not. As noted, CF epithelia hyperabsorb liquid resulting from dysregulation of ENaC and have lost the capacity to secrete liquid to restore ASL volume homeostasis on CF airway surfaces. As the result, the PGL layer is depleted and the mucus layer becomes concentrated and poorly transportable, thus impairing mucus clearance and predisposing the patient to airway infection. The consequence of PGL depletion in CF is profound. First, absence of the PGL layer eliminates the low viscosity solution for cilia beat, and hence slows cilial-dependent mucus clearance, and also eliminates the lubricant activity that prevents the mucus layer from interacting with the cell surface. Second, mucins in the

mucus layer become concentrated, greatly reducing the mesh diameter of the mucin network and increasing its adhesivity. The reduction in mesh size reduces bacterial motility, perhaps stimulating biofilm formation, and retards the ability of neutrophils to penetrate thickened mucus plaques to kill bacteria. Ultimately, the mucus layer becomes adherent to cell surfaces, making clearance via cough impossible as well. Adherent mucus plaques narrow and obstruct small airways and provide a nidus for infection. The cycle is then set for chronic infection, airway inflammation, mucin hypersecretion, and progressive airway obstruction and destruction. Importantly, restoration of PGL volume in CF cultures allows resumption of ciliary beating and mucus transport, giving hope that similar maneuvers will improve the CF condition in vivo. In addition to the described changes in ASL volume homeostasis, an ASL compositional change in CF has been proposed by investigators. Initial reports of differences in ASL sodium and chloride concentrations between normal and CF subjects have largely been refuted by direct and indirect measures of ASL ionic composition (demonstrating isotonicity of ASL) using a variety of techniques and experimental systems. However, ASL may be

255

CHAPTER 27 / CYSTIC FIBROSIS

relatively acidic in CF, owing to the absence of CFTR-dependent bicarbonate secretion. Lower ASL pH in CF could contribute to disease pathogenesis through adverse effects on mucus viscosity, as well as increased adhesion between secreted and tethered cellsurface mucins via electrostatic mechanisms. Further, acidification of the CF airway lumen might interfere with bactericidal activity, thus promoting further inflammation and airway damage. SECONDARY PATHOGENIC STEPS: MUCUS, PSEUDOMONAS, AND INFLAMMATION As described, the reduced ASL volume that produces progressive mucostasis and airway obstruction likely initiates the cascade of events that lead to clinically apparent CF lung disease. Studies are beginning to clarify the subsequent steps in this process (Fig. 27-2). First, thick mucus plaques adherent to airway surfaces paradoxically provide an anaerobic niche within the lung. Direct measurements in CF patients demonstrated that occluding mucopurulent plugs creates a dramatically hypoxic zone, with a pO2 of less than 2 torr. In vitro studies demonstrated that the combination of a thick mucus plaque and increased oxygen consumption by CF epithelia (owing to heightened Na+ transport) leads to the development of relatively hypoxic ASL in CF (vs normal and disease control) cultures. Importantly, P. aeruginosa contained in droplets added to the surface of these cultures can migrate into anaerobic zones within mucus plaques and grow by shifting to anaerobic metabolism. An increase in alginate production, characteristic of clinical CF pseudomonas isolates, is observed as well. Together, it has been speculated that the unique CF environment, initiated by mucostasis and hypoxia, triggers P. aeruginosa to convert to an anaerobic biofilm mode of growth. This shift in bacterial growth pattern can lead to increased resistance to secondary host defense mechanisms (e.g., neutrophils and soluble antimicrobials) and persistence of this infection. Indeed, persistent failure of certain defenses, for example, neutrophils, may lead to inadvertent damage to airway walls.

IMPLICATIONS FOR CURRENT AND FUTURE THERAPIES The ability to increase the quality and duration of life in CF may rest on the ability to target multiple steps in the disease pathogenesis cascade (Table 27-1). Cornerstones of CF therapy have included the physical removal of airway secretions via chest percussion and postural drainage and treatment of airway infections. Aerosolized antipseudomonal antibiotics have added significantly to the ability to treat patients with established lung disease, and new additions to the inhaled antibiotic armamentarium are expected. Utilization of aerosolized recombinant human DNase (rhDNase) (Pulmozyme®) to reduce the viscosity of CF airway secretions by cleaving intraluminal DNA (rhDNase), thus facilitating mucus clearance, also improves lung function. Early institution of this agent in children with minimal lung disease may also slow the progression of lung disease. Other agents that target mucus properties or mucin production are under investigation to determine whether further improvements may be gained. The concept of modulating the CF inflammatory response is not new and was tested through the use of high dose corticosteroids over long periods of time. This approach did slow the progression of lung disease, but was associated with unacceptably high toxicities (cataracts, osteoporosis, diabetes mellitus). Ibuprofen was subsequently used for its effects on neutrophil function/migration, and indeed greatly slowed the progression of lung disease in a subset of patients (age other tissues Ubiquitous Other tissues > neural Neuroendocrine Neural, platelets Ubiquitous Ubiquitous Liver, lung, kidney Blood cells Ubiquitous Ubiquitous

Increase AC, Ca2+ channel activity; activate c-Src tyrosine kinases Increase cGMP phoshodiesterase activity Increase phosphodiesterase activity Decrease AC; increase potassium channel activity; activate c-Src tyrosine kinases Decrease potassium channel activity; activate Rap1GAP1 Increase PLC-β1, PKC; activate LARGRhoGEF

Activate Rho proteins Increase AC, PLC, GIRK channels, MAPK, PI3K; inhibit GαI

The expression patterns and effectors of 15 subtypes of Gα subunits and Gβγ are shown. AC, adenylyl cyclase; cGMP, cyclic GMP; GAP, GTPase activating protein; GEF, guanine nucleotide exchange factor; PLC, phospholipase C; PKC, phosphokinase C; GIRK, G protein-regulated inwardly rectifying potassium channel; MAPK, mitogen-activated protein kinase; PI3K, phosphatidylinositol 3-kinase.

shutoff is mediated by PKA or PKC-induced phosphorylation of serine and threonine residues on the GPCR cytoplasmic tail, which leads to G protein uncoupling and formation of the inactive G protein heterotrimer. In addition, the cytoplasmic regions of GPCRs can be phosphorylated by G protein receptor kinases (GRKs). GRK-mediated phosphorylation can by itself lead to G protein uncoupling; however, GRK-phosphorylated GPCRs are also targets for arrestin binding, a process that further uncouples the receptor from G proteins. Arrestins in turn can mediate GPCR

internalization, often through clathrin-coated pits, to further shut off GPCR-mediated signaling. Depending on the GPCR, receptors are then either targeted to lysosomes for degradation, ubiquinated by intracellular factors, or dephosphorylated and returned to the cell surface in a process known as resensitization. Finally, GPCR shutoff is regulated by regulator of G protein signaling (RGS) proteins that function as GAPs to promote the conversion of GTP to GDP and Gα subunit shutoff. The extent of each of these mechanisms in regulating shutoff depends on the individual receptor as

299

CHAPTER 31 / MECHANISMS OF HORMONE ACTION

Table 31-3 Mechanisms of GPCR Shutoff Type of shutoff

Mediators

Mechanism

Second messenger

PKA, PKC

GRK/Arrestin-mediated

GRK 1-6 Visual and cone arrestins, β-arrestins 1 and 2, arrestins D and E

Trafficking RGS Proteins

Arrestins, clathrin, adaptin, caveolae >25 isoforms

Phosphorylation of GPCR, uncoupling of G proteins Phosphorylation of GPCR by GRK, arrestin binding to GPCR, GPCR Internalization, G protein uncoupling Internalization, lysosomal targeting, ubiquitinylation GTPase activating proteins (GAPs)

Shutoff of GPCR-induced signaling is mediated by several different mechanisms. Receptors can be phosphorylated by second messengers (PKA and PKC). Alternatively, GPCRs can be phosphorylated by G protein-receptor kinases (GRKs), followed by arrestin binding and receptor internalization and trafficking. Finally, intrinsic GTPase activity in the Gα subunit can be accelerated through interactions with regulator of G protein signaling (RGS) molecules.

well as the levels of the many GRK, arrestin, and RGS isoforms expressed in the target cell. Many disease states have been associated with mutations in either G proteins or GPCRs. For example, patients with McCuneAlbright syndrome have somatic mutations in the Gαs subunit that render it constitutively activated. Many of these mutations disrupt the intrinsic GTPase activity of the Gαs subunit, thus resulting in a Gαs protein that is resistant to inhibition by RGS proteins. Individuals with McCune-Albright syndrome have abnormalities that would be expected from overstimulation of Gαs-coupled receptors, including precocious puberty (LH receptor), thyroid tumors (TSH receptor), adrenal hyperplasia (adrenocorticotropic hormone receptor), growth-hormone secreting pituitary adenomas (growth hormone-releasing hormone receptor), café au lait spots (melanocyte-stimulating hormone receptor), and polyostotic fibrous dysplasia (PTH receptor). In contrast, patients with inactivating mutations in Gαs suffer from a complex array of syndromes known as pseudohypoparathyroidism. The clinical and biochemical phenotype of pseudohypoparathyroidism, which is inherited in an autosomal-dominant fashion, varies depending on the parent from which the mutated gene was inherited, ranging from hypocalcemia to the syndrome of Albright’s hereditary osteodystrophy. Many of the abnormalities associated with pseudohypoparathyroidism appear to be related to PTH resistance; however, resistance to other hormones that signal through Gαs have been reported as well. Diseases stemming from mutations in the GPCRs themselves are numerous; for example: 1. Inactivating mutations in the calcium sensor receptor that reduce the ability of the parathyroid glands to appropriately respond to serum calcium levels, resulting in increased PTH production, hypercalcemia, and hypophosphatemia. Heterozygotic mutations lead to the benign syndrome of familial hypocalciuric hypercalcemia, whereas homozygous mutations lead to neonatal severe hyperparathyroidism. 2. Activating mutations in the LH receptor in males that lead to precocious puberty and Leydig cell hyperplasia, whereas inactivating mutations lead to XY pseudohermaphroditism and Leydig cell hypoplasia.

3. Somatic activating mutations in the TSH receptor that result in hyperfunctioning thyroid nodules, and even thyroid cancer. Inactivating mutations have been associated with nonimmune hypothyroidism. 4. Inactivating mutations in the vasopressin receptor that result in accelerated receptor desensitization and internalization. These mutations are associated with nephrogenic diabetes insipidus, or vasopressin resistance.

INTRACELLULAR SIGNALING PATHWAYS As described, membrane receptors represent a critical cellular link to the extracellular environment. Within the cell, an intricate set of signaling pathways communicate receptor responses to other receptors, as well as to the cytoplasmic and nuclear compartments. For the most part, these signaling pathways consist of enzyme cascades, although there are exceptions, including gases such as nitric oxide (NO), lipids, and TFs (such as the STAT proteins). In many cases, receptors activate multiple signaling cascades. It appears that the use of combinations of pathways provides an important mechanism for generating specific cellular responses. Another characteristic of signal transduction pathways is the potential for “crosstalk” in which one pathway might activate or inhibit another pathway. Again, this generates the potential for diverse responses. In the next section, specific examples are provided for some of the better-studied signaling pathways central to endocrine responses. p21RAS AND MITOGEN-ACTIVATED PROTEIN KINASES Because growth factors induce a variety of cellular responses, including proliferation, there has been great interest in delineating the pathways that mediate growth factor responses. Tremendous progress has occurred in the elucidation of numerous cytoplasmic and nuclear kinases involved in growth responses. The number of different kinases is enormous, and they have been estimated to account for as many as 5–10% of transcribed genes. The family of MAPKs includes the extracellular signal-regulated kinases (ERKs), stress-activated protein kinases (SAPKs) or c-Jun N-terminal protein kinases, p38-kinases, and ERK5 (Fig. 31-7). As noted, tyrosine phosphorylation of growth factor receptors, such as EGF, creates a binding site for a variety of adaptor proteins. One adaptor,

300

SECTION IV / ENDOCRINOLOGY

Figure 31-7 Signal transduction cascades. Most cells express a wide variety of different growth factor receptors. Each of these receptors can activate an array of signaling pathways, not all of which are shown. Tyrosine kinases induce the binding of src-homology 2 (SH2) domain proteins, such as Grb-2. Grb-2 binds to guanine nucleotide exchange factors, such as son of sevenless (SOS). This pathway can also be activated by the insulin receptor through insulin-related substrate (IRS) or by cytokine receptors through JAKs. Although Gsα-coupled receptors act primarily through the camp pathway, they can also activate mitogen-activated kinases (MAPK) through their β- and γ-subunits. This figure emphasizes that different receptors can activate the ras pathway along with one or more downstream kinase cascades. Ras activates several downstream kinases in a cell type-specific manner. The preferential activation of a particular MAPK module, such as activation of the ERK rather than the SAPK pathway, provides a mechanism for specific responses. Many cellular targets, including these kinases and a variety of transcription factors, are involved in cellular growth responses.

Grb-2, binds to tyrosine-phosphorylated proteins through its central SH2 domain. Grb-2 contains flanking SH3 domains that recruit proline- rich guanine nucleotide exchange factors, such as SOS (Drosophila son of sevenless). After growth factor treatment (e.g., EGF), the Grb-2–SOS complex associates with the tyrosine-phosphorylated receptor, which leads to the activation of p21ras and induction of downstream kinases. The signal transduction pathways downstream of growth factor receptors involve a complex interplay of parallel, diverging, and converging pathways. The Ras proteins link growth factor receptor activation to protein phosphorylation and gene regulation through cascades of protein kinases. Induction of MAPKs requires dual phosphorylation on threonine and tyrosine residues by dual specificity MEKs. Several different protein kinases are capable of

functioning as MEKKs, including Raf-1 (see Fig. 31-7). In some cell types, the SAPKs and p38 kinases are activated by cellular stresses, including specific stimuli such as tumor necrosis factor (TNF)-α, ultraviolet (UV) irradiation, or genotoxic alkylating chemicals. The mechanisms by which one particular MAPK module is activated in response to a particular stimulus might depend on which combinations of kinases are present as well as interactions among various signaling cascades and specific scaffold molecules that bind to and organize the kinases participating in a particular MAPK module. TFs (e.g., c-Jun, c-Fos, activating transcription factor [ATF]-2, Elk-1) are one of the targets of MAPK cascades and provide an important mechanism for altering patterns of gene expression in response to extracellular signals. Phosphorylation of such TFs can alter interactions with transcriptional partners, change DNA binding affinity for a particular DNA sequence, or affect the activation surfaces that interact with the basal transcription apparatus. As an example, c-Jun is phosphorylated by several different kinases, including ERKs and SAPKs. Amino-terminal phosphorylation of c-Jun by the SAPKs enhances transactivation function, whereas phosphorylation of c-Jun near its DNA binding domain enhances DNA binding. As with many hormone regulatory systems, negative feedback is critical for dampening signal transduction systems once they have been activated. Because most of these pathways involve reversible protein phosphorylation, protein phosphatases provide a mechanism for temporally modulating specific signals. There might be as many as 1000 distinct protein phosphatase genes. The physiological importance of these phosphatases is exemplified by the ability of SV40 small t antigen to exert its cellular transforming effects by inhibiting protein phosphatase 2A, with consequent induction of MEK and ERK activity in the cell. Various growth factors induce phosphatases differentially. This might explain, in part, why treatment with one growth factor can result in sustained activation of ERK, whereas another factor causes transient stimulation of the kinase. Differences in the duration of signaling result in part through differential activation of G proteins in the Ras family. For example, rapid transient ERK activation is mediated by Ras coupling to Raf-1, whereas prolonged ERK activation involves an alternate small G protein family member known as Rap1, which couples to B-Raf in order to activate ERK. Additionally, signal strength and duration is translated into differential biological responses by altering the phosphorylation state, and thus the protein stability of short-lived TFs that are ERK substrates. Multisite, sequential phosphorylation of immediate early gene products, such as c-jun, c-fos, and c-myc, in response to sustained ERK activation stabilizes these molecules, resulting in their prolonged transcriptional activation, leading to increased cell proliferation. CALCIUM SIGNALING Many cellular responses involve alterations in intracellular calcium concentration as a mechanism for transmitting signals. There are two main pathways that initiate Ca2+ signaling. In excitable cells, the activities of voltage-dependent Ca2+ channels result in alterations of intracellular Ca2+ concentration. In nonexcitable cells, the slow inositol pathway (IP3) predominates, initiated either by receptor tyrosine kinases or by GPCRs of the seven transmembrane domain class. These receptors stimulate the hydrolysis of the membrane lipid phosphatidylinositol 4,5-bisphosphate by phospholipase C, leading to the production of diacylglycerol and IP3. IP3 binds to one of the

CHAPTER 31 / MECHANISMS OF HORMONE ACTION

301

Figure 31-8 cAMP stimulation of transcriptional responses. GPCRs provide a major pathway for the generation of cAMP by adenylate cyclase. Increased cAMP levels within the cell cause dissociation of the tetrameric protein kinase A (PKA) holoenzyme leading to the release of the active catalytic subunit of PKA. The active catalytic subunit is translocated to the nucleus, in which it phosphorylates transcription factor (TF) CREB, which binds as a homodimer to specific cAMP response elements (CREs) on target genes. The phosphorylation of CREB induces the binding of the coactivator, CREB-binding protein (CBP), which is also a substrate for phosphorylation. These transcription factors act by interacting with basal TFs such as TFIIB and TFIID. Several other kinases might phosphorylate CREB, including the calcium-regulated kinases.

members of the InsP3R family of integral endoplasmic reticulum (ER) membrane proteins, activating it as a Ca2+ channel to liberate stored Ca2+ from the ER lumen into the cytoplasm. The effects modulated by such changes in free intracellular Ca2+ are transient, owing to the actions of intracellular calcium binding proteins and pumps that serve to rapidly lower the concentration of free calcium. Such rises in intracellular calcium concentration regulate numerous processes by binding to intermediary proteins, such as calmodulin (CaM), which serves as an intracellular sensor of calcium concentration. CaM binds to a number of enzymes (CaM-dependent protein kinases), protein phosphatases, ion channels, phosphodiesterases, and adenylate cyclases to modify their activities. cAMP-DEPENDENT SIGNALING Alterations in intracellular cAMP concentration represent an important mechanism by which extracellular signals are transduced to regulate cellular function. cAMP is synthesized by nine members of the adenylate cyclase family from ATP. These membrane-associated enzymes are composed of conserved membrane-spanning and cytoplasmic catalytic domains and are regulated by the binding of hormones and drugs to GPCRs to modulate the levels of intracellular cAMP. cAMP levels can be further modulated by the effects of other influences, including processes intrinsic to the cell, such as the cell cycle, and the actions of members of the phosphodiesterase family. This large family of enzymes includes members grouped into 11 different protein families that serve to hydrolyze either cAMP and/or cGMP. Many of the pleiotropic array of cellular effects of cAMP are mediated by PKA, which acts on a number of cellular substrates, including enzymes, the cytoskeleton, and TFs. The effects of cAMP are mediated by its intracellular receptor PKA, which is made up of

two distinct subunits (regulatory [R] and catalytic [C]) in a tetrameric holoenzyme complex (R2C2). The regulatory subunits (RI and RII) contain two cAMP-binding domains at their carboxyl termini that regulate the assembly and activity of the complex. After the binding of cAMP, the holoenzyme dissociates into an R2 dimer and two free active catalytic kinase subunits (C), which phosphorylate cytoplasmic and nuclear target proteins (Fig. 31-8). An additional complicating factor is that the effects of cAMP in the cell might not be uniformly distributed. Instead, proteins such as the A-kinase anchoring proteins might lead to a gradient of cAMP activation, even within different regions of individual cells. Specific TFs are important targets of the PKA pathway. These include the CREBs and ATFs that are members of the bZIP class of TFs. Posttranslational modification of CREB by phosphorylation at specific serine residues induces conformational changes that alter the affinity of CREB for coactivator proteins, such as CREBbinding protein (CBP) or the TATA box-binding protein coactivator TAFII 110. CBP is thought to form a bridge between CREB and the basal transcription apparatus (Fig. 31-8). CBP also interacts with other bZIP proteins, such as c-Jun and c-Fos, other TFs, including c-Myb, as well as specific kinases. In addition to a clear role in cAMP signaling, CBP has also been implicated in mitogenic signaling and functions as a transcriptional coactivator with intrinsic histone-acetylase and ubiquitin ligase activities; both enzyme activities play key roles in transcription. Consistent with its role in multiple cell signaling pathways and transcriptional regulation, mutations of CBP cause Rubinstein-Taybi syndrome, which is associated with mental retardation, multiple congenital anomalies, and predisposition to malignancy. The bZIP dimerization structure of the CREB and ATF proteins provides the basis for

302

SECTION IV / ENDOCRINOLOGY

numerous combinations of different members of this family. The closely related cAMP response element modulator gene product can act as a dominant negative regulator of CREB transcriptional activity, either by forming inactive heterodimers with CREB or by binding to the CRE as a component of an inactive complex. Several other DNA regulatory elements and TFs are capable of stimulating gene transcription in response to increased cAMP and PKA activation. The TF, activator protein 2, also functions in basal, phorbol ester, and cAMP-mediated transcriptional induction. The CAAT enhancer binding protein (C/EBP) induces transcription of the adipocyte protein 2 gene promoter and stearoyl acyl CoA desaturase through a region that is also regulated by cAMP through PKA. cAMP-responsive regions of several genes appear to overlap with binding sites for NRs, some of which might also transduce cAMP effects. Despite the range of effects mediated by these pathways, some effects of cyclic AMP could not be reproduced by the effects of activated PKA, suggesting that cAMP might regulate cellular processes by distinct mechanisms. Cyclic nucleotide gated ion channels have been identified that are directly modulated by the binding of cAMP to a ligand binding pocket on the cytoplasmic surface of the channel. In addition, two novel cAMP receptors, Epac1 and Epac2 (exchange protein directly activated by cAMP; also called cAMP-GEFI and cAMP-GEFII), have been identified. EpacI and II are guanine nucleotide exchange factors that activate the small G proteins Rap1 and Rap2 following the binding of cAMP. cGMP-DEPENDENT SIGNALING In addition to the GCs that are structural components of membrane receptors (above), a smaller number of soluble GCs (also referred to as NO-sensitive GCs) have been identified. Two different α and two different βsubunits have been identified: α1, α2, β1, and β2. The active complexes are heteromeric, consisting of one α and one β-subunit. Each subunit consists of carboxyl terminal GC catalytic domains, which are conserved compared to the peptide receptor GCs. The amino termini of these proteins are less conserved and encode the segment responsible for coordinating a heme prosthetic group. This heme prosthetic group serves as the NO sensor and is responsible for the modulation of GC activity in response to changes in the intracellular concentration of NO. The NO is produced locally by members of the NO synthase family of enzymes, or pharmacologically, by agents such as nitroglycerine. As with members of the GC receptor family, the cellular effects of cGMP are mediated by effects on the activities of cGMP-dependent protein kinases, cyclic nucleotide-regulated ion channels, and cGMP-regulated cyclic nucleotide phosphodiesterases. Members of the phosphodiesterase family can further serve to modulate levels of cGMP; Sildenafil enhances the effect of NO in the regulation of vascular smooth muscle tone by inhibiting phosphodiesterase type 5.

NF-κB/REL FAMILY OF TRANSCRIPTION FACTORS NF-κB was originally identified as a protein in nuclear extracts of B cells that bound to the enhancer element of the κ light chain gene. This binding activity was demonstrated to be present in the cell in a “latent” fashion, which could be rapidly activated in the presence of protein synthesis inhibitors by a variety of stimuli, including TNF, lipopolysaccharide, interleukin-1, phorbol esters, growth factors, and UV irradiation. The basis for the rapid and protein synthesis-independent nature of NF-κB activation was found to reflect the association of the active binding moiety with

an inhibitory subunit, IκB. The dissociation of IκB from NF-κB results in the rapid appearance and translocation of NF-κB DNA binding activity to the nucleus. The active form of NF-κB was found to be dimeric, exerting its effects by binding to specific target sequences with the genome to regulate gene expression. Subsequent studies have considerably complicated this scenario. Five members of the NF-κB/Rel protein family have been identified, each with substantial sequence homology in what is termed the Rel homology domain. This segment is located at the amino terminus of each protein and contains elements critical for interaction with the inhibitory IκB proteins, dimerization, and DNA binding. Members of this protein family function as dimers, which may consist of identical or nonidentical subunits. In their inactive state, dimers of the NF-κB/Rel proteins are associated with monomers of the IκBα or IκBβ proteins. NF-κB activation is mediated by the IκB (IKK) complex, made up of catalytic components (IKKα and IKKβ) and a regulatory subunit (IKKγ or NEMO). Following stimulation by a number of important effectors, the IKK complex phosphorylates IκB on specific residues, promoting its ubiquitination and degradation. NF-κB activation is central to the regulation of immune response, owing to its role in the regulation of a host of cytokines and chemokines. It has become clear that NF-κB activation plays an important role in the regulation of apoptosis. Indeed, in some systems, activation of NF-κB has been demonstrated to antagonize the effects of chemotherapy and radiation in inducing apoptosis. Such findings have led to attempts to sensitize tumors to the actions of chemotherapy and radiation by blocking the activation of NF-κB. In addition to diseases in which abnormalities of NFκB expression occur, defects of the NF-κB signaling pathways have been identified. Incontinentia pigmenti is a rare X-linkeddominant disorder that affects ectodermal tissues resulting in abnormal skin pigmentation, retinal detachment, anodontia, alopecia, nail dystrophy and central nervous system defects. Over 90% of IP carrier females have a recurrent multiexon deletion of the IKKγ gene, which is required to activate the NF-κB pathway. In IP, mutations in IKKγ lead to the complete loss of NF-κB activation creating a susceptibility to cellular apoptosis in response to TNF-α. Although females, hemizygous as a result of X-chromosome inactivation, survive, this condition is lethal for males. CELL-CYCLE REGULATORY SIGNAL TRANSDUCTION PATHWAYS Phosphorylation plays an essential regulatory role in the cell cycle, and a large array of cyclin-dependent kinases (CDKs) has been identified (Chapter 4). The regulatory subunits of the CDKs, known as cyclins, form complexes with their catalytic partners to function as heterodimeric holoenzymes that phosphorylate specific proteins, including the tumor suppressor protein pRb (Chapter 4). Phosphorylation of pRb blocks its critical inhibitory function, allowing cell-cycle progression and differentiation to occur. Several proteins capable of binding cyclins and inhibiting CDK activity have been identified and are referred to as CDKIs. The CDKIs inhibit cell-cycle progression and inhibit tumor formation. Translocation of the cyclin D1 gene has been associated with certain cases of hyperparathyroidism (Chapter 36), and actually led to the identification of cyclin D1 in humans. In these cases, a somatic inversion of chromosome 11 brings cyclin D1 under the control of the PTH promoter. Consequently, cyclin D1 is overexpressed in the parathyroid cell and results in cellular proliferation. The proliferative and transforming effects of cyclin D1 are independent of its ability to activate cyclin-dependent protein kinases,

CHAPTER 31 / MECHANISMS OF HORMONE ACTION

but instead require its interaction with the CEBP/β TF, in which cyclin D1 relieves the transcriptional repressive activity of CEBP/β at multiple gene promoters. NUCLEAR RECEPTORS Nuclear Receptor Action The NR superfamily consists of structurally related proteins that are important modulators of gene expression in eukaryotes. The structures of the different family members are widely distributed in many species and are highly conserved, particularly in the segments encoding the DNA-binding domain (DBD) and ligand-binding domain (LBD) of the receptor proteins. In humans, 48 different members of this family have been identified, based on sequence analysis of the human genome. In some circumstances, the NRs mediate the physiological actions of small cell-permeable hormones, such as the sex steroids (estrogen, testosterone, and progesterone), cortisol, aldosterone, thyroid hormone, and vitamin D, as well as retinoids that are derived from dietary vitamin A. In other instances, no ligand has been identified. These receptors might not require the binding of a ligand for functional activation and as a group are referred to as “orphan” NRs. Structure and Classification of Nuclear Receptors From the first analyses of NR structure and function, it was apparent that these proteins were organized in a modular fashion. The DBD, made up of two centrally located “zinc fingers,” is the most highly conserved segment (Fig. 31-9). The structure of this segment has been well characterized using site-directed mutagenesis, X-ray crystallography, and nuclear magnetic resonance studies. Residues at the carboxyl terminal base of the first zinc finger are principal determinants of DNA binding specificity, where, as adjacent sequences at the base of the second zinc finger are critical to the formation of the dimerization interface. The DBDs of NRs that bind as monomers differ slightly in carboxyl terminal extensions of the DBD that contribute to the DNA binding by this element. Regions carboxy-terminal to the DBD specify sequences that localize the protein to the nucleus. Other functionally important segments of the NRs have been delineated. For members that bind ligand, the LBD is located in the carboxyl terminus of the receptor. In addition to its role in ligand binding, this region contains several overlapping domains, including regions required for dimerization of the receptor and for transcriptional activation and repression. The structures of the LBDs of a number of NR family members have been solved by Xray crystallography. These studies demonstrate that ligand binding induces important conformational changes that underlie the changes of activity of the NR in the regulation of gene activity. The amino-terminal segments of the NRs are the most variable and the least well characterized. In some cases, it contains additional transcription-activating domains and serves as a “docking site” for interactions with other proteins. NRs can be classified by several different schemes. One useful method employs the mechanisms by which these molecules recognize DNA (see Fig. 31-10). The classic steroid receptors share a similar DBD and bind to DNA sequences similar to the inverted palindromic sequence depicted. This group includes receptors for glucocorticoids, mineralocorticoids, progesterone, and androgens. The steroid receptors have been particularly well characterized and exist in the cell complexed with proteins of the heat shock family in the absence of ligand. Following the binding of ligand, the receptors dissociate from the heat shock proteins, translocate to the nucleus, and bind monomers to target DNA sequences within or adjacent to responsive genes.

303

Other NRs differ in the manner that they bind to target DNA sequences. The diverse patterns of recognition include the binding as monomer, heterodimers, and homodimers (see Fig. 31-10). The retinoid X receptors (RXRs) serve as heterodimeric partners for several different NRs, including retinoic acid receptors (RARs), thyroid hormone receptors, vitamin D receptor (VDR), and peroxisome proliferator-activating receptors (PPARs). The heterodimers typically bind to hormone response elements (HREs) that are arranged as direct repeats rather than in a palindromic manner. The spacing between these half-sites, and variations in some of the contextual bases, provide specificity to different receptors, providing a partial explanation for how so many different receptors can share a common binding motif but elicit distinct cellular responses. A number of members of the NR family were identified on the basis of sequence homology to the DBDs of characterized members of the NR family (e.g., the steroid receptors). This cloning exercise yielded cDNAs encoding proteins with structural similarities to the NRs (e.g., with the LBDs and DBDs), but for which no ligand had been defined. With time, ligands for several of these ‘orphan’ receptors have been identified. For example, the RXRs were initially classified as orphan receptors, but are known to bind 9-cis retinoic acid. The PPARs bind a variety of eicosanoids, as well as thiazolidinediones that are used to treat noninsulindependent diabetes. It remains to be determined whether ligands exist for all orphan members of the NR family. It appears likely that some members might not bind ligand, and instead might be constitutively active or regulated by other mechanisms, such as phosphorylation. In addition to the diversity in types of receptor the NR family, in several cases receptor isoforms encoded by different genes provide an additional level of control. For example, there are two separate genes for TRs, two ERs, three for all-trans RARs, and three RXRs. In addition, many of these genes generate multiple receptor subtypes by virtue of alternate mRNA splicing or alternate promoter or initiator methionine usage. The function of individual isoforms has been defined in only a small number of instances, but in the case of the PR, the A- and B-isoforms have physiologically distinct roles. Receptor subtype expression is highly regulated during development, and there is striking tissue-specific expression of various NR isoforms in adult organs. Therefore, like the peptide hormone receptors, tissue responsiveness to the hormone signal is regulated at its most fundamental level by selective expression of the NR in responsive tissues. Nuclear Receptor Function and the Modulation of Gene Expression NRs interact with specific DNA sequences that are frequently located within or adjacent to promoter regions of regulated genes. After the receptor binds to DNA, it is positioned to interact with other TFs to alter rates of gene transcription. Considerable progress has been made toward defining the mechanism(s) by which NRs modulate these effects. Several of the NRs physically interact with the basal TF, TFIIB. Using in vitro transcription analysis, the NRs have been shown to facilitate the rate of assembly and stabilize basal TFs in the preinitiation complex that forms at the sites of transcriptional initiation. An additional level of transcriptional control occurs through NR interactions with “coactivator” or “corepressor” proteins, which in turn interact with proteins of the basal TF complex. Several of these proteins have been identified on the basis of their ability to interact with NRs, using biochemical techniques or genetic screens in yeast. In general, coactivators recruit proteins that result

304

SECTION IV / ENDOCRINOLOGY

Figure 31-9 Modular structure of the NR family. Members of the NR family contain segments that are conserved to varying degrees among the family members. (A) A schematic structure of a prototypic NR is shown. The conserved ligand binding and DNA binding domains are indicated (hatched and filled rectangles, respectively). The relative positions of the activating functions (AF)-1 and -2 within the amino terminus and carboxyl terminus of the receptor proteins are shown. Sequences responsible for the nuclear localization of these proteins have been localized to the carboxyl terminal end of the DNA-binding domain. (B) Although the relative positions of the individual domains are in most instances maintained, the degree of sequence conservation varies widely. Members of the nuclear receptor family exhibit the highest degree of conservation when the amino acid sequences of the DNA binding domains of the receptor proteins are compared. Lesser degrees of homology are evident between the sequences that are made up of the LBDs of the receptors. In the example shown, the predicted amino acid sequences of three different members of the nuclear receptor family are compared. The degree of relatedness is shown for each of the two receptors when aligned with the predicted amino acid sequence of the human glucocorticoid receptor. The amino-terminal segments of the receptors differ considerably in size and sequence. The extent of homology is 20% of individuals evaluated by ultrasonography. In contrast, thyroid cancers only account for approx 1% of malignancies. Although the molecular defects underlying the pathogenesis of thyroid nodules and adenomas as well as papillary, follicular, and anaplastic thyroid cancers are far from being completely understood, new insights have been obtained with the tools of molecular biology (Table 35-8). A majority of thyroid nodules are of monoclonal origin making somatic mutations in growth-controlling gene a likely cause for their development. Two large groups of genes have been implicated in uncontrolled cell proliferation and clonal expansion of mutated cells. Protooncogenes are activated through gain-of-function mutations in the regulatory or coding sequence thus creating the oncogene. This results in overexpression of the normal product or an abnormal form of the protein. Tumor suppressor genes control cell proliferation under normal conditions. Unrestrained growth of cells and thus formation of tumors can occur through lossof-function of these genes. Among other important insights, the detection of certain molecular defects has modified diagnostic and therapeutic procedures. For example, mutations in the RET gene, which encodes the RET tyrosine kinase, are the molecular alterations found in medullary thyroid cancer (MTC) and the multiple endocrine neoplasia type 2 (MEN2) syndromes (Chapter 40). It is now possible to precisely detect carriers of the disease in early childhood and thus treat these patients accordingly. This chapter gives an overview on the variety of defects affecting thyroid function and growth at all levels of the axis.

THYROTROPIN-RELEASING HORMONE DEFICIENCY TRH is a hypothalamic tripeptide (Glu-His-Pro) generated from a large prohormone of 27 kDa. Besides its role as a stimulator

CHAPTER 35 / THYROID DISORDERS

339

Figure 35-2 Thyroid hormone synthesis in a follicular cell. For details see text. AC, adenylyl cyclase; IGF-I, IGF-I R, insulin-like growth factor I (receptor); MIT, DIT, mono- and diiodotyrosine; PKA, protein kinase A; PKC, protein kinase C; PLC, phospholipase C; TG, thyroglobulin; TPO, thyroperoxidase.

of the pituitary gland in which it releases TSH and prolactin, TRH is a neurotransmitter in numerous areas of the brain and is also found in some peripheral organs. A few patients with congenital hypothyroidism and isolated TRH deficiency without destructive hypothalamic lesions have been reported. The diagnosis of central hypothalamic hypothyroidism is made based on the constellation of low TSH and an increase thereof after administration of exogenous TRH. The molecular defect underlying these cases remains elusive and could affect synthesis or secretion of TRH. Targeted disruption of the TRH gene in mice led

to an overtly hypothyroid phenotype. Remarkably, the TSH levels were elevated in these mice but displayed diminished biological activity. Similar biochemical constellations with elevated TSH with reduced bioactivity have been found in some individuals with central hypothyroidism.

RESISTANCE TO TRH Resistance to TRH in pituitary thyrotrophs was discovered in a boy with isolated central hypothyroidism. His T4 was decreased, the TSH was normal, and there was no increase of TSH and prolactin

340

SECTION IV / ENDOCRINOLOGY

Figure 35-3 Structure of thyroxine (T4), triiododothyronine (T3) and reverse triiododothyronine (rT3). T4 is activated to T3 by deiodinase 1 and 2. T4 is inactivated into rT3 by deiodinase 3.

Figure 35-4 (A) Nuclear thyroid hormone action. Thyroxine (T4) enters the cell in which it is converted into triiododothyronine (T3) by 5′-monodeiodinases. In the nucleus, T3 interacts with thyroid hormone receptors. Thyroid hormone receptors bind to specific sequences, thyroid response elements (TRE) in the promoter regions of target genes in conjunction with other nuclear receptors such as the retinoid X receptor (RXR). (B) Schematic structure of the most abundant thyroid hormone receptor isoforms and their functional domains. The DNA binding domain consists of two zinc fingers. TRβ2 is most abundant in the pituitary gland in which it mediates negative regulation of thyrotropin (TSH). TRα2 does not bind T3; its exact physiological role is unknown. DBD, DNA binding domain; CoR box, interaction with corepressors; LBD, ligand binding domain; AF2, activation domain 2. (C) Model of thyroid hormone action on a positively regulated gene. In the absence of thyroid hormone, corepressors interact with thyroid hormone receptor homo- or heterodimers and silence transcription. Deacetylation of histones is associated with silenced transcription. Thyroid hormone relieves the interaction with corepressors and after binding of coactivators and factors of the basal transcription apparatus, gene expression is started, a process that involves chromatin remodeling through histone acetylation. RXR, retinoid X receptor; TR, thyroid hormone receptor; HDAC, histone deacetylase; CoR,Corepressor; CoA, Coactivator; CBP, CREB binding protein; HAT, histone acetyltransferase.

341

CHAPTER 35 / THYROID DISORDERS

Table 35-1 Main Clinical and Biochemical Findings in Hyper- and Hypothyroidism Hyperthyroidism

Hypothyroidism Signs

Hyperactivity Tachycardia/Arrhythmia Hyperthermia Increased perspiration Hyperreflexia Muscle weakness Tremor Weight loss Eyelid retraction Exophthalmos (Graves’ disease) Diffuse or nodular goiter

Lethargy Bradycardia Cold intolerance Dry skin Hyporeflexia Myxedema Hoarseness Weight gain

Intrauterine/neonatal Mental retardation possible Mental retardation Advanced bone age Neurological deficit Retarded bone age/growth Symptoms Nervousness Weakness Palpitation Increased appetite Irregular menses

Fatigue, sleepiness Depression Constipation Decreased appetite Irregular menses Paresthesia Laboratory

T4 ⇑, T3 ⇑, TSH ⇓ T4 ⇑, T3 ⇑, TSH ⇑

(a) primary (b) central

T4 ⇓, T3 ⇓, TSH ⇑ T4 ⇓, T3 ⇓, TSH ⇓

T3, triiododothyronine; T4, thyroxine; TSH, thyrotropin.

in response to TRH. Mutational analysis of the TRH receptor gene revealed compound heterozygous point mutations that inactivate the TRH receptor.

AUTONOMOUS PRODUCTION OF TSH TSH is a heterodimeric glycoprotein hormone. It shares a common α-chain with the pituitary follicle-stimulating hormone (FSH) and luteinizing hormone (LH) as well as the placental hormone choriogonadotropin (CG). Each of these hormones has a unique β-chain. Two novel human glycoprotein hormone-like genes α2 (A2) and β5 (B5) have been cloned. The heterodimer formed by these subunits, referred to as thyrostimulin, activates the human TSH receptor, but not the LH and FSH receptors. Thyrostimulin is expressed in the pituitary and, given that TSH receptors are also found in a subset of anterior pituitary cells, it has been proposed that it may act as a paracrine hormone. Its physiological role and relevance remain, however, unresolved. Hypersecretion of TSH by thyrotrope pituitary adenomas is a rare cause of hyperthyroidism. Biochemically, these patients are characterized by an inappropriate secretion of TSH and hyperthyroxinemia. Typically the secretion of free α subunit is excessive. Rarely, there is cosecretion of growth hormone. MRI and or CT-scan confirm the diagnosis. The molecular defect in the thyrotropes resulting in autonomous growth and function remains unclear. Clonal analysis using the X-inactivation technique in a very small number of tumors showed that they are of monoclonal origin suggesting that a somatic mutation or the loss of a tumor suppressor is at the onset of their development. Screening of several candidate genes (Gsα, Gqα, G11α, TRH receptor) in a small number of such tumors did not show alterations in these candidate genes. In a single TSHoma, a somatic mutation in the TRβ gene was identified (H435Y). This mutant TRβ has impaired T3 binding, and the T3-mediated negative regulation of the glycoprotein hormone α subunit and TSHβ genes is impaired suggesting that it may be responsible for the defect in negative regulation of TSH by thyroid hormone in the tumor.

Table 35-2 Hypothalamic and Pituitary Defects Gene/protein

Phenotype

TRH TRH receptor

Hypothalamic hypothyroidism Pituitary hypothyroidism No increase of TSH and prolactin in response to TRH TSH-deficient hypothyroidism Prolactin increase to TRH TSH-secreting adenomas with hyperthyroidism

TSH

POU1F1 (PIT1)

CPHD: GH, prolactin, TSH

PROP1 LHX3

CPHD: GH, prolactin, TSH, LH, FSH, (ACTH) CPHD: GH, prolactin, TSH, LH, FSH Rigid cervical spine CPHD: GH, TSH, ACTH Cerebellar and hindbrain defects CPHD with variable anterior pituitary hormone deficiencies SOD

LHX4 HESX1 (RPX)

Molecular defect

Inheritance

Chromosome

Unknown Inactivating mutations

AR AR

3q13.3-q21 8q23

Mutations in TSHβ-chain

AR

β-chain 1p22

Unknown TRβ mutation (1 case) Dominant negative or recessive mutations Inactivating mutations Inactivating mutations

Sporadic

Splice site mutation Haploinsufficiency? Homozygous or heterozygous point mutations

AD, AR

3p11

AR AR

5q 9q34.3

AD

1q25

AR, AD

3p21.2-p21.1

ACTH, adrenocorticotropic hormone; AD, autosomal-dominant; AR, autosomal-recessive; CPHD, combined pituitary hormone deficiency; FSH, follicle-stimulating hormone; GH, growth hormone; LH, luteinizing hormone; SOD, septo-optic dysplasia; TRH, thyrotropin-releasing hormone; TSH, thyrotropin.

342

SECTION IV / ENDOCRINOLOGY

Table 35-3 Thyroid Hormone Synthesis Defects Gene/protein

Phenotype

TSH-receptor

Familial nonautoimmune hyperthyroidism Sporadic congenital nonautoimmune hyperthyroidism Toxic adenomas Hypersensitivity to hCG Euthyroid hyperthyrotropinemia Hypothyroidism Toxic adenoma

GNAS1/Gsα

McCune-Albright syndrome

NIS (SLC5A5) Sodium-iodide symporter Thyroperoxidase PDS (SLC26A4)

Thyroglobulin

THOX2

Dehalogenase

Pseudohypoparathyroidism Ia: Hypothyroidism, hypogonadism, AHO Hypothyroidism with defective iodide uptake Congenital hypothyroidism, goiter Iodide organification defect Pendred’s syndrome: Sensoneurinal deafness, goiter, partial organification defect Goiter with compensated or overt hypothyroidism Classic endoplasmic reticulum storage disease with retention of mutated protein Transient or permanent hypothyroidism Iodide organification defect because of deficient H2O2 generation Congenital hypothyroidism, goiter, loss of iodide through secretion of DIT, MIT

Molecular defect Gain of function mutations in extracellular and transmembrane domain

Inheritance AD, Sporadic

Chromosome 14q31

SomaticAD Increased affinity for hCG Partially or completely inactivating mutations Somatic activating point mutations Mosaicism for gain of function mutations Inactivating point mutations on maternal allele Inactivating point mutations

AR

19p13.2-p12

Inactivating point mutations

AR

2p25

Inactivating mutations, small deletions

AR

7q31

Point mutations and splice site mutations

AR

8q24.2-q24.3

Monoallelic or biallelic mutations

Sporadic

15q15.3

Point mutations

AR

6q25.1

AR Sporadic

20q.13.2

Sporadic AD

AD, autosomal-dominant; AR, autosomal-recessive; DIT, diiodotyrosine, MIT, monoiodotyrosine; NIS, sodium-iodide symporter; TSH, thyrotropin.

Therapy consists in most cases in transsphenoidal removal of the pituitary adenoma. The somatostatin analog octreotide is useful in some patients unable to undergo surgery or with invasive tumors. Preoperatively, hyperthyroid patients should be controlled with thyreostatic drugs.

COMBINED PITUITARY HORMONE DEFICIENCY AND ISOLATED TSH DEFICIENCY TSH deficiency may occur in the setting of panhypopituitarism, most commonly owing to a pituitary adenoma. Genetic defects in the development and function of the pituitary gland can result in various forms of combined pituitary hormone deficiency (CPHD) (Table 35-2). Patients with CPHD present with impaired production and secretion of one or several anterior pituitary hormones that may include TSH. Some patients also display midline malformations. Mutations in the TSHβ gene can result in isolated TSH deficiency. Therapy aims at substituting T4, and, if deficient, other hormones. POU1F1 POU1F1 (traditionally referred to as PIT1) is a pituitary-specific transcription factor regulating the development of somatotropes, lactotropes, and thyrotropes as well as gene expression of growth hormone, prolactin and the β-subunit of TSH. It is a member of the POU family of transcription factors, an eponym that was created after the identification of the first three members of this group, Pit-1, Oct-1, and Unc-86. POU1F1 contains two

main functional domains, an activation domain and a DNA binding domain formed of a POU-specific and a homeodomain. The homeodomain is highly similar to the homeobox containing genes in Drosophila that control development of specific body segments, whereas the POU-domain is necessary for high affinity DNA binding. DNA sequences, which bind the transcription factor, are found in the promoters of the GH, prolactin, TSH-β, and POU1F1 genes. Several splice variants with variable activation domains have been reported and may define promoter specificity. Mutations in the murine homolog Pit1 have been identified in the Snell and Jackson dwarf mice, which have deficiencies in GH, prolactin, and TSH. Patients with identical hormone deficiencies have mutations in the human homolog POU1F1. POU1F1 mutations can be inherited in an autosomal-dominant or -recessive manner. Recessive mutations alter the transactivation or DNA binding properties of the transcription factor. The dominant mutations bind to DNA, but impair transcription by the wild-type allele by altering dimerization and or interaction with other nuclear transcription factors. A case of fetomaternal POU1F1 deficiency resulting from a heterozygous, dominant negative point mutation led to absolute fetal hypothyroidism in the fetus with dramatic delay in respiratory, cardiovascular, neurologic, and bone maturation. The absence of Pit-1 and the ensuing hypoprolactinemia in the mother resulted in puerperal alactogenesis. This case impressively illustrates the importance

343

CHAPTER 35 / THYROID DISORDERS

Table 35-4 Developmental Defects Gene/protein PAX8 TTF-1 (NKX2.1)

TTF-2 (FOXE1)

Phenotype

Molecular defect

Thyroid hypoplasia or ectopy Hypothyroidism Mild hypothyroidism Respiratory distress Choreoathetosis Mental retardation Bamforth-Lazarus syndrome: Thyroid agenesis Cleft palate Choanal atresia Bifid epiglottis Spiky hair

Inheritance

Chromosome

Point mutations (Haploinsufficiency?) Chromosomal deletions and point mutations: Haploinsufficiency

Sporadic AD Sporadic

2q12-q14 14q13

Inactivating point mutations

AR

9q22

AD, autosomal-dominant; AR, autosomal-recessive; TTF, thyroid transcription factor. Table 35-5 Binding Proteins Defects Gene/protein

Phenotype

TBG Thyroxine-binding globulin

TTR Transthyretin

Albumin

Decreased total T4 levels, euthyroidism Increased total T4 levels, euthyroidism Hyperthyroxinemia Familial amyloidotic polyneuropathy Familial dysalbuminemic hyperthyroxinemia Familial dysalbuminemic hypertriiodothyroninemia

Molecular defect

Inheritance

Chromosome

Deletions, point mutations Gene amplification

X-linked recessive

Xq22.2

Point mutations

Sporadic de novo germline mutations AD

18q11.2-q12.1

Point mutations

AD

4q11-13

Point mutations

AD

AD, autosomal-dominant; T4, thyroxine. Table 35-6 Other Genetic Defects Associated With Thyroid Dysfunction Gene/protein MCT8 (SCL16A2)

Deiodinase 3

KCNE3

Phenotype Hemizygous males: Elevated T3 and TSH Severe mental retardation Spastic quadriplegia Rotary nystagmus Impaired gaze and hearing Heterozygous females: only thyroid function abnormalities Consumptive hypothyroidism because of increased degradation of T4 and T3 in hemangiomas and other vascular tumors Thyrotoxic periodic hypokalemic paralysis

Molecular defect

Inheritance

Chromosome

Deletions, point mutations

AR, sporadic

Xq13.2

Overexpression

Sporadic

14q32

Monoallelic point mutations

Sporadic?

11q13-q14

AR, autosomal-recessive; T3, triiododothyronine; T4, thyroxine; TSH, thyrotropin.

of POU1F1 in the control of different endocrine axes and the importance of prenatal thyroid hormone for fetal maturation. PROP1 Characterization of the Ames dwarf mouse led to the cloning of the paired-like homeodomain factor Prop-1 (Prophet of pit1). PROP-1 is necessary for the expression of POU1F1 and is involved in ontogenesis, differentiation, and

function of somatotropes, lactotropes, thyrotropes, and possibly gonadotropes. Inactivating mutations in the human PROP-1 gene have been identified as a cause of an autosomal-recessive CPHD phenotype affecting these cell lineages. Of note, there is variability in the phenotypic expression among CPHD patients with PROP-1 mutations.

344

SECTION IV / ENDOCRINOLOGY

Table 35-7 Resistance to Thyroid Hormone Gene/protein TRβ

Unknown

Phenotype

Molecular defect

Resistance to thyroid hormone. Goiter, variable degrees of hypo- and or hyperthyroidism

Resistance to thyroid hormone without mutations in TRβ

Inheritance

Chromosome

Heterozygous point mutations: Dominant negative activity of mutated allele One kindred with total deletion of coding region of TRβ locus

AD, sporadic

3p24.3

Unknown

AD

AR

unknown

AD, autosomal-dominant; AR, autosomal-recessive. Table 35-8 Molecular Alterations in Benign and Malignant Thyroid Tumors Gene

Chromosomal location

Lesion

TSHR

14q31

Point mutations

Gsα

20q13.2

Point mutations

B-RAF RET/PTC

7q34 RET 10q11.2

PAX8/PPARγ

PAX8 2q12-q14

RET TRK RAS

p53

10q11.2 1q23-24 Hras 11p15.5 Kras 12p12.1 Nras 1p13.2 17p13

Point mutations Rearrangements RET/PTC1: paracentric inversion with H4 chromosome 10 RET/PTC2: reciprocal translocation with chromosome 17 with RIα PKA subunit RET/PTC3: intrachromosomal rearrangement with ELE1 gene Rearrangements t(2;3)(q13;p25) Point mutations Intrachromosomal rearrangements Point mutations

Rb

13q14.1-14

Point mutations Deletion, Insertion mRNA variants

p16 (MITS)

9p21

Deletion of coding region

p21/WAF

6p21.2

Reduced expression

MET C-MYC LOH

Overexpression Overexpression LOH

APC

7q31 8q24.12-q24 3p, 11q13 and multiple other loci 5q21-q22

PTEN

10q23.31

PRKAR1A

17q23-q24

Autosomal-dominant inheritance with somatic second hit Autosomal-dominant inheritance with somatic second hit Autosomal-dominant inheritance with somatic second hit

Tumor Toxic adenoma Differentiated carcinoma Adenoma, toxic adenoma, differentiated carcinomas PTC PTC

FTC MEN, MTC MNG, PTC Adenoma, differentiated thyroid carcinoma Differentiated carcinoma Anaplastic adenoma Differentiated carcinoma Anaplastic adenoma Differentiated carcinomas cell lines Differentiated and anaplastic carcinomas FTC Differentiated carcinoma FTC PTC in Gardner syndrome PTC in Cowden syndrome FTC in Carney complex type 1

FTC, follicular thyroid carcinomas; LOH, loss of heterozygosity; PPAR, peroxisome proliferators-activated receptor; PTC, papillary thyroid carcinomas; Rb, retinoblastoma susceptibility gene.

LHX3 Recessive mutations in LHX3, a LIM homeodomain transcription factor, also cause CPHD of all anterior pituitary hormones with the exception of adrenocorticotrophic hormone. In addition, these patients have a rigid cervical spine and a limited ability to rotate the head. In Lhx3–/– mice, the primordium of the

anterior pituitary is present, but with the exception of the corticotroph cells, there is no further growth and differentiation of the rostral part of the gland. LHX4 Heterozygous mutations in the human LIM homeobox transcription factor LHX4 lead to an autosomal-dominant pheno-

CHAPTER 35 / THYROID DISORDERS

type characterized by short stature, pituitary and cerebellar hindbrain defects, and abnormalities of the sella turcica. HESX1 HESX1, also referred to as RPX, is a member of the paired-like class of homeobox transcription factors. Familial septo-optic dysplasia (SOD), a syndromic form of CPHD associated with optic nerve hypoplasia and agenesis of midline structures in the brain, can be caused by homozygous mutations in HESX1. A similar phenotype was observed in Hesx1–/– mice. Interestingly, a small proportion of the mice heterozygous for a Hesx1 null allele also had a milder form of SOD. This discovery prompted further screening of patients presenting with a wide spectrum of congenital pituitary dysfunctions. A subset of these patients was heterozygous for HESX1 mutations. Phenotypically, heterozygous HESX1 mutations result in various constellations of pituitary hormone deficiencies and the phenotype is variable among family members with the same mutation. TSHβ MUTATIONS Congenital hypothyroidism caused by isolated hereditary TSH deficiency is a rare autosomal-recessive disease caused by mutations in the TSH β-chain. In these patients, TSH is unmeasurable or very low, and the administration of TRH does not result in a rise in serum TSH. The levels and the function of the other pituitary hormones are normal, including an adequate rise of prolactin in response to TRH. The first familial incidence was described in a Japanese family in two sisters with cretinism. Two other families from the same island were reported to harbor the same defect and it is thus likely that they originate from a common founder. Several independently recurrent mutations have now been reported in 18 families that are, in part, of distinct ethnic origin. They include point mutations, frame shift mutations and a splice site mutation. The C105V/114X frameshift mutation destroys a disulfide bond essential for normal protein conformation and bioactivity, other mutations lead to disruption of heterodimer formation with the β-chain (e.g., G29R), and the reported splice site mutation results in skipping of exon 2.

AUTONOMOUS FUNCTION OF THE TSH RECEPTOR TSH exerts its effects on thyroid follicular cells through the TSH receptor, a member of the G protein coupled seven transmembrane receptors. Together with the receptors for FSH and LH it forms a distinct subfamily defined by a large amino-terminal extracellular domain involved in binding of the hormone. The 744 amino acid receptor is encoded by a gene containing 10 exons and localized on chromosome 14q31. The TSH receptor is coupled to Gs and thus to the adenylyl cyclase cascade, which is the predominant signaling pathway for growth and function of the thyrocyte (see Fig. 35-2). It is, however, also coupled to Gq and activates the inositol phosphate pathway. Mutations in several G protein coupled seven transmembrane receptors confer constitutive activation to these receptors, a mechanism characterizing an important pathophysiological entity. For example, they have been found in rhodopsin as a cause of retinitis pigmentosa, in the LH/CG receptor in male limited precocious puberty, and in the TSH receptor they cause hyperthyroidism. Point mutations in the Ca2+-sensing receptor lead to autosomal-dominant hypocalcemia with hypercalcuric hypocalcemia, and in the parathyroid hormone receptor to metaphyseal chondrodysplasia. ACTIVATING SOMATIC TSH RECEPTOR MUTATIONS IN TOXIC ADENOMAS AND THYROID CARCINOMAS Somatic mutations in the TSH receptor are the main molecular

345

cause of toxic adenomas. Functionally, these mutations increase basal cAMP levels; some of the mutants also activate the phospholipase C cascade. In contrast to activating mutations in other seven transmembrane receptors, there is a striking diversity in the affected residues that are scattered over almost the entire transmembrane domain of the TSH receptor. Activating mutations in the TSH receptor have also been found in a small number of thyroid carcinomas with the unusual finding of increased hormone secretion. Remarkably, the mutations found in these tumors activate the cAMP cascade as well as the inositol triphosphate pathway. This supports the concept that concomitant activation of these two signaling cascades may promote transformation. ACTIVATING GERMLINE TSH RECEPTOR MUTATIONS IN FAMILIAL NONAUTOIMMUNE HYPERTHYROIDISM Activating mutations occurring in the germline give rise to familial nonautoimmune hyperthyroidism. Gain of function mutations are by definition dominant and one mutated allele is thus sufficient to result in disease. The subsequent activation of the adenylyl cyclase pathway increases function and growth of thyroid follicular cells resulting in hyperplasia and hyperthyroidism. The typical signs associated with autoimmune hyperthyroidism, i.e., exophthalmos, myxedema, stimulating autoantibodies, and lymphocytic infiltration of the thyroid gland are absent. Several families with nonautoimmune familial hyperthyroidism and documented TSH receptor mutation have been reported. The onset of hyperthyroidism may vary in carriers of the same mutation in a given kindred, suggesting that other factors such as genetic background and or iodine intake are modulating the phenotypic expression of the activated receptor. ACTIVATING GERMLINE TSH RECEPTOR MUTATIONS IN NEONATAL NONAUTOIMMUNE HYPERTHYROIDISM Congenital hyperthyroidism is usually caused by transplacental transfer of maternal antibodies in offspring of a mother with autoimmune thyroid disease. In this instance, the disease is transient and resolves within several weeks to months on clearance of the antibodies. However, a few cases show a persistent course with severe hyperthyroidism. De novo germline mutations in the TSH receptor have been found as cause of this unusual form of hyperthyroidism. The recognition of this entity of hyperthyroidism has clinical implications. In certain patients a more aggressive therapeutic approach (surgery, ablative radiotherapy) is indicated and, in families with nonautoimmune hyperthyroidism, molecular diagnostics allow an early diagnosis and treatment. TSH RECEPTOR MUTATIONS CONFERRING HYPERSENSITIVITY TO HUMAN CHORIONIC GONADOTROPIN A remarkable form of familial gestational hyperthyroidism is caused by a mutant TSH receptor displaying hypersensitivity to normal levels of human chorionic gonadotropin (hCG). The proband and her mother had a history of two miscarriages that were accompanied by hyperemesis. Subsequently, she had two pregnancies that were complicated by hyperthyroidism, severe nausea and vomiting. Analysis of the TSH receptor gene in the proband and her mother revealed a heterozygous point mutation resulting in the substitution of K183R in the extracellular domain of the TSH receptor. Functional studies in cells transfected with the mutated receptor documented hypersensitivity to hCG. Although the wild-type TSH receptor reacts only minimally to high doses of hCG, the K183R mutant is hypersensitive to hCG, although it is still 1000 times less

346

SECTION IV / ENDOCRINOLOGY

responsive to hCG than the LH/CG receptor. The K183R mutant does not differ from the wild-type in terms of membrane expression, or basal and TSH stimulated cAMP accumulation. Aside from explaining the recurrent hyperthyroidism in these two patients, the K183R TSH receptor mutation is unique because the sensitivity is only increased for hCG but remains unaltered for the cognate ligand TSH. Reduction of ligand specificity by naturally occurring mutations is not limited to the TSH receptor. Heterozygous mutations in the FSH receptor permitting stimulation by hCG have been identified in women with pregnancy-associated ovarian hyperstimulation syndrome.

INSENSITIVITY TO TSH AND INACTIVATING MUTATIONS OF THE TSH RECEPTOR TSH insensitivity results in a decrease in function and thus reduced synthesis and secretion of thyroid hormones and hypoplasia of the gland. Resistance to TSH may be caused by various molecular mechanisms, among them inactivating mutations in the TSH receptor that are partially or completely inactivating. The mode of inheritance is recessive and affected individuals are homozygous or compound heterozygous for inactivating mutations. Among these patients the phenotype encompasses a wide spectrum ranging from isolated TSH elevation to severe hypothyroidism, and there is a clear correlation between genotype and phenotype. The first human case with TSH resistance resulting from a defect in the TSH receptor was documented in three sisters, offspring of unrelated parents, who were found to have normal peripheral thyroid hormone but high TSH levels, a constellation referred to as euthyroid hyperthyrotropinemia. Both parents only showed discrete TSH elevations. None of the family members had clinical signs of hypothyroidism. The three affected siblings were found to be compound heterozygous for mutations in the extracellular TSHbinding domain of the receptor (P162A and I167N). In vitro studies documented that the I167N mutation had almost no biological activity, whereas P162A displayed reduced activity. Euthyroid hyperthyrotropinemia resulting from mutations in the TSH receptor gene has subsequently been reported in several other families. More pronounced or complete inactivation of both TSH receptor alleles leads to mild or severe congenital hypothyroidism. Because of absent uptake of the radioisotope, scintigraphic studies often do not reveal any thyroid tissue. However, ultrasound of the neck reveals a normally located hypoplastic gland. Intriguingly, many of these patients have normal or elevated TG levels. A loss of function mutation in the TSH receptor gene as a cause of TSH resistance was also discovered in the thoroughly studied hypothyroid hyt/hyt mouse. The phenotype of this inbred mouse strain is defined by congenital hypothyroidism, retarded growth, mild anemia, hearing loss, and infertility. The hypoplastic thyroids of these mutant mice are located in the proper position. Histologically, the thyroid follicular cells are developed, but incompletely differentiated, and the epithelial cells are not organized into structures recognizable as follicles. The hyt/hyt mouse and patients with TSH-resistant congenital hypothyroidism with correctly located hypoplastic glands confirm that development and migration of the thyroid is independent of TSH stimulation. This is consistent with the observation that the genes for TPO, TG, and the TSH receptor are only expressed once the gland has reached its pretracheal location. Although early events of thyroid development are not dependent on TSH and its signaling pathway, this cascade is essential for complete differentiation, growth and function of thyroid follicular cells. These findings have

been corroborated in the TSH receptor knock-out mouse, or mice overexpressing a dominant negative cAMP response element binding protein (CREB) in the thyroid. These studies exemplify how careful phenotyping of several naturally occurring mutants and genetically modified mice provide fundamental insights into the molecular mechanisms controlling organogenesis, differentiation and function in vivo. In other patients with sporadic or familial resistance to TSH, the TSH receptor gene was normal indicating locus heterogeneity owing to defects in other genes. Obvious candidate genes include genes encoding elements of the TSH-dependent signaling cascades or regulators of thyroid development and gene expression.

ACTIVATING AND INACTIVATING MUTATIONS IN GSA Analogous to the mutations in the TSH receptor, gain of function mutations in the GNAS1 gene encoding the α subunit of the stimulatory G protein (gsp mutations) lead to a constitutive activation of the cAMP pathway and subsequently to an increase of function and growth of cell types like the pituitary somatotropes or thyroid follicular cells. In the thyroid, somatic gsp mutations have been found with variable frequencies in nontoxic and toxic adenomas, as well as in differentiated thyroid carcinomas. The most commonly affected amino acids are arginine 201 and glutamine 227. The ensuing substitutions (R201C or R201H; Q227R or Q227K) impair the hydrolysis of GTP to GDP, resulting in an ongoing activation of adenylyl cyclase. The same molecular defect is found as somatic mutation in 35–40% of somatotrope tumors in acromegalic patients. Sporadic mutations in Gsα that occur early in development lead to the McCune-Albright syndrome. These patients are mosaic for the mutation and the clinical phenotype varies depending on its tissue distribution. The clinical manifestations include ovarian cysts that secrete sex steroids and cause precocious puberty, polyostotic fibrous dysplasia, café au lait skin pigmentation, GH-secreting pituitary adenomas, and hypersecreting autonomous thyroid nodules. In pseudohypoparathyroidism type Ia, the affected subjects show resistance not only to parathyroid hormone, but also to TSH and gonadotropins, in combination with the features of Albright’s hereditary osteodystrophy. The molecular cause is a reduction of expression of the stimulatory Gsα subunit by a variety of loss of function mutations. The disorder affects many cell types with the cardinal signs being stunted growth, obesity, skeletal abnormalities, and hypogonadism. At the level of the thyroid, it is typically associated with mild hypothyroidism. The GNAS1 gene is imprinted in a tissue-specific manner. Most tissues express both alleles, but the paternal allele is imprinted and thus inactivated in tissues displaying hormone resistance. In pseudohypoparathyroidism type Ia, the mutation disrupts the active maternal allele that subsequently results in a decrease in signaling and partial hormone resistance to the trophic hormone in tissues such as the proximal renal tubule, the thyroid and the gonads.

DEFECTIVE THYROID HORMONE SYNTHESIS The major steps involved in thyroid hormone synthesis are summarized in Fig. 35-2. After active transport of iodide into the thyroid follicular cell by the NIS, iodide is brought to the apical pole of the cells oriented toward the follicular lumen. At the apical membrane, pendrin, probably in conjunction with one or several other unidentified channels, is involved in transport of iodide into the follicular lumen. TG is secreted by exocytosis into the follicu-

CHAPTER 35 / THYROID DISORDERS

lar lumen. TPO, localized in the apical membrane, oxidizes iodide, and subsequently iodinates tyrosyl residues of the intrafollicular TG (organification or iodination) in the presence of hydrogen peroxide. Two elements of the NADPH-dependent H2O2-generating, system, THOX1, and THOX2, have been been identified. The iodotyrosines, mono- and diiodotyrosyl (MIT, DIT), are coupled to T4 or T3, a reaction that is also catalyzed by TPO (coupling). TG carrying T4, T3, and iodotyrosines is internalized into the follicular cell by fluid phase or receptor-mediated endocytosis and digested in lysosomes. Although the thyronines T4 and T3 are released into the bloodstream, MIT, and DIT are deiodinated by dehalogenase in the cell and the released iodide is recycled. A dehalogenase referred to as DEHAL1 has been cloned very recently. In patients with congenital hypothyroidism, defects have been identified at all major steps involved in hormonogenesis (Table 35-3). DEFECTS OF IODIDE TRANSPORT Normal iodide uptake at the basolateral membrane by the perchlorate-sensitive NIS is a rate-limiting step in thyroid hormone synthesis (see Fig. 35-2). Following the cloning of the NIS gene, several homozygous, or compound heterozygous mutations have been identified in individuals with hypothyroidism with iodine trapping defects. Most patients with iodide trapping defects have a diffuse or nodular goiter, little or no uptake of radioiodide, and a decreased saliva/serum radioiodine ratio. In children, the thyroid might be initially of normal size and often enlarges later in life. The precise molecular mechanisms by which NIS mutations directly cause iodide transport defects have been identified in a subset of cases. The T354P mutation, an alteration found in several patients from Japan, causes NIS to lose its functional ability to transport iodide. In contrast, the functional defects of two other NIS mutations (Q267E, S515X) are the consequence of defective cellular trafficking and failure of the mutant proteins to reach the plasma membrane. THYROPEROXIDASE TPO is a glycosylated hemoprotein that catalyzes several essential reactions of thyroid hormone synthesis: oxidation of iodide, the iodination of tyrosine residues in TG and the coupling of iodinated tyrosines to generate T4 and T3 (see Fig. 35-2). TPO is the protein historically referred to as microsomal antigen in autoimmune thyroid disease. It is anchored in the membrane and has its catalytic site in the follicular lumen. The enzyme is closely related to myeloperoxidase and it is thought that they share a common ancestor; their chromosomal localizations are, however, distinct. TPO defects are among the most frequent causes of inborn abnormalities of thyroid hormone synthesis. Because of the defective organification of iodide, these patients typically have a significant discharge of radioiodine after the administration of perchlorate. Mutations in the TPO gene have been reported in numerous families with a partial or total iodide organification defect. The overall incidence of congenital hypothyroidism is on average 1/3000, total iodide organification defect is thought to occur in approximately 1/66,000 of all infants with congenital hypothyroidism and almost all of these patients have homozygous or compound heterozygous mutations in the TPO gene. PENDRED SYNDROME Pendred syndrome is an autosomalrecessive disorder characterized by sensorineural deafness, goiter, and a positive perchlorate test. This disorder represents one of the most common forms of syndromic deafness, with an incidence estimated at 7.5–10/100,000 individuals. Although the classic presentation of the syndrome consists of the triad of deafness, goiter, and partial organification defect, the phenotypic expression of these components is

347

highly variable among families and even within the same family. Sensorineural hearing loss, in most instances profound prelingual deafness, is the hallmark of Pendred syndrome. More rarely, the hearing impairment manifests itself later in life as a progressive hearing loss. High resolution MRI of the inner ear reveals malformations of the vestibular aqueduct, endolymphatic duct, and endolymphatic sac in nearly 100% of individuals with a clinical diagnosis of Pendred syndrome. Goiter is the most variable component of the disorder, with some individuals developing very large goiters, whereas others present with minimal to no enlargement. Although many patients with Pendred syndrome are euthyroid, others have subclinical or overt hypothyroidism. Pendred syndrome is caused by mutations in the PDS gene, now officially designated SLC26A4. It encodes pendrin, a member of the solute carrier family 26A, which contains several anion transporters and the motor protein prestin. Pendrin is predominantly expressed in the thyroid, inner ear, and the kidney. In thyroid follicular cells, pendrin is inserted into the apical membrane and functional studies suggest that it is involved in apical iodide efflux from thyrocytes. These observations are consistent with the clinical phenotype, which is characterized by impaired iodide organification. In the kidney, pendrin is thought to act as a chloride-bicarbonate exchanger in β-intercalated cells of the cortical collecting duct, a subpopulation of cells that mediate bicarbonate secretion. The critical role of pendrin in the inner ear has been corroborated by targeted disruption of the Pds gene in mice. Pds–/– mice develop early onset deafness and exhibit signs of vestibular dysfunction. In line with the enlargement of the endolymphatic system observed in human patients, analysis of the inner ear in these mice reveals dilated endolymphatic ducts and sacs beyond E15, presumably as a consequence of defects in anion and fluid transport. During the last few years, more than 95 PDS gene mutations have been described, indicating marked allelic heterogeneity. The majority of PDS mutations are missense mutations and some of these mutants appear to be retained in the endoplasmic reticulum. A smaller number of mutations result in premature truncations or in alterations of splice donor or acceptor sites. Individuals with Pendred syndrome from consanguineous families are homozygous for PDS mutations, whereas sporadic cases typically harbor compound heterozygous mutations. Mutations in the PDS gene are not only found in patients with classic Pendred Syndrome, but also in individuals afflicted with familial enlarged vestibular aqueduct. The diagnosis of Pendred syndrome based on the clinical findings of deafness and goiter is not sufficient; it requires confirmation by imaging studies of the inner ear or/and molecular analysis of the PDS gene. HYDROGEN PEROXIDE GENERATION H2O2 is an essential factor in the iodination and coupling reactions. Although the ability of follicular cells to produce H2O2 has been known for more than three decades, the enzyme system remains only partially characterized. Two NADPH oxidases, THOX1 and THOX2 (also called LNOX or DUOX) that are part of this system have been cloned. Structurally, these proteins contain seven putative transmembrane domains, four NADPH binding sites, 1 FAD binding site, and in line with the predicted regulation by calcium, an everted finger motif. Heterozygous loss of function mutations in the THOX2 gene result in mild transient congenital hypothyroidism. Biallelic THOX2 mutations are associated with a severe phenotype and confirm that H2O2 is essential for iodide organification. There are no reported mutations in THOX1. Deficient H2O2 generation has been proposed to explain the phenotype in a few sporadic patients with euthyroid goiter and

348

SECTION IV / ENDOCRINOLOGY

abnormal iodide organification, and a family with two siblings presenting with hypothyroidism, goiter, and an iodide organification defect. The molecular defect in these patients remains unknown. THYROGLOBULIN TG is produced by thyroid follicular cells and secreted into the follicular lumen. Some of its tyrosine residues are iodinated by TPO (organification/iodination), and the tyrosines MIT and DIT are subsequently coupled to form T3 and T4 (coupling). TG is therefore considered to be a thyroid hormone precursor. Besides its importance for hormone synthesis, it allows storage of iodine and thyroid hormone and thus to adapt to scarce iodine supply. The monomer of TG is made up of a 19-amino acid signal peptide followed by 2749 residues containing 66 tyrosines. TG contains an average number of tyrosine residues, altogether 67, but only a minority of these residues localized in the carboxy- and aminoterminus are hormonogenic sites. Complete hydrolysis of iodinated TG yields only two to four molecules of the iodothyroxines T4 and T3. The mature protein is formed by two units in noncovalent linkage (19 S TG). Of the total weight, 10% are formed by carbohydrates, and glycosylation plays an important role in the structure of the protein. The TG monomer contains 20 glycosylation sites and extensive microheterogeneity has been reported in their use. The primary structure of the TG protein contains three regions with repetitive sequences with internal homology, and the carboxy-terminal part shares remarkable homology with acetylcholinesterase. This structure suggests the possibility of a convergent origin of the TG gene from two different ancestral DNA sequences. The human TG gene is large, spans approximately 270 kb and contains 48 exons. The transcription of the TG gene is controlled by transcription factors such as thyroid transcription factor (TTF)1 (NKX2.1), TTF-2 (FOXE1), and PAX 8. The full-length 8.5 kB mRNA sequence shows a 41-nucleotide 5′-untranslated segment preceding an open reading frame of 8307 bases and a 3′-untranslated segment ranging from 101 to 120 bp. There are numerous alternatively spliced RNA transcripts and the gene contains multiple polymorphisms, some of them silent, others with impact on the primary amino acid sequence. Defects of TG synthesis or secretion have been studied in several animal strains and human patients. TG gene defects are inherited in an autosomal-recessive manner. The phenotype is typically characterized by goitrous enlargement of the thyroid. The metabolic status is variable and, depending on the severity of the defect, patients are hypothyroid, subclinically hypothyroid, or euthyroid. Unless treated with levothyroxine, goiters are often remarkably large and display continuous growth. Symptoms can result from compression of adjacent neck structures. The radioiodine uptake is elevated indicating an activation of the iodine concentration mechanism, because of chronic stimulation of TSH. In patients evaluated with a perchlorate discharge test, there is no increased release of radioiodine after administration of the competitor, indicating that the organification process itself is not affected. Serum TG levels can vary from low to low normal, and the presence of an abnormal TG level in a goitrous individual may suggest a defective TG synthesis. An abnormal TG synthesis may also be suggested by the presence of abnormal iodoproteins in the serum. Because there is no normal intrathyroidal TG, albumin as well as other proteins are iodinated, generating iodotyrosines and iodohistidines. This is, however, an unspecific sign and also found in endemic and sporadic goiters and thyroids affected by autoimmune Hashimoto’s thyroiditis. Increased secretion of low-molecular weight iodinated material (>5%) in the

urine may also be helpful in establishing the diagnosis. Histological analysis often demonstrates scarce colloid, large follicular lumina, and cuboidal epithelial cells. In instances withimpaired TG export, TG-immunopositivity is found predominantly inside the cytoplasm. Although it is possible to diagnose TG defects at the molecular level, this is not a trivial task considering the large size of the gene; thus the data on molecular alterations are still relatively scarce. Further studies of patients with TG abnormalities may reveal a more detailed understanding of the structure-function relationship of TG. The first models with abnormal TG to be studied in detail at the molecular level were the Afrikaander cattle and the Dutch goat. The Afrikaander cattle is characterized by large goiters and an euthyroid metabolic status. A recessive point mutation in exon 9 leads to a premature stop. Interestingly, alternative splicing allows rescuing the transcription to some degree by producing a misspliced 7.3-kb message missing exon 9, a mechanism referred to as exon skipping. The original reading frame is maintained in this transcript and it is translated into a functional protein missing the part encoded by exon 9. Both transcripts are, however, only present at low levels. The Dutch goat is goitrous as well and, provided that iodine intake is high, euthyroidism can be maintained. A nonsense mutation (Y296X) in exon 8 results in a truncated protein. The fact that the animal may remain euthyroid despite this truncation is indirect evidence that the amino-terminal part contains a major hormonogenic site. Molecular analysis of several TG point mutations found in patients with congenital hypothyroidism and in the cog/cog mouse, which all present with goiters, reveal that at least some of these alterations result in a secretory defect and thus an endoplasmic reticulum storage disease. In contrast to these TG defects associated with goiter development, the recessive dwarf rdw/rdw rat displays a nongoitrous form of congenital primary hypothyroidism caused by a Tg gene mutation. The identification of a mutation in the Tg gene as a cause of nongoitrous hypothyroidism in the rdw/rdw rat challenges the previously held generalization that nongoitrous congenital hypothyroidism is caused by thyroid dysgenesis or defects in TSH-signaling. It has been proposed that TG mutations may be associated with nonendemic simple goiter. These results await, however, further confirmation; the TG gene contains multiple polymorphisms, and in the absence of functional data it remains unclear whether these reported alterations are causally involved in the abnormal phenotype. DEHALOGENASE DEFECT After entering the follicular cell, TG is hydrolyzed and T4 and T3 are secreted into the blood (see Fig. 35-2). The iodotyrosines, MIT and DIT, which are much more abundant in the TG molecule than in T4 and T3, are deiodinated by an intrathyroidal dehalogenase and recycled for hormone synthesis. An intrathyroidal dehalogenase referred to as DEHAL1, which may exert this function, has been cloned. Very recently, mutations in DEHAL1 have been reported in patients with a dehalogenase defect. In case of a defective dehalogenase system, MIT and DIT leak into the circulation and are excreted in the urine. This leads, especially if iodine is scarce, to a severe iodine loss and thus to hypothyroidism and goiter. Very recently, mutations in DEHAL1 have been reported in patients with a dehalogenase defect. Clinically, patients with a deiodinase defect present with congenital hypothyroidism and a goitrous gland. The diagnosis is established by administration of radiolabeled DIT. Normally, DIT will be deiodinated, whereas in the case of a defective dehalogenase, the majority will be secreted unaltered as DIT in the urine.

CHAPTER 35 / THYROID DISORDERS

Furthermore, administration of iodide in sufficient amounts to compensate for the increased loss will reestablish a euthyroid metabolic state. The disorder is inherited in an autosomal-recessive fashion. Although only homozygotes are clinically affected, biochemical testing in heterozygotes demonstrates an increased secretion of labeled DIT in the urine. The clinical and biochemical phenotype of several kindreds have been studied in detail. The classic report traces the family history of an inbred kindred originating from the marriage of first cousins through 160 yr.

DEVELOPMENTAL THYROID DEFECTS ASSOCIATED WITH MUTATIONS IN TRANSCRIPTION FACTORS In approximately 85% of all affected infants, congenital hypothyroidism is sporadic and associated with developmental defects referred to as thyroid dysgenesis. They include thyroid (hemi)agenesis, ectopic tissue, and thyroid hypoplasia. Molecular defects only explain a minority of cases of thyroid dysgenesis (Table 35-4). It is likely that a further subset of patients with thyroid dysgenesis have defects in other transacting proteins or elements of the signaling pathways controlling growth and function. In other instances, thyroid dysgenesis might be a polygenic disease or have a multifactorial basis. With further characterization of the molecular basis of congenital hypothyroidism, genetic testing, and counseling may become increasingly important in the future.

PAX8 PAX8 is a paired domain thyroid-specific transcription factor responsible for thyroid development and for TG and TPO gene expression. PAX8 binds predominantly to the TPO promoter and with less affinity to the TG promoter. Heterozygous mutations in PAX8 have been documented and characterized in sporadic and familial patients with thyroid hypoplasia or ectopy. In humans, the biochemical and morphological phenotype may vary among patients with the same PAX8 mutation. Underlying mechanisms may include incomplete penetrance, a phenomenon associated with mutations in other PAX genes. Alternatively, the phenotypic expression might be modulated by modifier genes. It is unclear why mutation of a single PAX8 allele is sufficient to result in congenital hypothyroidism in humans, a finding that contrasts with the observation that mice heterozygous for a disrupted Pax8 gene do not display a pathological phenotype. Proposed mechanisms explaining that one mutated allele is sufficient to cause disease in humans include haploinsufficiency, allele-specific expression in the thyroid and, less likely, a dominant negative effect. THYROID TRANSCRIPTION FACTOR-1 (TTF-1, NKX2.1) TTF-1 (NKX2.1, TITF-1, or thyroid specific enhancer-binding protein T/ebp), is a homeobox domain transcription factor of the NKX2 family involved in the development of the gland and in transcriptional control of the TG, TPO, and TSH receptor genes. It is expressed in the lung, forebrain, pituitary gland, and the thyroid. In the lung, TTF1 activates transcription of surfactant protein B. Mice with targeted disruption of both TTF1 alleles survive throughout gestation, but die at birth from respiratory failure. The lung is severely hypoplastic and consists of a sac-like structure without bronchioli, alveoli, or lung parenchyma. Both the thyroid gland and pituitary gland are completely absent, and the hypothalamus is severely malformed. The observation of a newborn with severe respiratory distress, a normally located thyroid gland, elevated TSH levels and a het-

349

erozygous deletion on chromosome 14q13 encompassing the TTF1 locus suggested that haploinsufficiency for TTF1 could be associated with impaired lung maturation and thyroid function. Haploinsufficiency, i.e., inability of a single functional allele to maintain normal function, is a frequently observed pathogenetic mechanism associated with mutations in transcription factors. The detection of a similar heterozygous deletion of chromosome 14q12-13.3 in two female siblings with congenital thyroid dysfunction and recurrent acute respiratory distress gave further support to this concept. A few additional patients with hyperthyrotropinemia, neonatal respiratory distress, and ataxia associated with missense or frameshift mutations, or chromosomal deletions of the TTF-1 gene have been reported. The TSH levels were only mildly elevated and the thyroid was normal in size and position. The hallmark of this phenotype is the neurological deficit, which includes ataxia or choreoathetosis, truncal apraxia, and mental retardation, in combination with neonatal respiratory distress. Heterozygous TTF-1 mutations have also been identified as the molecular cause of hereditary chorea. THYROID TRANSCRIPTION FACTOR-2 (TTF-2, FOXE1) TTF-2, official designation FOXE1, is a forkhead/winged-helix domain transcription factor activating the TG and the TPO gene promoters. Homozygosity for recessive mutations in TTF-2 results in a syndromic form of thyroid dysgenesis with the eponym BamforthLazarus syndrome. This phenotype, first described in two brothers from a consanguineous family, includes thyroid agenesis, cleft palate, choanal atresia, bifid epiglottis, and spiky hair. Mice homozygous for a disrupted Ttf2 gene die shortly after birth, and are profoundly hypothyroid. They exhibit either small lingual thyroid remnants or have complete thyroid agenesis, findings that support the important role of TTF-2 in thyroid development, and they also have cleft palates.

SERUM BINDING PROTEIN DEFECTS Thyroid hormones circulate bound to plasma proteins, the three major proteins being TBG, TTR (formerly referred to as T4-binding prealbumin) and albumin. To a minor degree, α and β lipoproteins bind T4. Under physiological conditions, only 0.03% of T4 and 0.3% of T3 circulate as free hormone. Abnormalities in transport proteins lead to a decrease or increase in total T4 or T3 levels; free hormone levels are within the normal range and patients are clinically euthyroid (Table 35-5). Failure to recognize these entities results in inappropriate treatment aimed at normalizing the thyroid hormone levels. TTR variants are of clinical importance because they are associated with various forms of amyloidosis. THYROXIN-BINDING GLOBULIN TBG is an acidic glycoprotein with a single binding site for T4 or T3. The TBG concentration is low (1–2 mg/dL), but the protein has a high affinity for thyroid hormones (T4 > T3) and it carries approx 75–80% of the bound thyroid hormones. The mature protein contains 395 amino acids with four heterosaccharide chains with 5–9 sialic acids. Like corticosteroid-binding globulin it shows a high homology with the proteases α1-antichymotrypsin and α1-antitrypsin and it is a member of the serine protease inhibitor (serpin) family of proteins, although it is not a protease inhibitor. The carbohydrates play a minor role in T4 binding, but loss of carbohydrates decreases its stability and elevates the hepatic clearance rate. An increase in sialylation, for example, induced by estrogens, lowers the hepatic clearance and thus increases the TBG levels. Multiple other drugs increase or

350

SECTION IV / ENDOCRINOLOGY

decrease TBG concentrations either by alteration of the synthesis rate and or the degree of sialylation. The single copy gene encoding TBG is localized on the X-chromosome (Xq11-q23). TBG abnormalities are classified according to the levels of TBG into complete or partial deficiencies or TBG excess. Besides their biochemical classification, several TBG variants have been characterized at the molecular level. Complete TBG-deficiency is defined as absence of TBG in the serum of hemizygous (XY) individuals. The heterozygote females in these kindreds have TBG levels of about half the normal amount because random X-inactivation results on average in a 50% reduction of the protein. The prevalence of complete deficiency has been estimated at 1/15,000. Partial deficiency is the most common form of TBG deficiency. In white and mixed populations it is found with a frequency of 1/4000, roughly 50% of Australian Aborigines have an abnormal TBG (TBG-A), and a TBG migrating slowly on electrophoresis (TBG-S) is found in African and Pacific Islands populations. Some of these TBG-variants have a reduced affinity for T4 (TBG-A, TBG San Diego). In some variants, an accelerated rate of degradation is responsible for their low serum concentration. The identified mutations in the TBG gene that result in complete or partial deficiency consist of point mutations, deletions, frameshift mutations, and splice site mutations. TBG excess is not as frequent as TBG deficiency with an incidence of approximately 1 in 25,000 births. TBG and T4 levels are elevated three- to fivefold in hemizygotes and two- to threefold in heterozygote females. The TBG increase is explained by overexpression of the protein because of gene amplification. TRANSTHYRETIN TTR is a homotetramer formed of subunits containing 127 amino acids and two T4 binding sites. Negative cooperativity allows only one site to be occupied at a given time. TTR has a substantially lower affinity for T4, but it is present in higher amounts than TBG (approx 25 mg/dL). TTR also binds retinol-binding protein and takes part in the transport of vitamin A. Binding of retinol does not influence T4 transport. The TTR gene contains four exons and is localized on chromosome 18. TTR is predominantly synthesized in the liver, but also in the plexus chorioideus. Although the exact role of TTR in the brain is not known, it has been postulated that T4 binds to TTR in the epithelial cells of the plexus chorioideus, is then secreted into the cerebrospinal fluid and subsequently distributed in the brain. More than 80 diseasecausing mutations in TTR have been reported. Their impact on T4 affinity varies and may result in unchanged, increased or reduced affinity. Many of the point mutations have been related to distinct forms of amyloidosis. The inheritance is autosomal-dominant in most instances. Although the clinical manifestations vary, most have polyneuropathy, thus the eponym familial amyloidotic polyneuropathy, and amyloid depositions in the heart. ALBUMIN Albumin is a monomer of 69 kDa associated with the transport of various endogenous and exogenous hydrophobic compounds. Albumin has a relatively low affinity for T4 and T3, but because of its high concentration (approx 3.5 g/dL) it binds up to 10% of T4 and 30% of T3. Familial dysalbuminemic hyperthyroxinemia is the most common cause of euthyroid hyperthyroxinemia. Familial dysalbuminemic hyperthyroxinemia is characterized by increased binding of T4 and thus elevated total T4, but normal free T4 levels. It is inherited in an autosomal-dominant manner. Studies of several families have led to the identification of two independently recurring point mutations in the albumin gene that lead to substitutions

of arginine at position 218 (R218H, R218P). Crystallographic analyses revealed that the two mutations result in conformational changes that favor binding of T4 on one of the four T4-binding sites on albumin. In a family presenting with high serum total T3, but not T4, an albumin mutation L66P was identified. The L66P albumin leads thus to familial dysalbuminemic hypertriiodothyroninemia with an affinity for T3 that is approximately 40 times higher than wildtype albumin. In contrast, the affinity for T4 is barely changed.

UPTAKE OF THYROID HORMONE INTO CELLS Cellular uptake of thyroid hormones appears to be mediated by amino acid channels. Point mutations and deletions in the X-chromosomal MCT8 gene have been identified in several male patients presenting with elevated T3 and TSH levels and a remarkable neurological phenotype that includes severe mental retardation, spastic quadriplegia, rotary nystagmus, and impaired gaze, and hearing. In one instance, the disorder was found in several males of a consanguineous mating, in the other cases, it was sporatic. Heterozygous females have discrete thyroid hormone abnormalities, but no neurological alterations. The abnormal T3 elevation may be due to impaired uptake into cells such as neurons and resembles in many aspects to the phenotype in neurological cretinism. It remains unclear whether MCT8 transports other amino acid derivatives that could be involved in the development of the complex phenotype. The observations indicate, however, that the uptake of thyroid hormones is, at least in part, mediated by channels. Of note, studies have revealed that the Allan-Herndon-Dudley syndrome, a recognized cause of X-linked mental retardation with a wide spectrum of neurological alterations, is caused by mutations in MCT8.

PERIPHERAL MONODEIODINATION In target tissues, T4 is metabolized into the more active compound T3 by intracellular 5′-monodeiodination or into the inactive metabolite rT3 by 5-monodeiodination (see Fig. 35-3). Because roughly 80% of T3 is generated by monodeiodination of T3, T4 is sometimes considered a prohormone. Monodeiodination of the outer and inner ring is catalyzed by three well-characterized deiodinases, which are unusual because they are selenoproteins. They contain the rare amino acid selenocysteine (Sec), which is encoded by UGA, a triplet that usually encodes a stop codon. The translation of the Sec codon requires specific stem loop sequences that are located in the 3′ untranslated region (SECIS element) of the mRNA. Type 1 iodothyronine deiodinase (DIO1) primarily converts T4 to T3 and is sensitive to propylthiouracil. In humans, it is expressed in many tissues, including liver, kidney, thyroid, and pituitary. DIO2 deiodinates exclusively the outer ring. Its expression is abundant in the human pituitary, brain, thyroid, and skeletal muscle; high levels are found in the brown adipose tissue of rodents in which it plays an important role in adaptive thermogenesis. DIO3 is the major enzyme inactivating T4 and T3 (see Fig. 35-3). In addition to its expression in tissues such as the central nervous system and skin, DIO3 is particularly abundant in the placenta and regulates circulating fetal thyroid hormone concentrations during gestation. Overexpression of DIO3 in infantile hemangiomas and other vascular tumors leads to consumptive hypothyroidism through inactivation of thyroid hormone at a rate that exceeds the maximal thyroid hormone synthesis.

CHAPTER 35 / THYROID DISORDERS

THYROID HORMONE ACTION Thyroid hormones exert their multiple cellular effects through nuclear TR, transcription factors that act by altering patterns of gene expression both as activators and repressors (see Fig. 35-4). The two TR genes, TRα and TRβ, cellular homologs of the viral erythroblastic leukemia oncogene v-erbA, were cloned based on their relationship to other members of the steroid receptor superfamily of nuclear receptors that share a characteristic modular domain structure with a central DNA binding domain and a carboxy-terminal ligand binding domain. In the TR, the carboxy-terminus of the receptor also contains nuclear localization signals, dimerization domains, and transactivation functions. The functional properties of the aminoterminal region of the receptor are less well characterized but seem also to be involved in transactivation. The two TRs, TRα and TRβ, are encoded by separate genes located on chromosomes 17 and 3, respectively. Although they bind thyroid hormones with high affinity, TRα and TRβ differ in their developmental patterns of expression, tissue distribution and the patterns of splicing to create additional isoforms. The TR can bind as monomer or with greater affinity as homodimers and heterodimers, particularly with retinoid X receptors to thyroid hormone response elements (TREs) in specific target genes. In most instances, unliganded TR represses transcription by recruiting corepressors, some of which have histone deacetylase activity (see Fig. 35-4C). Binding of T3 alters the conformation of the receptor resulting in release of the corepressor complex and recruitment of a coactivator complex that includes multiple histone acetyltransferases. The modification of histones is an important mechanism controlling chromatin structure and transcriptional events. Several TR-interacting coregulators act more directly on the basal transcriptional machinery suggesting that mechanisms independent of histone acetylation and deacetylation are also involved in TR action. Although there is considerable heterogeneity in the TREs of different target genes, most TREs contain two or more “half sites” that correspond to the minimal recognition motif for a receptor monomer. The consensus TRE half-site consists of the DNA sequence AGGTCA. These half-sites can be arranged as a palindrome, a direct repeat spaced by four nucleotides, or as an inverted repeat, also called lap (an eponym created by inversion of palindrome). Remarkably, TRs can bind to these TREs with different orientations as homodimers or heterodimers with cofactors, indicating flexibility in the determinants of the protein–protein interface. Receptor binding to these repeats is importantly influenced by neighboring nucleotides and by the spacing between the elements that provide one of the primary determinants of receptor specificity for different nuclear receptors like TRs, retinoid receptors and vitamin D receptors. Changes in gene transcription are reflected in alterations in mRNA levels followed by changes in protein biosynthesis. A large number of genes that respond to thyroid hormone have been identified. The crystal structure of the rat TRα1 ligand-binding domain bound to T3 has been solved and permits a better understanding of structure-function relationship. Knowledge about the structure may also permit the design of novel agonists and antagonists that may have differential effects on TRs and of use in conditions such as hyperthyroidism, lipid disorders, and obesity. Important insights into thyroid hormone action have been gained through systematic disruption of the TRα and TRβ• genes

351

and isoforms, independently or in combination. More than ten mutant strains have been characterized. These studies reveal that TRβ is essential for the development of the inner ear, color vision, and liver function. TRα, abundantly expressed in the heart, is essential for normal cardiac function. Moreover, it is important for maintaining basal energy metabolism and regulating enchondral bone formation. Evaluation of the combined TR knock-out indicates that hypothyroidism is associated with more significant abnormalities than receptor deficiency. This supports that the unliganded receptors have a repressive function of physiological relevance. Knock-in experiments have also been successfully used to model the syndrome of RTH and continue to reveal fundamental insights into TR-mediated gene regulation. For example, mice heterozygous for a knock-in of a dominant negative mutation into the TRα1 gene display a phenotype with marked visceral adiposity and insulin resistance, increased TSH levels and minimally elevated T4 and T3 levels suggesting that TRα plays an important role in regulating both lipogenesis and lipolysis by modulating adrenergic activity. Lastly, some thyroid hormone-mediated effects are rapid and not mediated by genomic actions. They involve effects on the cell membrane, mitochondria, and stimulation of signaling pathways. Of growing interest, these nongenomic effects await more thorough characterization.

RESISTANCE TO THYROID HORMONE RTH was first recognized in 1967 in two sibs of a consanguineous marriage presenting with goiter, high levels of proteinbound iodine, deaf-mutism, delayed bone age, and stippled epiphyses, but without signs of hyperthyroidism. Abnormalities in thyroid hormone itself or its transport into tissues were excluded, and it was thus postulated that resistance in peripheral tissues was the explanation for the absence of signs and symptoms of thyrotoxicosis. Following the cloning of the TRs, linkage analysis demonstrated that TRβ is tightly linked to RTH, whereas no association could be demonstrated to TRα. The ultimate proof that a defect in the receptor is the cause of RTH was provided in 1989 by the demonstration of mutations in the TRβ gene, a finding that has been confirmed in multiple reports. RTH is a rare disorder, but because of growing awareness of its existence more than 400 familial and sporadic patients have been studied. Biochemically, RTH is defined by elevated circulating levels of free thyroid hormones because of reduced target tissue responsiveness and normal, or elevated, levels of TSH. This “inappropriate” TSH elevation contrasts with the situation in hyperthyroidism, in which the pituitary secretion of TSH is suppressed. Patients with RTH typically present with goiter and signs of hypothyroidism in tissues expressing predominantly TRβ, but hyperthyroidism in organs with predominant TRα expression such as the heart. They can include short stature, delayed bone maturation, hyperactivity and learning disabilities, hearing defects and tachycardia. There is a striking clinical heterogeneity in patients with RTH, a phenomenon presumably caused by genetic background or nongenetic factors modulating thyroid hormone action. With the exception of the first studied kindred, a single sibship harboring a deletion of the entire coding sequence of the entire TRβ gene and a recessive pattern of inheritance, RTH is most commonly caused by heterozygous mutations of the TRβ gene. They can be inherited in an autosomal-dominant manner or occur as de novo mutations. The mutant receptors act in a dominant

352

SECTION IV / ENDOCRINOLOGY

negative fashion to block the activity of normal TRα and TRβ receptors. The mutations are clustered in three domains in the carboxy-terminal region of the receptor. Many mutations occur in CpG dinucleotide sequences, and consist in most cases of nucleotide substitutions that result in single amino acid substitutions. In a few cases, the mutations cause frameshifts, either altering the reading frame, or causing premature stop codons. No mutations have been found within the DNA binding domain or in the amino-terminus. In general, the mutations preserve some critical receptor functions such as dimerization and DNA binding, whereas inactivating other activities such as T3 binding and transcriptional activation. The dominant negative activity of mutant TRs involves several mechanisms. Mutant TRs, which have lost their transcriptional activity, can block wild-type TR from binding to TREs. Given the decreased or impaired T3 binding, this will favor interactions with corepressors (see Fig. 35-4C). Some mutations may also have altered dimerization properties that may disrupt interaction with coactivators. These naturally occurring mutations continue to provide important insights into the mechanisms of thyroid hormone action, molecular mechanisms of dominant negative activity, and structure-function relationship of the TRs, which is now facilitated by the available crystal structure for TRα. Several kindreds with a RTH phenotype without mutations in the TR and autosomal-dominant transmission of the disorder have been reported. It is likely that mutations in cofactors that are required for normal TR function are involved in the pathogenesis of RTH in these patients. Somatic mutations in the TRs have been identified in a subset of thyroid cancers, a TSH-secreting pituitary adenoma, renal cell cancers, and hepatomas. Their exact role in the development of these neoplasias is unknown.

AUTOIMMUNE DISORDERS Autoimmune thyroid disorders (AITD), Hashimoto thyroiditis and Graves’ disease, are by far the most common diseases affecting the thyroid gland. They are characterized by an immune response to thyroidal antigens, infiltration by T cells and production of antibodies. In Graves’ disease, thyroid-stimulating antibodies lead to activation of the TSH receptor and hyperthyroidism, a process involving mainly a Th2 cell response. In Hashimoto’s thyroiditis, a predominantly Th1-mediated chronic inflammation leads to progressive destruction and hypothyroidism. Both disorders are found more frequently in women than in men (4–10/1), a difference that is most commonly explained by influences of sex steroids on immunoregulatory mechanisms. The incidence of Graves’ disease in the general population has been estimated at 0.2–1%. Hashimoto’s thyroiditis has been reported to occur in up to 3–4.5%, and approx 15% of elderly women have thyroid autoantibodies although there is not necessarily a clinical correlate. Thought to be multifactorial, the AITD require a genetic predisposition in combination with environmental triggers. Epidemiological data support a strong genetic component in the development of AITD, further supported by twin studies. Graves’ disease occurs in approximately 3–9% of dizygotic and 30–60% of monozygotic twins. For Hashimoto’s thyroiditis, the concordance rate for monozygotic twins is approximately 40%, but very low for dizygotic twins. Of note, monozygotic twins are not identical in terms of their immune repertoire, which may explain that perfect concordance is absent. Genetic and molecular analyses have

allowed a partial understanding of the pathophysiological mechanisms underlying the two disorders, but their precise elucidation continues to form a major challenge. As in other autoimmune diseases, associations have been established between the disease and the presence of certain human leukocyte antigens (HLA) constellations. Graves’ disease is associated with HLA-DR3. However, most linkage studies have been negative. HLA correlations for Hashimoto’s thyroiditis differ from the ones reported for Graves’ disease and appear to include HLADR3, DR4, and DR5 haplotypes. Candidate gene analyses aiming at investigating associations between variants in genes such as the TSH receptor, TPO, and the T-cell receptor with AITD have been negative. In contrast, the TG and cytotoxic T-lymphocyte antigen, four genes appear to be important susceptibility genes for AITD. Whole genome screens using AITD families identified several loci or genes with evidence for linkage to AITD (GD-1 on chromosome 14; GD-2 on chromosome 20; HT-2 on chromosome 12; Tab-1 on chromosome 2; the TG gene on chromosome 8). It also became apparent that gene–gene interactions occur between some of these genes. The identified loci show, in part, linkage to Graves’ and Hashimoto’s disease suggesting that certain genes underlie the development of AITD, but that additional genetic and or nongenetic factors are required for the expression of the specific phenotype. ANTIBODIES AND AUTOIMMUNE THYROID DISEASE Because stimulating autoantibodies against the TSH receptor (TSAb) are the cause of Graves’ disease, their assessment can help establish the diagnosis. Most tests rely on measurement of the binding of these antibodies to the receptor but this test does not assess their bioactivity. Cloning of the TSH receptor led to the development of new bioassays that allow measurement of TSAb. They allow reliable distinction of TBAb; these blocking antibodies may also be present in other forms of AITD. For these new assay systems, the TSH receptor has been stably transfected into cell lines. Measurement of cAMP in response to a patient serum allows assessment of the presence and bioactivity of thyroid autoantibodies. Molecular biology established that the microsomal antigen in autoimmune thyroiditis is in fact TPO. The epitopes within the TPO enzyme, which are the targets of autoantibodies, are still a matter of debate, however. Recombinant TPO can now also be produced for the assay of autoantibodies.

THYROTOXIC HYPOKALEMIC PERIODIC PARALYSIS Periodic paralysis with hypokalemia can occur in patients with thyrotoxicosis. The clinical presentation is indistinguishable from familial hereditary hypokalemic paralysis without hyperthyroidism. Analysis of several genes encoding channels that are mutated in the familial form in patients with the thyrotoxic form led to the identification of a mutation in the KCNE3 potassium channel gene. No mutations could be found in other candidates such as the calcium channel CACN1AS or the sodium channel SCN4A.

THYROID CANCER Thyroid cancers are relatively infrequent neoplasms and account for approx 0.6–1.6% of all malignancies. The incidence of thyroid cancer is 1–10/100,000 in most countries. In the United States, this results in approximately 20,000 new cases and 1200 deaths per year. Thyroid cancer is about three times more frequent in females. Remarkably, thyroid cancer has shown the highest increase in

CHAPTER 35 / THYROID DISORDERS

353

Figure 35-5 Molecular alterations involved in the pathogenesis of benign and malignant thyroid tumors. (Redrawn and modified after Fagin JA.)

cancer incidence (+4.3/100,000) during the 1992–2002 survey period of the Surveillance, Epidemiology, and End Results of the National cancer institute (http://seer.cancer.gov/). Most thyroid malignancies are papillary thyroid carcinomas (PTC) or follicular thyroid carcinomas (FTC), both originating from follicular cells. PTC is by far the most common thyroid cancer (approximately 90%) in countries with sufficient iodine intake, FTC is comparatively uncommon. Their relative distribution is, however, variable and the incidence of FTC increases in regions with iodine deficiency. TG, only expressed in thyroid follicular cells, serves as an excellent tumor marker in patient follow-up. TG mRNA can be detected by reverse transcriptase PCR in serum samples and is being evaluated as a potential diagnostic tool for residual cancer. Besides the clinically relevant forms of PTC, occult PTC (50% of patients who have the Kenny–Caffey syndrome, an unusual syndrome characterized by short stature, osteosclerosis, basal ganglion calcifications, and ophthalmic defects. Studies indicate that the Kenny–Caffey syndrome is related to the Sanjad–Sakati syndrome, in which congenital hypoparathyroidism is associated with growth and mental retardation. Both of these autosomal-recessive disorders are linked to the same 2.6 cM region on chromosome 1q43-44, are allelic and are associated with defects in the TBCE gene, which encodes one of several chaperone proteins required for the proper folding of α-tubulin subunits and the formation of αβ-tubulin heterodimers. Cells from patients with these genetically related disorders show disturbances in subcellular organelles that require microtubules for membrane trafficking, such as the Golgi and late endosomal compartments. These findings provide evidence that the Sanjad–Sakati and Kenny–Caffey syndromes are chaperone diseases caused by a genetic defect in the tubulin assembly pathway, and establish a potential connection between tubulin physiology and the development of the parathyroid. Congenital hypoparathyroidism also occur as a feature of several generalized metabolic defects and in several mitochondrial neuromyopathies.

FAMILIAL ISOLATED HYPOPARATHYROIDISM Isolated hypoparathyroidism may be sporadic or familial, with inheritance of PTH deficiency by autosomal-dominant, autosomalrecessive, or X-linked modes of transmission. The age at onset covers a broad range (1 mo to 30 yr), but the condition is most commonly diagnosed during childhood. Isolated hypoparathyroidism has been associated with genetic defects that impair PTH synthesis (i.e., PTH gene defects) or secretion (i.e., CASR gene defects) as well as parathyroid gland development (e.g., GCMB gene defects). Defects in the PTH gene are an uncommon cause of hypoparathyroidism, with mutations reported in three families. One form of autosomal-dominant hypoparathyroidism has been attributed to a heterozygous mutation of the PTH gene consisting of a single base substitution (T → C) in exon 2. This mutation results in the substitution of arginine (CGT) for cysteine (TGT) in the leader sequence of preproPTH. The substitution of a charged amino acid in the midst of the hydrophobic core of the leader sequence inhibits processing of the mutant preproPTH molecule to proPTH by signal peptidase and is presumed to impair translocation of the mutant hormone and of the wild-type protein across the plasma membrane of the ER. Thus, this heterozygous mutation results in a dominant inhibitor phenotype that also prevents processing of the wild-type preproPTH molecule from the remaining normal PTH allele. This preproPTH mutation was the

CHAPTER 36 / DISORDERS OF THE PARATHYROID GLAND

first signal peptide mutation reported to cause human disease; signal peptide mutations have subsequently been found in other molecules such as preprovasopressin (causing diabetes insipidus) and Factor X (causing a coagulopathy). Mutations in the PTH gene are also the cause of autosomalrecessive hypoparathyroidism in two unrelated families. In one family, affected children were homozygous for a mutation in exon 2 that is predicted to disrupt normal processing of the preproPTH molecule. The mutant allele carries a T → C transition in the first base of codon 23 that results in the replacement of serine (TCG) by proline (CCG) at the –3 position of the signal peptide of preproPTH. This change is hypothesized to inhibit cleavage by signal peptidase at the normal position, and thereby lead to rapid degradation of the preproPTH protein in the rough ER. Affected patients who are homozygous for this allele present with symptomatic hypocalcemia within the first few weeks of life. In a second family, hypoparathyroidism occurred in members who were homozygous for a single base transversion (G → C) at the exon 2-intron 2 boundary. This mutation alters the invariant GT dinucleotide of the 5′ donor splice site that presumably affects annealing of the U1-snRNP recognition component of the nuclear RNA splicing enzyme. The use of a highly sensitive modification of reverse transcriptase-polymerase chain reaction allowed detection of very small amounts of preproPTH mRNA in cultured lymphoblasts from these patients, and revealed a PTH cDNA in affected subjects that was 90 bp shorter than the corresponding wild-type form. Nucleotide sequence analysis of the shortened cDNA revealed that exon 1 had been spliced to exon 3 in the mutant PTH mRNA, a process that resulted in the deletion of exon 2 from the mature transcript (i.e., exon skipping). The loss of exon 2 would eliminate both the initiation codon and the signal peptide sequence from the aberrant preproPTH mRNA, presumably explaining the molecular basis for autosomal-recessive hypoparathyroidism in this family. Isolated hypoparathyroidism can also arise because of failure of the parathyroid glands to develop properly during embryogenesis. Homozygous mutations in the GCMB gene lead to congenital hypoparathyroidism because loss of this transcription factor results in failure of the parathyroid glands to develop (i.e., parathyroid agenesis) during embryogenesis. Mice and humans who are deficient in the transcription factor gcm2 (glial cell missing, 2) develop hypoparathyroidism but lack defects in other tissues derived from the neural crest or pharyngeal pouches. These data implicate the GCMB gene, located at chromosome 6p23-24, as a cause of autosomalrecessive parathyroid aplasia in humans, but the prevalence of GCMB gene mutations in IH remains unknown. An X-linkedrecessive form of IH (OMIM 307700) has also been reported in two related multigeneration kindreds from Missouri, USA, and a deletion–insertion [del(X)(q27.1) inv ins (X;2)(q27.1;p25.3)] that could result in a position effect on SOX3 expression has been identified in affected subjects. Another mechanism underlying familial isolated hypoparathyroidism has been described. Heterozygous mutations in the gene encoding the CaR (CASR) that result in a gain of function have been identified in many subjects with autosomal-dominant hypocalcemia, a syndrome associated with low serum levels of PTH and relative hypercalciuria. In other cases, linkage of hypocalcemia to the chromosomal locus for the CaR (3q21-24) has provided indirect evidence for the involvement of this gene with familial hypoparathyroidism. Expression of mutant CaRs in

361

oocytes and mammalian cells leads to constitutive activation of the inositol phosphate signal transduction pathway. The host parathyroid cell, therefore, behaves as though it were being exposed to higher than normal concentration of extracellular calcium and appropriately fails to secrete PTH. Subsequent studies have identified similar activating mutations of the CaR gene in many patients with sporadic hypoparathyroidism. In both familial and sporadic cases, each affected propositus has demonstrated a unique mutation, suggesting that new mutations must sustain this disorder in the population. These results suggest that mutation of CaR gene may be the most common cause of genetic hypoparathyroidism. The CaR is expressed in the parathyroid gland and the kidney, in which it appears to play an important role in regulating calcium reabsorption. Thus, gain-of-function mutations in the CaR are likely to account for the increased calcium clearance and relative hypercalciuria noted in patients with autosomal-dominant hypocalcemia. These patients may therefore be at increased risk of nephrocalcinosis or nephrolithiasis. By contrast, loss of function mutations of the CaR in patients with familial hypocalciuric hypercalcemia (FHH) (see FHH and Neonatal Severe Hyperparathyroidism) are associated with decreased calcium clearance and relative hypocalciuria.

FHH AND NEONATAL SEVERE HYPERPARATHYROIDISM FHH, also called familial benign (hypocalciuric) hypercalcemia, is a disorder typically discovered incidentally on routine serum calcium screening. The hypercalcemia is usually mild (i.e., in the 10.5–12.0 mg/dL range), is lifelong, is generally not associated with the symptoms of hypercalcemia, and is associated with a reduction in the fractional urinary excretion of calcium. Historically, patients with FHH were confused with patients with typical sporadic parathyroid tumors, with the unfortunate consequence that affected patients underwent unnecessary partial parathyroidectomy, which would not alter the serum calcium, or equally unnecessary complete parathyroidectomy with the serious consequence of surgical hypoparathyroidism. Since the initial description of the syndrome in the 1970s, it has become clear that these patients manifest a form of biochemically defined primary hyperparathyroidism with slightly elevated or inappropriately normal serum PTH concentrations, that they have inappropriately efficient renal reabsorption of calcium, that the parathyroid glands have a defective ability to sense the hypercalcemia, and that multiple family members are typically involved, usually in an autosomal-dominant fashion. Inherited inactivating mutations in the extracellular CaR (encoded by the CASR gene) are responsible for the features of FHH in most affected kindreds (see Fig. 36-1). The responsible mutations generally involve either the extracellular domain (presumably adversely influencing calcium binding to the receptor) or one of the transmembrane loops (presumably interfering with conformational changes that in turn lead to the activation of signal transduction pathways). In general, affected patients have been heterozygous for the mutant calcium receptor allele. Notably, infants have been described who have a potentially life-threatening syndrome, characterized by severe hypercalcemia (sometimes in the 20 mg/dL range) and, in contrast to typical FHH, prominent parathyroid hypercellularity. This syndrome has been referred to as neonatal severe hyperparathyroidism (NSHPT), and may occur within typical FHH kindreds. NSHPT can result from a germline double dose of a mutant, inactivated calcium receptor gene, often

362

SECTION IV / ENDOCRINOLOGY

in the context of consanguineous parentage. Finally, germline heterozygosity for a calcium receptor gene bearing a specific missense mutation in the cytoplasmic tail was reported in familial parathyroid adenoma/hyperplasia with symptomatic hypercalcemia and hypercalciuria, suggesting a spectrum of possible phenotypic consequences of different germline CaR alterations. In support of the concept that inactive mutant forms of the calcium receptor are responsible for these syndromes, expression in Xenopus oocytes of calcium receptors containing the same mutations as those encountered in humans results in impaired ability to sense extracellular calcium. Also, an FHH-like phenotype is found in mice in which a single casr allele has been inactivated, and a more severe phenotype similar to NSHPT is seen in mice in which both casr alleles have been disrupted. FHH families have been defined whose responsible genetic locus maps to the CASR’s 3q locus but no mutations are apparent in the gene’s coding region. These findings suggest that mutations in the intronic or regulatory regions of the calcium receptor gene may explain additional kindreds. Mutations in other genes will undoubtedly be demonstrated in the future; for some families with FHH, the disease locus maps to locations distinct from 3q, namely 19p and 19q. It is presumed that the genes at these other loci encode other proteins that play a role in the calcium sensing machinery.

MOLECULAR GENETICS OF SPORADIC PARATHYROID GLAND NEOPLASIA Primary hyperparathyroidism is caused by excessive secretion of PTH, resulting in hypercalcemia. Patients with primary hyperparathyroidism have both an excessive parathyroid cell mass and a resetting of the set point by which PTH secretion is tightly coupled to the parathyroid cell’s ambient calcium level. In most (>80%) patients with primary hyperparathyroidism, a single benign parathyroid tumor (adenoma) is responsible, whereas multiple hypercellular glands are present in approx 15% (primary hyperplasia or “double adenomas”). Parathyroid carcinoma is rare, as is the ectopic secretion of PTH from nonparathyroid tumors. A comprehensive molecular pathophysiological description of parathyroid tumorigenesis will eventually need to fully explain the development of these types of tumors, as well as a variety of other special features such as the increased incidence of parathyroid tumors after exposure to neck irradiation and the disease’s epidemiological weighting toward postmenopausal women. Detailed molecular understanding will likely yield information of diagnostic, prognostic, preventative, or therapeutic importance. CLONALITY IN PARATHYROID TUMORIGENESIS The monoclonality or polyclonality of human tumors is an informative reflection of their underlying pathogenetic mechanism. Early data measuring isoforms of the X-chromosome-encoded protein G6PD in parathyroid adenomas in heterozygous women had indicated that apparently single parathyroid adenomas were polyclonal growths, likely to result solely from a generalized growth stimulus. These results, however, proved to be misleading, because modern molecular methods have now solidly established the monoclonality of typical parathyroid adenomas, both by X-chromosome inactivation analysis and by the direct demonstration of monoclonal genetic alterations in parathyroid adenomas (see below). Monoclonality highlights the concept that parathyroid adenomas are true neoplasms, consistent with clinical experience that surgical removal of the enlarged gland is generally curative. Neoplasia is a genetic disease, with most relevant DNA damage occurring somatically.

Figure 36-2 Schematic diagram illustrating the pericentromeric inversion of chromosome 11 deduced to have caused the observed rearrangement involving the PTH gene and the PRAD1 gene in a subset of parathyroid adenomas. The tumor’s other copy of chromosome 11, which contains an intact PTH gene, is not shown. (Reproduced with permission from Arnold A. Genetic basic of endocrine disease 5: molecular genetics of parathyroid gland neoplasia. J Clin Endocrinol Metab 1993;77: 1108–1112. Copyright 1993, The Endocrine Society.)

Monoclonality implies that the necessary accumulation of multiple mutations in a tumor progenitor cell occurs only rarely in a large population of cells within a tissue, conferring a selective growth advantage critical in tumor outgrowth or clonal evolution. The search for the specific oncogenes and tumor suppressor genes that are clonally activated or inactivated, respectively, in common sporadic parathyroid tumors is ongoing. Three notable successes and several important leads, described in the next section have emerged. The clonal status of parathyroid tumors other than adenomas has also been investigated. As expected, parathyroid carcinomas are monoclonal. More surprisingly, however, a substantial percentage of parathyroid tumors in the setting of primary hyperplasia and severe secondary hyperparathyroidism of uremia are monoclonal, indicating that somatic mutations have given selected cells a growth advantage over their already hyperplastic neighbors. It is conceivable that the “conversion” from polyclonality to monoclonality may be a key factor in the increasing autonomy of PTH secretion that develops in many hemodialysis patients, making them refractory to conventional medical therapy. Future identification of the specific molecular culprits in such clonal outgrowths may lead to rational new therapy or preventive measures for this important clinical problem. GENETIC DERANGEMENTS IN SPORADIC BENIGN PARATHYROID TUMORS The Cyclin D1/PRAD1 Oncogene Cyclin D1/PRAD1, the only oncogene implicated in sporadic parathyroid neoplasia, was discovered by virtue of its proximity to a clonal chromosomal breakpoint in a subset of parathyroid adenomas. This chromosomal inversion causes overexpression of the cyclin D1/PRAD1 gene by placing it in proximity to the strong tissue-specific enhancer of the PTH gene (Fig. 36-2). Cyclin D1 protein overexpression, resulting from gene rearrangement or other mechanisms, has been found in 20–40% of parathyroid adenomas. That such overexpression indeed causes parathyroid neoplasia has been experimentally validated by the demonstration of primary hyperparathyroidism in transgenic mice with parathyroid-targeted overexpression of cyclin D1. Cyclin D1 activation has also been incriminated in a variety

CHAPTER 36 / DISORDERS OF THE PARATHYROID GLAND

of other human tumors, including B-cell lymphoma, breast, and esophageal cancers. Cyclin D1/PRAD1 is recognized to have a crucial role in regulating progression through the G1 phase of the cell division cycle. To do so, the cyclin D1 protein is thought to act as an activating regulatory subunit for its partner cyclin-dependent kinase(s), cdk4 or cdk6. One action of active cdk4 or cdk6 may be phosphorylation of the retinoblastoma gene product pRB, moving the cell toward S phase, but this mechanism has not been established in parathyroid tissue. TUMOR SUPPRESSOR GENES Inactivation of both alleles of a tumor suppressor gene is typically necessary to adequately eliminate its antioncogenic product, and somatic deletion of a sometimes large stretch of DNA that includes the relevant gene is a common inactivating mechanism. Thus, identification of genomic regions that are clonally and nonrandomly lost in parathyroid adenomas can point to the locations of putative parathyroid tumor suppressor genes. Of sporadic parathyroid adenomas, approx 25–35% contain allelic losses of chromosome 11 DNA, often (but not always) including the region containing the MEN1 gene. This gene, responsible for familial multiple endocrine neoplasia type 1 (MEN1), was identified using positional cloning techniques (Chapter 39). Mutations in the affected individuals predict loss of function of the protein, indicating that MEN1 is a tumor suppressor gene. The gene product, menin, is a 610-amino acid protein located in the nucleus, but the precise molecular mechanisms responsible for its tumor suppressor activity remain uncertain. In sporadic parathyroid tumors, not associated with familial MEN1, biallelic somatic mutations inactivating MEN1 occur in 12–17%, or about half the tumors with allelic losses on 11q. It is not clear whether the tumor suppressor target in the remaining tumors with 11q loss is MEN1 or a different gene in its vicinity. Mutation of another tumor suppressor gene, HRPT2, was identified as the cause of the rare autosomaldominant hyperparathyroidism-jaw tumor (HPT-JT) syndrome, and somatic mutations in HRPT2 have been reported in a few sporadic parathyroid adenomas. However, the rarity of these mutations stands in marked contrast to the high frequency of HRPT2 mutations in parathyroid carcinoma (see Molecular Pathogenesis of Parathyroid Carcinoma). Finally, several other genomic regions of nonrandom clonal allelic loss in parathyroid adenomas highlight the locations of putative parathyroid tumor suppressor genes that remain unidentified, including 1p, 6q, 9p, and 15q. These data emphasize the molecular heterogeneity underlying parathyroid adenomatosis. OTHER GENETIC ASPECTS Some genes responsible for rare inherited predispositions to certain tumors have also proved important in more common, sporadic forms of the same tumors. The MEN1 gene, as mentioned, is an example. The discovery of RET proto-oncogene germline mutation in MEN2a (Chapter 40) made this gene a candidate for involvement in nonfamilial hyperparathyroidism. However, studies have failed to document somatic RET mutations in sporadic parathyroid adenomas. Similarly, inactivating mutations of the extracellular CaR have been sought but not found in sporadic adenomas, although secondary changes in CaR expression may well be important in determining the tumors’ altered sensitivity of PTH secretion to serum calcium. Finally, although the syndrome of familial isolated hyperparathyroidism can include phenotypic variants caused by germline mutations in MEN1, CASR, or HRPT2, additional genetic bases are likely to exist and, once identified, will be candidates for involvement in sporadic parathyroid tumors.

363

Figure 36-3 Molecular pathology of the ectopic production of PTH by an ovarian cancer. Schematic diagram of the normal PTH gene region (top) and the rearranged, amplified PTH gene region (bottom) in a PTHsecreting ovarian tumor. The bold “X” represents the breakpoint of the DNA rearrangement. (Reproduced with permission from Arnold A. Genetic basis of endocrine disease 5: molecular genetics of parathyroid gland neoplasia. J Clin Endocrinol Metab 1993;77:1108–1112.)

MOLECULAR PATHOGENESIS OF PARATHYROID CARCINOMA Apart from distant metastasis and extensive local invasion, most histopathological features of parathyroid carcinoma can overlap with parathyroid adenoma. Molecular insights that distinguish between them are thus of potential diagnostic and therapeutic importance. Because parathyroid carcinoma is overrepresented in the rare hereditary HPT-JT syndrome, the HRPT2 gene identified as responsible for this syndrome was examined as a candidate for involvement in sporadic parathyroid carcinoma. Inactivating mutations of HRPT2 were found in the majority of parathyroid cancers, and because noncoding mutations would not have been detected in these analyses, HRPT2 inactivation could be a factor in virtually all parathyroid cancers. Importantly, a subset of patients with apparently sporadic parathyroid carcinoma harbored unsuspected germline mutations of HRPT2, indicating that they may have HPT-JT or a phenotypic variant. The possibility of a familial disorder must be considered in any patient presenting with sporadic parathyroid carcinoma, and DNA diagnosis is being used for early diagnosis and treatment to prevent metastatic carcinoma in genetically susceptible family members. Recurrent clonal alterations have been reported that strongly suggest the involvement and chromosomal locations of other genes important in malignant parathyroid neoplasia. For example, recurrent losses have highlighted a region on chromosome 13 as the site of at least one such tumor suppressor, still to be identified. Several additional locations of frequent clonal DNA losses or gains have been reported as well. Importantly, several genomic regions frequently lost in parathyroid adenomas, including 11q (location of MEN1), 6q, and 15q, are rarely if ever lost in carcinomas, suggesting that parathyroid carcinomas arise de novo rather than from preexisting adenomas. Cyclin D1 may be overexpressed in many parathyroid cancers but larger sample sizes are needed.

ECTOPIC SECRETION OF PTH The ectopic secretion of PTH by nonparathyroid tumors is a rare cause of primary hyperparathyroidism. The molecular basis of ectopic PTH production in one such case, an ovarian carcinoma, was found to be a DNA rearrangement in the regulatory region of the tumor’s PTH gene (Fig. 36-3). Similar detailed molecular pathology has not been described in other examples of human ectopic hormone excess, and might involve analogous DNA

364

SECTION IV / ENDOCRINOLOGY

rearrangements or, alternatively, a change in the tumor tissue’s characteristic DNA-binding proteins. Preliminary reports indicate that ectopic production of PTH in some neoplasms may be associated with expression of GCMB, which may induce transdifferentiation of neoplastic cells to a parathyroid cell-like phenotype.

SELECTED REFERENCES Ahn TG, Antonarakis SE, Kronenberg HM, Igarashi T, Levine MA. Familial isolated hypoparathyroidism: a molecular genetic analysis of 8 families with 23 affected persons. Medicine 1986;65:73–81. Arnold A. The cyclin D1/PRAD1 oncogene in human neoplasia. J Investig Med 1995;43:543–549. Arnold A. Molecular basis of primary hyperparathyroidism. In: Bilezikian JP, Marcus R, Levine MA, eds. The Parathyroids, 2nd ed. San Diego, CA: Academic Press, 2001; pp. 331–347. Arnold A, Horst SA, Gardella TJ, Baba H, Levine MA, Kronenberg HM. Mutation of the signal peptide-encoding region of the preproparathyroid hormone gene in familial isolated hypoparathyroidism. J Clin Invest 1990;86:1084–1087. Baldini A. DiGeorge syndrome: an update. Curr Opin Cardiol 2004; 19:201–204. Bowl MR, Nesbit MA, Harding B, et al. An interstitial deletion–insertion involving chromosomes 2p25.3 and Xq27.1, near SOX3, causes X-linked recessive hypoparathyroidism. J Clin Invest 2005;115:2822–2831. Brown EM, Gamba G, Riccardi D, et al. Cloning and characterization of an extracellular Ca2+-sensing receptor from bovine parathyroid. Nature 1993;366:575–580. Carling T, Szabo E, Bai M, et al. Familial hypercalcemia and hypercalciuria caused by a novel mutation in the cytoplasmic tail of the calcium receptor. J Clin Endocrinol Metab 2000;85:2042–2047. Carpten JD, Robbins CM, Villablanca A, et al. HRPT2, encoding parafibromin, is mutated in hyperparathyroidism-jaw tumor syndrome. Nat Genet 2002;32:676–680. Chandrasekharappa SC, Guru SC, Manickam P, et al. Positional cloning of the gene for multiple endocrine neoplasia-type 1. Science 1997;276:404–407. Chattopadhyay N, Mithal A, Brown EM. The calcium-sensing receptor: a window into the physiology and pathophysiology of mineral ion metabolism. Endocr Rev 1996;17:289–307. Ding C, Buckingham B, Levine MA. Familial isolated hypoparathyroidism caused by a mutation in the gene for the transcription factor GCMB. J Clin Invest 2001;108:1215–1220. Dotzenrath C, The BT, Farnebo F, et al. Allelic loss of the retinoblastoma tumor suppressor gene: a marker for aggressive parathyroid tumors? J Clin Endocrinol Metab 1996;81:3194–3196.

Gunther T, Chen ZF, Kim J, et al. Genetic ablation of parathyroid glands reveals another source of parathyroid hormone. Nature 2000;406: 199–203. Hendy GN, D’Souza-Li L, Yang B, Canaff L, Cole DE. Mutations of the calcium-sensing receptor (CASR) in familial hypocalciuric hypercalcemia, neonatal severe hyperparathyroidism, and autosomal-dominant hypocalcemia. Hum Mutat 2000;16:281–296. Krebs LJ, Shattuck TM, Arnold A. HRPT2 mutational analysis of typical sporadic parathyroid adenomas. J Clin Endocrinol Metab 2005;90: 5015–5017. Kronenberg HM, Bringhurst FR, Segre GV, Potts JT Jr. Parathyroid hormone biosynthesis and metabolism. In: Bilezikian JP, Marcus R, Levine MA, eds. The Parathyroids, 2nd ed. San Diego, CA: Academic Press, 2001; pp. 17–30. Motokura T, Bloom T, Kim HG, et al. A novel cyclin encoded by a bcl1linked candidate oncogene. Nature 1991;350:512–515. Parfitt AM. Parathyroid growth: normal and abnormal. In: Bilezikian JP, Marcus R, Levine MA, eds. The Parathyroids, 2nd edition. San Diego, CA: Academic Press, 2001; pp. 293–329. Parkinson DB, Thakker RV. A donor splice site mutation in the parathyroid hormone gene is associated with autosomal-recessive hypoparathyroidism. Nat Genet 1992;1:149–153. Pollak MR, Brown EM, Chou Y-HW, et al. Mutations in the human Ca2+-sensing receptor gene cause familial hypocalciuric hypercalcemia and neonatal severe hyperparathyroidism. Cell 1993;75:1297–1303. Pollak MR, Brown EM, Estep HL, et al. Autosomal-dominant hypocalcaemia caused by a Ca2+-sensing receptor gene mutation. Nat Genet 1994;8:303–307. Shattuck TM, Välimäki S, Obara T, et al. Somatic and germline mutations of the HRPT2 gene in sporadic parathyroid carcinoma. N Engl J Med 2003;349:1722–1729. Sunthornthepvarakul T, Churesigaew S, Ngowngarmratana S. A novel mutation of the signal peptide of the preproparathyroid hormone gene associated with autosomal-recessive familial isolated hypoparathyroidism. J Clin Endocrinol Metab 1999;84:3792–3796. Thakker RV. Molecular basis of PTH underexpression. In: Bilezikian JP, Raisz LG, Rodan GA, eds. Principles of Bone Biology, 2nd ed. San Diego, CA: Academic Press, 2002; pp. 1105–1116. Van Esch H, Devriendt K. Transcription factor GATA3 and the human HDR syndrome. Cell Mol Life Sci 2001;58:1296–1300. Yagi H, Furutani Y, Hamada H, et al. Role of TBX1 in human del22q11.2 syndrome. Lancet 2003;362:1366–1373. Yamagishi H, Garg V, Matsuoka R, Thomas T, Srivastava D. A molecular pathway revealing a genetic basis for human cardiac and craniofacial defects. Science 1999;283:1158–1161.

37 Congenital Adrenal Hyperplasia ROBERT C. WILSON AND MARIA I. NEW SUMMARY Congenital adrenal hyperplasia is a family of autosomal-recessive disorders caused by mutations that encode for enzymes involved in one of the various steps of adrenal steroid synthesis. This chapter discusses adrenal steroidogenesis, 21-hyroxylase deficiency, 11β-hydroxylase deficiency, 3β-hydroxysteroid dehydrogenase deficiency, 17α hydroxylase/17,20-lyase deficiency, lipoid congenital adrenal hyperplasia, and various treatments. Key Words: Adrenal steroidogenesis; congenital adrenal hyperplasia (CAH); 3β-hydroxysteroid dehydrogenase deficiency; 11β-hydroxylase deficiency; 17α hydroxylase/17,20-lyase deficiency; 21-hydroxylase deficiency (21-OHD); lipoid congenital adrenal hyperplasia; salt wasting.

INTRODUCTION Congenital adrenal hyperplasia (CAH) is a family of autosomalrecessive disorders caused by mutations that encode for enzymes involved in one of the various steps of adrenal steroid synthesis (Table 37-1). These defects result in the absence or the decreased synthesis of cortisol from its cholesterol precursor. The anterior pituitary secretes excess adrenocorticotropic hormone (ACTH) via feedback regulation by cortisol, which results in overstimulation of the adrenals and causes hyperplasia. Symptoms owing to CAH can vary from mild to severe depending on the degree of the enzymatic defect. In the classical forms of CAH, defects in the cytochrome P450s 21-hydroxylase (21-OH) or 11β-hydroxylase (11β-OH) cause varying degrees of genital ambiguity in females owing to shunting of excess cortisol precursors to the androgen synthesis pathway (Fig. 37-1). Prenatal adrenal androgen excess causes virilization of female genitalia and postnatally results in advanced bone age and puberty in both females and males. Defects in androgen synthesis owing to defects in 3β-hydroxysteroid dehydrogenase (3β-HSD)/∆5,4-isomerase, in 17α-hydroxylase (17αOH)/17,20-lyase, and in the steroidogenic acute regulatory protein (StAR) result in inadequate prenatal virilization of males and depressed puberty in both sexes. Less severe, nonclassical forms of CAH present postnatally as signs of androgen excess.

ADRENAL STEROIDOGENESIS Aldosterone, cortisol (compound F) and testosterone are derived from cholesterol and utilize many of the same enzymes for From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

their synthesis in the adrenal cortex (see Fig. 37-1). Therefore, defects in any of the enzymes that are common to the synthesis pathway of these hormones can result in the loss of a combination of some or all of their production. Cortisol and aldosterone are synthesized in distinct zones of the adrenal cortex called the zona fasciculata and zona glomerulosa (ZG), respectively. Synthesis of these hormones is regulated by different mechanisms: cortisol synthesis is regulated by a negative feedback loop in which high serum levels of cortisol inhibit the release of ACTH from the pituitary, whereas low serum levels of cortisol stimulate the release of ACTH. This defines the hypothalamic–pituitary–adrenal axis (Fig. 37-2). Aldosterone, a hormone required for the regulation of sodium reabsorption across the tight epithelium of the renal distal tubule, controls fluid volume and is regulated by the renin-angiotensin system and by serum potassium. Angiotensin II (a potent vasoconstrictor) directly stimulates the secretion of aldosterone by the ZG when there is a reduction in renal perfusion resulting from an increase in plasma renin. Serum potassium also regulates the synthesis of aldosterone, though independently of volume status.

21-HYDROXYLASE DEFICIENCY Over 90% of cases with classical CAH are because of 21hydroxylase deficiency (21-OHD). There are three forms of this syndrome: classical salt wasting, classical simple virilizing, and nonclassical. CLASSICAL SIMPLE VIRILIZING Fetal adrenocortical function begins in the third month of gestation. At this time, reduction or loss of 21-OH activity results in the buildup of 17αhydroxyprogesterone (17-OHP). Because 17-OHP is a precursor of testosterone, elevated levels of 17-OHP are shunted to the androgen pathway, thus producing excess androgens. This excess adrenal androgen production coincides with the time of sexual development of the fetus and results in varying degrees of genital ambiguity in newborn females. In extreme cases, the urethra extends the full length of the phallus and cannot be distinguished from that of a normal male. In most cases, however, the excess androgens result in an enlarged clitoris with fusion of the labioscrotal folds; it can also result in a urogenital sinus. For these females, the internal genitalia are normal with normal development of ovaries and Müllerian structures. Males affected with 21-OHD are born with normal genitalia. After birth, both females and males develop signs of androgen excess such as precocious development of pubic and axillary hair,

365

366

SECTION IV / ENDOCRINOLOGY

Table 37-1 The Forms of Adrenal Hyperplasia Deficiency

Syndrome

Ambiguous Postnatal genitalia virilization

Salt metabolism

Steroids increased

Steroids decreased

Chromosomal location

None DHEA, 17-OHpregnenolone DHEA, 17-OHpregnenolone DOC, cortico-sterone 17-OHP, ∆4-A 17-OHP, ∆4-A 17-OHP, ∆4-A DOC, 11-deoxycortisol 11-deoxycortisol, ± DOC 18-OHcorticosterone

All Aldo, T, cortisol

8p11.2 1p13.1

StAR Lipoid hyperplasia 3β-Hydroxysteroid Classical dehydrogenase Nonclassical

Males Males

Yes

No salt wasting Salt wasting

No

Yes

Normal

17α-Hydroxylase

—-

Males

No

Hypertension

21-Hydroxylase

Salt wasting Simple virilizing Nonclassic Classical

Females Females No Females

Yes Yes Yes Yes

Salt wasting Normal Normal Hypertension

Nonclassical

No

Yes

Normal

Salt wasting

No

No

Salt wasting

11β-Hydroxylase

Corticosterone methyl oxidase type II

– Cortisol, T

10q24-25

Aldo, cortisol Cortisol —Cortisol, ± aldo

6p21.3 6p21.3 6p21.3 8q21-22

—-

8q21-22

Aldo

8q21-22

Aldo, adosterone; T, testosterone; ∆4-A, ∆4-androstenedione; DHEA, dehydroepiandrosterone; DOC, 11-deoxycorticosterone; 17-OHP, 17α-hydroxyprogesterone

Figure 37-1 Pathways of steroid biosynthesis. Enzymatic activities catalyzing each bioconversion are enclosed in boxes. For activities mediated by specific P450 cytochromes, systematic names of the enzymes (“CYP” followed by number) are listed in parentheses. Other bioconversions (*) are mediated by different enzymes in various tissues. The planar structures of cholesterol, aldosterone, cortisol, dihydrotestosterone, and estradiol are placed near the corresponding labels.

acne, phallic enlargement, rapid growth, and musculoskeletal development. Though initial growth in these patients is rapid, because of premature epiphyseal fusion, potential height is reduced and short adult stature results. Diagnosis may be delayed in males,

as the genital ambiguity that leads to diagnosis at birth in females is absent in males. Even if diagnosis is not delayed and adrenal androgen excess is controlled, patients do not generally achieve their target height.

CHAPTER 37 / CONGENITAL ADRENAL HYPERPLASIA

367

Figure 37-2 Feedback in the hypothalamic–pituitary–axis in (A) the normal individual and (B) the patient with classical CAH, formerly known as “adrenogenital syndrome.”

CLASSICAL SALT WASTING Salt-wasting 21-OHD makes up approx 75% of cases with classical 21-OHD. Depending on the severity of the loss of 21-OH function, adrenal aldosterone secretion may not be sufficient for regulating sodium reabsorption by distal renal tubules. Patients with insufficient aldosterone can suffer from salt-wasting 21-OHD. These patients exhibit the same symptoms as those with simple virilizing 21-OHD but with the potential of adrenal crisis (azotemia, vascular collapse, shock, and death) because of renal salt wasting. Adrenal crisis can occur as early as 1–4 wk of life. Although salt-wasting 21-OHD females may be diagnosed at birth owing to ambiguous genitalia, affected males are at high risk of salt-wasting adrenal crisis because their normal genitalia do not make the condition obvious. NONCLASSICAL Nonclassical 21-hydroxylase deficiency (NC21-OHD) may present at any time, in childhood, adolescence, or adulthood. Symptoms of NC21-OHD may include acne, premature development of pubic hair, advanced bone age, accelerated linear growth velocity, and as in classical 21-OHD, reduced adult stature owing to premature epiphyseal fusion. Symptoms have been observed to wax and wane. Females affected with NC21-OHD are born with normal genitalia, though postnatal symptoms may include hirsutism, temporal baldness, severe cystic acne, delayed menarche, menstrual irregularities, and infertility. A subset of female patients with NC21OHD develops polycystic ovaries. Boys manifesting NC21-OHD may have early beard growth, acne, early growth spurt, premature pubic hair, and an enlarged phallus. Proportionately small testes as compared with the phallus is a reliable indication of adrenal androgen excess as opposed to testicular androgen excess. Adrenal androgen excess in men is not easily detectable and may only be manifested by short stature or oligozoospermia and diminished fertility. A limited number of males and females who are affected with NC21-OHD remain asymptomatic, as discovered during family studies. However, biochemically such patients compare to affected individuals. EPIDEMIOLOGY Newborn screening worldwide of nearly 6.5 million babies has demonstrated an overall incidence of 1:15,000 live births for the classical form of 21-OHD. However, in two isolates the frequency is much higher (Yup’ik Eskimos in Alaska, 1/282, and the inhabitants of La Réunion island in France, 1/2141). The incidence of classical CAH in either homogenous

or heterogenous general populations is as high as 1/7500 live births (Brazil). It has been suggested that NC21-OHD is the most common autosomal-recessive disorder. In some ethnic populations, the frequency of NC21-OHD is so high it is regarded to be the most frequent autosomal-recessive defect. The highest ethnic-specific disease frequency was found among Ashkenazi Jews at 1/27. Other ethnicspecific frequencies were found to be 1/53 for Hispanics, 1/63 for Yugoslavs, 1/100 in a heterogeneous New York City population, and 1/333 for Italians. MOLECULAR GENETICS In 1977, molecular genetic studies of 21-OHD showed linkage with certain human leukocyte antigens (HLA), the human major histocompatibility complex, on the short arm of chromosome 6. For many years, HLA linkage was used for establishing the affected status in family studies. However, in some cases in which the parents shared the same HLA antigens, or if intra-HLA recombination occurred, HLA typing was not diagnostic. The gene for 21-OH, cloned in 1984, is termed CYP21, following the nomenclature for cytochrome P450. Southern blot analysis determined that the CYP21 gene was located within the serum complement component C4. This region of chromosome was duplicated, resulting in two isoforms of C4 (C4A and C4B) and what initially looked like two isoforms of CYP21. Sequence analysis of the two CYP21 genes revealed that one isoform contained a sequence that on translation resulted in a truncated nonfunctional protein, and it was therefore termed CYP21P for pseudogene. The nonfunctionality of CYP21P was confirmed in families without any hormonal abnormalities in which the CYP21P gene was completely missing. CYP21 and CYP21P are 96–98% homologous (Fig. 37-3) and are arranged in tandem within C4A and C4B, separated by approx 30 kb. The duplication of the locus containing CYP21 allows for misalignment of chromatids during meiosis, resulting in unequal crossing over. This results in a high frequency of deletions of the CYP21 and C4B genes. Duplications also occur, but without any clinical consequence. Deletions of CYP21 are found in approx 20% of the patients with classical 21-OHD. CYP21 and CYP21P consist of 10 exons spanning approx 5 kb. Most of the 21-OHD patients carry mutations found in CYP21P (Fig. 37-4). The generally accepted mechanism by which these deleterious mutations are transferred to the active CYP21 gene

368

SECTION IV / ENDOCRINOLOGY

Figure 37-3 The two homologues: CYP21 (the active gene) and CYP21P (the pseudogene). Noncorrespondent bases number less than 90 over a distance of 5.1 kb of DNA. Numbered are the pseudogene base changes frequently identified on mutant CYP21 genes responsible for 21-OHD through an apparent process termed gene conversion. (1) Missense mutation Pro-30 to Leu. (2) Point mutation in intron 2 causes new acceptor 3´-site to be recognized by intron splicing mechanism. (3) Eight basepair deletion shifts the reading frame. (4) Missense mutation Ile-172 to Asn. (5) Cluster of three nonconservative amino acid substitutions. (6) Conservative amino acid substitution Val-281 to Leu. (7) Single-base T insert shifts reading frame. (8) Nonsense mutation Gln to TAG. (9) Radical amino acid substitution (Arg-356 to Trp). KEY: BTE, basic transcriptional element; large boxes/line spaces, exons/introns (to scale); light shading, out-of-frame coding; half-height open boxes, stop codons; vertical lines not numbered, neutral amino acid polymorphisms; half-height vertical lines, silent mutations.

Figure 37-4

Mutations in the 21-hydroxylase gene (CYP21). CAH phenotypes: SW, salt wasting; SV, simple virilizing; NC, nonclassical.

from the homologous position in CYP21P is through gene conversion. Nine of these mutations can be transferred to the active CYP21 gene. Four of these mutations result in truncated proteins owing to premature termination of translation and are associated with the salt-wasting phenotype: (1) a point mutation in the second intron near the 3´ splice site results in aberrant splicing of 13 nucleotides upstream of the normal splice site; (2) in the third exon, an eight base deletion can occur, resulting in nonsense amino acids and the occurrence of a stop codon 20 amino acids toward the carboxyterminus; (3) in exon 7, one T nucleotide can be inserted into a run of seven T nucleotides, resulting in a frameshift that produces five aberrant amino acids and premature termination; and (4) a

point mutation in exon 8 in codon 318 changes a glutamine to a stop codon. The five other mutations that can be transferred from CYP21P to CYP21 result in amino acid substitutions. 1. A point mutation in exon 1 causes a substitution of proline by a leucine at codon 30 (P30L). This mutation had 30–60% of the activity of normal 21-OH when expressed in cell culture. The P30L mutation has been found to be associated with the nonclassical phenotype. 2. A point mutation in exon 4 causes a substitution of isoleucine by an asparagine at codon 172 (I172N) and is most often associated with the simple virilizing phenotype.

369

CHAPTER 37 / CONGENITAL ADRENAL HYPERPLASIA

Table 37-2 Mutations Causing CAH Exon/Intron Ex 1 Int 2 Ex 3 Ex 4 Ex 6 Ex 7 Int 7 Ex 8 Ex 8 Ex 8 Ex 10 Ex 10

Mutation

NT/Mutation

30-kb deletion Missense Aberrant splicing of intron 2 Frameshift Missense Cluster Missense Loss of splice donor site Nonsense Missense Missense Missense Missense

89 C∆T 656 A [or C] ∆G ∆708–715 1001 T∆A 1382 T∆A 1385 T∆A 1391 T∆A 1685 G∆T 1781 G∆C 1996 C∆T 2060 G∆A 2110 C∆T 2580 C∆T 2672 G∆C

AA P30L G110∆8nt I172N I236N, V237E, M239K V281L Q318X R339H R356W P453S R483P

Phenotype SW NC SW, SV SW SV SW NC SW SW NC SW NC SW

Ex, exon; Int, intron; ∆, deletion; SW, salt-wasting; SV, simple virilizing; NC, nonclassical; NT, nucleotide.

This mutation results in 9 kb of genomic DNA and encodes a 610 amino acid protein. The 1.83-kb coding region is organized into nine exons (exons 2–10) and eight introns (indicated by a line but not to scale). The sizes of the exons (boxes) range from 88 to 1312 bp and those of the introns range from 41 to 1564 bp. The start (ATG) and stop (TGA) sites in exons 2 and 10, respectively, are indicated. Exon 1, the 5´ part of exon 2 and 3´ part of exon 10 are untranslated (indicated by the hatched boxes). The locations of the three domains, which are formed by codons 1–40 (exon 2), 139–242 (exons 3 and 4), and 323–428 (exons 7–9), that interact with JunD are indicated by the stippled black boxes. The sites of the 623 germline mutations are indicated by the vertical lines above the gene and the sites of the 164 somatic mutations are represented below the gene. Mutations that have occurred more than four times (scale shown on the right) are indicated. (Reproduced by permission of the Society for Endocrinology. Pannet AAJ, Thakker RVT. Multiple endocrine neoplasia type1 [MEN1] gene. Endocr Relat Cancer 1996;6:449–473.)

mutations of the MEN1 gene have been identified, and most (>80%) of these are inactivating and consistent with its role as a tumor suppressor gene. These mutations are diverse in their types: approx 25% are nonsense mutations, approx 45% are frameshift deletions or insertions, approx 5% are in frame deletions or insertions, approx 5% are splice site mutations, and approx 15% are missense mutations (Fig. 39-3). More than 10% of the MEN1 mutations arise de novo and may be transmitted to subsequent generations. It is also important to note that 5–10% of MEN1 patients may not harbor mutations in the coding region of the MEN1 gene, and that these individuals may have mutations in the promoter or untranslated regions, which remain to be investigated. The mutations are not only diverse in their types but are also scattered throughout the 1830-bp coding region of the MEN1 gene (see Fig. 39-2) with no evidence for clustering as observed in MEN2 (see Chapter 40). Correlations between MEN1 mutations and the clinical manifestations of the disorder appear to be absent. This apparent lack of genotype–phenotype correlations, which contrasts with the situation in MEN2 (see Chapter 40), together with the wide diversity of mutations in the 1830 bp coding region of the MEN1 gene has made mutational analysis for diagnostic purposes in MEN1 time-consuming and expensive. Tumors from MEN1 patients and non-MEN1 patients have been observed to harbor the germline mutation together with a somatic mutation or LOH involving chromosome 11q13, as expected from Knudson’s model and the proposed role of the MEN1 gene as a tumor suppressor. The somatic mutations in tumors are scattered throughout the coding region (see Fig. 39-2) in a manner similar to that observed for the germline mutations. The 164 reported somatic mutations are also diverse in their types: approx 20% are nonsense mutations, approx 40% are frameshift deletions or insertions, approx 5% are inframe deletions or insertions, approx 5% are splice-site mutations, and approx 30% are missense mutations. Interestingly, somatic missense mutations have a significantly (p < 0.001) higher occurrence than germline missense mutations (see Fig. 39-3), although the reasons underlying this observation remain to be elucidated. Somatic MEN1 mutations also occur in sporadic (i.e., nonfamilial) tumors in patients

who do not have MEN1, and this is consistent with the Knudson hypothesis that predicts that mutations of a gene causing a hereditary cancer syndrome will also be involved in causing some nonhereditary forms of the cancer. Thus, MEN1 mutations have been reported in sporadic tumors of the parathyroids, pancreatic islet cells, anterior pituitary, and adrenal cortex (Fig. 39-4). Such MEN1 somatic mutations have also been observed in carcinoid tumors, lipomas, angiofibromas, and melanomas.

OTHER HEREDITARY ENDOCRINE TUMOR SYNDROMES AND GERMLINE MEN1 MUTATIONS The role of the MEN1 gene in the etiology of other endocrine disorders in which either parathyroid or pituitary tumors occur as an isolated endocrinopathy has been investigated by mutational analysis. Germline MEN1 mutations have been reported in 16 families with isolated primary hyperparathyroidism. The sole occurrence of parathyroid tumors in these families is remarkable and the mechanisms that determine the altered phenotypic expressions of these mutations remain to be elucidated. Mutational analysis studies in another inherited isolated endocrine tumor syndrome, that of familial isolated acromegaly, have not detected abnormalities of the MEN1 gene, even though segregation analysis in one family indicated that the gene was likely to be located on chromosome 11q13. However, nonsense mutations have been detected in MEN1 families with the Burin or prolactinoma variant, which is characterized by a high occurrence of prolactinomas and a low occurrence of gastrinomas.

FUNCTION OF MEN1 PROTEIN (MENIN) Initial analysis of the predicted amino acid sequence encoded by the MEN1 gene did not reveal homologies to any other proteins, sequence motifs, signal peptides, or consensus nuclear localization signal, and, thus, the putative function of the protein (menin) could not be deduced. However, studies based on immunofluorescence, Western blotting of subcellular fractions, and epitope tagging with enhanced green fluorescent protein, revealed that menin was located primarily in the nucleus. Furthermore, enhanced green

390

SECTION IV / ENDOCRINOLOGY

Figure 39-3 Frequency of germline and somatic multiple endocrine neoplasia type 1 mutations. A total of 623 germline mutations and 164 somatic mutations have been reported, and these are of diverse types, for example, nonsense, frameshifts, deletions, insertions, splice site, and missense mutations. The frequencies of each type of mutation in the germline and somatic group are similar with the exception of the missense mutations, which are found more frequently in tumors, i.e., the somatic group.

Figure 39-4 Frequencies of multiple endocrine neoplasia type-1 somatic mutations in nonfamilial (i.e., sporadic) tumors from nonMEN1 patients. The numbers (n) of tumors studied is indicated below for each group. Thus, MEN1 somatic mutations were observed in 18% of sporadic parathyroid tumors, 38% of gastrinomas (GAS), 14% of insulinomas (INS), 16% of nonfunctioning pancreatic islet cell tumors (NFPa), 3.5% of anterior pituitary (PIT) tumors, 35% of carcinoid tumors, and 2% of adrenocortical tumors (ADR[CT]). In addition, MEN1 somatic mutations have been observed in 57% of VIPomas (n = 7), 60% of glucagonomas (n = 5), 28% of lipomas (n = 7), 10% of angiofibromas (n = 19), and 2.5% of melanomas (n = 40) (data not shown).

fluorescent protein-tagged menin deletional constructs identified at least two independent nuclear localization signals (codons 479–497 and 588–608) that were located in the C-terminal quarter of the protein. Interestingly, the truncated MEN1 proteins that would result from the nonsense and frameshift mutations, if expressed, would lack at least one of these nuclear localization signals (see Fig. 39-2). Indeed menin is predominantly a nuclear protein in nondividing cells, but in dividing cells it is found in the cytoplasm. The function of menin remains to be elucidated but it does interact with a number of proteins involved in transcriptional regulation cell division, and genome stability. Thus, in transcriptional regulation, menin interacts with the activating protein-1, transcription factors JunD and C-jun, to suppress Jun mediated

transcriptional activation; members, for example, p50, p52, and p65, of the nuclear factor κB family of transcriptional regulators to repress nuclear factor κB mediated transcriptional activation; members of the Smad family, Smad3 and the Smad 1/5 complex, to inhibit the transforming growth factor-β and the bone morphogenetic protein-2 signaling pathways, respectively; and the mouse placental embryonic expression gene that encodes a homeobox containing protein. An interaction between menin and the nonmuscle myosin II-A heavy chain (NMHC II-A) indicates a role in cell division, as NMCH II-A may mediate alterations in cytokinesis and cell shape during cell division. A role for menin in controlling genome stability has been proposed because of its interactions with a subunit of replication protein (RPA2), which is a heterotrimeric protein required for DNA application, recombination, and repair; the tumor metastases suppressor NM23H1/nucleoside diphosphate kinase that induces GTPase activity; and the glial fibrillary acidic protein and vimentin, which are involved in the intermediate filament network. Thus, menin appears to have a large number of potential functions through interactions with proteins; whether these alter cell proliferation mechanisms independently or act via a single pathway remains to be elucidated.

MOUSE MODEL FOR MEN1 Mouse models for MEN1 have been generated through homologous recombination (i.e., knockout) of the mouse MEN1 gene. The mouse MEN1 gene consists of a 1833-bp open-reading frame that encodes a 611 amino acid protein. The mouse menin protein contains one more amino acid residue than the human menin, a glycine at 528. However, the mouse and human coding regions have 89 and 96% identities of the nucleotide and amino acid sequences, respectively, indicating a high degree of evolutionary conservation. One mouse knockout model for MEN1 was generated by introducing a floxed PGK-neomycin cassette into intron 2 and a third loxP site into intron 8, with the aim of deleting exons 3–8 in one allele. Heterozygous mice (+/–) developed parathyroid dysplasia and adenomas by 9 mo of age, pancreatic islet cell tumors that contained insulin by 9 mo of age, anterior pituitary tumors that contained prolactin by 16 mo of age, and adrenocortical carcinomas. The tumors, which had LOH at the MEN1 locus,

CHAPTER 39 / MULTIPLE ENDOCRINE NEOPLASIA TYPE 1

were not associated with any serum biochemical abnormalities, such as hypercalcemia, or hypoglycemia, but those +/– mice developing pancreatic islet cell tumors or hyperplasia had elevated serum insulin concentrations. Thus, these heterozygous (+/–) mice provide a model for the human MEN1 disease. However, in another study, heterozygous mice (+/–) surprisingly died as embryos in late gestation, with some embryos developing omphaloceles. Two studies have reported that homozygous (–/–) mice die in utero at embryonic d 11.5–13.5. In one study these –/– mice were developmentally delayed and significantly smaller, and 20% of them developed craniofacial abnormalities. The craniofacial abnormalities were because of dysplasia of the membranous skull bones, and this developmental pathway involves the bone morphogenetic protein2 signaling pathway. In another study, –/– mice were smaller in size, and developed extensive hemorrhage and edema. In addition, many of these –/– mice also had abnormalities of the neural tube, heart, and liver. Thus, many –/– mice had a failure of the closure of the neural tube, myocardial hypotrophy with a thin intraventricular septum, and decreased hepatic cellularity, which was associated with an altered organization and enhanced apoptosis. These results from the –/– mice reveal an important role for the MEN1 gene in the embryonic development of multiple organs.

FUTURE DIRECTIONS Combined clinical and laboratory investigations of MEN1 patients will improve patient management and treatment, and also facilitate institution of a screening protocol. In addition, these advances will provide a better understanding of the role of these mutations in causing the endocrine tumors. One of the challenges remaining is the remarkable tissue specificity of the MEN1 tumors. The MEN1 gene is ubiquitously expressed from an early embryonic stage, and yet the tumors form principally in three endocrine glands. The availability of the mouse model may help to answer this question, and the availability of cellular models will help elucidate the role of menin in regulating cell proliferation via transcriptional regulation and genome stability.

ACKNOWLEDGMENTS The author grateful to the Medical Research Council (MRC), UK for support; to B Harding for preparing the figures and to Miss Julie Allen for expert secretarial assistance.

SELECTED REFERENCES Agarwal SK, Guru SC, Heppner C, et al. Menin interacts with the AP1 transcription factor JunD and represses JunD-activated transcription. Cell 1999;84:730–735. Agarwal SK, Kester MB, Deblenko LV, et al. Germline mutations of the MEN1 gene in familial multiple endocrine neoplasia type 1 and related states. Hum Mol Genet 1997;6:1169–1175. Bassett JH, Rashbass P, Harding B, Forbes SA, Pannett AA, Thakker RV. Studies of the murine homolog of the multiple endocrine neoplasia type 1 (MEN1) gene, MEN1. J Bone Miner Res 1999;14:3–10. Bassett JHD, Forbes SA, Pannett AAJ, et al. Characterisation of mutations in patients with multiple endocrine neoplasia type 1 (MEN1). Am J Hum Genet 1998;62:232–244. Bertolino P, Radovanovic I, Casse H, Aguzzi A, Wang Z-Q, Zhang C-X. Genetic ablation of the tumor suppressor menin causes lethality at midgestation with defects in multiple organs. Mech Dev 2003;120: 549–560. Brandi ML, Gagel RF, Angeli A, et al. Guidelines for diagnosis and therapy of MEN type 1 and type 2. J Clin Endocrinol Metab 2001;86(12): 5658–5671. Chandrasekharappa SC, Guru SC, Manickam P, et al. Positional cloning of the gene for multiple endocrine neoplasia-type 1. Science 1997;276: 404–407.

391

Chandrasekharappa SC, Teh BT. Functional studies of the MEN1 gene. J Med 2003;253:606–615. Crabtree JS, Scacheri PC, Ward JM, et al. A mouse model of multiple endocrine neoplasia, type 1, develops multiple endocrine tumors. Proc Natl Acad Sci USA 2001;98:1118–1123. Darling TN, Skarulis MC, Steinberg SM, Marx SJ, Spiegel AM, Turner M. Multiple facial angiofibromas and collagenomas in patients with multiple endocrine neoplasia type 1. Arch Dermatol 1997;133:853–861. Guru SC, Goldsmith PK, Burns AL, et al. Menin, the product of the MEN1 gene, is a nuclear protein. Proc Natl Acad Sci USA 1998;95:1630–1634. Heppner C, Bilimoria KY, Agarwal SK, et al. The tumour suppressor protein and inhibits NF-kappaB-mediated transactivation. Oncogene 2001;20(36):4917–4925. Jensen RT. Management of the Zollinger-Ellison syndrome in patients with multiple endocrine neoplasia type 1. J Intern Med 1998;243:477–488. Kaji H, Canaff L, Lebrun JJ, Goltzman D, Hendy GN. Inactivation of menin, a Smad3-interacting protein, blocks transforming growth factor type beta signalling. Proc Natl Acad Sci USA 2001;98(7):3837–2842. Knudson AG. Antioncogenes and human cancer. Proc Natl Acad Sci USA 1993;90:10914–10921. Larsson C, Skogseid B, Oberg K, Nakamura Y, Nordenskjold MC. Multiple endocrine neoplasia type I gene maps to chromosome 11 and is lost in insulinoma. Nature 1988;332:85–87. Lemmens IH, Forsberg L, Pannett AA, et al. Menin interacts directly with the homeobox-containing protein Pem. Biochem Biophys Res Commun 2001;286:426–431. Lopez-Egido J, Cunningham J, Berg M, Oberg K, Bongcam-Rudloff E, Gobl A. Menin’s interaction with glial fibrillary acidic protein and vimentin suggests a role for the intermediate filament network in regulating menin activity. Exp Cell Res 2002;278:175–183. Marx SJ. Multiple Endocrine Neoplasia Type 1. In: Vogelstein B, Kinzler KW, eds. Genetic Basis of Human Cancer, New York: McGraw Hill, 1998; pp. 489–506. Norton JA, Fraker DL, Alexander R, et al. Surgery to cure the ZollingerEllison syndrome. N Engl J Med 1999;341:635–644. Ohkura N, Kishi M, Tsukada T, Yamaguchi K. Menin. A gene product responsible for multiple endocrine neoplasia type 1, interacts with the putative tumor metastasis suppressor nm23. Biochem Biophys Res Commun 2001;282:1206–1210. Olufenic SE, Green JS, Manikam P, et al. Common ancestral mutation in the MEN1 gene is likely responsible for the prolactinoma variant of MEN1 (MEN1 burin) in four kindreds from Newfoundland. Hum Mutat 1998;11:204–269. Pannett AAJ, Kennedy AM, Turner JJO, et al. Multiple endocrine neoplasia type 1 (MEN1) germline mutations in familial isolated primary hyperparathyroidism. Clin Endocrinol (Oxf) 2003;58:639–646. Pannett AAJ, Thakker RVT. Multiple endocrine neoplasia type 1 (MEN1) gene. Endocr Relat Cancer 1999;6:449–473. Scacheri PC, Crabtree JS, Novotny EA, et al. Bidirectional transcriptional activity of PGK—neomycin and unexpected embryonic lethality in heterozygote chimeric knockout mice. Genesis 2001;30:259–263. Skogseid B, Larsson C, Lindgren PG, et al. Clinical and genetic features of adrenocortical lesions in multiple endocrine neoplasia type 1. J Clin Endocrinol Metab 1992;75:76–81. Skogseid B, Oberg K, Benson L, et al. A standardized meal stimulation test of the endocrine pancreas for early detection of pancreatic endocrine tumors in multiple endocrine neoplasia type 1 syndrome: five years experience. J Clin Endocrinol Metab 1987;64:1233–1240. Sowa H, Kaji H, Canaff L, et al. Inactivation of menin, the product of multiple endocrine neoplasia type 1 gene, inhibits the commitment of multipotential mesenchymal stem cells into the osteoblast lineage. J Biol Chem 2003;23:21,058–21,069. Stewart C, Parente F, Piehl F, et al. Characterisation of the mouse MEN1 gene and its expression during development. Oncogene 1998;17: 2485–2493. Sukhodolets KE, Hickman AB, Agarwal SK, et al. The 32-kilodalton subunit of replication protein A interacts with menin, the product of the MEN1 tumour suppressor gene. Mol Cell Biol 2003;23:493–509. Teh BT, Kytola S, Farnebo F, et al. Mutation analysis of the MEN1 gene in multiple endocrine neoplasia type 1, familial acromegaly and familial isolated hyperparathyroidism. J Clin Endocrinol Metab 1998;83: 2621–2626.

392

SECTION IV / ENDOCRINOLOGY

Teh BT, Zedenius J, Kytola S, et al. Thymic carcinoids in multiple endocrine neoplasia type 1. Ann Surg 1998;228:99–105. Thakker RV. The molecular genetics of the multiple endocrine neoplasia syndromes. Clin Endocrinol (Oxf) 1993;39:1–14. Thakker RV. Multiple Endocrine Neoplasia Type 1 (MEN1). In: DeGroot LJ, Besser GK, Burger HG, et al. eds. Endocrinology, Philadelphia: WB Saunders, 2000; 2815–2831. The European Consortium on MEN1. Identification of the multiple endocrine neoplasia Type 1 (MEN1) gene. Hum Mol Genet 1997;6: 1177–1183. Tomassetti P, Migliori M, Caletti GC, Fusaroli P, Corinaldesi R, Gullo L. Treatment of type II gastric carcinoid tumors with somatostatin analogues. N Engl J Med 2000;343:551–554.

Trump D, Farren B, Wooding C, et al. Clinical studies of multiple endocrine neoplasia type 1 (MEN1) in 220 patients. Q J Med 1996;89: 653–669. Wolfe MM, Jensen RT. Zollinger-Ellison syndrome. Current concepts in diagnosis and management. N Engl J Med 1987;317:1200–1209. Yaguchi H, Ohkura N, Tsukada K. Menin, the multiple endocrine neoplasia type 1 gene product, exhibits GTP-hydrolyzing activity in the presence of the tumor metastasis suppressor nm23. J Biol Chem 2002; 277:38,197–38,204. Yamita W, Ikeo Y, Yamauchik K, Sakurai A, Hashizumek K. Suppression of insulin-induced AP-1 transactivation by menin accompanies inhibition of c-Fos induction. Int J Cancer 2003;103:738–744.

40 Multiple Endocrine Neoplasia Type 2 ROBERT F. GAGEL, SARAH SHEFELBINE, HIRONORI HAYASHI, AND GILBERT COTE SUMMARY Multiple endocrine neoplasia is a genetic endocrine tumor syndrome characterized by the presence of medullary thyroid carcinoma (MTC), pheochromocytoma, and hyperparathyroidism. There are two major variants. Multiple endocrine neoplasia (MEN) type 2A or Sipple syndrome has three clinical features: MTC, pheochromocytoma, and hyperparathyroidism; MEN2B has a different phenotype: MTC, pheochromocytoma, mucosal neuromas distributed throughout the mouth and gastrointestinal tract, and Marfanoid features including long thin arms and legs, an altered upper/lower body ratio, and pectus abnormalities. These clinical syndromes are caused by specific activating mutations of the RET tyrosine kinase receptor, important in neurological development of the gastrointestinal tract. MEN2 is one of a handful of genetic syndromes where the identification of a mutation leads to a specific action. Children with germline activating mutations of RET are treated in early childhood with a total thyroidectomy to prevent the development of metastatic MTC. Key Words: Activating mutation; C cell; chromaffin cell; genetic testing; hereditary cancer; hyperparathyroidism; marfanoid features; medullary thyroid carcinoma; MEN2A; MEN2B; mucosal neuromas; multiple endocrine neoplasia type 2; parathyroid cell; pheochromocytoma; RET proto-oncogene; thyroid cancer; tyrosine kinase receptor.

INTRODUCTION John Sipple first described components of the clinical syndrome that bears his name. Recognizing the association of thyroid cancer and bilateral pheochromocytomas in a patient at autopsy he reported this case with several others in a 1961 article. The type of thyroid carcinoma was more clearly defined and the clinical features of this syndrome, defined as multiple endocrine neoplasia (MEN) type 2, were separated from MEN1 (Table 40-1).

CLINICAL FEATURES MEN2 consists of medullary thyroid carcinoma (MTC), pheochromocytoma, and hyperparathyroidism inherited as an autosomaldominant trait. The nomenclature for this disease syndrome has evolved since its initial description; a classification system that reflects the clinical features and molecular causes is shown in From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

Table 40-1. The syndrome originally described by Sipple has been classified as MEN2A. MTC is found in nearly all gene carriers, pheochromocytoma in one-half, and hyperparathyroidism in 10–20%. Characteristics of all neoplastic components associated with MEN2 are bilaterality and multicentricity. This is especially important for MTC in which surgical cure is possible only by removal of all thyroid tissue. There are three variants of MEN2A: familial medullary thyroid carcinoma (FMTC); MEN2A with cutaneous lichen amyloidosis (MEN2A/CLA); and MEN2A with Hirschsprung disease. MTC in FMTC is inherited as an autosomal-dominant trait without other features of MEN2A. The MTC associated with FMTC tends to be less aggressive and is most likely to be confused with sporadic MTC because other manifestations of MEN2A are not present, and many affected family members may be asymptomatic. A diagnosis of FMTC should be made only in large, multigenerational families because of the incomplete penetrance (50% or less) for other manifestations of MEN2A. The MEN2A/CLA variant, found in approx 18 kindreds worldwide, is the association of MEN2A with a characteristic pruritic skin lesion located over the central upper back. The clinical features, other than the skin lesion, are identical to classic MEN2A, although hyperparathyroidism may be less common. Finally, a handful of families have been described with MEN2A and Hirschsprung disease. The penetrance of the Hirschsprung phenotype is variable, ranging from complete penetration to a single affected member. MEN2B is characterized by MTC and pheochromocytoma as well as a unique phenotype of marfanoid habitus and mucosal, oral and intestinal ganglioneuromatosis (see Table 40-1). This disorder is transmitted as an autosomal-dominant trait, although the majority of cases represent de novo mutations; the mutant allele is generally inherited from the father and the mutation is thought to occur during spermatogenesis. MTC associated with MEN2B is more aggressive than observed in MEN2A. Development of carcinoma during the first year of life is common and early death from metastatic MTC occurs in 30–40% of patients. The prognosis, however, is not universally poor. A number of multigeneration families suggest considerable variability in the outcome.

DIAGNOSIS MEN2 is a clinical syndrome that affects multiple endocrine organs. MTC and pheochromocytoma are the most common causes of morbidity and death. Hyperparathyroidism and its associated

393

394

SECTION IV / ENDOCRINOLOGY

Table 40-1 The Multiple Endocrine Neoplasia Syndromes Multiple endocrine neoplasia type 1 Pituitary tumors Parathyroid neoplasia Islet cell neoplasia Multiple endocrine neoplasia type 2 Multiple endocrine neoplasia type 2A (MEN2A) Medullary thyroid carcinoma (100%) Pheochromocytoma (50%) Parathyroid neoplasia (10–20%) Variants of MEN2A Familial medullary thyroid carcinoma MEN2A with cutaneous lichen amyloidosis MEN2A with Hirschsprung disease Multiple endocrine neoplasia type 2B Medullary thyroid carcinoma (100%) Pheochromocytoma (50%) No parathyroid disease Marfanoid habitus (nearly 100%) Intestinal ganglioneuromatosis and mucosal neuromas (nearly 100%)

findings of hypercalcemia, nephrolithiasis, and bone loss are less commonly a clinical problem. The C cell, the cell type comprising MTC, produces a small peptide hormone, calcitonin. In patients with MTC, serum calcitonin values are characteristically elevated and secretion may be stimulated 3- to 20-fold higher by intravenous pentagastrin, calcium, or by a combination of the two. Prospective studies utilizing provocative testing with pentagastrin or the combination of calcium and pentagastrin have been used successfully to identify MTC early in the course of the disease. Prospective pentagastrin screening and thyroidectomy has improved survival and quality of life in these kindreds, although not all children or young adults are cured utilizing this approach. As many as 10–15% of these patients have evidence of recurrent disease or elevated calcitonin values in long-term follow-up studies. Pheochromocytoma occurs in approx 50% of family members with MEN2. These tumors cause palpitations, headaches, and attacks of nervousness early in the course of tumor development; hypertension is infrequently found in patients with small tumors or hyperplasia of the adrenal. In patients with large or bilateral tumors, hypertension and cardiac arrhythmias may occur. Before the recognition of this syndrome in 1961 it is estimated that onehalf of the deaths in MEN2 kindreds occurred suddenly and were attributed to cardiac disease; many were likely related to pheochromocytoma. The goal of screening for pheochromocytoma in kindreds with MEN2 is to identify an adrenal tumor before it causes significant or life-threatening manifestations of catecholamine excess. One successful approach is the annual measurement of plasma or 12 or 24 h urine for catecholamines or metanephrines. Symptomatic patients frequently have an elevation of either plasma or urine catecholamines or metanephrines. MEN2-related pheochromocytomas are unique in that they preferentially produce disproportionate amounts of epinephrine and its metabolite metanephrine when compared with sporadic pheochromocytomas or those associated with von Hippel Lindau syndrome that produce predominately norepinephrine or normetanephrine. Patients with abnormal values should have imaging studies. CT

provides better definition of the adrenal medulla and is less expensive. MRI provides greater specificity if the adrenal medulla lights up on T2-weighted images. The major diagnostic problems occur at the transition between normal and abnormal. Patients may have symptoms suggestive of pheochromocytoma with normal catecholamine measurements and imaging studies. In this situation the differentiation between intermittent abnormal catecholamine secretion and an anxiety disorder can be difficult. A trial of adrenergic blockade or inhibition of catecholamine synthesis with α-methyl tyrosine may help separate these two possibilities. If the patient is improved by pharmacological intervention and surgery is contemplated, the choice of which adrenal to remove may be difficult. Higher resolution CT scans using fine cuts through the adrenal glands permits localization in most cases; in others octreotide or met-iodobenzyl guanidine scanning may provide insight. Another alternative would be to continue the patient on adrenergic blockade or catecholamine synthesis inhibition until a pheochromocytoma can be clearly identified. Successful experience with long-term management of malignant pheochromocytoma, a much more challenging situation, suggests that this approach could be used safely in a compliant patient. Earlier detection and improved surgical techniques have reduced the morbidity and mortality associated with pheochromocytoma to negligible levels. In fact, there is growing evidence that death from adrenal insufficiency may be a greater risk than death from pheochromocytoma; these concerns have led to a renewed interest and use of cortical sparing adrenalectomy, a technique in which the adrenal medulla is removed, leaving the cortex with intact vascularity.

GENETIC BASIS OF MEN2 MAPPING THE CAUSATIVE GENE The MEN2 gene was mapped to centromeric chromosome 10 in independent efforts in 1987. Subsequent work narrowed the region containing the MEN2 gene, leading to the identification of mutations of the RET protooncogene in 1993 (Fig. 40-1). The RET proto-oncogene was first discovered in 1985 when a rearranged form of this gene was shown to cause transformation. The RET proto-oncogene encodes a tyrosine kinase receptor. A naturally occurring rearrangement was subsequently identified in 10–35% of papillary thyroid carcinomas (PTC) and named the PTC oncogene. Multiple different PTC variants exist, each resulting from a chromosome 10 rearrangement that places the tyrosine kinase domain of the RET proto-oncogene under the transcriptional control of another constitutively expressed gene. There is compelling evidence that the RET rearrangement causes transformation of the thyroid follicular cell in animal models of PTC. The Glial Cell-Derived Neurotrophic FACTOR/RET/GFRA Receptor System A remarkable series of studies have defined a tyrosine kinase receptor signaling system important for the development of several components of the nervous system and normal kidney development. Individual components of this system include RET and a small protein ligand, glial cell-derived neurotrophic factor (GDNF), identified as a neuronal survival factor. The recognition that GDNF is a ligand for the RET tyrosine kinase receptor was not suspected until the mid-1990s when mice with RET or GDNF genes deleted were found to have nearly identical phenotypes, characterized by gastrointestinal neuronal features analogous to Hirschsprung disease, severe renal abnormalities, and developmental abnormalities of components of the

CHAPTER 40 / MULTIPLE ENDOCRINE NEOPLASIA TYPE 2

Figure 40-1 The RET tyrosine kinase receptor is a transmembrane receptor with a cadherin (CAD), cysteine-rich (CYS), and tyrosine kinase domain (TK). Interaction of GDNF, one of four ligands for RET, with the glial cell-derived neurotrophic factor receptor (GFRα-1) leads to dimerization of the receptor, autophosphorylation, and phosphorylation of downstream signaling substrates.

sympathetic nervous system. Detailed studies in these animals showed gastrointestinal neuronal features analogous to those found in Hirschsprung disease with incomplete development of the enteric nervous system. There was also a complete lack of renal development in these animals. Components of the sympathetic nervous system, including the superior cervical ganglion, were also underdeveloped. In an independent series of experiments, researchers discovered a receptor for GDNF, termed GDNF family receptor (GFR)-α and showed it to interact with the RET tyrosine kinase receptor. RET and GFRα-1, an extracellular protein tethered to the cell membrane by a glycosyl phosphatidylinositol linkage, together form a receptor for GDNF (see Fig. 40-1). Subsequent studies showed that targeted disruption of the mouse GFRα-1 gene causes a phenotype nearly indistinguishable from that observed for the mice in which GDNF and RET were deleted. Additional work has defined a complex receptor system. The RET tyrosine kinase receptor forms the transmembrane backbone for each of these variants. The four ligands (GDNF, artemin [ARTN], persephin [PSPN], and neurturin [NRTN]) belong to the transforming growth factor-β superfamily and are collectively known as GDNF family ligands (GFL). The presence of seven cysteine residues characterize them as members of the cystine knot protein family. These ligands interact with RET and one of four GFRα proteins (GFRα-1, GFRα-2, GFRα-3, GFRα-4). Perhaps more important than the components of the receptor system is the nature of the interaction between ligand and receptor system and how this interaction serves a developmental role. In a familiar theme of developmental biology, the ligand–receptor interaction serves to bring one tissue expressing the ligand into proximity with another expressing the receptor system. This interaction is perhaps best shown in the interactions between the ureteric bud and the metanephric blastema in kidney development. The ureteric bud (the developing renal ureteral collecting system) normally penetrates the metanephric blastema and branches to form the collecting system (Fig. 40-2). The peptide GDNF is expressed in the developing metanephric blastema and interacts with the RET receptor in the ureteric bud to promote branching of the ureteric bud into the developing kidney. In the absence of

395

Figure 40-2 Schematic view showing mouse embryonic kidney development. The ureteric bud, derived from the Wolffian duct, normally migrates into the developing metanephric blastema to form the collecting system. In either RET (RET –/–) or GDNF (GDNF –/–) knockouts there is a defect in migration of the ureteric bud into the metanephric blastema resulting in a failure of normal kidney development. (Derived from information contained in work published by several authors: Schilling T, Burck J, Sinn HP, et al., 2001; Pichel JG, Shen L, Sheng HZ, et al., 1996; Moore MW, Klein RD, Farinas I, et al., 1996.)

Figure 40-3 Migration of neural crest cells expressing the RET tyrosine kinase into the developing spinal cord and gastrointestinal tract. RET expressing cells form the superior cervical ganglion, parts of the dorsal root of the spinal cord, and the enteric neuronal plexus of the gastrointestinal tract.

either GDNF or RET there is a failure of this interaction, leading to an undeveloped, shrunken kidney. Available facts suggest a similar pattern of interaction between RET and GDNF in the developing enteric neuronal system. Clearly, RET is expressed in neural crest cells associated with somites 1–5. These RET positive cells migrate into the spinal cord and into the developing gastrointestinal tract during embryonic life (Fig. 40-3). The finding of incomplete intestinal neuronal development in RET, GDNF, and GFRα-1 knockout mice, provides compelling evidence for an interaction between the ligand/receptor systems of importance in neuronal innervation of the intestine.

396

SECTION IV / ENDOCRINOLOGY

Figure 40-4 Diagram of the glial cell-derived neurotrophic factor/ RET/GDNF family receptor-α1 receptor system showing intracellular tyrosine phosphorylation sites at codons 905, 1015, 1062, and 1096 with known interacting signaling linkage proteins.

In the broader perspective these studies provide insight into an important neural crest development pathway. Studies of RET expression in developing embryos have shown RET expression in neural crest derived from somites one to five. Between day 9 and 10.5 there is migration of RET positive cells from the neural crest into the developing spinal cord and gastrointestinal tract (see Fig. 40-3). It is possible to envision a tropic effect of GDNF interaction with the RET/GFRα-1 receptor system to entice neural crest cells into their normal developmental location. There is an evolving body of literature that GDNF, NRTN, ARTN, and PSPN function as survival factors. Indeed, GDNF was originally identified as a neurotrophic factor whose presence in a culture system prevented neuronal death. A substantial amount of data indicates that this receptor system functions in neurons and neuroendocrine cells to prevent apoptotic cell death. This is certainly evident in mice in which the GFL (GDNF, ATRN, PSPN, or NTRN), RET, or one of the four GFRα receptors have been deleted. In most of these single-gene deletion mice there are features that indicate a developmental loss of neurons at various developmental stages, arguing that this receptor system promotes neuronal survival. Postnatally, however, the receptor system appears to have a different set of functions that relate both to neuronal survival and specific neuronal function. The interaction of one of the four GFLs with the RET/GFRα family receptor activates a complex cascade of signaling systems. There are at least four different tyrosine phosphorylation sites on RET that could mediate downstream signaling (Fig. 40-4). There is a general agreement that Tyr 1062 of RET activates both mitogen-activated protein kinase and phosphatidyl inositol-3-kinase pathways (Fig. 40-5). There is considerable evidence that tyrosine 1062 is necessary for the neoplastic transformation events found in MEN2. Substitution of another amino acid at codon diminishes the transformation efficiency of mutant RET. MUTATIONS OF THE RET PROTO-ONCOGENE ASSOCIATED WITH MTC The mutations of RET in MEN2 predominantly affect two domains within the RET tyrosine kinase receptor. Mutations of a cysteine-rich region of the extracellular domain are the most common mutations found in MEN2 and its

Figure 40-5 Transformation of the C cell by the mutant RET receptor requires phosphorylation of tyrosine 1062 and involves activation of at least two signaling pathways, PI 3 kinase and MAP kinase. There is evidence that glial cell-derived neurotrophic factor interaction with the RET/GDNF family receptor-α1 receptor system is not required for transformation caused by mutation of extracellular cysteine residues, but may be involved for mutations involving the tyrosine kinase domains.

Figure 40-6 Mutations of the RET proto-oncogene associated with hereditary medullary thyroid carcinoma. The most common mutations affect two regions of the RET tyrosine kinase. Mutations of an extracellular cysteine domain (codons 609, 611, 618, 620, 630, and 634), important for dimerization of the receptor, cause MEN2A and its variants. Mutations of the tyrosine kinase domain are found in MEN2B (codons 883 and 918) and a few families with FMTC (768, 804, 891, 912). Abbreviations: multiple endocrine neoplasia type 2A, MEN2A; familial medullary thyroid carcinoma, FMTC; MEN2A/cutaneous lichen amyloidosis, MEN2A/CLA; MEN2A/Hirschsprung disease variant, MEN2A/Hirschsprung; and multiple endocrine neoplasia type 2B, MEN2B.

variants (Fig. 40-6). Each of these mutations converts a conserved cysteine at codons 609, 611, 618, 620, 630, or 634 to another amino acid. Of all RET mutations, 75% affect codon 634 and a single mutation, cys634arg, accounts for >50% of all mutations

CHAPTER 40 / MULTIPLE ENDOCRINE NEOPLASIA TYPE 2

associated with hereditary MTC. Genotype–phenotype correlation shows that a codon 634 mutation is most commonly associated with classic MEN2A or Sipple’s syndrome. Mutations of the codons found in exon 10 (609, 611, 618, 620) are most commonly found in FMTC. Despite these correlative efforts, any of the extracellular cysteine mutations can be associated with MEN2A or FMTC, a point of importance in the clinical management of kindreds with these mutations. A clinician should not conclude that an individual with an exon 10 mutation will not develop a pheochromocytoma unless there are multiple generations within a family that have not developed pheochromocytoma. All patients with the MEN2A/CLA syndrome have codon 634 mutations. MEN2 can also be associated with mutations of codon 790, 791, V804L, and 891. Germline mutations of the tyrosine kinase domain are found in MEN2B (codons 883 and 918) and FMTC (codons 768, V804M, and 912) (see Fig. 40-6). Of patients with MEN2B, 98% have the M918T substitution; a handful of MEN2B patients with the codon 883 mutation have been identified. The mutation in the remaining 1–2% has not been identified. In the handful of kindreds with the codon 533, 768, V804M, and 912 mutations the only clinical phenotype is familial MTC. Additional evidence for the oncogenicity of these mutations came from studies in which expression of mutant RET proto-oncogene in NIH 3T3 cells caused transformation. These studies showed that codon 634 mutation resulted in dimerization of the RET tyrosine kinase receptor in the absence of its ligand, leading to autophosphorylation and phosphorylation of downstream proteins. Transfection of the RET cDNA containing the met918thr mutation also causes transformation, although dimerization of the receptor does not occur and a different set of substrate proteins are phosphorylated. MUTATIONS OF THE RET PROTO-ONCOGENE ASSOCIATED WITH HIRSCHSPRUNG DISEASE Two independent lines of analysis led to the identification of mutations of the RET proto-oncogene in patients with Hirschsprung disease. The identification of a chromosome 10 deletion in a child with Hirschsprung disease led to mapping studies in familial Hirschsprung disease that localized the causative gene to proximal chromosome 10q. Subsequent investigations identified inactivating and presumed activating mutations of the RET tyrosine kinase in Hirschsprung disease. Analysis of kindreds in which MEN2A and Hirschsprung disease cosegregate (see Table 40-1) have identified codon 609, 618, or 620 mutations of the RET proto-oncogene. Mutations of the endothelin-β receptor gene and its ligand, endothelin-3, have also been identified in Hirschsprung disease. Mutation analysis is readily available from several commercial sources (www.genetests.org). Several techniques have been applied to detection of mutations including direct DNA sequencing and restriction analysis for specific mutations and denaturing gradient gel electrophoresis. Direct DNA sequencing has become the detection procedure of choice because it permits rapid sequencing of the seven exons known to be involved in MEN2. There are a few caveats. Genetic testing like most other forms of laboratory analysis will be associated with a small error rate caused by sample mix-up, contamination of the PCR reaction, failure of specific primers to amplify the mutant allele, or technician copying errors, collectively estimated in the range of 5%. MUTATIONS OF THE RET PROTO-ONCOGENE IN SPORADIC MTC Somatic met918thr mutations of the RET proto-oncogene have been identified in approx 25% of sporadic

397

MTCs. Evidence indicates that tumors with this particular mutation may pursue a more aggressive clinical course. It is important to consider RET proto-oncogene analysis in patients with apparent sporadic MTC because a compilation of available information indicates that 5–7% of patients with apparent sporadic MTC have germline mutations of the RET proto-oncogene.

MANAGEMENT/TREATMENT MANAGEMENT OF MTC Negative Genetic Test Results The availability of reliable genetic testing makes it possible to identify gene carriers with near certainty. Children with negative test results can be excluded from further screening studies. It is important to repeat the analysis on more than one blood sample to be certain of the test results. This is especially true of individuals who have a normal RET analysis and will receive no further screening. Positive Genetic Test Results Clinical decision making based on RET proto-oncogene testing has evolved over the past several years. In an international conference focused on MEN there was consensus that genetic testing should form the basis for a decision for thyroidectomy. There was a general agreement that children with the highest risk mutations (codon 883, 918 associated with MEN2B) should have thyroidectomy performed shortly after birth. Those with high-risk codons (611, 618, 620, 630, 634, and 891) should be considered for thyroidectomy around the age of 5 yr, the earliest age at which metastasis has been identified in a child with one of these mutations. Others have argued for earlier thyroidectomy based on the finding of microscopic MTC (without evidence of metastasis) in younger children. In children with lowrisk RET mutations (codons 790, 791, 768, 804, and 912), there are large kindreds in which there has never been a death caused by MTC. In these kindreds physicians are unwilling to recommend and parents are unwilling to accept a recommendation for early thyroidectomy. In patients with these mutations, thyroidectomy should be performed at some later age, perhaps between 10 and 15 yr of age. Some continue to perform pentagastrin or calcium testing for calcitonin and perform a thyroidectomy at the time of a positive test result. It is unclear whether earlier thyroidectomy would prevent the recurrences found in 15–20% of children thyroidectomized based on calcitonin testing in earlier series. A report of the European experience suggests that earlier thyroidectomy would be beneficial; in this large series metastasis to local lymph nodes was not observed until the mid-teenage years (although it is not clear how many lymph nodes were examined in each patient), suggesting earlier thyroidectomy may improve outcomes. At this point clinicians participating in the decision to perform early thyroidectomy should balance the risks and benefits of the procedure in each age group. The completeness of thyroidectomy may be important. It is difficult to perform a total thyroidectomy without damage to blood vessels associated with the posterior capsule of the thyroid gland. Such damage can cause a higher incidence of hypoparathyroidism, a situation most surgeons choose to avoid. A few normal C cells in residual thyroid tissue attached to the posterior capsule may be of little concern in a 40- to 50-yr-old patient, but there is the real possibility that a few normal cells expressing a mutant RET kinase in a 5-yr-old child may transform over a severaldecade period of follow-up. This concern has led some clinicians to recommend a total thyroidectomy including the posterior capsule, central lymph node dissection, and transplantation of

398

SECTION IV / ENDOCRINOLOGY

parathyroid tissue to the nondominant arm. Whether this approach will improve the already excellent long-term cure rate for this disease is unclear, but seems worthy of investigation. MANAGEMENT OF PHEOCHROMOCYTOMA The advent of genetic testing will have little impact on the management of pheochromocytoma in MEN2 other than to exclude 50% of family members with normal genetic tests from further screening. Patients with pheochromocytomas should have unilateral or bilateral adrenalectomy or a cortical sparing adrenalectomy, dependent on the specific clinical situation. Issues related to unilateral or bilateral adrenalectomy in this disease have been discussed in Chapter 38 and are not reviewed in this chapter. Patients with abnormal catecholamine secretion should have adrenal surgery performed prior to consideration of thyroidectomy. MANAGEMENT OF HYPERPARATHYROIDISM Hyperparathyroidism occurs infrequently in children who have received prophylactic thyroidectomy for MTC. In one series there have been no cases of hyperparathyroidism in children who received thyroidectomy at a mean age of 13 yr with a mean follow-up period in excess of 20 yr. There is debate about the appropriate management of parathyroid neoplasia in older patients. Either subtotal parathyroidectomy or total parathyroidectomy with transplantation of parathyroid tissue to the nondominant forearm has been advocated.

FUTURE DIRECTIONS Hereditary MTC is a rare disorder, but the specific molecular defect that causes this neoplastic syndrome provides a useful model in which to study strategies for inactivation of mutant RET tyrosine kinase activity. The fact that a single gene defect causes three different neoplastic manifestations makes it an interesting model for study. There are several small organic molecule tyrosine kinase inhibitors with activity against the RET kinase domain in phase I trials. The prolonged period during which transformation occurs makes MEN2 and hereditary MTC excellent models in which to assess the impact that prophylactic treatment with a drug designed to prevent RET activation would have on the natural history of the disease.

SELECTED REFERENCES Airaksinen MS, Saarma M. The GDNF family: signalling, biological functions and therapeutic value. Nat Rev Neurosci 2002;3:383–394. Angrist M, Bolk S, Thiel B, et al. Mutation analysis of the RET receptor tyrosine kinase in Hirschsprung’s disease. Hum Mol Genet 1995;4:821–830. Asai N, Iwashita T, Matsuyama M, et al. Mechanism of activation of the ret proto-oncogene by multiple endocrine neoplasia 2A mutations. Mol Cell Biol 1995;15:1613–1619. Berndt I, Reuter M, Saller B, et al. A new hot spot for mutations in the ret protooncogene causing familial medullary thyroid carcinoma and multiple endocrine neoplasia type 2A. J Clin Endocrinol Metab 1998;83: 770–774. Boccia LM, Green JS, Joyce C, et al. Mutation of RET codon 768 is associated with the FMTC phenotype. Clin Genet 1997;51:81–85. Bolino A, Schuffenecker I, Luo Y, et al. RET mutations in exons 13 and 14 of FMTC patients. Oncogene 1995;10:2415–2419. Borst MJ, Van Camp JM, Peacock ML, et al. Mutational analysis of multiple endocrine neoplasia type 2A associated with Hirschsprung’s disease. Surgery 1995;117:386–391. Brandi ML, Gagel RF, Angeli A, et al. Guidelines for diagnosis and therapy of MEN type 1 and type 2. J Clin Endocrinol Metab 2001;86:5658–5671. Cance WG, Wells SA Jr. Multiple endocrine neoplasia type IIa. Curr Probl Surg 1985;22:1–56. Carney JA, Go VL, Sizemore GW, et al. Alimentary-tract ganglioneuromatosis. A major component of the syndrome of multiple endocrine neoplasia, type 2b. N Engl J Med 1976;295:1287–1291.

Dang GT, Cote GJ, Schultz PN, et al. A codon 891 exon 15 RET protooncogene mutation in familial medullary thyroid carcinoma: A detection strategy. Mol Cell Probes 1999;13:77–79. DeLellis RA, Wolfe HJ, Gagel RF, et al. Adrenal medullary hyperplasia. A morphometric analysis in patients with familial medullary thyroid carcinoma. Am J Pathol 1976;83:177–196. Donis-Keller H, Dou S, Chi D, et al. Mutations in the RET proto-oncogene are associated with MEN2A and FMTC. Hum Mol Genet 1993;2: 851–856. Edery P, Lyonnet S, Mulligan LM, et al. Mutations of the RET protooncogene in Hirschsprung’s disease. Nature 1994;367:378–380. Eisenhofer G, Walther MM, Huynh TT, et al. Pheochromocytomas in von Hippel-Lindau syndrome and multiple endocrine neoplasia type 2 display distinct biochemical and clinical phenotypes. J Clin Endocrinol Metab 2001;86:1999–2008. Eng C, Clayton D, Schuffenecker I, et al. The relationship between specific RET proto-oncogene mutations and disease phenotype in multiple endocrine neoplasia type 2. International RET mutation consortium analysis. JAMA 1996;276:1575–1579. Farndon JR, Leight GS, Dilley WG, et al. Familial medullary thyroid carcinoma without associated endocrinopathies: a distinct clinical entity. Br J Surg 1986;73:278–281. Gagel RF, Levy ML, Donovan DT, et al. Multiple endocrine neoplasia type 2a associated with cutaneous lichen amyloidosis. Ann Intern Med 1989;111:802–806. Gagel RF, Marx S. Multiple endocrine neoplasia. In: Larsen PR, Kronenberg H, Melmed S, Polonsky KS, eds. Williams Textbook of Endocrinology. Philadelphia: WB Saunders, 2003:1717–1762. Gagel RF, Tashjian AH Jr, Cummings T, et al. The clinical outcome of prospective screening for multiple endocrine neoplasia type 2a: an 18-year experience. N Engl J Med 1988;318:478–484. Grieco M, Santoro M, Berlingieri MT, et al. PTC is a novel rearranged form of the ret proto-oncogene and is frequently detected in vivo in human thyroid papillary carcinomas. Cell 1990;60:557–563. Hofstra RM, Fattoruso O, Quadro L, et al. A novel point mutation in the intracellular domain of the ret protooncogene in a family with medullary thyroid carcinoma. J Clin Endocrinol Metab 1997;82:4176–4178. Hofstra RM, Landsvater RM, Ceccherini I, et al. A mutation in the RET proto-oncogene associated with multiple endocrine neoplasia type 2B and sporadic medullary thyroid carcinoma. Nature 1994;367:375–376. Jhiang SM, Sagartz JE, Tong Q, et al. Targeted expression of the ret/PTC1 oncogene induces papillary thyroid carcinomas. Endocrinology 1996;137:375–378. Jing S, Wen D, Yu Y, et al. GDNF-induced activation of the Ret protein tyrosine kinase is mediated by GDNFR-a, a novel receptor for GDNF. Cell 1996;85:1113–1124. Kakudo K, Carney JA, Sizemore GW. Medullary carcinoma of thyroid. Biologic behavior of the sporadic and familial neoplasm. Cancer 1985; 55:2818–2821. Lee JE, Curley SA, Gagel RF, et al. Cortical-sparing adrenalectomy for patients with bilateral pheochromocytoma. Surgery 1996;120:1064–1071. Lin LF, Doherty DH, Lile JD, et al. GDNF: a glial cell line-derived neurotrophic factor for midbrain dopaminergic neurons. Science 1993;260: 1130–1132. Mathew CG, Chin KS, Easton DF, et al. A linked genetic marker for multiple endocrine neoplasia type 2A on chromosome 10. Nature 1987; 328:527–528. Moore MW, Klein RD, Farinas I, et al. Renal and neuronal abnormalities in mice lacking GDNF. Nature 1996;382:76–79. Mulligan LM, Eng C, Attie T, et al. Diverse phenotypes associated with exon 10 mutations of the RET proto-oncogene. Hum Mol Genet 1994; 3:2163–2167. Mulligan LM, Kwok JB, Healey CS, et al. Germ-line mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A. Nature 1993;363:458–460. Nilsson O, Tisell LE, Jansson S, et al. Adrenal and extra-adrenal pheochromocytomas in a family with germline RET V804L mutation. JAMA 1999;281:1587, 1588. Pachnis V, Mankoo B, Costantini F. Expresssion of the c-ret proto-oncogene during mouse embryogenesis. Development 1993;119:1005–1017.

CHAPTER 40 / MULTIPLE ENDOCRINE NEOPLASIA TYPE 2

Pichel JG, Shen L, Sheng HZ, et al. Defects in enteric innervation and kidney development in mice lacking GDNF. Nature 1996;382:73–76. Romeo G, Ronchetto P, Luo Y, et al. Point mutations affecting the tyrosine kinase domain of the RET proto-oncogene in Hirschsprung’s disease. Nature 1994;367:377, 378. Sanchez M, Silos-Santiago I, Frisen J, et al. Newborn mice lacking GDNF display renal agenesis and absence of enteric neurons, but no deficits in midbrain dopaminergic neurons. Nature 1996;382:70–73. Santoro M, Carlomagno F, Romano A, et al. Activation of RET as a dominant transforming gene by germline mutations of MEN2A and MEN2B. Science 1995;267:381–383. Schilling T, Burck J, Sinn HP, et al. Prognostic value of codon 918 (ATG→ACG) RET proto-oncogene mutations in sporadic medullary thyroid carcinoma. Int J Cancer 2001;95:62–66. Schuchardt A, D’Agati V, Larsson-Blomberg L, et al. Defects in the kidney and enteric nervous system of mice lacking the tyrosine kinase receptor Ret. Nature 1994;367:380–383. Simpson NE, Kidd KK, Goodfellow PJ, et al. Assignment of multiple endocrine neoplasia type 2A to chromosome 10 by linkage. Nature 1987;328:528–530. Sipple JH. The association of pheochromocytoma with carcinoma of the thyroid gland. Am J Med 1961;31:163–166. Skinner MA, De Benedetti MK, Moley JF, et al. Medullary thyroid carcinoma in children with multiple endocrine neoplasia types 2A and 2B. J Pediatr Surg 1996;31:177–182. Steiner AL, Goodman AD, Powers SR. Study of a kindred with pheochromocytoma, medullary carcinoma, hyperparathyroidism and Cushing’s disease: Multiple endocrine neoplasia, type 2. Medicine 1968;47:371–409. Takahashi M, Cooper GM. ret transforming gene encodes a fusion protein homologous to tyrosine kinases. Mol Cell Biol 1987;7:1378–1385.

399

Takahashi M, Ritz J, Cooper GM. Activation of a novel human transforming gene, ret, by DNA rearrangement. Cell 1985;42:581–588. Trupp M, Arenas E, Falnzilber M, et al. Functional receptor for GDNF encoded by the c-ret proto-oncogene. Nature 1996;381:785–788. Tsuzuki T, Takahashi M, Asai N, et al. Spatial and temporal expression of the ret proto-oncogene product in embryonic, infant and adult rat tissues. Oncogene 1995;10:191–198. Vasen HFA, van der Feltz M, Raue F, et al. The natural course of multiple endocrine neoplasia type IIb: A study of 18 cases. Arch Intern Med 1992;152:1250–1252. Verdy MB, Cadotte M, Schurch W, et al. A French Canadian family with multiple endocrine neoplasia type 2 syndromes. Henry Ford Hosp Med J 1984;32:251–253. Wells SA Jr, Chi DD, Toshima K, et al. Predictive DNA testing and prophylactic thyroidectomy in patients at risk for multiple endocrine neoplasia type 2A. Ann Surg 1994;220:237–250. Williams ED, Brown CL, Doniach I. Pathological and clinical findings in a series of 67 cases of medullary carcinoma of the thyroid. J Clin Pathol 1966;19:103–113. Wohllk N, Cote GJ, Bugalho MMJ, et al. Relevance of RET proto-oncogene mutations in sporadic medullary thyroid carcinoma. J Clin Endocrinol Metab 1996;81:3740–3745. Web site for localizing sites for genetic testing: www.genetests.org. Zedenius J, Larsson C, Bergholm U, et al. Mutations of codon 918 in the RET proto-oncogene correlate to poor prognosis in sporadic medullary thyroid carcinomas. J Clin Endocrinol Metab 1995;80: 3088–3090. Zedenius J, Wallin G, Hamberger B, et al. Somatic and MEN2A de novo mutations identified in the RET proto-oncogene by screening of sporadic MTCs. Hum Mol Genet 1994;3:1259–1262.

41 Disorders of Sex Determination and Differentiation CHARMIAN A. QUIGLEY SUMMARY Abnormalities of sex determination and differentiation comprise two major clinical groups: disorders of gonadal development (i.e., disorders of sex determination: sex reversal, true hermaphroditism), with secondary effects on genital development, and defects of genital development in the presence of normal gonads (disorders of sex differentiation: male and female “pseudohermaphroditism”). This chapter will review the embryology of normal sex determination and differentiation, the genetic basis of normal and abnormal sex determination and differentiation, and the approaches to diagnosis of disorders of sex determination and differentiation. Key Words: Adrenogenital primordium; androgen insensitivity syndrome; congenital adrenal hyperplasia; gonadoblastoma; granulosa cell; hermaphroditism; labioscrotal fold; Leydig cell; Leydig cell hypoplasia; mosaicism; ovary determination; persistent Müllerian duct syndrome; primordial gonad; pseudohermaphroditism; Sertoli cell; sex differentiation; sex reversal; testis determination; theca cell; urogenital sinus.

INTRODUCTION Sex determination and differentiation are distinct, consecutive processes that follow the establishment of chromosomal sex at the time of gamete fertilization, subsequently requiring the coordinated expression of a specific set (or sets) of genes in a strict spatiotemporal manner. The term sex determination (alternatively called primary sex differentiation) refers to the development of gonadal sex––a process that occurs at approx 6–7 wk gestation in the human male fetus, and approx 10–11 wk gestation in the female fetus. As generally used, the term sex differentiation refers to the processes downstream of gonadal development––those regulated by gonadal secretions or lack thereof (also called secondary sex differentiation). In essence, the genetic complement endowed at fertilization determines gonad type and the latter determines the pattern of differentiation of the internal genital ducts and external genitalia. GENERAL CLINICAL FEATURES Abnormalities of sex determination and differentiation consists of two major clinical groups: disorders of gonadal development (i.e., disorders of sex determination: sex reversal, true hermaphroditism), with secondary effects on From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

genital development, and defects of genital development in the presence of normal gonads (disorders of sex differentiation: male and female pseudohermaphroditism). The disorders of sex determination and differentiation are summarized in Table 41-1. Approaches to diagnosis and management are discussed at the conclusion of the chapter. Sex Reversal and True Hermaphroditism The terminology is sometimes confusing: sex reversal refers to the condition in which the individual’s genetic sex opposes the gonadal (and therefore generally the phenotypic) sex. These are individuals with 46,XY karyotype who have no testes (in their place usually are streak gonads) and have a female (or ambiguous) phenotype, and those with 46,XX karyotype who have testes and varying degrees of genital masculinization. The archetypal and most common form of sex reversal is 46,XX maleness (approx 1 in 20,000 men has 46,XX karyotype). The phenotype of most 46,XX males is similar to that seen in Klinefelter syndrome (47,XXY): structurally normal (sometimes cryptorchid) testes are present and the internal and external genitalia are male. Testicular size and histology are normal in infancy. Pubertal virilization occurs to a greater or lesser degree, but like those with Klinefelter syndrome, affected postpubertal individuals have small testes and are azoospermic. Up to 20% of 46,XX males have subnormal masculinization, manifest by cryptorchidism, hypospadias or frank genital ambiguity. The testes have atrophic, hyalinized, seminiferous tubules and Leydig cell hyperplasia. The absence of spermatogenesis in 46,XX males may relate two factors, the presence of the extra X-chromosome, and the absence of Y-chromosomal genes involved in spermatogenesis. 46,XX males are taller than average for females, but shorter than 46,XY males, presumably because of absence of stature-determining genes located on Yq. There are two main subtypes of 46,XX maleness (and 46,XX true hermaphroditism, described later)—XXY+ and XXY–. The majority of patients with complete 46,XX maleness are XXY+, resulting from translocation of all or part of the distal end of the Y-chromosome (Yp), containing the sex-determining region of the Y-chromosome (SRY), to the short arm of an X-chromosome (Xp) during paternal gamete meiosis. In contrast, the majority of patients with 46,XX true hermaphroditism are XXY–. The molecular/ phenotypic inference from these findings is that the greater the amount of Y-chromosome material, the more complete the masculinization. Two specific hot spots for Yp-Xp recombination within areas of high X-Y sequence homology

400

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

401

Table 41-1 Genetic Disorders of Sex Determination and Differentiation Disorders of sex determination (development of gonad discordant from karyotype) 46,XX sex reversal/true hermaphroditism (46,XX karyotype with testes or ovotestes) XXY+ (SRY translocation to terminal Xp) XXY– (probable mutation of downstream regulator of testis development) 46,XY sex reversal/true hermaphroditism (46,XY karyotype with ovaries, ovotestes or streak gonads) SF1 deletion/mutation (homozygous, heterozygous) WT1 deletion/mutation (heterozygous) SRY deletion/mutation (Y-linked) SOX9 mutation (heterozygous) ATRX mutation (X-linked) DHH mutation (homozygous) Monosomy 9p (possible haploinsufficiency of DMRT1/2) DSS (duplication) WNT4 (duplication) Disorders of sex differentiation (development of phenotype discordant from gonad) 46,XY pseudohermaphroditism (46,XY karyotype with testes and female or ambiguous internal or external genitalia) Impaired testosterone production LHCGR mutation (homozygous, compound heterozygous) Defects of testosterone biosynthesis StAR mutation/deletion (homozygous, compound heterozygous) POR mutation (homozygous, heterozygous, compound heterozygous) CYP11A mutation (heterozygous, compound heterozygous) HSD3B2 mutation (homozygous, compound heterozygous) CYP17 mutation (homozygous, compound heterozygous) HSD17B3 mutation (homozygous, compound heterozygous) Impaired androgen response SRD5A2 deletion/mutation (homozygous, compound heterozygous) AR gene deletion/mutation (hemizygous) Impaired anti-Müllerian hormone production or action AMH mutation (homozygous, compound heterozygous) AMHR2 mutation (homozygous, compound heterozygous) 46,XX pseudohermaphroditism (46,XX karyotype with ovaries male or ambiguous internal or external genitalia) Fetal androgen excess Congenital adrenal hyperplasia POR mutation (heterozygous, compound heterozygous, homozygous) HSD3B2 mutation (homozygous, compound heterozygous) CYP21 mutation (homozygous, compound heterozygous) CYP11B1 mutation (homozygous, compound heterozygous) Aromatase deficiency CYP19 mutation (homozygous, compound heterozygous) Disorders affecting Müllerian structures HOXA13 mutation (heterozygous) Mayer-Rokitansky-Kuster-Hauser syndrome (unknown etiology) AMH, anti-Müllerian hormone; AMHR, anti-Müllerian hormone receptor; AR, androgen receptor; ATRX, α-thalassemia/mental retardation, X-linked gene; CG, chorionic gonadotropin; DHH, desert hedgehog; DSS, dosage-sensitive sex reversal; HOXA13, homeobox A 13; LH, luteinizing hormone; LHCGR, luteinizing hormone/chorionic gonadotropin receptor; POR, P450 oxidoreductase; SF1, steroidogenic factor 1; SOX, Sryrelated homeobox gene; SRY, sec-determining region of the Y-chromosome; StAR, steroidogenic acute regulatory; WNT, wingless-type MMTV integration site family member; WT1, Wilms tumor 1.

have been reported to account for >50% of such recombination events. However, many 46,XX males have no definable genetic change and because approx 10% of 46,XX males with testes are completely negative for all Y-encoded sequences, it is likely that non Y sequences are responsible for testis determination in these cases. For example, female-to-male sex reversal has been found in association with duplication of the region of chromosome 17 containing the SRY-related homeobox gene (SOX) 9. The converse of 46,XX maleness is complete gonadal dysgenesis in a 46,XY female (Swyer syndrome; XY gonadal dysgenesis),

however, it is much less common, occurring in only 1/100,000 females. Affected individuals have streak gonads with female internal and external genitalia, although significant phenotypic variation is found within affected families. Partial testicular dysgenesis is associated with genital development that essentially reflects the functional state of the gonads at the critical period of sex differentiation. Because of deficient estrogen production, breast development is poor and 46,XY females often present with delayed puberty or primary amenorrhea, accompanied by gonadotropin concentrations in the castrate range. Pubic hair is

402

SECTION IV / ENDOCRINOLOGY

usually present. There is a high incidence of gonadal neoplasia (gonadoblastomas and germinomas). Probably because of the presence of Y-chromosomal stature-determining genes, affected women are of normal to tall stature compared with 46,XX females. Individuals with deletion of Yp (including the SRY gene) in addition to the general features described earlier, may have certain features of Turner syndrome such as lymphedema, likely because of haploinsufficiency of a gene or genes on Xp. Male-to-female sex reversal has been reported in association with mutations of genes described in the following sections, including Wilms tumor 1 (WT1), steroidogenic factor (SF) 1, α-thalassemia/mental retardation, X-linked gene (ATRX), SRY and SRY homeobox-like gene 9 (SOX9) and also occurs in association with deletions of the region of chromosome 9 in which doublesex/ mab-related transcription factors (DMRT1 and DMRT2) reside, and with duplication of the chromosomal regions in which dosagesensitive sex reversal adrenal hypoplasia congenita locus on the X chromosome, gene 1 (DAX1; X chromosome) and wingless-type MMTV integration site family (WNT4; chromosome 1) reside. True hermaphroditism defines the condition in which, regardless of karyotype, there is coexistence of ovarian and testicular tissue in the same individual, either an ovary on one side and a testis on the other (~20% of patients), bilateral ovotestes (~30%) or one ovotestis and one testis or ovary (~50%). The testicular portions of the gonads are dysgenetic, with interstitial fibrosis and rare or absent spermatogonia, whereas the ovarian portions are histologically normal. Malignant degeneration of the gonad is reported in approx 5% of cases. Most affected invididuals have ambiguous genitalia, however, the phenotypic spectrum is broad: for example, a fully masculinized boy with bilaterally descended ovotestes has been reported. A hallmark, although not universal feature, is asymmetric genital development—Müllerian structures (that may include a full or hemiuterus) on one side, and Wolffian structures on the other; a gonad-containing hemiscrotum on one side (more often the right) with a flat, empty labium majorum on the other. In general the pattern of development reflects the predominant functional nature of the gonad on the ipsilateral side. In association with an ovotestis, the internal genitalia may show elements of both Müllerian and Wolffian origin on the same side. The pattern of pubertal development reflects the function of the gonads and fertility is not uncommon. Of affected individuals with true hermaphroditism, approx 70% have a 46,XX karyotype and some cases may represent a variant form of 46,XX maleness. Without thorough histological evaluation of the gonads it may be impossible to distinguish an undervirilized 46,XX male from a 46,XX true hermaphrodite. In addition, true hermaphroditism and complete sex reversal can coexist within the same family, so these disorders may be phenotypic variations of a certain genotype. The majority of patients with 46,XX true hermaphroditism are negative for any detectable Y-chromosomal sequences. About 20% of affected individuals have chromosomal mosaicism for 46,XX/46,XY or less commonly 46,XX/47,XXY; a small number of cases have a pure 46,XY karyotype and a few with 47,XXY karyotype have been reported. Although the molecular basis of this condition is unknown, genetic causes are implicated by the finding of familial cases of the disorder. Pseudohermaphroditism Pseudohermaphroditism (either male [46,XY] or female [46,XX]) refers to conditions in which the karyotype and gonad are congruous, however, there is a discrepancy between the gonadal sex and the phenotypic sex: individuals with

46,XY karyotype and testes whose phenotype is female or ambiguous (e.g., disorders of androgen production or action) and those with 46,XX karyotype and ovaries whose phenotype is male or masculinized (e.g., disorders resulting in excess androgen production, such as congenital adrenal hyperplasia [CAH]). Individuals of either karyotypic sex with these disorders have phallic development that ranges from a diminutive clitoris in a 46,XY individual to a completely formed penis in a 46,XX individual; the labioscrotal region may be fully fused and rugose, or bifid and smooth; the internal structures may be mainly female-type (Müllerian, in the absence of anti-Müllerian hormone [AMH] action), mainly male-type (Wolffian, in the presence of local androgen action), or a combination of the two. The internal and external genital morphology essentially reflects fetal production of, and response to, androgens and AMH during the critical period of gestation. OVERVIEW OF MOLECULAR PATHOPHYSIOLOGY Because of the multiplicity of enzymes, hormones, receptors, and transcription factors involved, there are numerous opportunities for the usually well-coordinated processes of sex determination and differentiation to go awry. Some general principles are worth elucidating before delving into the specific disorders: 1. Hormones have no intrinsic action and must act through specific receptors. Thus hormone deficiencies are typically manifest by lack of function of the corresponding receptor. In general, steroid hormone deficiencies, resulting from defects of genes encoding the biosynthetic enzymes responsible for generation of the respective hormones, typically manifest only in the presence of two defective gene alleles and are inherited as autosomal-recessive traits. 2. Receptors that mediate hormone action comprise two broad categories. (1) Membrane-associated receptors act as transducers of the hormone signal. Binding of the hormone (typically a peptide) to the extracellular domain of such a receptor sets off an intracellular signal cascade in which a downstream factor or second messenger eventually influences gene transcription. (2) Intracellular receptors, such as those for the steroid and thyroid hormones, function as nuclear transcription factors themselves by entering the nucleus and binding directly with target genes to modify their transcription, usually in a ligand-dependent fashion. A subclass of the nuclear transcription factor family referred to as orphan receptors (e.g., DAX1) function in the absence of known ligand. In addition, other nonreceptor transcription factors of different classes and families also function in a ligand-independent manner (e.g., SRY, SOX9, WT1). 3. Defects of membrane-associated transducer-type receptors that mediate peptide hormone action, such as the luteinizing hormone/chorionic gonadotropin receptor (LHCGR), cause an increase or a decrease in the activity of the system, depending on whether the mutation is an activating or an inhibitory one. This is illustrated by the contrasting effects of mutations in the LHCGR: inhibitory mutations cause the syndrome of leydig cell hypoplasia (LCH, see LHCG Receptor), whereas activating mutations are associated with the syndrome of familial male precocious puberty (see Chapter 43). Defects of these receptors are expressed only in the presence of two defective gene alleles. 4. In the intracellular receptor system, absence of ligand or defective binding of ligand to the receptor translates to

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

absent or reduced activity of the ligand-dependent transcription factor; complete absence of a transcription factor (e.g., when the encoding gene is deleted) results complete loss of target gene transcription. Mutations of the DNAbinding regions of these receptors alter protein-DNA interactions between transcription factors and their target genes, potentially resulting in decreased, increased or sometimes promiscuous transcriptional activation, the latter resulting from loss of DNA-binding specificity. In the case of a heterozygous mutation, the abnormal protein produced from the mutant allele may interfere with the action of the normal protein produced from the wild-type allele of the gene—a “dominant negative” effect. In this situation, the mutant protein may block access of the normal factor to its target DNA, by forming inactive dimers that are unable to bind the target DNA sequence, or sequester other critical transcription factors because of disturbed protein–protein interactions. Defects in genes encoding nuclear transcription factors may manifest as dominant conditions, with dysfunction occurring in the presence of a single mutant allele (e.g., thyroid hormone receptor), or as recessive conditions in which both alleles must be mutated to cause disease (e.g., vitamin D receptor). Defects of this receptor class manifest clinically as hormone resistance syndromes, because hormone action is diminished in the presence of increased hormone levels.

EMBRYOLOGY OF NORMAL SEX DETERMINATION AND DIFFERENTIATION Normal development of the gonads and genitalia has three major phases. First, in the earliest stages of gestation, the fetus develops a bipotential gonad, two sets of embryonic internal genitalia, and undifferentiated female-like external genitalia. The next step is the differentiation of the bipotential gonads into either ovaries or testes and the final phase is the differentiation and development of the internal and external genital primordia along male or female lines, depending on the nature of the hormonal products of the gonads (or lack thereof).

PHASE 1: DEVELOPMENT OF THE PRIMORDIAL STRUCTURES GONADS At approx 5 wk of human gestation, the intermediate mesoderm in the area that will become the kidney, adrenal, and gonad condenses into distinct regions. As development progresses, the urogenital ridges form on the dorsal wall of the body cavity. The urogenital ridge consists of the mesonephros (the forebear of the primitive kidney) located laterally, and the genital ridge, which will become the primitive gonad, located medially. Development of the genital ridge is accompanied by thickening and proliferation of the coelomic epithelium, which penetrates the underlying mesenchyme to form the primitive sex cords. Blood vessels grow into the developing gonad from the mesonephros, subsequently developing in a sexually dimorphic manner. The primitive genital or gonadal ridges initially contain no germ cells. Between weeks 5 and 6 of gestation, primordial germ cells migrate from the endoderm of the yolk sac along the dorsal mesentery of the hindgut into the indifferent gonad, invading the developing primary sex cords. At this stage the gonad consists of an outer cortex and inner medulla, and no morphological difference between the gonads of male and female fetuses can be detected until approx 7

403

wk gestation, at which time testicular development begins in the male fetus. INTERNAL GENITALIA Both the urinary and genital components of the urogenital system are derived, to a large extent, from the intermediate mesoderm, which becomes segmented into units termed nephrotomes. The lateral portions of the nephrotomes unite, forming a longitudinal duct on each side of the embryo, the mesonephric duct (later to become the Wolffian duct), by about week 4 of gestation. At approx 6 wk, the paramesonephric duct (to become the Müllerian duct) forms from the epithelium on the surface of the urogenital ridge, such that by 6 wk, both male and female fetuses are endowed with two sets of internal duct structures. EXTERNAL GENITALIA Just as the gonads and internal genitalia are indistinguishable between the sexes for the first few weeks of life, so it is for the external genital primordia (Fig. 41-1). In the fourth week of gestation the external genitalia of both sexes are represented simply by a midline protuberance––the genital tubercle. By week 6 (still indifferent), two medial folds, the urethral folds, flank the urogenital groove and two larger folds, the labioscrotal folds, are present laterally.

PHASE 2: SEX DETERMINATION (DEVELOPMENT OF TESTIS OR OVARY) TESTIS DEVELOPMENT Soon after the germ cells arrive, the gonad in the 46,XY fetus begins to differentiate; the first histologically discernible event in testis development is the appearance of primordial Sertoli cells, which differentiate from somatic cells of the coelomic epithelium at approx 7 wk gestation. The Sertoli cells proliferate, aggregate around the primitive germ cells and align into cord-like structures (medullary sex cords) that subsequently become the seminiferous tubules. The seclusion of germ cells within the tubules prevents meiosis and commits the germ cells to spermatogenic development. The prevention of meiosis may be the key event that directs gonadal development away from the ovarian pathway. This organizational process appears to be regulated by the Sertoli cells themselves. Germ cells are not required for this process, because morphological testis development occurs in their absence. About 1 wk later (approx 8 wk), steroidogenic Leydig cells differentiate from primitive interstitial cells of mesonephric origin, likely controlled by paracrine influences from Sertoli cells, possibly AMH. Another key event is the differentiation of peritubular myoid cells, thought to derive from the same interstitial cell lineage as Leydig cells. These myoepithelial cells are required for the development of the testis cords––the defining event in testicular organogenesis. Timing is critical as there is only a limited window during which these events can occur. OVARY DEVELOPMENT In contrast with testis development, normal ovarian differentiation specifically requires the presence of germ cells. Without the germ cell seeding of the 46,XX gonadal primordium, the tissue degenerates into a nonfunctional, mainly fibrous “streak.” Prior to week 10 of gestation the only histological feature that distinguishes an ovary is the absence of testicular features. Thereafter, ovarian structure becomes distinguishable, with the regression of the primary medullary sex cords, which are replaced by a vascular stroma. Secondary cortical sex cords that provide the supporting structure for the germ cells develop close to the surface of the gonad under the influence of the germ cell lineage. Within the sex cords the primary germ cells undergo vigorous mitotic replication to become oogonia; then the sex cords break up

404

SECTION IV / ENDOCRINOLOGY

Figure 41-1 Differentiation of the external genitalia. (Top) At week 6 of gestation the external genitalia are undifferentiated, consisting of the midline genital tubercle, two medial urethral folds flanking the urogenital slit, and two larger labioscrotal swellings laterally. (Left) In the absence of androgen action the genital tubercle becomes the clitoris and the urogenital sinus remains patent. The vesicovaginal septum forms to separate the urethra from the vagina, located posteriorly. The urethral folds develop as the labia minora, whereas the labioscrotal folds enlarge slightly, remaining unfused, as the labia majora. (Right) In the male fetus in the presence of androgens (mainly DHT), the genital tubercle elongates to become the body of the penis, and the urethral folds fuse in the midline to form the penile urethra. The labioscrotal swellings fuse and enlarge to become the scrotum. These processes are completed approximately between weeks 8 and 12 of gestation.

into clusters, becoming the primordial follicles at about week 16. The primordial follicles contain the diploid (i.e., 46 chromosomes) primary oocytes, which after entering into the first stage of meiosis (reduction division) remain quiescent until puberty. Follicular (granulosa) cells arise from the same somatic cell lineage as Sertoli cells. Theca cells represent the ovarian counterpart of Leydig cells.

There is a notable difference in the chronology of testicular vs ovarian development, the process of testis formation being completed by 8 wk gestation, at which time the process of ovarian development has not yet begun. In fact, ovarian development is not completed until after most of the processes of phenotypic sex differentiation, described below, have occurred. This, and the fact that phenotypic development is normal even in complete absence

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

of the ovary, highlights the lack of involvement of the ovary in the processes of normal female genital development.

PHASE 3: SEX DIFFERENTIATION (DEVELOPMENT OF MALE OR FEMALE GENITALIA) INTERNAL GENITALIA: MALE The primitive internal genital tracts are indistinguishable between the sexes until 7 wk gestation. From week 8 of gestation, hormonal secretions from the fetal testes induce masculinization of the internal genital structures in the 46,XY fetus. Initially the Wolffian ducts are stabilized (prevented from undergoing resorption) by the action (mainly local) of testosterone; between 9 and 13 wk gestation they undergo differentiation into the epididymides, vasa deferentia and seminal vesicles. Dihydrotestosterone (DHT) does not appear to mediate these processes, because the enzyme required for its production (5-α reductase 2) is not expressed in Wolffian tissues at the time of their differentiation. In parallel to the masculinization of the internal genitalia represented by Wolffian development, a process of “defeminization” of the redundant set of genital ducts occurs as the Müllerian ducts regress under the influence of locally acting AMH secreted by Sertoli cells. The Müllerian ducts are obliterated by week 11 of gestation, the only remnant of their existence in the male being the prostatic utricle. Absence of one testis results in retention of Müllerian structures and only limited Wolffian development on the ipsilateral side, indicating that the effects of AMH and testosterone are largely mediated in a paracrine fashion. INTERNAL GENITALIA: FEMALE In the absence of testicular secretions, as in the normal 46,XX fetus, the inverse set of genital tract developmental processes occurs. Without local androgen action, the Wolffian ducts regress. Meanwhile, absence of AMH allows the Müllerian ducts to develop. Their upper portions form the fallopian tubes; the lower sections fuse and differentiate as the uterus and upper part of the vagina. The lower portion of the vagina derives from the urogenital sinus (the ventral part of the embryonic mammalian cloaca formed by the growth of a fold dividing the cloaca where the gut and allantois meet), which remains patent in the absence of androgen action. EXTERNAL GENITALIA: MALE Under the influence of androgen action (primarily DHT), the genital tubercle elongates to form the body of the penis, and the urethral folds fuse ventrally from behind forward, to form the penile urethra. The labioscrotal folds grow toward each other, fusing in the midline to form the scrotum. DHT also induces the urogenital sinus to differentiate as the prostate, and inhibits the formation of the vesico-vaginal septum. These processes are completed by week 12 of gestation (Fig. 41-1). Between 12 and 24 wk gestation the testes migrate from their original lumbar location to the level of the internal inguinal ring above the scrotum. Descent of the testes through the inguinal ring and into the scrotum begins around week 28, and in most infants is completed by term. EXTERNAL GENITALIA: FEMALE In the absence of significant androgen action, such as in the normal female fetus, the genital tubercle elongates only slightly to form the clitoris (Fig. 41-1). The urogenital sinus remains open and the vesico-vaginal septum (a fold of tissue that separates the posterior wall of the bladder from the anterior wall of the vagina) forms, so that the urethra opens anteriorly and the vagina posteriorly. The vestibule of the urogenital sinus is bordered laterally by the urethral folds, which do not fuse and instead develop as the labia minora. Further lateral, the labioscrotal swellings enlarge somewhat but also remain unfused, forming the

405

labia majora. There is minor fusion posteriorly, forming the posterior commissure, and anteriorly, producing the mons pubis. These events occur from about week 7–12 of gestation.

GENETIC BASIS OF NORMAL AND ABNORMAL SEX DETERMINATION AND DIFFERENTIATION A complex interplay of genes, transcription factors, hormones, and receptors is required for normal sex determination and differentiation (Table 41-2). The primary event governing the path of morphological sex differentiation is the development of the gonad. In 1947 the elegant experiments of Jost determined that “maleness” was a state imposed on the fetus that would otherwise develop as a phenotypic female, leading to the theory that development of the ovary and female phenotype occur when the fetus is not exposed to the influences of specific “maleness-determining” genes. However, newer information, indicates that rather than being a passive, “default” process, female development likely also requires activation of specific, perhaps opposing, gene pathways. From these findings the concept has developed that sex determination represents the primordial “Battle of the Sexes”*––the dominance of one set of gonad-specific genes over another. The key event determining the winner of the battle appears to be whether the primordial germ cells that colonize the indifferent gonad enter meiosis. Testis cord development, under the influence of a number of SOX genes during a narrow developmental window, arrests the primordial germ cells in mitosis; in the absence of testis cord development, germ cells enter meiosis and ovarian development ensues. The genes involved in the regulation of sex determination and sex differentiation reside on both the sex chromosome and the autosomes. Based on studies of sex determination in other species it has been hypothesized that the mammalian sex chromosomes evolved from a homologous autosome pair. The initial event in sex determination––the development of the primordial, bipotential gonad––depends on a network of interacting factors encoded by at least a dozen, and possibly scores of genes. Subsequent development of the testis or the ovary is exquisitely regulated by a team of cooperative transcriptional activators and repressors that selectively up- or downregulate the genes required for sex-specific gonadal differentiation. BIPOTENTIAL GONAD AND PRIMORDIAL INTERNAL GENITAL DUCTS There is a relationship between renal, adrenal, and gonadal cellular precursors that underlies both normal and abnormal development, because these tissues all arise from the same regions of the primitive mesoderm and coelomic epithelium known as adrenogenital primordium. The adrenal cortex derives from mesenchymal cells attached to the coelomic cavity lining adjacent to the urogenital ridge within the intermediate mesoderm; similarly, the steroid-producing cells of the gonads (Leydig and theca cells) differentiate from mesenchymal stem cells. The specific ontogeny of steroidogenic tissues has particular relevance for the understanding of the genetic regulation of gonadal development. Because these tissues have common cellular ancestors, it is not surprising that they share aspects of their genetic makeup. Thus the roles and responsibilities of a number of transcription factors involved in the early stages of renal, adrenal *This term was coined by Blanche Capel and colleagues in “The Battle of the Sexes: Opposing Pathways in Sex Determination” in The Genetics and Biology of Sex Determination, Novartis Foundation Symposium 2002;244: pp. 187–202.

Table 41-2 Factors Involved in Sex Determination and Differentiation Gene name or pseudonyms

Human gene locus

Human protein name

Protein type

Factors involved in primordial gonad/reproductive tract formation 9q33 SF1 Orphan nuclear receptor/ SF1a Ftzf1 zinc finger transcription Ad4BP factor NR5A1

Genetic or cellular targets WT1, SRY, SOX9, DAX1, GnRHR, LHb, ACTHR, AMH, AMHR, StAR, P450scc, 21-OH’lase, 11b-OH’lase, oxytocin, SF1, others

Action

Activates transcription of many genes KO mice: no gonads or adrenals; in development of gonads, adrenals; retained Müllerian structures; regulates steroidogenesis. Synergizes abnormal hypothalamus. with WT1; antagonizes DAX1. Dose Haploinusfficient mice: reduced but dependent activity. not absent adrenal function. Homozygous human mutation: 46,XY sex reversal and adrenal hypoplasia. Heterozygous human mutation: 46,XX normal ovary, partial adrenal insufficiency. Represses transcription; activates XY homozygous deletion of transcription of Sry. Dose WT1+KTS isoform: male-to-female dependent effects sex reversal; XY homozygous deletion of WT1-KTS isoform: streak gonads in XX and XY. Human Denys-Drash syndrome: gonadal dysgenesis, congenital nephropathy, Wilms tumor. Expressed at stage of primitive streak. KO mice lack heads, kidneys Organizes development of anterior and genital ridges. neural tissues

11p13

WT1

Zinc finger transcription factor; tumor repressor

DAX1, IGF-II, type 1 IGF receptor, PDGF-A, Pax2, WT1, SRY

Lim1 Lhx1

11p12-13

LIM1

Not reported

Emx2a

10q26.1

EMX2

Homeodomain transcription factor with 2 LIM domains (4 zinc fingers) Homeodomain transcription factor

Wnt4, possibly Lim1

Similar to Lim1. Probably functions downstream of WT1

Lhx9

1q31-32

LHX9

LIM homeodomain transcription factor, similar to LIM1

Drives formation of sex cords

CBX2 M33

17q25, near Sox9

M33

Transcription repressor

Binds SF1 promoter. May have additive effect with -KTS isoform of WT1 in activating SF1 expression Possibly SRY

Wnt4a

1p35

WNT4

Cysteine-rich signaling molecule/secreted growth factor

406

WT1a

Mesonephric mesenchyme

Effects of over- or under-expression

Mediates changes in chromatin structure

Directs Müllerian duct formation

KO mice (XX or XY) lack kidneys, ureters, gonads and genital tracts and have brain defects. Human mutation causes schizencephaly; no urogenital phenotype reported. XX and XY KO mice have gonadal agenesis.

KO mice (XX or XY) have retarded development of gonadal ridges; XY mice have male-to-female sex reversal. XX and XY null mice have Müllerian agenesis (see “ovary” for details of human mutation).

c-Kit

4q12

Kit

Steel

12q22

Slf/KL

Transmembrane tyrosine kinase receptor; protooncogene Ligand or c-Kit

Factors involved in testis/male sex determination/differentiationb SRYa Yp11.3 SRY HMG box-containing transcription factor

407

Germ cell, hemopoetic and melanocyte precursors Unknown

Suppresses apoptosis, directs migration/ proliferation of stem cell populations As for c-Kit

Mouse mutations: white coat color, sterility, anemia. Human mutation: piebaldism, mast cell luekemia. Mouse mutations: white coat color, sterility, anemia.

SF1, SOX9, CYP19, AMH

Bends DNA; may antagonize SOX3

XX mice expressing transgenic Sry female-to-male sex reversal. Human SRY mutations: 46,XY sex reversal. Translocation of SRY to Xchromosome: 46,XX female-tomale sex reversal or true hermaphroditism. Odsex mice (deletion of regulatory locus upstream of Sox9): derepression of Sox9 expression in XX gonads → testis development. Human mutations cause α−thalassemia, mental retardation and genital anomalies → male-to-female sex reversal. Strain-specific effects: XY null mice have defective Leydig cell development and are feminized.

SOX9a

17q24-25

SOX9

HMG box-containing transcription factor of SRY family

Supporting cells of gonadal primordium

Stimulates differentiation of Sertoli cells

ATRXa XH2 XNP

Xq13.3

ATRX

Helicase; transcription factor

Gene regulation at interphase and chromosomal segregation at mitosis

Dhh

12q13.1

DHH

Signaling molecule

Widespread expression early in mouse embryogenesis, more restricted expression later Expressed only in testis

LHCGRa

2p21

LH/CG Receptor

G protein-coupled, 7-transmembrane peptide hormone receptor

Not applicable

StARa

8p11.2

StAR

Mitochondrial transport protein

Not applicable

Transports cholesterol to inner mitochondrial membrane

SRD5A1a ARa

5p15 Xq11-12

5α-reductase2 AR

Not applicable AMHR, ?CYP19

Converts testosterone → DHT Regulates transcription

AMHa

19p13.2-13.3 AMH

Not reported

Ligand for AMH-R

AMHRa

12q13

Mitochondrial enzyme Ligand-dependent nuclear receptor/transcription factor Glycoprotein homodimer of TGFβ family Transmembrane serine/ threonine kinase receptor

Mesenchymal and epithelial cells of müllerian ducts

Mediates apoptosis of Mullerian duct

Type II AMH receptor

Involved in interactions between Sertoli cells and germ cells. May regulate mitosis and meiosis in male germ cells Transduces LH signal to activate Gsα → cAMP. Required for Leydig cell testosterone production

Human mutation: Leydig cell hypoplasia → male pseudohermaphroditism. Mouse: normal sex differentaition. Males and females infertile. Human mutation: congenital lipoid adrenal hyperplasia and male pseudohermpahroditism. Male pseudohermaphroditism. Mouse, rat: testicular feminization (Tfm). Human mutations: androgen insensitivity syndromes. Persistent Müllerian duct syndrome. Persistent Müllerian duct syndrome.

(Continued)

Table 41-2 (Continued) Gene name or pseudonyms

Human gene locus

Human protein name

Protein type

Genetic or cellular targets

Action

Mesonephric mesenchyme

Directs initial Müllerian duct formation in both sexes; possible “anti-testis” factor in ovarian development

FoxL2a

3q23

FOXL2

Transcription factor

Not reported

Fst

5q11.2

Follistatin

Glycosylated protein related to TGFβ family

Granulosa cells

Expressed predominantly in ovary; earliest known marker of ovarian differentiation in mammals Antagonizes action of activins and BMP15 in steroidogenesis; involved in Wnt4 signaling

Gdf9

Unknown

GDF9

Secreted growth factor Ovarian somatic cells member of TGFβ family

Bmp15 Gdf9b

Xp11.2

BMP15

Secreted growth factor Granulosa cells member of TGFβ family

Hoxa13a

7p15-p14.2

HOXA13

Homeodomain transcription factor

Fgf8, Bmp7

Orphan nuclear receptor transcription factor

Retinoic acid receptor, Antagonizes SF1; regulates testis retinoid X receptor, StAR, cord organization. Dose-dependent P450scc, 3β-HSD, CYP17 effects.

408

Factors involved in ovary/female sex determination/differentiation 1p35 WNT4 Cysteine-rich signaling Wnt4a molecule/secreted growth factor

Factors with possible roles in either sex Xp21 DAX1 DAX1a Ahch NR0B1

Effects of over- or under-expression XX and XY null mice: Müllerian duct agenesis. Overexpression in XY: male-to-female sex reversal. Human: duplication of WNT4 associated with 46,XY male-tofemale sex reversal. Mutation in 46,XX: Müllerian regression, androgen excess. Goat: deletion associated with XX sex reversal. Human mutation: 46,XX ovarian dysgenesis. Overexpression in XX mice: small ovaries with block in folliculogenesis; Fst-null XX mice: development of testis-like vasculature Gdf9 null XX mice: block in oocyte differentiation

Secreted by oocytes as paracrine factor required for ovarian somatic cell function Paracrine stimulation of ovarian XX KO mice: subfertile; homozygous granulosa cell growth and mutation in XX sheep: premature proliferation; downregulates ovarian failure; XX human FSH receptor expression; heterozygous mutation: ovarian antagonized by FST dysgenesis Involved in epithelial-mesenchymal Mouse: XX null have hypoplasia of interactions required for cervix and vagina. Human mutation: morphogenesis of terminal gut and hand-foot-genital syndrome with urogenital tract, including Müllerian uterine malformation in 46,XX structures Strain-specific defects in XY mice: overexpression → testis maldevelopment and sex reversal; homozygous deletion → adrenal hypoplasia, normal testes. Human mutations: adrenal hypoplasia congenital, hypothalamic hypogonadism

DMRT1/2

9p24.3

DMRT1 and DMRT2

Doublesex-MAB domain transcription factors

Not reported

Expressed only in genital ridge Dose-dependent effects on postnatal testis development

GATA4a

8p23.1-p22

GATA4

Zinc finger transcription factor

“GATA” motif on target genes; AMH; genes for steroidogenic enzymes

Expressed early in both ovary and testis

WNT7a

3p25

WNT7a

Signaling molecule

Mesenchymal and epithelial cells of Müllerian ducts

XY: involved in Müllerian duct regression; XX: stimulates development of Müllerian duct

aGenes

XY null mice have normal prenatal testis development, but abnormal postnatal testis differentiation. Human monosomy 9p: 46,XY testis maldevelopment; 46,XX primary hypogonadism KO in mouse–embryonic lethal, no gonadal phenotype reported. Human mutation–cardiac defects; no gonadal phenotype reported XY null mice: retained Müllerian ducts; female Wnt7a deficient mice: defective, though not absent, development of oviducts and uterus

in which human mutations have been reported. see Table 4 for enzymes of steroidogenesis. A number of the genes listed have been implicated in gonadal/genital development only by studies in mice and defects have not been reported in humans. ACTHR, adrenocorticotropic hormone receptor; AMH, anti-Müllerian hormone; AMHR, anti-Müllerian hormone receptor; AR, androgen receptor; ATRX, α-thalassemia/mental retardation, X-linked gene; BMP15, bone morphogenetic protein 15; CG, chorionic gonadotropin; DAX1, dosage-sensitive sex reversal-adrenal hypoplasia congenita locus on the X-chromosome, gene 1; DHH, desert hedgehog; DHT, dihydrotestosterone; DMRT1/2, Doublesex and MAB-3-Related Transcription Factors 1 and 2; FSH, follicle stimulating hormone; FST, follistatin; FOXL2, forkhead transcription factor 2; GDF, growth differentiation factor; GnRH-R, gonadotropin-releasing hormone receptor; HMG, high-mobility group; HOXA13, homeobox A 13; HSD, hydroxysteroid dehydrogenase; IGF, insulin-like growth factor; KO, knockout; KTS, lysine/threonine/serine; LH, luteinizing hormone; LHCGR, luteinizing hormone/chorionic gonadotropin receptor; LHX9, LIM homeobox gene 9; LIM, Lin-11, Islet-1 and Mec-3; LIM1, LIM homeobox gene 1; PDGF, platelet-derived growth factor; SF1, steroidogenic factor 1; SOX, Sry-related homeobox gene; SRY, sex-determining region of the Y chromosome; StAR, steroidogenic acute regulatory protein; TGF-β, transforming growth factor-β; WNT, wingless-type MMTV integration site family member; WT1, Wilms tumor 1. bNote:

409

410

SECTION IV / ENDOCRINOLOGY

Figure 41-2 Genetic determinants of development of the primordial gonad. Transcription factors Emx2, Lim1, Wt1, and Wnt4 are required for development of the adrenogenital primordium, which forms following condensation of primitive mesoderm and coelomic epithelium. Thereafter, various combinations of transcription factors direct the fate of the undifferentiated primordial cells down one of a number of pathways to form the adrenal cortex, kidney, bipotential gonad and internal reproductive tract primordia. Factors highlighted in bold have been established to be involved in human development. Dax1, dosage-sensitive sex reversal-adrenal hypoplasia congenita locus on the X-chromosome, gene 1; Emx2, empty spiracles 2; Hoxa1, homeobox A1; Lhx9, LIM homeobox gene 9; Lim1, Lin-11, Islet-1, and Mec-3 homeobox gene 1; Sf1, steroidogenic factor 1; Wt1, Wilms tumor 1; Wnt4, wingless-type MMTV integration site family member 4.

and gonadal development overlap, when tissues are undifferentiated and the major developmental need of the organism is to increase cell mass. As development progresses, and populations of cells begin to differentiate along specific, irreversible paths, there is a requirement for a much more focused program of transcription factor action. This concept may help in the understanding of the diverse roles served by transcription factors such as SF1, WT1, and LIM (the acronym stands for Lin-11, Islet-1 and Mec-3––the three original members of the family) homeobox* gene 1 (LIM1), compared with the much more limited roles, expression, timing and cellular specificity of factors such as SRY and AMH. Factors implicated in the early development of the bipotential gonadal and reproductive tract primordia on the basis of defects observed in humans include SF1, WT1, and WNT4. In addition, LIM1, empty spiracles 2 (EMX2), and LIM homeobox gene 9 (LHX9) are implicated in these processes on the basis of murine studies. Furthermore the ligand/receptor pair Steel/c-Kit is vital, at least in mice, to the process of germ cell migration from the yolk sac to the gonadal primordium in both sexes. Exact relationships between these factors and the precise timing and order of their expression remain to be determined; a hypothetical scheme for their potential roles in regulating the processes of early development of the reproductive tract is shown in Fig. 41-2. In the next section, descriptions of the molecular biology of individual factors and their molecular defects in humans, are provided. Additional genes postulated to be involved in these processes based on studies in nonhuman species are described briefly at the end of this section. *The homeobox is the approx 180 bp sequence in the gene that encodes the approx 60 AA homeodomain of the respective protein.

Steroidogenic Factor 1 (SF1) Normal Function The orphan nuclear receptor transcription factor, SF1 (also referred to as adrenal 4-binding protein and officially termed nuclear receptor subfamily 5, group A, member 1) appears to be one of the earliest-acting and most critical factors in the primitive development of the reproductive tract. SF1 affects reproductive function at all three levels of the hypothalamic-pituitary-gonadal axis, as well as the adrenal gland, and subsequently regulates factors acting further down the pathway of gonadal/genital development in a male-specific fashion. Importantly, SF1 appears to act in a dose-dependent manner in both mice and humans. The human SF1 gene is located at chromosome 9q33, spans 30 kb of genomic DNA and contains seven exons including an initial noncoding exon. SF1 is a 461-amino acid (AA), 53-kDa protein containing two central DNA-binding zinc fingers (ZFs) typical of nuclear receptors, an activation domain and a C-terminal ligand-binding domain (for which no ligand has yet been identified, hence its designation as an “orphan” receptor). SF1 differs structurally from most nuclear receptors by lacking an N-terminal domain and differs functionally by binding as a monomer, rather than as the more usual dimer, to a nonpalindromic DNA sequence. Based on studies of human mutations, two regions of the protein appear particularly critical for function: the so-called P-box, located in the proximal portion of the first ZF, is responsible for interaction with the major groove of DNA; the A-box region downstream of the ZFs modulates monomeric binding to DNA. In the mouse Sf1 (also known as FtzF1, based on its similarity to the drosophila gene, fushi tarazu factor 1) is expressed in male

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

and female urogenital ridge at embryonic day (E)9-9.5, the earliest stage of organogenesis of the indifferent gonads. Consistent with its role in regulation of steroidogenesis, Sf1 is expressed prior to expression of the first enzyme of steroidogenesis, P450 side chain cleavage enzyme (p450scc). Because both Sf1 and Wt1 (discussed later) are expressed in the gonadal primordium at a similar time, before gonadal differentiation occurs, Sf1 interaction with Wt1 may play a role in proper development of the gonadal precursor. After a sex-specific gonad can be discerned, Sf1 expression becomes sexually dimorphic, presumably under the regulation of other sex-specific factors: in the fetal testis (specifically Sertoli and Leydig cells) Sf1 expression continues at a high level; in XX gonads Sf1 is expressed in somatic cells until E13 and disappears altogether between E13.5 and 16.5, in keeping with the lack of ovarian steroidogenic activity at that time. In adult mice Sf1 is expressed in all primary steroidogenic tissues, including all zones of the adrenal cortex, testicular Leydig and Sertoli cells, ovarian theca and granulosa cells and corpus luteum. Not only does Sf1 regulate gonadal and adrenal development, it also appears to control development of an important hypothalamic nucleus, and of pituitary gonadotropes. Sf1 mRNA is detectable in the developing mouse pituitary at E13.5-14.5 specifically in the gonadotropes. The expression of Sf1 and Dax1, another orphan nuclear receptor described later, overlap in their tissue distribution. Dax1 appears to repress or antagonize Sf1 activity, particularly the synergy between Sf1 and Wt1, probably via a protein–protein interaction with Sf1. On the basis of this finding it has been proposed that Sf1 and Dax1 functionally oppose each other. These interactions may be key for the differential expression of genes involved in female vs male development. SF1 is implicated in the regulation of numerous genes, as binding sites for this transcription factor have been located within the promoter regions of at least 20 target genes, including those encoding a number of other transcription factors, hormones, receptors, proteins, and enzymes of steroidogenesis, and the signaling molecule, Wnt4. SF1 binds to the AMH gene promoter and upregulates AMH expression in association with WT1. Notably, SF1 is unable to regulate AMH expression in a heterologous cell line unless its ligand-binding domain is deleted. This finding suggests existence of a possible ligand for SF1, perhaps a Sertoli cell-specific factor. Studies in transgenic null mice highlight the requirement for Sf1 in both sexes for development of the organs that derive from the adrenogenital primordium––primordial gonads and adrenal cortex––as well as the hypothalamus and pituitary. Male and female mice homozygous for deletion of Ftzf1 failed to develop steroidogenic tissues, having neither gonads nor adrenal glands. The mice had phenotypically female internal and external genitalia and died in the neonatal period of adrenal insufficiency. These mice did display mesenchymal thickening in the gonadal ridge area at the earliest stages of gonadal development (E10.5), but thereafter the cells of this region underwent apoptosis, suggesting that the role of Sf1 may be in maintaining rather than initiating gonadal development. The mice also had abnormal development of the ventromedial hypothalamic nucleus, important in control of pituitary gonadotropin secretion. Expression of proteins specific to gonadotropes, leutenizing hormone (LH), follicle stimulating hormone (FSH) and gonadotropin-releasing hormone (GnRH) receptor (GnRH-R) was absent, implicating Sf1 in gonadotrope development and LH/FSH expression. The presence of female internal genitalia in the null animals of both sexes indicates that the Müllerian ducts did not regress as

411

would normally occur in the males. Because Sf1 upregulates expression of Amh, deficiency of Sf1 in males would result in lack of Amh and therefore retention of Müllerian structures. Therefore, Sf1 is likely active earlier than Amh in the ontogeny of mouse sexual differentiation. Müllerian ducts develop normally in null female mice because Sf1 expression in the normal developing ovary is low, so the female is unaffected by its absence. In contrast to the complete absence of gonadal development in the null mice, Sf1 haploinsufficiency resulting from heterozygous Sf1 deletion is associated with gonads that apart from being slightly small, are essentially normal, indicating that a single active gene is adequate for this function. In addition, although significant defects in adrenal development and organization were observed, the mice were able to maintain adequate basal adrenal function by compensation via cellular hypertrophy and increased expression of steroidogenic acute regulatory protein (Star). Clinical Features and Molecular Defects The infrequent mutations identified in the human SF1 gene are instructive regarding the roles of SF1 in human, as compared to murine, development. In one report, a 46,XY phenotypically female infant presented with adrenal failure soon after birth. She had retained Müllerian structures, including a uterus, and small, intra-abdominal streak-like gonads containing poorly differentiated tubules. Gonadotropin responses to GnRH were normal but she had no testosterone response to human chorionic gonadotropin (hCG). A heterozygous mutation encoding a glycine to glutamic acid substitution in the first ZF of the SF1 DNA-binding domain was found, resulting in loss of DNA-binding activity of the receptor and therefore loss of transcription activating capacity. Because no dominant-negative activity was detected, it is unclear why this mutation caused gonadal dysgenesis in the heterozygous state. A second case involved a less severe missense mutation of SF1 manifest only in the homozygous state. Male-to-female sex reversal and adrenal hypoplasia were present in the homozygous 46,XY infant, but the heterozygous sister and consanguineous parents were clinically normal. A uterus was present on MRI scan. Histology of the gonads was not reported. The infant had massively elevated adrenocorticotropic hormone (ACTH), with extremely low cortisol and aldosterone. This mutation caused an arginine to glutamine substitution in the A-box region of the protein, associated with reduction but not complete loss of DNA binding and transcriptional activity. In the third case a 46,XX female presented at 14 mo of age with adrenal insufficiency. She had no apparent defect in ovary development. A heterozygous de novo mutation causing an arginine to leucine substitution in the SF1 hinge region was found, resulting in a transcriptionally inactive protein. This latter case in which a heterozygous mutation resulted in impaired, but not absent, adrenal development is similar to the phenotype of Sf1 haploinsufficiency in mice. Comparison of these clinical cases suggests the importance of gene dosage in humans. Double-copy expression of SF1 appears to be required for testis development in males, but not for ovarian development in females. Normal adrenal development occurred in the presence of a heterozygous SF1 mutation in one case and failed to occur in the other, suggesting that SF1 mutations may have varying effects on adrenal development depending on the location and type of mutation, the functional effects on the SF1 protein and perhaps the genetic background of the individual. Wilms Tumor 1 (WT1) Normal Function Like SF1, the transcription factor WT1, plays a pivotal role in early development of the primordial gonad and the bipotential internal genital duct systems prior to sex

412

SECTION IV / ENDOCRINOLOGY

Figure 41-3 Structure and function of transcription factor WT1. (A) Major isoforms of WT1. WT1 is a potent transcriptional activator or repressor, depending on the specific cellular and genetic context. There are four major isoforms of WT1, formed by differential splicing of exons 5 and 9 that either include or exclude a 17 amino acid sequence in the central region of the protein (encoded by exon 5) and a lysine, threonine, serine (KTS) triad between the third and fourth zinc fingers (encoded by exon 9) that alters the DNA-binding specificity of the protein. The –/– isoform, lacking both inserts, has greater transcriptional activation potential than the major isoform (+/+) that contains both inserts. (B) Wt1 binding at target genes. WT1 regulates transcription of numerous target genes involved in various aspects of sex determination and differentiation, including SRY, DAX1 and AMH. WT1 (–KTS) binds as a homodimer to the promoter of the SRY gene to upregulate its transcription. In contrast, at the AMH gene, WT1-KTS binds as a heterodimer with SF1, antagonizing binding of DAX1 to the promoter. AMH, anti-Müllerian hormone; DAX1, dosagesensitive sex reversal-adrenal hypoplasia congenita locus on the X-chromosome, gene 1; KTS, lysine/threonine/serine; SF1, steroidogenic factor 1; SOX, Sry-related homeobox gene; SRY, sex-determining region of the Y-chromosome; WT1, Wilms tumor 1.

determination, and in addition, a specific isoform appears to be required for subsequent development of the testis. WT1 is a member of the early growth response (EGR) family of transcription factors (proteins expressed early in the cell cycle, at G0–G1 transition) and acts as a transcriptional activator or repressor depending on the cellular or chromosomal context. There appears to be a general requirement for WT1 in the formation of organs derived from the intermediate mesoderm, particularly the differentiation of glomerular epithelial cells and gonadal primordium. In renal tissue WT1 acts as a tumor suppressor. The human WT1 gene is a complex locus at 11p13 that in fact consists of two genes, WT1 and WIT-1, expressed from opposite DNA strands. The function of the WIT-1 transcript is unknown, but a role as an antisense regulator of WT1 has been postulated. The highly conserved 50-kb WT1 gene itself contains 10 exons that can be alternatively spliced to yield four distinct mRNA

species of approx 3–3.5 kb each. The primary WT1 protein is a 429-AA (~50 kDa) transcription factor with 4 contiguous Cys2His2 ZF domains (encoded by exons 7–10) and an amino terminus rich in proline and glutamine, typical of certain transcription factors. Separate domains subserve transcriptional repression and activation: residues 85–124 encompass the repressor domain, and 181–250 the activator domain. These regions are distinct from the DNA-binding domain and their activities are probably mediated by protein–protein interaction. The four mRNA species encode four major proteins, designated WT1 (A–D), and perhaps as many as 32 different isoforms, differing mainly on the basis of the presence or absence of an additional 17 AA in the middle of the protein, and of a lysine/threonine/serine (KTS) triad between ZFs 3 and 4 that alters their spacing, thereby changing the DNA-binding specificity (Fig 41-3A). The –KTS and +KTS isoforms also have differential expression patterns within cell nuclei and appear to have distinct but

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

somewhat overlapping roles. The fact that all transcripts are expressed at similar levels suggests that each encoded protein makes a significant contribution to overall WT1 action, and interactions between the proteins, each of which may have distinct targets and functions, may be important in the control of cellular proliferation and differentiation exerted by WT1. Various forms of WT1 regulate SRY, DAX1, and AMH expression (Fig. 41-3B). WT1 can homodimerize and also binds several other proteins; WT1-KTS isoforms associate and synergize with SF1 to promote AMH expression and can activate the DAX1 promoter. However, DAX1 antagonizes the synergy between SF1 and WT1, most likely through a direct protein–protein interaction with SF1, suggesting that WT1 and DAX1 functionally oppose each other in testis development by modulating SF1-mediated transactivation. The fact that WT1 can upregulate expression of DAX1, which in turn antagonizes WT1/SF1-mediated stimulation of AMH expression suggests that the relative dosages of WT1-KTS and DAX1 and the timing of their expression during embyrogenesis are critical in the delicate balance of transcription factor activity required for gonadal/genital development. Wt1 expression in mouse embryogenesis also supports a role for Wt1 in development of urogenital tissues: Wt1 is expressed very early in fetal life, at about the same time as Sf1 and before Dax1. Wt1 is faintly detectable as early as day 9 and readily detectable in pronephric and mesonephric tissues on E10.5, at which time Sry expression is detectable in pre-Sertoli cells. By E11.5 the nephrogenic cord, condensing metanephric tissue and urogenital ridge display high levels of Wt1 message. In the developing gonad Wt1 expression localizes to the sex cords. Although the role for Wt1 in primordial gonad development is fairly clear, its roles in specific testis or ovary development are less well understood, and it seems likely that the various isoforms of Wt1 may function differentially in the different genetic contexts of testicular vs ovarian development. Consistent with this concept, Wt1 expression is differentially regulated during development depending on the sexual differentiation of the gonad. In mature gonads Wt1 expression is confined to the Sertoli cells and tunica albuginea of the testis and the granulosa and epithelial cells of the ovary. WT1 expression in human development between weeks 7–10 of gestation is similar to that seen in the mouse at the equivalent time. WT1 is expressed mainly in mesodermally derived tissues––kidneys, gonads and mesothelium––but is also expressed in spinal cord and brain, tissues of ectodermal origin. Midtrimester human embryos show strong expression in kidneys and gonads. WT1 expression is limited to Sertoli cells in adult testes. Transgenic mice homozygous for a knockout mutation of the entire Wt1 gene had failure of renal and gonadal development. At day 11 of gestation, cells of the metanephric blastema underwent apoptosis, the ureteric bud failed to grow out from the Wolffian duct, and the metanephric kidney did not form. A later study also demonstrated a requirement for Wt1 in the development of epicardium and adrenal gland (the reason for the difference in phenotype between studies is not clear). Different gonadal phenotypes are seen when the knockout targets a specific isoform of Wt1. Mice completely lacking the –KTS Wt1 isoforms had reduced Dax1 expression, tiny streak gonads (both males and females) and abnormal development of the internal genital ducts. Homozygous deletion of the +KTS isoforms in male mice caused complete XY sex reversal, the embryonic gonads having the morphological appearance of ovaries, associated

413

with a dramatic reduction of gonadal Sry and Sox9 expression and female-type Dax1 expression. These findings imply a key role for the +KTS isoforms in regulation of the critical testis-determining Sox genes in male gonadal development. As female mice lacking the +KTS isoforms had normal ovaries, this isoform appears to be unnecessary for ovarian development. Mice with reduced (but not absent) Wt1 +KTS levels had normal fetal urogenital development but developed glomerulosclerosis after 2–3 mo, representing a model for the human disease, Frasier syndrome (described later). This result indicates that Wt1+KTS dosage is important for normal postnatal renal function, but not for prenatal development, because the level of Wt1 (either isoform) expressed from a single normal Wt1 allele was adequate for normal urogenital development. These data demonstrate distinct functions for the Wt1 +/–KTS isoforms. It appears from these different transgenic experiments that some level of Wt1 expression is absolutely required for primordial gonad development, but that the subsequent development of the testis specifically requires the +KTS isoform. Clinical Features and Molecular Defects Three clinical syndromes are associated with allelic defects of WT1 (Table 41-3). The Denys-Drash syndrome comprises a triad of gonadal dysgenesis, congenital nephropathy and subsequent development of Wilms tumor. Gonadal and genital development is heterogeneous in these patients, ranging from streak gonads or ovotestes in either 46,XX or 46,XY individuals, to rudimentary or dysgenetic testes in 46,XY patients, or normal ovaries in a 46,XX patient. Perhaps because of an ascertainment bias, approx 90% of patients have 46,XY karyotype. Their external genitalia are usually ambiguous, sometimes female, occasionally male. The internal genitalia are often incongruous with respect to the external genitalia: Wolffian structures may be present in phenotypic females, Müllerian structures in phenotypic males. Even in 46,XX females, there may be anomalous internal genital development, and one 46,XY patient with dysgenetic testes had neither Müllerian nor Wolffian structures, compatible with a primary role of WT1 in early development of the bipotential internal duct systems. Gonadoblastomas and granulosa cell tumors have been reported. The severe congenital nephropathy (diffuse or focal mesangial sclerosis) generally leads to death by age 2. Wilms tumor is reported in over 50% of cases, most presenting with an abdominal mass in the second year of life. The prevalence of Wilms tumor would probably be higher if the patients survived longer, and most are found at autopsy to have persistent intralobar renal blastema, which may be the precursor of Wilms tumor. Notably, in patients with isolated bilateral Wilms tumor (distinct from the Denys-Drash syndrome) the occurrence of hypospadias and cryptorchidism is 10-fold higher in than in the general population. The Denys-Drash syndrome is associated with heterozygous WT1 mutations in >90% of cases. The disorder is genetically dominant, as no patient has been reported with mutations in both alleles of the gene. Mutations cluster within or near the ZF coding region (exons 7–10, particularly exon 9), most producing AA substitutions in ZF2 and ZF3. One mutation, encoding an arginine to tryptophan substitution at position 394 in ZF3, has occurred recurrently, being present in at least a dozen reported cases. Mutations in a nearby AA, aspartic acid 396 (either to asparagine or glycine), have been reported in other cases, suggesting an important role for these two AAs in ZF function. Another mutation that is reported in several instances is conversion of arginine 362 to a stop codon, resulting in a truncated protein lacking three of the four ZFs. Notably, there is significant phenotypic variation even among individuals with

414

SECTION IV / ENDOCRINOLOGY

Table 41-3 Clinical Syndromes Associated With Mutations in WT1 Syndrome

Mode of dysfunction

Mutation

Denys-Drash

Dominant negative. Possible reduced expression of SRY and AMH

Heterozygous amino acid substitutions in zinc fingers

Frasier (atypical subtype of Denys–Drash)

Hapolinsufficiency of +KTS isoform → imbalance in ratio of –KTS/+KTS forms

Heterozygous point mutation in intron 9 splice donor site.

WAGR (Wilms tumor, Aniridia, Genitourinary abnormalities, mental Retardation)

Hapolinsufficiency

Heterozygous gene deletion

Phenotype Kidney: diffuse mesangial sclerosis → glomerular nephropathy; Wilms tumor common. Gonads/genitalia: 46,XY–heterogeneous effects, from streak gonads with complete sex reversal to dysgenetic testes with partial or normal masculinization. Ovotestes have also been reported. 46,XX–streak gonads, ovotestes or ovaries. Internal genitalia may be discordant from external genitalia in both sexes. Kidney: focal and segmental glomerular sclerosis; later onset of renal failure than Denys–Drash Syndrome. Wilms tumor rare. Gonads/genitalia: 46,XY sex reversal, gonadoblastoma common; 46, XX–no reported phenotype. Kidney: Wilms tumor common. Gonads/genitalia: 46,XY–cryptorchidism, hypospadias; 46,XX–no reported phenotype.

Note: Homozygous deletion of WT1 in approx 10% of Wilms tumors. AMH, anti-Müllerian hormone; KTS, lysine/threonine/serine; SRY, sex-determining region of the Y-chromosome; WT1, Wilms tumor 1.

this severe truncation mutation. Other defects include a variety of nonsense, frameshift, splice-junction, and missense mutations, most of which appear de novo, as parental WT1 genes are normal. In one case, however, the normal father of an affected child was found to be heterozygous for the mutant allele, suggesting either mosaicism or reduced penetrance of the mutant gene. The primary molecular pathophysiology of WT1 mutations appears to be lack of normal DNA binding by the WT1 protein; mutant WT1 proteins containing nonconservative AA substitutions within the ZF region do not bind to the EGR1 consensus binding sequence bound by wild type WT1. Complete WT1 deletion produces milder genital abnormalities (cryptorchidism and/or hypospadias), than does a mutation that encodes expression of an abnormal WT1 protein, suggesting a dominant negative effect of mutant WT1 proteins by formation of inactive dimers with wildtype WT1. Mutations that affect zinc-coordinating cysteine or histidine residues likely prohibit DNA binding by disrupting proper spatial organization of the ZF. Although WT1 mutations are heterozygous at the germ cell level, tissue from the Wilms tumors of these patients generally demonstrates loss of heterozygosity for WT1, indicating that two mutant WT1 alleles may be required for loss of tumor suppressor function of WT1 (“two-hit” hypothesis). In a second WT1-related disease, Frasier syndrome (which in fact may be a variant of Denys-Drash), 46,XY patients have streak gonads, female external genitalia and frequently develop gonadoblastoma but not Wilms tumor. The renal failure is milder and of later onset than in the Denys–Drash syndrome, progressing to end-stage renal failure in adolescence or early adulthood. Frasier syndrome is caused by mutations in the donor splice site in intron 9 of WT1, with predicted loss of the +KTS WT1 isoform, resulting in a reduction of the normal +KTS/–KTS isoform ratio. The third WT1-related condition, the WAGR syndrome, includes Wilms tumor, aniridia, genitourinary abnormalities/gonadoblastoma and mental retardation. This syndrome likely results from heterozygous loss of genes contiguous with WT1 in the 11p13 region, accounting for the aniridia and mental retardation that accompany

the Wilms tumor and genitourinary abnormalities. In patients in which it has been examined, the mutant allele appears to have arisen in the paternally derived chromosome 11. GENES WITH PUTATIVE ROLES IN BIPOTENTIAL GONAD DEVELOPMENT The following genes are discussed briefly on the basis of significant evidence in nonhuman species for roles in development of the primordial gonad. Lim Homeobox Gene 1 (LIM1) LIM transcription factors constitute a large family (at least 40 members) of proteins that carry 2 tandem copies of the LIM domain, a unique cysteine-rich zincbinding domain involved in protein–protein interactions, followed downstream by a homeodomain that mediates DNA binding. LIM domains can interact with other proteins to form complexes that regulate transcription. Lim homeobox gene 1 (Lim1), encodes an early member of the family that appears to be a transcriptional activator. Lim1 plays a major role in organizing the development of the head in mice and is implicated in gonadal and renal development. Lim1 expression in mice occurs extremely early in the intermediate mesoderm and nephrogenic cord. Consequently, the action of Lim1 in gonad formation is likely at the level of the primordial gonad. Lim1 is also expressed later in development in several urogenital tissues such as mesonephros, Wolffian ducts, ureteric buds and definitive kidney. Mice homozygous for a null mutation of Lim1 or for mutations that alter conserved AAs required for structure of the LIM domains had no heads, kidneys, or gonads. Empty Spiracles 2 (EMX2) Another homeodomain transcription factor, empty spiracles 2 (Emx2), appears in mice to have similar functions to Lim1, being essential for the development of the dorsal telencephalon and components of the developing urogenital system. Emx2 is intensely expressed in the bipotential gonads and ovaries and in epithelial components of the developing urogenital system, specifically the pronephros and mesonephros, Wolffian and Müllerian ducts, ureteric buds and in early epithelial structures derived from metanephrogenic mesenchyme. Emx2 is also detected in brain, kidney, and uterus. Notably, expression of Lim1 in Emx2null mice was reduced and that of Wnt4 was absent, suggesting that

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

Emx2 may modulate their expression. In addition to developmental defects of the brain, Emx2-null mice lacked kidneys, ureters, gonads, and genital tracts. Degeneration of the Wolffian duct and mesonephric tubules was abnormally accelerated and Müllerian ducts did not form. These abnormalities imply a role for Emx2 in very early development of the renal, gonadal and internal genital primordia, probably at the stage of development of the mesonephros. Lim Homeobox Gene 9 (LHX9) Lhx9, encodes another member of the LIM family of transcription factors. Like Lim 1, Lhx9 also plays a role in primordial gonad development, evidenced by the finding of complete gonadal agenesis in Lhx9-null mice of both sexes, without apparent extra-gonadal defects (in contrast with the headless Lim1-null mutants). Gonadal agenesis appeared to result from failure of gonadal cell proliferation at about E12, rather than exaggeration of apoptosis. Lhx9 appears to be necessary for proliferation and invasion of the epithelial (somatic) cells of the genital ridge into the underlying mesenchyme and subsequent formation of sex cords––an essential step in formation of the gonads. Lhx9 expression is detectable at E9.5 in epithelial and subjacent mesenchymal cells of the early gonadal ridge. Expression localizes to the interstitial region of the developing testis as morphological differentiation occurs, then disappears as epithelial cells differentiate into Sertoli cells and begin to express Amh. In the fetal ovary Lhx9 is highly expressed in epithelial cords, then is downregulated as ovarian epithelial cells differentiate into granulosa cells. Thus it appears that Lhx9 expression is inversely correlated with the degree of differentiation of mesenchymal cells of the gonad. In human embryos LHX9 is expressed in the abdominal region of both sexes at the time of gonad formation. Human LHX9 mutations have not been reported, despite a careful search in at least one study in 27 patients with 46,XY gonadal agenesis or dysgenesis. All Lxh9-null mice were phenotypically female. They had atrophic uteri, vagina, and oviducts indicating that Lhx9 is not required for development of the internal genital primordia. As would be expected in the situation of gonadal agenesis, the mice had high FSH concentrations and no detectable testosterone or estradiol. A key role of Lhx9 may be in regulation of Sf1 expression. Lhx9 binds directly to the Sf1 promoter and may act synergistically with the –KTS isoform of Wt1 in activating Sf1 expression. This regulatory interaction appears to be specific to the gonad, evidenced in Lhx9-null mice by the finding of normal adrenal Sf1 expression despite absent expression in genital ridges. NORMAL AND ABNORMAL TESTIS DETERMINATION Male sex determination is synonymous with testis determination. Once development of the primordial gonadal and genital structures has occurred, the next steps diverge between the sexes. The process of testis development in the karyotypic male appears to be controlled by a switch-like mechanism that involves the crucial Y-chromosome-encoded transcription factor SRY, a related homodomain transcription factor, SOX9, and no doubt other genes and proteins that either regulate or are regulated by these factors. One of the earliest effects of SRY expression is the induction of somatic cell migration from the mesonephros into the XY gonad—a critical first step in preparation for development of testis cords. Subsequently, SOX9 directs the process of seminiferous tubule organization by Sertoli cells and regulates expression of the Sertoli cell glycoprotein AMH. AMH may in turn play a role in directing undifferentiated interstitial cells to develop as Leydig cells. Once testicular differentiation is established, other Y-encoded factors are required to maintain spermatogenesis. A number of other molecules have been identified as having involvement in testis development; however,

415

their exact positions in this pathway, their functions and the factors they regulate or by which they are regulated, remain to be elucidated. These include the +KTS isoform of WT1, the helicase enzyme ATRX, perhaps the intracellular signaling molecule, desert hedgehog (DHH) and gene(s) on chromosome 10q. Additional Xchromosomal sequences likely affect testis development, perhaps negatively, because the presence of one or more extra X-chromosomes (as found in Klinefelter syndrome and its variants) is associated with reduced testicular size. There are many hypothetical schemes for the interactions of the ever-expanding coterie of transcription factors involved in testis determination, and no definitive model for this process exists. Analyses of humans with gonadal dysgenesis, mouse models and in vitro assays have revealed that testis determination results from a complex network of interacting transcription factors and target genes in a nonlinear web of upregulation or activation and downregulation or repression steps. A hypothetical model, based primarily on work in mice, is as follows: 1. Establishment of the gonadal primordium (multiple factors, described earlier). 2. Activation of Sry by Wt1 and Gata4. 3. Sry repression of Wnt4 and perhaps forkhead transcription factor 2 (Foxl2) thus. 4. Allowing Sf1 to. 5. Stimulate expression of Sox9 (possibly by interfering with binding of a repressor factor to the Odsex locus upstream of the Sox9 gene) and perhaps Atrx, leading to. 6. Development of Sertoli cells followed by. 7. Wt1/Sf1/Sox9/Gata4-mediated induction of Amh expression. 8. Action of Amh via the Amh receptor(s) causing Müllerian duct regression. 9. Sf1-mediated stimulation of steroidogenic activity by fetal Leydig cells (Fig. 41-4). There may be significant species differences in the processes of testis development, so although the mouse is generally a convenient model, differences between mice and humans must not be overlooked and the details of sex determination and differentiation that have been elucidated in rodents cannot necessarily be extrapolated to human development. SRY-Related Homeobox (SOX) Genes There is an everexpanding family of DNA-binding, atypical transcription-regulating proteins related to each other by the presence of a central high mobility group (HMG) domain,* homologous (>60% AA identity) to the HMG domain of the founding member of the family, SRY. At least 20 SOX genes have been described, and a number of these appear to be involved in sex determination or have testisspecific expression, including SRY itself, SOX9, SOX3, SOX30, and perhaps SOX8. Only the most well characterized genes––SRY and SOX9––are discussed. Sex-Determining Region Of The Y (SRY) Gene Normal Function The existence of a Y-chromosomal “maleness-determining” gene was postulated in the 1930s, and in the 1960s was designated the “testis-determining factor.” Many candidates were proposed and rejected over the years, until 1990, *The HMG domain is a 70–80 AA DNA-binding motif comprised mainly of hydrophobic and charged AAs, shared by a group of architectural proteins involved in DNA transcription, replication, recombination, and repair.

416

SECTION IV / ENDOCRINOLOGY

Figure 41-4 Hypothetical model of testis determination. The bipotential gonad of an XY fetus expresses the key testis-determining gene Sry, which is upregulated by Wt1 and Gata4. Sry represses the “anti-testis” gene, Wnt4, and perhaps Foxl2 thus allowing Sf1 to stimulate Sox9 expression leading to differentiation of supporting cells into Sertoli cells, the primary organizers of the seminiferous tubule. The seclusion of germ cells within the tubules prevents germ cell meiosis and commits them to spermatogenic development. Wt1, Sf1, Sox9, and Gata4 induce Sertoli cell expression of Amh, which, acting via its type 2 receptor, causes Müllerian duct regression. Under the influence of Sertoli cells, interstitial cells differentiate as Leydig cells, subsequently secreting testosterone under the drive of Sf1 and Gata4 inducing development of Wolffian structures. Other factors that have less well-defined roles include Lim1, Emx2, Lhx9, Dax1, Dhh, Atrx, and Dmrt1/2. AMH, anti-Müllerian hormone; ATRX, α-thalassemia/mental retardation, X-linked gene; DAX1, dosage-sensitive sex reversal-adrenal hypoplasia congenita locus on the X-chromosome, gene 1; DHH, desert hedgehog; DMRT1/2, Doublesex and MAB-3-Related Transcription factors 1 and 2; EMX2, empty spiracles 2; LHX9, LIM homeobox gene 9; LIM1, LIM homeobox gene 1; SF1, steroidogenic factor 1; SOX, Sry-related homeobox gene; SRY, sec-determining region of the Y-chromosome; WNT, wingless-type; MMTV integration site family member; WT1, Wilms tumor 1.

when the existence of such a gene was confirmed, with the discovery of SRY. Insertion of the mouse Sry gene into fertilized XX mouse eggs resulted in development of testes and male genitalia (female-to-male sex reversal). These seminal studies demonstrated that Sry was the key Y-chomosomal gene sufficient to induce maleness. However, replacement of the homeobox of Sry with that of Sox3 or Sox9 also results in sex reversal in XX mice, indicating that Sox3 or Sox9 can functionally replace Sry and elicit development of testis cords, male patterns of gene expression, and male genital development. Human SRY is a 3.8-kb single-exon gene located just centromeric to the pseudoautosomal region of Yp (Yp11.3) that encodes a 204AA (24 kDa) protein. Approximately the middle one-third of the protein (79 AAs) represents the HMG domain that endows SRY with its sequence specific DNA binding to a target nucleotide sequence. This region is the most critical to protein function and almost all human mutations are located here. Outside the HMG domain the remainder of the protein is poorly conserved among mammals. The target genes of SRY are largely unknown, but may include those encoding AMH, SOX9, and the steroidogenic enzyme CYP11. Similarly, the exact mechanism of action and whether SRY has any direct transcriptional activating capacity, remain to be determined. Human SRY lacks a transcription activation domain (present in mouse), so it has been suggested that SRY function depends solely on the HMG domain and that it acts as an “architectural” transcription factor, by creating the spatial arrangement needed for the transcription “machinery” to work. Expression of Sry occurs in fetal mouse gonad at the earliest stage of specific testis formation, being first detected at E10.5 in pre-Sertoli cells in the developing gonadal ridge, to which its

expression is limited, about 1 d before testicular morphology can be discerned. This finding suggests that these pre-Sertoli cells are integral to the process of testis development. Indeed, one of the primary events in testis development is induction of Sertoli cell differentiation. Sry expression peaks at E11.5 and declines once testicular development is established, at E12.5. Sry expression and Sertoli cell differentiation are followed by testis differentiation, manifest by Sertoli cell-regulated organization of seminiferous tubules and Leydig cell differentiation. In human testes SRY expression begins at approx 6 wk gestation, just prior to specific testis development. The fundamental role of SRY appears to be the upregulation of SOX9 expression, which in turn stimulates development of the prime organizers of testicular architecture––the Sertoli cells. SRY may continue to act as a splicing factor in Sertoli cells and germ cells in the adult testis. Despite its preeminent role in testis determination, other transcription factors appear to regulate SRY expression, including WT1. Furthermore other upstream and downstream factors must be activated or repressed to allow testis development to occur. Evidence for this conclusion includes the following: 1. Sry expression during mouse gonadogenesis; occurs during a very brief time window. 2. The majority of 46,XY females with gonadal dysgenesis have an intact SRY gene. 3. Some 46,XX males with testes are completely Y negative, indicating that non-Y sequences must be responsible for testis determination in these cases. Clinical Features and Molecular Defects The study of individuals with the syndromes of 46,XX maleness and 46,XY gonadal

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

417

Figure 41-5 Mutations in human SRY.

dysgenesis was the catalyst for the eventual localization and characterization of the SRY gene. Of 46,XX males with unambiguous masculinization, >90% are SRY positive (this can be confirmed by PCR or fluorescent in situ hybridization analysis), whereas the converse holds for 46,XX males with genital ambiguity and 46,XX true hermaphrodites, of whom only a small number are SRY positive. Nonrandom inactivation of the X-chromosome carrying the translocated SRY is the postulated explanation for the presence of ovarian and testicular tissue in the same gonad in individuals with SRY-positive hermaphroditism. The coexistence of 46,XX complete maleness and 46,XX true hermaphroditism in a number of families may reflect variations in the pattern of inactivation of the SRY-bearing Xchromosome between affected individuals. In parallel to the finding of SRY sequences in the majority of 46,XX males, it would be predicted that most sex-reversed 46,XY females would have SRY defects. However, in general this has not proven to be the case as mutations in SRY have been detected in only 15–20% of 46,XY sex-reversed individuals with complete or partial gonadal dysgenesis (more common in complete dysgenesis). Suggested explanations for this finding include the theoretical presence of mutations outside the HMG box, perhaps in sequences important for regulation of SRY expression, inactivating mutations in upstream regulators of SRY, mutations that induce constitutive activity in genes usually negatively regulated by SRY, or activating mutations in factors that suppress or interfere with SRY. A variety of cytogenetic and molecular abnormalities have been reported in individuals with 46,XY sex reversal, including deletion of Yp, isolated deletion of SRY, nonsense mutations resulting in protein truncations, and at least 25 missense mutations within the SRY HMG box (Fig. 41-5). Only one missense mutation has been reported outside the HMG box, a serine to asparagine change in the N-terminal region. Most SRY mutations in 46,XY sex reversal produce nonconservative AA substitutions at highly conserved sites within the HMG domain. In vitro analysis of mutant SRY proteins reveals abnormal DNA binding and bending: some mutants bind DNA normally but less avidly, whereas others

bind with near-normal affinity but bend the DNA to a different angle. Less dramatic missense mutations cause partial, rather than complete, gonadal dysgenesis, as the mutant SRY protein likely retains some function. SRY mutations also have been reported in a few cases of 46,XY true hermaphroditism. One example was a postzygotic somatic mutation evidenced by the finding of both wild-type and mutant SRY alleles in gonadal DNA but only the wild-type SRY sequence in leukocyte DNA. There may be phenotypic variation between 46,XY individuals harboring the same SRY mutation, ranging from sex-reversed to unaffected, both within and between kindreds, perhaps because of variable penetrance or influences of the genetic background. The most severe SRY defects appear to arise de novo, whereas the milder defects are compatible with fertility and thereby transmission to offspring. Germline mosaicism for the SRY mutation may be associated with fertility in the fathers of affected individuals. SRY Homeobox-Like Gene 9 (SOX9) Normal Function Sox9, like Sry, can induce testis formation when inserted into XX mouse embryos and appears to be equally as important as its brother Sry, in the process of testis development. The human SRY gene is located at 17q24-25, in a region termed sex reversal autosomal 1 and encodes a 509-AA protein with features of a transcription regulator––a DNA-binding HMG domain and two transcriptional activation domains, including a proline and glutamine-rich domain in its C-terminal region. In vitro deletion of this latter region destroys the transactivating function of the protein. Unlike SRY, SOX9 displays strong sequence conservation throughout mammalian evolution. Furthermore, the sequence similarity between SOX9 and SRY suggests a relationship between the two that may represent evolution from a dosage-dependent autosomal sex determination system to a dominant Y-chromosomal system. Sox9 is initially expressed in the genital ridges of both sexes in mice at low levels and is subsequently downregulated in female genital ridges and upregulated in male genital ridges, its expression paralleling Sertoli cell differentiation, consistent with a role in testis determination. Based on the timing of expression and the

418

SECTION IV / ENDOCRINOLOGY

Figure 41-6 Mutations in human SOX9.

presence of a potential binding site for Sry within its promoter, Sox9 appears to be acting just downstream of Sry and is likely a target gene for Sry. However, there is probably not an absolute requirement for Sry regulation of Sox9, because Sox9 can induce testis development in the absence of Sry. Sox9 expression precedes that of Amh and Sox9 appears to play the pivotal role (in conjunction with Sf1, Wt1, and Gata4; Fig. 41-3B) in activating Amh expression, a role befitting a key maleness-determining factor. Sox9 may represent another “switch”-like mechanism, as it appears to act in a dominant fashion, similar to Sry. Indeed, Sox9 may be the pivotal maleness-determining factor, potentially displacing Sry in this role––a battle whose outcome is awaited. Human SOX9 expression is detectable in the testis at week 18 of gestation in the area of the rete testis and seminiferous tubules. In the adult, SOX9 is expressed most strongly in the testis, pancreas, prostate, kidney, brain, and the skeleton and at low level in most other adult tissues. Mice with heterozygous Sox9 deletion had skeletal malformations equivalent to those seen in the human disease, campomelic dysplasia, and died soon after birth. However, the testes were normal in the male mutant mice in contrast to many human SOX9 mutations, well known for causing male-to-female sex reversal in the heterozygous state, suggesting differences in SOX9/Sox9 function between mice and humans. Overexpression of Sox9 in XX gonads is associated with testis development in Odsex mice, providing additional evidence for the importance of Sox9 in testis determination. Clinical Features and Molecular Defects SOX9 was identified by cloning of a chromosomal translocation breakpoint from a sex-reversed patient with campomelic dysplasia, a rare syndrome featuring a distinctive form of skeletal malformation that causes severe limb bowing (campomelia), accompanied in approx 75% of affected 46,XY individuals by sex reversal (ovaries or streak gonads, and female internal and external genitalia). Affected 46,XX individuals have normal ovarian development, reflecting lack of involvement of SOX9 in this process. Death typically occurs in early childhood from respiratory compromise related to the skeletal dysplasia. A milder form of the condition, in which the limbs are not bowed is referred to as “acampomelic” campomelic dysplasia. Heterozygous mutations in SOX9 have been identified in 46,XY individuals with campomelic dysplasia and sex reversal (Fig. 41-6). In general SOX9 mutations occur de novo and affect only one SOX9 allele, although one case of compound heterozygosity has been reported. Most disease-causing mutations severely disrupt the protein (e.g., premature termination and frameshift mutations) and the phenotype appears to result from loss of function of the

transcription factor (haploinsufficiency). Affected sibling pairs with normal parents have been reported, as a consequence of gonadal mosaicism for the mutant gene. As with SRY mutations, variation in phenotype has been reported within the same family. For example, in one family with three children bearing a framsehift mutation, one 46,XY child had true hermaphroditism with ambiguous external genitalia, whereas a 46,XY sibling had sex reversal with bilateral ovaries and female genitalia. In patients with the milder “acampomelic” form, missense mutations resulting in AA substitutions within the HMG domain have been reported. The genital abnormalities are also milder than in those with complete SOX9 inactivation, manifest in one such patient by a bifid scrotum, perineal hypospadias, and undescended right testis. A number of patients with 46,XY autosomal-dominant sex reversal and campomelic dysplasia have chromosome 17 breakpoints at least 50–130 kb from the SOX9 locus, suggesting the presence of other genes responsible for the same phenotype, or perhaps a disturbance of SOX9 expression resulting from mutation of an upstream regulator. Positional cloning of the chromosome 17q breakpoint in one patient with sex reversal and “acampomelic” dysplasia identified a 3.5-kb complementary DNA that is expressed in testis but appears not to be translated. In addition, cases with chromosomal rearrangements involving 17q have been described most likely affecting regulatory elements upstream of SOX9. Partial 46,XX female-to-male sex reversal has been reported in an infant with mosaicism for a chromosomal rearrangement resulting in duplication of SOX9. The child had severe penile/scrotal hypospadias, a small penis with descended palpable gonads in a bifid scrotum and absence of uterus on ultrasound. This finding suggests that an extra dose of SOX9 is sufficient to initiate testis differentiation in the absence of SRY in humans as well as mice. α-Thalassemia/Mental Retardation, X-Linked (ATRX) Normal Function α-Thalassemia/mental retardation, X-Linked (ATRX) (also known as X-linked helicase-2) is a helicase—an enzyme that catalyzes the unwinding of doublestranded nucleic acids—and is a member of a family of proteins involved in DNA recombination and repair, chromatin remodeling, chromosome segregation and regulation of transcription. The large (>200 kb, 35 exon) ATRX gene is located at Xq13.3 (and is also on mouse X chromosome) and is subject to alternate splicing in different tissues. There is a homologous gene on the Y chromosome of marsupial mammals (not present in mouse or human) that is expressed specifically in the testis, giving additional weight to its role in male sex determination. The protein contains a nuclear localization signal and three ZFs in the N-terminal region; the C-terminal region contains six helicase domains and a glutamine

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

rich sequence common in transcription factors. ATRX is thought to function by binding to DNA via the ZF region then opening the double helix with the helicase region in an ATPase-dependent manner. The protein is associated with pericentromeric chromatin during interphase and mitosis, suggesting that ATRX may act as part of a protein complex that modulates chromatin structure. ATRX is expressed in a wide range of embryonic and adult human tissues, including developing brain and testis. Because of its X-chromosomal location, the gene undergoes X-inactivation in females to retain dosage equivalence with males. As a single active copy in normal females is not associated with testis development, although a single copy is adequate for testis development in males, ATRX likely functions downstream of a male-specific transcription factor such as SRY or SOX9, requiring activation by such a factor. Apart from the fact that human mutations are associated with sex reversal, little is known about the function of this protein in sex determination and differentiation, as comprehensive studies in mice are lacking. Clinical Features and Molecular Defects The ATRX syndrome is an X-linked disorder characterized in 46,XY individuals by variable severity of α-thalassemia, psychomotor retardation, dysmorphic features, gonadal dysgenesis and undermasculinization. In sex-reversed patients the gonads are streaks, however, Müllerian ducts are absent, indicating that AMH expression, and therefore Sertoli cell development were retained during the critical period. Consequently, the critical period for ATRX expression appears to be after Sertoli cell development. Diverse mutations in the ATRX gene have been reported in individuals with the ATRX syndrome and several clinical variants of the syndrome reflect different mutations in the gene, including Smith-Fineman-Myers, α-thalassemia myelodysplasia and Juberg-Marsidi syndromes. There appear to be some genotype–phenotype correlations: mutations associated with sex reversal have mainly been those causing truncation of the protein with loss of the C-terminal region including the “P-element” (a 15AA region homologous to that found in other transcriptionally active proteins) and the polyglutamine tract; mutations within a helicase domain are associated with severe mental retardation without thalassemia; and mutations that alter residues in carboxyl terminus outside the helicase domains result in the classic ATRX syndrome. As with other sex reversal disorders, phenotype may vary within a given family, likely reflecting influence of genetic background. Desert Hedgehog (DHH) Normal Function Desert hedgehog (Dhh) is a signaling molecule involved in interactions between Sertoli cells and germ cells whose primary role appears to be regulation of spermatogenesis. The 3-exon human DHH gene located at chromosome 12q13.1 encodes a 396-AA polypeptide. In mice Dhh is expressed only in the testis and not in the ovary. Expression is initiated in Sertoli cell precursors shortly after activation of Sry expression, and persists to adulthood. This system appears to regulate mitosis and meiosis in male germ cells, its role varying at different stages of development. In the embryonic testis Dhh regulates germ cell proliferation and Leydig cell development, whereas in the postnatal testis it directs germ cell maturation. The receptor for Dhh, Patched2, is expressed on Leydig cells and peritubular cells, and one of the key roles of Dhh appears to be in the proper development of peritubular tissue. Consequently, the mechanism by which Dhh regulates germ cell development may be indirect, via other cell types, and may be secondary to the more general effects on organization of testicular structure.

419

In the initial murine studies, homozygous null male mice had small testes and were infertile because of lack of mature sperm. However, apart from the deficiency of germ cells, the testes were structurally and microscopically normal––including the Dhhproducing Sertoli cells themselves. In a subsequent study in mice from a different genetic background, more than 90% of null male mice were feminized. They had small, undescended, ectopically located testes and poor Leydig cell development with low serum testosterone, female external genitalia and a blind vagina. The latter study demonstrated requirement for Dhh for normal development of peritubular myoid cells and the basal lamina and subsequent well-organized development of seminiferous tubules. Thus Dhh may or may not be required for testicular organogenesis depending on genetic background (contrast the effect of a human DHH mutation described below) and there is phenotypic heterogeneity of the null mutation even within a single genetic background. Clinical Features and Molecular Defects DHH is included within the section on testis determination primarily on the basis of a single 46,XY patient who had partial gonadal dysgenesis as a result of a homozygous mutation at the initiation codon in exon 1 of DHH, with predicted a failure of protein translation. The young woman, whose parents were first cousins, presented for evaluation of primary amenorrhea and had poor breast development, immature female external genitalia, a blind vagina, and a form of polyneuropathy. Laparoscopy revealed a testis on one side and a streak gonad on the other. Serum testosterone was low and FSH was elevated. NORMAL AND ABNORMAL OVARY DETERMINATION The factors directing ovarian development are less well characterized than those governing testis development. However, there is presumably an equally complex network of controls guiding this process, which likely requires action on two fronts: repression of autosomal testis-inducing genes such as SOX9 (but not SRY, as this is absent from the normal XX embryo), and either derepression (removal of an inhibitory influence) or activation (direct stimulation) of ovary-inducing genes. Importantly, any factor that functions as a “testis inhibitor” must be expressed and active prior to the time that expression of autosomal testis-determining factors would otherwise occur. One such testis inhibitor likely resides on the X chromosome and is probably a dosage sensitive locus, expressed from both X chromosomes in females. A former candidate for this role was DAX1, based on the finding of streak ovaries in place of testes in individuals with duplication of the X-chromosomal region containing the gene. However, DAX1 became less convincing as the “antitestis” gene when a female with homozygous DAX1 deletion was reported. Current prime candidates for roles as testis inhibitors are Wnt4, follistatin and “Odsex,” a locus on chromosome 17 upstream of Sox9 that may contain a gonad-specific Sox9 regulatory sequence. No candidate for the role of a positive “ovarydetermining factor” has been proposed; however, such a gene is probably also a dosage sensitive locus (therefore likely X chromosomal) required for germ cell survival in the ovarian milieu. This is suggested by the findings in individuals with Turner syndrome, who, in the absence of two functional copies of the X chromosome have ovarian regression because of early fetal demise of germ cells. It is therefore reasonable to speculate that initial ovarian differentiation occurs under the influence of a factor (or factors) expressed in the absence of the Y chromosome, and that subsequently, dosagesensitive X chromosomal genes (required in double copy) are responsible for ovarian maintenance. No clearly defined sequence of steps in the pathway of ovarian development has been established and this is likely a complex,

420

SECTION IV / ENDOCRINOLOGY

Figure 41-7 Hypothetical model of ovary determination. Wnt4 secreted by somatic cells increases expression of Fst, Dax1 and/or other factor(s) that antagonize Sf1 and repress Sox9 (perhaps by binding to the Odsex locus upstream of Sox9). Thus bipotential supporting cells do not differentiate as Sertoli cells, but instead as granulosa cells. Paracrine interactions between granulosa cells, secreting Foxl2 and Fst, and germ cells secreting Gdf9 and Bmp15, promote development of ovarian follicles. Wnt4 and Fst together inhibit activin-driven development of testicular (coelomic) vasculature, and Wnt4 inhibits differentiation of bipotential interstitial cells to Leydig cells. Because the XX fetal gonad does not produce sex steroids, the Wolffian ducts regress and external genitalia develop as female. Sertoli cells do not develop, therefore Amh is not produced, so Müllerian structures are maintained and subsequently differentiate under the influence of Hoxa13 and other factors. AMH, anti-Müllerian hormone; BMP15, bone morphogenetic protein 15; DAX1, dosage-sensitive sex reversal-adrenal hypoplasia congenita locus on the X-chromosome, gene 1; EMX2, empty spiracles 2; FOXL2, forkhead transcription factor 2; FST, follistatin; GDF, growth differentiation factor; Hoxa13, hoemobox A 13; LHX9, LIM homeobox gene 9; LIM1, LIM homeobox gene 1; SF1, steroidogenic factor 1; SOX, Sry-related homeobox gene; WNT, wingless-type MMTV integration site family member.

nonlinear, set of interconnected up- and downregulatory events. A speculative model, based on some of the known and hypothetical genetic events, mainly in mice, is as follows: 1. An unknown factor induces expression of Wnt4 by somatic cells. 2. Wnt4 increases expression of Dax1, Fst and/or other factor(s) that antagonize Sf1 and repress Sox9 (perhaps by binding to the Odsex locus). 3. Bipotential supporting cells do not differentiate as Sertoli cells, but instead as granulosa cells. 4. Paracrine interactions between granulosa cells, secreting Foxl2 and Fst, and germ cells secreting Gdf9 and bone morphogenetic protein 15 (Bmp15), promote development of ovarian follicles. 5. Wnt4 and Fst together inhibit activin-driven development of testicular (coelomic) vasculature. 6. Wnt4 inhibits differentiation of bipotential interstitial cells to Leydig cells. 7. The fetal gonad does not produce sex steroids, therefore the Wolffian ducts regress and external genitalia develop as female. 8. Because Sertoli cells do not develop, Amh is not produced. 9. Müllerian structures are maintained and subsequently differentiate under the influence of homeobox A 13 (Hoxa13) and other factors (Fig. 41-7). Wingless-Type MMTV Integration Site Family Gene 4 (WNT4) Normal Function Wnt genes belong to a family of protooncogenes with at least 16 known mammalian members

expressed in species ranging from Drosophila to man. These genes encode 38- to 43-kDa cysteine-rich glycoproteins with features typical of secreted growth factors (extracellular signaling molecules): a hydrophobic signal sequence and 21 conserved cysteines whose relative spacing is maintained. Transcription of Wnt genes appears to be developmentally regulated in a precise temporospatial manner. Interaction of Wnt factors with Frizzled receptors (members of the seven-transmembrane domain family) results in an intracellular signal cascade that in turn leads to transcriptional activation of target genes. By these interactions the Wnts regulate cell proliferation, morphology and fate. Members of this family are involved in sex determination and differentiation at various levels, including wingless-type MMTV integration site family gene 4 (WNT4) (ovary), WNT5A (external genitalia) and WNT7A (Müllerian duct regression, described later). Wnt4 is required for the initial stages of Müllerian duct formation in both sexes in mice and appears to be required for normal ovarian development in females, making it a prime candidate for an “anti-testis” gene, interacting with Fst and perhaps Dax1 in this role (Wnt4 regulates expression of both factors). Wnt4 expression is first detected in the mesonephric mesenchyme, which later forms the bipotential gonad, and is subsequently expressed in the indifferent gonads of both sexes. At the onset of sex-specific gonadal differentiation, Wnt4 expression is downregulated in the male (probably by Sry/Sox9), but in the female is maintained within somatic cell lineages throughout fetal life. Wnt4 is also strongly expressed in the mesenchyme underlying the Müllerian ducts, and in the adrenal cortex of both sexes, consistent with a role in regulation of steroidogenesis.

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

Homozygous Wnt4-null mice of both sexes had failure of differentiation of kidney mesenchyme and absence of Müllerian ducts in early gestation. Most striking were the specific effects in female mice, which had masculinization of the gonad and development of Wolffian ducts (the latter because of activation of gonadal steroidogenesis by Leydig cell precursors). Thus, Wnt4 acts in ovarian development by suppressing the differentiation of primordial interstitial cells into Leydig cells, thereby inhibiting testosterone synthesis and allowing female development to occur. Of note, the mice had no masculinization of the external genitalia, probably because testosterone was produced in insufficient quantities from the mutant gonads. Wnt4 further acts by stimulating Fst expression and these two factors act in concert to suppress/inhibit activin B, which would otherwise stimulate development of the testicular coelomic vasculature. The ovaries of the XX Wnt4-null mice had accelerated loss of oocytes to 40 patients with 17β-HSD deficiency. The majority are missense mutations resulting in AA substitutions in various regions of the protein; splice site mutations are also common. Phenotypic variability among individuals with the same mutation is well described. Identical mutations have been found within kindreds from widely distant regions of the world, suggesting genetic founder effects. Specific mutations tend to segregate with certain ethnicities, so knowledge of the patient’s ethnic background may be helpful in predicting the expected mutation. Like other autosomal-recessive disorders, rates of 17β-HSD deficiency are higher in populations with consanguinity. Compared with a rate of approx 1/140,000 in the Netherlands, Arabs in Gaza have a prevalence of 1/200–300. Homozygosity for a mutation that converts arginine 80 to glutamine was reported in 24 individuals from 9 such extended Arab families. This substitution impairs enzyme activity by increasing the NADPH cofactor binding constant 60-fold. One 46,XY phenotypic female had compound heterozygosity for two distinct AA substitutions and another had compound heterozygosity for a splice acceptor mutation and a missense mutation. When expressed in cultured mammalian cells in vitro the mutant enzymes displayed impaired enzyme activity. Proteins and Enzymes of Androgen Action* 5α-Reductase Normal Function The 5α-reductases are microsomal enzymes that catalyze the 5α-reduction of many C19 and C21 steroids, utilizing NADPH as a cofactor. There are two isozymes of 5α-reductase (they share approx 50% AA identity) encoded by separate genes, 5α-reductase 1 by SRD5A1 at 5p15, and the enzyme important in the context of sexual differentiation, 5αreductase 2, a 254-AA protein encoded by SRD5A2 located at 2p23. The genes are structurally similar, with five coding exons each. 5α-reductase 2 mediates the conversion of testosterone to more potent androgen, DHT in androgenic target tissues. The 5α-reductase isozymes have differential expression patterns: the type 1 isozyme is not detectable in the fetus, but is transiently expressed in newborn scalp and skin and postpubertal skin and is permanently expressed in the liver; the type 2 isozyme is expressed in the liver and in androgen target tissues, including the external genitalia, accessory sex organs, and prostate. It is expressed in the primordia of the prostate and external genitalia before their differentiation, but is not expressed in the Wolffian ducts untill after their differentiation, supporting the contention that testosterone rather than DHT is the critical androgen in this process. Expression is upregulated by androgens, as demonstrated by the marked increase in SRD5A2 mRNA level in the prostate of castrate animals following testosterone administration. Expression appears to be regulated in the opposite fashion in the liver. Clinical Features and Molecular Defects Deficiency of 5α-reductase 2 results in an autosomal-recessive form of male pseudohermaphroditism previously referred to as pseudovaginal *These proteins and their associated disorders are discussed in detail in Chapter 44.

perineoscrotal hypospadias. Clusters of affected patients have been reported in consanguineous populations in the Dominican Republic, Pakistan, Lebanon and New Guinea. Deficiency of 5αreductase 2 in the tissues of the fetal external genitalia and urogenital sinus results in inadequate local DHT concentrations. As in other forms of defective androgen production or action, there is subnormal masculinization of these structures. The genital phenotype varies widely among and even within affected kindreds, the most consistent findings being underdevelopment of the penis and prostate. Most patients have ambiguous external genitalia, with testes located in the inguinal region or a bifid scrotum, accompanied by a blind vaginal pouch or urogenital sinus. Because Sertoli cell AMH secretion is normal, Müllerian structures are absent. It is on the basis of these clinical findings, that the requirement for DHT in external genital masculinization was initially inferred. Because differentiation of the gonads and Wolffian ducts depends on high local concentrations of testosterone rather than DHT, the testes, epididymides, vasa deferentia, and seminal vesicles develop normally. Reduced conversion of testosterone to DHT in target tissues results in a marked increase in testosterone: DHT ratio diagnostic of 5α-reductase 2 deficiency. In normal infants (2 wk–6 mo of age), the testosterone:DHT ratio is 4.8 ± 2.2 (mean ± standard deviation) following stimulation with hCG. Affected infants have markedly increased testosterone:DHT ratios, in the 20–60 range or higher. Serum LH may be mildly increased. 5α-reductase 2 deficiency can be differentiated endocrinologically from 17βHSD deficiency (with which it shares phenotypic similarity) by the characteristically increased serum androstenedione in the latter condition. One of the most intriguing and well-documented features of this disorder is the striking virilization, including increased muscularity and deepening of the voice, that occurs at puberty in many affected individuals. In addition, the testes often enlarge, descend and develop Leydig cell hyperplasia, although spermatogenesis is absent or severely impaired. LH is normal or elevated, and testosterone levels increase into the adult male range, whereas DHT remains disproportionately low, but measurable. The testosterone to DHT ratio may be in the range of 30–80 (compared with the normal ratio of approx 9–15 at this age). The pubertal virilization is thought to result from both the high testosterone concentrations and from increased conversion of testosterone to DHT by 5αreductase 1 in liver and skin. Acne, facial hair, temporal hair recession, and prostatic enlargement do not develop, presumably because these events require higher concentrations of DHT. Gynecomastia does not develop because there is no increase in testicular estrogen production. Many affected individuals initially raised as females, change their gender role to male around the time of puberty. At least 30 different mutations have been identified throughout the SRD5A2 gene in affected families from >20 different ethnic groups, ranging from complete SRD5A2 deletion in a New Guinea kindred, to single base mutations resulting in gene splicing defects, premature termination or AA substitutions (the majority). Deletions, premature termination codons and splice junction defects prevent expression of a functional enzyme. AA substitutions may produce an unstable enzyme or impair binding of testosterone and/or NADPH. Substitutions at the N- or C-terminal ends of the molecule are associated with defective testosterone binding, whereas reduced NADPH binding affinity occurs only with

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

C-terminal substitutions. Differences in enzyme stability and affinity for testosterone and NADPH among kindreds reflect the genetic heterogeneity of the enzyme defects, but there is little relationship between the severity of the mutation and phenotype. Androgen Receptor (AR) Normal Function The AR is a ligand-activated nuclear transcription factor that mediates the effects of androgens in induction of target gene transcription in androgen-dependent tissues. The 8-exon AR gene at Xq11-12 encodes a 110-kDa, 910–919 AA protein consisting of three major functional domains: N-terminal (transcription-regulating), DNA-binding (ZF) and steroid-binding. The AR is expressed in a wide array of genital and nongenital tissues, reflecting its role as a widespread transcription factor. Activation of AR by androgen binding results in interaction of the receptor/ligand complex as a homodimer with androgen response element DNA sequences in the promoter regions of target genes, to regulate their transcription. The exact targets of the AR in genital development remain to be determined, although the AMH receptor appears to be one of them. In external genital tissues, testosterone is converted by 5α-reductase 2 to DHT, which has greater affinity for the AR. Nevertheless, the same molecular events occur following interaction of either testosterone or DHT with AR. The female fetus bears the same AR as the male, thus it is primarily the available concentration of androgens that is the major determinant of genital masculinization. Because Chapter 44 is devoted exclusively to the molecular biology and physiology of the AR, discussion of this important factor in this chapter is limited. Clinical Features and Molecular Defects Absence or defective function of the AR results in resistance to the effects of androgens, manifest clinically as AIS (previously referred to as testicular feminization), a heterogeneous condition thought to represent the single most common identifiable cause of male pseudohermaphroditism. Affected individuals have normal testicular function, but have variable defects of internal and external genital masculinization, associated with partial retention of Müllerian structures in some cases. In complete AIS (CAIS), which has a prevalence of approx 1/20,000 46,XY births, not only is the phenotype unequivocally female, the labia minora and majora and clitoris may be underdeveloped, suggesting involvement of low levels of androgen action in normal genital development in females. Infants with CAIS fail to demonstrate the postnatal surge of LH and testosterone seen in normal male infants (and retained in those with partial AIS) and have blunted LH responses to exogenous gonadotropin-releasing hormone. This failure of the normal infantile “mini-puberty” may reflect lack of prenatal “priming” of the hypothalamus due to absence of hypothalamic AR. At puberty, LH, testosterone and DHT increase to supranormal levels and serum estrogen concentrations are also enhanced, as a result of LH-driven testicular estrogen secretion and peripheral aromatization of testosterone. AMH levels are also supranormal during the first year of life and at puberty, a finding that may help to differentiate AIS from intersex conditions caused by testicular dysgenesis, in which AMH is low. Pubertal development in CAIS is entirely female, accompanied by absence of pubic and axillary hair. Infants with partial forms of AIS (PAIS) have widely varying degrees of genital masculinization and the clinical diagnosis is notoriously difficult, as hormonal profiles are often inconclusive.

437

The diagnosis of PAIS, should be suspected in a 46,XY infant with ambiguous genitalia if the classic elevations of LH and testosterone are present and if the responses of testosterone precursors, testosterone and DHT to exogenous hCG are normal or elevated. In contrast to those with CAIS, infants with PAIS have a vigorous “mini puberty” at approx 6–12 wk of age, as seen in normal male infants, with normal or supranormal LH and testosterone concentrations. Depending on the severity of the androgen resistance, varying degrees of virilization and/or feminization occur at puberty. The prevalence of PAIS is unknown. Demonstration of abnormal androgen binding in cultured genital skin fibroblasts or identification of a mutation in the AR gene of affected individuals confirms the diagnosis, but is generally impractical outside of research institutions. Furthermore, receptor levels in genital skin fibroblasts and androgen binding affinity often correlate poorly with the degree of masculinization and may be falsely negative if the defect falls outside the steroid-binding domain. In addition, in many cases of PAIS complete sequencing of the AR gene fails to reveal a mutation. Over 300 distinct mutations have been reported in the AR genes of hundreds of unrelated individuals with various forms of AIS*. Like the clinical spectrum, the molecular defects are highly heterogeneous. Protein-disrupting defects such as complete and partial gene deletions (rare), small insertions and deletions of a few base pairs, splice junction mutations and premature termination codons, are found in patients with CAIS. Missense mutations causing AA substitutions, the most common mutational type, are found in both CAIS and PAIS. The majority of missense mutations are located in the exons encoding the steroid-binding domain, with all but a few of the remainder reported in the exons encoding the DNA-binding domain. Mutations within the steroid-binding domain cause variable defects of androgen binding––absent, reduced (affinity or capacity or both), or qualitatively abnormal (increased thermolability of binding, ligand dissociation or altered binding specificity). Mutations within the DNA-binding domain typically alter the affinity, capacity or specificity of AR binding to androgen response element DNA sequences, without significant effect on androgen binding. Whether the mutation affects DNA or androgen binding, the final common pathway of AR defects is reduction/loss of ability of the mutant receptor to regulate transcription of androgen-dependent target genes. The AR gene appears to be particularly mutation-prone, and approx 25% of mutations arise de novo. A few individuals have been identified as having more than one AR gene mutation. Altlhough AR mutations can be demonstrated in almost all patients with CAIS, the same is not true for individuals with the clinical diagnosis of PAIS, of whom only approx 1/3 have a detectable alteration of the AR coding sequence. This may in part reflect inaccurate clinical differentiation between PAIS and other causes of male pseudohermaphroditism (e.g., 17α-hydroxylase deficiency) or may represent defects within noncoding regions of the AR gene or in cofactors required for normal AR action. Unfortunately, no genotype/phenotype correlation exists for AR mutations, and phenotype may differ widely among individuals with the same mutation, even within the same family (Fig. 41-8). Somatic mosaicism for both a mutant and a normal AR may account

* see www.mcgill.ca/androgendb.

438

SECTION IV / ENDOCRINOLOGY

Figure 41-10 Interactions between Amh, its receptor and other transcription factors in induction of Müllerian regression during male embryogenesis. Within the fetal testis, a synergistic combination of transcription factors including Wt1, Sf1, Sox9, and GATA4, stimulates Sertoli cell production of Amh. Meanwhile, Leydig cells produce testosterone, under the stimulatory influence of Sf1 and other factors. The primitive Müllerian ducts consist of epithelial and mesenchmal elements. Epithelial cells secrete Wnt7a––an extracellular signaling molecule. Wnt7a interacts with Sf1 and the androgen receptor (activated by testosterone secreted by Legdig cells) to stimulate transcription of the Amh receptor (Amhr) gene and its expression by the mesenchymal cells surrounding the duct. Binding of Sertoli cell-derived amh to the amhr stimulates apoptosis of the Müllerian epithelium leading to obliteration of the duct lumen, progressing in a cranio-caudal direction down the duct, mediated by a paracrine “death factor” that may be a product of the activated Amhr. This process occurs during weeks 9–11 of human male gestation. Amh, anti-Müllerian hormone; Amhr, anti-Müllerian hormone receptor; AR, androgen receptor; MMP2, matrix metalloprotease 2; Sf1, steroidogenic factor 1; Sox9, Sry-related homeobox gene 9; Wnt7a, wingless-type MMTV integration site family member 7a; Wt1, Wilms tumor 1.

for phenotypic variability in some cases, and may have clinical relevance, because there may be a significant response to exogenous androgens in the cell populations carrying the normal AR. More complete analysis of AR mutations is provided in Chapter 44. Mediators of Müllerian Duct Regression Anti-Müllerian Hormone (AMH) Normal Function Anti-Müllerian hormone (AMH or Müllerian inhibiting substance), produced by testicular Sertoli cells, is a member of the TGF-β family of growth factors that mediates regression of the Müllerian duct structures during normal male embryogenesis. The 5-exon, 2.75-kb AMH gene is located on human chromosome 19p13.3-13.2 and encodes a 560-AA glycoprotein that forms a 140-kDa homodimer. AMH is the first molecular marker after SRY specific for testis (Sertoli cell) differentiation, its expression in developing mouse testis beginning approx 2 d later than that of Sry, at approximately E13.5. Amh transcription appears to be initiated by Sox9 and subsequently upregulated by coordinated interactions at the Amh promoter between Sf1 and Sox9, Wt1, Gata4, and perhaps Dmrt1 (Fig. 41-3B). There is also some evidence suggesting upregulation of Amh expression by FSH and downregulation by androgens. The primary effect of AMH is to induce regression of Müllerian ducts during male sex differentiation (9–11 wk human gestation). These events occur in a paracrine fashion, with AMH secretion from one testis mediating only the regression of the ipsilateral Müllerian duct. Duct regression proceeds in a cranialto-caudal direction associated with a cranial-to-caudal gradient of AMHR2 (discussed later). This is followed by a wave of apoptosis and accumulation of β-catenin, an adherens junction protein (a factor that mediates cell–cell adhesion) within peri-Müllerian mesenchymal cells. Müllerian epithelial apoptosis seems to be

mediated by an extracellular proteinase—matrix metalloproteinase 2—a downstream product of AMH/AMHR2 signalling that functions as a paracrine “death factor” (Fig. 41-10). AMH is produced by Sertoli cells not only during the critical period of sex differentiation, but also in late gestation, after birth, and even, albeit at a much reduced rate, in adulthood, suggesting that AMH may have physiological roles other than its control of Müllerian duct regression. After puberty, AMH production is downregulated by androgens (this feature is absent in individuals with androgen insensitivity). AMH is not expressed prenatally in the ovary. However, low amounts are released into the follicular fluid by mature granulosa cells postnatally and one postulated role of ovarian AMH is to inhibit recruitment of primordial follicles into the pool of growing follicles, via effects on granulosa and theca cells, not on the oocytes themselves. The net effect is to allow for enhanced oocyte maturation prior to selection for ovulation. As expected from the known function of this protein, male mice with homozygous deletion of Amh had normal male reproductive tracts, but also had uteri and oviducts. However, mice with mutations in the Amh promoter that affect the regulation of Amh transcription had variable phenotypes. Mutation at the Sf1 binding site was associated with reduced Amh expression, but expression that was nevertheless adequate to induce Müllerian regression. In contrast, male mice homozygous for a mutant Sox9 binding site did not initiate Amh transcription, resulting in complete retention of Müllerian ducts. Female mice overexpressing human AMH had no uteri or fallopian tubes and blind-ending vaginas. The ovaries were devoid of germ cells and underwent reorganization into seminiferous tubulelike structures after birth. Males were also abnormal, with undervirilized external genitalia and impaired Wolffian duct development

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

associated with undescended testes notable for Leydig cell hypoplasia due to effects of excess AMH on Leydig cell development. Clinical Features and Molecular Defects Deficiency of AMH or abnormality of its receptor (described below) results in the persistent Müllerian duct syndromes (PMDS type 1 [AMH] and PMDS type 2 [AMHR2]), which are inherited in an autosomalrecessive, male-limited, pattern. Affected karyotypic males have normal male external genitalia, normal testicular histology and normal Wolffian duct differentiation, but have retained and differentiated Müllerian ducts and abnormalities of testicular descent. The typical case is that of a phenotypically normal male infant with unilateral ectopic testis and an inguinal hernia on the contralateral side. One testis descends into the scrotum and the ipsilateral fallopian tube and uterus follow it into the inguinal canal (hernia uteri inguinalis), bringing the contralateral testis along (transverse testicular ectopia). The Müllerian structures usually are detected only at the time of surgery for hernia repair. A less common clinical presentation is with bilateral cryptorchidism and inguinal hernias. In this situation the uterus is fixed in the pelvis and the testes are embedded in the broad ligament. These two clinical variants can occur within the same family. Serum testosterone is normal but AMH, which is readily measurable in the serum of normal males until puberty, is reduced. Because of the increased mobility of the testes, they are prone to torsion. In addition, there may be associated aplasia of the epididymis and upper part of the vas deferens. These problems, along with the often difficult task of attempting to relocate the testes, combine to impair fertility in most patients with PMDS. Because the testes are invariably undescended in PMDS it has been proposed that AMH is involved in the process of human testicular descent. However, this speculation is tempered by the finding of variable AMH levels in boys with simple cryptorchidism and of normal gubernacular development in Amh-null mice. Testicular tumors have occasionally been reported. Diverse AMH mutations have been identified in many patients with PMDS with low or undetectable serum AMH. Mutations tend to cluster in the exons encoding the N- and C-terminal regions of the protein (the carboxyl terminus is considered the biologically active region of the molecule) and are of various types including splicing mutations, nonsense mutations and missense mutations, which represent the majority. The fact that AMH molecules containing an AA substitution cannot be detected by standard enzyme-linked immunosorbent assay (ELISA) suggests that even a single AA substitution alters the structure of the protein enough to interfere with binding to the ELISA antibody. In the largest single study of 21 families, most patients were homozygous for AMH gene mutations, as the patients derived from consanguineous Arab or Mediterranean communities. About 40% of mutations have been reported in more than one family, but whether these represent a founder effect, or a mutational “hot spot” is not clear. Anti-Müllerian Hormone Receptor Type 2 (AMHR2) Normal Function Anti-Müllerian hormone receptor, type 2 (AMHR2) is a membrane-bound serine/threonine kinase similar to those for TGF-β and activin. The 11-exon, 8-kb gene encoding human the 573-AA AMHR2 protein is located at 12q13. Exons 1–3 encode a short signal sequence and the extracellular domain of the receptor, exon 4 encodes most of the transmembrane domain, and exons 5–11 encode the intracellular serine/threonine kinase domains. AMH binding to AMHR2 in target tissues leads to phosphorylation of a TGF-β type I receptor and the resulting intracellular signal cascade that leads to target gene activation and eventually to apoptosis of Müllerian cells. Candidates for the role

439

of the specific type I receptor with which the type 2 receptor interacts include activin receptor-like kinase 2 (ALK2) and the bone morphogenetic protein receptor 1a, also known as ALK3. Amhr2 is expressed in mesenchymal cells adjacent to the Müllerian ducts during mouse embryogenesis, consistent with the fact that the effects of Amh appear to be via changes in the mesenchyme surrounding the Müllerian ducts which in turn induces apoptosis of the Müllerian epithelium and duct regression. Amh causes regression of the cranial part of the Müllerian duct while it continues to grow caudally, associated with a cranial-to-caudal gradient of Amhr2 protein, followed by a wave of apoptosis spreading caudally along the Müllerian duct. Amhr2 expression can also be detected in fetal Sertoli and Leydig cells and in fetal and adult granulosa cells. There is evidence that AMH action on Leydig cells causes downregulation of testosterone production. Like AMH itself, the receptor is also expressed postpubertally. Expression of both ligand and receptor in the same cells suggests an autocrine action. AMHR2 expression is enhanced by SF1 (which also regulates expression of AMH itself) and WNT7a and appears to be downregulated by the AR. The receptor may also be regulated by its ligand. Expression of the receptor in female Müllerian mesenchyme does not result in duct regression, because Amh is absent in females prenatally. Male mice with homozygous deletion of Amhr2 had a similar internal genital phenotype to those with Amh deletion: a complete set of male and female internal ducts. The phenotype of Amh/Amhr2 double-knockout mutant males was indistinguishable from that of either single mutant. Clinical Features and Molecular Defects Apart from the fact that serum AMH concentrations are normal or high, the clinical phenotype of patients with AMHR2 mutations (PMDS type 2) is indistinguishable from that of patients with AMH deficiency (PMDS type 1). Indeed, approx 1/2 of patients with the clinical findings of PMDS have mutations in the ligand and half in the receptor. The single most common mutation in the AMHR2 gene is a 27-bp deletion in exon 10. Of the more than 20 different reported defects, others include nonsense mutations causing truncation of the receptor to varying degrees, and a variety of missense mutations causing AA substitutions in the extracellular or intracellular domain. As with AMH mutations, most AMHR2 gene mutations are found in the homozygous state and about 1/3 are recurrent, having been reported in more than one family. In a minority of patients, no mutation of either the hormone or its receptor can be detected, and these cases may represent good candidates for analysis of the type 1 AMH receptor required for effective AMHR2 signaling. Wingless-Type MMTV Integration Site Family Member 7a (WNT7a) WNT7a is another member of the WNT glycoprotein growth factor/signaling molecule family (page 420) involved in regulation of cell fate and patterning. Wnt7a is expressed along the length of the Müllerian epithelium in mice of both sexes but appears to have differing roles in male vs female development. Wnt7a is expressed in Müllerian epithelial cells from E12.5-14.5, whereas Amhr2 is expressed in the mesenchymal cells surrounding the Müllerian ducts at day 14.5. It is hypothesized that the Wnt7a signal from the epithelial cells regulates Amhr2 expression in the adjacent mesenchyme (Fig. 41-10). Wnt7a expression declines following Müllerian duct regression in the male. Male mice with homozygous deficiency of Wnt7a have retained Müllerian ducts, and do not express Amhr2 in the mesenchyme of the Müllerian ducts. Thus, Wnt7a deficiency results in

440

SECTION IV / ENDOCRINOLOGY

failure of Amhr2 expression and therefore failure of Müllerian duct regression in males. In contrast, female Wnt7a-deficient mice have defective, though not absent, development of the oviducts and uterus. The human gene is expressed in placenta, kidney, testis, uterus, fetal lung, and fetal and adult brain but no human disease has been reported in association with this gene. NORMAL AND ABNORMAL FEMALE SEX DIFFERENTIATION Beyond the fact that normal female external genital development requires no estrogen and minimal androgen, little is known about factors directly involved in female sex differentiation. Normal internal reproductive duct development requires persistence and differentiation of the Müllerian ducts into the uterus, upper vagina and fallopian tubes, and regression of the Wolffian ducts. Thus, AMH and testosterone action must be absent. The finding of familial disorders of reproductive tract development in females implies that genetic factors must indeed play a role in either stimulating Müllerian development, or inhibiting its repression. A small amount of information from murine and human studies of HOX genes has emerged. Individuals with 46,XX pseudohermaphroditism develop ovaries but have varying degrees of masculinization of internal and external genital structures. There are far fewer known causes of virilization of a female fetus than there are of undermasculinization of a male fetus. Essentially, disorders of female sex differentiation represent the converse of male pseudohermaphroditism, in general resulting from prenatal exposure to excessive androgen concentrations. In addition to abnormal fetal androgen secretion, maternal androgen excess caused by androgen secreting tumors, drugs or medications with androgenic activity can result in masculinization of the female fetus. These latter problems are not discussed. Abnormalities of Müllerian or vaginal development, such as the Mayer-Rokitansky-Küster-Hauser (MRKH) syndrome, vaginal atresia and Müllerian agenesis, although technically not forms of pseudohermaphroditism because there is no masculinization, are nevertheless significant disorders of female sex differentiation and are also discussed briefly. Disorders of Female Sex Differentiation Due to Excess Androgen Production The most common cause of excessive fetal androgen production in females is CAH due to deficiency of 21-hydroxylase, 11-hydroxylase, POR (discussed earlier), or 3β-HSD (discussed earlier). A much rarer cause of female pseudohermaphroditism is P450arom deficiency in which fetal virilization results from high levels of testosterone and androstenedione because of the inability of the placenta to convert these precursors to estrogens. 21-Hydroxylase (see Chapter 37 for additional details) Normal Function P450c21 catalyzes the hydroxylation of progesterone to DOC and 17-OHP to 11-deoxycortisol, using POR as a cofactor (Fig. 41-9). P450c21 is a 55-kDa protein encoded by the10-exon CYP21 gene (CYP21B or CYP21A2) located at chromosome 6 p21.3 within approx 2 centimorgans of the human leukocyte antigen complex, with which it has tight linkage. There is a homologous pseudogene, designated CYP21P (or CYP21A), located nearby, that contains three deleterious mutations: an 8-basepair deletion in exon 3, a T insertion in exon 7, and a stop codon in exon 8. Clinical Features and Molecular Defects Deficiency of P450c21 is the most common cause of CAH in both males and females. Inheritance is autosomal-recessive and the prevalence of the disorder ranges from 1/15,000 (some white populations) to 1/700 (Yupik Eskimo tribe). Hydroxylation is impaired in the zona

fasciculata of the adrenal glands, so that 17-OHP is not converted to 11-deoxycortisol, the immediate precursor of cortisol. Cortisol deficiency leads to a compensatory increase in ACTH secretion that in turn drives the adrenal gland, resulting in overproduction of cortisol precursors, particularly 17-OHP. These steroids are “shunted” down the remaining functional steroidogenic pathway, producing a surfeit of androgens (mainly testosterone) that result in virilization of 46,XX fetuses but have no untoward effect on the male fetus. In more than half of cases, 21-hydroxylase is also deficient in the zona glomerulosa and there is failure of conversion of progesterone to 11-DOC, resulting in deficiency of aldosterone. Shock or death may result from severe salt wasting and accompanying hypovolemia. There are four major clinical forms of 21-hydroxylase deficiency: salt-wasting (approx 75% of cases), simple virilizing, nonclassic late onset (also called attenuated or acquired), and cryptic (asymptomatic). Only the two most severe forms, those that produce genital ambiguity, are discussed. Female infants with the classic virilizing form of 21-hydroxylase deficiency have masculinization of the genitalia, ranging from limited clitoromegaly and posterior labial fusion in the milder cases, to intermediate phenotypes with labioscrotal fusion, a urogenital sinus and significant clitoral enlargement, to severely virilized genitalia that are indistinguishable from those of a cryptorchid male infant. The gonads are normal ovaries, and Müllerian duct derivatives develop normally, reflecting absence of AMH. Even in the most virilized infants, Wolffian development does not occur, perhaps because of differences in timing of adrenal vs gonadal androgen secretion. Untreated children have penile or clitoral enlargement, premature adrenarche, rapid linear growth and advanced skeletal maturation, ultimately leading to early epiphyseal closure and short stature. The virilizing form of 21-hydroxylase deficiency is the most likely diagnosis in an infant with genital ambiguity in the presence of 46,XX karyotype and should be suspected in any partially virilized infant in whom gonads cannot be located, either by palpation or by ultrasonography. The presence of a uterus enhances the likelihood of this diagnosis. Increased serum concentrations of 17-OHP are invariably present in infants with enzyme deficiency severe enough to cause virilization; this is accompanied by hyponatremia and hyperkalemia in those with the salt-wasting form. An ACTH stimulation test is not necessary to make this diagnosis in a virilized female infant. Molecular defects in CYP21B have been elucidated in a large number of patients with 21-hydroxylase deficiency. Many appear to have arisen as a result of recombination events between CYP21B and its homologous pseudogene, CYP21P, and cause either deletion of CYP21 (approx 20% of salt-wasting cases), or transfer of mutations from the pseudogene to the functional gene. This process, termed “gene conversion,” is suggested to account for the predominance of 21-hydroxylase deficiency over other forms of CAH. Single base mutations are also common. In a comprehensive study of 88 families the most common mutation was an A-G change in the second intron affecting mRNA splicing (26%); large deletions occurred in approx 21%; substitution of isoleucine 172 by asparagine was found in 16% and replacement of valine 281 by leucine in 11%. Homozygosity for severe mutations is present in approx 50% of those with classic saltwasting disease and compound heterozygosity is also common. The clinical and enzymatic findings of such patients reflect the combined effect of mutations in each allele. In vitro analysis of mutant 21-hydroxylase enzymes has been performed in a number of cases. A mutant containing threonine at

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

position 428 in place of cysteine has complete loss of enzymatic activity and heme binding. Of note, cysteine 428 is invariant among all P450 enzymes and is thought to be the heme-binding site. Other AA substitutions depress enzyme activity to varying degrees. Weak correlations exist between enzyme activity (phenotype) and mutation (genotype). 11β-Hydroxylase Normal Function There are two cytochrome P450 isozymes with 11β-hydroxylase activity involved in the latter steps of cortisol and aldosterone production. Regulated by ACTH, P450C11 (P450XIB1) catalyzes the final step in cortisol synthesis, 11β-hydroxylation of 11-deoxycortisol to cortisol in the zona fasciculata of the adrenal gland. The aldosterone synthase isozyme (P450AS, P450aldo, P450cmo, P450XIB2), regulated by the renin-angiotensin system, catalyzes the three-step conversion of 11-DOC to aldosterone (11-hydroxylation, 18-hydroxylation, 18-oxidation) in the zona glomerulosa. Both enzymes use POR as a cofactor. Two genes encode 93% homologous 479- to 503-AA enzymes: the 6.5-kb, 9-exon CYP11B1 gene localized to 8q22 encodes P450C11; a contiguous gene, CYP11B2, located approx 40 kb away within the same chromosomal locus, encodes P450AS. CYP11B1 and CYP11B2 are structurally homologous to CYP11A, which encodes P450scc (discussed earlier), and the trio represent a subfamily within the P450 superfamily. Clinical Features and Molecular Defects Deficiency of 11β-hydroxylase results in the second most common form of CAH, accounting for 5–8% of cases, depending on ethnic background. The condition occurs with relatively high frequency in consanguineous families in Saudi Arabia, and in Moroccan and Iranian Jews. Affected 46,XX infants present with variable genital virilization, similar to those with 21-hydroxylase deficiency, in the presence of normal ovaries and female internal genital structures. Some affected females have been so extremely virilized that they have been reared as males, the diagnosis delayed until puberty, when breast development and menses occurred. Driven by increased ACTH levels, 11-deoxycortisol is massively elevated and serum concentrations of 11-DOC, adrenal androgens and testosterone (by conversion from androstenedione) are also increased. The key clinical feature that distinguishes this condition from 21-hydoxylase deficiency is the presence of hypertension, induced by high concentrations of 11-DOC, whose mineralocorticoid activity causes sodium retention, variable hypokalemia and suppression of plasma renin activity. There is minimal correlation between the severity of the virilization and the hypertension. Precocious pseudopuberty and advanced skeletal maturation occur in untreated cases in both sexes. Mutations in CYP11B1 include frameshift mutations that delete the enzyme’s heme-binding domain and a number of missense mutations. The majority of molecular defects are de novo point mutations clustered in exons 6–8 of the gene, suggesting that this region encodes residues critical for enzymatic activity. The relatively high frequency of mutations in this gene is suggested to result from its high number of CpG dinucleotides, a well-recognized mutational “hot spot” in the human genome, rather than from recombination between CYP11B1 and CYP11B2, (as occurs between CYP21 and its adjacent pseudogene). Substitution of arginine by histidine at position 448 is the predominant defect in P450C11 in Moroccan Jews, suggesting a founder effect. This AA is highly conserved and thought to be required for heme binding, which is critical for enzymatic activity.

441

Aromatase Normal Function P450arom is a cytochrome P450 enzyme located in the endoplasmic reticulum of estrogen-producing cells. Using POR (described above) as a cofactor, it catalyzes conversion of C19 steroids (androstenedione and testosterone) to C18 estrogens (estradiol, estrone, estriol). Androgens produced by the fetal adrenal gland, then desulfated and aromatized by the placenta are the major source of circulating estrogens during pregnancy. P450arom is encoded by CYP19, a 9-exon, 70-kb gene located at chromosome 15q21.1. P450arom is expressed in a wide variety of human tissues, ranging from the preimplantation blastocyst to the placenta, ovarian granulosa and luteal cells, testicular Sertoli and Leydig cells, adipose tissue, brain, muscle, and liver. Expression is regulated in part by the use of tissue-specific alternative promoters; however, the same protein is expressed in all tissues. Clinical Features and Molecular Defects An apparently rare cause of female pseudohermaphroditism is the inability to convert fetal androgens to estrogens because of lack of placental P450arom activity as a result of mutations in the fetus. In two cases, the mothers developed progressive virilization in the latter part of pregnancy, associated with high androgen and low estrogen concentrations. Despite this, growth of the fetuses and placentas throughout gestation were normal. Following delivery maternal virilization resolved and in vitro assay of the placenta revealed negligible P450arom activity. Absence of maternal virilization in a third case suggested that some P450arom activity was retained. The affected 46,XX infants had male-appearing or ambiguous external genitalia with marked clitoral enlargement, rugation and fusion of labioscrotal folds, and a single meatus at the base of the phallic structure. The virilization results from placental inability to aromatize DHEA, which is therefore converted to androstenedione and testosterone, leading to massive elevations of the latter. An affected 46,XY infant had normal masculinization. The two affected 46,XX patients developed features of androgen excess at puberty, associated with ovarian cysts. Clitoral enlargement and facial acne were noted, the result of high adrenal androgens. In addition, the affected individuals had absence of breast development, attributed to deficiency of ovarian P450arom (and therefore low estrogens). Gonadotropins were modestly elevated, accompanied by high testosterone and low estradiol concentrations, which also resulted in delayed skeletal maturation and tall stature. There was also significant osteoporosis in the affected adult male. The defects in these cases were inherited in an autosomalrecessive manner and there was known parental consanguinity in one case. In the first case, the affected girl was homozygous for a splice junction point mutation that resulted in translation of an abnormal peptide containing an extra 29 AAs. The mutant enzyme retained only a minimal level of activity in vitro. In the second case, the affected individual had compound heterozygosity for two single base mutations that introduced two separate AA substitutions into the enzyme: arginine 435 to cysteine and two residues downstream, cysteine 437 to tyrosine. The mutant enzymes had extremely low activity in the presence of the R435C substitution, and complete absence of activity with the C437Y defect. Cysteine 437 is very highly conserved, and is apparently involved in heme binding, hence the destructive effect of the mutation on enzyme function. In the final case, the mutation introduced a cysteine in place of the highly conserved residue arginine 375, located in a region of the protein that may be involved in anchoring the

442

SECTION IV / ENDOCRINOLOGY

enzyme to the cell membrane. The mutant protein expressed in vitro had only 0.2% of the activity of wild-type P450arom. These cases, as well as the single reported case of an estrogen receptor mutation, and the evidence from estrogen receptor -deleted transgenic mice, indicate that contrary to longstanding belief, estrogens are not required for fetal survival. Disorders of Female Sex Differentiation Affecting Müllerian Structures Homeobox A 13 (HOXA13) and Related Genes Normal Function A member of the homeobox family of genes that encode the developmentally important homeodomain proteins, homeobox A 13 (HOXA13) and related genes, play important roles in the morphogenesis of the terminal part of the gut and urogenital tract and are involved in Müllerian development. HOXA13 is specifically discussed because in addition to urogenital abnormalities in Hoxa13-deficient mice, mutations affecting uterine development have been reported in humans. The human HOXA13 gene is located within a cluster of at least 8 homeobox genes on chromosome 7, at 7p15-p14.2. Within the protein certain residues and AA motifs are strongly conserved in fish, amphibian, reptile, chicken, and marsupial and placental mammals. In mice, Hoxa13 is expressed early in the epithelial cells of tissues of the developing hindgut in a manner that suggests a fundamental role in the epithelial-mesenchymal interaction necessary for tail growth and posterior gut/genitourinary patterning. In the development of the embryonic mouse Müllerian tract Hoxa9, Hoxa10, Hoxa11, and Hoxa13 are all expressed along the length of the paramesonephric duct. Later in development, expression of Hoxa13 is localized to the cervical and vaginal tissues. After birth, a spatial Hox axis is established, corresponding to the postnatal differentiation of this organ system: Hoxa9 is expressed in the fallopian tubes, Hoxa10 in the uterus, Hoxa11 in the uterus and cervix, and Hoxa13 in the upper vagina. In the developing mouse genital tubercle, Hoxa13 is essential for normal expression of Fgf8 and Bmp7 in the urethral plate epithelium. Mice with the semidominant mutation “hypodactyly” (Hd) have a 90-bp deletion within the first exon of Hoxa13. Homozygous (Hd/Hd) female mice have profound hypoplasia of the cervix and vaginal cavity, in addition to absence of digits. In male mutant mice, hypospadias occurs as a result of the combined loss of Fgf8 and Bmp7 expression in the urethral plate epithelium, as well as the ectopic expression of noggin in the flanking mesenchyme. Complete deletion of Hoxa13 causes more profound defects: Hoxa13(–/–) mutant fetuses have agenesis of the caudal portion of the Müllerian ducts, lack development of the presumptive urinary bladder and have premature stenosis of the umbilical arteries, which could account for the lethality of this mutation at midgestation. Clinical Features and Molecular Defects The hand-footuterus or hand-foot-genital syndrome is an autosomal-dominant disorder whose features include limb anomalies (short first metacarpals, small thumbs and great toes, short fifth fingers and fusion or delayed ossification of wrist bones) and disturbances of Müllerian fusion including a partially divided (bicornuate) or completely divided (didelphic) uterus; affected males have hypospadias of variable severity. Heterozygous mutations in HOXA13 have been found in a number of families with this condition. Such mutations have generally been severe, causing protein truncation that likely eliminates the DNA-binding capacity of the protein. It is of interest that mutations in HOXA13 have somewhat

similar effects to those of another, better characterized, homeobox gene – SOX9 – which also causes defects in both skeletal and sexual development. Mayer-Rokitansky-Kuster-Hauser Syndrome (MRKH) Although individuals with the Mayer-Rokitansky-Kuster-Hauser (MRKH) clearly are phenotypic females, there is nevertheless a failure of female sex differentiation of the internal genitalia. Affected girls usually present in their teens with primary amenorrhea despite otherwise normal secondary sexual development. Examination reveals absence or severe hypoplasia of the vagina; uterine agenesis is usual, however, some uterine development (uni- or bicornuate uterus) or occasionally a normal uterus, may be present. The ovaries are normal. There appear to be 2 subtypes of the disorder––the typical (isolated) and atypical forms, frequency being approximately equal. The typical form is characterized by presence of symmetric muscular buds (Müllerian remnants) and normal fallopian tubes. The atypical form has asymmetric aplasia of one or both buds, with or without dysplasia of the fallopian tubes. The atypical form may have associated anomalies including renal defects (agenesis or ectopia in approx 30–50% of patients) and skeletal abnormalities (vertebral malformations; Klippel-Feil anomaly) of variable severity. Laparoscopy is required to distinguish the typical from the atypical form. The most severe form of the disorder, MURCS, includes Müllerian duct aplasia, renal agenesis/ectopia, and cervical somite dysplasia (Klippel-Feil anomaly). Because of these associated anomalies, all girls and women with vaginal atresia should undergo skeletal radiographs and renal/ pelvic ultrasound. The MRKH syndrome is thought to result from failure of fusion of the lower Müllerian ducts during early embryogenesis. Cases in which there is associated skeletal dysplasia may represent a mesodermal malformation spectrum. The MRKH syndrome occurs in 1 in 4000–5000 females and represents the cause of primary amenorrhea in 15% of cases. Most cases are sporadic, however, a genetic defect likely underlies familial cases (~5%), which have been reported in patterns consistent with sex-limited autosomal-dominant or autosomal-recessive inheritance. Although one study suggested a potential relationship between a heterozygous mutation in the gene encoding galactose-1-phosphate uridyl transferase and vaginal agenesis in about half of patients studied, the data were inconclusive and have not been formally confirmed.

APPROACHES TO DIAGNOSIS Abnormalities of sex determination and differentiation in infancy require evaluation by an experienced team, including a pediatric endocrinologist, urologist, and geneticist. The principal aims of the diagnostic evaluation are to determine: (1) presence of potentially life-threatening adrenal steroid deficiencies and electrolyte derangements; (2) chromosomal sex; (3) type and functional status of the gonads; and (4) internal genital anatomy. The history should determine the presence of similarly affected siblings and of consanguinity between parents or of ethnic background in which inbreeding is common (suggestive of autosomalrecessive conditions such steroidogenic enzyme defects and 5αreductase 2 deficiency). A family history of sudden death in infancy, ambiguous genitalia, lack of pubertal development, amenorrhea or infertility may be found in some cases of CAH. A history in maternal relatives, of genital abnormalities, severe, persistent gynecomastia or infertility suggests an X-linked condition such as androgen insensitivity. Maternal virilization during pregnancy could suggest P450arom deficiency. Clinical examination,

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

aided by ultrasound and radiographic contrast studies, should examine the following: 1. Location of gonads. Gonads in the inguinal region are highly likely to be testes (ovarian herniation occurs exceptionally rarely), indicating that the infant is probably a genotypic male (or much less likely, another karyotype with presence of SRY). If the gonads are impalpable in a partially masculinized infant, ultrasound examination is required to locate the gonads, which may be anywhere from the abdomen to the inguinal ring, and to examine for associated developmental abnormalities of the renal tract. 2. External genital anatomy. Specifically, the size of the phallic structure (stretched penile length, excluding foreskin, and diameter at midshaft) and presence of chordee, location of the urethral meatus, presence of a separate vaginal orifice, size, fusion and rugation of labioscrotal folds, should be carefully documented and the genitalia photographed. Hyperpigmentation of the genitalia is evidence of ACTH excess, due to some form of CAH. 3. Internal genital anatomy. Cystoscopy or voiding cystourethrogram may be required to define the anatomy of the lower urogenital structures, including type of urethra (long, male-type vs short, female-type), presence of a vagina or more rudimentary structure such as a prostatic utricle, and presence/location of vaginal orifice (perineal vs urethral). The longer the urethra, smaller the vaginal structure, and higher its entry to the urethra, the greater the degree of androgenization that occurred in utero. 4. Presence or absence of a uterus. Because of exposure to maternal estrogens, the uterus is enlarged and readily detectable by rectal examination or ultrasonography in newborn females. Presence of a uterus reflects absence of AMH effect during gestation, indicating that either the gonads are not testes, or that if testes are present, AMH action was absent at the critical period during gestation (e.g., because of gonadal dysgenesis). 5. Presence of other dysmorphic features or intrauterine growth retardation. These findings may suggest a generalized disturbance of morphogenesis. Despite careful examination, it is almost impossible to differentiate between many of the disorders of sex determination and differentiation on clinical grounds alone. Detailed laboratory evaluation is important if dealing with an infant with ambiguous genitalia in whom the sex of rearing needs to be determined. Hormone secretion in the immediate postnatal period is dynamic, hormone concentrations changing rapidly over the first few days of life. Therefore, optimally, blood should be drawn within the first 36 h of life for hormone analysis. Primary care institutions should be instructed that for any infant with a concern regarding genital development, 10–15 mL of blood should be drawn immediately and the serum frozen for possible later assay. The following studies should be undertaken without hesitation: 1. Karyotype. In all cases a formal karyotype is required. When specifically requested to do so many laboratories can provide a rapid preliminary karyotype within 48 h. 2. Electrolytes and adrenal steroids. 17-OHP (for 21-hydroxylase deficiency); 11-deoxycortisol (for 11β-hydroxylase deficiency); 17-hydroxypregnenolone and DHEA (for 3β-HSD

3.

4.

5.

6.

7.

8.

443

deficiency), pregnenolone, progesterone, 17-hydroxypregnenolone and 17-OHP (for 17α-hydroxylase deficiency) and in most cases, plasma renin activity, should be measured on the initial blood sample. Other more discriminating steroids can be measured on stored, frozen serum once the karyotype and initial critical values are obtained. Normal electrolyte results in the immediate postnatal period do not exclude CAH, and until definitive results are obtained, the infant should be monitored for development of hyponatremia and hyperkalemia. Testosterone, DHT and testosterone precursors to look for evidence of a defect of testosterone biosynthesis or 5αreductase 2 deficiency. Serum testosterone is normally high (>200 ng/dl) in cord blood, but plummets to become almost undetectable at the end of the first week of life, rising again from approx 2–3 wk, to peak at approx 8–12 wk of age. It falls again from 12 to 16 wk, finally becoming essentially unmeasurable by 6 mo of age. Steroid analysis should therefore be performed during periods of maximal testicular activity, or, if this is not possible, following testicular stimulation with hCG (gonadal steroids and DHT are measured before, and 24 h after, administration of 1500 IU hCG every other day for three doses). Normal male infants respond with testosterone values >200 ng/dl, often much higher. An increased ratio of testosterone to DHT (>20/1) during periods of active testicular steroidogenesis or following hCG is found in 5α-reductase 2 deficiency. Hormonal profiles found in defects of testosterone biosynthesis and other disorders of sex differentiation are summarized in Table 41-5. Testosterone, DHT, and testosterone precursors are normal or elevated in infants with partial AIS. Infants with CAIS fail to demonstrate the normal postnatal LH/testosterone surge (although this is somewhat moot, because these infants typically do not have ambiguous genitalia). AMH measurement. Serum AMH is low in intersex states associated with defective testis development and is elevated in situations of defective testosterone synthesis or androgen insensitivity, in which Sertoli cells are normal. LH and FSH. LH concentration is increased in LCH and in most infants with PAIS; FSH is increased in those with gonadal dysgenesis. ACTH stimulation test. This is required to evaluate steroidogenic defects that affect the adrenals as well as the gonads (e.g., CLAH, 3β-HSD deficiency, 17α-hydroxylase deficiency). In virilized karyotypic females, an ACTH stimulation test is not required for diagnosis of 21-hydroxylase deficiency, but may aid diagnosis of other forms of CAH by exaggerating precursor steroid levels. Molecular genetic analysis. Analysis of genes such as those encoding the AR, 5α-reductase or 17α-hydroxylase may be helpful particularly in cases in which there is clinical ambiguity. Although expensive, genetic assays are becoming more readily available through specialized laboratories. Gonadal biopsy may be required in cases of suspected gonadal dysgenesis.

Despite a comprehensive work-up many patients with disorders of sexual development remain undiagnosed, reflecting a large number of remaining gaps in understanding the genetics and molecular biology of normal human sex determination and differentiation.

444

SECTION IV / ENDOCRINOLOGY

SELECTED REFERENCES Achermann JC, Ito M, Ito M, Hindmarsh PC, Jameson JL. A mutation in the gene encoding steroidogenic factor-1 causes XY sex reversal and adrenal failure in humans. Nat Genet 1999;22(2):125, 126. Achermann JC, Meeks JJ, Jameson JL. Phenotypic spectrum of mutations in DAX-1 and SF-1. Mol Cell Endocrinol 2001;185(1,2):17–25. Achermann JC, Ozisik G, Ito M, et al. Gonadal determination and adrenal development are regulated by the orphan nuclear receptor steroidogenic factor-1, in a dose-dependent manner. J Clin Endocrinol Metab 2002;87(4):1829–1833. Ahmed SF, Hughes IA. The genetics of male undermasculinization. Clin Endocrinol (Oxf) 2002;56(1):1–18. Auchus RJ. The genetics, pathophysiology, and management of human deficiencies of P450c17. Endocrinol Metab Clin North Am 2001; 30(1):101–119. Belgorosky A, Pepe C, Marino R, et al. Hypothalamic-pituitary-ovarian axis during infancy, early and late prepuberty in an aromatase-deficient girl who is a compound heterocygote for two new point mutations of the CYP19 gene. J Clin Endocrinol Metab 2003;88:5127–5131. Birk OS, Casiano DE, Wassif CA, et al. The LIM homeobox gene Lhx9 is essential for mouse gonad formation. Nature 2000;403(6772): 909–913. Boehmer ALM, Brinkmann AO, Sandkuijl LA, et al. 17β-Hydroxysteroid dehydrogenase-3 deficiency: diagnosis, phenotypic variability, population genetics, and worldwide distribution of ancient and de novo mutations. J Clin Endocrinol Metab 1999;84:4713–4721. Bose HS, Sato S, Aisenberg J, Shalev SA, Matsuo N, Miller WL. Mutations in the steroidogenic acute regulatory protein (StAR) in 6 patients with congenital lipoid adrenal hyperplasia. J Clin Endocrinol Metab 2000;85:3636–3639. Bose HS, Sugawara T, Strauss JF 3rd, Miller WL. The pathophysiology and genetics of congenital lipoid adrenal hyperplasia. N Engl J Med 1996;335(25):1870–1878. Bulun SE. Clinical review 78: Aromatase deficiency in women and men: would you have predicted the phenotypes? J Clin Endocrinol Metab 1996;81:867–871. Clark AM, Garland KK, Russell LD. Desert hedgehog (Dhh) gene is required in the mouse testis for formation of adult-type Leydig cells and normal development of peritubular cells and seminiferous tubules. Biol Reprod 2000;63(6):1825–1838. Cotinot C, Pailhoux E, Jaubert F, Fellous M. Molecular genetics of sex determination. Semin Reprod Med 2002;20(3):157–168. Fujieda K, Okuhara K, Abe S, Tajima T, Mukai T, Nakae J. Molecular pathogenesis of lipoid adrenal hyperplasia and adrenal hypoplasia congenita. J Steroid Biochem Mol Biol 2003;85(2–5):483–489. Geley S, Kapelari K, Johrer K, et al. CYP11B1 mutations causing congenital adrenal hyperplasia due to 11-beta-hydroxylase deficiency. J Clin Endocrinol Metab 1996;81:2896–2901. Gibbons RJ, Higgs DR. Molecular-clinical spectrum of the ATR-X syndrome. Am J Med Genet 2000;97(3):204–212. Goodfellow PN, Camerino G. DAX-1, an “antitestis” gene. EXS. 2001; (91):57–69. Graves JA. Evolution of the testis-determining gene—the rise and fall of SRY. Novartis Found Symp 2002;244:86–97. Guo JK, Hammes A, Chaboissier MC, et al. Early gonadal development: exploring Wt1 and Sox9 function. Novartis Found Symp 2002;244: 23–31. Harley VR, Clarkson MJ, Argentaro A. The molecular action and regulation of the testis-determining factors, SRY (Sex-Determining Region on the Y Chromosome) and SOX9 (SRY-Related High-Mobility Group [HMG] Box 9). Endocr Rev 2003;24:466–487. Heikkila M, Peltoketo H, Vainio S. Wnts and the female reproductive system. J Exp Zool 2001;290(6):616–623. Imperato-McGinley J, Zhu YS. Androgens and male physiology the syndrome of 5alpha-reductase-2 deficiency. Mol Cell Endocrinol 2002; 198(1,2):51–59. Jordan BK, Shen JH, Olaso R, Ingraham HA, Vilain E. Wnt4 overexpression disrupts normal testicular vasculature and inhibits testosterone synthesis by repressing steroidogenic factor 1/beta-catenin synergy. Proc Natl Acad Sci USA 2003;100(19):10,866–10,871.

Josso N, di Clemente N, Gouedard L. Anti-Müllerian hormone and its receptors. Mol Cell Endocrinol 2001;179(1,2):25–32. Katsumata N, Ohtake M, Hojo T, et al. Compound heterozygous mutations in the cholesterol side-chain cleavage enzyme gene (CYP11A) cause congenital adrenal insufficiency in humans. J Clin Endocrinol Metab 2002;87(8):3808–3813. Kawamoto T, Mitsuuchi Y, Toda K, et al. Role of steroid 11-beta-hydroxylase and steroid 18-hydroxylase in the biosynthesis of glucocorticoids and mineralocorticoids in humans. Proc Natl Acad Sci USA 1992;89: 1458–1462. Kim S, Kettlewell JR, Anderson RC, Bardwell VJ, Zarkower D. Sexually dimorphic expression of multiple doublesex-related genes in the embryonic mouse gonad. Gene Expr Patterns 2003;3(1):77–82. Koopman P. Sry, Sox9 and mammalian sex determination. EXS 2001;(91):25–56. Koopman P, Loffler KA. Sex determination: the fishy tale of Dmrt1. Curr Biol 2003;13(5):R177–R179. Lin D, Sugawara T, Strauss JF III, et al. Role of steroidogenic acute regulatory protein in adrenal and gonadal steroidogenesis. Science 1995;267:1828–1831. Lovell-Badge R, Canning C, Sekido R. Sex-determining genes in mice: building pathways. Novartis Found Symp 2002;244:4–18. Luu-The V. Analysis and characteristics of multiple types of human 17beta-hydroxysteroid dehydrogenase. J Steroid Biochem Mol Biol 2001;76(1–5):143–151. Merchant-Larios H, Moreno-Mendoza N. Onset of sex differentiation: dialog between genes and cells. Arch Med Res 2001;32(6):553–558. Merke DP, Bornstein SR, Avila NA, Chrousos GP. NIH conference. Future directions in the study and management of congenital adrenal hyperplasia due to 21-hydroxylase deficiency. Ann Intern Med 2002;136(4): 320–334. Miyamoto N, Yoshida M, Kuratani S, Matsuo I, Aizawa S. Defects of urogenital development in mice lacking Emx2. Development 1997;124 (9):1653–1664. Muscatelli F, Strom TM, Walker AP, et al. Mutations in the DAX-1 gene give rise to both X-linked adrenal hypoplasia congenita and hypogonadotropic gonadism. Nature 1994;372:672–676. Nilsson E, Skinner MK. Cellular interactions that control primordial follicle development and folliculogenesis. J Soc Gynecol Investig 2001;8(1 Suppl Proceedings):S17–S20. Online Mendelian Inheritance in Man. Johns Hopkins University. www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=OMIM. July 2005. Parker KL, Schimmer BP. Genes essential for early events in gonadal development. Ann Med 2002;34(3):171–178. Quigley CA, De Bellis A, Marschke KB, El-Awady MK, Wilson EM, French FS. Androgen receptor defects: Historical, clinical and molecular perspectives. Endocr Rev 1995;16:271–321. Raymond CS, Murphy MW, O’Sullivan MG, Bardwell VJ, Zarkower D. Dmrt1, a gene related to worm and fly sexual regulators, is required for mammalian testis differentiation. Genes Dev 2000;14(20):2587–2595. Renfree MB, Wilson JD, Shaw G. The hormonal control of sexual development. Novartis Found Symp 2002;244:136–152. Reutens AT, Achermann JC, Ito M, et al. Clinical and functional effects of mutations in the DAX-1 gene in patients with adrenal hypoplasia congenita. J Clin Endocrinol Metab 1999;84(2):504–511. Richter-Unruh A, Martens JW, Verhoef-Post M, et al. Leydig cell hypoplasia: cases with new mutations, new polymorphisms and cases without mutations in the luteinizing hormone receptor gene. Clin Endocrinol (Oxf) 2002;56(1):103–112. Scharnhorst V, van der Eb AJ, Jochemsen AG. WT1 proteins: functions in growth and differentiation. Gene 2001;273(2):141–161. Shawlot W, Behringer RR. Requirement for Lim1 in head-organizer function. Nature 1995;374(6521):425–430. Shen W-H, Moore CCD, Ikeda Y, Parker KL, Ingraham HA. Nuclear receptor steroidogenic factor 1 regulates the Müllerian inhibiting substance gene: a link to the sex determination cascade. Cell 1994;77:651–661. Simard J, Moisan AM, Morel Y. Congenital adrenal hyperplasia due to 3beta-hydroxysteroid dehydrogenase/Delta(5)-Delta(4) isomerase deficiency. Semin Reprod Med 2002;20(3):255–276.

CHAPTER 41 / DISORDERS OF SEX DETERMINATION AND DIFFERENTIATION

Simpson JL, Rajkovic A. Ovarian differentiation and gonadal failure. Am J Med Genet 1999;89(4):186–200. Sinclair A, Smith C, Western P, McClive P. A comparative analysis of vertebrate sex determination. Novartis Found Symp 2002;244:102–111. Suzuki T, Mizusaki H, Kawabe K, Kasahara M. Yoshioka H. Morohashi K. Concerted regulation of gonad differentiation by transcription factors and growth factors. Novartis Found Symp 2002;244:68–77. Teixeira J, Maheswaran S, Donahoe PK. Müllerian inhibiting substance: an instructive developmental hormone with diagnostic and possible therapeutic applications. Endocr Rev 2001;22(5):657–674. Tilmann C, Capel B. Cellular and molecular pathways regulating mammalian sex determination. Recent Prog Horm Res 2002;57:1–18. Tremblay JJ, Viger RS. Transcription factor GATA-4 enhances Müllerian inhibiting substance gene transcription through a direct interaction

445

with the nuclear receptor SF-1. Mol Endocrinol 1999;13(8): 1388– 1401. Tremblay JJ, Viger RS. Novel roles for GATA transcription factors in the regulation of steroidogenesis. J Steroid Biochem Mol Biol 2003;85(2–5):291–298. The Androgen Receptor Gene Mutations Database World Wide Web Server. The Lady Davis Institute for Medical Research and Sir Mortimer B. Davis-Jewish General Hospital. Montreal, Quebec, Canada. www.mcgill.ca/androgendb. Accessed 4/4/2005. Yao HH, Tilmann C, Zhao GQ, Capel B. The battle of the sexes: opposing pathways in sex determination. Novartis Found Symp 2002;244:187–198. Yao HH, Whoriskey W, Capel B. Desert Hedgehog/Patched 1 signaling specifies fetal Leydig cell fate in testis organogenesis. Genes Dev 2002;16(11):1433–1440.

42 Sex Chromosome Disorders ANDREW R. ZINN SUMMARY Sex chromosome disorders have held a special fascination for geneticists and endocrinologists because of the intimate association between the sex chromosomes, germ cells, and sex determination. Molecular studies are beginning to yield explanations for phenotypes associated with sex chromosome abnormalities, such as sex reversal with Yp deletions or Xp duplications, azoospermia with Yq deletions, and short stature in Turner syndrome. Even with completion of the human genome sequence, sex chromosome disorders remain a fertile topic for research in molecular medicine. Key Words: Azoospermia; gonadal dysgenesis; Klinefelter syndrome; sex chromosome disorders; Turner syndrome; Xp duplications; Yp deletions.

INTRODUCTION The clinical consequences of sex chromosome abnormalities are best understood by considering four principles. First, sex chromosomes have unique biological functions during gametogenesis and thus infertility is a frequent manifestation of sex chromosome abnormalities. Second, by virtue of their different sex chromosome constitutions, men and women have different copy numbers, or dosage, of X-linked genes. Because a twofold difference in expression of individual genes is sufficient in many cases to perturb mammalian growth and development, special mechanisms have evolved to balance the expression of X-linked genes between the sexes. Third, the Y chromosome has relatively few genes, most of which serve male-specific functions. Thus impaired male sexual development or reproductive capacity is a frequent sequela of various Y chromosome disorders. Last, sex chromosome disorders show extensive genetic and clinical variability, and the common diagnostic entities actually encompass a range of abnormalities. Table 42-1 summarizes the karyotypes, genotypes, and phenotypes associated with the disorders in this chapter.

STRUCTURE AND FUNCTION OF HUMAN SEX CHROMOSOMES Fig. 42-1 depicts an overview of the human sex chromosomes. The X chromosome is approx 160 million basepairs (Mb) in length. It is estimated to contain approx 1000 genes, encoding a variety of catalytic, structural, and regulatory proteins. Although From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

the Y chromosome is about one-third as large, it contains remarkably fewer genes, numbering only 34 in the Online Mendelian Inheritance in Man database. The Y chromosome is divided into two parts. The approx 30 Mb euchromatic portion encodes all known Y-linked genes. By contrast, the approx 20 Mb heterochromatic portion is made up of just simple repetitive sequences, and it can vary substantially in size among normal men. Because normal females do not have a Y chromosome or any Y-specific gene products, the Y chromosome is often thought to function only in male-specific processes, such as testis formation or spermatogenesis. Indeed, many Y-linked genes are expressed specifically in testis. However, a number of Y-linked genes are expressed in nontesticular tissues and are known to be important in processes that are not male-specific, including viability, growth, and morphogenesis. Normally the X and Y chromosomes do not recombine except at the tips, where the chromosomes pair during male meiosis. This pairing and recombination may be required for proper chromosome segregation. Genetic markers in these recombining regions do not show sex-linked inheritance; hence the regions are termed “pseudoautosomal”. The short arm (p) pseudoautosomal region spans approx 2.6 Mb, within which there are 13 known genes. The long arm (q) pseudoautosomal region is only 0.4 Mb in length and contains only four known genes. As might be expected because males and females both have two copies of the pseudoautosomal regions, genes within these regions do not show any predilection for sex-specific functions.

X INACTIVATION AND DOSAGE COMPENSATION The X and Y chromosomes began diverging from an ancestral autosome 240–320 × 106 yr ago, and over time, most Y-linked genes were lost. As a consequence, males as a rule have only one copy of X-linked genes, whereas females have two. However, this copy number difference is balanced in mammals by transcriptional inactivation of one X chromosome in females, a process termed dosage compensation or X inactivation. X inactivation occurs during early embryogenesis and is irreversible except in oogonia, where the inactive X is reactivated at or before the onset of meiosis. The inactive X chromosome becomes hypermethylated and late-replicating, and in most tissues it condenses into the Barr body, or sex chromatin. The choice of which X undergoes inactivation in each cell of the embryo is normally stochastic, but the inactive state is stably propagated to daughter cells through mitosis. Thus most women are mosaic with respect to which X is inactive in individual cells.

446

Table 42-1 Summary of Sex Chromosome Disordersa Disorder Gene deficiency Turner syndrome

XY gonadal dysgenesis (Swyer syndrome) Azoospermia

Karyotype

Molecular cytogenetics

45,X and variants

46,XY 46,XY or – 46,XYq–

47,XXY

Ring X mosaicism

45,X/46,X,r(X)

Dosage sensitive Normal or Xp21.2 sex reversal duplication Aberrant gene expression XX male 46,XX Gonadoblastoma

Various

Nongenic Premature ovarian failure associated with X-autosome translocations

46,X,t (Xq;autosome) with breakpoint between Xq13-q26

aDisorders

Ring is XIST negative in many severe cases

SRY translocated to X in most cases Presence of pericentromeric Y material (GBY locus)

discussed in this chapter. DAZ1, deleted in azoospermia; SRY, sex-determining region Y gene.

Gonads

Female

Streak

Female

Streak; risk of gonadoblastoma Azoospermia or oligospermia with variable histology

Male

Extragonadal features

Molecular defect

Short stature, webbed neck, lymphedema, aortic coarctation, cubitus valgus, nonverbal learning disability Tall stature

Haploinsufficiency of SHOX and other X-linked genes

Sometimes short stature

Male

Hyalinized testis; azoospermia

Female

Streak

Female

Streak

None

Male or hermaphrodite Female

Testes with azoospermia or ovotestis Streak + tumor

Klinefelter-like

Female

Ovaries with diminished follicles

447

Gene excess Klinefelter syndrome

SRY negative in some cases Yq11.23 deletion

Sexual phenotype

Poor to normal virilization, gynecomastia, long limbs, verbal deficits Mental retardation, multiple congenital anomalies



None

SRY mutation or deletion in some cases Deletion of DAZ1, USP9Y, or other spermatogenesis genes Overexpression of SHOX and other X-linked genes Functional disomy of pericentromeric X-linked genes Overexpression of NROB1 gene Presence of SRY, expression modified by X inactivation Not known

Failure of meiotic sex chromosome pairing or abnormal X inactivation?

448

SECTION IV / ENDOCRINOLOGY

Figure 42-1 Schematic depiction of human sex chromosomes and approximate location of key genetic loci discussed in the text. Symbols for cloned genes in parentheses. p, short arm; q, long arm.

Cytogenetic studies of aberrant X chromosomes pinpointed a cis-acting region of the proximal long arm, the X inactivation center (XIC), that is necessary and sufficient for X inactivation. Within the XIC, the X-inactive specific transcript (XIST) gene plays a key role in the initiation of X inactivation. XIST has a unique pattern of expression; it is transcribed only from the inactive X chromosome. Expression of XIST precedes chromatin condensation and transcriptional inactivation of other X-linked genes. The gene is transcribed as a large (>15 kb) polyadenylated, alternatively spliced RNA that does not encode a protein. The XIST RNA localizes within the nucleus to the inactive X chromosome, in which it recruits proteins that silence transcription. Once XIST is expressed and X inactivation has occurred, the inactive state of the chromosome is maintained by chromatin condensation, histone deacetylation, and DNA methylation. Although X inactivation occurs on a chromosome-wide scale, some specific genes escape X inactivation, i.e., they are transcribed from both the active and inactive X chromosomes. For some of these genes, Y-linked copies serve instead of X inactivation to equalize their dosage in males and females. For example, all short arm pseudoautosomal genes escape X inactivation and are present on both the X and Y chromosomes. By one estimate, as many as 1/4 of all X-linked genes are expressed from the inactive X in at least some tissue. However, the vast majority of these genes have no functional Y homolog, and the significance of their expression from the inactive X is unclear. The signals that determine whether a gene is subject to X inactivation are unknown, but some genes that escape X inactivation are clustered, suggesting that local chromatin structure is likely to be important.

DISORDERS RESULTING FROM GENE DEFICIENCY 45,X TURNER SYNDROME The loss of one sex chromosome (either an X or Y) results in monosomy X, or Turner syndrome, which occurs once in every 5000 births. The prenatal incidence of monosomy X is much greater: it is estimated that 2% of all human conceptuses are 45,X, but less than 1% of these survive to term. Paradoxically, the life expectancy of 45,X girls is only modestly reduced. Thus persons with Turner syndrome have the mildest manifestations of the underlying genetic disorder, complicating efforts to understand the molecular basis of the

phenotype. An adult woman with Turner syndrome is shown in Fig. 42-2. Characteristic features include growth retardation, ovarian failure, and anatomic abnormalities such as webbed neck, aortic coarctation, increased carrying angle of the elbows (cubitus valgus), short 4th metacarpals, lymphedema, and renal anomalies. Other metabolic and endocrine abnormalities associated with Turner syndrome include hypertension, glucose intolerance, and autoimmune thyroid disease. Girls with Turner syndrome are not usually mentally retarded, although they frequently have selective cognitive deficits, notably nonverbal learning disabilities such as impaired mathematical and visual-spatial skills. The pathophysiology of Turner syndrome is poorly understood. Growth failure is not because of growth hormone deficiency, although administration of pharmacological doses of recombinant human growth hormone during childhood accelerates growth and increases final stature. Ovarian failure is the result of accelerated oocyte atresia, beginning around 6 mo of gestation. By early infancy there are usually few or no remaining germ cells, and infertility is the rule, with rare exceptions. As in postmenopausal women, the loss of germ cells causes fibrosis of the ovaries and subsequent lack of sex steroid production. Most girls with Turner syndrome require hormone replacement therapy to induce puberty and to maintain cyclic menses. Extragonadal abnormalities, including webbed neck, lymphedema of the extremities, and perhaps aortic coarctation, might be caused by defective lymphatic vessel development. Alternatively, coarctation might result from a neural crest cell migratory defect. The mechanism of chromosome loss in Turner syndrome is not known. Meiotic nondisjunction accounts for only a minority of cases. For this reason, the risk of Turner syndrome is not associated with advanced maternal age, unlike Down syndrome. Most losses occur during mitosis in the early multicellular embryo, resulting in a high incidence of mosaicism, the presence of cells with different chromosomal constitutions within one individual. The additional cell line(s) in Turner syndrome mosaics may contain two or more X chromosomes or a structurally aberrant X chromosome. Embryos mosaic for cells with two or more copies of Xq have greater viability than nonmosaic 45,X embryos. Mosaicism for a Y-bearing cell line, most often 45,X/46,XY, deserves particular mention. The clinical phenotype depends on the proportion and tissue distribution of XY cells. Bilateral testes

CHAPTER 42 / SEX CHROMOSOME DISORDERS

Figure 42-2 Turner syndrome. (Used with permission from Connor JM, Ferguson-Smith MA, Essential Medical Genetics, 4th ed. Blackwell Science.)

may be present, or there may be unilateral testicular tissue with a contralateral streak gonad, a condition termed mixed gonadal dysgenesis (see Chapter 41). The presence of Y material in cells of a dysgenetic gonad confers a risk of gonadoblastoma, a malignant tumor. Prophylactic removal of streak gonads or histologically abnormal testes is recommended in any Turner syndrome patient whose karyotype includes Y material. The Y-linked gene predisposing to gonadoblastoma (GBY) is distinct from the sex-determining region Y gene (SRY), because XY females with deletion or mutation of SRY are still at risk. Available evidence suggests that GBY is near the Y centromere, although its precise identity is not known. Some studies using sensitive PCR assays have reported an alarmingly high prevalence of cryptic Y sequences in Turner syndrome. However, in the absence of testes or virilization, the clinical significance of Y material detected only by PCR is uncertain. The single X chromosome is maternal in approximately threefourths and paternal in approximately one-fourth of 45,X Turner

449

syndrome patients. The parental origin of the single X does not influence the physical phenotype, i.e., there is no imprinting. One study found that girls with a maternal X had poorer social and behavioral skills than those whose X was paternal; this finding has not been independently replicated. Several molecular mechanisms contribute to the phenotype of Turner syndrome. The unpaired X chromosome may trigger a meiotic checkpoint to cause oocyte apoptosis. Additionally, two copies of certain X-linked genes may be required for normal oocyte function. Anatomic abnormalities are evidence that proper dosage of genes that escape X-inactivation is also important for extragonadal development. In males, the Y chromosome presumably supplies the second copy of these genes. The molecular basis of Turner syndrome growth failure is beginning to be understood. Positional cloning identified a single pseudoautosomal gene, SHOX (see Fig. 42-1), the dosage of which is the major factor in Turner syndrome short stature. The lack of one copy of SHOX accounts for approximately two-thirds of the final height deficit in Turner syndrome; the remainder is presumably because of deficiency of other X-linked genes. As predicted, SHOX is expressed from both sex chromosomes in males and females. The gene encodes a member of the paired family of homeodomain transcription factors and is expressed in osteogenic cells during limb development. Its transcriptional targets are not yet known. Point mutations or deletions of SHOX are also implicated in three other growth disorders: idiopathic short stature, Leri-Weill dyschondrosteosis, and Langer mesomelic dysplasia, the homozygous form of dyschondrosteosis. Both dyschondrosteosis and Langer mesomelic dysplasia are characterized by disproportionate shortening of the forelimbs and Madelung deformity of the wrist, especially in females. This deformity usually develops around puberty and is rare in normal women but occurs in 2–7% of women with Turner syndrome. Madelung deformity may be because of premature closure of a portion of the radial epiphysis, suggesting that SHOX promotes normal bone growth and development by inhibiting epiphyseal closure. Estrogen induces epiphyseal closure, and estrogen deficiency may be the reason that most girls with Turner syndrome do not develop Madelung deformity. Fortunately, hormonal induction of puberty in girls with Turner syndrome does not increase the incidence of Madelung deformity, suggesting that there may be a window of vulnerability during prepubertal skeletal development. SHOX deficiency may also cause other Turner syndrome skeletal abnormalities such as cubitus valgus and short 4th metacarpal, which are also seen in some patients with dyschondrosteosis. 46,XY GONADAL DYSGENESIS Classic genetic studies in the fruit fly Drosophila melanogaster indicated that phenotypic sex is determined by the numerical ratio of X chromosomes to autosomes; the Drosophila Y chromosome is inert regarding sex determination. The same mechanism of sex determination was once thought to occur in mammals. However, with the advent of accurate karyotyping in the late 1950s it became clear that XO and XX humans are female, whereas XY and XXY persons are male. These findings established that the presence or absence of the Y chromosome determines human sex. Studies of rare exceptions, “sex reversed” XX males and XY females, were instrumental in elucidating the molecular basis of sex determination. These studies culminated in 1990 with the discovery of a single Y-linked gene, SRY, that directs the differentiation of the bipotential gonad into testes. SRY is situated near the pseudoautosomal region of the Y chromosome

450

SECTION IV / ENDOCRINOLOGY

short arm. The gene encodes a protein of 204 amino acids belonging to the high-mobility group family of DNA-binding proteins. The SRY gene is expressed transiently in the gonadal ridge during embryogenesis and then later in adult testis. SRY sits atop a cascade of genes involved in male sex determination and differentiation, such as AMH, the gene that encodes anti-Mullerian hormone (see Chapter 41). The discovery of SRY clarified the etiology of most XX males and some XY females. The SRY gene and adjacent Y chromosome sequences can be detected in most XX males, in whom it is generally translocated to the short arm of the X chromosome. The translocation results from aberrant recombination involving the nearby pseudoautosomal region. The translocated SRY gene may become subject to X inactivation, and mosaic SRY expression is hypothesized to explain some cases of incomplete sex reversal, for example, XX true hermaphroditism (presence of both testicular and ovarian tissue). Mutations that impair SRY’s function are detected in approx 10% of XY females. The approx 90% that remain unexplained are presumably caused by other genetic or environmental factors (see Chapter 41). AZOOSPERMIA Because SRY is on the short arm of the Y chromosome, large deletions of the long arm are still compatible with normal male sexual differentiation. However, defects in spermatogenesis are frequently associated with long arm deletions. The most severe defect is azoospermia, the complete absence of sperm in semen, and the Yq locus for this phenotype is denoted azoospermia factor (AZF) (see Fig. 42-1). Deletions large enough to detect by karyotype are rare, but molecular analysis reveals microdeletions of Yq in up to 10% of men with otherwise unexplained azoospermia. These deletions arise in many cases by recombination between repeated DNA sequences. Fifteen genes have been identified in the AZF region, many of these are expressed only in testis. Two AZF genes have been particularly well characterized. Each is present in multiple copies. The RNA binding motif (RBM) genes are members of a larger family of RNA-binding proteins. The Y chromosome has at least three functional RBM genes and numerous pseudogenes. Most RBM sequences are clustered near the AZF region on distal Yq, but related sequences are also present near the centromere on both arms of the Y chromosome. The RBM genes are expressed specifically in germ cells and may play a role in RNA splicing or packaging during spermatogenesis. The presence of multiple RBM genes makes it difficult to deduce their importance from studying patients with naturally occurring deletions, which typically affect only some of the copies or also affect other genes. For example, one functional RBM gene is absent in most normal Japanese men. The other well-characterized AZF gene is DAZ1 (deleted in azoospermia). Like RBM, DAZ1 is actually a family of genes that encode a protein with an RNA recognition motif expressed specifically in testis. DAZ1 is deleted in approx 10% of men with nonobstructive azoospermia or severe oligospermia. Interestingly, the testicular pathology associated with DAZ1 deletions ranges from the complete absence of germ cells (Sertoli-cell-only syndrome) to meiotic arrest with occasional mature condensed spermatids. In some cases even adjacent seminiferous tubules within the same man show different histological defects.

DISORDERS OWING TO GENE EXCESS 47, XXY KLINEFELTER SYNDROME The most common sex chromosome aneuploidy is Klinefelter syndrome (47,XXY; 46,XY/47,XXY; and variants), with an incidence of approx 1 in 500

Figure 42-3 Klinefelter syndrome. (Used with permission from Connor JM, Ferguson-Smith MA, Essential Medical Genetics, 4th ed. Essential Medical Genetics, Blackwell Science.)

male births. The extra X chromosome is usually the result of nondisjunction during paternal or maternal meiosis. The Klinefelter phenotype is variable but characterized by small, firm testes, azoospermia, tall stature, eunuchoid habitus, gynecomastia, and elevated gonadotropin levels (Fig. 42-3). Interestingly, patients with Klinefelter syndrome often have verbal learning disabilities, in contrast to the nonverbal cognitive deficits associated with Turner syndrome. Tall stature in Klinefelter syndrome is at least partially attributable to increased SHOX gene dosage and delayed epiphyseal closure. Azoospermia is caused by loss of spermatogenic cells, due either to abnormal gene dosage or to the presence of the unpaired X chromosome at the pachytene stage of meiosis. Hyalinization of seminiferous tubules and impaired Leydig cell testosterone

CHAPTER 42 / SEX CHROMOSOME DISORDERS

production follows loss of germ cells. Androgen replacement is the treatment of choice for testosterone deficiency, whereas treatment of severe gynecomastia is surgical. The phenotype of Klinefelter syndrome is mild compared to disorders involving an extra autosome, for example, trisomy 21 (Down syndrome). The reason is that the additional X chromosome in Klinefelter syndrome is inactivated, whereas the extra chromosome in Down syndrome or other autosomal aneuploidy syndromes remains active. Hence relatively few genes, namely those escaping X inactivation, have an increased functional dosage in Klinefelter syndrome. For this same reason, XXX females are usually normal. Individuals with greater numbers of X chromosomes, for example, 48,XXXY; 48,XXXX; 49,XXXXY; or 49, XXXXX karyotypes, often show more severe abnormalities such as mental retardation or skeletal malformations, even though all but one X chromosome is inactivated. The abnormalities are presumably caused by further increase in the dosage of genes that escape X inactivation. Similar phenotypes have been noted in men with severe Y aneuploidy, for example, 48,XYYY or 49,XYYYY karyotypes, suggesting that the culprit genes are present on both the X and Y chromosomes, for example, SHOX. RING X MOSAICISM Turner syndrome variants involving mosaicism for a ring X chromosome, or 45,X/46,X,r(X) karyotype, are sometimes associated with severe abnormalities such as mental retardation and multiple congenital anomalies. In general, the severity of the phenotype inversely correlates with the size of the ring; larger rings tend to be associated with a milder phenotype. The explanation for the severe phenotypes associated with small rings might lie with X inactivation and dosage compensation. Some small ring X chromosomes either lack the XIC region or fail to express the XIST gene; these rings express some genes that are normally X inactivated. The severe phenotype is probably caused by functional disomy, or expression of both copies, of one or more X-linked genes. DOSAGE-SENSITIVE SEX REVERSAL A few XY females have been reported with partial duplications of the short arm of the X chromosome that include band Xp21.1 (see Fig. 42-1). Because the duplicated X is not inactivated in the absence of a second complete X chromosome, sex reversal is caused by a twofold increase in expression of a gene or genes within the duplication. This locus has been designated dosage-sensitive sex reversal (DSS). Detailed mapping of partially overlapping duplications narrowed the DSS region to a small interval containing NROB1, which encodes a member of the nuclear hormone receptor superfamily. Loss-of-function mutations in NROB1 cause congenital adrenal hypoplasia and hypogonadotropic hypogonadism. Overexpression of NROB1 in transgenic mice causes XY sex reversal, confirming that NROB1 is DSS.

EFFECTS OTHER THAN GENE DOSAGE As mentioned, gonadal failure associated with Turner and Klinefelter syndrome may involve abnormal dosage of X-linked genes that escape inactivation. However, animal studies suggest that the presence of an unpaired X chromosome may cause germ cell apoptosis through a meiotic pachytene checkpoint mechanism. The effect is clearest for spermatogenesis, and explains the observation that men with balanced X-autosome translocations are almost invariably infertile. The presence of an analogous meiotic checkpoint during oogenesis is less certain. For example, gonadal function is usually normal in 47,XXX women. On the other hand, a meiotic checkpoint could explain why some X-autosome translocations are

451

associated with premature ovarian failure. In particular, women whose translocations break within a “critical region” of the X chromosome long arm, Xq13 to Xq26, often have primary or secondary amenorrhea. A number of these translocation breakpoints have been characterized in detail, and most do not appear to disrupt X-linked genes. Although not definitive, these data suggest that the ovarian failure associated with balanced X-autosome translocations is resulting from “nongenic” chromosomal effects such as interference with meiotic pairing or long-range perturbation of X inactivation.

SELECTED REFERENCES Bardoni B, Zanaria E, Guioli S, et al. A dosage sensitive locus at chromosome Xp21 is involved in male to female sex reversal. Nat Genet 1994;7:497–501. Binder G, Fritsch H, Schweizer R, Ranke MB. Radiological signs of LeriWeill dyschondrosteosis in Turner syndrome. Horm Res 2001;55: 71–76. Blaschke RJ, Rappold GA. SHOX in short stature syndromes. Horm Res 2001;55(Suppl 1):21–23. Burgoyne PS, Baker TG. Meiotic pairing and gametogenic failure. Symp Soc Exp Biol 1984;38:349–362. Carrel L, Cottle AA, Goglin KC, Willard HF. A first-generation X-inactivation profile of the human X chromosome. Proc Natl Acad Sci USA 1999;96:14,440–14,444. Chandley AC, Cooke HJ. Human male fertility—Y-linked genes and spermatogenesis. Hum Mol Genet 1994;3:1449–1452. Disteche CM. The great escape. Am J Hum Genet 1997;60:1312–1315. Ford CE, Jones KW, Polani PE, De Almeida JC, Briggs JH. A sex-chromosome anomaly in a case of gonadal dysgenesis (Turner’s syndrome). Lancet 1959;1:711–713. Graves JA, Wakefield MJ, Toder R. The origin and evolution of the pseudoautosomal regions of human sex chromosomes. Hum Mol Genet 1998;7:1991–1996. Hassold TJ. Chromosome abnormalities in human reproductive wastage. Trends Genet 1986;2:105–110. Lahn BT, Page DC. Functional coherence of the human Y chromosome. Science 1997;278:675–680. Lahn BT, Pearson NM, Jegalian K. The human Y chromosome, in the light of evolution. Nat Rev Genet 2001;2:207–216. LeMaire-Adkins R, Radke K, Hunt PA. Lack of checkpoint control at the metaphase/anaphase transition: a mechanism of meiotic nondisjunction in mammalian females. J Cell Biol 1997;139:1611–1619. Lippe B. Turner syndrome. Endocrinol Metab Clin North Am 1991;20: 121–152. Lyon MF. X-chromosome inactivation. Curr Biol 1999;9:R235–R237. Migeon BR, Luo S, Jani M, Jeppesen P. The severe phenotype of females with tiny ring X chromosomes is associated with inability of these chromosomes to undergo X inactivation. Am J Hum Genet 1994;55: 497–504. Miyabara S, Sugihara H, Maehara N, et al. Significance of cardiovascular malformations in cystic hygroma: a new interpretation of the pathogenesis. Am J Med Genet 1989;34:489–501. Page DC. Hypothesis: a Y-chromosomal gene causes gonadoblastoma in dysgenetic gonads. Development 1987;101(Suppl):151–155. Pennington BF, Bender B, Puck M, Salbenblatt J, Robinson A. Learning disabilities in children with sex chromosome anomalies. Child Dev 1982;53:1182–1192. Prueitt RL, Chen H, Barnes RI, Zinn AR. Most X;autosome translocations associated with premature ovarian failure do not interrupt X-linked genes. Cytogenet Genome Res 2002;97:32–38. Reijo R, Lee TY, Salo P, et al. Diverse spermatogenic defects in humans caused by Y chromosome deletions encompassing a novel RNAbinding protein gene. Nat Genet 1995;10:383–393. Ross JL, Scott C Jr, Marttila P, et al. Phenotypes associated with SHOX deficiency. J Clin Endocrinol Metab 2001;86:5674–5680. Sinclair AH, Berta P, Palmer MS, et al. A gene from the human sex-determining region encodes a protein with homology to a conserved DNAbinding motif. Nature 1990;346:240–244. Skuse DH, James RS, Bishop DV, et al. Evidence from Turner’s syndrome of an imprinted X-linked locus affecting cognitive function. Nature 1997;387:705–708.

452

SECTION IV / ENDOCRINOLOGY

Swain A, Narvaez V, Burgoyne P, Camerino G, Lovell-Badge R. Dax1 antagonizes Sry action in mammalian sex determination. Nature 1998; 391:761–767. Therman E, Laxova R, Susman B. The critical region on the human Xq. Hum Genet 1990;85:455–461. Therman E, Sarto G. Inactivation center on the human X chromosome. In: Sandberg A, ed. Cytogenetics of the Mammalian X Chromosome, vol. 3A. New York: Alan R. Liss, 1983; pp. 315–325.

Tsuchiya K, Reijo R, Page DC, Disteche CM. Gonadoblastoma: molecular definition of the susceptibility region on the Y chromosome. Am J Hum Genet 1995;57:1400–1407. Willard HF. X chromosome inactivation, XIST, and pursuit of the X-inactivation center. Cell 1996;86:5–7. Zinn AR, Page DC, Fisher EM. Turner syndrome: the case of the missing sex chromosome. Trends Genet 1993;9:90–93.

43 Disorders of Pubertal Development TOMONOBU HASEGAWA SUMMARY Puberty is the continuous maturation of growth and development to attain full sexual development and fertility by the hypothalamic–pituitary–gonadal axis and other complex endocrine systems. Although the physiology and molecular mechanism of normal puberty are not completely understood, we have made great progress in the management and treatment of disorders of pubertal development such as precocious puberty and delayed puberty. Moreover, the gene abnormalities of disorders of pubertal disorders have been elucidated one after another. This chapter mainly focuses clinical issues and single gene disorders of disorders of pubertal development. Key Words: Congenital adrenal hyperplasia; constitutional delay of puberty; delayed puberty; familial male limited precocious puberty; hypergonadotropic hypogonadism; hypogonadotropic hypogonadism; Kallmann syndrome; Klinefelter syndrome; McCune–Albright syndrome; precocious puberty; premature pubarche; premature; puberty; thelarche; Turner syndrome.

INTRODUCTION Puberty is the continuous maturation of growth and development to attain the full sexual development and fertility by the hypothalamic–pituitary–gonadal axis and other complex endocrine systems. It begins in late childhood and ends in early adulthood. At term, adolescence usually includes psychosocial changes during puberty. This chapter reviews the physiology of normal puberty and the pathophysiology of the disorders of pubertal development.

PHYSIOLOGY OF NORMAL PUBERTY PHYSICAL CHANGES The physical changes of puberty span a continuum (Tabel 43-1). In girls, the first physical change of puberty is usually breast development (thelarche), although pubic hair growth (pubarche) may be first in a minority of instances. Breast development is followed by pubic hair, then axillary hair growth within 1–2 yr. On an average, menarche occurs 2 yr from the onset of breast development. In boys, the first physical change of puberty is always testicular enlargement. The formal measurement of testicular volume is possible using Prader orchidometer. Testicular volume of 4 mL represents the onset of puberty. Testicular enlargement is followed by penile enlargement From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

and development of pubic hair. The development of axillary and facial hair, acne, and voice change occur in the later half of puberty. Adult testicular volumes and penile dimensions are generally achieved between the ages 14 and 18 yr. The pubertal growth spurt shows other sexually dimorphic features. Prepubertal growth is similar in both genders; the growth velocity is highest during infancy and decreases to a nadir before the pubertal growth spurt. With respect to the pubertal growth spurt, the later onset of the pubertal growth spurt in boys leads to approx 2 yr difference in peak height velocity, and boys have greater magnitude of peak height velocity. HORMONAL CHANGES Many dramatic hormonal changes occur during puberty. Two kinds of maturation play central roles. The first involves hypothalamic (gonadotropin releasing hormone; GnRH)–pituitary (gonadotropin)–gonadal axis and the second is the growth hormone (GH)-insulin-like growth factor (IGF)-1 axis. However, it should be emphasized that complex hormonal interactions characterize puberty. Hypothalamic (GnRH)–Pituitary (Gonadotropin)–Gonadal Axis In the prepubertal stage, the hypothalamic-pituitary-gonadal axis is dormant. Hypothalamus and pituitary activity are thought to be suppressed by neuronal restraint pathways and negative feedback by minute amounts of gonadal steroid hormones. Such neuronal restraint pathways are poorly understood. The signals that control the onset of puberty are also poorly characterized. It appears that the mechanisms responsible for the onset of puberty are extremely complex and likely involve the integration of numerous different signals. The onset of puberty is heralded by striking increases in luteinizing hormone (LH) secretion at night, manifested by an increase in amplitude and frequency of LH pulses. At least 1 yr before the clinical evidence of the onset of puberty, low serum concentrations of LH during sleep can be measured in serial serum samples obtained every 10–20 min. This sleep-entrained LH secretion is episodic owing to the pulsatile secretion of GnRH (GnRH pulse generator). By contrast, it is difficult to demonstrate the episodic secretion of follicle-stimulating hormone (FSH), as FSH clearance is slower than LH because of its higher sialic acid content. Gonadotropins are responsible for the maturation of the gonads. As pubertal maturation progresses, the amplitude and frequency of gonadotropin pulses increase during the day, in a pattern similar to that seen at night, until the final stage of sexual maturation is reached. In girls, the GnRH pulse generator ultimately establishes the regular cyclic variations of gonadotropins,

453

454

SECTION IV / ENDOCRINOLOGY

Table 43-1 “Tanner Stage” of Development in Secondary Sexual Characteristics Boys: genital (penis) development Stage 1 Prepubertal: testes, scrotum, and penis of about same size and proportion as in early childhood Stage 2 Enlargement of scrotum and testes. Skin of scrotum reddens and changes in texture Stage 3 Enlargement of penis, at first mainly in length. Further growth of testes and scrotum Stage 4 Increased size of penis with growth in breadth and development of glans. Testes and scrotum larger; scrotal skin darkened Stage 5 Genitalia adult in size and shape Girls: breast development Stage 1 Prepubertal: elevation of papilla only Stage 2 Breast bud stage: elevation of breast and papilla as small mound. Enlargement of areola diameter Stage 3 Further enlargement and elevation of breast and areola, with no separation of their contours Stage 4 Projection of areola and papilla to form a secondary mound above level of breast Stage 5 Mature stage: projection of papilla only, related to recession of areola to general contour of breast Both sexes: pubic hair Stage 1 Prepubertal: vellus over pubes is not further developed than over abdominal wall Stage 2 Sparse growth of long, slightly pigmented, downy hair, straight or slightly curled, chiefly at base of penis or along labia Stage 3 Considerably darker, coarser, and more curled hair. Hair spreads sparsely over junction of pubes Stage 4 Hair now adult in type, but area covered is still considerably smaller than in adult. No spread to medial surface of thighs Stage 5 Adult in quantity and type with distribution of horizontal (or classically “feminine” pattern) Stage 6 Spread up linea alba (male-type pattern)

estrogen, and progesterone of the menstrual cycle. In boys, the same GnRH pulse generator establishes a pattern characterized by relatively constant levels of gonadotropins and testosterone, with minimal diurnal variation. Undoubtedly, androgen is the principal effector of the physical changes of sexual maturation during puberty in boys, whereas estrogen drives the development of secondary sexual characteristics in girls. Several clinical observations support that estrogen is the main contributor to the process of skeletal maturation, growth spurt, and skeletal mineralization in both genders. First, effective suppression of the rapid skeletal maturation in boys with GnRHindependent precocious puberty requires inhibition of aromatase activity to reduce serum estradiol concentrations, in addition to the use of antiandrogens to interfere with androgen action. Second, patients with aromatase deficiency of both genders exhibit delayed skeletal maturation and have no pubertal growth spurts. Third, patients with complete androgen insensitivity (46,XY) exhibit pubertal growth spurts, suggesting that the growth spurt in boys is not mediated via the androgen receptor, but is instead controlled indirectly via the estrogen receptor, after the conversion of testosterone to estrogen. GH-IGF-1 Axis The secretion of GH increases two- to threefold during puberty. GH is produced by the somatotrophs of the anterior pituitary gland and its secretion is regulated mainly by the effects of GH-releasing hormone and somatostatin. Treatment of late pubertal boys with an estrogen receptor antagonist diminished GH secretion, suggesting the critical role of estrogen to increase GH secretion during puberty, even in boys. The mechanism by which estrogen enhances GH secretion is not completely understood. The effect of GH is primarily through hepatic synthesis of IGF-1. During puberty, increased GH secretion leads to increased serum IGF-1 concentrations, which correlate better with pubertal Tanner stages than chronological age. Adrenal Androgens Before puberty, the increased secretion of adrenal androgens (dehydroepiandrosterone [DHEA] and dehydroepiandrosterone-sulfate [DHEA-S]) by the zona reticularis (“adrenarche”) occurs. Although an adrenal androgen-stimulating factor has been postulated to induce adrenarche, no such factor

has been identified. The physiological role of adrenarche has not been elucidated, although adrenal androgen may be functioning in development of scrotal and labial hair. Importantly, the timing of adrenarche and puberty are independent. Leptin Leptin produced by adipose tissue has a key role in regulating the onset of puberty and/or pubertal development. In boys, serum leptin concentrations increase just before the onset of puberty and decrease after the initiation of puberty. In girls, serum leptin concentrations increase after the onset of puberty, remain stable in midpuberty, and further increase in late puberty. Judging from animal studies, leptin alone is insufficient to induce or promote puberty. However, individuals with congenital leptin deficiency or congenital leptin resistance resulting from mutations of the leptin (LEP) or leptin receptor (LEPR) genes exhibit hypogonadotropic hypogonadism. Leptin treatment to a 9-yr-old girl with congenital leptin deficiency induced an early pubertal pattern of LH response in GnRH stimulation tests.

THE PATHOPHYSIOLOGY OF DISORDERS OF PUBERTAL DEVELOPMENT PRECOCIOUS PUBERTY Precocious puberty has been generally defined as the onset of pubertal development before the age of 8 yr in girls and before the age of 9 yr in boys. This definition is arbitrary; however, because of the variation in the age at which puberty begins among different ethnic groups (see Table 43-2). In 1999, the Lawson Wilkins Pediatric Endocrine Society proposed new guidelines not to evaluate girls with either breast or pubic hair development after the age of 7 yr in Caucasians and after the age of 6 yr in African Americans. This has not been universally accepted, as some authors believe that the definition of precocious puberty being pubertal development before the age of 8 yr in girls may identify girls with onset of puberty between 6 and 8 yr who may benefit from treatment. Incomplete (or partial) precocious puberty is isolated manifestations of pubertal development without other signs of puberty such as premature thelarche and premature pubarche (described in Incomplete [or Partial] Precocious Puberty [Variation of Normal Puberty]).

CHAPTER 43 / DISORDERS OF PUBERTAL DEVELOPMENT

Table 43-2 Mean Age (Years) at the Onset of Development of Breast, Pubic Hair, and Menarche in Different Ethnic Groups

Caucasian African-American Mexican-American

Breast

Pubic hair

Menarche

10.3 9.5 9.8

10.5 9.5 10.3

12.7 12.1 12.2

Classification Precocious puberty is classified as GnRHdependent or GnRH-independent. GnRH-dependent precocious puberty is because of the early activation of the normal pubertal hypothalamic–pituitary–gonadal axis. GnRH-gonadotropin activates the gonads leading to increasing gonadal steroid hormone secretion and progressive sexual maturation. Conversely, GnRH-independent precocious puberty is not because of the early activation of normal puberty. GnRH-dependent precocious puberty is sometimes called gonadotropin-dependent, central, or true precocious puberty. GnRH-independent precocious puberty is also called gonadotropinindependent, peripheral, or pseudoprecocious puberty. The terms “GnRH-dependent” and “GnRH-independent” are employed in this chapter. This terminology is preferred as GnRH is essential for the activation of the gonadotropin–gonadal axis in normal puberty. Furthermore, precocious puberty because of an intracranial human chorionic gonadotropin (hCG)-secreting tumor is definitely GnRH-independent, but gonadotropin dependent and central precocious puberty. The second classification of precocious puberty is isosexual or heterosexual (contrasexual). “Heterosexual” refers to feminization in boys or virilization in girls. GnRH-dependent precocious puberty is always isosexual by definition, whereas GnRH-independent precocious can be either isosexual or heterosexual. General Evaluation Evaluation of patients with early pubertal development should always begin with a detailed clinical history. The sequence of sexual developmental changes is most important in differentiating GnRH-dependent and GnRH-independent forms. History should focus on central nervous system (CNS) abnormalities (trauma, infection, neoplasia, irradiation, and so on). Growth charts are indispensable to discern a pattern of rapid linear growth. On physical examination in girls, determining whether maturation is isosexual or heterosexual is critical. In boys, testicular volume should be measured. Any pubertal development without testicular enlargement (less than 3 mL) suggests GnRH-independent precocious puberty. In both sexes, neurological abnormality that suggests CNS disorders should be assessed. Diagnostic evaluation starts with the assessment of bone age and the measurements of serum gonadotropins, particularly LH. Advanced bone age indicates gonadal steroid hormone action. Sensitive immunometric assays for serum LH such as the immunochemiluminometric assay are widely available. Using such assays, serum LH concentrations are detectable in more than half of children with GnRH-dependent precocious puberty. A random LH more than 0.3 mIU/mL by immunochemiluminometric assay is highly suggestive of GnRH-dependence. By contrast, gonadotropins are undetectable in prepubertal children and children with GnRH-independent precocious puberty. Measurements of estradiol and testosterone should be obtained. Undetectable estradiol, however, does not exclude the absence of precocious puberty, as the detection limits of the current assays are not

455

Table 43-3 Causes of GnRH-Dependent Precocious Puberty Idiopathic CNS disorders CNS tumors Hypothalamic hamartoma, astrocytoma, ependymoma, craniopharyngioma, optic glioma, and others Other CNS lesions Static encephalopathy (a result of infection, neonatal asphyxia, hypoxia, severe head trauma, and others) Low-dose cranial radiation Hydrocephalus Arachnoid cyst Maternal uniparental disomy for chromosome 14 Combined GnRH-dependent and GnRH-independent Treated congenital adrenal hyperplasia McCune–Albright syndrome, late Familial male limited precocious puberty, late Functional ovarian cyst, late

sufficiently sensitive. Serial measurements of serum LH concentration, usually every 10–20 min, during nocturnal sleep have greater diagnostic power than single random waking measurements. Pulsatile secretion of LH during nocturnal sleep is well illustrated in GnRH-dependent precocious puberty. Measurement of both LH and FSH concentrations following a GnRH stimulation test is the standard to differentiate GnRHdependent from GnRH-independent precocious puberty. LH predominance or a peak LH to FSH ratio >1.0 in the GnRH stimulation test is diagnostic for GnRH dependence in any assays. FSH predominance, however, can be present in the early phase of GnRH-dependent precocious puberty. Once GnRH-dependent precocious puberty is confirmed, cranial CT and/or MRI is warranted for the CNS disorders (Table 43-3). Additional useful imaging studies are pelvic ultrasonography and MRI. Bilateral ovarian volume more than 2 mL together with large bilateral cysts (>9 mm) suggests GnRH-dependence. A uterine length more than 3.5 cm and a fundus to cervix ratio more than 1.0 indicate estrogen activity. GnRH-Dependent Precocious Puberty GnRH-dependent precocious puberty is more than 10-fold more frequent in girls than boys, although the reason for this skewed sex ratio is not clear. GnRH-dependent precocious puberty is most often idiopathic in girls, whereas a CNS disorder is demonstrated in 25–75% of boys. Clinically the sequence of sexual development is preserved. In girls, the first sign is breast development, followed by the appearance of pubic hair, axillary hair, and withdrawal bleeding. In boys, enlargement of testes is the first, although most children do not notice this change. This is followed by enlargement of the penis, the appearance of pubic hair, and facial acne. Erections and nocturnal emissions are not uncommon. Causes Of cases of GnRH-dependent precocious puberty in girls, two-thirds are idiopathic (Table 43-3). The mechanism of the early activation of GnRH activity in this category is largely unknown. CNS disorders are well-recognized causes of GnRH-dependent precocious puberty. Precocious puberty with any neurological signs and symptoms, including developmental delay, suggests GnRH-dependence. Although less common, CNS tumors are important causes of precocious puberty, particularly in girls younger than 6 yr of age

456

SECTION IV / ENDOCRINOLOGY

and boys of any age. CNS tumors causing GnRH-dependent precocious puberty are rarely malignant, and can be viewed as causing precocious puberty by one of two distinct mechanisms. First, tumors may act as ectopic GnRH pulse generators that have escaped the normal inhibitory influences exerted in the prepubertal period. Hypothalamic hamartoma is representative of this type of disorder and is the most common CNS lesion associated with precocious puberty. It is more frequently found in boys and the onset is usually before 3 yr of age in girls. The association of gelastic (laughing) episodes or seizures is classic, but rare. Hypothalamic hamartomas are visualized as iso-intense masses on MRI. Histologically, such hamartomas are benign congenital CNS malformations made up of disorganized but otherwise normal neuronal and glial elements. In other instances, tumors may interrupt the normal inhibitory pathways exerted in the prepubertal period. Astrocytomas, ependymomas, craniopharyngiomas, and optic gliomas have all been identified in such cases. The possibility, however, cannot be excluded that focal derangements of the cellular environment in the vicinity of GnRH neurons may be causally related to premature activation of the GnRH pulse generator. Other CNS lesions are relatively common. Static encephalopathy is a result of infection, neonatal asphyxia, hypoxia, severe head trauma, and other causes during the neonatal period, infancy, or early childhood. These lesions can interrupt the normal tonic inhibitory pathways exerted in the prepubertal period. Cranial radiation has dose-dependent dual effects on hypothalamic– pituitary–gonadal axis. Low doses of cranial radiation (18–24 Gy) may cause precocious puberty. Low-dose cranial radiation has been widely used in the CNS prophylactic treatment of acute lymphoblastic leukemia and has been associated with a downward shift in the distribution of ages at pubertal onset and menarche in girls who have received such treatment. Precocious puberty in girls owing to prior low-dose cranial radiation is increasing in frequency. By contrast, precocious puberty is rare in boys treated in this manner. High doses of cranial radiation (more than 50 Gy to the hypothalamic-pituitary area) may cause gonadotropin deficiency. The chance of finding CNS disorders in both sexes may be inversely proportional to the age of the child, with the greatest yield in children younger than 4-yr old. Routine MRI is less likely to have positive findings in girls whose pubertal development began after 6 yr of age. In contrast, in a series of 4000 children referred to the National Institutes of Health (NIH), one-third of the girls and more than 90% of the boys had identifiable CNS disorders visible on CT or MRI. This high prevalence of CNS disorders may be biased by the composition of the referral population. Maternal uniparental disomy for chromosome 14 (UPD14) may cause GnRH-dependent precocious puberty and intrauterine growth retardation. GnRH-dependent precocious puberty in maternal UPD14 results from the loss of the functionally active paternally derived allele. Evidence for a maternal imprinting (paternal expressive) gene(s) on chromosome 14 includes the presence of GnRH-dependent precocious puberty and intrauterine growth retardation in cases with either maternal heterodisomy or isodisomy, and that paternal UPD14 results in a very different phenotype. Treatment The most important but difficult issue is to determine when treatment is indicated. Treatment goals are to improve adult height and to prevent psychosocial problems, including ageinappropriate treatment. Natural history of GnRH-dependent precocious puberty is the early completion of physical growth and

sexual maturation. Despite early acceleration ahead of peers, early epiphyseal fusion usually leads to compromised adult height compared with genetic height potential. Major psychosocial problems are the disruption of familial, educational, and peer relationships, because adults may relate to these children in response to their sexual maturity rather than their chronological age. Some “adolescent” behavior may be exhibited, although sexual behavior such as increased fondling and masturbation are infrequent. Advanced sexual maturity may place these children at increased risk of being sexually abused. Thus, when compromised adult height is highly likely or any psychosocial problems occur, treatment is absolutely required. On the contrary, when compromised adult height is less likely and psychosocial problems do not occur, treatment should be postponed and the child closely observed. Not all children with GnRH-dependent precocious puberty require treatment. Very slowly progressing sexual precocity is possible and may not require treatment. This group can be characterized by the following: the onset age is near 8 yr in girls and 9 yr in boys, and the bone age is less than 2 yr advanced in comparison to chronological age. The first choice of treatment is long-acting synthetic agonist analogues of GnRH (GnRHa). GnRHa induces “hypogonadism” by the continuous nonpulsatile GnRH action on the pituitary gonadotrophs. The first demonstrable endocrinological change effected by this treatment is reduction in basal and GnRH-stimulated LH and FSH. GnRHa treatment results in suppression of pubertal physical changes, a decrease in growth velocity, and the disappearance of psychosocial problems. The majority of treated girls experience no increase in breast development, and one-third show regression to an earlier Tanner stage. Effects on pubic hair are less predictable, although most children show either no progression or a minor degree of regression. When treated appropriately to prepubertal endocrinological status, the slowing of growth velocity is accompanied by slowing of bone age maturation. Preservation of or increase in adult height can be achieved. The length of time that treatment is continued depends on bone age and estimates of adult height in individual cases. The adult height by GnRHa treatment is higher than predicted height at the start of treatment, but lower than target height and predicted height at the end of treatment. Most studies suggest that the prognosis of adult height is better if bone ages are relatively young at the start of treatment, indicating the importance of early diagnosis and intervention. The combination of GH and GnRHa treatment may improve the prognosis of adult height more than GnRHa treatment alone. Clinical trials addressing this issue are ongoing. Two major concerns have been raised regarding potential side effects of GnRHa treatment. First, GnRHa treatment may suppress the increase of bone mineral density (BMD), an important feature of normal puberty. Despite this, it has been reported that bone mineral density was not compromised in girls treated with GnRHa for GnRH-dependent precocious puberty who had completed treatment and had subsequently attained a bone age of greater than 14 yr. Second, some have questioned the reversibility of the suppression of pituitary–gonadal axis (“hypogonadism”) following cessation of GnRHa treatment. The available data show that pubertal development resumed after withdrawal of treatment. Theoretically, fertility should not be hampered by such treatments, but longitudinal follow-up is necessary. GnRH-Independent Precocious Puberty GnRH-independent precocious puberty is about one-fifth as common as

457

CHAPTER 43 / DISORDERS OF PUBERTAL DEVELOPMENT

Table 43-4 Causes of GnRH-Independent Precocious Puberty Boys and girls Late onset congenital adrenal hyperplasiaa McCune–Albright syndrome Peutz–Jeghers syndrome Adrenal carcinoma associated with Cushing syndromea Primary hypothyroidism, prolonged and untreated Iatrogenic Boys Testicular disorders Familial male limited precocious puberty Leydig cell adenoma Human chorionic gonadotropin-secreting tumors Androgen-secreting tumors Girls Ovarian disorders Granulosa cell tumors Functional ovarian cysts Other estrogen-secreting tumors aIsosexual

in boys but heterosexual in girls.

GnRH-dependent forms. This form of precocious puberty is clinically characterized by sexual development of unusual sequence, no history of CNS abnormality, and no rapid linear growth. From an endocrine standpoint, GnRH-independent forms are characterized by increased production of gonadal steroid hormones in the absence of activation of the hypothalamic–pituitary axis (Table 43-4). GnRHindependent precocious puberty includes conditions that mimic the effect of pituitary gonadotropins on gonadal function, such as hCG from germ cell tumor. Molecular mechanisms underlying two forms of GnRH-independent precocious puberty, McCune–Albright syndrome (MAS) and familial male limited precocious puberty (FMPP), are discussed later. McCune–Albright Syndrome MAS originally was defined by the clinical triad of precocious puberty, polyostotic fibrous dysplasia of bone, and café au lait spots of skin. It was advocated that MAS’ definition should be broader than the triad; fibrous dysplasia plus at least one of the typical hyperfunctioning endocrinopathies and/or café au lait spots. Reasons why a broadening of the definition has been proposed include that although MAS is rare, clinical manifestations vary individually and almost any combination of the clinical features (Table 43-5) is possible. Second, other hyperfunctioning endocrinopathies can be found in association with precocious puberty, including hyperthyroidism, GH excess, Cushing’s syndrome, and so on. Third, clinical manifestations other than those within the triad are possible, such as liver dysfunction, idiopathic pancreatitis, and sudden death. Fourth, fibrous dysplasia is not rare and is the most common component of MAS. Fibrous dysplasia can involve a single skeletal site or multiple sites. Fifth, the clinical features of MAS may develop over time. Prevalence of the different clinical features of MAS in the NIH series is summarized in Table 43-5. Precocious puberty in MAS is GnRH-independent, although the coexistence of GnRH-dependent and GnRH-independent forms is possible in longstanding cases. The prevalence of precocious puberty is less in boys, although that of gonadal abnormalities between boys and girls is similar. In affected girls, the first

Table 43-5 Prevalence of Clinical Features of MAS Clinical findings Fibrous dysplasia Café au lait spots Gonadal abnormalities (boys and girls) Boys: precocious puberty and/or abnormal testes on US or Leydig cell hyperplasia Girls: precocious puberty Thyroid abnormalities (total) Hyperthyroidism Abnormal gland on US only Renal phosphate wasting Growth hormone excess Cushing’s syndrome Primary hyperparathyroidism Idiopathic pancreatitis

Prevalence (%) 99 82 82 77 78 69 34 34 49 18 7 4 4

sign of sexual development is usually vaginal bleeding or spotting. A waxing and waning course of precocious puberty is not uncommon. Ultrasonography and/or MRI scan may reveal ovarian cysts. The peculiar bone lesion of MAS is fibrous dysplasia. Fibrous dysplasia can occur in any bone, with the common locations being the skull base and proximal femurs. The clinical manifestations can be pain, limp, or pathological fractures. Alternatively, fibrous dysplasia in craniofacial bones can cause a painless lump or facial asymmetry. Radiographs show expansile lesions. Sarcomatous degeneration occurs rarely. Café au lait spots are usually present at birth or shortly thereafter. These are large and irregular and have serrated outlines, called “coast-of-Maine.” Common sites of café au lait spots are the forehead, the neck or upper back, the shoulder and upper arm, the lumbosacral region, and the buttocks, and often limited to half of the body. Constitutive active mutations of the GNAS1 gene have been identified in mosaic fashion (somatic mosaicism) in MAS (Fig. 43-1A,B). GNAS1 encodes the α-subunit of the stimulatory G protein (Gsα), which regulates adenylate cyclase activity. G protein has a key role in intracellular signal transduction of G protein-coupled receptors (GPCR). In these membrane receptors, ligands bind GPCR, activating the adenylate cyclase system, the membrane-bound enzyme that catalyzes the formation of the intracellular second messenger, cAMP. A heterozygous substitution was identified in exon 8 of the GNAS1 gene, predicting the replacement of arginine by histidine or cysteine at position 201 of the mature Gsα protein. This substitution causes a marked decrease in the intrinsic GTPase activity of Gsα, prolonging the survival of the active conformation of GTP and resulting in constitutive activation of adenylate cyclase activity. The consequent increased production of cAMP explains the hyperfunctioning of multiple endocrine organs where GPCRs are expressed. Moreover, more mutant allele was expressed than wild type in tissues histologically most affected. This mutation was detected in bone as well as skin lesions. This somatic mutation may occur very early in gestation leading to prenatal lethality, relatively early in gestation leading to typical MAS because of abnormal monoclonal cell population, or later in life leading to hyperfunction in single endocrine organ. Because these mutations arise somatically, MAS is not inherited.

458

SECTION IV / ENDOCRINOLOGY

Figure 43-1 In each panel, the parallel horizontal lines are the plasma membrane with the extracellular space above and the intracellular space below. GPCR is G protein-coupled receptor and AC is adenylate cyclase. (A) GPCR in normal state. In the basal state, the G protein is a heterotrimer (α-, β-, and γ-subunits) with guanosine diphosphate (GDP) tightly bound to the α-subunit (Gsα). When ligand binds to the G proteincoupled receptor, the interaction with the G proteins leads to the dissociation of guanosine diphosphate, the binding of GTP, and the dissociation of the Gsα and the βγ-subunits. GTP-bound Gsα activates AC activity, inducing signal transduction. Under physiological conditions, the regulation of AC activity is transient and is terminated by the hydrolysis of the bound GTP by the intrinsic GTPase activity of Gsα. The Gsα-subunit with bound GDP reassociates to form the αβγ–heterotrimer. (B) GPCR in McCune–Albright syndrome (MAS). In MAS, constitutive active mutations of the GNAS1 gene encoding Gsα have been identified in mosaic fashion. The mutated Gsα (αMUT) has a markedly reduced intrinsic GTPase activity. When ligand binds GPCR, activated GPCR induces the release of GDP, the binding of GTP, the activation of Gsα, its dissociation from the βγ-dimer, and the regulation of AC activity. Owing to the decreased intrinsic GTPase activity of the Gsα, activated Gsα is not inactivated efficiently, leading to constitutive induction of signal transduction.

For GnRH-independent precocious puberty in MAS, the use of an aromatase inhibitor is recommended. In boys, an aromatase inhibitor is frequently combined with an antiandrogen agent. Surgery is the mainstay for treatment of fibrous dysplasia. Bisphosphonate has been used to reduce pain, although no evidence is available to indicate whether such treatments can alter the natural history of fibrous dysplasia. Effective treatment for café au lait spots has not been described. Familial Male-Limited Precocious Puberty FMPP or familial testotoxicosis is an autosomal-dominant, male-limited form of GnRH-independent precocious puberty. Histologically, Leydig cell hyperplasia is evident. FMPP is caused by a heterozygous constitutive active mutation of the gene encoding luteinizing hormone/ chorionic gonadotropin (CG) receptor (LHCGR). FMPP is characterized by 1. GnRH-independent precocious puberty. 2. Male-limited autosomal-dominant inheritance. 3. The onset of puberty before 4 yr of age. A family history of precocious puberty and/or short adult height in males is usually obtained. Sporadic cases have also been reported. Affected males present with progressive virilization by age 4 yr. The clinical hallmark is the relatively mild enlargement of the testes for the degree of virilization. Testosterone production in testes can occur autonomously with increased serum testosterone concentration. Typically, affected subjects have advanced bone maturation, leading to short adult height. Females carrying the same heterozygous mutation are not affected, because activation of both LH and FSH receptors is required for estrogen production in the ovarian follicle. The molecular pathogenesis of FMPP has been traced to heterozygous constitutively active mutations within specific segments of the LHCGR gene. The LH/CG receptor is a member of the GPCR family. In vitro studies of such mutant receptors have

demonstrated marked increase in cAMP production in the absence of ligand, consistent with GnRH-independent precocious puberty. Treatment aims to reduce testosterone production (in the testes) and androgen action (peripherally). Androgen receptor antagonists, aromatase inhibitors, inhibitors of 17, 20 lyase, or the combination of these agents have been employed. Late-Onset Congenital Adrenal Hyperplasia Congenital adrenal hyperplasia (CAH) is an autosomal-recessive disorder characterized by deficiencies of one of the enzymes critical to normal adrenal steroidogenesis (Chapter 37). In CAH, compensatory hypersecretion of adrenocorticotropic hormone (ACTH) stimulates the increased adrenal androgen. GnRH-independent precocious puberty can be induced by this increased adrenal androgen. Late onset (or nonclassic) CAH should be considered in any child with premature pubarche, especially in association with enlarged penis without testicular enlargement in boys (isosexual precocious puberty) and in association with clitoromegaly in girls (heterosexual precocious puberty). When untreated, rapid linear growth and advanced bone age are likely. Deficiency of 21-hydroxylase accounts for more than 90% of CAH, which is caused by homozygous or compound heterozygous CYP21A2 mutations. Deficiency of 11β-hydroxylase owing to homozygous or compound heterozygous CYP11B1 mutations is less frequent. hCG-Secreting Tumor hCG-secreting tumors, such as germinoma, hepatoblastoma, or pinealomas, may lead to GnRH-independent isosexual precocious puberty exclusively in boys. Girls with hCG-secreting tumors do not develop precocious puberty, just as girls with LHCGR mutation of FMPP do not. The intracranial pineal region is the common site for germinoma. Mediastinal germinoma producing hCG have been reported in Klinefelter syndrome (described in Klinefelter Syndrome). Precocious puberty caused by hCG secreting tumors is definitely GnRH-independent, as no hypothalamus-pituitary activation is evident. Because of the extensive structural homology of the β-subunits of hCG and LH, at

CHAPTER 43 / DISORDERS OF PUBERTAL DEVELOPMENT

high levels, hCG can stimulate the production of testosterone via activation of the LH/CG receptor in Leydig cells. Hormone Production From Tumors in Ovary, Testis, or Adrenal Glands The most common clinical presentation of granulosa cell tumors of the ovary is GnRH-independent precocious puberty, as most can produce estrogen. Granulosa cell tumors are the most common type of ovarian tumors in children, although the etiology is unknown. Tumors are often associated with a large cyst or solid mass, which are easily visible on pelvic ultrasound. They are benign and noninvasive. Surgical resection is the treatment of choice. Peutz–Jeghers syndrome is an autosomal-dominant disorder characterized by 1. Melanocytic macules of the lips, buccal mucosa, and digits. 2. Multiple gastrointestinal hamartomatous polyps. 3. An increased risk of various neoplasms. Ovarian theca cell tumors in girls with this syndrome can cause GnRH-independent precocious puberty. Gynecomastia (GnRHindependent heterosexual precocious puberty) has been reported in boys with Peutz–Jeghers syndrome having a testicular sex-cord tumor with increased aromatase activity. The occurrence of ovarian tumors in girls, however, far exceeds that of testicular tumors in boys. A deletion of exons 4 and 5 and an inversion of exons 6 and 7 in an STK11 encoding a widely expressed serine/threonine kinase was found in a large family with Peutz–Jeghers syndrome. Other heterozygous mutations in STK11 have been identified in this disorder. The germline mutations in STK11 in conjunction with somatic mutations in a second allele have been proposed to cause neoplasms in this syndrome. Leydig cell adenomas are associated with GnRH-independent precocious puberty in boys. In boys with Leydig cell adenomas, signs of isosexual precocious puberty typically appear between the ages of 5 and 9 yr, later than FMPP. The somatic mutation of LHCGR gene in these tumors has been reported. This mutation was proven to be constitutively active, analogous to those seen in FMPP, leading to GnRH-independent precocious puberty. Surgical resection of the tumor results in arrest of pubertal development. Adrenal carcinoma associated with Cushing syndrome sometimes causes GnRH-independent precocious puberty, whereas adrenal adenoma rarely does. Adrenal carcinoma can produce adrenal androgen, leading to isosexual precocious puberty in boys, and heterosexual precocious puberty in girls. Serum concentrations of DHEA-S, DHEA, and androstenedione are usually elevated. Small adrenal carcinoma is curative by surgery, but larger ones may require additional treatments such as chemotherapy. Incomplete (or Partial) Precocious Puberty (Variation of Normal Puberty) PREMATURE THELARCHE Premature thelarche is an isolated breast development without other signs of puberty in girls that commonly occurs in the first 2 yr. Rapid linear growth and advanced bone age are absent. The natural course is spontaneous regression often within 18 mo or no additional change in breast size. Premature thelarche is typically associated with a degree of FSH secretion. If performed, a GnRH stimulation test usually reveals an FSH predominance response. Serum estradiol is undetectable. The prevalence of ovarian microcysts detected by ultrasonography is increased. The etiology of premature thelarche has

459

not been elucidated, although it has been attributed to an increased sensitivity of the breast tissue to estrogen. Careful follow-up for the appearance of other pubertal signs and linear growth is indicated. Premature thelarche can develop to GnRH-dependent precocious puberty in approx 10–15% of cases. PREMATURE PUBARCHE Premature pubarche is the isolated development of pubic hair, which is usually a benign condition resulting from early adrenarche. A report of the characteristics of 171 subjects with premature pubarche included abnormalities of steroidogenesis (late-onset 21-hydroxylase deficiency, and so on) in 12% of patients as diagnosed by ACTH stimulation tests. Lateonset CAH should, therefore, be considered in any child with premature pubarche. Premature pubarche must be followed closely to monitor for developing GnRH-dependent precocious puberty. Some girls with premature pubarche may develop functional hyperandrogenism in the midteenage years associated with polycystic ovarian syndrome. DELAYED PUBERTY Delayed puberty is defined as the failure to mature sexually at an appropriate age. Both delayed onset of puberty and failure to complete pubertal development are considered as delayed puberty. In the United States, the ages of 12 or 13 yr in girls and 14 yr in boys serve as practical guidelines to determine the need for evaluation for delayed onset of puberty. In girls, if more than 5 yr have elapsed between the beginning of breast development and menarche, failure of complete pubertal development is considered. In boys, 3–4 yr are normal to complete pubertal development. Classification Delayed puberty is classified as hypogonadotropic (or secondary) hypogonadism, hypergonadotropic (or primary) hypogonadism, and constitutional delay of puberty. Hypogonadotropic hypogonadism is failure of maturation of hypothalamic-pituitary axis, which is the stimulation signal for the gonads. Hypergonadotropic hypogonadism is the failure of gonads to respond to hypothalamic-pituitary stimulation. Constitutional delay of puberty is delayed maturation of hypothalamicpituitary–gonadal axis (variation of normal puberty). General Evaluation Evaluation of patients with delayed pubertal development begins with a detailed clinical history of the onset and pattern of puberty. Recording growth charts (heights and weights) throughout childhood is mandatory. Consistent growth delay (short stature) suggests constitutional delay of puberty. A history of CNS tumors and their treatment, ability to smell, hypopituitarisim, any medication causing hyperprolactinemia as well as general medical history is relevant in both genders. Past history should be focused on cryptorchidism, hypospadias, or micropenis in boys and coarctation of the aorta in girls. Family history is important such as growth, onset and patterns of puberty, infertility, and ability to smell. A family history of delayed onset of puberty suggests constitutional delay of puberty. Physical examination starts with measurements of height, weight, upper– lower segment ratio, and arm span. The evaluation of Tanner stage as well as pubertal development should be performed. In boys, testicular size must be measured. Careful general examination is important to determine whether any major or minor anomalies are present. Neurological examination including the ability to smell should be done routinely. The initial diagnostic evaluation usually involves the assessment of bone age and measurements of basal serum LH, FSH, and gonadal steroid hormones (testosterone in boys and estradiol in girls). A bone age more than 13.5 yr in boys or 11.5 yr in girls

460

SECTION IV / ENDOCRINOLOGY

suggests an active hypothalamic-pituitary axis. Very high concentrations of serum LH and FSH with low gonadal steroid hormones indicate hypergonadotropic hypogonadism. In such cases, a GnRH stimulation test is unnecessary as the hyper response in hypergonadotropic hypogonadism can be predicted by high basal gonadotropins. The difficult task is the differentiation between hypogonadotropic hypogonadism and constitutional delay of puberty. In prepubertal ages, neither basal gonadotropins nor gonadal steroid hormones can differentiate between the two. It should be noted that gonadotropin responses following GnRH stimulation are seldom helpful. GnRH stimulation tests yield overlapping responses that do not permit the differentiation of hypogonadotropic hypogonadism and constitutional delay of puberty. In boys, a hCG stimulation test may provoke a similar response in both hypogonadotropic hypogonadism and constitutional delay of puberty. Serial measurements of serum LH during nocturnal sleep may reveal pulsatile patterns suggestive of puberty, but this is not useful to differentiate hypogonadotropic hypogonadism from constitutional delay of puberty at prepubertal ages. Once hypogonadotropic hypogonadism is confirmed, cranial CT and/or MRI is warranted for exclusion of CNS disorders (Table 43-6). MRI readily visualizes olfactory bulbs in healthy children even in neonates. Olfactometry or an intravenous olfactory stimulation test may be required to reveal hyposmia. Hypogonadotropic Hypogonadism Hypogonadotropic hypogonadism refers to the deficiency of pulsatile release of gonadotropins, which may result from a variety of hypothalamic and pituitary disorders. In the presence of a hypothalamic defect, or absence of GnRH-secreting neurons, failure of GnRH secretion results in a lack of stimulation of pituitary gonadotrophs. In contrast, pituitary disorders, such as tumors or hypophysitis, cause direct failure of pituitary gonadotropin secretion. In practice, it is not uncommon to encounter difficulty in determining whether the origin of hypogonadotropic hypogonadism is hypothalamic or pituitary. Causes No human GnRH deficiency owing to GNRH mutation has been reported, although a mouse Gnrh mutation (hpg mouse) exists (Table 43-6). Molecular analysis of the hpg mouse revealed a deletion in the hpg genome of at least 33.5 kb removing the two exons of the Gnrh gene. This deletion resulted in a transcriptionally active gene that leads to the synthesis of a truncated GnRH protein and a hypogonadotropic hypogonadism in homozygous state. GnRH resistance owing to homozygous or compound heterozygous mutations in the GnRH receptor gene has been described in up to 20% of patients with “idiopathic” hypogonadotropic hypogonadism. The clinical features of GnRH receptor gene mutations are variable. Male patients may have prepubertalsized or adult-sized testes. All female patients had primary amenorrhea, but showed no or Tanner stage III breast development. In one female patient, pulsatile administration of GnRH induced ovulation and allowed pregnancy. Kallmann’s syndrome is discussed later. One human LHβ deficiency resulting from LHB gene mutation has been reported. The affected individual had delayed puberty and arrested spermatogenesis. Laboratory data showed a high serum concentration of immunoreactive LH with low serum testosterone. A homozygous loss of function missense mutation Q54R was identified in the LHB gene. This mutation alters an

Table 43-6 Causes of Hypogonadotropic Hypogonadism GnRH or GnRH receptor deficiency GnRH deficiency (GNRH mutation)a GnRH resistance (GNRHR mutation) Others Kallmann’s syndrome KAL1 mutation FGFR-1 mutation Others Isolated gonadotropin deficiency LHβ deficiency FSHβ deficiency GPR54 deficiencyb Others Hypogonadotropic hypogonadism with other endocrine abnormalities due to single gene disorder PC1 deficiency α-subunit deficiencya Congenital leptin deficiency (LEP mutation) Congenital leptin resistance (LEPR mutation) DAX1 mutation Others Congenital combined pituitary hormone deficiency HESX1 mutation LHX3 mutation LHX4 mutation PROP1 mutation EGR1 mutationa Others Panhypopituitarism with invisible or thin pituitary stalk CNS disorders Tumor Irradiation Surgery Anatomic defect Autoimmune disease Others Functional gonadotropin deficiency Anorexia nervosa Excessive exercise Chronic systemic illness Emotional stress Malnutrition Hyperprolactinemia Others aNo

human mutations have been reported so far. level of abnormality, hypothalamus or pituitary, is not fully understood. bThe

amino acid that is conserved in all β-subunits of the glycoprotein hormones. CHO cells were transfected with cDNAs encoding αglycoprotein subunit (GSU) and either the wild or mutant LHB to assess the biological effects of the mutation. When LH concentrations were measured in culture medium by radioimmunoassay and radioreceptor assay, LH concentrations in the medium by radioimmunoassay were higher in cells transfected with αGSU and mutant LHB cDNAs compared with those transfected with αGSU and wild LHB cDNAs. However, the mutant LH was undetectable by radioreceptor assay, indicating that the absence of biological activity of mutant LH was because of its inability to bind its receptor.

CHAPTER 43 / DISORDERS OF PUBERTAL DEVELOPMENT

A small number of women have been reported to have FSHβ deficiency resulting from homozygous or compound heterozygous FSHB gene mutations, V61X/V61X and V61X/C51G. These women presented with delayed puberty and primary amenorrhea. Laboratory data showed an undetectable serum concentration of FSH and high LH. Exogenous FSH treatment resulted in follicular maturation, ovulation, and fertility. V61X caused a deletion of the C-terminus, including residues 90-110, that was essential for heterodimer formation. The C51G substitution mutates a residue that is part of a motif thought to be essential to the organization of the core of the protein. The functional effects of V61X and C51G were studied by coexpressing αGSU and mutant FSHB genes. Immunoradiometric assay of culture media showed undetectable concentrations of mutant FSH, consistent with patients’s laboratory data. Two men have been reported to have a homozygous FSHB gene mutation, V61X/V61X or C82R/C82R, presenting with azoospermia. Other patients have been reported to have isolated gonadotropin deficiency. However, in most instances the LHB and FSHB gene analyses have not been reported. Homozygous or compound heterozygous loss of function mutation of the GPR54 gene causes hypogonadotropic hypogonadism. GPR54 is a GPCR whose ligand is a 54 amino acid peptide derived from the KiSS1 protein. Affected subjects carried a homozygous deletion of 155 nucleotides encompassing the splicing acceptor site of intron 4-exon 5 junction and part of exon 5, homozygous L102P, homozygous L148S, or compound heterozygous R331X/X339R. The transfection of COS-7 cells with L148S, R331X, or X339R plus introduced stop codon after polyA tail showed decreased accumulation of inositol phosphate by ligand stimulation. The patient with R331X/X339R had attenuated secretion of endogenous GnRH. The Gpr54 knockout mice had isolated hypogonadotropic hypogonadism, but they showed responsiveness to both exogenous gonadotropins and GnRH and had normal levels of GnRH in hypothalamus. These studies established the central role for GPR54 in GnRH secretion. Hypogonadotropic hypogonadism with other endocrine abnormalities resulting from single gene disorder has been reported, consistent with the complex interactions within the hypothalamicpituitary–gonadal axis and other endocrine systems being absolutely required for normal puberty. For example, prohormone convertase (PC)-1 deficiency because of a PC1 gene mutation causes hypogonadotropic hypogonadism, obesity, and hypocortisolemia. Although PC1 regulates posttranslational modification of prohormones and neuropeptides, it has not been elucidated why PC1 deficiency causes hypogonadotropic hypogonadism. Individuals with congenital leptin deficiency or congenital leptin resistance because of LEP or LEPR gene mutation show hypogonadotropic hypogonadism associated with obesity. Male patients with X-linked adrenal hypoplasia congenita because of hemizygous DAX1 mutation exhibit primary adrenal insufficiency in infancy or early childhood and hypogonadotropic hypogonadism. Some studies have revealed that hypogonadotropic hypogonadism in DAX1 mutation represents defects at the pituitary gonadotrophs as well as the hypothalamus, consistent with the expression of DAX1 at both sites. No fertile male patient with hemizygous DAX1 mutation has been described following GnRH or gonadotropin treatment, suggesting critical roles of DAX1 for the process of spermatogenesis. Interestingly, delayed puberty has been described in one female with a heterozygous DAX1 mutation in one family, and hypogonadotropic hypogonadism without adrenal insufficiency has been

461

described in one female with a homozygous DAX1 mutation. Analysis of DAX1 gene in more than 100 patients with “idiopathic” hypogonadotropic hypogonadism failed to identify any mutations. Patients with congenital combined pituitary hormone deficiency can have hypogonadotropic hypogonadism when the mutated genes regulate gonadotroph differentiation. Two children with homozygous HESX1 mutation with septo-optic dysplasia and congenital combined pituitary hormone deficiency have impaired gonadotropin secretion, although these children are prepubertal. Homozygous LHX3 mutations have been reported in patients with congenital combined pituitary hormone deficiency including hypogonadotropic hypogonadism. These patients have normal ACTH secretion and limited head rotation because of cervical spine rigidity. Heterozygous LHX4 mutation has been reported in familial combined pituitary hormone deficiency including hypogonadotropic hypogonadism. Patients with PROP1 mutation may develop hypogonadotropic hypogonadism, although the onset of hypogonadism varies. These patients have GH and thyroid-stimulating hormone deficiency in childhood and may develop progressive ACTH deficiency. Panhypopituitarism with an invisible or thin pituitary stalk can produce hypogonadotropic hypogonadism. These patients usually have a history of complicated delivery as breech presentation or neonatal asphyxia. The invisible or thin pituitary stalk reflects its injury (transection) at birth or the primary abnormality of the stalk. Hypogonadotropic hypogonadism as well as GH, thyroid-stimulating hormone, and ACTH deficiency may gradually develop. CNS disorders may cause hypogonadotropic hypogonadism by impairing hypothalamic or pituitary function. Sellar or suprasellar tumors (e.g., craniopharyngioma) commonly disturb the processes of pubertal development, causing either precocious puberty (as described earlier) or delayed puberty. Such tumors are often associated with other kinds of pituitary defects, of which GH deficiency is by far the most common. Anorexia nervosa is notorious for causing functional gonadotropin deficiency in adolescent females. The prevalence of anorexia nervosa is definitely increasing in the United States and other industrialized countries. Anorexia nervosa is characterized by severe weight loss, a distorted body image, obsessive fear of obesity, and multiple endocrine abnormalities including functional gonadotropin deficiency. The recovery of most of endocrine abnormalities after weight gain suggests that endocrine abnormalities are secondary to weight loss. However, amenorrhea or infertility may persist for months to years after weight gain. Higher brain dysfunction in anorexia nervosa may hamper the complete recovery of functional gonadotropin deficiency. Increasing evidence shows the link between excessive exercise and functional gonadotropin deficiency, especially in adolescent females. Long-distance runners, swimmers, ballet dancers, figure skaters, and gymnasts are well-known athletes who may develop amenorrhea. Excessive exercise itself or strict weight control can cause functional gonadotropin deficiency. Hyperprolactinemia may cause functional gonadotropin deficiency. Hyperprolactinemia is because of pituitary prolactinoma, hypothyroidism, or drugs such as neuroleptics, antihypertensives, dopamine receptor antagonists, and antidepressants. Galactorrhea may or may not be present. The mechanism by which hyperprolactinemia causes functional gonadotropin deficiency is most likely multifactorial. Increased prolactin may inhibit GnRH secretion and/or decrease responsiveness of gonadotrophs to GnRH.

462

SECTION IV / ENDOCRINOLOGY

Treatment The goals of treatment of hypogonadotropic hypogonadism are to induce and maintain puberty and establish fertility. Gonadal steroid hormone replacement can induce and maintain puberty, testosterone in boys and estrogen in girls, when patients are at pubertal ages (about 13 yr in boys and 12–13 yr in girls). Gonadal steroid hormones should be gradually increased to an adult dosage to mimic normal pubertal development. In girls, cyclic therapy using estrogen and progesterone is necessary to establish regular withdrawal bleeding. Pulsatile GnRH treatment or the combination of gonadotropins is effective both to induce and maintain puberty and to establish fertility. It is uncertain whether prolonged gonadal steroid hormone replacement in male hypogonadotropic hypogonadism causes irreversible damage to spermatogenesis before the induction of the treatment to establish fertility. Treatment of underlying diseases should focus on CNS disorders and functional gonadotropin deficiency. Kallmann’s Syndrome Kallmann’s syndrome is characterized by the association of hypogonadotropic hypogonadism and anosmia (or hyposmia). Hypogonadotropic hypogonadism is hypothalamic in origin, and anosmia (or hyposmia) results from agenesis or hypoplasia of the olfactory bulbs and tracts (see Etiology). Three modes of transmission have been described: X-linked (KAL1), autosomaldominant (KAL2), and autosomal-recessive (KAL3). The prevalence of Kallmann’s syndrome in boys is four times that in girls. Clinical Features Typical clinical features of Kallmann’s syndrome are delayed puberty and no or reduced sense of smell (often unrecognized by the patient). Boys with Kallmann’s syndrome may have micropenis and/or bilateral or unilateral cryptorchidism in infancy. Both genders fail to develop puberty. Patients may have other clinical features depending on the mode of transmission. Renal agenesis is observed in up to 40% of KAL1, but not in KAL2. Synkinesia (mirror image movements) is associated principally with KAL1, but has also been observed in KAL2. Cleft lip/palate is present mostly in association with KAL2, but may also be seen in KAL1. MRI can clearly demonstrate the absence of olfactory bulbs and hypoplasia of the olfactory sulci, even in newborn infants. Olfactometry or the intravenous olfactory stimulation test is helpful to identify or confirm anosmia or hyposmia. In contrast to the delayed puberty caused by CNS tumors or constitutional delay of puberty, patients with Kallmann’s syndrome usually have appropriate or tall stature for their age. Untreated adults and individuals of pubertal age commonly have eunuchoid proportions. Some subjects with KAL1 or fibroblast growth factor receptor (FGFR)-1 (see below) mutations manifest varying combinations of these features or do not have any of the clinical features. In the reported KAL2 families with FGFR-1 mutation, the prognosis of fertility was not compromised without treatment or by the combination of gonadotropins. Etiology During embryogenesis, olfactory neurons and GnRH-secreting neurons originate in the olfactory placode in the nose and migrate into the CNS. The olfactory nerves associate with the terminal nerve and vomeronasal nerve to produce a bridge between the olfactory epithelium and the forebrain. The cells that will become GnRH-secreting neurons arise within the region of the olfactory placodes and migrate from the nasal epithelium, through the cribriform plate of the nose and then along the olfactory tract–forebrain axis to reach the preoptic and hypothalamic areas, where they differentiate to become the GnRH-secreting neurons. Given the developmental connection between GnRHsecreting and olfactory neurons, abnormalities in this developmental stage can cause Kallmann’s syndrome.

An X-linked form of Kallmann’s syndrome, KAL1, is caused by the deletion of or mutation of KAL1, which is located at Xp22.3. Examination of the expression of the KAL1 gene during embryogenesis demonstrates that it can be detected in various neuronal populations of the CNS, including cells of the olfactory bulbs. This gene partially escapes X inactivation and encodes a 680 amino acid protein, anosmin-1, an extracellular matrix component. An inactive homologous pseudogene, KALP, is located on the Y chromosome at Yq11. The derived amino acid sequence of KAL1 predicts a protein that contains a leader peptide with protease inhibitor domain that is followed by 4 fibronectin type III repeats; no transmembrane or anchoring regions are present. The protein, anosmin-1, is N-glycosylated, secreted in the cell culture medium, and localized at the cell surface. Several lines of evidence indicated that heparan-sulfate chains of proteoglycans are involved in the anosmin-1 to the cell surface. Anosmin-1 is thought to have a dual branch-promoting and guidance activity and is involved in the patterning of mitral and tufted cell axon collaterals to the olfactory cortex. Anosmin-1 may bind by means of a heparan sulfate proteoglycan to its cognate receptor or by other extracellular cues to induce axonal branching and axon misrouting. The role of KAL1 in events such as kidney formation and migration is obscure, although the expressed KAL1 in the Wolffian duct might be involved in renal development by interaction with the metanephric mesenchyme. Heterozygous loss of function mutations in FGFR-1 have been identified in an autosomal dominant form of Kallmann’s syndrome (KAL2). FGFR-1 is located at 8p11.2-p12 and encodes the FGFR-1; and its gain of function has been shown to cause craniosynostosis. Several lines of evidence suggest that KAL1 and FGFR-1 interact functionally to effect the normal migration of GnRH-secreting and olfactory neurons. Anomsin-1 may directly participate in FGF signaling through interactions with the FGFR-1. Notably, nonpenetrance of the disease in some mutation carriers can simulate autosomal-recessive transmission. An autosomal-recessive form of Kallmann’s syndrome (KAL3) also exists. The responsible gene for KAL3 has not been identified. Constitutional Delay of Puberty (Variation of Normal Puberty) Constitutional delay of puberty is a common, benign condition that represents a variant of normal puberty. The time of onset of puberty is delayed compared with the general population, although the pattern of pubertal development is normal. Boys are referred for evaluation of this condition significantly more often than girls. Together with their delayed pubertal development, these children often have short stature, approximately two to three standard deviations below the mean, called constitutional delay of growth and puberty. The bone age is usually delayed 2–4 yr behind chronological age. These children demonstrate pubertal development that is more commensurate with their bone age than their chronological ages. Adult height and complete pubertal maturation are achieved significantly later, some young men reporting continued linear growth in their late teens or early twenties. Adult height is generally appropriate for the genetic background, but is commonly in the low normal range. The family history often reveals other individuals with suspected constitutional delay of puberty, most often in males. Constitutional delay of puberty usually does not require treatment. Hypergonadotropic Hypogonadism Hypergonadotropic (or primary) hypogonadism is failure of the gonads to respond to hypothalamic–pituitary stimulation. Hypergonadotropic hypogonadism is associated with elevated serum concentration of

CHAPTER 43 / DISORDERS OF PUBERTAL DEVELOPMENT

Table 43-7 Causes of Hypergonadotropic Hypogonadism Chromosome abnormalities Klinefelter syndrome Turner syndrome Other sex chromosome abnormalities Autosomal chromosome abnormalities Isolated gonadal dysgenesis Perrault syndrome Others Gonadotropin resistance LH/CG resistance because of LHCGR mutation FSH resistance because of FSH resistance mutation LH and FSH resistance because of GNAS1 mutation (pseudohypoparathyroidism, type 1A) Others Enzymatic defects in gonadal steroid hormone biosynthesis Syndromes associated with hypergonadotropic hypogonadism Gonadal destruction Vanishing Trauma Torsion Autoimmunity Surgery Chemotherapy Infections Radiation Others Exposure to environmental estrogen in utero (boys)a Unknown Galactosemia Others aProposed

but not proved.

gonadotropins owing to the absence of the negative feedback effects of gonadal steroid hormones. Causes The most common causes of hypergonadotropic hypogonadism are chromosome abnormalities such as Klinefelter syndrome, Turner syndrome and Down syndrome (Table 43-7). Why hypergonadotropic hypogonadism is common in chromosomal abnormalities is not completely understood, although pairing failure of homologous chromosomes in meiocytes is proposed to cause germ cell loss. Klinefelter syndrome and Turner syndrome are discussed later. An uncommon form of hypergonadotropic hypogonadism is isolated gonadal dysgenesis. This condition appears to be genetically heterogeneous and both sporadic and familial cases have been reported. Affected patients having female external genitalia, either with 46,XX or 46,XY, typically present with no pubertal development. They do not have any other abnormalities such as short stature and minor anomalies. One familial variant of isolated gonadal dysgenesis is Perrault syndrome, gonadal dysgenesis in 46,XX with sensorineural hearing loss. This syndrome is autosomal recessive with obligatory ovarian dysgenesis in 46,XX homozygotes and facultative sensorineural hearing loss in 46,XX and 46,XY homozygotes. The molecular defect has not been identified. The receptors for the gonadotropins LH/CG and FSH are both members of the GPCR family. Selective defects in the gonadal

463

response to gonadotropins have been traced in several pedigrees to mutations of the genes encoding the LH/CG and FSH receptors. LH/CG resistance is resulting from loss of function mutations in the LHCGR gene. Male patients are classically associated with a form of male pseudohermaphroditism termed Leydig-cell hypoplasia, rather than being associated with delayed puberty. This disorder is characterized by female phenotype in the presence of a 46,XY karyotype, low serum concentration of testosterone, increased LH, and lack of testosterone secretion in response to a hCG stimulation test. Homozygous or compound heterozygous loss of function mutations in the LHCGR gene has been identified. The identification of a missense mutation of the LHCGR gene in a phenotypic male infant evaluated for a small but normally formed penis suggests that the range of altered phenotypes associated with LH/CG resistance because of loss of function mutations in the LHCGR gene may be broader than initially identified. A 46,XX subject with homozygous K354E was a phenotypically normal adult female with primary amenorrhea and cystic ovaries, suggesting that female patients with LH/CG resistance had some pubertal development, but impairment in the normal ovarian cycle. Follicle-stimulating hormone resistance is caused by a loss of function mutation in the FSHR gene. A homozygous loss of function missense mutation in FSHR gene was found in Finnish female patients associated with 46,XX hypergonadotropic ovarian dysgenesis, which was originally classified as isolated gonadal dysgenesis. The disorder is common in this population (1:8300 females) and is inherited in an autosomal-recessive manner. Males homozygous for this mutation had a normal phenotype with variable degrees of spermatogenetic failure, but surprisingly, did not show azoospermia or absolute infertility. LH and FSH resistance can be because of GNAS1 mutations (pseudohypoparathyroidism, type 1A [PHP1A]). GNAS1 encodes Gsα and displays imprinting (see precocious puberty, MAS in this chapter). Maternally inherited loss of function heterozygous mutation in GNAS1 causes PHP1A. PHP1A can be associated with resistance to multiple hormones, including hypergonadotropic hypogonadism. Patients with pseudohypoparathyroidism, type 1B (PHP1B) exhibit parathyroid hormone resistance, typically without other endocrine abnormalities. PHP1B is most likely caused by mutations in regulatory regions of the GNAS1 gene inherited from the mother that are predicted to interfere with the parent-specific methylation of GNAS1. Other causes include enzymatic defects in gonadal steroid hormone biosynthesis, syndromes associated with hypergonadotropic hypogonadism, gonadal destruction, and unknown etiologies (see Table 43-7). Treatment The goal of treatment of hypergonadotropic hypogonadism is to induce and maintain puberty. Gonadal steroid hormone replacement (see Treatment of Hypogonadotropic Hypogonadism) can induce and maintain puberty. Gonadal steroid hormone replacement therapy in Turner syndrome is more complex, requiring coordination of timing with respect to GH therapy to optimize adult height. Unfortunately, other than sperm or egg donation, treatment to establish fertility in any hypergonadotropic hypogonadism is unsatisfactory. Some successful pregnancies have been reported, however, by testicular sperm extraction/microsurgical epididymal sperm aspiration combined with in vitro fertilization/intracytoplasmic sperm injection. Klinefelter Syndrome The most common form of male hypergonadotropic hypogonadism is Klinefelter syndrome

464

SECTION IV / ENDOCRINOLOGY

(47,XXY karyotype), which occurs with a prevalence of 1:2500 adult males. Patients with Klinefelter syndrome may demonstrate a slowing or arrest of pubertal development as gonadal function declines after the age of puberty, although gonadal function remains relatively normal until that age. The hallmark of physical findings is small testes, and infertility is the rule. Subjects are relatively tall owing to the additional copy of the X-linked SHOX gene. Mean full-scale I.Q. is between 85 and 90. When a patient with Klinefelter syndrome develops precocious puberty, germinoma of the mediastinum is highly likely. Turner Syndrome The most common form of female hypergonadotropic hypogonadism is Turner syndrome (45,X karyotype, typically), which occurs with an incidence of 1 in 2000 live-born females. Turner syndrome is a well-defined X chromosome abnormality characterized by short stature, Turner somatic stigmata, and hypergonadotropic hypogonadism. With respect to hypergonadotropic hypogonadism, adult patients usually have “streak” gonads consisting of fibrous tissue without germ cells. However, germ cells may be present in gonads in fetal life and infancy. Short stature is primarily ascribed to SHOX haploinsufficiency in addition to nonspecific growth disadvantage caused by chromosome imbalance. Turner somatic stigmata, either surface (webbed neck, high arched palate, ptosis, and so on) or visceral (coarctation of the aorta, horseshoe kidney, and so on), may be because of haploinsufficiency of postulated “lymphogenic gene(s)” on the short arm of X chromosome and nonspecific developmental disadvantage caused by chromosome imbalance. GH treatment improves growth rates as well as adult height. Recommendations for diagnosis, treatment, and management of Turner syndrome have been published by the fifth international symposium on Turner syndrome.

SELECTED REFERENCES Achermann JC, Gu WX, Kotlar TJ, et al. Mutational analysis of DAX1 in patients with hypogonadotropic hypogonadism or pubertal delay. J Clin Endocrinol Metab 1999;84:4497–4500. Ahima RS, Dushay J, Flier SN, Prabakaran D, Flier JS. Leptin accelerates the onset of puberty in normal female mice. J Clin Invest 1997;99: 391–395. Aittomäki K, Lucena JLD, Pakarinen P, et al. Mutation in the folliclestimulating hormone receptor gene causes hereditary hypergonadotropic ovarian failure. Cell 1995;82:959–968 Balducci R, Boscherini B, Mangiantini A, Morellini, Toscano V. Isolated precocious pubarche: an approach. J Clin Endocrinol Metab 1994;79:582–589. Beranova M, Oliveira LM, Bedecarrats GY, et al. Prevalence, phenotypic spectrum, and modes of inheritance of gonadotropin-releasing hormone receptor mutations in idiopathic hypogonadotropic hypogonadism. J Clin Endocrinol Metab 2001;86:1580–1588. Chehab FF, Mounzih K, Lu R, Lim ME. Early onset of reproductive function in normal female mice treated with leptin. Science 1997;275: 88–90. Clement K, Vaisse C, Lahlou N, et al. A mutation in the human leptin receptor gene causes obesity and pituitary dysfunction. Nature 1998; 392:398–401. Counts DR, Pescovitz OH, Barnes MK, et al. Dissociation of adrenarche and gonadarche in precocious puberty and in isolated hypogonadotropic hypogonadism. J Clin Endocrinol Metab 1987;64:1174–1178. Dattani MT, Martinez-Barbera JP, Thomas PQ, et al. Mutations in the homeobox gene HESX1/Hesx1 associated with aepto-optic dysplasia in human and nouse. Nat Genet 1998;19:125–133. De Roux N, Genin E, Carel J-C, Matsuda F, Chaussain J-L, Milgrom E. Hypogonadotropic hypogonadism due to loss of function of the KiSS1-derived peptide receptor GPR54. Proc Natl Acad Sci USA 2003;100:10,972–10,976.

De Roux N, Young J, Misrahi M, et al. A family with hypogonadotropic hypogonadism and mutations in the gonadotropin-releasing hormone receptor. N Engl J Med 1997;337:1597–1602. Dode C, Levilliers J, Dupont J-M, et al. Loss-of-function mutations in FGFR1 cause autosomal dominant Kallmann syndrome. Nat Genet 2003;33:463–465. Dunkel L, Perheentupa J, Virtanen M, et al. Gonadotropin-releasing hormone test and human chronic gonadotropin test in the diagnosis of gonadotropin deficiency in prepubertal boys. J Pediatr 1985;107:388–392. Farooqi IS, Jebb SA, Langmack G, et al. Effects of recombinant leptin therapy in a child with congenital leptin deficiency. N Engl J Med 1999;341:879–884. Feuillan PP, Foster CM, Pescovitz OH, et al. Treatment of precocious puberty in the McCune-Albright syndrome with the aromatase inhibitor testolactone. N Engl J Med 1986;315:1115–1119. Fokstuen S, Ginsburg C, Zachmann M, Schinzel A. Maternal uniparental disomy 14 as a cause of intrauterine growth retardation and early onset of puberty. J Pediatr 1999;134:689–695. Heger S, Partsch CJ, Peter M, Blum WF, Kiess W, Sippell WG. Serum leptin levels in patients with progressive central precocious puberty. Pediatr Res 1999;46:71–75. Heger S, Partsch CJ, Sippell WG. Long-term outcome after depot gonadotropin-releasing agonist treatment of central precocious puberty: Final height, body proportion, bone mineral density, and reproductive function. J Clin Endocrinol Metab 1999;84:4583–4590. Hemminki A, Markie D, Tomlinson I, et al. A serine/threonine kinase gene defective in Peutz-Jeghers syndrome. Nature 1998;391:184–187. Jackson RS, Creemers JW, Ohagi S, et al. Obesity and impaired prohormone processing associated with mutations on the human proormone convertase 1 gene. Nat Genet 1997;16:303–306. Jenne DE, Reimann H, Nezu J, et al. Peutz-Jeghers syndrome is caused by mutations in a novel serine threonine kinase. Nat Genet 1988;18:38–43. Kaplowitz PB, Oberfield SE. Reexamination of the age limit for defining when puberty is precocious in girls in the United States: implications for evaluation and treatment. Drug and Therapeutics and Executive Committees of the Lawson Wilkins Pediatric Endocrine Society. Pediatrics 1999;104:936–941. Kawate N, Kletter GB, Wilson BE, Netzloff ML, Menon KMJ. Identification of constitutively activating mutation of the luteinizing hormone receptor in a family with male limited gonadotrophin independent precocious puberty (testotoxicosis). J Med Genet 1995;32: 553, 554. Kosugi S, Van Dop C, Geffner ME, et al. Characterization of heterogeneous mutations causing constitutive activation of the luteinizing hormone receptor in familial male precocious puberty. Hum Mol Genet 1995;4:183–188. Latronico AC, Anasti J, Arnhold IJP, et al. Testicular and ovarian resistance to luteinizing hormone caused by inactivating mutations of the luteinizing hormone. N Engl J Med 1996;334:507–512. Laue L, Chan WY, Hsueh AJ, et al. Genetic heterogeneity of constitutively activating mutations of the human luteinizing hormone receptor in familial male-limited precocious puberty. Proc Natl Acad Sci USA 1995;92:1906–1910. Laue L, Wu SM, Kudo M, et al. A nonsense mutation of the human luteinizing hormone receptor gene in Leydig cell hypoplasia. Hum Mol Genet 1995;4:1429–1433. Layman LC, Cohen DP, Jin M, et al. Mutations in gonadotoropin-releasing hormone receptor gene cause hypogonadotropic hypogonadism. Nat Genet 1998;18:14, 15. Laymen LC, Lee EJ, Peak DB, et al. Delayed puberty and hypogonadadism caused by mutations in the follicular-stimulating hormone β-subunit gene. New Engl J Med 1997;337:607–611. Lazar L, Kauli R, Pertzelan A, Phillip M. Gonadotropin-suppressive therapy in girls with early and fast puberty affects the pace of puberty but not total pubertal growth or final height. J Clin Endocrinol Metab 2002;87:2090–2094. Levine MA, Downs RW Jr, Moses AM, et al. Resistance to multiple hormones in patients with pseudohypoparathyroidism; association with deficient activity of guanine nucleotide regulatory protein. Am J Med 1983;74:545–556.

CHAPTER 43 / DISORDERS OF PUBERTAL DEVELOPMENT

Liu G, Duranteau L, Carel J-C, Monroe J, Doyle DA, Shenker AS. Leydigcell tumors caused by an activating mutation of the gene encoding the luteining hormone receptor. New Eng J Med 1999;341:1731–1736. Liu J, Litman D, Rosenberg MJ, Yu S, Biesecker LG, Weinstein LS. A GNAS1 imprinting defect in pseudohypoparathyroidism type 1B. J Clin Invest 2000;106:1167–1174. Mantzoros CS, Flier JS, Rogol AD. A longitudinal assessment of hormonal and physical alterations during normal puberty in boys. V. Rising leptin levels may signal the onset of puberty. J Clin Endocrinol Metab 1997; 82:1066–1070. Matthews CH, Borgato S, Beck-Peccoz P, et al. Primary amenorrhoea and infertility due to a mutation in the β-subunit of follicle-stimulating hormone. Nat Genet 1993;5:83–86. Merke DP, Tajima T, Baron J, Cutler GBJ. Hypogonadotropic hypogonadism in a female caused by an X-linked recessive mutation in the DAX1 gene. N Engl J Med 1999;340:1248–1252. Nachtigall LB, Boepple PA, Pralong FP, Crowley WF Jr. Adult-onset idiopathic hypogonadotropic hypogonadism—a treatable form of male infertility. N Engl J Med 1997;336:410–415. Neely EK, Bachrach LK, Hintz RL, et al. Bone mineral density during treatment of central precocious puberty. J Pediatr 1995;127:819–822. Netchine I, Sobrier ML, Krude H, et al. Mutaions in LHX3 results in a new syndrome revealed by combined pituitary hormone deficiency. Nat Genet 2000;25:182–186. Nishi Y, Hamamoto K, Kajiyama M, Kawamura I. The Perrault syndrome: clinical report and review. Am J Med Genet 1988;31:623–629. Ogilvy-Stuart AL, Clayton PE, Shalet SM. Cranial irradiation and early puberty. J Clin Endocrinol Metab 1994;78:1282–1286. Ozata M, Ozdemir IC, Licinio J. Human leptin deficiency caused by a missense mutation: multiple endocrine defects, decreased sympathetic tone, and immune system dysfunction indicate new targets for leptin action, greater central than peripheral resistance to the effects of leptin, and spontaneous correction of leptin-mediated defects. J Clin Endocrinol Metab 1999;84:3686–3695. Patten JL, John DR, Valle D, et al. Mutation in the gene encoding the stimulatory protein of adenylate cyclase in Albright’s hereditary osteodystrophy. New Engl J Med 1990;322:1412–1419. Paul D, Conte FA, Grumbach MM, Kaplan SL. Long term effect of gonadotropin-releasing hormone agonists therapy on final and near-final height in 26 children with true precocious puberty treated at a median age of less than 5 years. J Clin Endocrinol Metab 1995;80:546–551. Pescovitz OH, Hench KD, Barnes KM, Loriaux DL, Cutler GB Jr. Premature thelarche and central precocious puberty: the relationship between clin-

465

ical presentation and gonadotropin response to luteinizing hormonereleasing hormone. J Clin Endocrinol Metab 1988;67:474–479. Phillip M, Arbelle JE, Segev Y, Parvari R. Male hypogpnadism due to a mutation in the gene for the b-subunit of follicular-stimulating hormone. N Engl J Med 1998;338:1729–1732. Rao E, Weiss B, Fukami M, et al. Pseudoautosomal deletions encompassing a novel homebox gene cause growth failure in idiopathic short stature and Turner syndrome. Nat Genet 1997;16:54–63. Saenger P, Wikland KA, Conway GS, et al. Fifth international symposium on Turner syndrome. Recommendations for the diagnosis and management of Turner syndrome. J Clin Endocrinol Metab 2001;86:3061–3069. Schwindinger WF, Francomano CA, Levine MA. Identification of a mutation in gene encoding the subunit of the stimulatory G protein of adenylyl cyclase in McCune-Albright syndrome. Proc Natl Acad Sci USA 1992;89:5152–5156. Seminara SB, Seminara SB, Messager S, et al. The GPR54 gene as a regulator of puberty. New Engl J Med 2003:349;1614–1627. Shenker A, Laue L, Kosui S, Merendino JJ Jr, Minegishi T, Cutler GB Jr. A constitutively activating mutation of the luteinizing hormone receptor in familial male precocious puberty. Nature 1993;365:652–654. Smith EP, Boyd J, Frank GR, et al. Estrogen resistance caused by a mutation in the estrogen-receptor gene in a man. N Engl J Med 1994;331: 1056–1061. Tapanainen JS, Aittomaki K, Min J, Vaskivuo T, Huhtaniemi IT. Men homozygous for an inactivating mutation of the follicle-stimulating hormone (FSH) receptor gene present variable suppression of spermatogenesis and fertility. Nat Genet 1997;15:205, 206. Tauber M, Berro B, Delagnes V, et al. Can some growth hormone (GH)deficient children benefit from combined therapy with gonadotropinreleasing hormone analogs and GH? Results of a retrospective study. J Clin Endocrinol Metab 2003;88:1179–1183. Weinstein LS, Shenker A, Gejm an PV, Merino MJ, Friedman E, Spiegel AM. Activating mutations of the stimulatory G protein in the McCuneAlbright syndrome. N Engl J Med 1991;325:1688–1695. Weiss J, Axelrod L, Whitcomb RW, Harris PE, Crowley WF, Jameson JL. Hypogonadism caused by a single amino acid substitution in the β subunit of luteinizing hormone. N Engl J Med 1992;326:179–183. Wu T, Mendola P, Buck GM. Thinic differences in the presence of secondary sex characteristics and menarche among US girls: The Third National Health and Nutrition Exanimation Survey, 1988–1994. Pediatrics 2002;110:752–757. Wu W, Cogan JD, Pfaffle RW, et al. Mutations in PROP1 cause familial combined pituitary hormone deficiency. Nat Genet 1998;18:147–149.

44 Defects of Androgen Action MICHAEL J. MCPHAUL SUMMARY The development of the male phenotype is a complex process that involves the active participation of genes involved at many levels, from those specifying gonadal differentiation to the androgen receptor (AR) itself. Defective virilization can be caused by defects anywhere along the pathway. Considering the clinical syndromes caused by known defects of genes such as steroid 5α-reductase II and the AR, the pathogenesis of a large proportion of defects in virilization remains unexplained. It is conceivable that some might represent defects in genes required for normal AR function or at steps beyond the site of action of the AR itself (e.g., coactivators, or defects in genes activated by the AR). Key Words: Amino; androgen receptor (AR); 5α-dihydrotestosterone; 5α-reductase; Reifenstein’s syndrome; sex-determining region Y; testosterone.

INTRODUCTION Since 1935 it has been recognized that the principal androgen secreted by the testes is testosterone. Although testosterone is the most abundant circulating androgen, 5α-dihydrotestosterone is the predominant hormone found complexed to the androgen receptor (AR) in the nuclei of target cells, such as the prostate. This finding opened a new perspective on androgen physiology and focused attention on the 5α-reductase enzyme(s) that catalyze this conversion as potential modulators of androgen action in selected tissues. As described later, this inference has been confirmed by the recognition that some abnormalities of male sexual development can be traced to defects in the conversion of testosterone to 5α-dihydrotestosterone.

STRUCTURE AND FUNCTION OF THE AR Abundant biochemical and genetic data demonstrated that the actions of androgen were mediated by a single receptor that was encoded on the X-chromosome, and that defects in this receptor protein, the AR, could result in a range of abnormalities of male phenotypic sexual development. The isolation of cDNAs encoding the AR revealed it to be a member of a large group of related transcription factors, the nuclear receptor family. This family includes members that are ligand responsive, such as the steroid, thyroid hormone, and retinoid receptors, and others that are thought to be constitutively active or modulated by other influences, such as From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

phosphorylation. All exhibit a modular structure (displayed for the AR in Fig. 44-1) consisting of a highly conserved DNA-binding domain, a less highly conserved carboxyl-terminal ligand-binding domain, and an amino-terminal segment that is poorly conserved between individual family members, both in terms of primary amino acid sequence, and length. The DNA-binding domain is made up of two elements (termed “zinc fingers”) that mediate the sequence-specific DNA binding of the AR. This segment is the most highly conserved region between members of this gene family. The carboxy-terminus is approx 250 amino acids long and encodes the portion of the protein that binds androgens with high affinity. The amino-terminus of the AR is somewhat atypical in that it contains three segments made up of repeated amino acids (one of repeated glutamine residues, one of repeated proline residues, and one of repeated glycine residues). Polymorphisms of these regions appear to have little effect on AR function in normal individuals, but expansions of the glutamine repeat have been implicated in the pathogenesis of spinal and bulbar muscular atrophy (SBMA, Kennedy’s syndrome, see AR Mutations: Special Cases). The nonligand-bound AR is thought to exist in the cell in association with a number of ancillary proteins, particularly members of the heat shock protein family. After the binding of hormone, the receptor dissociates from these proteins and binds to specific DNA sequences within or adjacent to androgen-responsive genes. This ligand-activated receptor interacts with components of the transcriptional apparatus to stimulate and stabilize active transcription complexes. In some models, the AR appears to modulate genes in a negative fashion or to alter mRNA stability, but these phenomena have been less well characterized.

AR DEFECTS CLINICAL FEATURES A spectrum of phenotypes can result from defects of AR function (Table 44-1). Patients completely unresponsive to the actions of androgen (referred to as complete testicular feminization or complete androgen insensitivity) have a 46,XY karyotype, but an external phenotype that is completely female in appearance, despite normal or elevated levels of circulating androgens. Owing to the secretion and action of Müllerian inhibitory substance by the functional testes present in these patients, the uterus and fallopian tubes are absent. Such individuals are usually raised as females and might first seek attention for evaluation of primary amenorrhea. Gonadectomy is often performed, because the intraabdominal testes show an increased rate

466

467

CHAPTER 44 / DEFECTS OF ANDROGEN ACTION

Figure 44-1 Mutations of the human AR that cause abnormalities of AR function (above). A schematic representation of the human AR is shown. The approximate boundaries of the DNA and LBDs are indicated, as are the locations of the repeated stretches of glutamine, proline, and glycine residues within the amino terminus (below). A selected grouping of mutations causing androgen resistance are grouped according to the type of genetic lesion (amino acid substitution or premature termination codon, left margin) or the effect that the mutation has receptor function (right margin). Also comprised in this figure are a selected number of mutations reported in literature that have been detected in prostate cancer specimens that appear to alter the binding characteristics and activities of the AR. The expansion of the glutamine repeat that causes spinal bulbar muscular atrophy is also represented. (Adapted from McPhaul MJ, Marcelli M, Zoppi S, Griffin JE, Wilson JD. Genetic basis of endocrine disease. J Clin Endocrinol Metab 1993;76[1]:17–23.) Table 44-1 Phenotypes Associated With Defects in the Genes Encoding the Androgen Receptor and 5α-Reductase II Gene AR

Syndrome Complete female feminization Complete androgen insensitivity Incomplete testicular feminization Reifenstein syndrome

5α-Reductase II

Under-virilized, fertile male Infertile male 5α-Reductase deficiency

External genitalia

Wolffian duct derivatives

Müllerian duct derivatives

Completely female

Not virilized

Absent

Predominantly female; some clitoromegaly, closure of the posterior fourchette Substantial phallic development, but with severe forms of hypospadias Male phenotype with small phallus Male phenotype Can be variable: ranging from female external genitalia at birth to a predominantly male phenotype with hypospadias

Usually partially virilized Variable defects

Absent

Male Male Normal male

Absent Absent Absent

Absent

468

SECTION IV / ENDOCRINOLOGY

of malignant tumor development. With estrogen replacement, these individuals often lead completely normal lives as women, although they are infertile. Partial androgen insensitivity encompasses a wide range of abnormalities of male phenotypic development. Affected individuals have phenotypes ranging from predominately female to minor defects of male development. The term incomplete testicular feminization has been applied to individuals with nearly complete forms of androgen resistance who demonstrate only slight evidence of virilization (such as clitoromegaly). These patients are usually managed in a fashion similar to that of patients with complete testicular feminization. Reifenstein’s syndrome is a constellation of features that includes severe defects of male urogenital development (perineal or penoscrotal hypospadias) and gynecomastia. Reifenstein’s syndrome represents a far more difficult challenge. As noted clinically, the developmental abnormalities are substantial and surgical correction of these defects often requires multiple separate surgical procedures. Such efforts are further hampered by the small size of the genitalia of affected children. After surgical correction, most individuals with this phenotype are raised as males, although many exhibit difficulty with gender identity. At the mildly affected end of the spectrum, a small number of individuals have been described in which normal development of the external genitalia has occurred, but in which some degree of undervirilization is clinically evident. In some, this phenotype has been associated with azoospermia and infertility without undervirilization. In others, normal sperm density and fertility is identified in men with varying degrees of virilization. Patients with these phenotypes have been identified as having AR defects on the basis of a family history and abnormalities of in vitro ligand-binding assays of the AR. Although specific AR defects have been reported in association with these syndromes, it is not clear how frequently such phenotypes are caused by AR defects in the general population. Information derived from the analysis of groups of individuals with infertility would suggest that AR mutations can be identified in only a small subset of such patients. An unusual variation of the undervirilized phenotype is that presented by patients with spinal and bulbar muscular atrophy (SBMA) (see Chapter 116). Patients with this syndrome have normal male development and normal male secondary sexual characteristics throughout much of their lives but develop signs of clinical androgen resistance beginning in middle age. Although these signs of androgen resistance are unmistakable (usually gynecomastia), the difficulties presented by the symptoms of anterior motor neuron degeneration pose a far more serious threat to the health of such patients and represent the usual cause of death in affected individuals. Interestingly, expansions of this same element have been identified in a subset of patients with infertility. AR MUTATIONS AND DEFECTIVE AR FUNCTION Molecular defects of the AR that cause the syndromes of androgen resistance have been studied by a number of groups. The causative mutations have been identified in more than 100 pedigrees and include all of the major clinical syndromes. A database listing the mutations causing androgen resistance is accessible through the internet. Deletions or insertions of the AR gene occur with a frequency of approx 5–10% of patients with androgen resistance and range in size from single or multiple nucleotides to deletions of the entire gene. Because such mutations disturb the open reading frame of the AR, patients with this type of mutation do not express an intact receptor protein and, with few exceptions, lack detectable androgen

binding in their cells, and tissues. In addition, because the intact AR is not present in the tissues of such patients, AR function is absent and affected individuals always demonstrate the phenotype of complete androgen insensitivity (complete testicular feminization). In contrast to the frequency of deletions and insertions in the AR gene, single nucleotide substitutions are much more frequent and account for the bulk of patients with androgen resistance as a result of AR mutations. In some cases, such substitutions result in large-scale alterations of receptor structure, as when they result in alterations of AR mRNA splicing or introduction of premature termination codons within the AR open reading frame. These instances, as with gene deletion or insertion mutations that interrupt the primary sequence of intact AR protein, are usually associated with a lack of detectable androgen binding in tissues and a phenotype of complete androgen insensitivity. Single nucleotide substitutions that result in single amino acid changes within the AR protein are the most frequent type of mutation in the AR. These defects fall into two general categories: those within the DNA-binding domain and those that have been localized to the ligand-binding domain of the receptor. Unlike the other mutation categories previously described (deletions, insertions, premature termination), the effect of these substitutions on AR function can vary, and the entire spectrum of androgen-resistant phenotypes has been traced to single amino acid substitutions within the AR protein. In most instances, the principal effect of the amino acid substitution is on AR function and major effects on the level of AR abundance, as measured in patient fibroblasts, are uncommon. Amino acid substitutions in the DNA-binding domain have little effect on the binding of ligand by the mutant receptor. Despite the capacity of these mutant ARs to bind ligand normally, receptor function is reduced after ligand stimulation, compared with the activity of the normal AR. This reduced function is caused by a decreased capacity of the mutant AR to bind to target sequences within responsive genes. In this category of mutation, the degree of impairment of receptor binding to DNA appears to have a direct relationship to the degree that receptor function is reduced. Similar conclusions hold true for larger alterations of receptor structure occurring within the DNA-binding domain that maintain the reading frame of the receptor (e.g., deletions of single amino acid residues within the DNA-binding domain or the single reported instance caused by in-frame deletion of exon 3). Amino acid substitutions in the ligand-binding domain (LBD) can result in a variety of changes in the capacity of the AR to bind its ligand. In a surprisingly small proportion of cases, the amino acid replacement renders the receptor completely unable to bind ligand. Under these circumstances, it is presumed that as a result the alteration of the LBD structure the binding pocket is radically distorted. This conformational change blocks the capacity of the AR to bind its ligand and the mutant receptor is thus “locked” into an inactive conformation. This type of mutant AR is functionally equivalent to mutations that result in the synthesis of truncated forms of the AR, as even pharmacological doses of potent synthetic androgens are unable to restore receptor function in vivo or in functional assays performed in vitro. Far more frequently, however, amino acid substitutions in the LBD lead to the synthesis of mutant ARs that exhibit the capacity to bind hormone, but with properties that are abnormal compared to the normal AR (e.g., bind ligand with a reduced affinity or stability). Although the exact type of ligand-binding abnormality differs depending on the nature and location of the amino acid substitution within the ligand-binding domain, it appears that the formation and

CHAPTER 44 / DEFECTS OF ANDROGEN ACTION

469

Figure 44-2 Comparison of the structure of human steroid 5α reductases I and II and mutations that alter the binding of testosterone and nicotinamide adenine dinucleotide phosphate (above). Alignment of the predicted protein sequences of steroid 5α-reductase I (259 amino acids) and steroid 5α-reductase II (254 amino acids). The degree of sequence identity is shown for each of the five coding exons (boxed percentages) (below). Mutations identified within the coding sequence of the steroid 5α-reductase II gene in patients with 5α-reductase deficiency. Although more than 50 mutations been identified that cause 5α-reductase deficiency, a much smaller proportion have been analyzed biochemically. In this figure, only the mutations that cause alterations in the binding of testosterone or NADPH are indicated. Comprised in this figure is an amino acid substitution A49T that has been suggested to be associated with an increased risk of prostate cancer. (Drawn after Russell DW, Wilson JD. Steroid 5α-reductase: two genes/two enzymes. Annu Rev Biochem 1994;63:25–61.)

stability of the hormone-receptor complex is the final determinant of the degree of impairment of mutant AR function. In vitro studies of such mutant receptors demonstrate that the use of multiple high doses of physiological ligands (testosterone or dihydrotestosterone) or the use of potent nonmetabolizable androgen agonists can overcome the defective function of many mutant ARs of this type. In addition to its mechanistic implications, this finding might have substantial clinical implications as well, because it suggests that the pharmacological manipulation of many mutant ARs might be possible. Studies have investigated the levels of expression and the function of normal and mutant ARs. In most circumstances the major effect of most mutations in the AR is not at the level of AR abundance, but at the level of AR function. This is not true in all cases, however, with the most obvious exceptions being mutations that result in truncation of the receptor protein (e.g., termination codons or alterations of mRNA splicing).

GENOTYPE–PHENOTYPE CORRELATIONS AND PHENOTYPIC MODIFIERS The breadth of mutations that has been identified has permitted the recognition that the effect of an AR mutation depends on the extent to which AR expression and function are altered. In considering the large number of identified mutations, there is generally good agreement in terms of the phenotype observed when the same mutation is identified in different pedigrees. This agreement is clearest in patients with complete forms of androgen insensitivity and a greater degree of variation is observed in patients with partial forms of androgen insensitivity. This is likely because at the time of sexual development, relatively small variations in AR levels or activity can have discernible effects on the degree of observed virilization. Two different mechanisms have been reported as potential modifiers of the phenotype in different individuals with the same genotype. The first, differences in the level of 5α-reductase activity, contributes to differences between patients with distinctive partial phenotypes of androgen insensitivity. The second mechanism, somatic mosaicism, could account for even larger variations in observed phenotype.

AR MUTATIONS—SPECIAL CASES The androgen-resistance syndromes described are caused by mutations that impair receptor function to varying degrees. Two additional types of AR mutation have been identified that result in an alteration of AR responsiveness or an apparent “gain” of function. The first example of a “gain of function” mutation in the AR is the genetic defect causing Kennedy’s syndrome. This disease, also known as SBMA, is characterized by signs and symptoms caused by a progressive degeneration of spinal and bulbar motor muscles. These features are associated with clinical signs compatible with mild androgen insensitivity, which also appear in middle age. The pathogenesis of this extraordinary disease was traced to the expansion of a triplet nucleotide repeat (CAG) encoding a repeated sequence of glutamine residues within the amino-terminus of the receptor (see Fig. 44-1). The expansion of this glutamine repeat, the first of an increasing number of similar diseases that have also been traced to the expansion of trinucleotide repeats (Chapter 116), is thought to have two different effects on AR function. First, it is clear that the expansion of this segment of the AR diminishes the capacity of the AR to activate responsive genes after androgen stimulation. This diminished AR function can be demonstrated using a variety of functional assays and is presumably the cause of the subtle signs of androgen resistance observed in affected individuals. It is clear from the study of rare patients with complete deletions of the AR that the SBMA phenotype (progressive death of motor neurons in bulbar nuclei and in the spinal cord) is not caused by a simple lack of functional AR. Instead, this disease appears to be caused by a type of toxic “gain of function” that is caused by the expression of ARs containing the expanded glutamine repeats in specific cell types. Analogy to results obtained from the study of the Huntington disease gene (also caused by a glutamine repeat expansion) would suggest that the glutamine expansion might permit interaction of the mutant AR with intracellular targets in spinal motor neurons that somehow mediate the observed cell-type-specific toxicity. Alterations of axonal transport have been identified as one neuron-specific function that might contribute to the pathogenesis of this disorder.

470

SECTION IV / ENDOCRINOLOGY

mutant receptors are in the progression of prostate cancers toward the androgen-independent phenotype, the preponderance of analyses suggest that such mutations are found most frequently in advanced stage tumors, particularly in patients treated with AR antagonists, such as hydroxyflutamide.

Table 44-2 Distribution of 5α-Reductases I and II Expression in Rat and Human Tissues Rat

Skin Testes Epididymis Vas deferens Seminal vesicle Ventral prostate Prostate Ovary Adrenal Brain Colon Heart Intestine Kidney Liver Lung Muscle Spleen Stomach Pons Cerebral Hypothalmus

Human

Type I

Type II

Type I

Type II

— D* D*e D* D Dd*

— 1+* 5+*e 1+* 1+* Dd*

+* ND† ND† — ND†

+a† +b†,* — +†

ND†,*

+c†,*

2+* 3+* 3+* 3+* ND* 3+* 3+* 4+* 2+* ND* D* D* — — —

ND* D* ND* D* ND* D* ND* ND* ND* ND* ND* ND* — — —

+†,*

+†,*

ND ND ND

+ + +

aHistochemical studies indicate that the expression of 5α-reductase II is localized to the cells of the dermal papilla. bStaining localized to epithelial cells in histochemical studies. cStaining localized to basal epithelial and stroma cells. dType I is detected in basal epithelial cells and type II in stroma cells in in situ hybridization studies of the regenerating rat central prostate. eTypes I and II are localized to the epithelial cells using in situ hybridization. Summary of studies measuring 5α-reductases I and II expression using measurements of the corresponding RNA (*) or protein (†) are derived from published studies of Russell and coworkers. The rating scales (1+ least, 5+ highest) used in summary are designed to convey a sense of the relative abundance of the 5α-reductase isozymes in the two species—rat and human—that have been studied most carefully. Because the rat and human have been conducted separately, they can only be compared with one another in a qualitative fashion. D is detectable (i.e., at the limits of detection). ND, is not detected; —, values not reported. In addition to the results tabulated here, the studies of Thigpen et al. clearly indicate that in the human substantial changes in abundance can be demonstrated in the expression of the types I and II isoenzymes in a tissue at different times in development.

The second type of AR mutation that causes a gain or alteration of receptor function is the mutations identified in human prostate cancer specimens. First recognized in the AR gene of a prostate tumor cell line, LNCaP, amino acid substitutions have been detected in a number of clinical prostate cancer specimens. Although a number of these mutations have been found, fewer have been completely studied concerning their effect on the ligand responsiveness of the mutated AR. In instances in which detailed studies have been performed, tests of receptor function using androgenresponsive reporter genes demonstrate that the mutant ARs can be stimulated by ligands that cannot ordinarily activate the normal AR. Such ligands include adrenal androgens and even compounds (such as hydroxyflutamide) that act to antagonize the function of the normal AR. Although it is not possible to conclude how important these

5α-REDUCTASE DEFICIENCY PHYSIOLOGICAL AND CLINICAL STUDIES The observation that 5α-dihydrotestosterone was the principal hormone bound to the AR in the nuclei of target cells suggested the potential importance of 5α-reductase in androgen physiology. The subsequent identification of rare patients with specific defects of virilization that formed reduced quantities of 5α-reduced androgen metabolites emphasized the importance of this metabolic step. The normal virilization of other tissues, such as the epididymis in patients with clinical 5α-reductase deficiency, led to the concept that the action of testosterone was sufficient to effect the actions of androgen in selected tissues, but that the formation of 5α-dihydrotestosterone was crucial in others, such as the prostate. This dichotomy has not yet been completely explained at a mechanistic level. CLINICAL PHENOTYPE The clinical features of many infants with deficiencies of 5α-reductase are consistent with marked defects of androgen action (see Table 44-1). The external genitalia are characterized by a microphallus, severe hypospadias, and a bifid scrotum. A blind vaginal pouch is present and opens either directly onto the perineum or onto a urogenital sinus. Owing to the production of Müllerian inhibitory substance by the functional testes, no Müllerian structures are present. It is of interest that in association with the pubertal rise in testosterone levels, several changes may take place in the phenotype of affected individuals. The phallus enlarges and some male secondary sexual characteristics appear, such as changes in voice and muscle mass. Notably, this increase in testosterone has not been reported to result in acne, prostate growth, or male pattern baldness. It has also been reported that in some individuals, raised initially as females, gender identity may change after the pubertal rise in androgen levels. The ability to identify individuals with 5α-reductase II deficiency at the molecular level has permitted the recognition that this syndrome also includes individuals with less severely affected phenotypes.

5α-REDUCTASE STRUCTURE AND MECHANISM OF ACTION Attempts to purify 5α-reductase using classic techniques of protein purification failed and its structure remained elusive until expression cloning in Xenopus oocytes was employed to isolate a cDNA encoding a steroid 5α-reductase from the rat prostate. This advance permitted a number of studies and resulted in the isolation of cDNAs encoding two related isozymes (Fig. 44-2) encoded by two distinct, related genes from humans and from all vertebrate species examined to date. Inspection of the structures of the cDNAs indicates substantial sequence divergence between the two isozymes, both within and between different species. Both enzymes are extremely hydrophobic and are thought to be imbedded in the nuclear membrane of cells, accounting for the failure of extensive efforts to purify the enzyme in an active form. Structural variations between the two isozymes confer substantial differences in their physical properties that have been exploited to develop compounds that inhibit one or the other isozyme preferentially. Studies of the distribution of the two isozymes (Table 44-2) have demonstrated that the patterns of expression of these two proteins

CHAPTER 44 / DEFECTS OF ANDROGEN ACTION

471

Figure 44-3 Schematic of the pathways controlling the development of the normal male phenotype. Male sexual development is a complex cascade of events that can be affected by lesions in a number of genes. Some, such as lesions in the AR or the sex-determining region Y gene, can cause disturbances of virilization of all androgen-responsive tissues. The effects of defects in other genes, however, (such as mutation of 5α-reductase II), are manifest only in selected tissues. It is likely that defects in other genes can contribute to defects of virilization, as a large proportion of subjects with abnormalities of male phenotypic development cannot be accounted for by defects in the genes most carefully studied to date, such as the AR and steroid 5α-reductase 2 genes.

differ, both in terms of tissue, cell type, and developmental stage. These disparities suggest substantially different physiological roles for the two enzymes. GENETIC DEFECTS OF STEROID 5α-REDUCTASE II The genetic lesions causing clinical 5α-reductase deficiency have been identified in pedigrees from around the world. In all instances, when a genetic defect has been detected, it has been localized to the 5α-reductase II gene. Unlike the AR, the gene encoding 5α-reductase II is autosomal and each individual possesses two copies. For this reason, mutations causing clinical 5α-reductase deficiency are recessive and the inheritance of two defective 5α-reductase II gene alleles is required for the defects to be manifest. 5α-reductase II deficiency is infrequently caused by deletions or insertions in the gene. More often, defects in the gene are caused by mutations that result in premature termination of the protein or single amino acid substitutions within the open reading frame. Identification of the locations of these substitutions and determination of the physical properties of the mutant enzymes have permitted the identification of amino acid residues important for binding of steroid substrates and for the binding of nicotinamide adenine dinucleotide phosphate, a cofactor required for the reduction reaction. These studies have also identified sites that have been mutated repeated in apparently unrelated pedigrees. These regions are apparently more susceptible to mutation. As a consequence of the recessive nature of clinical 5α-reductase II deficiency, it would be expected that the genetic basis of this rare trait would be traced most frequently to homozygosity for a single defec-

tive allele (e.g., consanguinity). Surprisingly, a high proportion of patients (approx 30%) have been found to be compound heterozygotes, suggesting either an unexpectedly high mutation rate of the 5α-reductase II gene or a high frequency of individuals in the general population that carry single defective alleles. This apparent paradox has not been resolved. GENETIC DEFECTS OF STEROID 5α-REDUCTASE I An abnormal phenotype caused by a deficiency of 5α-reductase I in humans has not been described. The differential expression of steroid 5α-reductase I in selected tissues (see Table 44-2) suggests that specific phenotypes in humans that result from lesions in this gene or abnormalities of its expression might be identified in the future. Because of the lack of mutations in the human population, the first insights into the nature of defects that might be expected in humans have come from experiments in mice in which steroid 5α-reductase I gene has been disrupted by homologous recombination. Male mice homozygous for this targeted null allele develop normally, both prenatally and postnatally. Unexpectedly, although female mice homozygous for this same null allele develop normally into adulthood, when pregnant, such mice fail to initiate parturition normally. Experiments in which different steroids were administered to these steroid 5α-reductase I-deficient mice suggested that the synthesis and action of 5α-reduced androgens were important elements of the normal parturition process in mice. Although the mechanisms and generality of these findings remain to be determined, such results suggest that

472

SECTION IV / ENDOCRINOLOGY

5α-reductase I might play important—and even unexpected— roles in human physiology as well. ADDITIONAL LESSONS FROM MURINE MODELS Based on the phenotypes of human patients with defects of 5α-reductase II, it was anticipated that defects of virilization would be observed in animals in which the 5α-reductase II gene had been disrupted. Surprisingly, male mice in which both 5α-reductase II genes had been inactivated showed no defects of external genitalia development and only modest reductions in the size of fully formed seminal vesicles and prostates. These findings have been attributed to the observation that in animals with disruptions of either the 5α-reductase II or both the 5-reductase I and II genes, testosterone levels in target tissues are dramatically elevated. These results support the concept that 5α-reductase serves to amplify weak androgenic signals in tissues in which androgen concentrations are limiting (Fig. 44-3).

ACKNOWLEDGMENT The original work is in this chapter and was supported by NIH grants DK03892 and DK52678 and by grant I-1090 from the Robert A. Welch Foundation.

SELECTED REFERENCES Boehmer AL, Brinkmann AO, Nijman RM, et al. Phenotypic variation in a family with partial androgen insensitivity syndrome explained by differences in 5alpha dihydrotestosterone availability. J Clin Endocrinol Metab 2001;86:1240–1246. Boehmer AL, Brinkmann O, Bruggenwirth H, et al. Genotype versus phenotype in families with androgen insensitivity syndrome. J Clin Endocrinol Metab 2001;86:4151–4160. Giovannucci E, Stampfer MJ, Krithivas K, et al. The CAG repeat within the androgen receptor gene and its relationship to prostate cancer. Proc Natl Acad Sci USA 1997;94:3320–3323. Gottlieb B. Androgen Receptor Gene Mutation Data Base. The Lady Davis Institute for Medical Research, Sir Mortimer B.Davis-Jewish General Hospital. Montreal, Quebec, Canada. http://www.androgendb.mcgill.ca/. Accessed on March 29, 2006. Hardy DO, Scher HI, Bogenreider T, et al. Androgen receptor CAG repeat lengths in prostate cancer: correlation with age of onset. J Clin Endocrinol Metab 1996;81:4400–4405.

Hiort O, Sinnecker GH, Holterhus PM, Nitsche EM, Kruse K. The clinical and molecular spectrum of androgen insensitivity syndromes. Am J Med Genet 1996;63:218–222. Holterhus PM, Bruggenwirth HT, Brinkmann AO, Hiort O. Post-zygotic mutations and somatic mosaicism in androgen insensitivity syndrome. Trends Genet 2001;17:627, 628. Holterhus PM, Bruggenwirth HT, Hiort O, et al. Mosaicism due to a somatic mutation of the androgen receptor gene determines phenotype in androgen insensitivity syndrome. J Clin Endocrinol Metab 1997;82:3584–3589. Holterhus PM, Wiebel J, Sinnecker GH, et al. Clinical and molecular spectrum of somatic mosaicism in androgen insensitivity syndrome. Pediatr Res 1999;46(6):684–690. http://www.androgendb. mcgill.ca/. Accessed on March 29, 2006. Imperato-McGinley J. 5 Alpha-reductase-2 deficiency. Curr Ther Endocrinol Metab 1997;6:384–387. Lumbroso S, Lobaccaro JM, Vial C, et al. Molecular analysis of the androgen receptor gene in Kennedy’s disease. Report of two families and review of the literature. Horm Res 1997;7:23–29. Mahendroo MS, Cala KM, Hess DL, Russell DW. Unexpected virilization in male mice lacking steroid 5 alpha-reductase enzymes. Endocrinology 2001;142:652–662. Mahendroo MS, Cala KM, Russell DW. 5α-Reduced androgens are required for parturition in mice. Mol Endocrinol 1996;10:380– 392. Mahendroo MS, Russell DW. Male and female isoenzymes of steroid 5alpha-reductase. Rev Reprod 1999;4:179–183. Marcelli M, Zoppi S, Wilson CM, Griffin JE, McPhaul MJ. Amino acid substitutions in the hormone-binding domain of the human androgen receptor alter the stability of the hormone receptor complex. J Clin Invest 1994;94:1642–1650. Quigley CA, DeBellis A, Marschke KB, El-awady MK, Wilson EM, French FS. Androgen receptor defects: historical, clinical, and molecular perspectives. Endocr Rev 1995;16:271–321. Thigpen AE, Silver RI, Guileyardo JM, Casey ML, McConnell JD, Russell DW. Tissue distribution and ontogeny of steroid 5α-reductase isozyme expression. J Clin Invest 1993;92:903–910. Wigley WC, Prihoda JS, Mowszowicz I, et al. Natural mutagenesis study of the human steroid 5 alpha-reductase 2 Isozyme. Biochemistry 1994;33:1265–1270. Wilson JD, Griffin JE, Russell DW. Steroid 5α-reductase 2 deficiency. Endocr Rev 1993;14:77–93.

45 Molecular Endocrinology of the Testis MARCO MARCELLI, GLENN R. CUNNINGHAM, JOSÉ M. GARCIA, KIRK C. LO, AND DOLORES J. LAMB

SUMMARY This chapter reviews genetically influenced causes of male hypogonadism. Acquired cases of hypergonadotropic hypogonadism and acquired forms of gonadotropin deficiency from functional causes or systemic diseases have not been included because they do not have a genetic etiology. Key Words: Alström syndrome; Dystrophia myotonica 1 (DM1); hypergonadotropic; Kallmann syndrome (KS); Klinefelter syndrome; 46,XX; 46,XY; reproductive system; Smith–Lemli–Opitz syndrome (SLOS); testis.

INTRODUCTION Two distinct functions of the male reproductive system are essential for the survival of human species. First is the continuous production, nourishment, and storage of male gametes (spermatozoa). The second function is synthesis and secretion of the androgenic hormones necessary for male sexual differentiation and function. The testis is controlled by signals from the hypothalamuspituitary–gonadal (HPG) axis, in addition to local paracrine and autocrine signals to fulfil these functions.

MALE HYPOGONADISM Disorders of testicular function result from abnormalities involving the endocrine (Leydig cells) and/or reproductive (germ cell maturation) compartments of the testis. Low testosterone (T) production in adult males is usually accompanied by decreased libido, abnormal secondary sexual characteristics, and infertility, as normal spermatogenesis requires normal production of testicular androgens (hypogonadism with undervirilization and infertility). However, there are situations in which hypogonadism is restricted to the germinal compartment (hypogonadism with infertility and normal virilization). The serum level of gonadotropins indicates at what site within the HPG axis the defect is localized. In instances in which gonadal function is abnormal and the hypothalamic-pituitary (HP) structures are normal (hypergonadotropic or primary hypogonadism), gonadotropin levels are elevated. When the primary defect lies at the From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

level of the HP structures (hypogonadotropic or secondary hypogonadism), gonadotropin levels are inappropriately normal or low. Depending on the serum level of gonadotropins, hypogonadism can be classified as hypergonadotropic or hypogonadotropic (if resulting from testicular or HP disorders, respectively) (Table 45-1) and eugonadotropic (if the abnormality lies only in the germinal cells and serum T levels are normal). Both congenital and acquired forms of Leydig cell dysfunction have been described, and the clinical picture that results when T synthesis is impaired is different depending on whether androgen deficiency developed prenatally, before puberty, or after puberty (Table 45-2). If the developing fetus was not exposed to an adequate level of androgen because of testicular failure during fetal development, the infant manifests pseudohermaphroditisms, characterized by a large spectrum of potential abnormalities, including ambiguous genitalia, micropenism, and rudimentary testes. If reduced T production developed before puberty, but testicular function was normal during embryogenesis, androgen-related somatic changes normally observed at puberty are absent or incomplete and the patient develops an eunuchoid habitus because of failure of the epiphyses to close, and poor development of skeletal muscles and body hair. If reduced T production developed after puberty, the first symptom of the patient is impotence, whereas loss of secondary sexual characteristics may take several years to become complete. Hypogonadism manifested by abnormal sperm production can be associated with normal virilization, because many infertile men produce a normal or only a minimally abnormal amount of T. However, in other cases infertility is associated with a reduced production of T and clinical signs of androgen deficiency (infertility from hypergonadotropic or hypogonadotropic disorders [see Table 45-1]). In such cases, infertility may be a direct consequence of the reduced T production.

HYPERGONADOTROPIC DISORDERS Hypergonadotropic abnormalities of the testes are owing to primary testicular failure (see Table 45-1). They can be classified in two main groups: congenital or acquired disorders. Some genetic syndromes lead to primary hypogonadism as a result of chromosomal abnormalities, enzymatic defects, or receptor defects. Others are congenital multiorgan diseases with associated hypogonadism, and it is not always clear why patients affected by some

473

474

SECTION IV / ENDOCRINOLOGY

Table 45-1 Hypogonadotropic and Hypergonadotropic Disorders (A) HYPERGONADOTROPIC DISORDERS Genetically inherited Chromosomal abnormalities Klinefelter 46,XX males 46,XY pure gonadal dysgenesis Multi-organ diseases Alström syndrome Dysptrophia myotonica 1 Dystrophia myotonica 2 Cardiomyopathy with hypogonadotropic hypogonadism Alopecia-mental retardation syndrome with convulsions and hypergonadotropic hypogonadism Insulin-resistant diabetes with acanthosis nigricans, hypogonadism, pigmentary retinopathy, deafness, and mental retardation Testicular regression syndrome (bilateral anorchia) Hypergonadotropic hypogonadism in autoimmune polyglandular syndromes I and II Receptor defects LH resistant testis Syndromes of androgen resistance

Defects of steroidogenesis 7-dehydrocholesterol reductase deficiency (7-DHCR-D) or Smith–Lemli–Opitz syndrome Lipoid CAH 3β-hydroxysteroid dehydrogenase 2 deficiency 17α-hydroxylase 17,20-lyase (CYP17) deficiency 17β-hydroxy steroid dehydrogenase deficiency Acquired Viral Medications Trauma Environmental Autoimmunity Ionizing radiations

(B) HYPOGONADOTROPIC DISORDERS Genetically inherited Associated with gonadotropin deficiency GPR54 deficiency Kallmann syndrome Fertile eunuch syndrome Hypogonadotropic hypogonadism from biologically inactive molecules of the hypothalamic–pituitary axis Hypogonadotropic hypogonadism from mutations of genes causing combined pituitary hormone deficiencies Hypogonadotropic hypogonadism from mutations of genes associated with obesity Leptin and leptin receptor Prohormone convertase-1

Associated with central nervous system disorders Prader–Lahart–Willi Laurence–Moon–Biedl Möbius Borjeson–Forssman–Lehman syndrome Leopard syndrome Rud’s syndrome Loewe’s syndrome Carpenter syndrome Associated with adrenal insufficiency Adrenal hypoplasia congenita

Acquired Associated with anatomic disorders Pituitary apoplexy Primary or metastatic tumors of pituitary or adjacent structures Infection Infiltrative disorders (Hemochromatosis, sarcoidosis, histiocytosis, lymphocytic hypophysitis) Empty sella Functional Secondary to Cushing’s syndrome to ethanol to exercise to empty sella to hyperprolactinemia Associated with systemic diseases AIDS Liver diseases Renal diseases Hemochromatosis Neurological diseases

7-DHCR-D, 7-dehydrocholesterol reductase deficiency; CAH, congenital adrenal hyperplasia; LH, luteinizing hormone.

CHAPTER 45 / MOLECULAR ENDOCRINOLOGY OF THE TESTIS

Table 45-2 Manifestations of Leydig Cell Dysfunction Prenatal leydig cell dysfunction: External genitalia: female, ambiguous; male, hypoplastic Wolffian ducts derivatives: absent—rudimentary Müllerian ducts derivatives: absent Gonads: Small—rudimentary testes, which can be located within the abdomen, inside the inguinal canal or in the scrotum Prepubertal Leydig cells dysfunction: External genitalia: small penis 1/40 in 21%, antismooth muscle antibody with a titer >1/40 in 21%, and antiliver kidney microsomal antigen in 5%).

DIAGNOSTICS A number of immunological and molecular assays have been developed for the detection and assessment of hepatitis C status. The immunological or serological tests identify the presence of antiHCV antibodies, which indicates previous exposure to the virus without differentiating between acute, chronic or resolved infection. In contrast, the molecular or virological assays detect specific viral

547

nucleic acid sequences (HCV RNA), which indicate ongoing presence of the virus. Although serological assays are typically used for screening and first-line diagnosis, molecular/virological assays are needed in certain situations with the aim of confirming infection and/or monitoring therapy. SEROLOGICAL TESTS There are two types of anti-HCV assays, the enzyme immunoassay (EIA) and the recombinant immunoblot assay (RIBA). Both detect antibodies against different HCV antigens from the core and nonstructural proteins and were developed using recombinant HCV antigens. Serological assays are typically used for screening and first-line diagnosis. Three generations of EIAs have been developed with increasing sensitivity. As first-generation EIAs lack sensitivity and specificity, confirmatory RIBAs were systematically used in samples positive in EIA. With the newer assays and the increased use of molecular assays, confirmatory RIBAs are seldom needed. Screening Assays The latest EIA-3 assay uses recombinant antigens from the core, NS3/4A (c200) and NS5 (EIA-2 does not have this capture antigen) regions to provide better sensitivity for the detection of antibody to HCV. Even with these new assays, specificity might need to be confirmed in some populations, especially, those with low-risk. In this setting, confirmatory tests, typically molecular testing, are generally helpful. Confirmatory Assays The RIBA tests use the same antigens as the corresponding EIA tests but in a strip or immunoblot assay format. Their sensitivity is generally lower than for EIA assays, but their specificity is better because the reactivity specific for the capture antigen can be visualized. The 4-antigen RIBA-2 (RIBA HCV 2.0 Strip Immunoassay, Chiron Corporation) was approved by the United States Food and Drug Administration in June 1993. This assay incorporated antigens 5-1-1, c100-3, c22-3, c33c and SOD (superoxide dismutase, as a control). With this assay, a specimen is considered positive if two bands or more are detectable, representing at least the presence of antibodies to two different HCV-derived capture antigens. If a specimen reacts with bands from only one region, such as 5-1-1 and c100-3 (both from the NS4 region), it is considered indeterminate. If a specimen reacts with the SOD band in addition to two or more HCV antigen bands, the result is also considered indeterminate. The third generation assay, RIBA-3 (RIBA HCV 3.0 Strip Immunoassay, Chiron Corporation) includes two recombinant antigens (c33c and NS5) and synthetic peptides from the nucleocapsid (c22) and the NS4 (c100-3) regions. It has been approved for use by blood banks as a supplemental test for EIA-3-positive samples. This new RIBA format is able to identify correctly HCV-negative and HCVpositive results in those samples that are RIBA-2 indeterminate. VIROLOGICAL TESTS As HCV cell culture systems have not been established, detection of viral RNA in serum has been used as a direct marker of the virus. Qualitative HCV RNA assays are optimized to produce maximal signal regardless of the amount of input HCV RNA. In comparison, quantitative assays are developed to yield reproducible and accurate quantification values regardless of the input concentration of virus. Thus, these two types of assays are developed differently. Viral detection (i.e., qualitative result) is accomplished by target amplification methods such as the PCR-based AMPLICOR assay (Cobas Amplicor HVC Kit, Roche Diagnostics) or by the TMA-based VERSANT assay (Versant HCV RNA 3.0 Assay, Bayer Diagnostics). The qualitative PCR test for HCV might detect as few as 50 International Units (IU)/mL. The VERSANT assay employing TMA has a sensitivity

548

SECTION VI / GASTROENTEROLOGY

of 10 IU/mL. To achieve these results, specific procedures directed to preserving the integrity of HCV RNA should be followed carefully, including separation of serum or plasma from whole blood within 4 h of venipuncture, and rapid storage of specimens at –70°C. Qualitative assays for HCV RNA are used for confirmation of diagnosis and for assessment of viral clearance at the end of therapy and at follow-up to therapy (i.e., 6 mo after cessation of therapy). Ninety-five percent of viral loads at baseline are within the range of 33,500–3,700,000 IU/mL; therefore, the sensitivity of either of the qualitative assays should be sufficient to detect HCV RNA in a confirmatory application. In comparison, the TMAbased assay detected HCV RNA in (7–30% of) relapsers at the end of treatment when the PCR-based assay detected none. In this setting, very high sensitivity is important in detection of residual HCV RNA at the end of therapy. The quantitative HCV RNA tests allow measurement of viral load and they are useful in monitoring the effectiveness of antiviral therapy and in evaluating the course of clinical disease. The branched DNA (bDNA) VERSANT assay (Bayer Diagnostics) uses capture and target probes directed to the conserved 5′ UTR and core regions in a microwell format. HCV RNA captured to the microwell supports the formation of very large hybridization complexes that include synthetic branched DNA oligonucleotides (signal amplification). These large hybrids contain many copies of probes that are labeled with alkaline phosphatase. Following addition of a chemiluminescent substrate, light is emitted in proportion to the number of HCV RNA molecules in the original specimen; the quantity of HCV RNA is determined by comparison of chemiluminescence from the specimen with a standard curve containing calibrators with known amount of target. The third generation assay, registered with the FDA, has been shown to measure down to 615 IU/mL with a dymanic range of over 4 logs (upper limit of detection 7.7 × 106 IU/mL). The Amplicor HCV Monitor, v 2.0 quantitative assay provides a semiquantitative measurement of viral RNA. HCV RNA is reverse transcribed and amplified by end point PCR (i.e., with a specific number of amplification cycles) using primers from the most conserved region (in 5′ UTR) of the HCV genome. This reaction produces very large numbers of double-stranded DNA that has the same sequence as a portion of the HCV RNA in the specimen. The assumption is that the amount of amplicon produced is directly proportional to the amount of HCV RNA in the specimen. A quantitation standard (with a known number of HCV RNA targets) is added to each specimen and coamplified with the target. An internal control also is added to each specimen and is used to indicate if there are inhibitors of the RT-PCR steps in the reaction. A microwell plate detection system with chemiluminescent label is used for quantification of amplicons produced. The amount of viral RNA present in the amplified sample is calculated based on the signal from HCV RNA in the specimen and on the signal generated from the quantification standard. The lower limit of detection of the AMPLICOR assay is approx 600 IU/mL. Recently, kinetic or real-time PCR methodologies have been applied to quantification of HCV RNA. Amplification by PCR is geometric and nucleotide and/or enzyme might become rate limiting during the reaction; the rate of accumulation of amplicons might decrease during the reaction and with higher input numbers of HCV RNA. Thus, end point PCR assays have a limited dynamic range of 2.5–3 log10 units. In comparison, real-time PCR assays

are carried out like end point assay except that signal from amplicons is detected during amplification. The fractional PCR cycle in which signal exceeds a critical threshold value (Ct), marks the start of the exponential phase of target amplification. A lower Ct indicates a higher input concentration of HCV RNA. The quantity of HCV RNA in a specimen is calculated based on the Ct for the specimen and the Ct for the Quantification Standard (with a known number of HCV target molecules). Alternatively, quantification might be based on the Ct for the specimen and the Ct’s for a series of calibrators. Fluorescence energy transfer is used in many probe systems, including TaqMan, employed with the real time PCR technology. The assay is designed to avoid the problems associated with rate limitation resulting from enzyme and/or nucleotide depletion and, thus, is reported to be linear. Roche Molecular Systems and Abbott Diagnostics employ real time PCR in their Analyte Specific Reagents or kits. Bayer Diagnostics currently is developing assays that employ real time technologies. Quantitative assays are employed in monitoring the efficacy of therapy often by showing changes in viral load between baseline (i.e., start of therapy) and during therapy. Linearity of assay response (i.e., degree to which the assay standard curve approximates a straight line) provides the clinician and patient accurate results regardless of the level of virus; changes in viral load reflect actual changes in the patient. The bDNA-based assay is linear (slopes of about 1) whereas the Amplicor HCV Monitor assay might exhibit nonlinear performance at viral loads greater than 500,000 IU/mL. Real-time PCR-based assays are linear. An immunoassay that provides quantification values for HCV core antigen has been introduced as an indirect measure of viral load. This assay has a sensitivity of approx 20,000 IU/mL. Thus, the 2-log10 stopping rule at week 12 of therapy is employed in assessing nonresponse could be employed only in patients with viral loads of ≥2,000,000 IU/mL at baseline. To date, this assay has limited application because it is an indirect measure of virus and it is considerably less sensitivity than are the quantitative assays. To permit universal standardization of HCV RNA quantitation units, a recent collaborative study established the World Health Organization (WHO) International Standard for HCV RNA quantitation. A lyophilized genotype 1 sample was accepted as the candidate standard and was assigned a titer of 105 IU/mL. This standard has been used to calibrate HCV RNA panels and by industrial manufacturers to express HCV RNA load in IU/mL in their assays (226). The goals of standardization are to allow for generalization of findings from clinical studies with one assay to patient management with all of the assays and to permit clinicians to use results from any of the assays in monitoring patients. HCV GENOTYPING Differentiation of the HCV genotypes is accomplished by direct sequencing, differential PCR using typespecific primers, detection of PCR products with type-specific probes (available commercially, the Line Probe Assay), restriction fragment length polymorphism analysis, and serotyping with typespecific antibody. These techniques use targets in the 5′ UTR, NS4, NS5, and core regions. Of the many methods available for genotyping, direct sequencing coupled with phylogenetic tree analysis is regarded as the gold standard. However, sequencing methods are technically demanding and time-consuming. A study comparing sequencing, type-specific primers, serotyping, and restriction fragment length polymorphism analysis, demonstrated more than 90% concordance between methods.

CHAPTER 52 / HEPATITIS C

SELECTION OF SEROLOGICAL AND VIROLOGICAL TESTS Initial diagnostic testing of HCV infection is currently made by detecting specific antibody by EIA. For low-risk patients, a negative EIA test is sufficient to rule out infection. A positive EIA for low-risk patients should be confirmed by RIBA assay or molecular testing. For high-risk patients, individuals with recent exposure or immunocompromised patients (including HIV, chronic hemodialysis patients, transplant patients), a negative EIA should be confirmed and molecular test such as PCR or TMAbased qualitative test is the choice. SCREENING IN BLOOD BANKS With the latest EIA and RIBA assays, the number of false positive or indeterminate test results is very low. One major limitation of these assays however is the existence of a window period by which the patient might transmit the virus before antibodies are detectable. With recent assays, the average delay is approx 80 d. For this reason, many blood banks throughout the world have instituted testing all donors with qualitative PCR or TMA assays. DIAGNOSIS IN IMMUNOCOMPROMISED PATIENTS Antibody tests underestimate the prevalence of HCV infection in immunocompromised populations and detection of viral RNA is frequently required for diagnosis. Antibody assays also lack sensitivity in the transplant setting, as these patients are immunosuppressed. Molecular/virological assays for HCV RNA are the assays of choice. LIVER HISTOLOGY The range of histological findings in patients with chronic HCV infection is broad, from minimal periportal lymphocytic inflammation to active hepatitis with bridging fibrosis, hepatocyte necrosis, and frank cirrhosis. Steatosis, lymphoid aggregates, and bile duct damage are frequently found in the liver biopsies of patients with this infection, but there is considerable overlap with the histological findings in patients with chronic HBV infection and occasionally with autoimmune hepatitis. The role of a liver biopsy in the management of HCVinfected patients is still not settled although most physicians perform a liver biopsy before initiating treatment. This recommendation is waning somewhat for patients with genotypes 2 and 3, in which the viral response is much higher with a shorter course of therapy. The potential risks associated with a liver biopsy (1:3000 chance of bleeding, 1:12,000 chance of death) contrast with the importance of the information generated by this means. Recent consensus conferences have stated that a liver biopsy is recommended in patients with chronic hepatitis C and abnormal serum ALT levels so that correct staging and grading can be performed. This information is particularly relevant when considering antiviral therapy or when other causes of liver disease might be present. A posttreatment liver biopsy is not necessary as most trials have demonstrated that viral sustained response is generally associated with stable or improved histological findings. In patients with normal serum ALT levels, liver biopsy might also provide information about the stage of disease, particularly if antiviral therapy is being considered. Although fibrosis progression appears to be slow in these patients, only 20% have an absolutely normal liver, and a small percentage (50 g/d) have been documented as associated variables. Immunosuppression is clearly linked with more aggressive disease. Patients with humoral immunodeficiency (hypo γ-globulinemic patients) or cellular immune impairment (liver or kidney transplant recipients, HIV infected patients with low CD4 count) have shown rates of liver disease progression significantly faster than those observed in immunocompetent patients. For patients with compensated HCV-related cirrhosis, the prognosis of patients in the short-term might be good. Actuarial survival is 91% at 5 yr and 79% after 10 yr in the absence of clinical decompensation. Survival drops to 50% at 5 yr among those who develop clinical decompensation. The cumulative probability of developing an episode of clinical decompensation is approx 5% at 1 yr, increasing to 30% at 10 yr from the diagnosis of cirrhosis. The risk of developing HCC is 1–4%/yr once cirrhosis is established. For patients with persistently normal serum ALT levels (around one-third of the patients), most will have some degree of histologicaly proven chronic liver damage ranging from mild chronic hepatitis to cirrhosis. Disease progression appears to be slower in these patients than in those with abnormal serum ALT levels, and progression to cirrhosis is unlikely. For renal transplant recipients, elevated serum ALT levels are more common in anti-HCV positive than in anti-HCV negative patients (median 48 and 14%, respectively). There are also reported differences in patient and graft survival between antiHCV positive and anti-HCV negative kidney transplant recipients, the increased mortality being related to liver dysfunction. HCV is evolving as an important issue in liver transplantation. Chronic HCV infection is the most common indication for liver transplantation in the United States. Recurrence of HCV following liver transplantation is universal. Viremia level increased by one log after liver transplantation. In a small proportion of patients (5%), an accelerated course of liver injury leading to rapid development of liver failure, reminiscent of that previously described in HBV-infected recipients with fibrosing cholestatic hepatitis has been observed. Disease progression is significantly faster than that observed in immunocompetent patients, and the time to develop cirrhosis is short in this population. Although histological evidence of liver injury will develop in approximately half of the patients within the first year posttransplantation, severe graft dysfunction in the short-term is infrequent. With longer follow-up (5–7 yr), a significant proportion of patients, ranging from 8 to

550

SECTION VI / GASTROENTEROLOGY

30%, develop HCV related graft-cirrhosis. The risk of decompensation once the transplant patient has developed clinically compensated cirrhosis is approx 40% at 1 yr. For patients coinfected with HCV and human immunodeficiency virus undergoing active antiretroviral therapy, chronic hepatitis C is a growing cause of morbidity and mortality in this population. It was also observed that hepatic decompensation develop more rapidly in coinfected patients than in those with HCV infection alone but the mechanism responsible for this clinical observation was not known.

TREATMENT The primary goal of therapy is to eradicate infection early in the course of the disease to prevent progression to end-stage liver disease and eventually to HCC. However, this is a long-term goal and cannot be measured in short-term clinical studies. In this context, normalization of serum ALT levels, loss of HCV RNA, and improvements in histological findings have been employed as the current standard therapeutic end points. Current data suggest that the achievement of these end points will translate into long-term benefit from therapy, as measured by reduction in the rate of disease progression, reduction in the need for liver transplantation, reduction in rate of development of HCC and improvement in survival. Indeed, several studies have shown that in patients with sustained biochemical and virological responses, the durability of the response is more than 95% for up to 10 yr. A lesser goal of therapy is to reduce the secondary spread of infection by eradicating the viremia. Interferon-α based regimens constitute the cornerstone of our current antiviral therapies. There are several types of interferon-α available, including recombinant forms (interferon α-2a, α-2b, and consensus interferon) and naturally occurring forms (lymphoblastoid interferon). Interferons are naturaly occurring proteins, which exert a wide array of antiviral, antiproliferative, and immunomodulatory effects. For years, interferon-α given in monotherapy was the standard treatment in patients with chronic hepatitis C, with results considered modest at best. A modified version of interferon, pegylated interferon-α, which consists of interferon-α bound to a molecule of polyethylene glycol of varying length and with different order of branches has been developed. This process of pegylation increases the half-life of the molecule and reduces the volume of distribution. This altered pharmacokinetic profile provides more uniform and prolonged plasma levels, which in turn have led to weekly dosing of this product. Such a profile with better drug exposure is leading to better therapeutic efficacy compared to the unmodified interferon-α. Ribavirin is a broad-spectrum antiviral agent with activity against both DNA and RNA viruses. In immunocompetent patients with chronic HCV infection, ribavirin resulted in improved serum ALT levels but the effect was transient and no significant direct antiviral activity was observed. Ribavirin is known to induce hemolysis, a side effect that is dose limiting. Because of this side effect, ribavirin is contraindicated in patients with a past history of myocardial infarction or cardiac arrhythmia. Ribavirin is also potentially teratogenic and patients and partners are required to avoid pregnancy during therapy and for 6 mo after cessation of treatment. Ribavirin has a long cumulative half-life. It is excreted by the kidney and cannot be removed by hemodialysis. Hence, ribavirin is contraindicated in patients with a serum creatinine of greater than 1.5 mg/dL.

Although neither interferon-α nor ribavirin monotherapy is very effective in the treatment of chronic HCV infection, the combination of these two drugs improves the treatment response rate by many folds and is currently the gold standard for the treatment of chronic HCV infection in individuals who have not received any previous treatment (i.e., treatment naive). The current recommendation for patients with HCV genotype 1 infection is pegylated interferon-α injection (dosing depends on the preparation: 180 µg weekly for pegylated interferon-α2a; and 1.5 µg/kg weekly for pegylated interferon-α2b) for 48 wk in combination with oral ribavirin (1000–1200 mg daily with pegylated interferon-α2a and 800 mg daily with pegylated interferon-α2b). Trials using pegylated interferon-α2a have compared safety and efficacy of 800 mg vs 1000–1200 mg of ribavirin daily, and for genotype 1 infection, results suggest superiority of the higher dose. Given these findings, it is likely that 1000–1200 mg of ribavirin is necessary for optimizing outcomes in these patients whether given in combination with pegylated interferon-α2b or -α2a. The recommendation for patients with HCV genotypes 2 and 3 infection is similar to patients with HCV genotype 1 infection except that 24 wk of therapy, and a dose of ribavirin at 800 mg appear adequate. Overall, the long-term sustained response (defined as normal serum ALT and with no detectable HCV RNA 6 mo after cessation of therapy) is around 55%. For patients with HCV genotype 1 infection, the long-term sustained response is around 40% and for patients with genotypes 2 and 3 infection, the long-term sustained response is around 80%. Adherence to full dose and duration of therapy can improve the sustained virological response rate by another 10%. Factors associated with a better treatment outcome include low pretreatment serum HCV RNA levels, genotypes 2 and 3 infection, absence of cirrhosis, female gender and age less than 40 yr. African-Americans have generally been shown to respond less well to treatment than Caucasians, whether the treatment is interferon or pegylated interferon plus ribavirin. It should also be noted that with the current combination of pegylated interferon-α and ribavirin therapy, patients with HCV-related cirrhosis and patients with “transition to cirrhosis” have shown similar response rate as noncirrhotic patients, although there have been few studies focusing specifically on outcomes in patients with advanced liver disease. Thus, the presence of clinically compensated cirrhosis should not exclude patients from antiviral therapy. Regarding indications, theoretically all patients with ongoing HCV infection with persistent elevation of serum ALT levels are potential candidates of antiviral therapy. However, given that (1) our therapies are still imperfect in terms of efficacy, and are associated with significant side effects; and (2) the natural history of hepatitis C is benign in the majority of patients, both physicians and patients need to evaluate the potential benefit, the costs, and the risks before initiating therapy. Ironically, patients with the lowest likelihood of progression are precisely the ones who are most likely to respond. In contrast, those with the least likelihood of responding or with the least tolerance of side effects of treatment are often the ones with the greatest need. In patients with a low chance of disease progression (immunocompetent, less than 40 yr of age, female gender, no alcohol consumption, minimal histological inflammation, and fibrosis), there are as many arguments to recommend starting treatment as there are to defer treatment until safer and more effective therapies are available. There are certain situations in which the guidelines regarding treatment are not yet well developed, in part because studies in

CHAPTER 52 / HEPATITIS C

certain patient populations are still underway and complete results have not yet been published in the peer-reviewed literature. These include patients with active substance abuse (alcohol and injection drug use), those with clinically decompensated cirrhosis, those with very mild liver disease and those with coinfection with other viruses such as HIV and/or hepatitis B. As the need for treatment is high in many of these subpopulations, results from carefully conducted clinical trials are greatly needed. Clearance of virus in a larger percentage of patients might ultimately require the development of HCV therapies with different mechanisms of action to those that are currently available. At present, combination therapy using the nonpegylated interferon in combination with oral ribavirin has been approved for the treatment of children with chronic HCV infection. Patients with HIV coinfection are frequently being treated in the context of randomized controlled trials. Preliminary data suggest that although sustained viral clearance is achievable, overall responses to peginterferon plus ribavirin are lower than those observed in HIV-negative patients. Interferon has produced short-term improvement in the signs and symptoms of cryoglobulinemia in those patients demonstrating a virological response, although recurrence of viremia and cryoglobulinemia is the rule when treatment is discontinued. Interferon has been used successfully in the treatment of HCVassociated membranoproliferative glomerulonephritis. There is a theoretical rationale for treating patients with acute HCV infection as the majority will develop persistent infection and chronic liver disease. The few studies that have been conducted in patients with acute HCV have used different dosage and duration of interferon therapy. Whereas, most are uncontrolled case series, high rates of remission have also been reported. The Manns study showed clearance of virus in 98% of a small number of treated patients; these patients were selected for the presence of icterus. In patients who have relapsed after a course of interferon monotherapy, combination therapy is the treatment of choice. There are no FDA-approved treatments for nonresponders to previous treatment with interferon alone, or interferon-α/ribavirin combination therapy. If retreatment is undertaken with peginterferon plus ribavirin in those who have failed a previous course, viral clearance is achievable, particularly in patients with genotype non-1 infection, those who have failed previous interferon monotherapy rather than interferon plus ribavirin combination therapy and those who have demonstrated some degree of treatment sensitivity with a reduction in HCV RNA levels during the first course of therapy. Monitoring before and during therapy is important in the overall management of patients with chronic HCV infection. Before starting therapy, blood tests should be conducted to establish baseline status, including liver biochemistry, complete blood count, viral load (with a quantitative HCV RNA test), and thyroid-stimulating hormone level. A pregnancy test is needed to rule out pregnancy before initiating ribavirin therapy in women. Strongly worded counseling needs to take place for partners who potentially could be involved in conception during ribavirin treatment to utilize a dual method of contraception to prevent pregnancy. HCV genotyping will help in selecting the best treatment strategy (i.e., dosage of ribavirin and duration of therapy). During the first month of therapy, a complete blood count should be done weekly as most side effects occur within this period of time, particularly hemolytic anemia related to ribavirin therapy. Approximately 9–23% of patients, depending on the ribavirin dose, will have a fall in hemo-

551

globin to less than 10 g/dL with a mean decrease of approx 3 g/dL. As dose reductions of ribavirin have the potential to compromise response to combination therapy, adjunctive treatments such as erythropoietin are being increasingly used to manage side effects in patients receiving HCV treatment. After the first month, liver biochemistry and a complete blood count should be performed monthly, and thyroid-stimulating hormone level every 3 mo. Modifications of doses should be done according to the severity of side effects. The recent NIH Guidelines recommend that viral load should be assessed at week 12 of therapy. Several studies of therapy with pegylated interferon plus ribavirin showed that patients in whom the viral load did not decrease by at least 2 log10 almost invariably (97–100% negative predictive value) were nonresponders with the full course of therapy. At the end of treatment and at 6 mo postdiscontinuation of treatment, serum HCV RNA testing with sensitive, qualitative assay should be carried out to assess treatment response. If sustained response is achieved, it is recommended that serum HCV RNA testing be performed annually for at least 2 yr after completion of therapy. A repeat liver biopsy following treatment is rarely necessary. Newer therapies are being developed for anti-HCV infection. A number of the HCV viral peptides in the replication machinery are good targets for novel therapy development. Current efforts are focusing on targeting HCV internal ribosomal entry site, NS3 protease, NS3 helicase, and NS5B RNA-dependent RNA polymerase. Molecular approaches utilizing antisense, ribozyme, and small inhibitory RNA directed against the HCV genome are also being tested. A number of immunomodulators are also being tested for their ability to booster the host immune response. These include HCV-specific immune stimulation, i.e., therapeutic vaccine, as well as non-HCV specific immunomodulation using various cytokines. Efforts are also directed at developing a safer version of ribavirin, a prodrug of ribavirin that is predominately released by the deaminase enzyme system in the liver.

PREVENTION GENERAL MEASURES As there is no effective prophylactic vaccine and no effective postexposure prophylaxis, major efforts should be placed on counseling both HCV-infected patients and those at risk of infection on the general measures. Adequate sterilization of medical and surgical equipment is mandatory. Efforts should also be made to modify injection practices involved in folk medicine, rituals, and cosmetic procedures. Persons at high-risk of HCV infection such as recipients of blood/blood product transfusions before 1990, drug users, and sexual partners of persons infected with HCV should be tested for antiHCV. HCV-infected patients should be instructed to avoid sharing razors and toothbrushes and to cover any open wounds. Given the low rate of vertical transmission, pregnancy is not contraindicated in HCV-infected women. No recommendations have been issued regarding the mode of delivery. Breast-feeding is not contraindicated. It is recommended that HCV-infected patients be vaccinated against hepatitis A and B, given the increased risk of severe liver disease if superinfection with these viruses occurs. PASSIVE IMMUNOPROPHYLAXIS Immune serum globulin has not been studied clinically in the prevention of HCV infection. Nevertheless, there are several lines of evidence suggesting that such measures are unlikely to be effective. Studies from the 1970s of immune serum globulin as a measure of prophylaxis against posttransfusion NANBH failed to demonstrate any signif-

552

SECTION VI / GASTROENTEROLOGY

icant benefit. The neutralizing immune response to HCV infection even in healthy adults appears to be weak and hence it would seem unlikely that immune serum globulin would contain sufficient neutralizing antibody to be effective. Finally, HCV has an inherently high nucleotide substitution rate that would predictably facilitate rapid escape from humoral immune recognition. Despite this scientific rationale, polyclonal immunoglobulins containing anti-HCV are being tested to decrease the incidence of recurrent HCV viremia 1-yr post-transplantation, preliminary results have not been encouraging and long-term protection has not been reported. Whether such immune sera will be protective in the setting of exposure to virus remains to be determined. ACTIVE IMMUNOPROPHYLAXIS Vaccine development for HCV appears to encounter the same difficulties encountered in the development of human immunodeficiency virus vaccine. The absence of natural protective immunity in chimpanzees that recovered from HCV infection to challenges with both heterologous and homologous strains of virus bodes poorly for vaccine development. Nevertheless, neutralizing antibodies to HCV can be elicited during natural infection and have been demonstrated to be effective at preventing viral transmission to experimental animals, when used to neutralize virus in vitro. Substantial efforts are currently directed to the development of prophylactic HCV vaccines. “Science has radically changed the conditions of human life on earth. It has expanded our knowledge and our power but not our capacity to use them with wisdom” J. William Fulbright (1905–1995) In: Old Men and New Realities

ACKNOWLEDGMENTS The authors thank all their colleagues who share their data, knowledge, and passion that drives the advancement of basic and clinical science with them. Many scientists and clinicians secured our knowledge on HCV. Neither space nor time has allowed us to be comprehensive. For those significant contributions that we failed to mention, we apologize.

SELECTED REFERENCES Alberti A, Chemello L, Benvegnu L. Natural history of hepatitis C J Hepatol 1999;31:17–24. Alter HJ, Seeff LB. Recovery, persistence, and sequelae in hepatitis C infection: a perspective on long-term outcome. Semin Liv Dis 2000; 20:17–35. Alter MJ, Kniszon-Moran D, Nainan OV, et al. The prevalence of hepatitis C virus infection in the United States, 1999 through 1994. N Engl J Med 1999;341:556–562. Bellentani S, Tiribelli C. The spectrum of liver disease in the general population: lesson from the Dionysos study. J Hepatol 2001;35:531–537. Berenguer M, Prieto M, Rayon JM, et al. Natural history of clinicaly compensated HCV-related graft cirrhosis following liver transplantation. Hepatology 2000;32:852–858. Berebguer M, Lopez-Labrador FX, Wright TL. Hepatitis C and liver transplantation. J Hepatol 2001;35:666–678. Bouvier-Alias M, Patel K, Dahari H, et al. Clinical utility of total HCV core antigen quantification: a new indirect marker of HCV replication. Hepatology 2002;36:211–218. Bukh J, Miller RH, Purcell RH. Genetic heterogeneity of hepatitis C virus: quasispecies and genotypes. Semin Liver Dis 1995;15:41–63. Bukh J, Forns X, Emerson SU, Purcell RH. Studies of hepatitis C virus in chimpanzees and their importance for vaccine development. Intervirology 2001;44:132–142. Centers for Disease Control and Prevention. Recommendations for prevention and control of hepatitis C virus (HCV) infection and HCVrelated chronic disease. MMWR 1998;47(No. RR-19).

Comanor L, Hendricks DA. Hepatitis C virus RNA tests: performance attributes and their impact on clinical utility. Expert Rev Mol Diagn 2003;3:689–702. Davis GL, Lau JYN. Choice of appropriate endpoints of response to interferon-α therapy in chronic hepatitis C virus infection. J Hepatol 1995;22(suppl 1):110–114. Davis GL, Lau JYN. Factors predictive of response to interferon. Hepatol 1997;26:122S–127S. Davis GL, Esteban-Mur R, Rustgi V, et al. Interferon alfa-2b alone or in combination with ribavirin for the treatment of relapse of chronic hepatitis C. N Engl J Med 1998;339: 1493–1499. Di Bisceglie AM, McHutchinson J, Rice CM. New therapeutic strategies for hepatitis C. Hepatology 2002;35:224–231. Elbeik T, Surtihadi J, Destree M, et al. Multicenter evaluation of the performance characteritics of the Bayer VERSANT HCV RNA 3.0 assay (bDNA). J Clin Micro 2004;42:563–569. Fang JWS, Chow V, Lau JYN. Virology of hepatitis C virus. Semin in Liver Dis 1997;1:493–514. Farci P, Alter HJ, Wong D, et al. A long-term study of hepatitis C virus replication in non-A, non-B hepatitis. N Engl J Med 1991;325: 98–104. Gish RG. Standards of treatment in chronic hepatitis C. Semin Liver Dis 1999;19:35–47. Gretch DR. Diagnostic tests for hepatitis C. Hepatology 1997;26:43S–47S. Hadziyannis SJ, Sette H Jr, Morgan TR, et al. Pegasys International Study Group. Peginteron alpha2a and ribavirin combination therapy in chronic hepatitis C: a randomized study of treatment duration and ribavirin dose. Ann Intern Med 2004;1(140):346–355. Hoofnagle JH. Therapy for acute hepatitis C. N Engl J Med 2001;345: 1495–1497. Jonas MM. Hepatitis C in children. In: Hepatitis C, Liang TJ and Hoofnagle JH, eds. Biomed Res Rep. San Diego, CA: Academic Press, 2000, pp. 389–404. Koziel MJ. Cytokines in viral hepatitis. Semin Liver Dis 1999;19:157–169. Krajden M, Ziermann R, Khan A, et al. Qualitative detection of hepatitis C virus RNA: comparison of analytical sensitivity, clinical performance, and workflow of the COBAS Amplicor HCV test version 2.0 and the HCV RNA transcription-mediated amplification qualitative assay. J Clin Micro 2002;40:2903–2907. Lau JYN, Mizokami M, Kolberg JA, et al. Application of six hepatitis C virus subtyping systems to sera of patients with chronic hepatitis C in the United States. J Infect Dis 1995;171:281–289. Lau JYN, Davis GL, Prescott LE, et al. Distribution of hepatitis C virus genotypes in United States patients with chronic hepatitis C seen in tertiary referral centers. Ann Intern Med 1996;124:868–876. Lau JYN, Standring DN. Development of novel therapies for hepatitis C. In: Hepatitis C, Liang TJ, Hoofnagle JH, eds. Biomed Res Rep 2000: San Diego, CA: Academic Press: pp. 453–467. Lauer GM, Walker BD. Hepatitis C virus infection. N Engl J Med 2001;345:41–52. Lee SG, Antony A, Lee N, et al. Improved version 2.0 qualitative and quantitative AMPLICOR reverse transcription-PCR tests for hepaitis C virus RNA: calibration to international units, enhanced genotype reactivity, and performance characteristics. J Clin Micro 2000;38: 4171–4179. Lindenbach BD, Rice CM. Evasive maneuvers by hepatitis C virus. Hepatology 2003;38:3669–3679. McHutchison JG, Gordon SC, Schiff ER, et al. Interferon alfa-2b alone or in combination with ribavirin as initial treatment for chronic hepatitis C. International Hepatitis Interventional Therapy Group. N Engl J Med 1998;339:1485–1492. McHutshinson JG, Poynard T. Combination therapy with interferon plus ribavirin for the initial treatment of chronic hepatitis C. Semin Liver Dis 1999;19:57–65. Mizokami M, Gobojori T, Lau JYN. Molecular evolutionary Virology - its application to the study of hepatitis C virus. Gastroenterology 1994;107:1181, 1182. Morishima C, Chung M, Ng KW, et al. Stengths and limitations of commercial tests for hepatitis C virus RNA quantification. J Clin Micro 2004;42:421–425. National Institutes of Health Consensus Develeopment Conference Statement: Management of Hepatitis C. Hepatology 2002;36(Suppl): S3–S20.

CHAPTER 52 / HEPATITIS C

Nelson DR, Lau JYN. Host immune response to hepatitis C virus. Viral Hepatitis Reviews 1996;2:37–48. Nelson DR, Marousis CG, Davis GL, et al. The role of hepatitis C virusspecific cytotoxic T-lymphocytes in chronic hepatitis C. J Immunol 1997;158:1473–1481. Nelson DR, Lau JYN. Pathogenesis of chronic hepatitis C virus infection. Antiviral Therapy 1998;3:25–35. Pawlotsky JM, Bouvier-Alias M, Hezode C, et al. Standardization of hepatitis C virus RNA quantitation. Hepatology 2000;32:654–659. Pawlotsky JM. Use and interpretation of hepatitis C diagnostic assays. Clin Liver Dis 2003;7:127–137. Poynard T, Bedossa P, Opolon P, for the OBSVIRC, METAVIR, CLINIVIR, and DOSVIRC groups. Natural history of liver fibrosis progression in patients with chronic hepatitis C. Lancet 1997;349: 825–832. Poynard T, Marcellin P, Lee SS, et al. Randomized trial of interferon alpha 2b plus ribavirin for 48 weeks or for 24 weeks versus interferon alpha 2b plus placebo for 48 weeks for treatment of chronic infection with hepatitis C virus. International Hepatitis Interventional Therapy Group (IHIT) Lancet 1998;352:1426–1432. Robert W, McMurray. Hepatitis C-Associated Autoimmune Disorders. Rheumatic Diseases Clinics of North America 1998;24:353–374. Reed KE, Rice CM. Overview of hepatitis C virus genome structure, polyprotein processing, and protein properties. Curr Top Microbiol Immunol 2000;242:55–84. Rossi SJ, Wright TL. New developments in the treatment of hepatitis C. Gut 2003;52:756, 757. Robertson B, Myers G, Howard C, et al. Classification, nomenclature, and database development for hepatitis C virus (HCV) and related

553

viruses: proposals for standardization. Arch Virol 1998;143: 2393–2403. Sarrazin C, Hendricks DA, Sedarati F, Zeuzem S. Assessment, by transcription-mediated amplification, of virologic response in patients with chronic hepatitis C virus treated with peginterferon α-2a. J Clin Micro 2002;39:2850–2855. Seeff LB, Buskell-Bales Z, Wright EC, et al. Long-term mortality after transfusion-associated non-A, non-B hepatitis. The National Heart, Lung, and Blood Institute Study Group. N Engl J Med 1992;327:1906–1911. Seeff LB. Natural history of hepatitis C. Am J Med 1999;107:10S–15S. Sherman KE, Rouster SD, Horn PS. Comparison of methodologies for quantification of hepatitis C virus (HCV) RNA in patients coinfected with HCV and human immunodeficiency virus. Clin Infect Dis 2002;35:482–487. Simmonds P, Alberti A, Bonino F, et al. Nomenclature of genotypes for hepatitis C virus (correspondence). Hepatology 1994;19: 1321–1324. Shiffman ML. Management of interferon therapy non-responders. Clin Liver Dis 2001;5:1025–1043. Thomas DL, Astemborski J, Rai J, et al. The natural history of hepatitis C virus infection: host, viral, and environmental factors. JAMA 2000; 34:809–816. Walker MP, Appleby TC, Zhong W, Lau JYN, Hong Z. Hepatitis C virus therapies: current treatments, targets and future perspectives. Antiviral Chem Chemother 2003;14:1–21. Zignego A, Bréchot C. Extrahepatic Manifestations of HCV Infection: Facts and controversies. J Hepatol 1999;31:369–376.

53 Molecular Diagnostics in Hepatitis B SCOTT BOWDEN AND STEPHEN LOCARNINI SUMMARY An estimated 400 million people are chronically infected with hepatitis B virus (HBV) and many will suffer serious liver disease as a consequence. HBV is an enveloped, partially double-stranded DNA virus, which replicates via reverse transcription of an RNA intermediate. The error rate of the reverse transcriptase generates a heterogeneous population of variants, endowing HBV with the ability to evade immune or antiviral selection pressure. Whereas HBV infection can be diagnosed by serological assays, the introduction of new antiviral agents to treat chronic infection requires monitoring by sensitive DNA amplification assays designed for quantification of HBV DNA in serum and liver tissue. Key Words: Antiviral therapy; drug resistance; HBV ccc DNA; HBV genome organization; HBV genotypes; HBV viral load.

INTRODUCTION More than 400 million people worldwide are chronically infected with the hepatitis B virus (HBV), making hepatitis B one of the most common infectious diseases. Of these chronic carriers, only 1–2% will annually spontaneously clear hepatitis B surface antigen (HBsAg). Ultimately, more than half of the HBsAg-positive patients will die of hepatocellular carcinoma (HCC) or liver failure. Host factors such as age, immunosuppression, and gender can contribute to patient outcome. Over the past decade there has been increasing interest in the role of virological factors such as load and specific mutations in particular regions of the viral genome on the clinical course and outcome of persistent infection. HBV is an enveloped, partially double-stranded DNA virus, which is the prototype member of the family Hepadnaviridae. The virus replicates its DNA genome via an RNA intermediate, an unusual strategy that generates a heterogeneous population of genetic variants during the normal course of infection and provides enormous potential for adaptation to changing environments and response to particular selection pressures. This property endows the virus with the ability to evade host immunity and resist conventional antiviral strategies. However, the extreme genetic economy of the HBV genome, achieved by the use of overlapping reading frames, is a major constraint on HBV evolution and as such could be exploited to improve existing therapeutic approaches, in particular by the use of combination treatments targeting different parts of the virus life cycle. Studies on the pathogenesis of chronic hepatitis B From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

(CH-B) have revealed a complex interaction between the virus, the hepatocyte, and the host’s immune response. This chapter describes the organization of the HBV genome, the role of the proteins it encodes, and the common variants produced by exogenous selection pressure. New antiviral therapies available for the treatment of CH-B are discussed with the molecular technologies necessary to assess their efficacy. Finally, promising techniques for investigating the intricacies of the HBV life cycle in the liver are highlighted.

THE VIRUS AND ITS GENOME VIRION MORPHOLOGY Three types of virus-associated particles are found in serum of an HBV-infected individual. (1) HBV virions measuring 42 nm in diameter, comprising an outer envelope formed by the HBsAg. This envelope surrounds an inner nucleocapsid consists of the hepatitis B core antigen (HBcAg) that contains the packaged viral genome and an endogenous viral DNA polymerase; (2) abundant spherical particles of approx 22 nm in diameter that are in a 104–106-fold excess over the HBV virions; and (3) filamentous structures of approx 20–22 nm in diameter and of variable length. The latter two forms of subviral particles consist of virus-derived glycoproteins (GP) of HBsAg and do not contain the HBV genome and are thus not infectious. The purified 22 nm particles are highly immunogenic and when administered to normal individuals in the form of the hepatitis B vaccine, are able to induce a neutralizing anti-HBs antibody response. HBV GENOME The HBV genome is a circular, partially double-stranded relaxed circular DNA molecule of approx 3200 nucleotides (nt) in length (Fig. 53-1). The two linear DNA strands are held in a circular configuration by a 226-bp cohesive overlap between the 5′ ends of the two DNA strands that contain two 11-nt direct repeats called DR1 and DR2. All known complete HBV genomes are gapped, nicked, and circular, comprising 3181–3221 nt depending on the genotype (Table 53-1). Within the virion, the minus strand of the genomic DNA has a fixed length with defined 5′ and 3′ ends, and a terminal redundancy of 8–9 nt. The minus strand is not a closed circle and has a nick near the 5′ end of the plus strand. The viral DNA polymerase is covalently bound to the 5′ end of the minus strand. The 5′ end of the plus strand contains an 18-base long oligoribonucleotide, which is capped in the same manner as mRNA. The 3′ end of the plus strand is not at a fixed position so most viral genomes contain a single-stranded gap region of variable size from 20 to 80% of the genomic length, which may be filled in by the endogenous viral DNA polymerase.

554

CHAPTER 53 / MOLECULAR DIAGNOSTICS IN HEPATITIS B

Figure 53-1 The organization of the hepatitis B virus genome. The inner circles represent the complete minus-sense strand of the genomic DNA with the viral polymerase attached to the 5′ end and the incomplete plus-sense strand with the capped RNA oligonucleotide at the 5′ end. The direct repeat sequences are designated DR1 and DR2. The four open reading frames (core [Pre-C/C], envelope [Pre-S/S], polymerase [POL], and X) are shown as the discrete thicker arrows around the genomic DNA. The outer thin circular lines depict the viral RNA transcripts, the core and pgRNA (3.5 kb), the Pre-S mRNA (2.4 kb), the S mRNA (2.1 kb) and the X mRNA (0.7 kb). Note that the transcripts all end at a common polyadenylation site (AAA).

The minus strand of HBV encodes four major open reading frames (ORFs) that carry all the protein-coding capacity of the virus: the envelope (pre-S1, pre-S2, and S), the core, the X protein and the polymerase. These overlap in a frame-shifted manner with one another, so that the minus strand is read 1.5 times. The longest ORF encodes the viral polymerase (Pol) whereas the ORF for the envelope (Pre-S/S) gene is completely located within the Pol ORF. The ORF for the core (Pre-Core/Core–Pre-C/C) and X genes partially overlap the Pol ORF. HBV is able to encode more than one protein from an ORF by using multiple internal AUG or start codons, creating additional start sites for protein biosynthesis. Thus, nested sets of proteins with different N-termini are synthesized. This gene overlap, although genetically economical, places constraints on the mutation and evolution rates. Regulatory Elements Controlling Viral Replication Multiple regulatory elements involved in the regulation and expression of each individual HBV gene are located throughout the whole length of the viral genome. Because every region of the HBV genome is a protein-encoding sequence, all cis-acting regulatory elements reside within gene sequences. HBV genomic expression is regulated by two enhancers (EnhI and II), four promoters, a glucocorticoid response element, a negative regulatory element, and a CCAAT element. The four promoters, consists of the basal core promoter (BCP), the Pre-S1, Pre-S2/S, and X promoters, control the corresponding four major mRNA species of 3.5 kb (Pre-C/C and pregenomic [pg] RNA), 2.4 kb (Pre-S1 mRNA), 2.1 kb (Pre-S2/S

555

mRNA), and 0.7 kb (X mRNA) (see Fig. 53-1). HBV transcription is dependent to a large extent on liver-enriched transcription factors. All of these mRNA species are capped, unspliced, and share a common polyadenylation signal. Three other regulatory elements that also control HBV replication are located on the viral RNAs: the polyadenylation signal, posttranscriptional regulatory element, and the encapsidation signal (ε). The pgRNA serves as the template for reverse transcription. Regulation of viral gene expression also occurs at the level of translation. The pgRNA serves as the mRNA not only for the viral core protein but also for the viral polymerase, which initiates from an AUG located in the distal portion of the core gene, although not in the same reading frame as core. Reading of the polymerase ORF appears to be inefficient compared to that of the core ORF. However, because core particles are assembled from 240 core subunits and only one or perhaps two polymerase proteins, pgRNA may serve as an mRNA for the translation, on average, of approx 200–300 core polypeptides before allowing the translation of a polymerase polypeptide. Because the polymerase preferentially binds to the 5′ end of its own mRNA to initiate reverse transcription and packaging, synthesis of the polymerase is probably sufficient to stop further translation of the pregenome. Pre-S/S ORF HBsAg comprises the small (SHBs), medium (MHBs), and large (LHBs)-sized proteins (Fig. 53-2), all of which exist in two forms differing in their degree of glycosylation. N-linked glycosylation and glucosidase processing are necessary for virion, but not subviral particle secretion. The 22-nm spherical form contains approx 89% of SHBs, 10% of MHBs, and 1% of LHBs. The filamentous forms consist of the same ratio of proteins as the virion envelope, comprising approx 70% of SHBs, 10% of MHBs, and 20% of LHBs. The 2.4-kb mRNA transcript encodes the LHBs whereas the 2.1-kb mRNA transcript encodes the MHBs and SHBs (see Fig. 53-2). The synthesis and assembly of the viral envelope, the small particles, and filaments occur in the endoplasmic reticulum (ER) membranes with the resultant assembled viral proteins budding into the ER lumen. The SHBs domain is 226 amino acids long and is the most abundant protein of the three HBV-associated particles. The SHBs are found in glycosylated (GP27) and nonglycosylated (P24) forms. They contain a high number of cysteine residues that are crosslinked with each other, forming a conformational loop that is the major antigenic determinant of the HBsAg (from amino acid residues 98–170), referred to as the major hydrophilic region. The most characterized mutation associated with vaccine escape, sG145R, is located within this region. The MHBs containing the Pre-S2 domain are a minor component of the virion or subviral particles and consists of SHBs with a 55 amino acid N-terminal extension. The MHBs are either doubly or singly glycosylated (GP36 and GP33) and are required for virus secretion. However, this protein is not required for infectivity or virus assembly. The MHBs are considerably more immunogenic than SHBs, and Pre-S2-containing HBs particles generated from animal cell lines have been used in some countries as a prophylactic vaccine. The LHBs contain an additional 108 or 119 amino acids (depending on the subtype/genotype; see Table 53-1) in comparison to MHBs at its N terminus. The LHBs are mainly glycosylated (GP42) and myristylated, and are essential for viral infection, assembly, and release. A small amount of this protein is nonglycosylated (P39). A number of B- and T-cell epitopes have been mapped to LHBs, MHBs, and SHBs. The Pre-S1 region contains

556

SECTION VI / GASTROENTEROLOGY

Table 53-1 Overview of the Eight Major Genotypes of HBV HBV Proteins (no. of amino acids)

Frequency of mutationa

Genome length (nt)

PreS1b

Pol

Core

PC Uncommon (C1858) Common (T1858) Common Low

Common

Genotype

Subtype

BCP

A

adw2, ayw1

3221

119

845

185

B

adw2, ayw1

3215

119

843

183

Bj Ba

adw2, aywl adw2, aywl

3215 3215

119 119

843 843

183 183

C

adw2, adr, ayr

3215

119

843

183

Common T/C1858

Common

D

ayw

3182

108

832

183

Common

E F

ayw adw, ayw

3212 3215

118 119

842 843

183 183

G

adw

3248

108

842

195

H

adw

3215

119

843

183

Common T1858 ND Uncommon (C1858) Very common (insertion) ND

Global distribution

ND

Western Europe, USA, Central Africa, India Japan, Taiwan, Indonesia, China, USA Japan China, Taiwan, Indonesia, Vietnam East Asia, Taiwan, Korea, China, USA, Japan, Polynesia Mediterranean region, India, USA West Africa Central and South America, Polynesia USA, Europe

ND

Central, South America

Common Uncommon Uncommon

ND ND

HBV, hepatitis B virus; ND, not described; Pol, polymerase. aPC, Precore mutations such as G1896A; BCP, basic core promoter mutations such as A1762T, G1764A; common (up to 50% of isolates); uncommon (less than 10% of isolates); very common (most isolates). bPre-S2, 55aa; S, 226aa.

Figure 53-2 Pre-S1/Pre-S2/S open reading frames. The Pre-S/S open reading frame can be divided into Pre-S1, Pre-S2, and S domains by three in-frame start codons. The proteins encoded by the open reading frame are expressed from two mRNA transcripts. The larger 2.4-kb Pre-S1 mRNA leads to expression of the large-sized proteins, whereas the Pre-S2/S 2.1-kb mRNA encodes medium-sized proteins and small-sized proteins. MHBs are translated using the first start codon and SHBs from the internal start codon. LHBs can be detected as a nonglycosylated species (P39) but its major form is glycosylated (GP42). SHBs are found in both glycosylated (GP27) and nonglycosylated (P24) forms, whereas MHBs are glycosylated either at one (GP33) or two (GP36) sites.

two important epitopes. One at amino acid residues 58–100 is thought to be recognized by antibodies involved in viral clearance. The other, at residues 21–47 is thought to be the hepatocyte binding domain. The N terminus of Pre-S2 can bind to the fibronectin found in liver sinusoids, providing some degree of host tissue specificity. Interestingly, most of the Pre-S2 is not essential for viral replication or virion infectivity whereas only residues 109–113 are indispensable.

There are four different alleles within the S gene: r, d, y, and w. The determinants d/y and w/r are mutually exclusive, thus forming two allelic groups. The “a” determinant (amino acid residues 107–149) is part of all HBs subtypes and can be divided into two alleles that differ at amino acid 126 being either threonine or isoleucine. The main subtypes are designated ayw, ayr, adw, and adr. The clinically most important determinant of HBsAg is the “a” determinant (Fig. 53-3). During the natural course of infection,

CHAPTER 53 / MOLECULAR DIAGNOSTICS IN HEPATITIS B

557

Figure 53-3 Schematic model of hepatitis B surface antigen. The two loops of the “a” determinant are stabilized by disulphide bridges (s–s) between cysteine residues. The C-terminal end of the protein has two alpha helices (shown as cylindrical structures). Serotype-specific epitopes are located in the “a” determinant. A point mutation that converts K122 to R122 (K122R) changes the d subtype to y and a similar mutation at K160R changes subtype w to r. These produce the four major subtypes of hepatitis B virus—adw, adr, ayw, and ayr. The location of the glycine residue that is modified to arginine (G145R) in virus associated with vaccine escape is shown. (Modified from Wallace et al., 1997.)

antibodies against all the determinants are produced, but only the antibodies against the “a” determinant are protective against a challenge with other HBV subtypes. The “antigenic” conformation of the “a” determinant is regarded as a two-loop structure stabilized by disulphide bridges between Cys-107 and Cys-138 (first loop) and Cys-139 and Cys-147 (second loop) whereas Cys-137 binding to Cys-149 stabilizes the interaction (see Fig. 53-3). If this conformation is altered, then anti-HBs produced previously against the native “a” determinant are no longer protective. Pre-C/C ORF The Pre-C/C ORF encodes the core protein (P21), which is the major polypeptide of the nucleocapsid and is expressed as the HBcAg (Fig. 53-4). The HBc protein is 183, 185, or 195 amino acids long, depending on the genotype of the virus (see Table 53-1). ORF C is preceded by a short, upstream, inphase ORF known as the precore region from which the soluble hepatitis Be antigen (HBeAg) is synthesized (see Fig. 53-4). Both the core (HBcAg) and precore (HBeAg) proteins are targets for immune-mediated viral clearance mechanisms. Translation of the precore region from the 3.5-kb Pre-C/C mRNA results in a 25-kDa polypeptide. The first 19 amino acids of the precore protein represent a secretion signal that allows for the translocation of the precore protein into the lumen of the ER (see Fig. 53-4). The signal sequence is cleaved off by a host cell signal peptidase and the protein is secreted through the ER and Golgi apparatus. A further modification of the C terminus results in the secretion of a heterogeneous population of proteins (15–18 kDa), which is serologically defined as HBeAg. The precore protein is not essential to viral replication and thus can be regarded as an “accessory protein.” The HBc protein is translated in the cytosol from the 3.5-kb pg mRNA after which these proteins initially form dimers, followed by multimerization to form the nucleocapsid. The HBc protein has

been crystallized and the nucleocapsid protein has been mapped using cryoelectron microscopy. The multimerization of the HBc protein can occur independently of the encapsidation of the pgRNA–Pol complex. The core protein can be divided into two major domains, the N-terminal assembly domain (up to amino acid 144) and the C-terminal arginine-rich region involved in RNA–DNA binding interactions. The core protein can self-assemble into capsids, and two icosahedral shells of different sizes have been observed: particles with a T = 3 symmetry containing 90 homodimers of 32 nm, and particles with a T = 4 symmetry consisting of 120 homodimers of 36 nm. Pol ORF The Pol gene is the longest ORF, spanning almost 80% of the genome, and overlaps the three remaining ORFs (see Fig. 53-1). Codons 834–845 in the Pol ORF have sequence homology to known reverse transcriptases (rts) and most parts of the ORF are essential for viral replication. The Pol protein is translated from the pgRNA (see Fig. 53-4). The 90-kDa product of the Pol ORF is a multifunctional protein that has at least four domains: (1) the N-terminal domain that encodes the terminal protein, which is covalently linked to the 5′ end of the minus strand of virion DNA. This part of the Pol ORF is necessary for priming minus-strand synthesis; (2) an intervening domain with no specific recognized function, referred to as the spacer or tether region; (3) the third domain that encodes the RNA- and DNA-dependent DNA polymerase activities, i.e., the rts; and (4) the C-terminal domain that encodes RNase H activity that cleaves the RNA in the RNA/DNA hybrids during the reverse transcription process. The terminal protein’s role in protein priming reverse transcription includes the provision of the substrate tyrosine at amino acid 63 of the HBV Pol for the formation of the covalent bond between the enzyme and the first nucleotide (G) of the minus-strand DNA. The DNA polymerase domain contains the amino-acid motif YMDD,

558

SECTION VI / GASTROENTEROLOGY

Figure 53-4 Core (Pre-C/C) open reading frame. The pregenomic RNA acts as a bicistronic mRNA. It codes for the polymerase protein and, by using an internal start codon, it also produces a translation product corresponding to the core protein (hepatitis B core antigen or P21). Multiple copies of the core protein can self-assemble into particles. Initiation of translation from the precore mRNA leads to production of a precore polypeptide (P25). The N terminus of this polypeptide has a signal sequence that targets it to the secretory pathway. The signal sequence is cleaved off by a host cell signal peptidase and the C terminus of the polypeptide is further cleaved by Golgi proteases to produce a heterogeneous population of proteins (P16, P18, and P20) known collectively as hepatitis Be antigen.

which is essential for rt activity. The RNase H domain, as well as being involved in degrading the RNA template, has other functions, including playing a role in viral RNA packaging, optimizing the priming of minus-strand DNA synthesis and in elongation of the minus-strand viral DNA. Several cytotoxic T-lymphocyte (CTL) epitopes have been mapped to the polymerase protein. Six CTL epitopes within the rt or the RNase H domains of the viral polymerase have been described in patients with acute hepatitis. The epitope (GLSRYVARL) appears most relevant for interferon responsiveness and viral clearance. It has been suggested that Ser-457, Ala-461, or Arg-462 are contact sites for the T-cell receptor. If these critical amino acids are changed by mutation, then the T-cell receptor binding activity will be reduced and the CTL-mediated immune response attenuated and/or blunted. The advent of nucleoside per nucleotide analog treatment has resulted in the outgrowth of otherwise minor quasispecies with mutations in the HBV Pol gene. Antiviral resistance to lamivudine (LMV) has been mapped to the YMDD locus in HBV Pol. Using the newly described nomenclature system, the major mutations within the rt gene that were selected during LMV therapy results in changes designated rtM204I/V (domain C, methionine at position 204 replaced by isoleucine or valine) with or without rtL180M (domain B). The domain B mutation rtL180M is also selected during famciclovir treatment. Recently, the development of adefovir (ADV)-resistant HBV with changes in the domain D (rtN236T) and domain B (rtA181T/V) of the viral polymerase has been described as has entecavir resistant virus with changes in domains B and C (rtS184G and rtS202I). X ORF The X ORF produces a 0.7-kb mRNA transcript encoding a polypeptide 154 amino acids in size (HBx) with a predicted molecular weight of 16–19 kDa (see Fig. 53-1). When first identified, the function of the X protein was unknown and thus, the X designation was assigned. HBx is the second accessory protein

of HBV and is dispensable for virus production in vitro, but is a critical component of the infectivity process in vivo. The C-terminal portion of the protein seems to be associated with the transactivating properties of X whereas the N terminus appears to act as a negative regulator of transactivation. HBx can act in the cytosol and in the nucleus. HBx behaves as a transcriptional transactivator for a number of viral and cellular gene promoters through direct interaction with transcription factors such as RPB5 subunit of RNA polymerase II, TATA-binding protein, and ATF/CREB (nuclear pathways). The cytoplasmic pathways are divided into protein kinase C (PKC)-dependent and -independent mechanisms. HBx interacts directly with inactive cytosolic PKC thereby activating it. In the PKC-independent cytosolic pathway, HBx is involved in the activation of signal transduction pathways, such as the Ras/Raf/MAP kinase cascade. The Ras pathway leads to the phosphorylation and activation of the transactivating domain of c-jun. HBx is a multifunctional viral regulator that modulates transcription, cell responses to protein degradation, and signaling pathways. Such regulation of viral and cellular genes affects viral replication and viral proliferation, directly or indirectly. HBx has been implicated in the development of HCC and affects cell cycle checkpoints, cell death, and carcinogenesis. Although the precise mechanism remains uncertain, the HBx-associated transactivation activity may lead to alterations in cellular gene expression that contribute to cell transformation. HBx can bind to and inactivate the transcription factor and tumor suppressor, p53. In addition, the HBx can dampen the immune response by interfering with the ubiquitin-proteolysis pathway following binding with a 26S proteasome complex.

VIRAL GENOTYPES There are eight recognized genotypes of HBV designated A to H, which vary by 8% at the nucleotide level over the entire genome.

CHAPTER 53 / MOLECULAR DIAGNOSTICS IN HEPATITIS B

These HBV genotypes have unique insertions or deletions. For example, HBV genotype A varies from the other genotypes by an insertion of 6 nt in the terminal protein region of the polymerase gene, whereas HBV genotype D has a 33-nt deletion in the spacer region of the polymerase gene, and HBV genotypes E and G have a 3-nt deletion in the same region (see Table 53-1). HBV genotype G also has a 36-nt insertion in the N terminus of the core gene. Genotype H has been defined and is most similar to genotype F. The HBV genotype designation is based on the entire genomic sequence. Thus, it is more reliable than the serological subtype nomenclature used previously, which was based on the immunoreactivity of particular antibodies to a limited number of amino acids in the envelope protein. The relationship between the four major HBV subtypes (adw, adr, ayw, and ayr) and genotypes has been determined (see Table 53-1). Important pathogenic and therapeutic differences do exist among HBV genotypes. For example, genotype C is associated with more severe liver disease than genotype B in Taiwan, whereas genotype D is associated with more severe liver disease than genotype A in India. Genotypes C and D are associated with a lower response rate to interferon therapy compared with genotypes B and A. Recombination between two HBV genotypes has been reported for genotypes B and C and genotypes A and D generating even more diversity. Furthermore, there is evidence for the widespread distribution of related hepatitis viruses in primates, implying that these viruses may have a common origin and that cross-species transmission of hepadnaviruses may have occurred among hominoids.

COMMON HBV MUTANTS Viral and host factors, as well as exogenous selection pressures, typically define the predominate species in an infected individual. Exogenous pressures include nucleoside per nucleotide analog treatment, as well as immune-based intervention such as hepatitis B immunoglobulin, and vaccination. The presumably immune-based selection pressure that results in the reduction or loss in HBeAg and eventual elimination of HBsAg are probably responsible for the selection of particular mutants such as those associated with CH-B. Viral rts lack proofreading function and are inherently errorprone, which ensures that HBV populations exist in the host as heterogeneous mixtures known as quasispecies. The HBV mutation frequency has been estimated to be approx 1.4–3.2 × 10–5 nt substitutions/site/yr, approx 10-fold higher than other DNA viruses. The magnitude and rate of virus replication are also important in the process of mutation generation, with the total viral load in serum frequently approaching 1011 virions/mL. Most estimates place the mean half-life of the serum HBV pool at approx 1–2 d, so that the daily rate of de novo HBV production may be as great as 1011 virions. The high viral loads and turnover rates, coupled with poor replication fidelity, all influence mutation generation and the extent of the HBV quasispecies pool. However, these are less than for other retroviruses, mainly because of the constraints imposed by the overlapping reading frames. MUTATIONS IN THE BCP REGION AND PRECORE GENE Two major groups of mutations have been identified that result in reduced or blocked HBeAg expression. The first major group includes a translational stop codon mutation at nt 1896 (codon 28: TGG; tryptophan) of the precore gene. The single base substitution (G-to-A) at nt 1896 gives rise to a translational stop codon (TGG to TAG; TAG = stop codon) in the second last codon (codon 28) of the precore gene located in the ε structure. The ε structure is a highly

559

conserved stem loop structure critical in viral replication, with the nucleotide G1896 forming a bp with nt 1858 at the base of the stem loop. In HBV genotypes B, D, E, G, and some strains of genotype C, the 1858 is a thymidine (T) (see Table 53-1). Thus, the stop codon mutation created by G1896A stabilizes the ε structure (T–A). In contrast, the precore stop codon mutation is rarely detected in HBV genotype A, F, and some strains of HBV genotype C, as the nucleotide at position 1858 is a cytidine (C), maintaining the preferred Watson–Crick basepairing (G–C). Other less common mutations have been found such as precore codon 17 (G1862T), which can alter the processing of the precore proteins in the Golgi apparatus. Interestingly, the mutation at the precore codon 29 (G1899A) has been found in isolation or in association with the G1896A or G1862T, but its significance is unclear as it has no translational effect. The second major group of mutations affects the BCP at nt1762 and nt1764 resulting in a transcriptional reduction of the Pre-C/C mRNA. Mutations in the BCP, such as A1762T plus G1764A, may be found in isolation or in conjunction with precore mutations, depending on the genotype. The double mutation of A1762T plus G1764A results in a decrease in HBeAg production, but not disappearance and interestingly, an increase in viral load. In general, this pattern of mutation is often found in genotype A-infected individuals. BCP mutants display reduced binding of liver-specific transcription factors resulting in less Pre-C/C mRNA transcripts and, thus, less precore protein. However, this mutation does not affect the transcription of pgRNA or the translation of the core or polymerase protein. By removing the inhibitory effect of the precore protein on HBV replication, the BCP mutations appear to enhance viral replication by suppressing Pre-C/C mRNA relative to pgRNA. MUTATIONS IN THE CORE GENE The core protein can be divided into two major domains, the N-terminal domain up to amino acid position 144, and the functionally important argininerich C-terminal domain. The C-terminal domain (up to amino acid 164) is required for encapsidation of the pgRNA and seems to stabilize the capsid by protein–nucleic acid interactions. The remaining C terminus (up to amino acid 173) appears important for the synthesis of plus-strand DNA and subsequent genomic replication. The core protein contains a number of important B-cell and CTL epitopes. During the elimination phase of CH-B, escape mutations within those epitopes are readily selected. These “hot spots” have been linked to major CTL (aa 18–30) and T-helper (TH) cell (aa 50–70) regions and to two B-cell (HBc/eI and HBc/e2) epitopes at residues 75–90 and 120–140, respectively. The HBcAg and HBeAg are highly cross-reactive at the T-cell level. The HBeAg probably presents immunogenic epitopes to the T cells, thus “protecting” the HBcAg-expressing hepatocytes against the immune system. After selection of the HBeAg-negative mutant(s), the epitopes of the HBcAg come under intense pressure from the immune system. Thus, the frequency of core gene mutations is typically associated with the presence of precore stop codon mutations, HBeAg-negativity, and active liver disease. MUTATIONS IN THE X GENE The X-ORF overlaps the C terminus of the Pol gene and the N terminus of the C gene (see Fig. 53-1). Depending on the extent of the mutation(s) within the X region, mutations can affect three genes at once. X-region deletions can create fusion proteins between the pol and the core proteins (PCproteins) or between the pol and a 3′-truncated HBx (PX-proteins). Typically, mutations in the X region involve the regulatory elements that control replication. Because the BCP encompasses nts

560

SECTION VI / GASTROENTEROLOGY

1742–1802 and overlaps with the X gene in the concomitant reading frame, the A1762T plus G1764A BCP mutations also cause changes in the X gene at xK130M and xV131I. Additionally, nearly all deletions/insertions in the BCP shift the X-gene reading frame and lead to the production of truncated X proteins. These shortened X proteins lack the domain in the C terminus (aa 130–140) that is required for the transactivation activity of HBxAg. The two direct repeat sequences, DR-1 and DR-2, lie within the BCP sequence and are considered important in reverse transcription. Between DR-1 and DR-2 lies the cohesive end region of the HBV genome and this appears to be a preferred site for viral integration into host DNA. Interestingly, the cohesive end region near DR-1 contains a preferred cleavage site for topoisomerase I that can catalyze illegitimate recombinations between HBV and cellular DNA; following viral integration, deletions in this region have been found. MUTATIONS IN THE ENVELOPE GENE The Pre-S sequences exhibit the highest heterogeneity of the HBV genome. Point mutations, deletions, and genetic recombinations within the Pre-S genes have been identified in HBV DNA sequences analyzed from the sera of chronic carriers. Deletions within the Pre-S region can affect the S promoter causing an imbalance in envelope assembly resulting in intracellular retention and accumulation of HBsAg. Viral genomes, which cannot synthesize Pre-S2 proteins, occur frequently, and can be the dominant virus populations in chronic carriers. The Pre-S2 region overlaps the spacer region of the Pol protein that is not essential for enzyme activity. It has been speculated that the loss of the highly immunogeneic Pre-S2 protein reflects an immunological escape mechanism. All hepatitis B vaccines contain the major HBsAg, and an immune response to the major hydrophilic region located from residue 99–170 induces protective immunity. Mutations within this epitope have been selected during vaccination and also following treatment of liver transplant patients with hepatitis B immunoglobulin. Most isolates have a mutation from glycine to arginine at residue 145 of HBsAg (sG145R) or aspartate to alanine at residue 144 (sD144A). The sG145R mutation has been associated with vaccine failure. Mutations that abolish or alter the two-loop structure of the “a” determinant can affect anti-HBs binding. Such effects include changes in the hydrophobicity, the electrical charge, the acidity of the loops, addition of possible additional N-glycosylation sites (glycine-130 to Asn), or changing the stability of the disulphide bridge (Cys-147 to Gly). Finally, mutations near the “a” determinant, such as an insertion of eight amino acids between Thr123 and Cys-124, two amino acids between Cys-121 and Lys-122 or of arginine between Pro-120 and Cys-121 can disturb the secondary structure of the “a” determinant. POLYMERASE MUTATIONS: ANTIVIRAL DRUG RESISTANCE The advent of nucleoside per nucleotide analog treatment has resulted in the outgrowth of otherwise minor quasispecies with mutations in the HBV Pol gene. Antiviral resistance to LMV has been mapped to the YMDD locus, the catalytic site in the C domain of HBV polymerase, whereas resistance to ADV is associated with mutations in the D domain of the enzyme. Mutations within the rt gene that were selected during LMV therapy cause changes that are designated rtM204I/V/S (Domain C) +/–rtL180M (Domain B), whereas rtN236T (D-domain) and rtA181T are the major changes associated with ADV-resistant virus. Lamivudine Resistance LMV resistance increases progressively during treatment at rates between 14 and 32% annually,

approaching 100% after 48 mo treatment. Factors that increase the risk of development of resistance include high pretherapy serum HBV DNA levels and ALT levels and incomplete suppression of viral replication. LMV resistance does not confer cross-resistance to ADV. Mutations that confer LMV resistance decrease in vitro sensitivity to LMV from at least 20-fold to greater than 100-fold. The rtM204I substitution has been detected in isolation, but rtM204V and rtM204S are found only in association with other changes in the B or A domains. Numerous other secondary changes in the rt sequence also occur in conjunction with rtM204V/I/S, and some of these are probably compensatory. Lamivudine resistance is mainly owing to steric hindrance resulting from the replacement of the methionine residue in the YMDD motif with either of the branched amino acids valine or isoleucine, which decrease the ability of the polymerase to bind the nucleoside analog in the deoxynucleotide triphosphate binding pocket relative to the natural substrate. Adefovir Dipivoxil Resistance HBV resistance to ADV occurs less frequently (approx 2% after 2 yr, 5% after 3 yr, and 16% after 4 yr) than resistance to LMV. ADV resistance is conferred by substitution of threonine for asparagine at codon 236 (rtN236T), which is located in the D domain of the polymerase a change that does not significantly affect sensitivity to LMV, and/or the rtA181V/T change, which is located in the B domain of the enzyme. The mechanism of resistance by N236T has been determined to be indirect perturbation of the triphosphate binding site whereas indirect steric hindrance accounts for resistance owing to rtA181V/T. Both mutations, either together or separately, decrease sensitivity to the drug by less than 10-fold. Entecavir Resistance Resistance to entecavir has been observed in two patients who had already failed to respond to LMV. The viral load increase was associated with clinical deterioration and rising levels of serum ALT. Sequencing of the HBV pol gene at the time of viral breakthrough confirmed LMV resistance and identified several unique mutations associated with entecavir resistance. • Patient A: rtI169T (B-domain) and rtM250V (E-domain). • Patient B: rtT184G (B-domain) and rtS202I (C-domain). These changes have been confirmed to be associated with entecavir resistance by using in vitro phenotypic assays. Telbivudine Resistance Telbivudine is the β-L isomer of thymidine. After 24 wk of therapy, it results in a 4–6 log drop in HBV DNA levels. Resistance occurred at the rate of 5% per annum and resulted in the selection of the rtM204I mutation, indicating cross-resistance with LMV. Pol-Env Link The envelope gene lies completely within the Pol gene but in a frame-shifted manner. Thus, mutations in one gene have the potential to affect the product(s) of the other gene. For example, some mutations in the Pol gene that confer drugresistance affect the immunogenicity of the envelope gene products (Fig. 53-5). Conversely, immune selection can influence antiviral drug sensitivity. Cases of transmission of drug-resistant HBV have been reported. The clinical and public health importance of drug resistance is clear; improved treatment strategies are urgently required to ensure prevention of resistance. Pathogenicity of Drug Resistant HBV: Role of Compensatory Mutations Compensatory mutations that partly or wholly restore the level of viral fitness have been documented during therapy for HIV infections and similar scenarios have been described for HBV.

CHAPTER 53 / MOLECULAR DIAGNOSTICS IN HEPATITIS B

561

Figure 53-5 The Pol–Env link. The genomic organization of HBV is such that the Pre-S/S open reading frame (ENV) is completely overlapped by the polymerase (POL) open reading frame. Resistance to nucleoside/nucleotide analogs is caused by changes in the reverse transcriptase domain of the polymerase and this has the potential to select viruses with changes in the overlapping S gene, which may affect the a determinant. Similarly, changes to the S gene caused by immune selection may have the ability to modify drug sensitivity by altering the Pol gene and affecting the RT domain.

A high frequency of the co-occurrence of rtL80V/I (domain A) and rtL180M has been observed in conjunction with the M204V/I changes that confer LMV resistance in treated Japanese patients (genotype C). The presence of double and triple changes was associated with higher viral loads, increased LMV resistance, and disease exacerbation. Longitudinal studies showed that the mutations responsible for these sequence changes occurred almost simultaneously, just before viral breakthrough and also that the mutants were displaced by wild-type genotype C HBV after completion of therapy. Similar observations have been recorded for liver transplant patients who developed life-threatening HBV recurrence. HBV isolates from these patients had compensatory mutations that enhanced their in vitro replication efficiency in the presence of LMV. Even greater enhancement and drug dependency occurred when mutations resulting in sG145R or sP120T, key changes in the envelope protein, were also present. Mutations that abolish or decrease the expression of HBeAg affect replication efficiency. HBeAg-negative CH-B, which is common in the Mediterranean region, is usually characterized by lower serum HBV DNA levels than those found during HBeAg-positive CH-B. A group of HBeAg-negative CH-B patients have been studied who were infected with genotype D HBV that contained both BCP and precore stop codon mutations. Development of LMV resistance in these patients was associated with relatively rapid increases in viremia, frequently culminating in biochemical breakthrough, severe hepatic flares, and disease progression. In vitro studies confirmed that the presence of the typical precore mutation (G1896A) could compensate for the replication deficiency in LMVresistant HBV quasispecies. Several case studies have confirmed that drug resistant HBV mutants are capable of causing severe, even fatal, disease, challenging the notion that drug resistant HBV mutants, and other minority quasispecies are benign. These reports and the observed enhanced replication of HBV in the presence of LMV as well as the precore mutant have led to the conclusion that multiple mutations are continually selected

during inadequate antiviral therapy and that these can act as compensatory mutations having the potential to restore replication competence.

MOLECULAR DIAGNOSTICS—SERUM COMPARTMENT The natural history of CH-B can be divided into replicative and nonreplicative stages. As there is no practical measure of HBV infectivity, assays of serum HBV DNA by sensitive molecular methods have provided an indication of the level of viral replication. Also, the detection of particular viral proteins in an infected liver by immunohistochemical techniques can indicate the presence of active replication. Development of potent chemotherapy to control CH-B, primarily using nucleoside per nucleotide analogs, has made it necessary to use more sophisticated methods for monitoring their effectiveness than traditional biochemistry and serology markers. Sensitive nucleic acid tests (NATs) are available to detect HBV DNA, employing signal or target amplification technologies, and each has advantages and disadvantages. Generally, the signal amplification assays have reduced sensitivity compared to the target amplification assays, which in turn suffer from a restricted upper range. Furthermore, the various assays may express the results in different units making comparisons difficult and highlighting the need for standardization. New generation amplification assays have become commercially available that have been standardized to the World Health Organization international reference standard. These include an improved chemiluminescent signal amplification assay (Bayer VERSANT HBV 3.0 test) and assays exploiting the greater dynamic range intrinsic to kinetic or real-time PCR (Roche COBAS TaqMan HBV test and Artus RealArt HBV LC PCR reagents) (Fig. 53-6). BAYER VERSANT HBV 3.0 TEST This amplification assay employs a series of sequential nucleic acid hybridization steps to produce a chemiluminescent signal that is proportional to the amount of HBV DNA in the original sample. The first step involves denaturation of the partially double-stranded HBV DNA so that

562

SECTION VI / GASTROENTEROLOGY

Figure 53-6 Dynamic range (expressed in IU/mL) of commercially available quantitative hepatitis B virus DNA assays. The Bayer VERSANT 3.0 bDNA assay relies on signal amplification whereas the Roche TaqMan and Artus RealArt assays rely on target amplification by real-time polymerase chain reaction.

target synthetic DNA oligonucleotides (probes) can bind to singlestranded portions of the HBV DNA. The target probes are a mix of two separate oligonucleotides and each oligonucleotide has two binding regions. Both have a region for specific hybridization to HBV DNA; one probe also has region to hybridize to a capture probe that is immobilized on the plastic surface of a microwell plate and the other probe has a portion complementary to a highly branched synthetic oligonucleotide (hence bDNA), which is subsequently added to the mix. The branches of the bDNA oligonucleotides are designed to have repeat sequences that allow multiple hybridization of another oligonucleotide (label probe) that is conjugated with alkaline phosphatase. On addition of a chemiluminescent substrate, a signal is generated and by comparison to a standard curve, a viral load can be determined. The assay has a wide dynamic range reporting in both copies/mL and International Units (IU)/mL. The lower limit of the assay is 357 IU/mL (2 × 103 copies/mL) and the upper limit is 17,587,100 IU/mL (108 copies/mL). Signal amplification assays do offer some advantages over the target or nucleic acid amplification assays. There is no need for an extensive sample preparation to purify the DNA and because no amplicons (amplified subunits of DNA) are generated, there is no need for the stringent contamination control procedures employed for target amplification assays such as PCR. ROCHE COBAS TAQMAN HBV TEST This commercial assay supercedes the previous COBAS AMPLICOR HBV MONITOR assay, which relied on coamplification of a modified HBV DNA standard of known copy number to allow quantification. The MONITOR end point PCR assay suffered from a limited upper dynamic range mainly because of the saturation effects inherent in such technology. The development of real-time or kinetic PCR has overcome many of these deficiencies. In kinetic PCR, product is detected as it accumulates by the use of fluorescent-based detection chemistry. This allows a signal to be monitored following each thermal cycle without having to sample directly from each tube, which also reduces the contamination risks. The Roche COBAS TaqMan HBV test uses a dual fluorophore labeled probe that is designed to anneal to one strand of the HBV sequence generated from the PCR primers. The reporter fluorophore’s emissions are extinguished by the nearby downstream quencher fluorophore,

resulting in no signal being produced. However, as the Taq DNA polymerase generates the complementary strand, the endonuclease activity of the enzyme cleaves the reporter off the probe and its emissions are no longer quenched, resulting in a measurable signal. The signal is proportional to the amount of amplicons generated, which is proportional to the amount of original target. Quantification is achieved using a linearized HBV plasmid Quantitation Standard DNA, which is incorporated with each sample and is carried through sample preparation, amplification and detection steps. The linear range of the assay is from approx 30 IU/mL to an upper limit of 1.1 × 108 IU/mL. ARTUS REALART HBV LC PCR REAGENTS This kit contains ready-for-use reagents for the quantification of HBV DNA by real-time PCR specifically using the Roche LightCycler™ instrument (Roche Diagnostics, Mannheim, Germany). The assay uses the fluorescence resonance energy transfer technology to monitor the fluorescence intensity. Dual probes are used where the 3′ end of one probe has a donor fluorophore and the 5′ end of the second probe has an acceptor fluorophore. The probes are designed to hybridize adjacent to one another on a strand of HBV DNA. If specific HBV DNA is present, on excitation the donor fluorophore transfers energy to the acceptor that emits a signal of defined wavelength for detection. The kit also provides an internal control, which is added to the lysed sample, to monitor DNA purification and to check for any inhibition of the PCR. Quantification is achieved by reference to a standard curve generated from supplied controls. The dynamic range of the assay is from 2 × 102 to an upper limit of 109 IU/mL. VIRAL LOAD—CLINICAL SIGNIFICANCE HBV DNA can be detected in acute infection before any serological marker and this may be of some use in the investigation of nosocomial outbreaks. In practice, acute infection can be diagnosed by serological detection of HBcIgM and usually HBsAg is also present. In chronic infection, the presence of HBeAg is a reliable indicator of active replication. Seroconversion from HBeAg positive to anti-HBe was presumed to be part of the immunoelimination phase of CH-B that preceded virus clearance and loss of HBsAg. Instead, it became evident that many patients without HBeAg had HBeAg-negative CH-B in which disease was ongoing, although characteristically it was associated with a lower viral load than the HBeAg-positive CH-B. A number of workers have attempted to examine whether there is a threshold level of HBV DNA that can be used to discriminate between active and inactive disease, because markers such as HBeAg and serum transaminase levels have proved inadequate. It has been proposed some patients could be characterized as having inactive disease when they have detectable HBsAg, but without HBeAg and no transaminase elevations, and have HBV DNA levels below 105 copies/mL (approx 2 × 104 IU/mL). However, as previously demonstrated, many cases of HBeAg-negative CH-B have transaminase flares, meaning that there must be a greater reliance on HBV DNA levels. Evaluation of a large group of untreated patients with HBeAg-negative CH-B using regular testing of transaminase levels and histological examination when indicated, led to the conclusion that HBV DNA levels of 3 × 104 copies/mL (approx 6 × 103 IU/mL) represented a better cut off to classify a patient into the inactive carrier state. The ability to measure HBV DNA levels has also allowed strategies to be designed to reduce the risk of virus transmission from HBV-infected healthcare workers (HCWs) to patients. Clearly, if all HCWs who were HBsAg positive were excluded from performing exposure prone procedures, there would be little

CHAPTER 53 / MOLECULAR DIAGNOSTICS IN HEPATITIS B

opportunity for HBV transmission. However, it can be argued that the loss of expertise of highly qualified and experienced HCWs does affect the health care system, particularly in developing countries where replacement personnel may not be available. Thus, efforts have been made to establish guidelines for the management of HBV-infected HCWs based on HBV DNA levels. Many countries exclude HBeAg-positive HCWs from certain health care duties but HBV transmission has been documented to take place from HCWs negative for HBeAg. A European consensus panel agreed that each country could individually determine a HBV DNA level cut off for HCWs performing exposure prone procedures but proposed that a maximum level of 104 copies/mL (approx 2 × 103 IU/mL) would provide a balanced risk management strategy. This value would take into account that transmission rarely occurred with viral loads greater than 105 copies/mL (approx 2 × 104 IU/mL) and provided a 1 log safety margin to account for possible intermittent flares of activity, assay variation or early development of resistance while on anti-HBV therapy. Others consider this to be a conservative estimate and counter propose that a level of 105 copies/mL (approx 2 × 104 IU/mL) is associated with a minimal increase in risk. VIRAL LOAD—USES IN ANTIVIRAL THERAPY Histological examination of liver tissue and determination of serum transaminase levels are both important measures to use in assessing a patient’s eligibility for antiviral therapy. Importantly, detection and quantification of serum HBV DNA should also be one of the prerequisites for eligibility because it provides the clinician with a practical tool to monitor the effectiveness of treatment. Kinetics of Viral Load Decay on Therapy One of the most promising areas in the evaluation of antiviral therapy is the examination of the kinetics of decline in viral load. Serial viral load measurements from patients prior to therapy show that viremia is maintained at an approximately constant level, even though that level can vary from patient to patient. This indicates that there is a dynamic equilibrium between virus production and clearance that maintains this steady-state relationship. The equilibrium is disturbed by the introduction of antiviral therapy and the analysis of the rate of decline in viral load allows predictions to be made about drug efficacy using mathematical models. In simplified models, there are at least two phases of viral load decay. The initial phase corresponds to clearance of circulating virus; the rate, and magnitude of the decline reflects the drug’s efficacy. The second slower phase is thought to correspond to the eradication of infected cells. For HBV, this most likely represents immune elimination of infected cells, possibly aided by natural cell turnover. Unlike the viral dynamics of other blood-borne viruses, such as HIV or HCV, there appears to be greater variation in the phases with HBV infection and more complex mathematical models have been developed to explain these features. It is hoped that application of these models will provide predictive data on treatment outcome and allow the design of more effective treatment regimes. Monitoring Effectiveness of Treatment Treatment of chronic HBV infection with either LMV or ADV produces a rapid decrease in viremia, corresponding to the first phase of viral decay. A small proportion of patients show loss of HBeAg and normalization of serum transaminase levels. However, during the second slower phase, drug-resistant virus can appear. The emergence of drug resistance may be signaled by clinical deterioration or increasing serum transaminases. A superior indicator of the development of resistance is a relapse of HBV DNA levels in the face of ongoing

563

therapy. Development of resistant virus can be confirmed by sequence analysis of the polymerase gene after PCR amplification in which characteristic mutations have been reported for many of the newer anti-HBV agents. A sensitive reverse hybridization line probe assay is also available but this detects only the common mutations that confer LMV resistance. Although empirically an increase in HBV DNA levels on therapy may be presumed to be resulting from the development of resistant virus, it is also important to verify that it is not because of noncompliance. Defining End Points The development of nucleoside per nucleotide analogs as anti-HBV agents has prompted reappraisal of the traditional end points used to assess drug efficacy. For treatment with interferon, HBeAg loss or seroconversion 6 mo after therapy usually defines successful response. Treatment with the new anti-HBV agents produces a rapid decline in viremia, often to levels undetectable by even the new generation NATs. HBeAg seroconversion is uncommon and may be reversible when treatment ceases and is thus an unreliable marker. Histological examination of liver tissue can be useful but biopsies are invasive and cannot be performed on a frequent basis. Consideration has been given to using serum HBV DNA levels as an efficacy end point and although it has proven a useful tool, there is insufficient correlation with treatment responses to justify its use as a sole predictor. OCCULT INFECTION Several case reports have described how some patients show clinical evidence of CH-B infection, despite remaining serologically HBsAg-negative. The use of sensitive NATs demonstrated that in many instances the presence of HBV DNA could be detected. Cases such as these were defined as having occult HBV infection. Several explanations have been proposed to account for the contradictory testing profile. One possibility was that mutations in the surface gene were sufficient to allow HBsAg to evade serological detection. However, studies employing sequence analysis have shown there are few instances of changes to the relevant epitopes of HBsAg. Furthermore, although mutations to regulatory elements in the HBV genome may have an indirect effect on HBsAg synthesis, these too appear to be rare. The most significant finding in most studies of occult infection is that of a very low level of virus replication. This may be because of an incomplete immune response that allows a low level of virus replication or even an enduring cellular response that requires a low level of persistent virus replication. The clinical significance of occult hepatitis B has not been determined, although it appears that transmission of infection from patients with occult infection can occur. QUASISPECIES Rt, an essential enzyme in the life cycle of HBV, lacks proofreading capacity and results in an estimated mutation rate of 10–3–10–5 misincorporations per nucleotide copied. Mathematical analysis of viral load decay following antiviral therapy predicts greater than 1011 virus particles are produced per day. These features of HBV replication mean that heterogeneous genetic variants are generated, which are termed quasispecies. Quasispecies generation is a continuous process with competition and selection resulting in evolutionary flexibility. For HBV, the quasispecies diversity is constrained by overlapping reading frames, meaning that a mutation modifying a protein encoded by one reading frame can have a deleterious effect in the overlapping reading frame. Nevertheless, the high rates of virus production ensure that a large number of viable mutants are produced. Under selection pressure, such as from the host immune system or antiviral therapy, high quasispecies variability allows a

564

SECTION VI / GASTROENTEROLOGY

rapid response resulting in a new dominant quasispecies that can be further refined by selection. QUANTIFICATION OF HBsAg HBV infection leads to overexpression of HBsAg, which is found in vast excess over virions in the blood. Before development of automated assays, quantifying HBsAg was carried out by modifying radioimmunoassays or by Laurell electrophoresis. In acute hepatitis B, peak concentrations of HBsAg decrease within 16–20 d in resolving infection as the host immune system responds. In patients with acute hepatitis B who become carriers, no such decrease is seen. Automated assays have been adapted to the quantification of HBsAg, allowing greater throughput of samples. Using such assays, HBsAg levels were found to be higher in HBeAg-positive CH-B carriers than in HBeAg-negative CH-B carriers. Importantly, HBsAg levels in both sets of patients showed a good correlation with serum HBV DNA levels, potentially providing an inexpensive alternative to the HBV DNA NATs, worthy of further investigation.

MOLECULAR DIAGNOSTICS—LIVER COMPARTMENT IMMUNOHISTOCHEMISTRY Components of HBV that reflect replication can be demonstrated in liver biopsy sections using specialized techniques. Various HBV DNA and RNA replicative intermediates can be demonstrated by PCR and/or in situ hybridization. Viral proteins such as HBsAg, HBcAg, HBeAg, HBx, and Pre-S proteins can all be demonstrated by immunohistochemistry. Immunodetection of HBsAg and HBcAg in hepatocytes can provide useful information about the replicative status of HBV and is usually performed as part of a histopathological diagnosis of patients with CH-B. The pattern of antigen distribution relates somewhat to the stage of infection. Patients in the nonreplicative phase have little or no demonstratable HBcAg in the liver. In the replicative phase, HBcAg is found in the hepatocyte nuclei, and when replication is very high, in the cytoplasm as well. Cytoplasmic HBcAg tends to correlate with the activity of the liver disease. HBsAg is found in the hepatocyte cytoplasm, but it tends to be inversely related to the amount of HBcAg and to the disease activity. Patients with abundant HBcAg generally have less HBsAg overall and less per cell than those with little or no HBcAg. Patients in the nonreplicative phase of infection, with integration of HBV DNA into the hepatocyte genome, often have hepatocytes with abundant cytoplasmic HBsAg. These cells have uniformly pale, eosinophilic cytoplasm with a “ground-glass” appearance on routine hematoxylin-eosin stains. In immunostains they appear dark and discrete, compared to the pale and diffuse staining pattern of the replicative phase. Both surface and core antigens typically have a patchy distribution within the liver, and more than one pattern may be found in the same biopsy. The cellular localization of HBcAg and HBeAg generally coincide but there are important differences at a subcellular level. HBeAg is usually detected in the nucleus and/or cytoplasm and intense or strong cytoplasmic HBeAg-staining is associated with a high serum HBV DNA level and inactive liver disease. HBcAg can also be detected in the nucleus and/or cytoplasm, but strong cytoplasmic HBcAg expression is typically associated with active liver disease. The proportion of hepatocytes expressing HBsAg correlates inversely with viremia and HBsAg and HBcAg are usually expressed independently within the lobule. Thus, cytoplasmic HBcAg (and not HBeAg) is the target for immune system-mediated cytolysis of hepatocytes. Cytoplasmic HBeAg is not associated with

liver damage, but instead with high levels of HBV replication. Finally, the degree of expression of HBcAg in the hepatocyte nucleus does reflect the level of viral replication in CH-B INTRAHEPATIC HBV COVALENTLY CLOSED CIRCULAR DNA HBV infection is not directly cytopathic, so HBV DNA can persist in the hepatocyte as long as the cell survives. In an interesting study, a commercially available HBV PCR assay was modified to investigate the levels of HBV DNA in frozen liver tissue. Subgroups of patients with CH-B were examined, including those with active replication, as defined by detection of serum HBV DNA and liver HBcAg reactivity, and those with “suppressed” replication, which included patients on antiviral therapy and patients coinfected with hepatitis D virus. Although patients with active virus replication had the highest levels of intrahepatic viral DNA, patients with suppressed HBV activity maintained relatively high levels of intrahepatic HBV DNA. The mean difference in serum HBV DNA levels between the two groups was more than 300-fold, whereas the mean difference in intrahepatic DNA levels was only approx 3-fold. It is likely that the major component of the intrahepatic HBV DNA in the suppressed patient groups is the covalently closed circular (ccc) replicative form found in the nucleus. This viral form is produced from the partially double-stranded genomic HBV DNA found in the viral nucleocapsid. On infection, the genomic DNA is transported to the hepatocyte nucleus in which it is converted by host cell enzymes into a fully double-stranded form. From this, a ccc DNA molecule is generated that associates with cellular histones and other nuclear proteins to form a viral minichromosome. This viral replicative form, the HBV ccc DNA, remains in the nucleus and serves as the transcriptional template for HBV RNA production. The hepatocyte RNA polymerase II transcribes all the viral mRNA from the HBV ccc DNA, including the terminally redundant primary transcript, the pgRNA. This RNA molecule is encapsidated in precursors of the virus core and is reverse transcribed by the viral polymerase to form the singlestranded minus-sense DNA. The RNA is degraded by the viral RNase H and the remaining minus-strand DNA becomes the template for the synthesis of a plus-strand DNA of variable length. Assembled nucleocapsids containing the partially doublestranded HBV DNA can follow either of two pathways. They can become enveloped by the HBV surface proteins and be secreted from the cell or be recycled back to the nucleus in which the HBV DNA is converted into ccc DNA as part of an intracellular conversion pathway to maintain a ccc DNA pool. Once a pool of 5–50 copies has been established, nucleocapsids associate with envelope proteins, rather than being recycled, and leave the cell as virions. The determinant of which pathway the nucleocapsid follows appears to be the amount of pre-S protein. The greater the number of viral ccc DNA molecules, the greater the level of RNA transcripts, resulting in more pre-S protein available to associate with nucleocapsids. The pool of viral ccc DNA molecules in the cell nucleus is the likely reason for the relapse of viremia after treatment with nucleoside and nucleotide analogs. HBV replication does not use a semiconservative mechanism, thus, antiviral therapy with such agents can only affect newly synthesized DNA and not the pre-existing HBV ccc DNA pool. Elimination of viral ccc DNA might be achieved if production of new virions could be completely blocked; ccc DNA would be eventually lost by either decay or death of the infected hepatocyte. Two major means of removal of viral ccc DNA have

CHAPTER 53 / MOLECULAR DIAGNOSTICS IN HEPATITIS B

been proposed, a cytokine-induced noncytolytic clearance of virus from infected cells, and the killing of infected cells by virusspecific cytotoxic T cells. The critical role played by intrahepatic viral ccc DNA has motivated researchers to develop assays for its detection and quantification. Assays for hepadnaviral ccc DNA have been developed for the woodchuck and duck hepatitis B models and applied to the investigation of the mechanism of clearance of infection. Studies using the woodchuck model showed that in transient infection a high proportion of infected hepatocytes were destroyed and replaced by uninfected cells, derived by cell division from the remaining infected hepatocytes. Some support that this mode of eradication of infection also occurs with antiviral therapy has been shown in chronically infected woodchucks treated with nucleoside analogs. The amount of viral ccc DNA decreased substantially whereas the amount of residual integrated viral DNA remained the same, suggesting a turnover of infected hepatocytes and compensatory accumulation of uninfected cells. Furthermore, in chronically infected ducks treated with a cytotoxic antiviral drug, the loss of infected hepatocytes was more rapid and showed in surviving animals that cell death and regeneration of uninfected cells play a role in recovery from infection. For the assessment of HBV ccc DNA, a selective PCR assay was devised. Techniques that may have been applicable with animal models may not necessarily apply to patients in which the amount of sample available is limited, usually biopsy tissue, and the copy number of viral ccc DNA is low. Assays have been adapted to the real-time PCR format that provides for a greater dynamic range and more accurate quantification. In contrast to some of the studies in animal models, investigation of 48 wk treatment with the nucleotide analog ADV showed that there was a significant decrease in HBV ccc DNA levels, yet the number of immunostaining cells remained similar to that of pretreatment. This suggests a role for a noncytolytic mechanism of clearance rather than destruction of infected cells. Consistent with this is the finding of a more rapid elimination of HBV ccc DNA than that of HBsAg-positive hepatocytes in HBV infected chimpanzees. However, this phase of the elimination was followed by a second phase in which there was a peak of hepatocyte turnover and a surge of interferon-γ CD8+ T cells associated with loss of remaining viral ccc DNA. The study illustrates that the two mechanisms of eradication of infection, noncytopathic clearance of intracellular virus and immune-mediated destruction of hepatocytes, are not mutually exclusive. PCR assays for HBV ccc DNA can have many uses. In patients being treated for CH-B, assessing the level of viral ccc DNA may indicate drug efficacy and help decide treatment duration. Insights into the mechanisms of elimination of HBV ccc DNA may identify potential targets for intervention. Assays require further standardization and validation before they can be considered part of a patient management strategy.

CONCLUSIONS Hepatitis B is a vaccine preventable disease, yet it remains the commonest form of viral hepatitis and a worldwide public health problem. Acute infection can produce symptoms including fever, jaundice, and malaise that may spontaneously resolve within 6 mo. A number of people do not resolve the infection and become chronic carriers; this group is at risk of developing severe complications such as cirrhosis and HCC. Chronic carriers also represent a pool of infectious virus. Infection acquired early in life is likely

565

to lead to chronic infection, so in countries hyperendemic for hepatitis B, transmission from chronically infected mothers to their babies helps perpetuate the carrier cycle. Despite the availability of an efficacious vaccine that prevents HBV infection, help for those who already have CH-B relies on control of their disease by antiviral chemotherapy. The first effective antiviral agent used to treat CH-B was α-interferon. It is effective in eliminating HBV infection in a proportion of patients but is associated with considerable side effects. This has led to the search for other less toxic agents for the control of CH-B. HBV replicates by reverse transcription of an RNA transcript. This provides the opportunity for inhibition of replication by nucleoside and nucleotide analogs, many of which have been previously evaluated for the treatment of HIV infection. Antiviral therapy with agents such as LMV and ADV result in a dramatic drop in serum HBV DNA levels, which has required the development of sensitive NATs to assess their efficacy. The replication strategy of HBV also means that a heterogeneous population of genetic variants, or quasispecies, is generated although this is constrained to some degree by the genomic organization of overlapping reading frames, which allows the virus to respond to the selection pressure placed on it by antiviral therapy. Here too technology can play a role in being able to detect clinically significant variants, particularly detection of the early development of resistance. Combination therapy with nucleoside per nucleotide analogs has been an effective approach to combat problems associated with resistance in the treatment of HIV infection and this has already been initiated against CH-B in preliminary clinical trials. Treatment with interferon has been revisited with the development of pegylated versions that reduce the frequency of injection and appear to show improved efficacy. The use of sensitive and standardized NATs will be important in determining optimal combination therapy and may play some role in defining therapeutic end points. Assessment of liver damage can be determined by the histological examination of tissue from a liver biopsy. Immunohistochemical demonstration of viral proteins can also provide some indication of the stage of HBV replication. Molecular technology has been applied to the investigation of HBV replication in liver tissue. A number of assays have been developed to measure the replicative intermediate HBV ccc DNA. These assays need to be validated and standardized before their full potential can be realized. The ability to measure this important intermediate may provide information on treatment efficacy and an understanding of the mechanisms involved in the elimination of viral ccc DNA. Further studies of HBV replication may shed light on the nature of other chronic viral infections. Analysis of the results from patients treated with combination therapy could provide a basis for improved combination therapy for treatment of HIV infection, rather than vice versa, as is the current situation. Likewise, an understanding of how HBV causes immune-mediated liver disease may aid in the prevention of complications such as cirrhosis and HCC caused by hepatitis C virus. Clearly, many questions remain to be answered with this fascinating pathogen.

SELECTED REFERENCES Alberti A. Can serum HBV-DNA be used as a primary end point to assess efficacy of new treatments for chronic hepatitis B? Hepatology 2003;38:18–20. Bartenschlager R, Schaller H. Hepadnaviral assembly is initiated by polymerase binding to the encapsidation signal in the viral RNA genome. EMBO J 1992;11:3413–3420.

566

SECTION VI / GASTROENTEROLOGY

Beasley R, Lin C, Hwang L, Chien C. Hepatocellular carcinoma and hepatitis B virus. Lancet 1981;2:1129–1133. Birnbaum F, Nassal M. Hepatitis B virus nucleocapsid assembly: primary structure requirements in the core protein. J Virol 1990;64:3319–3330. Brechot C, Thiers V, Kremsdorf D, Nalpas B, Pol S, Paterlini-Brechot P. Persistent hepatitis B virus infection in subjects without hepatitis B surface antigen: clinically significant or purely “occult”? Hepatology 2001;34:194–203. Carman WF, Jacyna MR, Hadziyannis S, et al. Mutation preventing formation of hepatitis B e antigen in patients with chronic hepatitis B infection. Lancet 1989;2:588–591. Carman WF, Zanetti AR, Karayiannis P, et al. Vaccine-induced escape mutant of hepatitis B virus. Lancet 1990;336:325–329. Chang LJ, Pryciak P, Ganem D, Varmus HE. Biosynthesis of the reverse transcriptase of hepatitis B viruses involves de novo translational initiation not ribosomal frameshifting. Nature 1989;337:364–368. Crowther R, Kiselev N, Bottcher B, et al. Three-dimensional structure of hepatitis B virus core particles determined by electron cryomicroscopy. Cell 1994;77:943–950. Das K, Xiong X, Yang H, et al. Molecular modeling and biochemical characterization reveal the mechanism of hepatitis B virus polymerase resistance to lamivudine (3TC) and emtricitabine (FTC). J Virol 2001; 75:4771–4779. Fattovich G, Brollo L, Giustina G, et al. Natural history and prognostic factors for chronic hepatitis type B. Gut 1991;32:294–298. Ganem D, Schneider R. Hepadnaviridae: The Viruses and Their Replication. In: Knipe DM, Howley PM, eds. Fields Virology, vol. 2, Philadelphia: Lippincott-Raven, 2001; pp. 2923–2970. Gauthier J, Bourne EJ, Lutz MW, et al. Quantitation of hepatitis B viremia and emergence of YMDD variants in patients with chronic hepatitis B treated with lamivudine. J Infect Dis 1999;180:1757–1762. Goodman Z. Histopathology of hepatitis B virus infection. In: Lai CLS, ed. Human Virus Guides, vol. 1, London: International Medical Press, 2002:131–143. Guidotti LG, Rochford R, Chung J, Shapiro M, Purcell R, Chisari FV. Viral clearance without destruction of infected cells during acute HBV infection. Science 1999;284:825–829. Gunther S, Fischer L, Pult I, Sterneck M, Will H. Naturally occurring variants of hepatitis B virus. Adv Virus Res 1999;52:25–137. Hadziyannis SJ, Lieberman HM, Karvountzis GG, Shafritz DA. Analysis of liver disease, nuclear HBcAg, viral replication, and hepatitis B virus DNA in liver and serum of HBeAg vs. anti-HBe positive carriers of hepatitis B virus. Hepatology 1983;3:656–662. Hino O, Ohtake K, Rogler CE. Features of two hepatitis B virus (HBV) DNA integrations suggest mechanisms of HBV integration. J Virol 1989;63:2638–2643. Jung M, Pape G. Immunology of hepatitis B infection. Lancet Infect Dis 2002;2:43–50. Kann M, Gerlich W. Hepadnaviridae: structure and molecular virology. In: Zuckerman A, Thomas H, eds. Viral Hepatitis. London: Churchill Livingstone, 1998:77–105. Koike K, Tsutsumi T, Fujie H, Shintani Y, Kyoji M. Molecular mechanism of viral hepatocarcinogenesis. Oncology 2002;62(Suppl. 1):29–37. Lok AS, Akarca U, Greene S. Mutations in the pre-core region of hepatitis B virus serve to enhance the stability of the secondary structure of the pre-genome encapsidation signal. Proc Natl Acad Sci USA 1994;91:4077–4081. Lok AS, Heathcote EJ, Hoofnagle JH. Management of hepatitis B: 2000— summary of a workshop. Gastroenterology 2001;120:1828–1853. Milich DR, McLachlan A, Stahl S, et al. Comparative immunogenicity of hepatitis B virus core and E antigens. J Immunol 1988;141:3617–3624.

Naoumov NV, Portmann BC, Tedder RS, et al. Detection of hepatitis B virus antigens in liver tissue. Gastroenterology 1990;99:1248–1253. Newbold JE, Xin H, Tencza M, et al. The covalently closed duplex form of the hepadnavirus genome exists in situ as a heterogeneous population of viral minochromosomes. J Virol 1995;69:3350–3357. Nowak MA, Bonhoeffer S, Hill AM, Boehme R, Thomas HC, McDade H. Viral dynamics in hepatitis B virus infection. Proc Natl Acad Sci USA 1996;93:4398–4402. Richman DD. The impact of drug resistance on the effectiveness of chemotherapy for chronic hepatitis B. Hepatology 2000;32:866,867. Rossner MT. Review: hepatitis B virus X-gene product: a promiscuous transcriptional activator. J Med Virol 1992;36:101–117. Saldanha J, Gerlich W, Lelie N, Dawson P, Heermann K, Heath A. An international collaborative study to establish a World Health Organization international standard for hepatitis B virus DNA nucleic acid amplification techniques. Vox Sang 2001;80:63–71. Schaller H, Fischer M. Transcriptional control of hepadnavirus gene expression. Curr Top Microbiol Immunol 1991;168:21–39. Schlicht HJ, Bartenschlager R, Schaller H. Biosynthesis and enzymatic functions of the hepadnaviral reverse transcriptase. In: McLachlan A, ed. Molecular Biology of the Hepatitis B Viruses. Boca Raton, FL: CRC Press, 1991:171–180. Seeger C, Mason WS. Hepatitis B virus biology. Microbiol Mol Biol Rev 2000;64:51–68. Shaul Y. Regulation of hepadnavirus transcription. In: McLachlan A, ed. Molecular Biology of the Hepatitis B Virus. Boca Raton, FL: CRC Press, 1991:193–211. Stuyver LJ, Locarnini SA, Lok A, et al. Nomenclature for antiviral-resistant human hepatitis B virus mutations in the polymerase region. Hepatology 2001;33:751–757. Summers J, Mason WS. Replication of the genome of a hepatitis B-like virus by reverse transcription of an RNA intermediate. Cell 1982;29: 403–415. Summers J, Mason WS. Residual integrated viral DNA after hepadnavirus clearance by nucleoside analog therapy. Proc Natl Acad Sci USA 2004;101:638–640. Summers J, Smith PM, Horwich AL. Hepadnavirus envelope proteins regulate covalently closed circular DNA amplification. J Virol 1990;64:2819–2824. Summers J, Jilbert AR, Yang W, et al. Hepatocyte turnover during resolution of a transient hepadnaviral infection. Proc Natl Acad Sci USA 2003;100:11,652–11,659. Tuttleman JS, Pourcel C, Summers J. Formation of the pool of covalently closed circular viral DNA in hepadnavirus-infected cells. Cell 1986; 47:451–460. Wallace WA, Carman WF. Surface variation of HBV: scientific and medical relevance. Viral Hepat Rev 1997;3:5–16. Wang GH, Seeger C. Novel mechanism for reverse transcription in hepatitis B viruses. J Virol 1993;67:6507–6512. Werle B, Bowden D, Locarnini S, et al. Persistence of viral cccDNA during the natural history of chronic hepatitis B and slow decline with adefovir dipivoxil therapy. Gastroenterology 2004;126:1750–1758. Will H, Reiser W, Weimer T, et al. Replication strategy of human hepatitis B virus. J Virol 1987;61:904–911. Wu TT, Coates L, Aldrich CE, Summers J, Mason WS. In hepatocytes infected with duck hepatitis B virus, the template for viral RNA synthesis is amplified by an intracellular pathway. Virology 1990;175: 255–261. Wynne SA, Crowther RA, Leslie AG. The crystal structure of the human hepatitis B virus capsid. Mol Cell 1999;3:771–780. Zoulim F, Saputelli J, Seeger C. Woodchuck hepatitis virus X protein is required for viral infection in vivo. J Virol 1994;68:2026–2030.

54 Hereditary Hemochromatosis FRANCESCA FERRARA, ELENA CORRADINI, AND ANTONELLO PIETRANGELO SUMMARY Iron is a major component of the Earth’s crust, but its own chemistry greatly limits utilization and also sets the basis for its toxicity. Hereditary hemochromatosis (HH) is the most common cause of iron overload in humans. For much of the twentieth century, HH was regarded as a monogenic disorder characterized by excess tissue deposits of iron inevitably producing organ damage. This view has been shattered by the identification of similar phenotypes associated with mutations of at least four different ironmetabolism genes (HFE, TfR2, HAMP, HJV) and the increasing appreciation of the disease’s multifactorial nature. Key Words: Hereditary hemochromatosis (HH); HFE; hepcidin; hemojuvelin; iron; TfR2.

INTRODUCTION Iron is a major component of the Earth’s crust, but its own chemistry greatly limits utilization and also sets the basis for its toxicity. The capacity of readily exchanging electrons in aerobic conditions makes iron essential for fundamental cell functions, such as DNA synthesis, transport of oxygen and electrons, and cell respiration. However, because humans have no means to control iron excretion, excess iron, regardless of the route of entry, accumulates in parenchymal organs, and threatens cell viability. A number of disease states are caused by excessive accumulation of iron in the body (Table 54-1). The term “primary” identifies diseases owing to a recognizable hereditary defect in proteins directly involved in iron homeostasis, as opposed to conditions in which iron overload is secondary to known factors or diseases. Hereditary hemochromatosis (HH) is the most common cause of iron overload in humans. The discussion of other rarer primary iron-overload disorders is beyond the scope of this chapter. Ferroportin-associated iron overload, erroneously classified as type-4 hemochromatosis in genetic tassonomy is likely the most common cause of hereditary hyperferritinemia beyond classic HH. The disorder, clinically recognized in 1999, has distinctive genetic, pathological, and clinical features (Table 54-2).

HISTORICAL ASPECTS The term “hemochromatosis” was introduced in 1889 to describe autopsy findings of massive tissue deposits of iron assoFrom: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

ciated with multiple organ damage, particularly cirrhosis. In 1935 a critical review of reported cases concluded that the syndrome was hereditary and that excess tissue iron was the cause of the organ pathology. By the mid 1970s, HH had been recognized as an autosomal-recessive disorder linked to the HLA-A3 region of the short arm of chromosome 6 (4, 5), and in 1996 it was attributed to the newly discovered “hemochromatosis gene” (HFE).

GENOTYPES AND PHENOTYPE The basic features of the HH syndrome can be produced by pathogenic mutations of four different iron metabolism genes (see Table 54-2). Depending on the gene involved, and its physiological role in iron homeostasis, the HH phenotype varies, ranging from massive early-onset iron loading with severe organ disease (e.g., associated with homozygous mutations of HAMP or HJV) (historically known as juvenile hemochromatosis), to the milder adult-onset phenotype characterizing classic HFE and transferrin receptor 2 (TfR2)-related forms. Phenotypic penetrance of “juvenile” HH genes appears to be high, whereas that of “adult” HH genes is lower. In general, adult HH phenotypes are also particularly subject to the modifying effects of host-related and environmental factors, such as alcohol abuse or chronic liver insult (Fig. 54-1). “Intermediate phenotypes” can result from combined heterozygous or homozygous mutations of multiple hemochromatosis genes. Increased risk of clinically expressed disease may indeed occur in patients with heterozygous mutations of both adult and juvenile HH genes. The variety of genotypes that can produce an HH phenotype highlights the importance of defining the disease clinically as a unique entity and avoiding classification into subtypes based exclusively on genetic diversities. ADULT ONSET HH: HFE- AND TFR2-RELATED Classic HH is the most frequently inherited metabolic disorder found in whites, with a genetic prevalence ten times higher than that of cystic fibrosis. It is associated with mutation of the HFE gene located on chromosome 6, in most cases a single-base transition leading to substitution of tyrosine for cysteine at position 282 (C282Y) of the HFE protein. The clinical impact of another less common HFE mutation, H63D (aspartic acid for histidine at position 63), appears to be limited, and yet, 1–2% of C282Y/H63D compound heterozygotes seem predisposed to disease expression. C282Y is thought to have originated by chance in a single ancestor inhabiting northwestern Europe some 2000 yr ago. The genetic defect caused no serious obstacle to reproduction and may even have conferred some advantages such as resistance to dietary iron deficiencies

567

568

SECTION VI / GASTROENTEROLOGY

Table 54-1 Human Iron Overload Disorders Primary iron overload disorders Hereditary hemochromatosis (adult and juvenile onset forms) Ferroportin-related iron overload Aceruloplasminemia A(hypo)transferrinemia Secondary iron overload disorders Dietary iron overload Parenteral iron overload Hematological disorders associated with inefficient erythropoiesis Long-term hemodialysis Chronic liver disease Viral hepatitis Alcoholic cirrhosis Nonalcoholic fatty liver disease Porphyria cutanea tarda Dysmetabolic iron overload syndrome Other iron overload disorders Iron overload in sub-Saharan Africans Neonatal hemochromatosis

and certain infectious diseases. It was passed on and spread through population migration. C282Y homozygosity is found in approx 5/1000 of individuals of northern European descent. All C282Y homozygotes are genetically predisposed to a chain of events that may culminate in severe multiple organ damage, but it is impossible to predict whether and to what extent this risk will be expressed (see Fig. 54-1). A small percentage of C282Y homozygotes never present signs of altered iron metabolism. Symptomatic patients generally present during midlife, often with nonspecific complaints (e.g., unexplained fatigue, joint pain). Liver disease (with manifestations ranging from hepatomegaly and/or moderate hypertransaminasemia to cirrhosis and even hepatocellular carcinoma) usually predominates but endocrine (diabetes, hypogonadotropic hypogonadism, impotence, hypothyroidism), cardiac (arrhythmias, heart failure), and joint disease (destructive arthritis) is also common. Although very few cases have been described, the iron-overload phenotype associated with mutations of TfR2 appears to be similar to that of classic HFE-related HH. JUVENILE-ONSET HH: HEPCIDIN- AND HEMOJUVELINRELATED The juvenile onset hereditary hemochromatosis (J-HH) phenotype is much more severe (see Table 54-2). Plasma iron loading and tissue iron excess (reflected, respectively, by increased transferrin saturation and serum ferritin values) are evident early in life in both sexes. Symptomatic organ disease occurs as early as the first decade, and, although liver involvement is a constant, diabetes, hypogonadotropic hypogonadism, cardiomyopathy, arrhythmias, and heart failure are much more evident. Death (usually because of intractable heart failure) before the age of 30 yr is not uncommon. J-HH is rare. Cases have been linked to homozygous mutations of the HAMP gene, which encodes hepcidin, an antimicrobial peptide synthesized by hepatocytes that plays a major role in human iron metabolism, but most cases of J-HH are because of mutations of the HJV (formerly HFE2) gene located on chromosome 1. The most common mutation is a G320V substitution in the HJV gene product, hemojuvelin, whose function is unknown.

PATHOGENESIS A gradual and progressive expansion of the plasma iron compartment (reflected by high saturation of circulating transferrin) is the earliest detectable biochemical abnormality in all forms of HH. This alteration is evident at birth and progressive, but, at least for the adult HH form, does not cause problems during childhood and adolescence in male and in menstruating females owing to high growth demands and iron losses, respectively. These protective mechanisms do not seem to take place in the juvenile form where uncontrolled iron expansion leads to rapid tissue iron overload. Although increase of plasma iron pool in enteral or transfusioninduced iron overload is normally followed by increased iron excretion, this compensatory response fails in HH, probably because the intestinal cells that are eliminated contain little or no iron or ferritin. As a result, the total iron entering the plasma largely exceeds loss and, as plasma iron concentrations reaches critical levels, iron begins to accumulate in the parenchymal cells of the liver, where stores more than 25,000–30,000 mg can be found (normal value is 1000) Hypercalcemia Trauma Blunt abdominal trauma Penetrating ulcer Postoperative/post-ERCP Infections Mycoplasma pneumoniae Virus (mumps, coxsackie, CMV, HSV, HIV) Bacteria Mycobacterium avium-intracellulare Vascular Vasculitis (e.g., connective tissue diseases) Hypotension (postoperative: e.g., CABG) Infarction Abdominal aortic aneurysm Idiopathic Unknown

(SPINK1). This specific trypsin inhibitor is regulated as an acute phase protein and the ratio of SPINK1 to trypsin is therefore dependent on the status of the immune system and inflammation. The third mechanism is trypsin autolysis. The trypsin molecule has two globular domains connected by a peptide chain containing an arginine at codon 122. This arginine serves as target for a second trypsin and therefore serves as a self-destruct site when R122 is exposed. Cleavage at this site permanently inactivates trypsin (Fig. 55-1). Accessibility of R122 to cleavage appears to be regulated by calcium. Low calcium levels (e.g., inside the acinar cell) favor autolysis whereas high levels (e.g., inside the pancreatic duct

CHAPTER 55 / PANCREATIC EXOCRINE DYSFUNCTION

575

Figure 55-1 Model of trypsin self-destruct mechanism preventing pancreatic autodigestion. (A) Autoactivation and enzymatic activation of trypsinogen generate trace amounts of active trypsin within pancreatic acinar cells. Active trypsin is inhibited by a limited supply of trypsin inhibitor. If trypsin activity exceeds the inhibitory capacity of PSTI, then proenzymes, including mesotrypsin and enzyme Y, are activated. These enzymes feed back to inactivate wild-type trypsinogen, trypsin, and other zymogens. (B) Activation of mutant (HP) trypsin in amounts that exceed the inhibitory capacity of PSTI results in activation of proenzymes. Because the Arg 117 cleavage site for trypsin-like enzymes is replaced by His in the HP mutant trypsin, trypsin continues to activate trypsinogen and other zymogens unabated, leading to autodigestion of the pancreas and pancreatitis.

and duodenum) prevent autolysis. Thus, trypsin activity regulates zymogen activation and calcium levels regulate trypsin survival. If calcium levels are improperly regulated within the pancreatic acinar cell then conditions favor prolonged survival of trypsin, leading to zymogen activation, pancreatic autodigestion, and pancreatitis. Because the acinar cells use calcium as the primary second messenger system, intracellular hypercalcemia can occur with hyperstimulation of the pancreas, with hypercalcemia during normal pancreas stimulation, with impaired calcium removal as can occur with mitochondrial injury or other problems such as occurs with significant alcohol exposure. If trypsinogen is released from the pancreas and becomes activated, the high calcium levels in the pancreatic juice eliminate the autolysis protection mechanism. However, the ductal system is protected both by SPINK1 inhibition of small amounts of active trypsin and by rapid flushing of the duct with bicarbonate-rich fluid secretions generated by ion efflux from the duct cells through CF transmembrane conductance regulator (CFTR). Conditions that dampen recognition of trypsin activation that inhibit CFTR and duct cell function, or that decrease SPINK1 inhibitory capacity increase susceptibility to acute pancreatitis. Finally, if active digestive enzymes escape from the pancreatic acinar cells and ducts into the intracellular space or plasma, they are inactivated by hepatocyte-derived α2-macroglobulin and α1-protease inhibitors. Thus, multiple protective strategies exist to prevent trypsinogen from being activated and to prevent trypsin from activating zymogens and initiating pancreatic autodigestion. Acute pancreatitis from zymogen activation and autodigestion likely cause limited injury compared with the severe effects of the inflammatory response. The release of the products of digestion trigger the immune system with the release of a cascade of proinflammatory mediators including interleukin (IL)-1, IL-2, IL-6, IL-8, tumor necrosis factor-α, and platelet activating factor, which often leads to a systemic inflammatory response syndrome (SIRS). If pancreatic injury is mild, then the inflammatory response is self-limited

and the process is contained. However, if pancreatic injury is severe, then a more intense inflammatory response ensues. This SIRS is responsible for many of the local and systemic complications of acute pancreatitis including a hypermetabolic state, decreased peripheral vascular resistance with increased cardiac output, pulmonary complications of hypoxia and adult respiratory distress syndrome, and renal injury. After the acute phase the pancreas begins to heal. However, it may take months before pancreatic function returns to normal. Thus, acute pancreatitis progresses through three general phases: an acute injury resulting from a variety of causes, a characteristic inflammatory response that depends on the severity of injury and intensity of the immune response, and finally a healing phase. GENETIC PREDISPOSITION A growing number of genes are being associated with susceptibility to acute pancreatitis and recurrent acute pancreatitis. In addition, modifier genes (including genes that regulate the immune response) may determine clinical course of acute pancreatitis. Many of the primary genetic defects that cause recurrent acute pancreatitis eventually lead to chronic pancreatitis. These include mutations in the trypsinogen gene (PRSS1), the SPINK1 gene, and mild-variable mutations in the CFTR gene (see Chronic Pancreatitis and Pancreatic Insufficiency). Genetic predisposition to hypertriglyceridemia and hypercalcemia also lead to recurrent acute and chronic pancreatitis. HP is inherited as an autosomal-dominant trait with 80% penetration and variable expression. Hundreds of kindreds have been identified since the genetic nature of this disorder was recognized by Comfort and Steinberg in 1952. The cause of HP was determined after the gene was localized to chromosome VII and then identified as cationic trypsinogen (protease, serine 1, PRSS1) by sequencing of candidate genes. Cationic trypsinogen is the most common isoform of trypsinogen. It accounts for two-thirds of trypsin activity, with anionic trypsinogen (PRSS2) accounting for another one-third and mesotrypsinogen (PRSS3) accounting for

576

SECTION VI / GASTROENTEROLOGY

less than 5%. An arginine to histidine (R > H) substitution caused by a mutation in codon 122 (R122; initially reported as R117H using the standard chymotrypsinogen numbering system), was found in the initial five families used to map the gene. The mutation was identified at the critical trypsin autolysis site, and was therefore hypothesized to be a gain-of-function mutation that prevented prematurely activated trypsin from being eliminated, leading to zymogen activation and pancreatitis. Since then, other PRSS1 mutations have been identified in hereditary and occasionally sporadic pancreatitis, and these mutations may either enhance trypsinogen activation or diminish autolysis. The R122H and N29I mutations are, by far, the most common trypsinogen mutations. The SPINK1 gene codes for a specific trypsin inhibitor, PSTI. The SPINK1 N34S single-nucleotide polymorphism (SNP) is linked with several other SNPs and, therefore, forms a haplotype. The N34S haplotype is common in most populations, ranging from 1 to 4% of control alleles. Homozygous SPINK1 N34S mutations markedly increase the risk of recurrent acute pancreatitis and chronic pancreatitis, especially in children. Several other mutations, including M1T, P55S, R65Q, and R67C and a number of intronic or promoter region mutations are also occasionally seen. SPINK1 mutations are common in patients with childhood onset pancreatitis and in tropical pancreatitis, although in the majority of cases only heterozygous mutations are identified. It is likely that one or more additional gene mutations and/or specific environmental factors are required to cause pancreatitis in these cases, suggesting in many cases that recurrent acute and chronic pancreatitis represent complex genetic disorders. The CFTR gene codes for an anion channel located on the apical surface of pancreatic duct cells. This ion channel promotes the movement of chloride, bicarbonate and fluid into the duct lumen. Fluid secretion at this site is critical for flushing the zymogen proteins out of the pancreas and into the intestine. Severe mutations in both CFTR alleles lead to minimal fluid secretion and CF (see Genetic Basis). Milder mutations in one allele with a severe mutation in the other provide enough CFTR function for borderline pancreatic function, but there may not be enough reserve for clearing the ducts if trypsinogen activation is significant, or if the outflow tract has high resistance (e.g., in pancreatic divisum), and, therefore, may predispose patients to recurrent acute pancreatitis. The three genes previously mentioned are the best studied examples of genes that increase susceptibility to acute pancreatitis. Other genes are under investigation that do not increase the risk of pancreatitis, but may alter the severity of the clinical course, and therefore act as modifier genes. MANAGEMENT Mild pancreatitis tends to be a self-limited process. Treatment is supportive and includes pancreatic rest (nothing by mouth), intravenous hydration, and analgesics. Patients usually recover within 3 d, and can be discharged from the hospital when oral intake is adequate. By contrast, severe pancreatitis can result in life-threatening complications requiring aggressive management. Because these patients are at risk for sudden death, intensive care unit admission is often indicated during the initial phase of severe acute pancreatitis with SIRS. To predict which patients with acute pancreatitis are most likely to develop life-threatening complications, several prognostic schemes have been developed (Table 55-3). These schemes are useful for identifying the subset of patient with acute pancreatitis for whom intensive monitoring is most important. The major scoring systems and markers of severity are

Table 55-3 Prognostic Criteria in Acute Pancreatitis Ranson criteriaa On admission Age over 55 yr Leukocyte count over 16,000/mm3 Blood glucose over 200 mg/dL Serum LDH over 350 IU/L Serum GOT (SGOT/AST) over 250 IU/dL First 48 h Hematocrit decreases over 10% BUN rises over 5 mg/dL Serum calcium below 8 mg/dL Arterial PO2 below 60 mmHg Base deficit over 4 mEq/L Estimated fluid sequestration more than 6 L Glasgow criteriaa On admission Age over 55 yr WBC count over 15,000/µL Glucose over 180 mg/dL BUN over 45 mg/dL PO2 over 60 mmHg Albumin less than 3.2 g/dL Calcium less than 8 mg/dL LDH more than 600 IU/L aEach factor is one point. Three or more points suggest severe acute pancreatitis.

directed at quantitating the inflammatory response to pancreatitis or assessing multiorgan system injury rather than quantifying amount of pancreatic injury. The best tools for predicting a severe clinical course include the APACHE-II score of more than 7 or IL-6 more than 400 pg/mL at admission, neutrophil elastase more than 300 mcg/L or Trypsinogen 2 > 35 nmol/L at 24 h, or Ranson/ Glasgow score more than 2 or C-reactive protein more than 150 mg/L at 48 h. Early recognition of a severe course is a key to the management of the most common life-threatening complications of severe acute pancreatitis. Cardiovascular collapse and respiratory failure each may occur rapidly and require aggressive support. Multiorgan failure may develop as a consequence of the hypermetabolic state as nutritional stores are depleted, and early nutritional support should be considered. Nutrition can be delivered by total parenteral nutrition or preferably via a nasojejunal feeding tube. Sepsis and infected pancreatic necrosis also commonly occur and are life-threatening complications. Prophylactic broad spectrum antibiotics such as imipenem or cefuroxime may reduce the frequency of these serious infectious complications. Surgery is needed if pancreatic necrosis becomes infected. FUTURE DIRECTIONS Because the injury phase of acute pancreatitis is usually brief and early recognition is difficult, prevention is important. Understanding the sites and mechanisms of early enzyme activation, the factors controlling the inflammatory response and the resulting pathological events may be helpful in limiting pancreatic damage when pancreatic injury occurs. Insights into the mechanism regulating the immune system and identification of SNPs that predict a severe course will be invaluable in determining clinical management strategies.

CHAPTER 55 / PANCREATIC EXOCRINE DYSFUNCTION

Because genetic susceptibility genes have been identified, transgenic animal models can be developed to aid in the study of human pancreatitis, and strategies to prevent HP and other forms of pancreatitis can also be developed.

CHRONIC PANCREATITIS AND PANCREATIC INSUFFICIENCY Chronic pancreatitis is a destructive process characterized by the permanent loss of exocrine function. Typical features include persistent pancreatic inflammation leading to fibrosis, pancreatic atrophy, and ductal dilatation and distortion. In adults, the most common cause of chronic pancreatitis and of pancreatic insufficiency is alcohol. In children, pancreatic insufficiency often occurs without inflammation; the leading causes of this clinical presentation are CF and SDS. Pancreatic insufficiency occurs when digestive enzymes fail to reach the intestine. This can occur because of pancreatic destruction (e.g., chronic pancreatitis), blockage of the pancreatic duct (e.g., pancreatic cancer), surgical removal of the pancreas, or failure of enzyme synthesis (e.g., SDS). As noted, mutations in PRSS1, SPINK1, and CFTR markedly increase susceptibility to recurrent acute pancreatitis. Molecular epidemiology and family studies demonstrate that in these cases, recurrent acute pancreatitis leads to chronic pancreatitis. The same pattern of recurrent acute pancreatitis in alcoholics appears to lead to alcoholic chronic pancreatitis, and susceptibility to alcoholic pancreatitis may also have a genetic link. The prototype disease causing pancreatic insufficiency is CF. CF is a common disease with an incidence of 1/3000 in US Caucasians. Of patients with CF, approx 85% develop pancreatic insufficiency, and pancreatic dysfunction is such a prominent feature of CF that the disease was initially termed fibrocystic disease of the pancreas. CLINICAL FEATURES The most important manifestations of pancreatic insufficiency result from malabsorption of incompletely digested nutrients. Malabsorption of fat and fat-soluble vitamins is usually prominent, leading to steatorrhea associated with delayed growth in children or weight loss in children or adults. In patients with CF, meconium ileus or pulmonary symptoms often suggest the diagnosis before pancreatic insufficiency is evident. The clinical features of CF are discussed in Chapter 27. In adults, the most common cause of pancreatic insufficiency is chronic pancreatitis. Prominent symptoms include chronic midepigastric pain (often postprandial), weight loss largely because of food avoidance, and steatorrhea. In addition, diabetes mellitus sometimes accompanies pancreatic insufficiency resulting from CF or advanced chronic pancreatitis. DIAGNOSIS A diagnosis of pancreatic insufficiency is suggested by the combination of steatorrhea (malabsorption of at least 7%, and typically of 20–50%, of ingested fat during a 72-h stool collection) with normal bowel mucosal morphology (by biopsy) and function (e.g., D-xylose absorption). The diagnosis is supported by a clear response to pancreatic enzyme replacement therapy. When the diagnosis is uncertain, the reference standard for documenting pancreatic insufficiency is the “tubed” secretin test. In this test, the duodenum is intubated to measure the amount of bicarbonate and fluid secreted by the pancreas in response to secretin. Pancreatic insufficiency can also be detected by noninvasive tests such as the dual-label Schilling test, the Bentiromide test, Sudan stain of stool, or fecal human elastase testing. Because the

577

available elastase test distinguishes between human and nonhuman forms of the enzyme, this test can detect pancreatic insufficiency in patients who are receiving pancreatic enzyme replacement therapy, but it is less useful for monitoring responses to this therapy. In children and young adults with pancreatic insufficiency, a diagnosis of CF is most commonly established by the sweat test (quantitative pilocarpine iontophoresis) or by CF gene testing. In children with pancreatic insufficiency, the differential diagnosis includes SDS (an autosomal-recessive condition associated with neutropenia—see below) and rare enzyme deficiencies (e.g., congenital lipase deficiency). In adults, pancreatic insufficiency most often results from chronic pancreatitis. Although the most common cause of chronic pancreatitis is alcohol abuse, a substantial minority of cases are associated with an identifiable genetic, environmental, or surgical etiology, or remain idiopathic. The diagnosis of chronic pancreatitis is usually supported by evidence of pancreatic atrophy, ductal dilatation or calcification, or a mass/pseudocyst as detected by abdominal imaging procedures (sonogram, CT, MRI, magnetic resonance cholangiopancreatography, or endoscopic retrograde cholangiopancreatography). It is noteworthy that serum amylase and lipase levels are elevated in acute pancreatitis, but that these tests are often within or below normal limits in patients with pancreatic insufficiency or chronic pancreatitis, even during episodes of acute pain. GENETIC BASIS The genetics of CF is presented in Chapter 27. Briefly, the inheritance pattern of CF is autosomal-recessive, the gene causing CF was mapped to chromosome VII by linkage analysis, and this gene was identified by positional cloning. The protein product of this gene was named the CFTR based on the anticipated function of the protein. The CFTR gene contains 27 exons, spans 250 kb, and encodes an integral membrane protein containing 1480 amino acids. In many types of epithelial cells, CFTR regulates ion fluxes across the plasma membrane by functioning as a cAMP-regulated anion channel and as a regulator of other ion transport proteins. As an anion channel, CFTR conducts chloride, bicarbonate, or both, depending on the tissue. Since the CF gene’s discovery in 1989, roughly 1000 different CFTR mutations have been associated with CF. In European Caucasians, one mutation, ∆F508, accounts for 60–70% of CFcausing alleles. About half of US CF patients are ∆F508 homozygotes; these individuals have low levels of residual CFTR function (CT and 258+2T>C. The penetrance and phenotypic expression of patients with SBDS mutations is variable. In severe cases, pancreatic insufficiency with malabsorption, steatorrhea, or failure to thrive may develop in the first year of life. Mild cases of SDS are seen more often in young men, probably because of greater parental concern about short stature among males. Nearly half of SDS patients showed moderate age-related improvement in pancreatic function leading to pancreatic sufficiency. Cyclic neutropenia is also common

and typically leads to infections (e.g., recurrent otitis media, sinusitis, pneumonia, osteomyelitis, urinary tract infections, skin infections, and lymphadenitis). Less common findings include myelodysplastic syndromes, acute leukemias, and a variety of skeletal abnormalities. MOLECULAR PATHOPHYSIOLOGY CF pancreatic disease is associated with characteristic pathological and physiological abnormalities. Obstruction of the intralobular pancreatic ducts is an early pathological event and impaired pancreatic secretion of bicarbonate is an early physiological defect. These findings suggest that CFTR may normally function to prevent obstruction of the intralobular ducts and to promote bicarbonate secretion. These concepts are supported by available data concerning the localization of CFTR in human pancreas. The predominant site of CFTR mRNA and protein in pancreas is in the cells lining the proximal ductules. Within these duct epithelial cells, CFTR protein is predominantly detected as a component of the apical plasma membrane. The distribution of CFTR in pancreas supports the model shown in Fig. 55-2, which postulates an important role for CFTRassociated anion channels in the regulation of ductal secretion. According to this model, secretory stimuli act via cAMP to activate CFTR and thereby open Cl–/bicarbonate channels at the apical membrane of the duct cell. Because of the driving force provided by the Na+,K+-ATPase at the basolateral membrane, opening apical anion channels results in the movement of Cl– and bicarbonate across the apical membrane, leading to the movement of Na+ and water into the lumen, and thereby diluting the pancreatic juice. This model suggests that CFTR in the duct cell contributes to the normal dilution and alkalinization of pancreatic juice. In humans with CF, the ultimate result of the absence of CFTR at this site is a defect in the dilution and alkalinization of the protein-rich acinar secretions in the duct lumen, leading to failure to flush zymogens out of the duct and formation of protein plugs, thereby to pancreatic injury in CF. Uncertainty remains concerning the precise mechanism by which pancreatic injury results from defective dilution and alkalinization of the pancreatic juice in CF. Dilution and alkalinization

CHAPTER 55 / PANCREATIC EXOCRINE DYSFUNCTION

both may help to prevent precipitation of proteins as the pancreatic juice flows through the ducts. Beyond this, data from in vitro models suggests that alkalinization of the lumen of the pancreatic duct system may also be important for maintaining normal acinar cell function during exocytosis. Finally, ductal fluid secretion tends to flush zymogen proteins out of the pancreas. In CFTR-related idiopathic chronic pancreatitis, inadequate ductal CFTR function also probably plays an early role in pathogenesis. Maintaining flow through the ducts may be especially important in individuals who have increased risk for intraductal trypsin activation (e.g., resulting from PSTI mutations). When these individuals also have reduced CFTR function (e.g., resulting from CFTR mutations), flow through the pancreatic duct is slow and this promotes injury by prolonging pancreatic exposure to any activated trypsin that occurs in the duct lumen. Thus, mutations of the CFTR and PSTI genes tend to have additive (independent) impact on pancreatitis risk. This agrees well with the distinct sites in the exocrine pancreas for the protein products of these genes: CFTR is ductal whereas PSTI is acinar. In SDS, the primary site of pancreatic injury is the acinus rather than the duct. Acinar replacement is prominent; inflammation and gland destruction are atypical. Thus, the exocrine insufficiency of SDS mainly affects enzyme secretion rather than bicarbonate secretion, whereas the opposite pattern is seen in CF. MANAGEMENT Pain control is often the predominant issue in treating chronic pancreatitis. Structural causes for pain include ductal obstruction resulting from strictures, tumor, pseudocysts, or stones, and a parenchymal compartment syndrome in which exocrine tissue is encased by fibrosis. Pain resulting from structural causes is typically intermittent and often occurs when serum cholecystokinin levels increase after a fat- or protein-rich meal. Medical therapy for this type of pain can include using gastric acid suppression, pancreatic enzyme supplements, and narcotics to reduce pancreatic secretion, whereas surgical options include procedures designed to relieve areas of ductal obstruction or fibrotic encasement. Chronic pancreatitis can also cause pain because of inflammation. This pain can be severe and unremitting, and surgical resection and celiac ganglion ablation are important management options. Pancreatic insufficiency and diabetes mellitus are also common problems in chronic pancreatitis. Pancreatic enzyme replacement is the cornerstone of therapy for pancreatic insufficiency. Widely used enzyme preparations include a combination of lipase, amylase, and proteases. Because lipase is rapidly degraded at low pH, replacement enzymes are commonly administered either as enteric-coated microspheres or in combination with an H2-receptor antagonist. Moderate enzyme doses are usually sufficient to achieve partial correction of steatorrhea with marked improvement in nutrition. Complete correction of steatorrhea is usually not possible even when very high enzyme doses are administered. High enzyme doses should be used with caution because they have been associated with colonic strictures. The management of diabetes is reviewed in Chapter 32. The high prevalence of CFTR and PSTI mutations in idiopathic pancreatitis suggests that many patients with this condition could benefit from genetic testing. Potential benefits include helping patients understand their disease, promoting prompt detection of extrapancreatic CFTR-related conditions, identifying CF carriers for genetic counseling, and identifying occasional CF patients in whom pancreatitis occurs before the onset of lung disease. CFTR gene testing may also improve medical

579

care in cases in which physicians might otherwise suspect undisclosed alcoholism as the cause of pancreatitis. CFTR mutation testing is more likely to be positive in younger CF patients who have extrapancreatic evidence of reduced CFTR function (e.g., borderline sweat testing). False-negative tests are more likely in non-Caucasians because the assays widely used for CF carrier screening are less sensitive in these populations. It is also unclear if the limited number of CFTR gene mutations that are included in CF screening panels is sufficient for determining risk of pancreatitis. FUTURE DIRECTIONS Research is moving toward developing new methods to prevent idiopathic pancreatitis in highly susceptible individuals. Gene testing can already identify asymptomatic individuals in whom the risk of developing pancreatitis is increased by several 100-fold. As more is learned about the subsets of CFTR and PSTI genotypes associated with highest pancreatitis risk, the predictive power of gene testing will continue to improve, enabling identification of highly susceptible individuals prior to the onset of their first pancreatitis symptoms. Once identified, these individuals can be studied to learn about early events in the pathogenesis of this condition as a key step toward developing and evaluating preventive therapy. Specific questions being addressed by ongoing research include the following: Is the risk of pancreatitis increased in CF carriers who have one normal CFTR allele? Should CFTR-related idiopathic pancreatitis be classified as a form of CF? Can CFTR or PSTI gene testing be used to guide therapy in patients with pancreatitis? Do individuals with CFTR or PSTI mutations have increased susceptibility to alcoholic or drug-related pancreatitis?

SELECTED REFERENCES Bachem MG, Schneider E, Gross H, et al. Identification, culture, and characterization of pancreatic stellate cells in rats and humans. Gastroenterology 1998;115:421–432. Bhatia E, Choudhuri G, Sikora SS, et al. Tropical calcific pancreatitis: strong association with SPINK1 trypsin inhibitor mutations. Gastroenterology 2002;123:1020–1025. Boocock GR, Morrison JA, Popovic M, et al. Mutations in BDS are associated with Shwachman–Diamond syndrome. Nat Genet 2003;33:97–101. Cipolli M. Shwachman–Diamond syndrome: clinical phenotypes. Pancreatology 2001;1:543–548. Cohn JA, Friedman KJ, Noone PG, Knowles MR, Silverman LM, Jowell PS. Relation between mutations of the cystic fibrosis gene and idiopathic pancreatitis. N Engl J Med 1998;339:653–658. Durie PR. Inherited and congenital disorders of the exocrine pancreas. Gastroenterologist 1996;4:169–187. Etemad B, Whitcomb DC. Chronic pancreatitis: diagnosis, classification, and new genetic developments. Gastroenterology 2001;120:682–707. Frey CF. Current management of chronic pancreatitis. Adv Surg 1995; 28:337–370. Gorry MC, Gabbaizedeh D, Furey W, et al. Mutations in the cationic trypsinogen gene are associated with recurrent acute and chronic pancreatitis. Gastroenterology 1997;113:1063–1068. Kingsnorth A. Role of cytokines and their inhibitors in acute pancreatitis. Gut 1997;40:1–4. Kloppel G, Maillet B. Pseudocysts in chronic pancreatitis: A morphological analysis of 57 resection specimens and 9 autopsy pancreata. Pancreas 1991;6:266–274. Kusske AM, Rongione AJ, Reber HA. Cytokines and acute pancreatitis. Gastroenterology 1996;110:639–642. Marino CR, Matovcik LM, Gorelick FS, Cohn JA. Localization of the cystic fibrosis transmembrane conductance regulator in pancreas. J Clin Invest 1991;88:712–716. Mews P, Phillips P, Fahmy R, et al. Pancreatic stellate cells respond to inflammatory cytokines: Potential role in chronic pancreatitis. Gut 2002;50:535–541.

580

SECTION VI / GASTROENTEROLOGY

Noone PG, Zhou Z, Silverman LM, Jowell PS, Knowles MR, Cohn JA. Cystic fibrosis gene mutations and pancreatitis risk: Relation to epithelial ion transport and trypsin inhibitor gene mutations. Gastroenterology 2001;121:1310–1319. Papachristou GI, Whitcomb DC. Predictors of severity and necrosis in acute pancreatitis. Gastroenterol Clin North Am 2004;33:871–890. Pfutzer RH, Barmada MM, Brunskill AP, et al. SPINK1/PSTI polymorphisms act as disease modifiers in familial and idiopathic chronic pancreatitis. Gastroenterology 2000;119:615–623. Ranson JH, Rifkind KM, Roses DF, Fink SD, Eng K, Spencer FC. Prognostic signs and the role of operative management in acute pancreatitis. Surg Gynecol Obstet 1974;139:69–81. Rinderknecht H. Pancreatic secretory enzymes. In: Go VLW, Dimagno EP, eds. The Pancreas: Biology, Pathobiology, and Disease, 2nd ed. New York: Raven, 1993; pp. 219–251. Sainio V, Kemppainen E, Puolakkainen P, et al. Early antibiotic treatment in acute necrotising pancreatitis. Lancet 1995;346:663–667. Schneider A, Suman A, Rossi L, et al. SPINK1/PSTI mutations are associated with tropical pancreatitis and type II diabetes mellitus in Bangladesh. Gastroenterology 2002;123:1026–1030 Sharer N, Schwarz M, Malone G, et al. Mutations of the cystic fibrosis gene in patients with chronic pancreatitis. N Engl J Med 1998;339: 645–652.

Slaff J, Jacobson D, Tillman CR, Curington C, Toskes P. Protease-specific suppression of pancreatic exocrine secretion. Gastroenterology 1984;87:44–52. Stern RC, Eisenberg JD, Wagener JS, et al. A comparison of the efficacy and tolerance of pancrelipase and placebo in the treatment of steatorrhea in cystic fibrosis patients with clinical exocrine pancreatic insufficiency. Am J Gastroenterol 2000;95:1932–1938. Sutton R, Criddle D, Raraty MG, Tepikin A, Neoptolemos JP, Petersen OH. Signal transduction, calcium and acute pancreatitis. Pancreatology 2003;3:497–505. Whitcomb DC, Ermentrout DB. A mathematical model of the pancreatic duct cell generating high bicarbonate concentrations in pancreatic juice. Pancreas 2004;29:e30–e40. Whitcomb DC, Gorry MC, Preston RA, et al. Hereditary pancreatitis is caused by a mutation in the cationic trypsinogen gene. Nat Genet 1996;14:141–145. Whitcomb DC, Lowe M. Pancreatitis: acute and chronic. In: Walker WA, Durie PR, Hamilton JR, Walker-Smith JA, Watkins JB, eds. Pediatric Gastrointestinal Disease: Pathophysiology, Diagnosis, Management. Hamilton, BC: Decker, 2000; pp. 1584–1597. Witt H, Luck W, Hennies HC, et al. Mutations in the gene encoding the serine protease inhibitor, Kazal type 1 are associated with chronic pancreatitis. Nat Genet 2000;25:213–216.

56 Small and Large Bowel Dysfunction DEBORAH C. RUBIN SUMMARY The study of inflammatory bowel diseases (IBD) has advanced greatly, with the identification of the first disease susceptibility gene for Crohn's disease, the tentative identification of other susceptibility loci for IBD, the creation of new animal models of IBD using gene knock out and transgenic technology, and the application of cellular and molecular techniques to develop novel therapies for IBD. This chapter summarizes the important molecular and genetic developments relevant to the study of IBD, celiac sprue, lactase deficiency, glucose–galactose malabsorption, and abetalipoproteinemia as several prominent examples of how molecular technology can help advance the practice of gastrointestinal medicine. Key Words: Abetalipoproteinemia; celiac sprue; Chrohn's disease; glucose–galactose malabsorption; inflammatory bowel diseases (IBD); lactase deficiency; ulcerative colitis.

INTRODUCTION The application of molecular biological and genetic techniques to the study of the luminal gastrointestinal tract epithelium has led to exciting advances in several broad areas. The study of inflammatory bowel diseases (IBD) has advanced greatly, with the identification of the first disease susceptibility gene for Crohn’s disease, the tentative identification of other susceptibility loci for IBD, the creation of new animal models of IBD using gene knockout and transgenic technology, and the application of cellular and molecular techniques to develop novel therapies for IBD. The molecular mechanisms underlying intestinal digestion and absorption of nutrients have been further clarified by the cloning and characterization of enterocytic genes that play critical roles in the digestion and absorption of luminal nutrients. Their chromosomal localization, structure, and regulation of expression have been at least partially characterized. These studies have provided the groundwork for understanding the pathophysiology of several rare but welldescribed genetic disorders of small intestinal transport, and have begun to clarify the mechanisms underlying more common gastrointestinal problems such as lactase deficiency. This chapter summarizes the important molecular and genetic developments relevant to the study of IBD, celiac sprue, lactase deficiency, glucose–galactose malabsorption, and abetalipoproteinemia as several prominent From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

examples of how molecular technology can help advance the practice of gastrointestinal medicine.

INFLAMMATORY BOWEL DISEASE CLINICAL PRESENTATION The IBDs include Crohn’s disease and ulcerative colitis. These are chronic inflammatory disorders of the gastrointestinal tract that are manifested by symptoms of diarrhea and abdominal pain. Although the clinical presentation may be quite similar, important differences in their underlying pathophysiology result in unique complications and long-term sequelae. In ulcerative colitis, the colon is the primary affected organ, but Crohn’s disease can occur in any part of the gastrointestinal tract from mouth to anus. Ulcerative colitis is characterized by a predominantly superficial mucosal inflammation. In Crohn’s disease, the inflammatory process is transmural and granulomatous, affecting all layers of the gastrointestinal tract, from mucosa to serosa. The etiology of these disorders is unknown. The identification of C-terminal caspase recruitment domain (CARD15/NOD2) as a Crohn’s disease susceptibility factor has supported the roles of genetic factors and immune responses to luminal gut flora in its pathogenesis (see IBD Gene Discovery). A variety of disease-susceptibility chromosomal loci have been identified by linkage analysis of families with IBD. Several of these contain immune regulatory gene clusters. Studies of mice with targeted disruption of cytokine or T-cell receptor genes, or of chimeric transgenic mice in which there is disruption of the N cadherin gene or overexpression of tumor necrosis factor (TNF)-α, have provided promising new models of IBD. These novel systems may help clarify the etiology and pathogenesis of these disorders and aid in the design of new therapeutic agents. EPIDEMIOLOGY There are marked geographic differences in the incidence of IBD. The highest incidence rates are found in northern Europe and North America among Caucasians, ranging from 2 to 10/100,000. Prevalence of ulcerative colitis in these regions is 37–246 cases/100,000 person-years, and Crohn’s disease is 26–199 cases/100,000. The incidence of IBD in southern Europe, Asia, and the developing world is low but has been rising. IBD appears to be more common in Jews of European or Ashkenazic origin, but not of Sephardic (African and Mediterranean) origin. The peak age of onset is from 15–25 yr, and a second peak between 55 and 65 yr of age has been suggested. There is a slight female predominance in the prevalence of both disorders. Although the disease is most common in Caucasians, the incidence in African Americans appears to be increasing, and in some pediatric studies reaches that of Caucasians.

581

582

SECTION VI / GASTROENTEROLOGY

Asian and Hispanic Americans continue to have a lower incidence of IBD. Environmental risk factors include cigarette smoking, which increases the risk of Crohn’s disease but appears to be protective in ulcerative colitis. Other studies support the possible involvement of mycobacteria in Crohn’s disease pathogenesis. CROHN’S DISEASE Location of Disease, Signs, and Symptoms Crohn’s disease can affect any part of the gastrointestinal tract. The most frequently involved sites are the ileocolon (40%), the small intestine alone (30%), or the colon alone (25%). Other regions of the gastrointestinal tract may also be involved, including the mouth, esophagus, and stomach. Anal involvement is common. Inflammation may involve the bowel in a segmental fashion, with “skip” areas of normal small and large intestine intermixed with patches of inflamed mucosa. Symptoms resulting from the inflammatory process include abdominal pain (especially in the right lower quadrant owing to terminal ileal involvement) and diarrhea. Other frequent symptoms include fever, weight loss, and malaise. A variety of extraintestinal manifestations including arthritis, uveitis/iritis, and skin diseases such as erythema nodosum and pyoderma gangrenosum are associated with IBD. Signs of active Crohn’s disease include right lower quadrant abdominal fullness, mass or tenderness, and a chronically ill appearance. There are multiple causes of diarrhea in Crohn’s disease. Enhanced secretion occurs in the small bowel and colon because of inflammation. Patients with ileal disease or ileal resection may malabsorb bile salts, as specific receptors and transport proteins are expressed only in the ileum. Unabsorbed bile acids then enter the colon in which they produce a secretory diarrhea. Extensive terminal ileal involvement or resection of greater than 100 cm of ileum results in fatty acid diarrhea because of excessive loss of bile salts that cannot be replaced by enhanced hepatic production. Fat malabsorption may also result from bacterial overgrowth because of strictures of the small bowel leading to deconjugation of bile salts. Crohn’s disease has also been associated with lactase deficiency, resulting in maldigestion of milk and milk products. Undigested lactose then produces an osmotic diarrhea. Complications The transmural inflammatory process characteristic of Crohn’s disease leads to many complications. Inflammation may lead to ulceration through all layers of the bowel, resulting in the formation of fistulous tracts. Fistulas can form between the gut and any neighboring organ. Enteroenteric, enterocolonic, and enterocutaneous fistulas are frequently found in Crohn’s disease. Less common are enterovesicular and enterovaginal fistulae. Strictures of the small or large bowel may result from chronic inflammation and can lead to intestinal obstruction. If severe and unremitting despite conservative management, strictures may require surgical resection to permit the continuation of normal enteral nutrition. Perforation of the small bowel or colon may precipitate the onset of an acute abdomen requiring immediate surgical intervention. Alternatively, perforations and fistulas may produce intra-abdominal abscesses, another common complication of Crohn’s disease. Abscesses may occur anywhere in the peritoneal cavity and must be drained to control sepsis. Patients who have undergone surgical resection or who have extensive small bowel inflammation, fistulas, or strictures may suffer from short bowel syndrome with excessive diarrhea, volume loss, and malabsorption. These patients frequently require treatment with total parenteral nutrition to maintain a normal nutritional and fluid and electrolyte status. As indicated, loss of

functional ileum may lead to fat malabsorption. Because of these complications, surgery is avoided whenever possible. Kidney stones are most frequently made up of oxalate and are found in patients with ileal disease and fat malabsorption. Calcium, which normally binds to oxalate to produce insoluble calcium oxalate, is instead bound by fatty acids. This leads to the formation of sodium oxalate, which is readily absorbed through the colon, leading to hyperoxaluria and stone formation. ULCERATIVE COLITIS Location of Disease, Signs, and Symptoms Ulcerative colitis is characterized by a continuous mucosal inflammatory process affecting the colon. Inflammation may be confined to the rectum alone (ulcerative proctitis), may extend from the rectum to the left colon, or may involve the entire colon. Unlike Crohn’s disease, at least part of the rectum is always involved and the inflammatory process is continuous in affected areas (i.e., there are no “skip” regions). Abdominal pain, diarrhea, and gastrointestinal bleeding are common clinical presentations of this disorder. Anorexia and weight loss are occasionally present but are not common. Unlike Crohn’s disease, which frequently follows a low-grade, chronic course, ulcerative colitis patients may present with fulminant colitis necessitating urgent colectomy. Physical findings include abdominal tenderness and grossly bloody stool on rectal examination. In fulminant cases, abdominal distention, massive hemorrhage, fever, and tachycardia may be present. Complications Toxic megacolon may occur in patients with acute fulminant colitis. The colon becomes atonic and dilated resulting from severe inflammation and can perforate if untreated. Toxic megacolon requires emergent surgical consultation and in many cases, removal of the colon. Strictures may result from chronic inflammation. Colon cancer is an important complication of long-standing ulcerative colitis. Although the precise risk of colon cancer is not clear, it is increased in patients with disease for greater than 8–10 yr, and is higher in patients with pancolitis compared with left-sided colitis alone. To prevent colon cancer either prophylactic colectomy or yearly colonoscopic screening for dysplasia/cancer is recommended depending on the patient’s preference and disease activity. An important advance has been the development of the ileoanal anastomosis. Instead of an ileostomy, bowel continuity is maintained by creating an ileal reservoir in place of the rectum. This has led to greater acceptance for surgical options in the treatment of ulcerative colitis. Extraintestinal Manifestations There are multiple extraintestinal manifestations of IBD, including arthritis, skin and ocular diseases, liver involvement, kidney stones, and osteoporosis. The arthritic complaints include colitic arthritis, ankylosing spondylitis, and sacroiliitis. The large joints are generally involved in colitic arthritis. Arthritic complaints generally respond to adequate treatment of the bowel disease. Symptoms resulting from ankylosing spondylitis and sacroiliitis may be severe and more refractory to treatment. Skin manifestations of IBD include pyoderma gangrenosum and erythema nodosum. Ocular diseases are predominantly uveitis and episcleritis. Liver involvement may manifest as sclerosing cholangitis or pericholangitis. Cholesterol gallstones occur in Crohn’s disease if there is ileal involvement or after resection, resulting in bile salt malabsorption. Finally, kidney stones may form in patients with small intestinal Crohn’s disease and fat malabsorption, and are usually made up of calcium oxalate or urate. Osteoporosis may result from chronic fat malabsorption leading to calcium and vitamin D deficiency and from chronic corticosteroid therapy.

CHAPTER 56 / SMALL AND LARGE BOWEL DYSFUNCTION

DIAGNOSIS History and Physical Examinations A history of bloody diarrhea, abdominal pain, and fever should lead to a strong suspicion for ulcerative colitis. Symptoms in Crohn’s disease are frequently more indolent and chronic, including intermittent diarrhea and fevers, abdominal pain, and weight loss. Physical findings in ulcerative colitis may include abdominal tenderness, and in fulminant disease, distention, decreased bowel sounds and an acute abdomen. In Crohn’s disease, abdominal tenderness and an inflammatory abdominal mass may be present, particularly in the right lower quadrant. The perianal and perirectal regions should be examined for signs of fistulization and abscess. Signs of extraintestinal disease, including arthritis, ocular inflammation, and skin nodules or ulcers may be detected. Laboratory and Pathological Examinations There is no specific laboratory test that is diagnostic of ulcerative colitis or Crohn’s disease. An elevated white blood cell count can be seen in active disease or may be an indicator of sepsis because of perforation or abscess formation in Crohn’s disease. Anemia may result from gastrointestinal bleeding or nutrient deficiencies (e.g., vitamin B12 owing to ileal Crohn’s disease, decreased iron absorption in duodenal Crohn’s disease). Erythrocyte sedimentation rate may be elevated in active disease. Nutritional parameters such as blood total protein and albumin levels may also be deranged in Crohn’s disease. Pathological specimens (endoscopic biopsies or surgical resection) may reveal evidence of active inflammation. Characteristic of colitis are crypt abscesses, in which neutrophils invade the crypts. Granulomas are more characteristic of Crohn’s disease. Crypt atrophy and distortion are also seen, as is acute and chronic inflammation with lymphocytes and eosinophils, and mucin depletion. Ulceration of the mucosa occurs in ulcerative colitis, but the presence of transmural inflammation is most characteristic of Crohn’s disease. Radiological and Endoscopic Studies Evaluation of the mucosa for the presence of active disease can be achieved by endoscopic examination. In patients with symptoms of colitis, colonoscopy may aid in the initial diagnosis and is required to determine the extent of disease activity. The mucosal surface is observed for the presence of erythema, edema, granularity, friability, and ulceration. The extent of colonic involvement may help differentiate Crohn’s disease from ulcerative colitis. The presence of “skip” areas, in which normal mucosa is intermixed with inflamed mucosa, is most characteristic of Crohn’s disease, whereas in ulcerative colitis, inflammation is continuous. During colonoscopy, the terminal ileum may be entered and evaluated for involvement by Crohn’s disease. Direct visualization of the upper gastrointestinal tract by routine upper endoscopy or small bowel enteroscopy may also help to establish the presence of Crohn’s disease. Radiological examination is an important part of the diagnostic evaluation of patients with IBD. Contrast studies are useful for detecting strictures and fistulae, as well as assessing disease activity in regions of the small intestine that are inaccessible to routine endoscopy. Patients with ulcerative colitis may demonstrate typical findings on barium enema, including straightening and shortening of the colon, loss of normal haustrations, and inflammatory polyps or malignancy. Mucosal detail is more difficult to assess with radiography, and endoscopic examination is generally preferred. In patients with very active colitis, radiographic examination is avoided because of the fear of inducing toxic megacolon owing to the introduction of air and barium. Similarly, colonoscopic

583

testing is performed with caution and usually in a limited manner. Computerized axial tomographic scanning is an important diagnostic modality particularly for abscess detection in Crohn’s disease. If accessible, CT-directed drainage of abscess collections may be performed. GENETIC BASIS As confirmed by the discovery of the CARD15/NOD2 gene mutation that confers susceptibility to Crohn’s disease (see IBD Gene Discovery), genetic factors clearly play a role in the pathogenesis of IBD. Genetic predisposition to Crohn’s disease was suggested by studies indicating an increased incidence in families, specific ethnic groups, and in twins. Of patients with IBD, approx 10–20% have a family history, and the risk for first degree relatives to develop IBD is 2–5%. Twin studies indicate that monozygotic twins have a higher concordance rate than dizygotic twins. The complex clinical manifestations of these diseases suggest a heterogeneous genetic basis. MOLECULAR PATHOPHYSIOLOGY The etiology of the IBDs is unknown. One widely accepted hypothesis is that IBD results from an abnormal host immune response to normal gut components such as microflora, nutrients, or cellular constituents. Intestinal bacterial flora is postulated to be particularly important in activating the mucosal immune system in IBD. This may also be accompanied by a primary defect in gut permeability that allows antigens to nonspecifically “leak” through the mucosa and thereby stimulate an inflammatory response. IBD GENE DISCOVERY: CARD15 AND IBD SUSCEPTIBILITY GENES A major advance in the study of Crohn’s disease pathophysiology has been the identification of CARD15/NOD2 as the first Crohn’s susceptibility gene. CARD15/NOD2: encodes an intracellular protein that is involved in bacterial detection via peptidoglycan recognition. CARD15/NOD2: may play an important role in generating a protective response following exposure to bacteria. There are three genetic variants associated with Crohn’s disease; 10–30% of patients with Crohn’s disease are heterozygotes for one of these variants, and a smaller percentage of patients are homozygous or compound heterozygotes for these mutations. Yet, even homozygotes have only a 1 in 25 risk for developing disease, indicating the requirement for additional genetic and/or environmental factors. Interestingly, CARD15/NOD2: mutations are most consistently associated with the presence of ileal disease. Other susceptibility loci for IBD have been identified by whole genome scanning techniques. There are presently seven loci, termed IBD1-7, that meet linkage criteria. These chromosomal regions include, for example, human leukocyte antigen (HLA) and cytokine gene clusters, which have been postulated to fit the role of disease susceptibility genes. However, the specific genes and mutations have not been identified. Animal Models of IBD The specific role of a gene can be determined in the whole animal in “loss of function” experiments that utilize embryonic stem cell techniques. Mutations in specific target genes can be created in these cells to delete the gene or produce a sequence that encodes a nonfunctional protein. The mutated stem cells are injected into a mouse blastocyst and grown in a host mouse. Heterozygous mice are then backcrossed to create homozygous mutants. In this manner, targeted deletion of various immune mediator genes has been performed. Interestingly, when interleukin (IL)-2, IL-10, major histocompatibility complex (MHC) class II or T-cell receptor α- or β-genes are deleted, gross inflammation of the bowel is produced in mice. IL-2 is produced by activated T cells and is an important regulator of immune responses. It increases T-cell

584

SECTION VI / GASTROENTEROLOGY

Table 56-1 Comparison of Crohn’s Disease and Ulcerative Colitis Clinical feature Site of involvement Pathology Extraintestinal manifestations

Complications

Treatment

Crohn’s disease

Ulcerative colitis

Entire gastrointestinal tract Transmural inflammation, granulomas Skin

Colon only Mucosal inflammation Skin

Eyes Joints Strictures, fistulae, abscesses

Eyes Joints Strictures, colon cancer, toxic megacolon Surgical removal of diseased colon is curative

Surgical removal of diseased segments avoided because of disease recurrence

proliferation, leads to differentiation of B cells, and activates macrophages, natural killer cells, and lymphokine-activated killer cells. IL-10 is produced by T-helper cell subset 2, macrophages, keratinocytes, Ly-1 B cells, and thymocytes. It increases MHC class-II expression on B cells, and suppresses macrophage activation in vitro. IL-10 inhibits IL-1, IL-6, and TNF-α production by activated macrophages. Deletion of the IL-2 or IL-10 gene produces inflammation of the intestine, but with different clinical features. Deletion of IL-2 led to colonic disease, with diarrhea, rectal prolapse, and bleeding. Pathological examination of the colon showed crypt abscesses, ulceration, goblet cell depletion, and epithelial dysplasia. MHC class-II expression, normally limited to the small intestine, was seen in the colon. In contrast, deletion of IL-10 led to a chronic enterocolitis affecting the entire gastrointestinal tract with growth retardation and iron deficiency anemia. Mucosal inflammation with villus loss, mucosal gastrointestinal erosions, and crypt hyperplasia were found in small intestine and colon. MHC class-II expression was aberrantly enhanced in a small bowel and colonic epithelium. The onset of disease in the majority of mice was relatively rapid; by 3–4 wk, growth retardation was noted. Interestingly, in both models, maintenance in a pathogen-free environment greatly diminished the intestinal inflammation. IL-10 mice had disease limited to the proximal colon only, and IL-2 mice were free of disease. A third model of IBD was generated in T-cell receptor α- or β-mutants, which lack αβ-T cells. These mice developed colonic inflammation but no diarrhea or bleeding. TNF has been implicated as a key pathogenic mediator in Crohn’s disease and a model of TNF overexpression has been developed. These TNF∆ARE mice bear an endogenous deletion in the 3′AU rich region in the TNF gene resulting in high circulating TNF levels. These mice develop a specific Crohn’s disease-like phenotype with chronic transmural ileitis. Not only can immune cell defects produce intestinal inflammation, but disruption of the protective epithelial barrier may also lead to IBD. A dominant-negative N-cadherin mutant was linked to an intestine specific promoter and thus was selectively disrupted in the gut, in villus as well as crypt cells. This resulted in markedly decreased expression of E-cadherin with gross disruption of the epithelial barrier and severe mucosal inflammation. The small intestine was selectively affected with lymphoid hyperplasia, blunted villi, cryptitis and crypt abscesses, mucosal ulcerations, and eventually adenoma formation.

Although these models of IBD are diverse, they support several hypotheses regarding the etiology of these disorders. Disruption of the epithelial barrier can lead to intestinal inflammation even in the presence of a normal immune system. Selective immune defects that lead to the unopposed action of B cells can produce inflammation of the gut. Also the intestine is clearly vulnerable to inflammation because of constant external antigen exposure and the presence of luminal substances. MANAGEMENT Medical Therapy Because the underlying cause of IBD is still unknown, therapy is presently directed toward controlling the intestinal inflammatory response (Table 56-1). One of the mainstays of treatment is oral and enema compounds containing 5-aminosalicylic acid (5-ASA), available in several different forms. These compounds are effective in the treatment of mild-tomoderate ulcerative or Crohn’s colitis as single agents. They are used in conjunction with corticosteroids in the treatment of severe disease. Although their precise mechanism of action is unknown, they are thought to inhibit leukotriene synthesis via their inhibition of neutrophil lipoxygenase, particularly leukotriene B4. Sulfasalazine was the first of these agents and consists of 5-ASA bound to a sulfapyridine residue to prevent its premature absorption in the small intestine. Colonic bacteria are required to release the 5-ASA component. However, many side effects result predominantly from the sulfapyridine component, including doserelated effects, such as headache, nausea, vomiting, anorexia, fever, and rash. Idiosyncratic, hypersensitivity responses include rash, hemolytic and aplastic anemia, and agranulocytosis. Also, long-term usage in young men is frequently limited by its adverse effects on sperm count and morphology. Therefore, several compounds have been formulated containing 5-ASA alone, with various modifications to facilitate its delivery to the small intestine and colon, or preferentially to the ileum and colon, without excessive systemic absorption before reaching its preferred site. These include Pentasa, Asacol, and Dipentum. Pentasa and Asacol are released by pH-dependent mechanisms. Pentasa is an ethylcellulose-coated form of mesalamine (5-ASA) that is released throughout the small intestine and colon. Asacol is coated with an acrylic-based resin that dissolves at pH 7.0 or greater, facilitating its release only at the terminal ileum and colon. Dipentum consists of two 5-ASA molecules linked by an azo bond that is cleaved in the colon. These newer formulations clearly have fewer side effects than sulfasalazine and, thus, have

CHAPTER 56 / SMALL AND LARGE BOWEL DYSFUNCTION

led to an expanded pharmacological armamentarium. Mesalamine in enema form (Rowasa enemas) is a valuable treatment for ulcerative proctitis. Corticosteroids are important in the therapy of moderate to severe IBD because of their broad anti-inflammatory actions. They are used in the treatment of active ulcerative colitis as well as Crohn’s disease of the small and large bowel. Owing to the numerous side effects attendant with chronic glucocorticoid usage, longterm therapy for IBD is avoided. There is no data supporting its use in maintaining disease remission, unlike 5-aminosalicylate containing compounds in colitis. Antibiotics are critical for the treatment of infectious complications of Crohn’s disease, such as sepsis, abscess formation, and small bowel bacterial overgrowth. Metronidazole has been used successfully to treat perianal disease and colonic Crohn’s disease. With the recognition of the importance of bacterial flora in Crohn’s pathogenesis, antibiotics have increasingly been used as treatment for active disease. Immunosuppressive agents such as azathioprine and its metabolite, 6-mercaptopurine, are used in Crohn’s disease and to a lesser extent, in ulcerative colitis, for steroid-dependent and poorly controlled patients. Treatment is generally begun in patients who cannot be tapered from corticosteroids because of recurrent symptoms and who have major side effects, such as osteoporosis, hypertension, cataracts, glucose intolerance, and so on. Alternatively, immunosuppressive agents may be added to treat the patient with refractory disease inadequately treated with corticosteroid and 5-ASA compounds. Immunosuppressive agents are also used in patients with ulcerative colitis and refractory proctitis, but less commonly. Other agents for active ulcerative colitis and Crohn’s disease include methotrexate and cyclosporine. There are data that support the use of intravenous cyclosporine for severe colitis as a last-ditch medical effort to avoid surgical intervention. Methotrexate may be used as an alternative medication for steroid-dependent patients to induce and maintain a steroid-free remission. The addition of infliximab, a chimeric monoclonal antibody to human TNF-α, has remarkably expanded the Crohn’s disease treatment repertoire and is an excellent example of how understanding the molecular pathophysiology of disease leads to improved treatment regimens. As indicated, TNF is a major proinflammatory immune regulator with a prominent role in Crohn’s disease. Infliximab infusion can induce remission in refractory Crohn’s disease and can heal fistulous disease. Optimal treatment regimens, designed to maximize efficacy and minimize side effects, are being developed. This drug has been approved to maintain remission in Crohn’s disease. Surgical Treatment In fulminant ulcerative colitis that is refractory to medical therapy, total colectomy may be unavoidable. In addition, surgical resection may be required for corticosteroid-dependent patients with ulcerative colitis who have major steroid-induced side effects. Colectomy may also be performed in patients with colonic dysplasia or carcinoma, or prophylactically to prevent colon cancer in patients with >10 yr history of active ulcerative colitis. As an alternative to prophylactic colectomy, yearly colonoscopic screening for colon cancer may be performed after 8–10 yr of disease. The major problems with this approach are that cancers in ulcerative colitis arise from flat mucosa and, therefore, may be missed and accurate grading of dysplasia is difficult. Colectomy is curative in ulcerative colitis and ileoanal anastomotic procedures eliminate the need for an

585

ileostomy. This surgical procedure is not an option for patients with Crohn’s disease, as there is a high incidence of recurrent disease in the ileal pouch leading to diarrhea, bleeding, pain, severe malfunction of the pouch, and eventually, the need for revision. In addition, surgery is not curative in Crohn’s disease because it recurs commonly in all parts of the intestinal tract. Therefore, the approach to patients with Crohn’s disease is very cautious and surgical intervention is avoided whenever possible. However, limited resection is often performed to relieve obstruction resulting from strictures to repair perforations or for abscess drainage. FUTURE DIRECTIONS Significant research developments include the generation of animal models that recapitulate many of the features of human IBD, the identification of CARD15 mutations in CD, and the localization of new chromosomal susceptibility loci for IBD. The novel mouse models have proven useful for elucidating the pathogenesis of these disorders and for designing more effective therapies. The identification of the gene(s) that are responsible for the familial risk of IBD will undoubtedly continue to be a major research focus, as the key to developing new diagnostic and therapeutic approaches.

CELIAC SPRUE Celiac sprue (also called celiac disease or gluten-sensitive enteropathy) is a chronic disease of the small intestinal mucosa resulting from exposure to dietary proteins contained in wheat, rye, and barley. The alcohol soluble component of wheat-derived gluten, or gliadin, and similar proteins derived from rye, barley, and oats, are toxic to the mucosa of susceptible individuals. In classic celiac sprue, the small bowel demonstrates complete villus atrophy and marked crypt hyperplasia, resulting in malabsorption and diarrhea. It is clear that specific host genetic factors are required in combination with exposure to gliadin to damage the epithelium. Linkage to histocompatibility antigen loci has been recognized for many years. The antigen recognized by the antiendomysial antibody has been identified as tissue transglutaminase. Data suggest that deamidation of gliadin proteins by tissue transglutaminase is required for the generation of epitopes that are recognized by CD4+ T cells. A specific 33 amino acid peptide has been identified that is derived from alpha gliadin by digestion with gastric and pancreatic enzymes, is not further proteolyzed into smaller fragments despite continued exposure to these enzymes, reacts with tissue transglutaminase and induces gut-derived T cells from patients with celiac sprue. These discoveries may have therapeutic implications to permit the development of a novel, antigen-specific immunotherapy. CLINICAL PRESENTATION Patients with celiac sprue exhibit a broad spectrum of clinical presentations. In children, the disease tends to occur in the first 3 yr of life. In adults, the peak incidence is in the third decade, but may occur at any age. In the classic clinical presentation, extensive loss of normal villi throughout the small intestine leads to gross malabsorption and steatorrhea, severe weight loss, wasting, multiple vitamin deficiencies, and abdominal distention. The vitamin deficiencies may lead to anemia (B12, folate, iron), or a bleeding diathesis (vitamin K). Neurological symptoms such as peripheral neuropathy and paresthesias are also common. In other cases, patients may be asymptomatic or complain only of mild abdominal discomfort, malaise, bloating, and diarrhea. Those with more patchy proximal intestinal involvement may present simply with iron or folate deficiency, or osteomalacia. Relatives of patients with celiac sprue who are screened for the disease often have mild illness or may be completely asymptomatic.

586

SECTION VI / GASTROENTEROLOGY

Celiac sprue is frequently associated with other disorders. Dermatitis herpetiformis is characterized by severely pruritic papulovesicular lesions of the skin. The majority of afflicted patients show intestinal biopsy findings characteristic of celiac sprue. These patients are often only mildly affected by the gut lesion and may even be asymptomatic. Isolated nutrient deficiencies may be the only clinical manifestation. Both the intestinal and skin lesions resolve with gluten withdrawal. Other diseases that are associated with celiac sprue include insulin-dependent diabetes mellitus (50-fold higher incidence of celiac disease in patients with diabetes), thyroid disease, and selective IgA deficiency, which share similar HLA haplotypes. Complications of celiac sprue include the development of T-cell lymphoma or other gastrointestinal malignancies. Ulcerative jejunoileitis is a serious complication resulting in ulceration and stricture of the small bowel with bleeding, perforation, and obstruction. Finally, patients may become refractory to removal of gluten from the diet and may require treatment with corticosteroids or immunosuppressive agents. EPIDEMIOLOGY Celiac sprue is most prevalent in Caucasians, particularly in Northern Europe. The highest prevalence has been reported in Ireland at a frequency of 1/300 but the more typical incidence in Europe is 1–2/10,000 population. Asians from India and Pakistan are also frequently affected, but the disease is rare in Japanese, Chinese, and Africans. A study of Minnesota residents who were symptomatic and had celiac sprue documented by biopsy and response to a gluten-free diet revealed an incidence of celiac sprue of 1.2/100,000 person-years in the United States, with a prevalence of 21.8/100,000. Studies have indicated a decrease in the prevalence of childhood celiac sprue in several European countries, perhaps because of changes in feeding or other unidentified environmental factors. DIAGNOSIS A valuable test for the diagnosis of celiac sprue is the small intestinal biopsy. Forceps biopsies of the duodenum or proximal jejunum can be obtained during routine upper endoscopy or peroral capsule biopsy techniques may be used. Biopsy should be performed before the initiation of gluten withdrawal and can also be repeated after removal of gluten from the diet to document an adequate response. Typical findings on biopsy are absence or atrophy of villi, crypt elongation and hyperplasia, and increased immune cell infiltrate. There is a marked increase in intraepithelial lymphocytes and infiltration of the lamina propria with lymphocytes, eosinophils, and macrophages. Because the biopsy findings in celiac sprue are not unique and can be seen in tropical sprue, lymphoma, viral or eosinophilic gastroenteritis, common variable immunodeficiency, or parasitic infection, the clinical response to removal of gluten in conjunction with biopsy is critical for definitively making the diagnosis. Rechallenge with gluten and rebiopsy is rarely recommended (except in equivocal cases). Antibody tests are used in conjunction with small bowel biopsy to confirm the diagnosis. Serum antigliadin antibodies are positive in approx 90% of patients with untreated sprue but are also found in other disorders such as sarcoidosis and rheumatoid arthritis. Antiendomysial antibodies, which are recognized to be directed against the autoantigen tissue transglutaminase, are also elevated in patients with histological abnormalities on biopsy, detecting more than 95% of untreated patients. Tissue transglutaminase antibodies are virtually pathognomonic of celiac disease. GENETIC BASIS Genetics plays an important role in celiac sprue, but in conjunction with environmental and other factors.

Approximately 8–12% of family members of patients with celiac sprue are also affected, as determined by intestinal biopsy. Interestingly, only 50% of all family members with abnormal biopsies manifest symptoms of the disease. In identical twins, there is 70% concordance for celiac sprue, indicating that factors other than genetics play a role in disease susceptibility. The linkage of celiac sprue with specific MHC genes has been recognized for years. There is a strong association with HLA class-II haplotypes, particularly HLA DQ2 and HLADQ8. In Northern Europeans, the DR w17 and DQ w2 serological specificities are very common. HLA DQ2 is present in 90% of patients with celiac sprue. Molecular analysis indicates that a specific DQ α- and β-chain molecule encoded by DQA1*0501 and DQB1*0201 is found in 99% of Scandinavians with sprue. In southern European populations, the same HLA-DQ molecule is expressed on an HLA-DR7 and DR5 (or DR11/DR7, DR12/DR7) haplotype, as opposed to HLA DR3, DQ2 (or HLADR17) in northern Europeans. These molecules are necessary but not sufficient for developing the disease, because HLA identical siblings have only 40–50% concordance for this disease, and only a small percentage of the population that carries these alleles develops sprue. In addition, some populations do not demonstrate the same very high percentage of patients with HLA-DQ2, but instead are DR4 positive. Sequence analysis of the HLA DQ alleles from patients with celiac disease revealed no mutations when compared with normal individuals. This suggests that molecules encoded in regions outside of this HLA locus are required for generating the disease and there may be different genetic subgroups. MOLECULAR PATHOPHYSIOLOGY The biochemistry of gluten and gliadin has been intensively studied to identify the protein moieties that are responsible for toxicity. The alcohol soluble component of gluten, gliadin, appears to be responsible for the disease. Several reports have shown that there are three peptides derived from alpha gliadin (located between amino acids 57 and 75, with the sequences PFPQPQLPY, PQPQLPYPQ, and PYPQPQLPY) that are deamidated by tissue transglutaminase and are recognized by CD4+ T cells from patients with celiac disease. Reports suggest that deamidation is required for T-cell stimulation, as it results in the production of negatively charged amino acids that are required for binding to the HLA-DQ2 molecule. A 33 amino acid peptide was identified that is derived from alpha gliadin by digestion with gastric and pancreatic enzymes, is not further proteolyzed into smaller fragments despite continued exposure to these enzymes, reacts with tissue transglutaminase and induces gut-derived T cells from patients with celiac sprue. However, it is likely that other peptides can also induce disease, as indicated by the diversity of antigenic responses in children. Based on its sequence homology to A-gliadin, another environmental factor that may play a role in the pathogenesis of celiac sprue is a history of adenovirus 12 infection. There is a region of homology between the adenovirus serotype 12 E1b protein to A-gliadin that spans 12 amino acids, with eight identical residues. In addition, almost 90% of patients with untreated celiac sprue had serological evidence of prior adenovirus 12 infection, compared with only 17% of disease controls. These data suggest that mechanisms of molecular mimicry may be involved in the pathogenesis of this disorder, in which T cells recognize peptides in common to gliadin and viral proteins. However, adenovirus 12 DNA has not been detected in intestinal samples, and peptide challenge with an

CHAPTER 56 / SMALL AND LARGE BOWEL DYSFUNCTION

oligopeptide derived from the homologous region has not been shown to produce intestinal injury. MANAGEMENT Removal of all gluten from the diet is the primary therapy for celiac sprue. Wheat, rye, oats, and barleycontaining products must be eliminated from the diet because gluten and other related prolamins can produce injury. This restricted diet must be maintained lifelong. Patients may initially exhibit lactose intolerance because of villus atrophy and brush border enzyme deficiency, but this should resolve with treatment.

DISORDERS OF SMALL INTESTINAL ABSORPTION AND TRANSPORT The brush border of the small intestinal enterocyte contains disaccharidases, peptidases, and transport proteins that are responsible for the absorption of ingested nutrients. Progress has been made in clarifying the molecular basis of several disorders characterized by selective malabsorption of carbohydrates or lipids. This section presents a summary of advances in understanding the pathophysiology underlying several of the more common of these defects. LACTASE DEFICIENCY Clinical Features and Epidemiology Lactase is one of the best-studied of the brush border enzymes. It is responsible for the hydrolysis of lactose into glucose and galactose. Several clinical categories of lactase deficiency have been described. Congenital lactase deficiency is a rare autosomal-recessive disorder, resulting in milk intolerance from birth. Much more common is the onset of diminished enzyme activity in childhood or adolescence and lasting into adulthood, also known as adult-onset hypolactasia. Loss of brush border lactase activity may also result from other small intestinal disorders, such as Crohn’s disease, celiac sprue, viral illnesses, or other infections. Loss of lactase activity beginning in childhood/adolescence occurs in the majority of humans throughout the world except for Caucasian populations. In particular, most white northern Europeans and their descendants in North America and Australia (populations that were traditionally cattle farmers and thus persistent milk drinkers) retain high lactase activity lifelong. The prevalence of lactase deficiency throughout the world supports the hypothesis that the decline in lactase activity with age is a normal event in response to weaning, common to all land mammals. Persistence of lactase activity has therefore been postulated to be the result of a genetic mutation that occurred in milk-drinking populations, leading to a survival advantage because milk-drinkers might be healthier. Ingestion of milk and milk products in deficient individuals leads to abdominal pain, diarrhea, bloating, and excessive flatulence because of unabsorbed lactose in the lumen. This results in increased small bowel fluid. Fermentation of lactose in the colon produces hydrogen, methane, and carbon dioxide gas. Removal of lactose-containing products in the diet leads to resolution of these symptoms. Diagnosis A lactose tolerance test may be performed, which consists of administering 50 g of lactose orally. If sufficient lactase activity is present, blood glucose levels will rise by more than 20 mg/dL during a 2-h period. Lactase deficiency is diagnosed by the lack of serum glucose elevation and frequently, the onset of symptoms of malabsorption, with gas, diarrhea, bloating, and abdominal pain. If the serum glucose increases yet symptoms also occur, one must consider occult diabetes mellitus. Alternatively, a

587

trial of dietary lactose avoidance may be attempted. After restriction of oral intake of milk and milk products, the patient is assessed for symptom resolution. Finally, small bowel biopsy allows for definitive diagnosis, but is unnecessary given the adequacy of the other two approaches. Genetic Basis and Molecular Pathophysiology As determined by population and family studies, adult-onset hypolactasia is an autosomal-recessive trait, and persistence of lactase activity is autosomal-dominant. The molecular basis for lactase persistence/ nonpersistence has been the subject of intensive investigation. In the majority of studies, lactase-specific activity and protein levels correlate very well with mRNA, indicating transcriptional regulation of lactase expression. A single-nucleotide polymorphism located at –14 Kb upstream from the start of lactase transcription, in the intron of another neighboring gene known as minichromosome maintenance (which plays a role in cell-cycle control), is highly associated with lactase persistence in Finnish populations. Individuals with low lactase activity were homozygous for a cytidine at this site, whereas lactase persistent individuals had either C/T or T/T. Another association was noted at a single-nucleotide polymorphism at –22018, with a G to A substitution. These substitutions were also associated with higher levels of lactase mRNA expression. Thus cis-acting elements appear to be responsible for the persistence/nonpersistence phenotype, based on differences in level of expression of lactase from one allele compared with the other. Studies have also focused on characterizing transacting factors that regulate lactase expression, because, for example, levels of NF-LPH-1, an intestine-specific transactivator that binds to a sequence close to the TATA box, are high in newborn pigs with high lactase activity and low in adult pigs with low lactase activity. Still other data suggest that there is a second phenotype of lactase deficiency, in which post-translational processing of lactase-phlorizin is altered. Thus, the precise molecular mechanisms underlying lactase persistence/nonpersistence are under investigation. Management The treatment of symptomatic lactase deficiency consists of removal of lactose-containing nutrients from the diet (Table 56-2) and/or the use of oral enzyme replacement therapy. It is preferable to continue some milk and dairy food ingestion because of their rich calcium stores, especially in children and in women who are prone to osteoporosis. Lactase oral preparations are available in capsules or tablets and can be chewed. Replacement pills are taken with meals and titrated to diminish symptoms. Lactase may also be added directly to milk and ingested after an overnight incubation, or pretreated milk can be purchased. Future Directions The ready availability of oral lactase capsules and the ease of treatment of symptoms by dietary avoidance or supplementation makes more sophisticated approaches to treatment, such as gene therapy, unnecessary. The major research interest still lies in determining how epidemiological observations are linked to genetics; i.e., what are the genetic alterations that lead to lactase persistence in the milk-drinking populations of the world, and are there differences among different ethnic or racial groups. These are questions of great interest to population and evolutionary geneticists. GLUCOSE–GALACTOSE MALABSORPTION Clinical Features Glucose–galactose malabsorption is a rare congenital disorder in which enterocytic transport of glucose and galactose is absent, because of mutation of the gene encoding the

588

SECTION VI / GASTROENTEROLOGY

Table 56-2 Medical Therapy for Inflammatory Bowel Disease 5-Aminosalicylic acid compounds Sulfasalazine (Azulfidine) Mesalamine Oral (Pentasa, Asacol, Dipentum) Enema (Rowasa) Suppository (Rowasa) Glucocorticoids Oral Enema Suppository Intravenous Antibiotics Metronidazole Immunosuppressive agents Azathioprine 6-Mercaptopurine Cyclosporine Oral Enema Intravenous Methotrexate Intramuscular Oral Infliximals

Na+/glucose cotransporter sodium glucose transporter (SGLT)-1. This disease is characterized by the onset of profuse watery diarrhea in newborns that responds to the removal of dietary sources of glucose and galactose. Diarrhea usually develops within 4 d of birth. The diarrhea is so severe that dehydration is frequent. As in other diseases of carbohydrate absorption, the presence of unabsorbed sugars leads to an increase osmotic load in the small intestine with increased luminal fluid. Laboratory findings include an acidic stool pH because of colonic bacterial metabolism of unabsorbed sugar. In addition, sugar is found in the urine, consistent with the expression of SGLT-1 in the kidney. Oral monosaccharide tolerance tests reveal that absorption of glucose and galactose but not fructose is impaired. Genetics and Molecular Pathophysiology Glucose– galactose malabsorption is an autosomal-recessive disease. The precise incidence of this disorder has not been determined but it is clearly rare, as indicated by the few case reports in the literature. Frequently, there is a history of consanguineous marriage. Expression cloning of the cDNA that encodes SGLT-1, the Na+/glucose cotransporter provided a major breakthrough in understanding pathophysiology of this disorder. Specific mutations in SGLT-1 lead to loss of transporter function in humans, resulting in glucose and galactose malabsorption. Analysis of cDNA reverse transcribed from RNA from intestinal biopsies of two affected sisters showed a single missense mutation, with a single base change from G to A at position 92, changing an aspartate to an asparagine. Their parents were heterozygous for this mutation. The autosomal-recessive inheritance of this disorder was confirmed. The mutant SGLT1 mRNA was expressed in Xenopus laevis oocytes and compared with the normal SGLT1 mRNA for ability to transport a glucose analog, α-methyl-D-glucopyranoside. As anticipated, the mutant SGLT1 demonstrated no detectable transport activity.

Subsequently, 30 other patients have been screened for mutations in SGLT1. The mutations included missense, nonsense, and frameshift mutations, and some occurred in splice sites at intron–exon boundaries. Two areas were identified as “hot spots” for missense mutations, between residues 289–304 and 369–405. Analysis of the mutant proteins indicated that loss of transporter activity was owing either to the production of truncated or nonfunctional protein, to defective trafficking of the transporter, or to defects in the transport cycle. The SGLT1 gene has been localized to chromosome 22 on the distal q arm. Management The treatment of glucose–galactose malabsorption is the immediate removal of lactose, glucose, and galactose containing substances from the diet. Fructose is well tolerated, and a fructose-based formula is substituted with rapid resolution of symptoms. There are reports of increasing tolerance to glucose and galactose with age. Rechallenge with small amounts of these sugars may be possible. Future Directions The cloning of the SGLT-1 cDNA and the identification of naturally occurring mutations has already led to notable advances in the knowledge of the basic mechanisms of sugar transport. The regulation of expression of this and other cotransporters, their mode of insertion in the membrane, and structure–function correlation are all possible to address in the research laboratory. With the identification of the SGLT1 gene, gene therapy may eventually be considered. However, in the near future, this is unlikely because of the relative ease of treatment by elimination diets as well as the rarity of this disorder. ABETALIPOPROTEINEMIA Clinical Features Abetalipoproteinemia is a rare disorder, characterized by the absence of apolipoprotein (apo) B containing lipoproteins in plasma. Patients present in infancy with diarrhea resulting from severe fat malabsorption, poor weight gain, and acanthocytosis. Atypical retinitis pigmentosa and spinocerebellar degeneration become apparent at older ages. Ocular symptoms include decreased night and color vision, and neurological signs of ataxia, spastic gait and dysmetria, loss of deep tendon reflexes, and decreased vibratory and proprioceptive sense appear in the second decade. Spontaneous hemorrhage characteristic of a bleeding diathesis may occur because of vitamin K deficiency. Diagnosis The hallmark of this disorder is an abnormal plasma lipid profile in association with signs of spinocerebellar degeneration and a pigmented retinopathy. Total cholesterol and triglyceride plasma levels are extremely low and there is an absence of apo B containing plasma lipoproteins, including chylomicrons, very low-density lipoproteins, and low-density lipoproteins. Most patients are anemic because of enhanced hemolysis of abnormal red blood cells (acanthocytes). Fat malabsorption leads to deficiency of fat soluble vitamins. In particular, vitamins E, K, and A are affected as they are transported in chylomicrons. Vitamin E deficiency is most severe because its absorption from the intestine, transport to the liver, and circulation in the plasma are dependent on apo B containing lipoproteins. Vitamin E deficiency is clinically important because it is thought to be responsible for the development of neurological and retinal sequelae. Intestinal biopsy reveals lipid laden enterocytes but normal villi, and is useful for distinguishing this disorder from other intestinal illnesses associated with fat malabsorption. Genetics and Molecular Pathophysiology Abetalipoproteinemia is a rare autosomal-recessive disorder. Consanguinity is common in these families. An absence of plasma lipoproteins

CHAPTER 56 / SMALL AND LARGE BOWEL DYSFUNCTION

containing apo B led to the hypothesis that this disorder is because of a defect in the apo B gene. However, linkage between apo B and abetalipoproteinemia in families has not been established by restriction fragment length polymorphism analysis and defects in the apo B gene have not been identified. Because assembly or secretion of apo B containing lipoproteins were most likely defective in this disorder, the microsomal triglyceride transport protein (MTP) was examined in affected individuals. This gene, located on chromosome 4q22-24, encodes a protein that catalyzes the transport of triglyceride, cholesteryl ester, and phospholipid between membranes. It is a heterodimer made up of a protein disulphide isomerase and a unique 97-kd subunit that appears to be responsible for the lipid transport activity. This larger subunit shares homology with lipovitellin 1, a member of the lipid-binding lipovitellin complex that serves as a source for lipid in the developing embryo. MTP activity and particularly the 97-kd subunit, normally present in microsomes from liver and intestine, are absent in intestinal biopsies from patients. Sequence analysis of cDNA and genomic DNA from patients with abetalipoproteinemia revealed homozygous frameshift and nonsense mutations in the gene encoding the 88-kd subunit, resulting in truncated proteins. Two missense mutations have also been identified. Analysis of additional patients indicated frameshift and splice site mutations, most of which lead to the formation of a truncated protein. Transfection of Cos-1 cells with mutant MTP cDNAs lacking part of the carboxy terminus shows decreased lipid transfer activity, supporting the hypothesis that alterations of this protein are the primary cause of abetalipoproteinemia. Management A low-fat diet and supplementation with high doses of vitamin E as well as vitamins A and K are required. In particular, it is critical to avoid vitamin E deficiency as this appears to be the primary cause of the severe and irreversible neurological and ocular changes characteristic of this disease. Future Directions The cloning of the cDNA encoding the MTP raises the possibility of gene therapy in the future. Although many of the devastating neurological and ocular sequelae may be avoided by high doses of vitamin E, there may be a role for gene replacement in refractory patients. Also, rapid, definitive diagnosis is crucial, so that prenatal screening of siblings may be useful. Finally, because patients with MTP deficiency have very low cholesterol and triglyceride levels, identification of selective inhibitors of its activity may prove a valuable addition to the treatment of hypercholesterolemia.

SELECTED REFERENCES Ahmad T, Tamboli C, Jewell D, Colombel JF. Clinical relevance of advances in genetics and pharmacogenetics of IBD. Gastroenterology 2004;126(6):1533–1549. Anderson RP, Dgano P, Godkin AJ, Jewell DP, Hill AVS. In vivo antigen challenge in celiac disease identifies a single transglutaminasemodified peptide as the dominant A-gliadin T-cell epitope. Nat Med 2000;6:337–342. Berriot-Varoqueaux N, Aggerbeck LP, Samson-Bouma ME, Wetterau JR. The role of the microsomal triglyceride transport protein in abetalipoproteinemia. Annu Rev Nutr 2000;20:663–697. Boll W, Wagner P, Mantei N. Structure of the chromosomal gene and cDNAs coding for lactase-phlorizin hydrolase in humans with

589

adult-type hypolactasia or persistence of lactase. Am J Hum Genet 1991;48:889–902. Cho JH, Nicolae Dl, Gold LH. Identification of novel susceptibility loci for inflammatory bowel disease on chromosomes 1p, 3q, and 4q: Evidence for epistasis between 1p and IBD1. Proc Natl Acad Sci USA 1998;95:7502–7507. Enattah NS, Sahi T, Savilahti E, Terwilliger JD, Peltonen L, Jarvela I. Identification of a variant associated with adult-type hypolactasia. Nat Genet 2002;30:233–237. Evans L, Grasset E, Heyman M, Dumontier AM, Beau JP, Desjexu JF. Congenital selective malabsorption of glucose and galactose. J Pediatr Gastroenterol Nutr 1985;4:878–886. Grand RJ, Montgomery RK, Chitkara DK, Hirschhorn JN. Changing genes; losing lactase. Gut 2003;52:621, 622. Green PHR, Jabri B. Coeliac disease. Lancet 2003;362:383–391. Hediger MA, Coady MJ, Ikeda TS, Wright EM. Expression cloning and cDNA sequencing of the Na/glucose cotransporter. Nature 1987;330: 379–381. Hugot JP. Genetic origin of IBD. Inflamm Bowel Dis 2004,10:S11–S15. Hugot JP, Laurent-Puig P, Gower-Rousseau C, et al. Mapping of a susceptibility locus for Crohn’s disease on chromosome 16. Nature 1996; 379(6568):821–823. Kagnoff MF, Harwood JI, Bugawan TL, Erlich HA. Structural analysis of the HLA-DR, -DQ, and -DP alleles on the celiac disease-associated DR3 (Drw17) haplotype. Proc Natl Acad Sci USA 1989;86:6274–6278. Loftus EV. Clincial epidemiology of inflammatory bowel disease: incidence, prevalence and environmental influences. Gastroenterology 2004;126: 1504–1517. Martin MG, Turk E, Lostao P, Kerner C, Wright EM. Defects in Na/glucose cotransporter (SGLT1) trafficking and function cause glucosegalactose malabsorption. Nat Genet 1996;12:216–220. Mowat AM. Coeliac disease—a meeting point for genetics, immunology, and protein chemistry. Lancet 2003;361:1290–1292. Pizarro TT, Arseneau KO, Bamias G, Cominelli F. Mouse models for the study of Crohn’s disease. Trends Mol Med 2003;9:218–222. Rader DJ, Brewer HB. Abetalipoproteinemia: new insights into lipoprotein assembly and vitamin E metabolism from a rare genetic disease. JAMA 1993;270:865–869. Sadlack B, Merz H, Schorle H, Schimpl A, Feller AC, Horak I. Ulcerative colitis-like disease in mice with a disrupted interleukin-2 gene. Cell 1993;75:253–261. Shamir R. Advances in celiac disease. Gastroenterol Clin North Am 2003;32:931–947. Shan L, Molberg O, Parrot I, et al. Structural basis for gluten intolerance in celiac sprue. Science 2002;297:2275–2279. Sharp D, Blinderman L, Combs KA, et al. Cloning and gene defects in microsomal triglyceride transfer protein associated with abetalipoproteinemia. Nature 1993;365:65–69. Sollid LM. Coeliac disease: dissecting a complex inflammatory disorder. Nat Rev Immunol 2002;2:647–655. Strober W, Ehrhardt RO. Chronic intestinal inflammation: an unexpected outcome in cytokine or T cell receptor mutant mice. Cell 1993;75:203–205. Swallow DM. Genetics of lactase persistence and lactose intolerance. Annu Rev Genet 2003;37:197–219. Troelsen JT, Olsen J, Noren O, Sjostrom H. A novel intestinal trans-factor (NF-LPH-1) interacts with the lactase-phlorizin hydrolase promoter and co-varies with the enzymatic activity. J Biol Chem 1992;267: 20,407–20,411. Turk E, Zabel B, Mundlos S, Dyer J, Wright EM. Glucose/galactose malabsorption caused by a defect in the Na+/glucose cotransporter. Nature 1991;350:354–356. Wetterau JR, Aggerbeck LP, Bouma M, et al. Absence of microsomal triglyceride transfer protein in individuals with abetalipoproteinemia. Science 1992;258:999–1001.

57 The Molecular Mechanisms

of Helicobacter pylori-Associated Gastroduodenal Disease PETER B. ERNST

SUMMARY In 1984, Barry Marshall and Robin Warren proposed a role for Helicobacter pylori in gastroduodenal disease leading to an avalanche of research intended to prove or disprove their theory. The result has been a series of advances that have updated our understanding of these diseases and completely modernized the clinical approach to their management. In recognition of this impact, Marshall and Warren received the Nobel Prize for Physiology or Medicine in 2005. This chapter summarizes some important findings that have advanced our knowledge in this field with emphasis on the host response and the immunopathogenesis of gastroduodenal disease. Key Words: Adaptive immunity; B cells; bacterial colonization; cytokines; epithelium; gastric; immunopathogenesis; innate immunity; persistence; T cells.

INTRODUCTION In 1984, Marshall and Warren proposed a role for Helicobacter pylori in gastroduodenal disease. The result has been a series of advances that have updated the understanding of these diseases and completely changed the clinical approach to their management. Many aspects of the pathogenesis of these diseases have been dissected to the molecular level with key pathogenic mechanisms being validated by the association of some genes with the development of gastric cancer. There has been particular emphasis on understanding the molecular structures associated with the organism and their role in modifying the host responses. The gastric immune and inflammatory response have emerged as key elements in the pathogenesis of gastritis and epithelial cell damage that are accepted as being an integral component in the disease process. This chapter summarizes some of the important findings that have advanced the knowledge in this field, with special emphasis on the host response. Gastric secretions have evolved to facilitate the acid and peptic degradation of luminal contents. This environment precludes the successful colonization by most micro-organisms but H. pylori From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

has evolved to survive in this niche. Although H. pylori colonizes the majority of humanity, the lack of colonization by a more diverse flora may have limited the selective pressures shaping the evolution of gastric immune and inflammatory responses. Therefore, most research has focused on the interaction between a single species of bacteria and the host response. This approach has forced investigators to compare their findings in the stomach to observations in other tissues infected with a range of organisms. Epidemiological studies have provided convincing evidence that H. pylori is associated with, and indeed responsible for, several gastroduodenal diseases ranging from a generally benign, subclinical gastritis to peptic ulcers and several types of gastric cancer. Advances from experiments manipulating the organism and examining its interaction in cell lines or animal models have shown that infection activates multiple pathways leading to changes in epithelial cell morphology, tight junctions, cytokines, and epithelial cell turnover. Clearly, the molecular pathogenesis involves many factors and complementary interactions among the bacteria and various elements of the host response.

EPIDEMIOLOGY The majority of new H. pylori infections occur in children; however, in the absence of specific clinical signs associated with infection, the details concerning the mode of transmission are less clear. Vomitus, saliva, and feces are the presumed sources of direct transmission particularly in crowded housing conditions. Generally, infection correlates inversely to socioeconomic conditions as lifetime infection rates in countries with a strong economy have dropped toward 10%, although rates in countries with emerging economies approach 80–90%. Once established, the infection persists for life unless antibiotic therapy is administered. Most infected individuals experience subclinical gastritis although gastroduodenal ulceration may occur in 10–15% of the infected population. Gastric cancer is less frequent with approx 1% of infected individuals developing adenocarcinoma whereas even less experience gastric maltoma. In fact, H. pylori has been recognized as a type-I carcinogen by the World Health Organization. One study has even suggested that H. pylori is a necessary condition for noncardia gastric cancer. The geographic distribution of these

590

CHAPTER 57 / THE MOLECULAR MECHANISMS OF H. PYLORI-ASSOCIATED GASTRODUODENAL DISEASE

591

Figure 57-1 Proposed pathogenesis for gastric cancer associated with Helicobacter pylori infection. Originally, a multi-step pathogenesis was proposed for gastric cancer that began with gastritis and progressed to atrophy, metaplasia, dysplasia, and sometimes cancer. With the recognition of H. pylori as a major cause of gastric cancer, the etiology has been identified. Because not everyone infected with H. pylori develops gastric cancer, several modifiers have also been identified. The perception is that gastric cancer arises from multiple “hits” that include oxidative stress and environmental toxins that increase mutation rates. Diet, bacterial factors, and genes regulating the host response likely affect the degree of oxidative stress and DNA damage. Once key genes are mutated, the enhanced epithelial growth associated with infection drives the tumor proliferation. OipA, outer inflammatory protein A; cag PAI, cag pathogenicity island; VacA, a vacuolating cytotoxin.

diseases suggest that gastric cancer is more common in Asian and Latin American countries whereas duodenal ulcers occur more frequently in Western Europe and the United States or Canada. Paradoxically, chronic infection in Africa is not associated with cancer and remains the “African Enigma.” These interesting patterns imply that regional differences in bacterial strain or environmental factors have a significant impact on the outcome of infection (Fig. 57-1). Studies examining the impact of strains on inflammation and disease have yielded several interesting associations (discussed later) but the predictive value of specific bacterial factors remains limited as most “virulent” strains can also be found in many asymptomatic subjects. The variation in the host response also affects outcome as genetic studies in humans suggest that mutations leading to an increase in gastric immune/inflammatory responses are associated with gastric cancer. Thus, the collective evidence points to a multifactorial process in the pathogenesis of the diseases associated with this infection.

PATHOGENESIS BACTERIAL PATHOGENESIS Colonization H. pylori is a Gram-negative, flagellated organism that produces a number of enzymes including catalase and urease, which help neutralize host responses and favor colonization. The ability of the organism to adapt to the harsh pH in the gastric lumen has been illuminated by investigators who have examined the changes in H. pylori gene expression following exposure to low pH. A major change was observed in the expression of genes that control motility. Investigators were able to observe that the increase in the expression of genes encoding proteins involved in the motility apparatus led to enhanced movement. In addition, there was an increased expression of the genes encoding urease and the other proteins associated with the optimal function of this enzyme. The urease produced by H. pylori displays an adaptation to the gastric milieu in that it has two different pH values at which its function is optimal—one being the predictable pH 7.2 and the other pH 3.0. Because this enzyme functions well at a low pH, it was originally proposed that urease reached the extracellular environment in which it could metabolize urea into NH3 that in turn buffers protons in the gastric lumen. An elegant study suggested that the enzyme functions mostly within the bacteria to generate ammonia ions that will buffer the H+ ions as they reach the cytoplasm of the organism.

Thus, on infection, H. pylori adapts to the gastric pH to produce the molecular machinery required to neutralize the acid as well as to escape quickly to the relatively safe confines beneath the mucus blanket. In this niche, H. pylori enjoys some of the same cytoprotective mechanisms that the host epithelium uses for the maintenance of its structural integrity and function. Moreover, the organism is well equipped to continue to adapt to changes in its niche. Clearly, understanding the detailed molecular machinery that controls the expression and function of bacteria 1 genes such as those encoding the flagellar or urease proteins will have the potential to provide novel therapeutic targets that could impair the survival of the organism in the acidic gastric environment. Attachment H. pylori shows a strict tropism for gastric epithelium that could reflect positive selection by gastric epithelial cells, for example, the tissue-specific expression of receptors. Alternatively, intestinal epithelial cells could negatively select for colonization because of the production of antimicrobial factors that discourage H. pylori growth. The latter possibility would be of interest as the controlled expression of any putative antimicrobial factor by epithelial cells could provide a novel therapeutic approach for vaccine-based therapies. In fact, several naturally produced proteins that exhibit an antimicrobial effect have been examined in the infected gastric mucosa H. pylori also decreases the expression of the antibacterial molecule “secretory leukocyte protease inhibitor.” Thus, it is not clear if the effects of H. pylori on the expression of these molecules in the gastric niche limit the concentration of H. pylori to tolerable levels or whether they confer a selective advantage by inhibiting the growth of other species. The CAG Pathogenicity Island Once H. pylori has established itself in its preferred niche, the organism must attach to host cells and possibly damage them enough to ensure that sufficient nutrients can be obtained in the subsequent inflammatory exudate or transudate (Fig. 57-2). Consequently, studies have been directed toward understanding the receptors for this organism and their effects on epithelial cell signaling. The most studied interaction involves a stretch of DNA referred to as the cag pathogenicity island (cag PAI). Genes within the cag PAI encode proteins that provide a type-IV secretion apparatus that allows bacterial macromolecules to translocate into the host cell. The intact cag PAI of H. pylori plays a significant role in virulence in humans as evidenced by researchers who described the role of cagE, a gene that appears to be essential for a functioning bacterial secretion system. This study showed an increase in

592

SECTION VI / GASTROENTEROLOGY

Figure 57-2 Potential interactions between H. pylori and gastric epithelial cells. This figure summarizes the best-known interactions between bacterial products and host cell receptors that have been described in the text. Any of these potential interactions could enhance bacterial binding although secreted bacterial products may also engage the receptors. These interactions allow the epithelial cell to transduce a signal to the host indicating that a luminal infection is present that may represent some danger. One aspect that is evident is that few bacterial/epithelial cell interactions have been validated in vivo or linked to specific signaling pathways that result in a known epithelial cell response. It is also evident that no single pathway is responsible for all changes in epithelial cell responses. Furthermore, it is clear that the bacterial effects on epithelial cell signaling are insufficient to explain the magnitude of the gastritis and the local or systemic consequence that are attributed to H. pylori. Hence, these interactions complement the effects of cytokines, neuroendocrine influences, and other signals emanating from the lamina propria. cag PAI, cag pathogenicity island; IL, interleukin; OipA, outer inflammatory protein A; PG, peptidoglycan Vac A, a vacuolating cytotoxin.

duodenal ulceration in children carrying strains expressing cagE in association with higher levels of gastric IL-8. This report validates a study in vitro in which mutations of the gene encoding picB (now referred to as cagE) abrogated the ability of H. pylori to induce IL-8. Furthermore, infecting gerbils with a mutated strain that lacked CagE production reduced the severity of gastritis substantially. Gastric ulceration developed in virtually all of the gerbils infected with the wild-type strain but in none of the animals infected with the cagE mutant. Finally, intestinal metaplasia and gastric cancer developed in some gerbils with wild-type infection but not in animals infected with the cagE-negative mutant. These findings were confirmed and extended by showing that a cag-positive H. pylori strain induced severe gastritis, apoptosis, atrophy, and gastric ulceration in gerbils. The investigators then used the H. pylori whole genome microarray to identify differences in gene content between specific H. pylori strains. Naturally occurring or experimentally induced mutations of the H. pylori cag region were associated with decreased gastric mucosal inflammation in vivo, and this was correlated to reduced potency in activating IL-8 gene expression or apoptosis in vitro. Attachment Via Lewis Antigens Le antigens expressed by host cells may serve as the major receptor for bacterial binding. A technically sophisticated experiment has shown that mice genetically engineered to express increased levels of Leb had more severe gastritis although there was no effect on bacterial load. Other experiments have demonstrated that specific bacterial gene products, most notably BabA, act as the bacterial ligand for the Leb receptor. Another adhesion molecule expressed by H. pylori, SabA, binds weakly to sialyl Lex (sLex) antigens. This interaction may be significant in the maintenance of persistent infection as the sLex antigens emerge on gastric epithelial cells in response to inflammation.

The reciprocal interaction between bacterial Le antigens and the host has also been examined. The expression of Lex structures by H. pylori lipopolysaccharide (LPS) promotes adhesion to the gastric epithelium. However, isogenic mutants of the bacteria that do not express the fucosyl transferases necessary for Lex and Ley expression infected mice and bound human gastric epithelial cells comparably. Although some of the approaches supporting a role for the Le antigens on host cells are very elegant, the hypothesis has been challenged. In addition, there are data showing that the binding of H. pylori to epithelial cells freshly isolated from human gastric biopsy specimens is unaffected by the Le antigen expression whereas less than half of the strains tested in infected subjects express the SabA adhesion that binds sLex. Another approach to establish the role of the interaction between Leb and BabA has been epidemiological studies. Some investigators have data suggesting that the babA2 genotype may be found preferentially in strains of H. pylori that are more likely to be associated with inflammation, duodenal ulcer, and gastric cancer. However, the importance of Leb as a receptor has been discounted by several reports that have not observed an increase in infection in subjects with the Leb phenotype. Moreover, individuals who do not express Leb can clearly be infected with H. pylori and other studies report that the majority of strains infecting the subjects studied expressed the babA2 gene and did not induce ulcers or cancer. It is possible that the product of the babA2 gene has other functions besides binding Leb that would contribute to its association with disease. The lack of direct evidence that molecules bearing the glycoconjugates mediate signaling events or facilitate other signaling mechanisms weakens any argument for a significant role in the pathogenesis of the infection. Hence, Le antigens can only be considered as a subset of receptors within a broad array of molecules that can bind H. pylori.

CHAPTER 57 / THE MOLECULAR MECHANISMS OF H. PYLORI-ASSOCIATED GASTRODUODENAL DISEASE

Attachment Via Molecules Associated With Immune/ Inflammatory Responses H. pylori can also bind to the complex of invariant chain and class II major histocompatibility complex (MHC) molecules expressed on the surface of gastric epithelial cells. The class II MHC molecules were the first epithelial cell receptor shown to directly affect signaling in host cells as urease binds to epithelial cells via class II MHC and is sufficient to induce apoptosis. Using transfected cells lines, investigators showed that class II MHC molecules were sufficient to allow host cells to bind H. pylori. Furthermore, these molecules are expressed in situ by gastric epithelial cells at increased levels during infection and their expression is increased in response to interferon (IFN)-γ. Because H. pylori is also well known for inducing cytokine production in the epithelium, these studies were extended and showed that both cytokine production and apoptosis are mediated through class II molecules, suggesting the class II MHC complex serves an important role in binding and signaling. A class of molecules known as pathogen-associated molecular receptors (PAMPs) have also been examined for a role in binding H. pylori. These receptors include the Toll-like receptors (TLRs). Eleven such TLRs have been identified and it appears that each has a relative specificity for various bacterial molecules. For example, TLR4 is able to recognize the LPS of many bacteria. Cytokines and H. pylori particles increase the expression of TLR4. H. pylori LPS stimulates monocytes and gastric epithelial cell responses in guinea pigs, presumably via TLR4. TLR5 is another PAMP that binds bacterial flagellins and similarly induces a signaling response that would trigger acute inflammation. H. pylori produces flagellin that binds this ligand and activates a response in vitro. Other PAMPs, including TLR2 are activated by highly purified H. pylori LPS, and together these receptors may bind bacterial products in order to enhance bacterial binding as well as cell signaling. It should be remembered that gastric epithelial cell lines that do not express TLRs still respond well to H. pylori as evidenced by IL-8 production. Thus, it remains to be shown if these receptors and their required accessory molecules are expressed and functional in situ. Other Receptors Several other binding mechanisms have been explored. For example, many strains of H. pylori express a vacuolating cytotoxin, VacA, which can attach to epithelial cells via an interaction with a protein tyrosine phosphatase. Volumes have been written on the structure and function of VacA and its association with disease. For example, specific VacA alleles (S1, m1) are associated with disease and the induction of epithelial cell apoptosis so the importance of the interaction between VacA and its receptor appear to have significant importance in the pathogenesis of gastroduodenal disease. A gastric trefoil protein known as trefoil factor 1 (TFF1) serves as a receptor for H. pylori. This molecule is expressed predominantly in the gastric mucosa and may account for some of the gastric tropism. Interestingly, when the expression of TFF1 is decreased in genetically engineered mice, the animals develop a spontaneous, antral adenoma suggesting that TFF1 provides an element of control over gastric epithelial cell growth. The attachment of H. pylori to this molecule could impair its function (mimicking the decreased expression in the animal models) thus favoring the proliferation of epithelial cells and the subsequent development of cancer in some infected subjects. Collectively, these studies suggest that H. pylori uses many distinct receptors and all of these receptors likely provide complementary attachment mechanisms that facilitate colonization and

593

enhance other interactions, such as the function of the type-IV secretion system. As a consequence of these interactions, a series of signaling processes initiate the host response to this infection. The ensuing inflammation may favor bacterial survival by increasing bacterial receptors, providing serum transudate as a source of nutrients and inducing antibacterial proteins that impair competition from other organisms that may try to colonize the more accommodating niche provided by a chronic infection with H. pylori. The relative role of the various receptors will emerge only when there is a more comprehensive understanding of the distinct binding interactions and their respective impact on epithelial cell signaling. THE IMMUNOPATHOGENESIS OF H. PYLORI William Osler suggested that “patients seem to die not of their infection but of the body’s response to it.” With H. pylori, sterilizing immunity does not occur following natural infection despite the presence of a robust, inflammatory response. The importance of the host response in disease is supported by the observations that heterogeneity in the regions of the genome that control the magnitude of inflammation is associated with gastric cancer linked with H. pylori infection. These associations began with the report that polymorphisms in the regions controlling IL-1β were associated with an increased incidence of hypochlorhydria—a presumptive precursor associated with gastric cancer. IL-1 has been previously reported as a potent inhibitor of gastric acid secretion, so an increase in IL-1 not only drives inflammation but leads to a physiological state that precedes gastric cancer development. Importantly, this study has been replicated in a Japanese population in which IL-1β polymorphisms were not only associated with gastric cancer but also a decrease in duodenal ulcer recurrence whereas others reported increased gastric IL-1, more severe gastritis, gastric atrophy, and colonization with H. pylori strains that are associated with gastric cancer. If the hypothesis is correct, then other proinflammatory gene polymorphisms could be predicted to be associated with increased gastritis as well as cancer and indeed, this appears to be the case. Other genes that regulate the magnitude of the inflammatory response, including IL-10, tumor necrosis factor (TNF)-α and IL-8 have also been associated with the sequence of events leading to cancer. Thus, the host response and tissue remodeling following infection is directly responsible for a significant component of the disease associated with H. pylori. The most compelling of these studies provides evidence that the gene is not only mutated, but the mutation changes the expression of the cytokine and the tissue levels in affected subjects reflect the predicted change. More of these studies are being performed and additional insight into the role of the immune response as well as specific immune molecules should emerge. How Epithelial Cells Recognize and Respond to “Danger” Because the overwhelming bulk of the antigenic mass associated with H. pylori infection resides in the lumen, the gastric epithelial cells have been examined for the mechanisms by which they “register” that an infection has occurred. The epithelial response to infection is complicated as it is driven by several molecular component parts of the bacterium, the signaling linked to specific receptors that recognize these components as well as the local concentrations of hormones, neurotransmitters, immune/inflammatory cytokines, and mediators or stromal responses. The epithelial cell responses include changes in epithelial cell morphology (the hummingbird phenotype), disruption of the tight-junctional complexes, the production of cytokines, increased epithelial cell proliferation, increased rates of epithelial cell death via apoptosis, and the induction of numerous genes that reflect the stress

594

SECTION VI / GASTROENTEROLOGY

Figure 57-3 An overview of signaling responses of gastric epithelial cells during H. pylori infection. Several signaling pathways have been studied using intact organisms or specific bacterial products. Substantial crosstalk exists between the respective pathways triggered by bacteria or host response molecules. Some of the major events that have been characterized include the pathways involving the translocated CagA in the development of the hummingbird phenotype; the differential role of the outer inflammatory protein A (OipA) and the cag pathogenicity island in signal transduction; the effects of intact H. pylori on extracellular ERK; P38 as well as Jun n-terminal kinase mitogen-activated protein kinases; and the role of oxidative stress in modifying transcription factors by reduction via redox factor-1 (ref-1), which reduces C-Jun to C-Jun-R. Together, these pathways provide specific targets for novel therapies that interfere with the control of phosphorylation, the transactivation by transcription factors or the control of oxidative stress. IFN, interferon; TNF, tumor necrosis factor.

encountered in response to infection. It should be appreciated that, similar to other multi-factorial diseases, it is a significant scientific challenge to dissect the various pathogenic processes that culminate in peptic ulcers or gastric cancer and it is still premature to rank some pathways over others in terms of their relative importance. Changes in Gene Expression The best overview of the epithelial cell response can be ascertained from studies examining the broad changes in the expression of hundreds of genes as assayed with high-throughput gene expression systems. These techniques provide a powerful approach to evaluate changes in the expression of a large number of genes in response to specific conditions, such as bacterial strain. When you combine the effects of bacteria with proinflammatory mediators found in gastric tissue during infection, well in excess of 1000 genes are affected. Detailed analysis of the effects of infection, different cell lines and various inflammatory mediators is needed over a wide range of time points that would be relevant to the condition epithelial cells face in vivo. The genes of immediate interest are those that regulate the immune/inflammatory response, epithelial cell turnover including apoptosis and proliferation, as well as those affecting physiological properties in the stomach. Eventually, the genes that are less obviously connected to the understanding of epithelial cell homeostasis may emerge as being the most important. Although a detailed study of these genes is interesting on its own, the mechanisms regulating these changes may be more relevant for identifying points in the pathogenesis that can be targeted for therapeutic intervention (Fig. 57-3). Effects of Infection on Transcription Factor Activation The expression of genes in epithelial cells stimulated with H. pylori is regulated by transcription factors that in turn are controlled by a series of signaling mechanisms. Although many transcription factors are activated in response to infection, nuclear factor (NF)-κB, and

activator protein 1 (AP-1) have been studied in the most detail. These transcription factors regulate the expression of a wide variety of proinflammatory cytokines and cellular adhesion molecules that are induced in response to the infection or the local cytokine milieu. H pylori activates NF-κB in gastric epithelial cells both in vitro and in vivo. In patients with H. pylori gastritis, gastric epithelial cell NF-κB activity was markedly enhanced and this correlated with the intensity of neutrophil infiltration and IL-8 protein levels. This pathway is of particular interest given the association between polymorphisms in the il-8 gene that lead to increased mucosal IL-8 expression, inflammation, and other premalignant changes associated with gastric cancer. In addition, several strains of H. pylori are associated with both disease and increased IL-8 production, so understanding the transcription factors controlling IL-8 provides important insight into the pathogenesis of disease. One of the novel concepts just emerging in the context of H. pylori infection is the role of oxidative stress in the control of gene expression. Reactive oxygen and nitrogen species are generated in response to infection (described later). Although their role as antimicrobial mediators is understood, their function in the control of gene transcription deserves attention. One molecule that plays a key role in regulating redox-sensitive signaling is apurinic/apyrimidinic endonuclease-1 or redox factor-1 as it is also called. This molecule is expressed in gastric epithelial cells during infection with H. pylori, and studies show that it is important for the full activation of AP-1 and NF-κB as well as IL-8 production in response to H. pylori. This pathway is important in the optimal activation of many other transcription factors including p53 and activating transcription factor so it is likely that the significance of redox-sensitive signaling in the pathogenesis of diseases associated with H. pylori will become more evident.

CHAPTER 57 / THE MOLECULAR MECHANISMS OF H. PYLORI-ASSOCIATED GASTRODUODENAL DISEASE

Signaling Pathways Activated in Response to H. Pylori H. pylori infection is linked to the activation of NF-κB and AP-1 in gastric epithelial cell lines by several signaling mechanisms. These include p21-activated kinase 1, which associates with and activates Nck-interacting kinase thereby activating the kinases inhibitor of kappa kinase α and inhibitor of kappa kinase β leading to phosphorylation and degradation of the inhibitory protein IκBα and the release of activated NF-κB. Mitogen-activated protein (MAP) kinase cascades regulate a range of cell functions including proliferation, inflammatory responses and cell survival. Cag-positive H. pylori activates the extracellular signal-regulated kinase (ERK), Jun n-terminal kinase and p38 MAP kinase pathways, and ERK and p38 regulate IL-8 production in gastric epithelial cells. The VacA toxin is sufficient to induce the phosphorylation of p38 and ERK1/2 although this signaling pathway seems to be independent of the vacuolization induced by this toxin. Kinases such as ERK and p38 can participate in the activation of the transcription factor AP-1 and the regulation of several genes that encode proteins that drive inflammation. FR167653, a p38α MAP kinase inhibitor, reduced both neutrophil infiltration and gastric mucosal injury in H. pylori-infected Mongolian gerbils. Thus, not only is p38 MAP kinase important in regulating the host inflammatory response to H. pylori infection, but this and other kinases can become therapeutic targets that attenuate inflammation and the ensuing tissue damage associated with infection. Although infection with intact bacteria stimulates several signaling pathways, more attention is given to the role of specific virulence factors such as those associated with the cag PAI because H. pylori that carry the cag PAI are associated with increased IL-8 expression in gastric mucosal biopsy specimens. The only identified effector protein able to translocate via the cag PAI is the CagA protein. Following its translocation into the cytoplasm, CagA is tyrosine phosphorylated by host Src kinases and interacts with SHP-2—a phosphatase involved in cell signaling and regulating epithelial cell morphology—as well as Grb2 and C-terminal SRC kinase (Csk). Interaction with Grb2 results in the activation of the Ras/ERK pathway that contributes to an increase in proliferation whereas the activation of C-terminal SRC kinase may provide a negative feedback on CagA phosphorylation. Heterogeneity in the phosphorylation sites within CagA may vary by geographic distribution and results in an altered ability to interact with SHP-2 and modify intracellular signaling. It is intriguing to speculate that heterogeneity in the CagA protein may account for some of the geographic differences in disease. Researchers would suggest that interference with SHP-2 signaling favors gastric hyperplasia. Together, these studies point to fascinating molecular interactions to be investigated in the context of gastric cancer. H. pylori can stimulate IL-8 directly via other bacterial factors. One such factor is designated outer inflammatory protein A (oipA), which induces epithelial cell IL-8 production, and does not appear to depend on the H. pylori cag PAI. Strains that express oipA were related to bacterial density, mucosal IL-8 levels, neutrophil infiltration, and clinical presentation. However, little activation of epithelial cell signaling via NF-κB occurs in the absence of the cag PAI. Subsequent studies have shown that specific bacterial products led to the activation of different transcription factors that collaborate to enhance IL-8 production. For example, the cag PAI and oipA are both required for the full activation of IL-8 but the former achieves this through both NF-κB and AP-1, whereas the latter was required to recruit activation by a STAT1-IRF1-ISRE response element.

595

Although the cag PAI facilitates the translocation of CagA, this bacterial protein per se has no effect on IL-8 and it is likely that other bacterial proteins translocate and contribute to the cag PAIdependent signaling leading to IL-8 production. Together, these experiments illustrate the level of complexity that is emerging as individual bacterial factors are being aligned with overlapping or distinct signaling pathways that will define the epithelial cell phenotype during infection. Beyond the Epithelium—Host Responses in the Lamina Propria Although several epithelial receptors for H. pylori have been described, they do not appear to fully account for the magnitude of the inflammatory response to an organism that resides predominantly in the lumen. Some bacteria may infect epithelial cells whereas significant amounts of bacterial material may “leak” around epithelial cells and reach the lamina propria in which it can activate underlying phagocytes including neutrophils and macrophages. One candidate bacterial factor is the H. pylori neutrophil-activating protein (HP-NAP). This is a 150-kDa decamer protein that stimulates chemotaxis of monocytes and neutrophils. It also promotes neutrophil adhesion to endothelial cells and NADPH oxidase complex assembly at the plasma membrane, leading to subsequent production of reactive oxygen intermediates. In the context of the inflammatory environment during H. pylori gastritis, an additional observation made by investigators was that TNF-α and IFN-γ primed neutrophils and potentiated the effect of HP-NAP. With the recruitment and activation of macrophages and neutrophils, other inflammatory mediators are released. As mentioned, reactive oxygen intermediates, NO, and some of its metabolites are important in the host defense against microbes. Increased expression of iNOS, the inducible isoform of NO synthase, is observed in the gastric mucosa during infection with H. pylori. NO and O2–, which may be produced by infiltrating neutrophils, react to form peroxynitrite, ONOO–, a potent oxidant and reducing agent. Although these products have potent antimicrobial effects, uncontrolled or inappropriate production, however, could play a role in the gastric mucosal damage observed during H. pylori infection. For instance, the catabolism of urea by urease provides CO2 that rapidly neutralizes the bactericidal activity of the peroxynitrate and forms ONO–OCO2. Thus, urease may sustain bacterial colonization, however, this reaction enhances the nitration potential of ONOO– and may favor mutagenesis. Cytokines from the epithelial cells complement those released in the lamina propria. For example, neutrophils are not only activated by IL-8 but also by other chemokines such as ENA-78 and Gro-α that can be produced by the epithelium, the adjacent myofibroblasts, or the macrophages within the lamina propria. Other cytokines are induced in macrophages, for example, urease induces TNF-α and IL-6 whereas heat shock protein 60 induces IL-6. Intact bacteria can induce the production of chemokines that recruit T cells as well as IL-12 and IL-18—two cytokines that favor the selection of T-helper (Th)1 cells. Thus, intact bacteria or bacterial factors trigger a broad cytokine response within the lamina propria as well. Gastric T-Cell Responses As adaptive responses develop, different Th cell subsets emerge as defined by characteristic patterns of cytokine secretion. At the risk of oversimplifying many complicated signaling events, Th1 cells promote cell-mediated immune responses mainly through the production of IFN-γ and TNF-α, whereas Th2 cells produce IL-4, IL-5, IL-10, and

596

SECTION VI / GASTROENTEROLOGY

Figure 57-4 The immunopathogenesis of gastritis and epithelial cell damage. This figure shows the various means by which the gastric immune and inflammatory responses mediate epithelial cell damage. First, through direct effects, Th1 cells can induce epithelial cell death via Fas/Fas ligand interactions. T-cell-derived cytokines not only enhance the induction of apoptosis but also upregulate the expression of receptors such as class II major histocompatibility complex increase bacterial binding and favor the induction of apoptosis by H. pylori. In addition, Th1 cells complement the effects of H. pylori by stimulating epithelial cells to produce cytokines that recruit and activate neutrophils. In turn, activated neutrophils impart an oxidative stress that damages epithelial cells as well as cellular DNA. Finally, gastric T cells can modulate B-cell responses possibly leading to the production autoantibodies including those of the IgG class that can activate complement (C’) and contribute to immunecomplex-mediated inflammation. From this model, it is clear how genetic polymorphisms that increase the expression of cytokines that drive inflammation (i.e., IL-1β, tumor necrosis factor-α, IL-8) or decrease the expression of those that prevent inflammation (i.e., IL-10) contribute to an increase in gastritis. As gastritis is believed to favor the development of adenocarcinoma, it makes sense that these polymorphisms are associated with increased rates of gastric cancer. IFN, interferon, IL, interleukin; TNF, tumor necrosis factor.

transforming growth factor-β. The Th2 cells can promote mucosal IgA or IgE responses as well as diminish the inflammation caused by Th1 cytokines. Th1 Cells Predominate in the Stomach Studies suggest that the gastric mucosa is preconditioned to favor Th1 cell development because IL-12- and IFN-γ-producing cells are present at baseline and increase during infection. One hypothetical explanation is that infection selectively blocks Th2 development. In fact, H. pylori can interfere with Stat6 activation by IL-4, which could impair Th2 development. H. pylori infection is also associated with increased gastric IL-18 transcripts in the antral mucosa. This effect is independent of the cag-status of the infecting strain. Thus, IL-12 and IL-18 may contribute to the predominant Th1 response observed during H. pylori infection. Other cytokines that enhance Th1 responses, such as IL-23 and IL-27, remain to be studied in human tissue. Interactions between bacteria, epithelial cells, and intercellular mediators play an important role in controlling the host response. This principle was demonstrated in a study characterizing the IL-17 response to infection. Biologically active IL-17, a cytokine produced by activated CD4+ T lymphocytes, is increased in the mucosa of H. pylori-infected patients. IL-17 is capable of activating NF-κB and MAP kinases that can initiate IL-8 expression by gastric epithelial cells thereby providing a positive feedback to epithelial cells and additional neutrophil recruitment. In addition, the activation of transcription factors by IL-17 may contribute to the increased levels of numerous other proinflammatory cytokines and enzymes observed during H. pylori infection, such as IL-1β, TNF-α, and cyclooxygenase-2. Similarly, Th1 cells produce IFN-γ and TNF-α that can increase the expression of many genes in the epithelium including

IL-8. Thus, the gastric T-cell response can complement the direct effects of the bacteria on the epithelium. Pathogenic Effects of Th1 Cells The inability of Th1 responses to clear infection suggests that T-cell activation may contribute to more severe gastroduodenal diseases (Fig. 57-4). This idea is supported by animal studies showing that protective immunity induced by vaccines for Helicobacter spp rely on Th cells other than Th1 cells possibly including Th2 cells. Other studies have implicated the Th1 cytokines IFN-γ and TNF-α in the activation of epithelial cell gene expression including IL-8 production. In addition, these cytokines enhance bacterial binding, apoptosis, as well as inflammation, atrophy and dysplasia in an animal model. Th1 cytokines, such as IFN-γ and TNF-α, also augment H. pylori-induced gastric epithelial death via apoptosis. Further, TNF-α, IFN-γ, and IL-1β upregulate gastric mucosal Fas antigen expression. Because Th1 cells express higher levels of FasL than Th2 cells, the relative increase in Th1 cells during H. pylori infection may induce epithelial cell death through Fas–FasL interactions. Although B cells have been studied the most in the context of autoreactivity, proton-pump-specific Th1 cells in the gastric mucosa may act as effector cells in the targeted destruction of the H+,K+-ATPase in autoimmune gastritis. However, during H. pylori infection, in the absence of any organ-specific autoimmunity, T cells reactive to H+,K+-ATPase were not detected. This led to the conclusion that T-cell autoreactivity arises more selectively. Potential Beneficial Effects of Other Th Subsets In contrast to the pathogenic effects of Th1 cells, the anti-inflammatory effects of cytokines associated with Th2 cells, or possibly other regulatory subsets of Th cells can attenuate gastric inflammation. More direct evidence suggests that IL-4 can decrease gastritis and

CHAPTER 57 / THE MOLECULAR MECHANISMS OF H. PYLORI-ASSOCIATED GASTRODUODENAL DISEASE

the effects of this cytokine may be mediated by the release of somatostatin. These interesting studies illustrate the important convergence of the flora, the host response, and other physiological mediators to modulate mucosal homeostasis. Additional studies will determine if the manipulation of T-cell responses can control the changes in epithelial cell atrophy and dysplasia in these animal models. The fact that the gastric response can be modified by Th2 cells raises the question regarding the role of other T-cell subsets, such as regulatory T cells (Treg), in the pathogenesis of disease associated with H. pylori infection. Depletion of Treg in neonatal mice leads to autoimmune gastritis implying that Treg play an important role in the control of gastritis. The report that infection with H. pylori improves the autoimmune gastritis induced in neonatal mice suggests that infection may indeed stimulate a subset of antiinflammatory T cells that impair excessive inflammation that could otherwise lead to the spontaneous clearance of the organism. There are several precedents for this in immunity and future studies may prove such a mechanism exists for prolonging H. pylori infection. Gastric B-Cell Responses Normally, antibodies in the gastrointestinal tract are of the IgA isotype, as IgA antibodies are highly adapted for mucosal protection. IgA has structural features that allow them to be transported selectively into the lumen with mucosal secretions, resist acid proteolysis and enzymatic degradation, and block bacterial attachment or neutralize toxins in the lumen. Moreover, they confer protective immunity without activating complement and stimulating deleterious amounts of inflammation. During infection with H. pylori, IgA producing cells are increased; however, IgG and IgM are also detected along with activated complement. Therefore, local immune complexes may be contributing to gastroduodenal inflammation and tissue damage during infection. Direct evidence for autoantibody production in response to H. pylori requires the identification of the self-antigens that are recognized by these antibodies. Given the structural homology between Le antigens on H. pylori and host cells, an obvious hypothesis is that these antigens will contribute to the described autoreactive responses. Indeed in the mouse models, immunization with H. pylori induces antibody-producing cells that recognize Lex/y that are capable of inducing gastritis. Furthermore, an antibody response directed to gastric parietal cells during H. pylori infection correlates with increased corpus atrophy. In addition, ferrets naturally infected with H. mustelae generate antigastric antibodies that recognize parietal cells. The cellular target for these antibodies in mice appears to be the β-chain of the gastric H+,K+-ATPase localized in the parietal cell canaliculi in the corpus. Anti-Le antibodies have been described in humans and occur independently of the Le phenotype of the host. However, they do not appear to account for autoreactivity. Thus, it seems logical that autoantibodies induced in mice may recognize different targets within the gastric mucosa and may even cross-react with human gastric tissue. However, the autoantibodies induced in humans may have a completely different specificity. Independent reports have shown that gastric antibodies can resemble rheumatoid factors because they recognize other antibodies. Thus, a complex of antibodies recognized by adjacent antibodies could trigger immune complex-mediated inflammation. Other investigators have shown that antibodies associated with H. pylori infection recognize gastric epithelial cells.

597

Moreover, monoclonal antibodies that recognize H. pylori crossreact to both human and murine gastric epithelial cells. Adoptive transfer of monoclonal antibodies to recipient mice induces gastritis as does the transfer of B-cells recognizing heat shock proteins from subjects with maltoma. Two groups have suggested that the level of autoantibodies in humans correlates with the severity of the gastritis. Thus, direct and indirect evidence supports the idea that autoantibodies can be induced during infection. Is There an Immunological Basis for Persistence? With few exceptions, infection with H. pylori persists for the life of the host unless there is some intervention with antibiotics. This has led investigators to ask if a form of immunological avoidance or tolerance exists that prevents immunity from developing. Although infection leads to the recruitment and activation of innate responses that could lead to potentially protective, adaptive responses, there is evidence that H. pylori disrupts the processes that could contribute to immunity. As discussed, several bacterial factors including catalase and urease antagonize innate host responses. Virulent strains of H. pylori also impair phagocytosis. The VacA toxin impairs antigen presentation by macrophages by inhibiting the Iidependent pathway mediated by newly synthesized class II MHC molecules. Moreover, H. pylori express antigenic molecules that mimic host molecules, such as Lewis antigens, and self-antigens normally stimulate T cells to release cytokines that protect from autoimmune reactions. However, the cytokine profile associated with H. pylori infection does not obviously resemble the expected version in a tolerant environment. For example, IL-4, IL-10, and transforming growth factor-β (which could mediate an antiinflammatory effect) are not expressed to the same levels as the proinflammatory cytokines like IFN-γ and TNF-α. That is not to say that the expression of potentially anti-inflammatory cytokines that exists is not sufficient to impair immune responses and favor persistence but some additional data are required to support this model. Additionally, the infected gastric mucosa is characterized by chronic, active inflammation. Thus, tolerance, if it has occurred, may be relative. Some evidence for a role in immune avoidance for the persistence of H. pylori can be found in animal models in which infection is cleared by immunization procedures that induce a substantial amount of gastritis. It remains to be determined if immunity is resulting from a higher degree of inflammation than what is observed in response to “natural” infection or whether protection results from a specific and selected response induced by the immunization. Another perspective is that the T-cell responses may not be inappropriate but they may lack a level of coordination necessary to achieve immunity. Some groups have demonstrated that the bacteria are able to inhibit the growth of T cells and actually lead to the induction of apoptosis. This process has been demonstrated using cell lines and attributed to very specific patterns of gene expression including the induction of FasL and Fas on T cells that would enable apoptosis to proceed. Although studies suggest this can occur directly in a redox-sensitive manner, other reports suggest that peptides from H. pylori activate monocytes to produce oxygen radicals that impair the expression of CD3 zeta chain of the T-cell receptor complex as well as induce T-cell apoptosis. Support for the loss of T cells by apoptosis is given by the presence of apoptotic T cells in the gastric mucosa as well as the expression of Fas and FasL that was detected on biopsy specimens by immunohistochemistry.

598

SECTION VI / GASTROENTEROLOGY

VacA impairs T-cell growth and function independently of apoptosis in a T-cell line by inhibiting nuclear translocation of a transcription factor activation required for IL-2 production. Related studies illustrate the complexity of these events as the effects on cell lines differ from T cells isolated from human peripheral blood in that the toxin impairs IL-2-dependent expansion in freshly isolated T cells but not IL-2 production nor their survival. Others have shown that exposure of human T cells to H. pylori decreases their proliferative response by decreasing the expression of the CD3 zeta chain of the T-cell receptor complex. These reports support the model that H. pylori interferes with normal T-cell activation in several ways; however, the answer may lie in the T cells within the gastric mucosa because cell lines and peripheral blood are a model that is often not replicated within the tissue. For example, some antigen-specific T-cell responses can be induced in the gastric mucosa. Nonetheless, it is still possible that the effect of the organism as well as the cytokine milieu disrupt the coordination that is required for the development of an effective, antigen-specific T-cell response.

CONCLUSION The evidence presented above illustrates several changes associated with H. pylori infection that contribute to gastritis and alterations in epithelial cell biology. It is clear that these responses are induced by many different factors, which points to the importance of several pathways and the lack of an exclusive target that deserves preferential attention. There is the chance that a “final common pathway” may emerge as a key target that advances the understanding of the complex pathogenesis of gastroduodenal diseases. Insights at the molecular level will improve the understanding of the pathogenesis of these disorders and provide novel opportunities for diagnosis, treatment and modeling of gastrointestinal diseases consequential to chronic inflammation.

ACKNOWLEDGMENTS The author and some of the work referred to are supported by the National Institutes of Health Grants DK50980, DK56703, and RR00175.

SELECTED REFERENCES Aihara M, Tsuchimoto D, Takizawa H, et al. Mechanisms involved in Helicobacter pylori-induced interleukin-8 production by a gastric cancer cell line, MKN45. Infect Immun 1997;65:3218–3224. Allen LA, Schlesinger LS, Kang B. Virulent strains of Helicobacter pylori demonstrate delayed phagocytosis and stimulate homotypic phagosome fusion in macrophages. J Exp Med 2000;191:115–128. Amieva MR, Vogelmann R, Covacci A, Tompkins LS, Nelson WJ, Falkow S. Disruption of the epithelial apical-junctional complex by Helicobacter pylori CagA. Science 2003;300:1430–1434. Andersen-Nissen E, Smith KD, Strobe KL, et al. Evasion of Toll-like receptor 5 by flagellated bacteria. Proc Natl Acad Sci USA 2005; 102:9247–9252. Appelmelk BJ, Faller G, Claeys D, Kirchner T, Vandenbroucke-Grauls CM. Bugs on trial: the case of Helicobacter pylori and autoimmunity. Immunol Today 1998;19:296–299. Appelmelk BJ, Simoons-Smit I, Negrini R, et al. Potential role of molecular mimicry between Helicobacter pylori lipopolysaccharide and host Lewis blood group antigens in autoimmunity. Infect Immun 1996; 64:2031–2040. Argent RH, Kidd M, Owen RJ, Thomas RJ, Limb MC, Atherton JC. Determinants and consequences of different levels of CagA phosphorylation for clinical isolates of Helicobacter pylori. Gastroenterology 2004;127:514–523.

Atherton JC, Cao P, Peek RM, Tummuru MK, Blaser MJ, Cover TL. Mosaicism in vacuolating cytotoxin alleles of Helicobacter pylori. Association of specific vacA types with cytotoxin production and peptic ulceration. J Biol Chem 1995;270:17,771–17,777. Bamford KB, Fan XJ, Crowe SE, et al. Lymphocytes in the human gastric mucosa during Helicobacter pylori have a T helper cell 1 phenotype. Gastroenterology 1998;114:482–492. Basak C, Pathak SK, Bhattacharyya A, Pathak S, Basu J, Kundu M. The secreted peptidyl prolyl cis,trans-isomerase HP0175 of Helicobacter pylori induces apoptosis of gastric epithelial cells in a. J Immunol 2005;174:5672–5680. Berstad AE, Brandtzaeg P, Stave R, Halstensen TS. Epithelium related deposition of activated complement in Helicobacter pylori associated gastritis. Gut 1997;40:196–203. Beswick EJ, Das S, Pinchuk IV, et al. Helicobacter pylori-induced IL-8 production by gastric epithelial cells up-regulates CD74 expression. J Immunol 2005;175:171–176. Betten A, Bylund J, Cristophe T, et al. A proinflammatory peptide from Helicobacter pylori activates monocytes to induce lymphocyte dysfunction and apoptosis. J Clin Invest 2001;108:1221–1228. Blaser MJ. The biology of cag in the Helicobacter pylori-human interaction. Gastroenterology 2005;128:1512–1515. Blaser MJ, Atherton JC. Helicobacter pylori persistence: biology and disease. J Clin Invest 2004;113:321–333. Bliss CM Jr, Golenblock DT, Keates S, Linevsky JK, Kelly CP. Helicobacter pylori lipopolysaccharide binds to CD14 and stimulates release of interleukin-8, epithelial neutrophil-activating peptide 78, and monocyte chemotactic protein 1 by human monocytes. Infect Immun 1998;66:5357–5363. Boncristiano M, Paccani SR, Barone S, et al. The Helicobacter pylori vacuolating toxin inhibits T cell activation by two independent mechanisms. J Exp Med 2003;198:1887–1897. Boren T, Falk P, Roth KA, Larson G, Normark S. Attachment of Helicobacter pylori to human gastric epithelium mediated by blood group antigens. Science 1993;262:1892–1895. Brandt S, Kwok T, Hartig R, Konig W, Backert S. NF-kappaB activation and potentiation of proinflammatory responses by the Helicobacter pylori CagA protein. Proc Natl Acad Sci USA 2005;102:9300–9305. Brenner H, Arndt V, Stegmaier C, Ziegler H, Rothenbacher D. Is Helicobacter pylori infection a necessary condition for noncardia gastric cancer? Am J Epidemiol 2004;159:252–258. Ceponis PJ, McKay DM, Menaker RJ, Galindo-Mata E, Jones NL. Helicobacter pylori infection interferes with epithelial Stat6mediated interleukin-4 signal transduction independent of cagA, cagE, or VacA. J Immunol 2003;171:2035–2041. Chiou CC, Chan CC, Sheu DL, Chen KT, Li YS, Chan EC. Helicobacter pylori infection induced alteration of gene expression in human gastric cells. Gut 2001;48:598–604. Claeys D, Faller G, Appelmelk BJ, Negrini R, Kirchner T. The gastric H+, K+ -ATPase is a major autoantigen in chronic Helicobacter pylori gastritis with body mucosa atrophy. Gastroenterology 1998;115:340–347. Clyne M, Drumm B. Absence of effect of Lewis A and Lewis B expression on adherence of Helicobacter pylori to human gastric cells. Gastroenterology 1997;113:72–80. Clyne M, Dillon P, Daly S, et al. Helicobacter pylori interacts with the human single-domain trefoil protein TFF1. Proc Natl Acad Sci USA 2004;101:7409–7414. Correa P. Human gastric carcinogenesis: a multistep and multifactorial process—First American Cancer Society Award Lecture on Cancer Epidemiology and Prevention. Cancer Res 1992;52:6735–6740. Correa P. Bacterial infections as a cause of cancer. J Natl Cancer Inst 2003;95:E3. Covacci A, Telford JL, Del Giudice G, Parsonnet J, Rappuoli R. Helicobacter pylori virulence and genetic geography. Science 1999; 284:1328–1333. Cover TL. The vacuolating cytotoxin of Helicobacter pylori. Molecular Microbiology 1996;20:241–246. Cover TL, Krishna US, Israel DA, Peek RM Jr. Induction of gastric epithelial cell apoptosis by Helicobacter pylori vacuolating cytotoxin. Cancer Res 2003;63:951–957.

CHAPTER 57 / THE MOLECULAR MECHANISMS OF H. PYLORI-ASSOCIATED GASTRODUODENAL DISEASE

Cox JM, Clayton CL, Tomita T, Wallace DM, Robinson PA, Crabtree JE. cDNA array analysis of cag pathogenicity island-associated Helicobacter pylori epithelial cell response genes. Infect Immun 2001;69:6970–6980. Crabtree JE. Role of cytokines in pathogenesis of Helicobacter pyloriinduced mucosal damage. Dig Dis Sci 1998;43:46S–55S. Crabtree JE, Shallcross TM, Heatley RV, Wyatt JI. Mucosal tumour necrosis factor α and interleukin-6 in patients with Helicobacter pylori associated gastritis. Gut 1991;32:1473–1477. Crabtree JE, Covacci A, Farmery SM, et al. Helicobacter pylori induced interleukin-8 expression in gastric epithelial cells is associated with CagA positive phenotype. J Clin Pathol 1995;48:41–45. Croinin TO, Clyne M, Appelmelk BJ, Drumm B. Antigastric autoantibodies in ferrets naturally infected with Helicobacter mustelae. Infect Immun 2001;69:2708–2713. Day AS, Jones NL, Lynett JT,et al. cagE is a virulence factor associated with Helicobacter pylori-induced duodenal ulceration in children. J Infect Dis 2000;181:1370–1375. D’Elios MM, Bergman MP, Azzurri A, et al. H(+),K(+)-ATPase (proton pump) is the target autoantigen of Th1-type cytotoxic T cells in autoimmune gastritis. Gastroenterology 2001;120:377–386. D’Elios MM, Manghetti M, Almerigogna F, et al. Different cytokine profile and antigen-specificity repertoire in Helicobacter pylori-specific T cell clones from the antrum of chronic gastritis patients with or without peptic ulcer. Eur J Immunol 1997;27:1751–1755. D’Elios MM, Manghetti M, De Carli M, et al. T helper 1 effector cells specific for Helicobacter pylori in gastric antrum of patients with peptic ulcer disease. J Immunol 1997;158:962–967. Ding SZ, O’Hara AM, Denning TL, Crowe SE. Helicobacter pylori and H2O2 increases AP endonuclease-1/redox factor-1 expression in human gastric epithelial cells. Gastroenterology 2004;127:845–858. Eaton KA, Krakowa S. Chronic active gastritis due to Helicobacter pylori in immunized gnotobiotic piglets. Gastroenterology 1992; 103:1580–1586. Eaton KA, Mefford M, Thevenot T. The role of T cell subsets and cytokines in the pathogenesis of Helicobacter pylori gastritis in mice. J Immunol 2001;166:7456–7461. Eaton KA, Ringler SR, Danon SJ. Murine splenocytes induce severe gastritis and delayed-type hypersensitivity and suppress bacterial colonization in Helicobacter pylori-infected SCID mice. Infect Immun 1999;67:4594–4602. Eck M, Schmausser B, Scheller K, et al. CXC chemokines Gro(alpha)/IL-8 and IP-10/MIG in Helicobacter pylori gastritis. Clin Exp Immunol 2000;122:192–199. Edwards NJ, Monteiro MA, Faller G, et al. Lewis X structures in the O antigen side-chain promote adhesion of Helicobacter pylori to the gastric epithelium. Mol Microbiol 2000;35:1530–1539. Elliott SN, Ernst PB, Kelly CP. The year in Helicobacter pylori 2001: Molecular Inflammation. Curr Opin Gastroenterol 2002;17:s12–s18. El-Omar E, Carrington M, Chow W-H, et al. Interlekin-1 polymorphisms associated with increased risk of gastric cancer. Nature 2000;404: 398–402. El-Omar EM, Rabkin CS, Gammon MD, et al. Increased risk of noncardia gastric cancer associated with proinflammatory cytokine gene polymorphisms. Gastroenterology 2003;124:1193–1201. Ernst PB, Gold B. The disease spectrum of H. pylori: the immunopathogenesis of gastroduodenal ulcer and gastric cancer. Palo Alto Annu Rev 2000:615–640. Ernst PB, Ryan K, Goldberg J. What is the exact role of Lewis antigens and autoantibodies in Helicobacter related disease. In: Hunt RH and Tytgat GNJ, eds. Helicobacter pylori: Basic Mechanisms to Clinical Cure 2002. London: Kluwer, 2003; pp. 73–82. Falk P, Roth KA, Boren T, Westblom TU, Gordon JI, Normark S. An in vitro adherence assay reveals that Helicobacter pylori exhibits cell lineage-specific tropism in the human gastric epithelium. Proc Natl Acad Sci USA 1993;90:2035–2039. Faller G, Steininger H, Eck M, Hensen J, Hann EG, Kirchner T. Antigastric autoantibodies in Helicobacter pylori gastritis: Prevalence, in-situ binding sites and clues for clinical relevance. Virchows Arch 1996;427:483–486.

599

Fan XG, Chua A, Fan XJ, Keeling PW. Increased gastric production of interleukin-8 and tumour necrosis factor in patients with Helicobacter pylori infection. J Clin Pathol 1995;48:133–136. Fan XJ, Crowe SE, Behar S, et al. The effect of class II MHC expression on adherence of Helicobacter pylori and induction of apoptosis in gastric epithelial cells: a mechanism for Th1 cell-mediated damage. J Exp Med 1998;187:1659–1669. Fan X, Gunasena H, Cheng Z, et al. Helicobacter pylori urease binds to class II MHC on gastric epithelial cells and induces their apoptosis. J Immunol 2000;165:1918–1924. Figueiredo C, Machado JC, Pharoah P, et al. Helicobacter pylori and interleukin 1 genotyping: An opportunity to identify high-risk individuals for gastric carcinoma. J Natl Cancer Inst 2002;94: 1680–1687. Foryst-Ludwig A, Naumann M. p21-activated kinase 1 activates the nuclear factor kappa B (NF-kappa B)-inducing kinase-Ikappa B kinases NF-kappa B pathway and proinflammatory cytokines in Helicobacter pylori infection. J Biol Chem 2000;275:39,779–39,785. Fox JG, Beck P, Dangler CA, et al. Concurrent enteric helminth infection modulates inflammation and gastric immune responses and reduces helicobacter-induced gastric atrophy. Nat Med 2000;6:536–542. Franco AT, Israel DA, Washington MK, et al. Activation of beta-catenin by carcinogenic Helicobacter pylori. Proc Natl Acad Sci USA 2005; 102:10,646–10,651. Furuta T, El-Omar EM, Xiao F, et al. Interleukin 1beta polymorphisms increase risk of hypochlorhydria and atrophic gastritis and reduce risk of duodenal ulcer recurrence in Japan. Gastroenterology 2002;123:92–105. Gebert B, Fischer W, Weiss E, Hoffmann R, Haas R. Helicobacter pylori vacuolating cytotoxin inhibits T lymphocyte activation. Science 2003;301:1099–1102. Gerhard M, Lehn N, Neumayer N, et al. Clinical relevance of the Helicobacter pylori gene for blood-group antigen-binding adhesin. Proc Natl Acad Sci USA 1999;96:12,778–12,783. Gerhard M, Schmees C, Voland P, et al. A secreted low-molecular-weight protein from Helicobacter pylori induces cell-cycle arrest of T cells. Gastroenterology 2005;128:1327–1339. Gewirtz AT, Yu Y, Krishna US, Israel DA, Lyons SL, Peek RM Jr. Helicobacter pylori flagellin evades toll-like receptor 5-mediated innate immunity. J Infect Dis 2004;189:1914–1920. Gobert AP, Bambou JC, Werts C, et al. Helicobacter pylori heat shock protein 60 mediates interleukin-6 production by macrophages via a toll-like receptor (TLR)-2-, TLR-4-, and myeloid differentiation factor 88-independent mechanism. J Biol Chem 2004;279:245–250. Gobert AP, Mersey BD, Cheng Y, Blumberg DR, Newton JC, Wilson KT. Cutting edge: urease release by Helicobacter pylori stimulates macrophage inducible nitric oxide synthase. J Immunol 2002;168: 6002–6006. Graham DY, Malaty HM, Evans DG, Evans DJ Jr, Klein PD, Adam E. Epidemiology of Helicobacter pylori in an asymptomatic population in the United States. Effect of age, race, and socioeconomic status. Gastroenterology 1991;100:1495–1501. Greiner A, Marx A, Heesemann J, Leebmann J, Schmausser B, MullerHermelink HK. Idiotype identity in a MALT-type lymphoma and B cells in Helicobacter pylori associated chronic gastritis. Lab Invest 1994;70:572–578. Guillemin K, Salama NR, Tompkins LS, Falkow S. Cag pathogenicity island-specific responses of gastric epithelial cells to Helicobacter pylori infection. Proc Natl Acad Sci USA 2002;99:15,136–15,141. Guruge JL, Falk PG, Lorenz RG, et al. Epithelial attachment alters the outcome of Helicobacter pylori infection. Proc Natl Acad Sci USA 1998;95:3925–3930. Haeberle HA, Kubin M, Bamford KB, et al. Differential stimulation of interleukin-12 (IL-12) and IL-10 by live and killed Helicobacter pylori in vitro and association of IL-12 production with gamma interferon-producing T cells in the human gastric mucosa. Infect Immun 1997;65:4229–4235. Harris PR, Mobley HLT, Perez-Perez GI, Blaser MJ, Smith PD. Helicobacter pylori urease is a potent stimulus of mononuclear phagocyte activation and inflammatory cytokine production. Gastroenterology 1996;111:419–425.

600

SECTION VI / GASTROENTEROLOGY

Hase K, Murakami M, Iimura M, et al. Expression of LL-37 by human gastric epithelial cells as a potential host defense mechanism against Helicobacter pylori. Gastroenterology 2003;125:1613–1625. Hellmig S, Hampe J, Folsch UR, Schreiber S. Role of IL-10 promoter haplotypes in Helicobacter pylori associated gastric inflammation. Gut 2005;54:888. Heneghan MA, McCarthy CF, Janulaityte D, Moran AP. Relationship of anti-Lewis x and anti-Lewis y antibodies in serum samples from gastric cancer and chronic gastritis patients to Helicobacter pylori-mediated autoimmunity. Infect Immun 2001;69:4774–4781. Hida N, Shimoyama T Jr, Neville P, et al. Crabtree JE. Increased expression of IL-10 and IL-12 (p40) mRNA in Helicobacter pylori infected gastric mucosa: Relation to bacterial cag status and peptic ulceration. J Clin Pathol 1999;52:658–664. Higashi H, Nakaya A, Tsutsumi R, et al. Helicobacter pylori CagA induces Ras-independent morphogenetic response through SHP-2 recruitment and activation. J Biol Chem 2004;279:17,205–17,216. Higashi H, Tsutsumi R, Fujita A, et al. Biological activity of the Helicobacter pylori virulence factor CagA is determined by variation in the tyrosine phosphorylation sites. Proc Natl Acad Sci USA 2002; 99:14,428–14,433. Higashi H, Tsutsumi R, Muto S, et al. SHP-2 tyrosine phosphatase as an intracellular target of Helicobacter pylori CagA protein. Science 2002;295:683–686. Hisaeda H, Maekawa Y, Iwakawa D, et al. Escape of malaria parasites from host immunity requires CD4(+)CD25(+) regulatory T cells. Nat Med 2004;10:29, 30. Hwang IR, Kodama T, Kikuchi S, et al. Effect of interleukin 1 polymorphisms on gastric mucosal interleukin 1beta production in Helicobacter pylori infection. Gastroenterology 2002;123:1793–1803. Ilver D, Arnqvist A, Ogren J, et al. Helicobacter pylori adhesin binding fucosylated histo-blood group antigens revealed by retagging. Science 1998;279:373–377. Isomoto H, Mizuta Y, Miyazaki M, et al. Implication of NF-kappaB in Helicobacter pylori-associated gastritis. Am J Gastroenterol 2000; 95:2768–2776. Israel DA, Salama N, Arnold CN, et al. Helicobacter pylori strain-specific differences in genetic content, identified by microarray, influence host inflammatory responses. J Clin Invest 2001;107:611–620. Ito M, Haruma K, Kaya S, et al. Role of anti-parietal cell antibody in Helicobacter pylori-associated atrophic gastritis: evaluation in a country of high prevalence of atrophic gastritis. Scand J Gastroenterol 2002;37:287–293. Itoh T, Wakatsuki Y, Yoshida M, et al. The vast majority of gastric T cells are polarized to produce T helper 1 type cytokines upon antigenic stimulation despite the absence of Helicobacter pylori infection. J Gastroenterol 1999;34:560–570. Karttunen R, Karttunen T, Ekre H-PT, MacDonald TT. Interferon gamma and interleukin 4 secreting cells in the gastric antrum in Helicobacter pylori positive and negative gastritis. Gut 1995;36:341–345. Kavermann H, Burns BP, Angermuller K, et al. Identification and characterization of Helicobacter pylori genes essential for gastric colonization. J Exp Med 2003;197:813–822. Kawahara T, Kuwano Y, Teshima-Kondo S, et al. Toll-like receptor 4 regulates gastric pit cell responses to Helicobacter pylori infection. J Med Invest 2001;48:190–197. Kawahara Y, Yokota K, Mizuno M, et al. Antibodies to human gastric epithelial cells and heat shock protein 60 in Helicobacter pylori positive mucosa associated lymphoid tissue lymphoma. Gut 1999; 45:20–23. Keates S, Hitti YS, Upton M, Kelly CP. Helicobacter pylori infection activates NF-kB in gastric epithelial cells. Gastroenterology 1997;113: 1099–1109. Keates S, Keates AC, Warny M, Peek RM Jr, Murray PG, Kelly CP. Differential activation of mitogen-activated protein kinases in AGS gastric epithelial cells by cag+ and cag– Helicobacter pylori. J Immunol 1999;163:5552–5559. Koyama S. Apoptotic depletion of infiltrating mucosal lymphocytes associated with Fas ligand expression by Helicobacter pylori-infected gastric mucosal epithelium: human glandular stomach as a site of immune privilege. Dig Dis Sci 2000;45:773–780.

Kuwahara H, Miyamoto Y, Akaike T, et al. Helicobacter pylori urease suppresses bactericidal activity of peroxynitrite via carbon dioxide production. Infect Immun 2000;68:4378–4383. Lefebvre O, Chenard MP, Masson R, et al. Gastric mucosa abnormalities and tumorigenesis in mice lacking the pS2 trefoil protein. Science 1996;274:259–262. Lindholm C, Quiding-Jarbrink M, Lonroth H, Hamlet A, Svennerholm AM. Local cytokine response in Helicobacter pylori-infected subjects. Infect Immun 1998;66:5964–5971. Lohoff M, Rollinghoff M, Sommer F. Helicobacter pylori gastritis: a Th1 mediated disease? J Biotechnol 2000;83:33–36. Lu H, Hsu PI, Graham DY, Yamaoka Y. Duodenal ulcer promoting gene of Helicobacter pylori. Gastroenterology 2005;128:833–848. Luzza F, Parrello T, Monteleone G, et al. Up-regulation of IL-17 is associated with bioactive IL-8 expression in Helicobacter pylori-infected human gastric mucosa. J Immunol 2000;165:5332–5337. Macarthur M, Hold GL, El-Omar EM. Inflammation and Cancer II. Role of chronic inflammation and cytokine gene polymorphisms in the pathogenesis of gastrointestinal malignancy. Am J Physiol Gastrointest Liver Physiol 2004;286:G515–G520. Machado JC, Figueiredo C, Canedo P, et al. A proinflammatory genetic profile increases the risk for chronic atrophic gastritis and gastric carcinoma. Gastroenterology 2003;125:364-371. Maeda S, Yoshida H, Ogura K, et al. H. pylori activates NF-kappaB through a signaling pathway involving IkappaB kinases, NF-kappaB-inducing kinase, TRAF2, and TRAF6 in gastric cancer cells. Gastroenterology 2000;119:97–108. Mahdavi J, Sonden B, Hurtig M, et al. Helicobacter pylori SabA adhesin in persistent infection and chronic inflammation. Science 2002; 297:573–578. Mannick EE, Bravo LE, Zarama G, et al. Inducible nitric oxide synthase, nitrotyrosine and apoptosis in Helicobacter pylori gastritis: effect of antibiotics and antioxidants. Cancer Res 1996;56:3238–3243. Marshall BJ, Warren JR. Unidentified curved bacilli in the stomach of patients with gastritis and peptic ulceration. Lancet 1984;8390: 1311–1315. Mentis A, Blackwell CC, Weir DM, et al. ABO blood group, secretor status and detection of Helicobacter pylori among patients with gastric or duodenal ulcers. Epidemiol Infect 1991;106:221–229. Merrell DS, Goodrich ML, Otto G, Tompkins LS, Falkow S. pH-regulated gene expression of the gastric pathogen Helicobacter pylori. Infect Immun 2003;71:3529–3539. Meyer-Ter-Vehn T, Covacci A, Kist M, Pahl HL. Helicobacter pylori activates mitogen-activated protein kinase cascades and induces expression of the proto-oncogenes c-fos and c-jun. J Biol Chem 2000;275: 16,064–16,072. Michetti P, Corthesy-Theulaz I, Davin C, et al. Immunization of BALB/c mice against Helicobacter felis infection with Helicobacter pylori urease. Gastroenterology 1994;107:1002–1011. Mizushima T, Sugiyama T, Komatsu Y, Ishizuka J, Kato M, Asaka M. Clinical relevance of the babA2 genotype of Helicobacter pylori in Japanese clinical isolates. J Clin Microbiol 2001;39:2463–2465. Mobley HL, Island MD, Hausinger RP. Molecular biology of microbial ureases. Microbiological Reviews 1995;59:451–480. Mobley HLT, Hu L-T, Foxall PA. Helicobacter pylori urease: properties and role in pathogenesis. Scand J Gastroenterol 1991;26(Suppl 187):39–46. Mohammadi M, Nedrud J, Redline R, Lycke N, Czinn SJ. Murine CD4 T-cell response to Helicobacter infection: TH1 cells enhance gastritis and TH2 cells reduce bacterial load. Gastroenterology 1997;113: 1848–1857. Molinari M, Salio M, Galli C, et al. Selective inhibition of Ii-dependent antigen presentation by Helicobacter pylori toxin VacA. J Exp Med 1998;187:135–140. Monteiro MA, Chan KH, Rasko DA, Taylor DE, Zheng PY, Appelmelk BJ. Simultaneous expression of type 1 and type 2 Lewis blood group antigens by Helicobacter pylori lipopolysaccharides: Molecular mimicry between H. pylori lipopolysaccharides and human gastric epithelial cell surface glycoforms. J Biol Chem 1998;273:11,533–11,543.

CHAPTER 57 / THE MOLECULAR MECHANISMS OF H. PYLORI-ASSOCIATED GASTRODUODENAL DISEASE

Moss SF, Sordillo EM, Abdalla AM, et al. Increased gastric epithelial cell apoptosis associated with colonization with cagA+ Helicobacter pylori strains. Cancer Res 2001;61:1406–1411. Nagata K, Yu H, Nishikawa M, et al. Helicobacter pylori generates superoxide radicals and modulates nitric oxide metabolism. J Biol Chem 1998;273:14,071–14,073. Nakayama M, Kimura M, Wada A, et al. Helicobacter pylori VacA activates the p38/activating transcription factor 2-mediated signal pathway in AZ-521 cells. J Biol Chem 2004;279:7024–7028. Naumann M, Crabtree JE. Helicobacter pylori-induced epithelial cell signalling in gastric carcinogenesis. Trends Microbiol 2004;12:29–36. Naumann M, Wessler S, Bartsch C, et al. Activation of activator protein 1 and stress response kinases in epithelial cells colonized by Helicobacter pylori encoding the cag pathogenicity island. J Biol Chem 1999;274:31,655–31,662. Negrini R, Lisato L, Zanella I, et al. Helicobacter pylori infection induces antibodies cross-reacting with human gastric mucosa. Gastroenterology 1991;101:437–445. Negrini R, Savio A, Poiesi C, et al. Antigenic mimicry between Helicobacter pylori and gastric mucosa in the pathogenesis of body atrophic gastritis. Gastroenterology 1996;111:655–665. Niv Y, Fraser G, Delpre G, et al. Helicobacter pylori infection and blood groups. Am J Gastroenterol 1996;91:101–104. Noach LA, Bosma NB, Jansen J, Hoek FJ, van-Deventer SJ, Tytgat GN. Mucosal tumor necrosis factor-alpha, interleukin-1 beta and interleukin-8 production in patients with Helicobacter pylori. Scand J Gastroenterol 1994;29:425–429. Ogura K, Maeda S, Nakao M, et al. Virulence factors of Helicobacter pylori responsible for gastric diseases in Mongolian gerbil. J Exp Med 2000;192:1601–1610. Oh JD, Karam SM, Gordon JI. Intracellular Helicobacter pylori in gastric epithelial progenitors. Proc Natl Acad Sci USA 2005;102: 5186–5191. O’Hara AM, Mifflin RC, Ryan KA, et al. Silencing of apurinic/apyrimidinic endonuclease-1/redox factor-1 by RNA interfence inhibits Helicobacter pylori-induced transcription factor activation and interleukin-8 secretion in human gastric epithelial cells. Gastroenterology 2004;126:A118. Okazaki K, Ohana M, Oshima C, et al. Interaction of Helicobacter pyloriinduced follicular gastritis and autoimmune gastritis in BALB/c mice with post-thymectomy autoimmune gastritis. J Gastroenterol 2003;38:1131–1137. O’Neil DA, Cole SP, Martin-Porter E, et al. Regulation of human betadefensins by gastric epithelial cells in response to infection with Helicobacter pylori or stimulation with interleukin-1. Infect Immun 2000;68:5412–5415. Parsonnet J. The incidence of Helicobacter pylori infection. Aliment Pharmacol Ther 1995;9(Suppl 2):45–51. Paziak-Domanska B, Chmiela M, Jarosinska A, Rudnicka W. Potential role of CagA in the inhibition of T cell reactivity in Helicobacter pylori infections. Cell Immunol 2000;202:136–139. Peek RM Jr. IV. Helicobacter pylori strain-specific activation of signal transduction cascades related to gastric inflammation. Am J Physiol Gastrointest Liver Physiol 2001;280:G525–G530. Phadnis SH, Parlow MH, Levy M, et al. Surface localization of a Helicobacter pylori urease and heat shock protein homolog requires bacterial lysis. Infect Immun 1995;64:905–912. Prinz C, Schoniger M, Rad R, et al. Key importance of the Helicobacter pylori adherence factor blood group antigen binding adhesin during chronic gastric inflammation. Cancer Res 2001;61:1903–1909. Rad R, Gerhard M, Lang R, et al. The Helicobacter pylori blood group antigen-binding adhesin facilitates bacterial colonization and augments a nonspecific immune response. J Immunol 2002;168: 3033–3041. Rad R, Prinz C, Neu B, et al. Synergistic effect of Helicobacter pylori virulence factors and interleukin-1 polymorphisms for the development of severe histological changes in the gastric mucosa. J Infect Dis 2003;188:272–281. Rieder G, Einsiedl W, Hatz RA, Stolte M, Enders GA, Walz A. Comparison of CXC chemokines ENA-78 and interleukin-8

601

expression in Helicobacter pylori-associated gastritis. Infect Immun 2001;69:81–88. Sachs G, Weeks DL, Melchers K, Scott DR. The gastric biology of Helicobacter pylori. Annu Rev Physiol 2003;65:349–369. Sakaguchi S, Fukuma K, Kuribayashi K, Masuda T. Organ-specific autoimmune diseases induced in mice by elimination of T cell subset. I. Evidence for the active participation of T cells in natural self-tolerance; deficit of a T cell subset as a possible cause of autoimmune disease. J Exp Med 1985;161:72–87. Satin B, Del Giudice G, Della BV, et al. The neutrophil-activating protein (HP-NAP) of Helicobacter pylori is a protective antigen and a major virulence factor. J Exp Med 2000;191:1467–1476. Scott DR, Weeks D, Hong C, Postius S, Melchers K, Sachs G. The role of internal urease in acid resistance of Helicobacter pylori. Gastroenterology 1998;114:58–70. Segal ED, Cha J, Lo J, Falkow S, Tompkins LS. Altered states: involvement of phosphorylated CagA in the induction of host cellular growth changes by Helicobacter pylori. Proc Natl Acad Sci USA 1999; 96: 14,559–14,564. Sepulveda AR, Tao H, Carloni E, Sepulveda J, Graham DY, Peterson LE. Screening of gene expression profiles in gastric epithelial cells induced by Helicobacter pylori using microarray analysis. Aliment Pharmacol Ther 2002;16(Suppl 2):145–157. Sharma SA, Tummuru MK, Blaser MJ, Kerr LD. Activation of IL-8 gene expression by Helicobacter pylori is regulated by transcription factor nuclear factor-kappa B in gastric epithelial cells. J Immunol 1998;160:2401–2407. Sherburne R, Taylor DE. Helicobacter pylori expresses a complex surface carbohydrate, Lewis X. Infect Immun 1995;63:4564–4568. Shimoyama T, Everett SM, Dixon MF, Axon AT, Crabtree JE. Chemokine mRNA expression in gastric mucosa is associated with Helicobacter pylori cagA positivity and severity of gastritis. J Clin Pathol 1998;51:765–770. Smith H, Lou YH, Lacy P, Tung KS. Tolerance mechanism in experimental ovarian and gastric autoimmune diseases. J Immunol 1992;149: 2212–2218. Smith MG, Hold GL, Rabkin CS, et al. The IL-8-251 promoter polymorphism is associated with high IL-8 production, severe infammation and increased risk of pre-malignant changes in H. pylori positive subjects. Gastroenterology 2004;126:A23. Smith MF Jr, Mitchell A, Li G, et al. TLR2 and TLR5, but not TLR4, are required for Helicobacter pylori-induced NF-kappa B activation and chemokine expression by epithelial cells. J Biol Chem 2003;278: 32,552–32,560. Smythies LE, Waites KB, Lindsey JR, Harris PR, Ghiara P, Smith PD. Helicobacter pylori-induced mucosal inflammation is Th1 mediated and exacerbated in IL-4, but not IFN-gamma, gene-deficient mice. J Immunol 2000;165:1022–1029. Stark MA, Huo Y, Burcin TL, Morris MA, Olson TS, Ley K. Phagocytosis of apoptotic neutrophils regulates granulopoiesis via IL-23 and IL-17. Immunity 2005;22:285–294. Su B, Ceponis PJ, Lebel S, Huynh H, Sherman PM. Helicobacter pylori activates Toll-like receptor 4 expression in gastrointestinal epithelial cells. Infect Immun 2003;71:3496–3502. Suerbaum S, Michetti P. Helicobacter pylori infection. N Engl J Med 2002;347:1175–1186. Sundrud MS, Torres VJ, Unutmaz D, Cover TL. Inhibition of primary human T cell proliferation by Helicobacter pylori vacuolating toxin (VacA) is independent of VacA effects on IL-2 secretion. Proc Natl Acad Sci USA 2004;101:7727–7732. Takahashi S, Keto Y, Fujita T, Uchiyama T, Yamamoto A. FR167653, a p38 mitogen-activated protein kinase inhibitor, prevents Helicobacter pylori-induced gastritis in Mongolian gerbils. J Pharmacol Exp Ther 2001;296:48–56. Takata T, El-Omar E, Camorlinga M, et al. Helicobacter pylori does not require Lewis X or Lewis Y expression to colonize C3H/HeJ mice. Infect Immun 2002;70:3073–3079. Tanaka J, Suzuki T, Mimuro H, Sasakawa C. Structural definition on the surface of Helicobacter pylori type IV secretion apparatus. Cell Microbiol 2003;5:395–404.

602

SECTION VI / GASTROENTEROLOGY

Taylor DE, Rasko DA, Sherburne R, Ho C, Jewell LD. Lack of correlation between Lewis antigen expression by Helicobacter pylori and gastric epithelial cells in infected patients. Gastroenterology 1998;115: 1113–1122. Tebbutt NC, Giraud AS, Inglese M, et al. Reciprocal regulation of gastrointestinal homeostasis by SHP2 and STAT-mediated trefoil gene activation in gp130 mutant mice. Nat Med 2002;8:1089–1097. Tomita T, Jackson AM, Hida N, et al. Expression of interleukin-18, a Th1 cytokine, in human gastric mucosa is increased in Helicobacter pylori infection. J Infect Dis 2001;183:620–627. Tsutsumi R, Higashi H, Higuchi M, Okada M, Hatakeyama M. Attenuation of Helicobacter pylori CagA x SHP-2 signaling by interaction between CagA and C-terminal Src kinase. J Biol Chem 2003;278:3664–3670. Tummuru MKR, Sharma SA, Blaser MJ. Helicobacter pylori picB, a homologue of the Bordetella pertussis toxin secretion protein, is required for induction of IL-8 in gastric epithelial cells. Mol Microbiol 1995;18:867–876. Uehara A, Okumura T, Sekiya C, Okaura K, Takasugi Y, Namiki M. Interleukin-1 inhibits the secretion of gastric acid in rats: Possible involvement of prostaglandin. Biochem Biophys Res Commun 1989;162:1578–1584. Uemura N, Okamoto S, Yamamoto S, et al. Helicobacter pylori infection and the development of gastric cancer. N Engl J Med 2001;345: 784–789. Valnes K, Brandtzaeg P, Elgjo K, Stave R. Quantitative distribution of immunoglobulin-producing cells in gastric mucosa: relation to chronic gastritis and glandular atrophy. Gut 1986;27:505–514. van Den Brink GR, ten Kate FJ, Ponsioen CY, et al. Expression and activation of NF-kappa B in the antrum of the human stomach. J Immunol 2000;164:3353–3359. van Doorn LJ, Figueiredo C, Megraud F, et al. Geographic distribution of vacA allelic types of Helicobacter pylori. Gastroenterology 1999; 116:823–830. Velin D, Bachmann D, Bouzourene H, Michetti P. Mast cells are critical mediators of vaccine-induced Helicobacter clearance in the mouse model. Gastroenterology 2005;129:142–155. Wagner S, Beil W, Westermann J, et al. Regulation of epithelial cell growth by Helicobacter pylori: Evidence for a major role of apoptosis. Gastroenterology 1997;113:1836–1847. Wang J, Brooks EG, Bamford KB, Denning TL, Pappo J, Ernst PB. Negative selection of T cells by Helicobacter pylori as a model for

bacterial strain selection by immune evasion. J Immunol 2001;167: 926–934. Wang J, Fan XJ, Lindholm C, et al. Helicobacter pylori modulates lymphoepithelial cell interactions leading to epithelial cell damage through Fas/Fas Ligand interactions. Infect Immun 2000;68:4303–4311. Wex T, Treiber G, Nilius M, Vieth M, Roessner A, Malfertheiner P. Helicobacter pylori-mediated gastritis induces local downregulation of secretory leukocyte protease inhibitor in the antrum. Infect Immun 2004;72:2383–2385. Yahiro K, Wada A, Nakayama M, et al. Protein-tyrosine phosphatase alpha, RPTP alpha, is a Helicobacter pylori VacA receptor. J Biol Chem 2003;278:19,183–19,189. Yamaoka Y, Kwon DH, Graham DY. A M(r) 34,000 proinflammatory outer membrane protein (oipA) of Helicobacter pylori. Proc Natl Acad Sci USA 2000;97:7533–7538. Yamaoka Y, Kikuchi S, El Zimaity HM, Gutierrez O, Osato MS, Graham DY. Importance of Helicobacter pylori oipA in clinical presentation, gastric inflammation, and mucosal interleukin 8 production. Gastroenterology 2002;123:414–424. Yamaoka Y, Kudo T, Lu H, Casola A, Brasier AR, Graham DY. Role of interferon-stimulated responsive element-like element in interleukin-8 promoter in Helicobacter pylori infection. Gastroenterology 2004;126:1030–1043. Yasumoto K, Okamoto S, Mukaida N, Murakami S, Mai M, Matsushima K. Tumor necrosis factor α and interferon γ synergistically induce interleukin 8 production in a human gastric cancer cell line through acting concurrently on AP-1 and NF-kappaB-like binding sites of the interleukin 8 gene. J Biol Chem 1992;267:22,506–22,511. Yokota K, Kobayashi K, Kawahara Y, et al. Gastric ulcers in SCID mice induced by Helicobacter pylori infection after transplanting lymphocytes from patients with gastric lymphoma. Gastroenterology 1999; 117:893–899. Yokoyama K, Higashi H, Ishikawa S, et al. Functional antagonism between Helicobacter pylori CagA and vacuolating toxin VacA in control of the NFAT signaling pathway in gastric epithelial cells. Proc Natl Acad Sci USA 2005;102:9661–9666. Zabaleta J, McGee DJ, Zea AH, et al. Helicobacter pylori arginase inhibits T cell proliferation and reduces the expression of the TCR zeta-chain (CD3zeta). J Immunol 2004;173:586–593. Zavros Y, Rathinavelu S, Kao JY, et al. Treatment of Helicobacter gastritis with interleukin-4 requires somatostatin. Proc Natl Acad Sci 2003;100:12,944–12,949.

NEPHROLOGY VII SECTION EDITOR:

WILLIAM E. MITCH

Abbreviations VII. NEPHROLOGY AA ACE ACEi Ach ADAM-2 ADCC ADMA AGEs AICD AIF AIPs Akt Ang AP-1 Apaf-1 APC AQP2 ARF AT1-R ATPase AVP AVPR2 BH4 Bk BMP BP Ca++ CAAX box

cAMP CAT cGMP cNOS Co-Smad COX COX-1 COX-2 CTL DAG DCs dDAVP DI DTH ECM

arachidonic acid angiotensin-converting enzyme angiotensin-converting enzyme inhibitors acetylcholine a disintegrin domain and metalloproteinase antibody-dependent cell-mediated cytotoxicity asymmetric dimethyl arginine advanced glycation end products activation-induced cell death apoptosis inducing factor aldosterone-induced proteins protein kinase B angiotensin activator protein-1 apoptotic protease-activating factor-1 antigen-presenting cell aquaporin-2 acute renal failure angiotensin II type 1 receptors adenosine triphosphatase vasopressin, also called antidiuretic hormone arginine vasopressin 2 receptor the reduced form of tetrahydrobiopterin bradykinin bone morphogenetic protein blood pressure calcium a methylated cysteine residue followed by two aliphatic residues, followed by any residue at the C-terminal end of a protein cyclic adenosine monophosphate cationic amino acid transporter cyclic guanosine monophosphate constitutive NOS Smad4 cyclooxygenase cyclooxygenase-1 cyclooxygenase-2 cytotoxic T-lymphocytes diacylglycerol dendritic cells desmopressin diabetes insipidus delayed type hypersensitivity extracellular matrix

EDRF Em ENaCs Endo G eNOS ERK GC GEF GFR G-kinase GTP H2O2 HGF HLA IAPs ICAM-1 ICCa++ ICMT IFN IL IL-8 iNOS IP3 IκB I-Smads JNK L L-NAME LTα–/– LTβR–/– MAP MAPK MCP-1 MEK MEK-1 MEKK-1 MHC MHC-II Mn MREs MURF1 NADPH NDI Nedd4

NF-κB NK

604

endothelium-derived relaxing factor membrane potential epithelial Na channels endonuclease G endothelial or type III NOS extracellular signal regulated kinase guanyl cyclase guanine nucleotide exchange factor glomerular filtration rate CGMP kinase guanidine triphosphatase hydrogen peroxide hepatocyte growth factor human leukocytic antigens inhibitor of apopotosis intracellular adhesion molecule 1 intracellular calcium isoprenyl-cysteine-O-carboxy-methyl-transferase interferon interleukin interleukin-8 inducible or type II NOS 1,4,5-inositol trisphosphate inhibitor of κB inhibitory Smads Jun N-terminal kinase ligand L-Nitro arginine methyl ester splenectomized, lymphtoxin α-deficient mice lymphotoxin β-receptor-deficient mice mitogen-activated protein mitogen-activated protein kinase monocyte chemoattractant protein-1 mitogen-activated protein MAP kinase/ERK kinase mitogen-activated protein kinase kinase 1 major histocompatibility complex major histocompatibility complex II mitochondrial mineralocorticoid response elements muscle ring finger nicotine adenine dinucleotide phosphate nephrogenic diabetes insipidus a specific ubiquitin ligase: neural-precursorcell-expressed-developmentally downregulated protein nuclear factor-κB natural killer

CHAPTER 58 / NITRIC OXIDE SYNTHASE AND CYCLOOXYGENASE

nNOS NO NO2– NOS NOS-2 O2·– ONOO– PAMPS PDE-II PDE-III PDK1/2 PDZ PG PGE2 PGH2 PGI2 PGI2-S PI-3-kinase PIP2 PIP3 PKC PKG PLA2 RAGE RANTES RBCs ROS

neuronal or type I NOS nitric oxide nitrate nitric oxide synthase inducible NOS superoxide anion peroxynitrite pathogen-associated molecular patterns phosphodiesterase-II phosphodiesterase-III phosphatidylinositol-dependent kinases PSD-95, Discs-large, 20-1 prostaglandin prostaglandin E2 prostaglandin H2 prostacyclin prostacyclin synthase phosphatidylinositol 3-kinase phosphatidylinositol bisphosphate phosphatidylinositol trisphosphate protein kinase C protein kinase G phospholipase A2 receptor for advanced glycation end products regulated on activation, normal T-cell expressed red blood cells reactive oxygen species

R-Smads SAH SDMA SGK SHR SLC Smads SOD SR TβRI TβRII TCR TCRz TGF TGF-β Th TLR TNF TNFR TRAIL (DR4) Treg TxA2 UbC UT-B V2R VSMCs

605

receptor-regulated or receptor-associated Smads S-adenosyl-homocysteine symmetric dimethyl arginine serum-and-glucocorticoid-dependent kinase spontaneously hypertensive rat secondary lymphoid chemokine signaling pathways of TGF-β superoxide dismutase sarcoplasmic reticulum type I serine/threonine kinase receptors type II serine/threonine kinase receptors T-cell receptor for antigen T-cell receptor 2 chain transforming growth factor transforming growth factor-β CD4 helper toll-like receptor tumor necrosis factor tumor necrosis factor receptor TNF-related apoptosis inducing ligand the CD4+ T-cell constitutively expressing CD25+ on its surface thromboxane A2 ubiquitin C3 subunit gene the RBC-facilitated urea transporter V2-receptors vascular smooth muscle cells

58 Nitric Oxide Synthase

and Cyclooxygenase in the Kidneys CHRISTOPHER S. WILCOX

SUMMARY Three isoforms of nitric oxide synthase (NOS) and two of cyclooxygenase (COX) are expressed in the normal kidney. Neuronal (type I) NOS is expressed heavily in the macula densa, where it blunts the tubuloglomerular feedback response that increases afferent arteriolar tone in proportion to distal NaCl delivery and in the cortical and inner medullary collecting ducts where it impairs tubular Na+ entry. Endothelial (type III) NOS is expressed in endothelial cells where it mediates the endothelium-derived relaxing factor response and in the thick ascending limb, where it inhibits tubular NaCl reabsorption. Inducible (type II) NOS is widely expressed after cytokine challenge. It participates in cell death following sepsis or ischemia of the kidney. COX-1 is expressed in endothelial and smooth muscle cells of renal vasculature and in the medullary collecting ducts where it antogonizes arginine vasopressin and inhibits Na+ reabsorption. COX-2 is expressed in the macula densa cells and adjacent thick ascending limb where it enhances renin section and inhibits tubular NaCl reabsorption. NOS and COX isoforms interact extensively to fulfill their major roles of adapting renal function to changes in salt and water intake or blood pressure. Key Words: Tabuloglomerular; feedback; hypertension; nitric oxide; prostaglandin; oxidative stress.

INTRODUCTION Nitric oxide synthase (NOS) and cyclooxygenase (COX) are molecular targets that mediate important actions in the kidney. They are coexpressed in many vascular and epithelial cells, and mediate vasoconstriction, tubular reabsorption, renin secretion, and mediator release. Interactions are discussed at the end of this chapter.

NITRIC OXIDE SYNTHASE ISOFORMS NOS catalyzes the two-step oxygenation of arginine via hydroxyarginine to citrulline and NO. The enzyme has a Km for L-arginine in the low micromolar range, consumes molecular O2, oxidizes NADPH, and has a requirement for the reduced form of tetrahydrobiopterin (BH4), flavine adenine dinucleotide, and flavine mononucleotide. Under conditions of reduced availability of substrate or BH4, O2 is incompletely reduced with the preferential From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

formation of superoxide anion (O2⋅–) rather than NO. This converts NOS from a source of the antihypertensive, anti-inflammatory NO to the prohypertensive, proinflammatory O2⋅–. Three NOS isoforms have been cloned from the kidney. Two of these, neuronal (n) or type-I NOS and endothelial (e) or type-III NOS, are expressed constitutively. A third isoform, inducible (i) or type-II NOS, has low or absent basal expression in most organs, but is subject to major induction during sepsis, inflammation, or cytokine stimulation. However, in the kidney, constitutive (c) NOS isoforms have transcriptional and post-transcriptional regulation and sufficient iNOS is expressed basally to regulate normal renal function. A second important distinction between cNOS and iNOS concerns their activation. Both enzymes have an absolute requirement for calcium (Ca2+) and calmodulin, but, although changes in intracellular calcium (IC Ca 2 +) regulate cNOS, Ca2+ and calmodulin are tightly bound to the active site of iNOS whose regulation is effectively independent of changes in IC Ca 2 +. These differences provide a molecular explanation for the different functions of these two classes of enzymes. cNOS is preformed in cells, subject to rapid activation/inactivation via changes in IC Ca 2 + or enzyme phosphorylation mediated via Akt, and generates discrete, low-level (nanomolar) bursts of NO, which function in an autocrine or paracrine manner. In contrast, renal iNOS has some basal activity, but is capable of substantial transcriptional upregulation during stimulation with cytokines or growth factors to produce large (micromolar) sustained concentrations of NO that dictate changes in whole organ function that persist for hours or days. The kidney is unusual in its high basal rate of generation of cytokines, which may be the explanation for iNOS expression in the normal kidney. NO GENERATION AND METABOLISM NO is a mediator of the endothelium-derived relaxing factor (EDRF) response. Investigators showed that EDRF is inactivated rapidly at high pO2, but this is prevented by superoxide dismutase (SOD), which metabolizes O2– to hydrogen peroxide. They concluded that EDRF is a labile mediator subject to rapid oxidative bioinactivation. Chemiluminescence was used to demonstrate unequivocally that NO from the aorta is released by acetylcholine and functions as an EDRF. NO is the source of endogenous nitrate (NO2–) generation. Molecular Targets for NO in Renal Blood Vessels Signaling pathways have been defined whereby NO activates guanyl cyclase (GC) to generate cyclic guanosine monophosphate (cGMP). This

606

CHAPTER 58 / NITRIC OXIDE SYNTHASE AND CYCLOOXYGENASE

607

Figure 58-1 Cell diagram showing some pathways for NOS regulation in endothelial cells and for NO action in VSMCs. For details, see Molecular Targets for NO in Renal Blood Vessels.

activates cGMP kinase (G kinase) to stimulate protein kinase G (PKG) (Fig. 58-1). PKG activates myosin light chain phosphatase that dephosphorylates one of the light chains of each myosin head, which is also called the regulatory chain. This dephosphorylation dissociates the cross-bridges between the myosin head and the actin filaments, and prevents the attachment–detachment cycling of the myosin head with the actin filament. This leads to relaxation of the vascular smooth muscle cells (VSMCs) that is detected in pharmacological studies as an EDRF response. NO also reduces IC Ca 2 + in VSMCs. NO directly activates (1) a Ca2+ ATPase on the membrane of the sarcoplasmic reticulum, thereby sequestering cytoplasmic Ca2+ in the SR; (2) a Ca2+-dependent K+ channel in the cell membrane, thereby increasing the membrane potential (Em); (3) inactivating L-type voltage-gated Ca2+-channels; and (4) sodium–potassium ATPase. In addition, NO inhibits phospholipase C, causing decreased turnover of phosphoinositides.

NO can interact with many other heme-centered enzymes. For example, NO changes the affinity of hemoglobin for O2. NO binds to thiol groups, forming S-nitrosothiols that modify the actions of the target protein and stabilize NO by protecting it from oxidative attack, thereby prolonging its biological action. The interaction of NO with O2⋅– forms peroxynitrite (ONOO–) which effectively terminates many of NO’s actions. However, ONOO– is a potent oxidative and nitrosative agent that attacks tyrosine epitopes to form nitrotyrosine modification of proteins. These can be detected immunologically to provide insight into O2⋅– and NO interactions in the kidney. Nitrotyrosine modification can also alter function. For example, prostacyclin synthase (PGI2-S) is inactivated by ONOO– at a concentration of only 50 nM. Molecular Targets for NO in the Tubular Cells NO interacts with heme-centered enzymes, such as GC, expressed in tubular and juxtaglomerular cells to generate cGMP. cGMP can inhibit

608

SECTION VII / NEPHROLOGY

Figure 58-2 Cell diagram showing some pathways whereby NO generated by nNOS in the MD or eNOS in the vascular endothelium affects the tubuloglomerular feedback response or afferent arteriolar tone. For details, see Molecular Targets for NO in the Tubular Cells.

phosphodiesterase (PDE)-II in the cells of the thick ascending limb of the loop of Henle and the macula densa (MD). PDE-II metabolizes and inactivates cAMP (Fig. 58-2). The ensuing decrease in cAMP decreases the phosphorylation of the Na+/K+/2Cl– cotransporter in the luminal cell membrane, and thereby leads to its inactivation. NO also activates Ca2+-activated K+channels that increase Em and thereby enhance Na+ entry and reabsorption. Similar events in the juxtaglomerular cells of the afferent arteriole have complex effects on renin secretion. Renin secretion initially is decreased after NO by actions of cGMP to activate G-kinase and PKG. Later, renin secretion is stimulated via actions of cAMP to enhance protein kinase A. This occurs because NO inhibits PDE-III, thereby leading to a gradual increase in cAMP accumulation. NO Measurement NO released into the bloodstream reacts rapidly with hemoglobin to form nitrosohemoglobin which, on reduction, releases NO2– which can be further oxidized to NO3–. Cellular release of NO2– + NO3– (NOx) and renal excretion of NOx during consumption of a NOx-deficient diet are methods to quantitate NO generation. NOS activity in the tissues can be studied directly from the release of labeled citrulline from labeled arginine, from the partial pressure of NO measured with a porphyrinic NO-sensitive electrode, by the use of fluorescent dyes, such as diaminofluorescein-2, which are trapped within cells and fluoresce on reaction with NO, by spectrophotometric changes that occur when NO interacts with test proteins such as hemoglobin, by generation of NO signaling molecules such as cGMP, and from functional responses to blockade of NOS with competitive antagonists such as L-nitroarginine methyl ester. Distribution of NOS Isoforms in the Kidney Type-I (neuronal) NOS is heavily expressed in MD cells. Other sites include the cortical and inner medullary collecting ducts and Bowman’s

capsule of the glomerulus. There is a lower level of expression in the efferent arteriole. Type-III (endothelial) NOS is expressed primarily in the endothelial cells of the arterioles and capillaries of the kidney. eNOS is expressed also in the cells of the thick ascending limb, collecting ducts, and proximal tubules. Type-II (inducible) NOS can be widely expressed in the glomeruli, vessels, and tubules on challenge with lipopolysaccharide or cytokines. Basal expression is apparent in cells of the thick ascending limb, and the proximal and distal nephron. Regulation of NO Generation and Action NO generation is closely regulated. Cellular arginine uptake is mediated via cationic amino transporters (CATs) expressed on cell membranes (Figs. 58-1 and 58-2). There is competition for transport among the cationic amino acids, such as ornithine, lysine, and homoarginine, and with the dimethyl arginines, asymmetric dimethyl arginine (ADMA), and symmetric dimethyl arginine. All of these can inhibit arginine uptake and NO action (Fig. 58-1). Endothelial NOS in vascular endothelial cells is activated by a rise in IC Ca 2 + by agonists, such as acetylcholine, acting on M-3 receptors or bradykinin (Bk) acting on B2 receptors. Sheer stress is transduced by integrins expressed in caveolae, which activates Akt-1 which phosphorylates and activates eNOS. NOS activity in tubular or vascular cells can be inhibited by lack of availability of essential cofactors, notably BH4, which, during oxidative stress, is converted to the inactive form, dihydrobiopterin. ADMA is an endogenous inhibitor of NOS and CAT. ADMA (but not symmetric dimethyl arginine, which is ineffective as a NOS inhibitor) is metabolized to citrulline by dimethylarginine dimethylaminohydrolase (DDAH), which is widely expressed in the kidney at sites of NOS expression. The tubuloglomerular feedback response is a unique renal mechanism, whereby delivery and reabsorption of Cl– at the MD

CHAPTER 58 / NITRIC OXIDE SYNTHASE AND CYCLOOXYGENASE

segment leads to the elaboration of a mediator that vasoconstricts the adjacent afferent arteriole. Mediators include adenosine acting on type-1 receptors and adenosine triphosphate acting on purinergic type-2 receptors. The tubuloglomerular feedback response contributes to renal autoregulation and to adaptation of renal hemodynamics to changes in salt intake. Defects in tubuloglomerular feedback underlie salt-sensitive hypertension. Neuronal NOS also is activated in the MD during Cl– reabsorption. The ensuing increase in NO generation blunts the tubuloglomerular feedback response both by inhibiting vasoconstriction of the adjacent afferent arteriole and by inhibiting the generation of the tubuloglomerular feedback signal by blocking the luminal Cl– transporter (Fig. 58-2). An overactive tubuloglomerular feedback response in the spontaneously hypertensive rat (SHR) or the Dahl salt-sensitive rat can be ascribed to a blunted NO action in the juxtaglomerular apparatus because of enhanced inactivation of NO by O2⋅–. This overactive tubuloglomerular feedback response likely contributes to renal vasoconstriction and hypertension. Activation mechanisms for nNOS have been studied in the MD. Basal activity is negligible, but becomes substantial during delivery and reabsorption of NaCl at this segment. Reabsorption via the luminal Na+/K+/2Cl– cotransporter increases cell [Na+], which activates basolateral Na+/Ca2+ countertransport, thereby increasing IC Ca 2 + and activating NOS. nNOS is coexpressed with a protein having a postsynaptic density of 93 along the basolateral aspect of the MD cell. Its interaction with the PDZ domain on nNOS dimerizes and activates the enzyme (Fig. 58-2). NO is inactivated rapidly in renal tubular epithelial cells by interaction with O2⋅–. NADPH oxidase is responsible for much of the O2⋅– generation in the renal cortex. As studied most fully in the leukocyte, this enzyme requires assembly of a complex of five protein subunits with a small guanidine triphosphatase activating protein rac-1. When activated, NADPH oxidase reduces O2 to O2⋅–. All the major protein components of NADPH oxidase are expressed in VSMCs and endothelial cells of the renal vessels and in the distal nephron, most prominently in the MD. Other prominent sites of NADPH oxidase expression include interstitial fibroblasts and mesangial cells (Fig. 58-2). Expression of NADPH oxidase along the luminal aspect of MD and thick ascending limb cells may protect the luminal cotransporters from phosphorylation and inactivation by NO, whereas expression in the interstitium, VSMC, and endothelial cells limits the access of NO generated in the MD or the endothelial cells to GC expressed in the VSMCs of the adjacent afferent arteriole (Fig. 58-2). Thus, the expression and activation of NADPH oxidase are important factors that limit the biological actions of NO in the juxtaglomerular apparatus. Regulation of NOS Expression A low salt intake increases the expression of nNOS in the MD and renal cortex, but reduces its expression in the cortical collecting ducts. Despite its upregulation during salt restriction, the expression of nNOS in the MD is unresponsive to angiotensin (Ang) II infusion or to blockade of Ang II type-1 receptors (AT1-Rs). The upregulation of nNOS may be a response to reduced bioactive NO, because NO inhibits NOS activity and expression. The expression of nNOS in the MD is downregulated in the reduced renal mass model of chronic renal insufficiency, but is upregulated in several models of hypertension, including the poststenotic kidney of rats with Goldblatt hypertension and the SHR. However, oxidative stress that accompanies hypertension in these models enhances ONOO– deposition and prevents the action of NO to vasodilate the afferent arteriole.

609

In contrast, nNOS is upregulated in the MD in models of early insulinopenic diabetes mellitus, where it mediates glomerular hypertension and hyperfiltration via blunting of the vasoconstrictive tubuloglomerular feedback response. Expression of eNOS in endothelial cells of the afferent arteriole also is upregulated by salt restriction but, unlike nNOS in the MD, this can be ascribed to Ang II action on AT1-R, because it is prevented by blockade of AT1-R with losartan. eNOS expression is upregulated in models of early insulinopenic diabetes mellitus and in some models of hypertension, such as the SHR. Dietary salt restriction upregulates nNOS expression in the MD. However, its activity in the regulation of afferent arteriolar tone is absent, but can be restored by microperfusion of the NOS substrate, L-arginine, into the MD. The limitation of NOS activity in the MD during salt restriction by arginine availability can be ascribed to a reduced plasma arginine concentration and a reduced cellular uptake of L-arginine from the tubular lumen. Hepatic cells metabolize available arginine to ornithine and urea and thereby limit plasma levels of arginine. During dietary salt restriction there is an upregulation of the high-capacity CAT transporter for arginine expressed on hepatic cells. This may enhance hepatic arginine uptake from the portal system sufficiently to facilitate its metabolism and restrict its plasma level. The reduced plasma levels of arginine are accompanied by increases in ornithine and in urea appearance. Dietary salt restriction reduces the expression of CAT-2A in the thick ascending limb and MD cells of the loop of Henle. This restricts the uptake of arginine from the tubular fluid into these cells and contributes to the reduced activity of nNOS during salt restriction. These adaptations may be important because reduced NOS activity could contribute to the reduced renal blood flow and the enhanced renal NaCl reabsorption during salt deficiency. Despite overexpression of nNOS in the MD, and of eNOS in the endothelium, the renal blood flow, and the tubuloglomerular feedback of the SHR are independent of NOS, as indicated by little or no response to blockade of NOS with drugs such as L-nitroarginine methyl ester. NOS activity in the juxtaglomerular apparatus of the SHR is not limited by the availability of arginine or BH4 but can be restored in full by microperfusion into the juxtaglomerular apparatus of the antioxidant SOD mimetic, tempol. The renal oxidative stress in the SHR and the Ang II infused rat or mouse has been attributed to an upregulation of NADPH-oxidase that is mediated by AT1-R. Ang II upregulates the p22phox and Nox-1 components of NADPH oxidase in the kidney, but downregulates Nox-4 and extracellular and mitochondrial-SOD. Ang II type-2 receptors have contrary effects and therefore are protective of the development of oxidative stress. Renovascular Actions of NO The renal preglomerular (afferent) and postglomerular (efferent) arterioles are vasodilated tonically by NO, as indicated by a sharp increase in their resistances during blockade of NOS. An enhanced renal NO generation contributes to renal vasodilatation in normal pregnancy, during infusion of amino acids, and in early diabetes mellitus. Conversely, a reduced NO bioactivity contributes to renal vasoconstriction in many models of hypertension, such as prolonged Ang II infusion in the rat, the SHR, lead-induced hypertension, and the Dahl saltsensitive rat and in models of chronic renal failure. There is a reduced total body NO generation in humans with essential hypertension and chronic renal failure. The NO that dilates the renal afferent arteriole may derive either from the adjacent vascular endothelial cell or the adjacent MD cell (Fig. 58-2). Despite these

610

SECTION VII / NEPHROLOGY

widespread effects of NO to regulate afferent arteriole tone, NO is not implicated in renal autoregulation. Role of NO in Tubular Function Most investigators have concluded that NO reduces tubular NaCl and fluid reabsorption. However, this is often complicated by secondary effects. For example, the increase in blood pressure (BP) that accompanies NOS inhibition systems reduces tubular NaCl reabsorption. Effects of NO on segmental NaCl tubular reabsorption are controversial. Both stimulation and inhibition have been reported in the proximal tubule and collecting ducts. Blockade of NOS in the proximal tubule can impair luminal Na+/H+ exchange leading to impaired reabsorption of Na+ and HCO3– and development of a metabolic acidosis contrast. Stimulation of NO generation from eNOS in the adjacent peritubular capillary endothelium by Bk enhances basolateral HCO3– uptake. Higher concentrations of NO in the proximal tubule inhibit Na+/K+ ATPase. In the collecting ducts, NO, via cGMP, inhibits the luminal entry of Na+ via the epithelial sodium channel, but also paradoxically can enhance Na+ entry secondary to an increase in the basolateral potassium conductance that hyperpolarizes the cell. Studies in the thick ascending limb of the loop of Henle uniformally demonstrate inhibition owing to blockade by NO of the luminal Na+/K+/2Cl– cotransporter, Na+/H+ exchanger, and K+ channel and the basolateral Na+/K+ ATPase. NO increases cGMP and cAMP, which promote phosphorylation and inactivation of the luminal transporters (Fig. 58-2). NO likely derived from nNOS within mitochondria in tubular epithelial cells competes with O2 for cytochrome-c reductase and diminishes cellular O2 usage. A reduction in NO in the kidney increases O2 usage and reduces the efficiency with which the kidney uses O2 for tubular Na+ reabsorption. Consequently, there is a fall in the pO2 of the kidney cortex. Effects of NO on Renin Secretion NO has site- and timedependent effects on the secretion of renin from the afferent arteriole. NO can stimulate renin secretion from the juxtaglomerular cells via inhibition of PDE-III and elevation of cAMP, or may inhibit it via the cGMP/PKG pathway. Activation of nNOS in the MD by luminal perfusion of L-arginine enhances the secretion of renin. This may represent the effects of MD NO to activate COX-2 whose product, prostacyclin (PGI2), stimulates renin secretion. In contrast, acute activation of eNOS in endothelial cells leads to cGMP-dependent inhibition of renin secretion. However, during prolonged stimulation of eNOS, the initial inhibition is followed by enhanced release. Role of NO in Pathophysiology and Disease A decrease in intrarenal NO activity in blood vessels, the MD and renal tubules accompanies many models of hypertension and contributes to renal vasoconstriction and enhanced tubular NaCl reabsorption. Inhibition of NOS in normal animals shifts the relationship between BP and salt so as to induce salt sensitive hypertension. In contrast, correction of oxidative stress in hypertensive models with an SOD mimetic that enhances NO bioactivity shifts the relationship back to the normal state, thereby correcting salt sensitivity and lowering BP. An increase in NO in the kidney contributes to hyperfiltration in models of early diabetes mellitus. In models of postischemic or endotoxic renal insufficiency, where oxidative stress is prominent, massive overproduction of NO in the kidney is implicated in the associated tubular damage that may be mediated by ONOO–.

CYCLOOXYGENASE COX is a fatty acid mono-oxygenase. It is widely distributed in the kidney and elsewhere. On cell activation by increased sheer

stress, transmural pressure, peptides such as Ang II, Bk, or endothelin, cytokines or hypoxia, phospholipase A2 hydrolyzes membrane phospholipids to free arachidonic acid that is metabolized by COX to prostaglandin (PG) endoperoxides. The initial COX product is prostaglandin G2 but this is rapidly rearranged to the stable prostaglandin H2 (PGH2), which not only activates thromboxane-prostanoid receptors, but also is the substrate for three important classes of enzymes. Prostaglandin E2 synthase generates prostaglandin E2 (PGE2), PGI2-S generates PGI2, and thromboxane A2 synthase generates TxA2. COX ISOFORMS IN THE KIDNEY Two COX isoforms are expressed in the kidney. COX-1 is widely expressed constitutively in body tissues. In most organs, COX-1 is responsible for basal PG synthesis. COX-2 is termed inducible because in most organs there is negligible basal COX-2 expression until stimulation with lipopolysaccharide or cytokines when COX-2 becomes the predominant source of PGs. However, as with NOS, the kidney is unusual in having high basal expression of the inducible isoform, which contributes importantly to basal renal PG productions. COX-1 is located in the endoplasmic reticulum of the endothelial and VSMCs of the afferent and efferent arterioles and larger arteries, and in the glomerulus, the medullary collecting ducts, and medullary interstitial cells. Major sites of COX-2 expression include the MD cells, some adjacent thick ascending limb cells and medullary interstitial cells. COX-2 is expressed more widely in the embryonic kidney. Mice with targeted disruption of the gene for COX-2, but not COX-1, suffer nephron hypoplasia and renal atrophy, implying an important role for COX-2 products in renal development. REGULATION OF COX ACTIVITY COX activity is enhanced by peroxides, but high concentration of O2⋅– leads to its irreversible inactivation. In low concentrations, NO stimulates COX activity and vascular PGI2 generation. However, in high concentrations, as during iNOS induction, NO can inactivate COX. COX-1 activity is determined by the availability of arachidonate. The net effect of inhibition of COX-1-derived PGs in the blood vessels is usually vasoconstriction, likely owing primarily to inhibition of PGI2 synthesis in resistance vessels. COX-1 activity in the collecting ducts is increased by arginine vasopressin (AVP). An ensuing increase in PGE2 activates adenylate cyclase to generate cAMP which antagonizes the effects of AVP to enhance water permeability and terminates other effects of AVP. PGE2 also inhibits the activity of the luminal Na+ entry channel. Thus, blockade of COX-1 enhances tubular reabsorption of fluid and NaCl. COX-2 activity is determined by enzyme expression, which is stimulated by serum, growth factors, lipopolysaccharide, tumor promoters, and peptides that include Ang II, Bk, and endothelin, whereas glucocorticoids inhibit expression. COX-2 expression in the MD is stimulated by dietary salt restriction. Remarkably, this effect is enhanced during blockade of AT1-R which indicates that Ang II has a negative feedback effect on COX-2 expression at this site. In contrast, Ang II stimulates expression of COX-2 in the thick ascending limb and blood vessels. A reduction in extracellular [Cl–] at the thick ascending limb stimulates COX-2 expression by increasing p38 mitogen-activated protein kinase activity. Negative feedback inhibition of COX-2 expression in the MD by Ang II is signaled via nuclear transcription factor-κB and is reinforced by Ang II-induced aldosterone secretion, which also inhibits COX-2 expression. Expression of COX-2 in the interstitial cells of the renal medulla is enhanced by dehydration or an increase in interstitial osmolality. The ensuing increase in PGE2

CHAPTER 58 / NITRIC OXIDE SYNTHASE AND CYCLOOXYGENASE

611

Figure 58-3 Cell diagram showing some pathways whereby nNOS and COX-2 in the MD interact to regulate afferent arteriolar tone. For details, see Interactions Between NOS and COX in the Kidney.

buffers the effects of AVP to enhance free water reabsorption in the collecting ducts. COX-2 expression in the MD is reduced by blockade of NOS and is increased by NO donors. However, a clear picture has not emerged of the inter-relationship of MD nNOS and COX-2 expression because the regulation of COX-2 by salt intake is relatively intact in the nNOS knockout mouse. Renovascular Effects of COX Products The regulation of renal vascular resistance by COX-1 is of acknowledged importance, but is complicated by opposite effects of the vasoconstrictor PGs, PGH2, and TxA2 that activate the thromboxane-prostanoid receptors and of the vasodilator PG, PGI2 that activates the PGI2-R. Moreover, the actions of PGE2 are transduced via four discrete receptors linked to vasodilator or vasoconstrictor pathways. The response to nonselective inhibition of COX generally is to reduce renal blood flow and the glomerular filtration rate (GFR) and to raise BP. This likely reflects predominantly inhibition of PGI2 synthesis by COX-2 in the renal and systemic resistance vessels. During short-term Ang II infusion, the increased renal cortical and medullary vascular resistances are enhanced during blockade of COX-1, but are blunted during blockade of COX-2. These acute responses to Ang II likely represent phospholipase-A2 activation with arachidonic acid release and metabolism to predominantly vasodilator PGI2 by COX-2 in the vascular endothelium. However, nonselective inhibition of COX during Cl– loading to stimulate tubuloglomerular feedback, in models of renovascular hypertension, reduced renal mass hypertension and during prolonged infusion of Ang II, increases the renal blood flow and GFR and/or reduces the BP. These contrary consequences likely involve predominant effects of COX-2 blockade to inhibit PGI2- and PGE2-induced renin secretion, and to inhibit the generation of PGH2 and TxA2 that mediate vasoconstriction and enhanced

tubuloglomerular feedback mediated via the thromboxaneprostanoid receptors. Importantly, nonselective inhibition of COX or selective inhibition of COX-2 either does not change, or more often increases the renal vascular resistance and the BP in normal human subjects or those with essential hypertension. As in animal models, nonselective inhibition of COX reduces BP in patients with renovascular hypertension. Effects of COX Products on Tubular Reabsorption The predominant effect of nonselective inhibition of COX is a reduction in NaCl and fluid excretion and an increase in urine concentration. This likely reflects predominantly COX blockade in the collecting ducts that prevents the buffering by PGE2 of AVPinduced increases in water permeability and Na+ reabsorption. Effects of COX Products on Renin Secretion Nonselective blockade of COX inhibits renin secretion profoundly. This likely represents predominant inhibition of COX-2-derived PGs that mediate MD-stimulated renin secretion. Indeed, COX-2 is upregulated by dietary NaCl restriction and in models of renal artery stenosis and mediates the accompanying increases in renin secretion. Role of COX in Disease COX products have relatively little effect on renal function during normal salt intake. In contrast, nonselective COX inhibition leads to renal vasoconstriction and reduced GFR during dietary salt restriction, and to renal fluid and salt retention with edema or hypertension during dietary salt loading. The latter effect likely involves predominantly inhibition of COX-1-mediated reductions in NaCl and fluid transport in the collecting ducts. However, inhibition of COX-2 can increase BP in individuals with essential hypertension. In contrast, aspirin given to patients with renovascular hypertension reduces BP and renin secretion. Overproduction of COX products contributes to renin

612

SECTION VII / NEPHROLOGY

secretion in patients with Bartter’s syndrome. Nonselective inhibition of COX in human subjects or selective inhibition of COX-2 in animals impairs the natriuretic and diuretic response to loop diuretics. Overproduction of COX products in the glomerulus during glomerulonephritis, or the nephrotic syndrome, contributes to increased glomerular protein permeability and proteinuria. Enhanced generation of medullary COX products during severe potassium depletion or hypercalciuria leads to polyuria, dehydration, and failure to concentrate the urine, likely reflecting enhanced buffering of the effects of AVP in the collecting ducts by PGE2. Following ureteral obstruction in the rat, there is initially overproduction of vasodilatory PGs that maintain the renal blood flow. However, during the subsequent days, cellular infiltration generates PGs with predominant vasoconstrictor action mediated through the thromboxane-prostanoid receptors. Renal vein constriction enhances renal PG synthesis that maintains renal blood flow. COX products contribute to the hyperfiltration in models of early diabetes mellitus, but during the established phase with the development of glomerulopathy, the predominant COX products are TxA2 and PGH2, which mediate protein kinase C-induced renal vasoconstriction, proteinuria, glomerular sclerosis, and renal insufficiency via the thromboxane-prostanoid receptors. Overall, COX products generally are not unique mediators of pathophysiology, but modulate the response to the disease process.

INTERACTIONS BETWEEN NOS AND COX IN THE KIDNEY Although interaction between NOS and COX is widely acknowledged, its details remain controversial. Low levels of NO directly activate COX, enhance vascular PGI2 generation, and transcriptionally upregulate COX-2 in the MD. In contrast, high levels of NO contribute to irreversible inactivation of COX. Within the MD cells (Fig. 58-3) an increase in NOS-1-derived NO can activate COX-2 to generate PGs that contribute to renin secretion from the afferent arteriole and modulation of the tubuloglomerular feedback response. Because both inducible NOS (NOS-2) and COX-2 are highly inducible, these two isoforms are widely expressed in the tissues and kidneys during postischemic, nephrotoxic and shock syndromes, where they contribute to hypotension and renal insufficiency. The coincident production of large quantities of NO and O2⋅– will lead to widespread protein nitrosylation secondary to ONOO– formation, which likely contributes to structural and functional abnormality. Blockade of NOS-2 in a model of renal ischemia limits tubular damage whereas blockade of COX-2 in a model of chronic renal failure limits proteinuria and progressive renal damage. However, any beneficial effects of blockade of COX-2 must be weighed closely against the evidence that nonselective, or COX-2 selective, antagonists have been linked to a wide range of renal insults, including acute renal insufficiency, membranous glomerulonephritis, interstitial nephritis, and papillary necrosis. Therefore, these drugs are generally contraindicated in patients predisposed to these conditions.

CONCLUSION NOS and COX fulfill many critical roles in the kidney in adapting normal renal function to changes in salt intake or BP. Although they can function individually, increasing emphasis is placed on interaction between these two systems. The outcome of NOS or

COX activation is complex because it depends on isoforms and their expression within discrete nephron segments and, in the case of COX, can lead to products with widely divergent actions.

ACKNOWLEDGMENTS Christopher S. Wilcox, MD, PhD, was supported by grants from the NIDDK (Grant Nos. DK-36079 and DK-49870) and the NHLBI (Grant No. HL HL68686-01) and by funds from the George E. Schreiner Chair of Nephrology. Tina Chabrashvili, MD, PhD, provided valuable advice on the preparation of the chapter, which was prepared by Sharon Clements.

SELECTED REFERENCES Agmon Y, Brezis M. NO and the medullary circulation. In: Goligorsky MS, Gross SS, eds. Nitric oxide and the kidney: physiology and pathophysiology. New York: Chapman & Hall, 1997:250–270. Breyer RM, Bagdassarian CK, Myers SA, Breyer MD. Prostanoid receptors: subtypes and signaling. Annu Rev Pharmacol Toxicol 2001;41:661–690. Breyer MD, Breyer RM. G protein-coupled prostanoid receptors and the kidney. Annu Rev Physiol 2001;63:579–605. Breyer MD, Harris RC. Cyclooxygenase 2 and the kidney. Curr Opin Nephrol Hypertens 2001;10:89–98. Breyer MD, Hao C, Qi Z. Cyclooxygenase-2 selective inhibitors and the kidney. Curr Opin Crit Care 2001;7:393–400. Ferreri NR, An SJ, McGiff JC. Cyclooxygenase-2 expression and function in the medullary thick ascending limb. Am J Physiol 1999;277 F360–F368. FitzGerald GA, Patrono C. The coxibs, selective inhibitors of cyclooxygenase-2. N Engl J Med 2001;345:433–442. Harris RC, Breyer MD. Physiological regulation of cycooxygenase-2 in the kidney. Am J Physiol Renal Physiol 2001;281:F1–F11. Imig JD, Kitiyakara C, Wilcox CS. Arachidonate metabolites. In: Seldin DW, Giebisch G, eds. The Kidney: Physiology and Pathophysiology. Raven, New York, 2000, pp. 875–889. Komers R, Anderson S, Epstein M. Renal and cardiovascular effects of selective cyclooxygenase-2 inhibitors. Am J Kidney Dis 2002;38:1145–1157. Kone BC, Baylis C. Biosynthesis and homeostatic roles of nitric oxide in the normal kidney. Am J Physiol 1997;272:F561–F578. Kurtz A, Wagner C. Role of nitric oxide in the control of renin secretion. Am J Physiol 1998;275:F849–F862. Manjunath G, Tighiouart H, Coresh J, et al. Level of kidney function as a risk factor for cardiovascular outcomes in the elderly. Kidney Int 2003; 63:1121–1129. Mardini IA, FitzGerald GA. Selective inhibitors of cyclooxygenase-2. A growing class of anti-inflammatory drugs. Mol Interv 2001;1: 30–38. Ortiz PA, Garvin JL. Role of nitric oxice in the regulation of nephron transport. Am J Physiol Renal Physiol 2002;282:F777–F784. Pallone TL, Mattson DL. Role of nitric oxide in regulation of the renal medulla in normal and hypertensive kidneys. Curr Opin Nephrol Hypertens 2002;11:93–98. Pfeilschifter J, Muhl H. NOS in mesangial cells: physiological and pathophysiological roles. In: Goligorsky MS, Gross SS, eds. Nitric Oxide and the Kidney: Physiology and Pathophysiology. Chapman & Hall, New York, 1997, pp. 198–215. Schnermann J. Cyclooxygenase-2 and macula densa control of renin secretion. Nephrol Dial Transplant 2001;16:1735–1738. Schnermann J, Briggs JP. Role of NO in the function of the juxtaglomerular apparatus. In: Goligorsky MS, Gross SS, eds. Nitric Oxide and the Kidney: Physiology and Pathophysiology. New York: Chapman & Hall, 1997:176–197. Welch WJ, Wilcox CS. Tubuloglomerular feedback and macula densaderived NO. In: Goligorsky MS, Gross SS, eds. Nitric Oxide and the Kidney: Physiology and Pathophysiology. Chapman & Hall, New York, 1997, pp. 216–232. Welch WJ, Wilcox CS. What is brain nitric oxide synthase doing in the kidney? Curr Opin Nephrol Hypertens 2002;11:109–115. Wilcox CS. L-arginine-NO pathway. In: Seldin DW, Giebisch G, eds. The Kidney: Physiology and Pathophysiology. Raven, New York, 2000, pp. 849–871.

59 Hypertension and Sodium Channel Turnover

DOUGLAS C. EATON, BELA MALIK, AND HE-PING MA SUMMARY Hypertension is an increase of blood pressure to levels greater than normal that arises because of a mismatch between the volume of the vascular tree and the volume of blood. Blood volume depends on total body sodium content, which is a balance between sodium intake and output. Total body sodium is controlled by variable excretion of sodium by the kidneys. To regulate sodium balance, the primary variable that the kidney monitors is not total body sodium, but rather systemic blood pressure. Renal regulation of blood pressure is via the release of the peptide hormone, renin from specialized renal cells. Release of renin ultimately leads to the production of angiotensin II. Angiotensin II increases total peripheral resistance and blood pressure and also leads to an increase in aldosterone. Aldosterone is a steroid hormone that increases sodium reabsorption in the distal nephron by activating epithelial Na channels (ENaCs). Thus, Hypertension is a defect in one of these elements that control total body sodium balance. Key Words: Aldosterone; angiotensin II; degradation; hypertension; inositol lipids; KRas; “Liddle’s” Syndrome; Nedd4; small G proteins; “ubiquitin” salt sensitive hypertension.

INTRODUCTION Hypertension is a chronic increase of blood pressure to levels >140/90 mm Hg. Between 1988 and 1994, 23% of all Americans aged 20–74 were found to have hypertension. Hypertension is more common among African Americans; in the United States, nearly one in three African Americans has hypertension compared with one in four whites. African Americans develop high blood pressure earlier in life compared with whites and at every age have more severe consequences of high blood pressure. The estimated prevalence of high blood pressure for non-Hispanic, African Americans age 20 and older is 35% for men and 34.2% for women. For whites, the estimated prevalence of high blood pressure for men is 24.4 and 22% for women. Hypertension predisposes individuals to develop cardiovascular disease, renal disease, and cerebral vascular injury. According to the Centers for Disease Control and Prevention stroke alone killed 158,448 people (61.4% among women) in 1998 and accounted for about 1 of almost 15 deaths in the United States (http://www.cdc.gov/washington/overview/ heartstk.htm). From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge, C. Patterson © Humana Press, Inc., Totowa, NJ

HYPERTENSION AND TOTAL BODY SODIUM BALANCE The average American diet contains about 10 g of salt per day, but the amount of ingested salt can vary enormously. Despite large daily variations in salt intake, normal values of plasma sodium deviate little from the normal range of 135–145 mmol/L and, under normal circumstances, there is a critical balance of excretion that is exactly equal to the intake. There is some sodium in sweat and feces; the amount in sweat is incidental to maintaining thermal balance and the output in feces under physiological circumstances is generally small. Consequently, the control of total body sodium is due to variations in excretion by the kidneys. Urinary sodium concentrations are usually in the range of 10–40 mmol/L but can range from undetectable to >150 mmol/L. Thus, the kidney not only is the primary organ responsible for removing sodium from the body but it also maintains total body sodium balance and a normal blood concentration of sodium. REGULATION OF TOTAL BODY SODIUM BALANCE BY THE KIDNEY The observation that the sodium content of the urine varies directly with the amount of ingested sodium shows that the kidney is responsible for sodium balance, but two questions remain: How does the kidney sense total body sodium and what are the mechanisms by which the kidney can adjust the urinary output of sodium to match ingested sodium? Surprisingly, the primary variable that the kidney monitors in regulating sodium balance is not total body sodium, but rather systemic blood pressure. Arterial blood pressure (sensed by carotid and aortic arch baroreceptors) and venous blood pressure (or volume) sensed by distension of atria alter sympathetic activity (decreased blood pressure/or extracellular volume increases sympathetic activity). This in turn, varies blood pressure in the afferent arteriole supplying the nephron, the primary functional unit of the kidney (Fig. 59-1), and is sensed by pressure-sensitive cells within the nephron. Changes at any of these sites are interpreted as a change in total body sodium because blood pressure, blood volume, and total body sodium are closely linked. Thus, regulation of total body sodium balance by the kidney is intimately related to and ultimately depends on renal regulation of blood pressure. RENAL REGULATION OF SYSTEMIC BLOOD PRESSURE The primary site in the nephron that responds to both sympathetic activity and systemic blood pressure is the juxtaglomerular apparatus

613

614

SECTION VII / NEPHROLOGY

Figure 59-1 A schematic diagram of the nephron, the functional unit of the kidney. The nephron consists of the renal tubule and its associated blood supply. Blood entering the kidney from the renal artery (on left) enters the glomerulus through the afferent arteriole. The glomerulus acts as a filter that delivers about 20% of the blood volume to the proximal nephron as a cell-free, ultrafiltrate of plasma. Most of the glomerular filtrate is reabsorbed along the nephron so that only approx 1% of the original filtered amount remains to be excreted; however, the primary discretionary control of the amount of sodium present in the final urine is by control of sodium reabsorption in the distal nephron particularly aldosterone-mediated sodium reabsorption by the collecting tubule. The juxtaglomerular apparatus is the primary site at which the kidney senses systemic blood pressure and is the site for production of renin.

(see Fig. 59-1). Renal regulation of blood pressure primarily centers around release of the peptide hormone, renin, from specialized granular cells within the juxtaglomerular apparatus. When sympathetic activity increases owing to a fall in systemic blood pressure or if arterial pressure in the kidney decreases, the granular cells secrete renin. Release of renin initiates a complicated series of biochemical events. Renin is a proteolytic enzyme that splits circulating angiotensinogen (produced in liver) to form angiotensin I. Angiotensin-converting enzyme (ACE) present in endothelial cells of capillaries (especially in the pulmonary vasculature) cleaves angiotensin I to produce angiotensin II. Angiotensin II is a potent vasoconstrictor and increases total peripheral resistance and blood pressure. All of these events occur rapidly in response to a decrease in blood pressure. These responses represent a kidney-mediated, short-term correction for a decrease in blood pressure. ALDOSTERONE REGULATION OF SODIUM BALANCE AND MEAN SYSTEMIC BLOOD PRESSURE In addition to this potential for short-term correction of blood pressure by angiotensin II, a sustained decrease in blood pressure is often associated with a decrease in blood volume suggesting a decrease in both total body sodium and water. The correction of a sustained decrease in blood pressure requires a decrease in excretion with

increased renal reabsorption of salt and water until the blood volume returns to normal. Part of this type of correction is a direct response to changes in blood pressure; a reduction in blood pressure leads to a lower volume of blood filtered by the glomerulus. This leads to less sodium filtered, an increase in the reabsorption of filtered sodium, and less Na excretion. Primary control of total body sodium is due to the action of the hormone, aldosterone. Aldosterone is a steroid hormone produced by adrenal cortical cells. Its primary effect is to increase sodium reabsorption in the distal nephron, particularly the collecting tubule (Figs. 59-1 and 59-2). The most important physiological factor controlling circulating levels of aldosterone is the circulating level of angiotensin II. As noted earlier, the kidney responds to a decrease in blood pressure by producing renin, a hormone that produces an initial short-term correction of blood pressure by promoting production of a vasoconstrictive agent, angiotensin II. Angiotensin II also stimulates the adrenal cortex to produce aldosterone. Aldosterone targets the distal nephron to increase sodium reabsorption and, thereby, increase total body sodium and blood volume producing a long-term correction of low blood pressure. HYPERTENSION: A DISEASE OF ABNORMAL SODIUM BALANCE Hypertension is almost always associated with a blood volume and total body sodium content that are too high for

CHAPTER 59 / HYPERTENSION AND SODIUM CHANNEL TURNOVER

615

Figure 59-2 Mechanism of Aldosterone’s Action and ENaC structure. (A) Aldosterone mechanism and pathology. Aldosterone enters principal cells (A) and interacts with cytosolic aldosterone receptors. The aldosterone-bound receptors interact with nuclear DNA to promote gene expression. The aldosterone-induced gene products activate sodium channels (B) and sodium pumps to increase sodium reabsorption. (A) Abnormalities of the adrenals and certain tumors can lead to primary hyper- or hypo-aldosteronism. (B) Loss of function mutations in sodium channels can lead to both pseudophypoaldosteronism; gain of function mutations produce pseudohyperaldosteronism (Liddle’s syndrome). (C) Defects in 11-β-hydroxysteroid dehydrogenase (C) produces Apparent Mineralocorticoid Excess because glucocorticoids can then interact with aldosterone receptors to stimulate sodium channels. (B) Structure of ENaC. Each subunit has only two predicted membrane spanning domains, an unusual topology for an ion channel. Topologic analyses suggest that the amino and carboxy terminal regions of the ENaCs are cytoplasmic, and that each subunit has a large extracellular loop present among the membrane spanning regions. The α-, β-, and γ-ENaC have cysteine rich regions within putative extracellular loops, which are postulated to form intersubunit disulfide bonds that are important in oligomerization of Na channels. The three ENaCs also possess proline rich regions within their intracellular C-terminal domains that are important for interaction with accessory proteins that regulate the stability of functional channels in the apical membrane of renal cells. The three homologous ENaC subunit proteins form a heteromultimeric channel protein apparently made up of two α-, one β-, and one γ-subunit.

the physiological condition. In some cases, the reason for the excess blood volume is clear. For example, glomerular diseases often lead to inappropriate release of renin and an increase in angiotensin II and aldosterone levels with increases in collecting tubule sodium reabsorption, and ultimately, an increase in blood pressure. A tumor of the adrenal cortex can produce excess aldosterone causing an increase in blood pressure. A specific gain of function mutation in the sodium reabsorptive mechanism in the collecting duct also leads to excess sodium reabsorption and profound hypertension. Three examples illustrate types of defects that can complicate the control mechanism for maintaining total body sodium and mean blood pressure. First, excess renin production is a problem with the sensing mechanism for the system. Second, excess production of aldosterone is a defect in the signaling mechanism that lies between the sensor (pressure sensors in the large vessels and the kidney) and the effector (distal nephron sodium reabsorption); and the third example is a defect in the effector system (distal nephron sodium reabsorption). In each case, the defect is easily identified and correcting the underlying pathology usually corrects the hypertension. Table 59-1 provides a list of identifiable genetic defects in the regulation of sodium balance that lead to increases or decreases in blood pressure. Despite the relatively extensive list of known genetic defects, the etiology of hypertension is unclear. In most cases, renin angiotensin and aldosterone levels are normal or even reduced

and yet blood pressure is elevated as if the set-point of the control loop sensing blood pressure and varying sodium reabsorption is inexplicably high. The relatively normal levels of circulating renin, angiotensin II, and aldosterone imply that the defect in the regulation of sodium reabsorption lies beyond aldosterone interacting with the cells of the collecting tubule. Hypertension research has attempted to define the components controlling sodium reabsorption in response to aldosterone.

MOLECULAR BIOLOGY AND STRUCTURE OF EPITHELIAL NA CHANNELS cDNAs for epithelial Na channels (ENaCs) were originally described by investigators who used expression cloning methods to isolate three separate cDNAs from rat distal colon. They designated the three cDNAs α-, β-, and γ-rat ENaC. Coinjection into Xenopus oocytes with cRNAs prepared from all three cDNAs induces expression of amiloride-sensitive Na channels that possess characteristics virtually identical to those observed in renal cortical collecting tubule principal cells. The three subunits show limited sequence homology to each other although, across species, the same subunit has significant (70–90%), amino acid sequence homology (some regions do have only limited homology). Each subunit has only two predicted membrane spanning domains, an unusual topology for an ion channel. Topologic analyses suggest that the amino and carboxy-terminal regions of the ENaCs are cytoplasmic,

616

SECTION VII / NEPHROLOGY

Table 59-1 Genetic Bases for Changes in Blood Pressure Genetic bases for hypertension Syndrome

Defect

Apparent mineralocorticoid excess

Loss of function mutation of 11-β-hydroxysteroid dehydrogenase

Liddle’s

Gain of function mutation in the β- or γ-subunit of ENaC

T594M β–ENaC

Single amino acid mutation in β-subunit of ENaC with moderate gain of function

Glucocorticoid-remediable aldosteronism

Elevated aldosterone owing to an abnormal activation of aldosterone synthase by ACTH Gain of function leading to increased circulating aldosterone

Aldosterone synthase polymorphisms

11-β-hydroxylase deficiency

Loss of function mutation leading to abnormal accumulation of 11-deoxycorticosterone that activates aldosterone receptors

Pseudohypoaldosteronism Type II (Gordon’s syndrome)

Loss of function mutation in With No K [lysine] kinases

α-Adducin polymorphism

Structural protein that activates Na K-ATPase Mutation in mineralocorticoid receptor that makes receptor more sensitive to aldosterone

Hypertension exacerbated in pregnancy

SGK

Two activating mutations

Various single nucleotide polymorphisms in the genes for renin, ACE, and angiotensin II receptor

Presumably slightly altered function of the genes controlling angiotensin production and sensitivity

Characteristics Aldosterone receptors in renal principal cells are not only activated by aldosterone, but can be activated by normal levels of circulating glucocorticoids promoting constitutive sodium reabsorption with profound early-onset hypertension and growth arrest. Responsive to mineralocorticoid (aldosterone) receptor antagonists, for example, spironolactone, eplerenone. Constitutive sodium reabsorption regardless of circulating aldosterone levels with profound earlyonset hypertension Responsive to Midamor (amiloride HCl). Increased collecting duct sodium reabsorption with mild to moderate hypertension. Most prevalent in individuals of African descent. Responsive to Midamor (amiloride HCl). Early-onset hypertension that can be controlled by administration of glucocorticoids (to reduce ACTH production). Increased sodium reabsorption with mild hypertension. Most prevalent in men. More prevalent in individuals of Japanese descent. Responsive to mineralocorticoid (aldosterone) receptor antagonists, for example, spironolactone, eplerenone. Activation of aldosterone receptors leads to sodium retention and early-onset hypertension. Associated accumulation of androgens leads to masculinization of females in utero and both sexes postnatally. May be responsive to the aldosterone receptor antagonist, eplerenone, but spironolactone would exacerbate androgenic effects. Loss of With No K [lysine] kinase (WNK) function leads to activation of distal nephron thiazide-sensitive sodium chloride cotransporter causing sodium retention, hypertension, and hyperkalemia. Responsive to thiazides. Increased Na pump activity modestly increases sodium reabsorption to produce mild hypertension. Early-onset hypertension greatly exaggerated by pregnancy because progesterone also binds to the mutant receptor and promotes additional sodium reabsorption. The aldosterone receptor antagonist, spironolactone, acts as an agonist for the mutant receptor; effect of eplerenone has not been examined. SGK inhibits ENaC degradation so activation causes accumulation of ENaC and an increase in renal sodium reabsorption with consequent hypertension. Possibly responsive to Midamor (amiloride HCl). Statistically significant increases in blood pressure. Some single nucleotide polymorphisms are more prevalent. in African Americans, others more prevalent in European Americans, and a few prevalent in both populations. Responsive to ACE inhibitors (e.g., captopril, and lotensin) and angiotensin II receptor inhibitors (e.g., losartan), and volume reduction with thiazide diuretics. (Continued)

CHAPTER 59 / HYPERTENSION AND SODIUM CHANNEL TURNOVER

617

Table 59-1 (Continued) Genetic bases for hypotension Syndrome

Defect

Characteristics

Pseudohypoaldosteronism type I (dominant) Pseudohypoaldosteronism type I (recessive)

Loss of function mutation of the mineralocorticoid receptor Partial loss of function mutation of the α-subunit of ENaC

Aldosterone synthase

Loss of function leading to decreased circulating aldosterone

Steroid 21-hydroxylase

Loss of function leading to decreased synthesis of aldosterone Loss of function of distal nephron thiazide-sensitive NaCl transporter

Gitelman’s syndrome

Salt wasting, hypotension, severe neonatal symptoms. Salt wasting, hypotension, hyperkalemia, excess fluid secretion from airways, loss of salt taste, serious symptoms at all ages. Decreased sodium reabsorption with severe hypotension and occasionally shock because of reduced intravascular volume. Decreased sodium reabsorption with severe hypotension and other endocrine abnormalities. Salt wasting and mild hypotension in homozygotes.

ACTH, adrenocorticotropin hormone.

and indicate that each subunit has a large extracellular loop present between the two membrane spanning regions. The α-, β-, and γENaC have cysteine rich regions within putative extracellular loops that are postulated to form intersubunit disulfide bonds necessary for oligomerization of Na channels. The three ENaCs also possess proline-rich regions within their intracellular C-terminal domains that are important for interactions between the subunits and accessory proteins that regulate the stability of functional channels in the apical membrane of kidney cells.

CELLULAR REGULATION OF ENaC Sodium transport across the Na+ reabsorbing, tight epithelia such as the distal nephron or colon is the major factor determining total body Na+ content and thus, long-term blood pressure regulation. Sodium reabsorption in the distal nephron is a two step process (Fig. 59-2). First, sodium enters renal cells from the tubular lumen through ENaCs that are positioned in the apical membrane. Sodium is then actively transported out of the cell by Na K-ATPase on the basolateral membrane. ENaC ACTIVITY IS IMPORTANT FOR CONTROLLING BLOOD SODIUM LEVELS AND MEAN BLOOD PRESSURE ENaC is usually the rate-limiting step for Na+ transport; and so represents the primary point for control of total body sodium balance and mean blood pressure. In fact, abnormalities in ENaC function can lead to disorders of total body Na+ homeostasis, blood volume, blood pressure, and lung clearance (e.g., Liddle’s syndrome and Pseudohypoaldosteronism Type I). An increase in ENaC activity or gain of function such as found in Liddle’s syndrome leads to excessive salt reabsorption by the renal distal tubule, body fluid retention, and increased blood pressure. It has been suggested that in some African Americans with salt-sensitive hypertension, excessive sodium reabsorption is associated with ENaC hyperactivity. The clinical profile of these subjects includes low circulating levels of renin and angiotensin II and is consistent with this hypothesis. Also consistent is the observation that patients respond to treatment with diuretics like amiloride or triamterene whereas ACE inhibitors are ineffective. The control of ENaC activity, therefore, is important for understanding the pathophysiology of these disorders and hypertension in general.

REGULATION OF NA CHANNELS BY SMALL G PROTEINS The small G proteins are single subunit proteins ranging in size from approx 20–30 kDa. There are several major categories of small G proteins that can participate in many aspects of cellular function including cell growth and differentiation. The first small G proteins described were the Ras proteins: there are three major categories of Ras proteins, H-, N-, and K-Ras proteins. All Ras proteins are posttranslationally modified by the attachment of lipophilic groups (usually isoprenyl residues) to the C-terminal end plus subsequent methyl esterification of the C-terminus; these modifications are necessary for the biological function of the protein. After they are conjugated to methyl and isoprenyl groups, Ras proteins can be activated by GTP. Interestingly, methylation of isolated patches of membrane causes an increase in the activity of ENaC. This methylation-induced activation of ENaC is augmented simply by raising the cytosolic content of GTP. Thus, methylation of apical membranes excised as cell-free patches leads to an increase in ENaC activity. The response to methylation is augmented by GTP indicating that the small G protein regulates ENaC activity. Evidence shows that one specific Ras protein, K-Ras2A, is a target for methylation in kidney cells and that Ras methylation is induced by aldosterone. K-Ras does contain the consensus amino acid sequence for methylation, a CAAX box (a methylated cysteine residue followed by two aliphatic residues, followed by any residue at the C-terminal end of a protein). The process requires both GTP and methylation before K-Ras activity rises. Inhibition of K-Ras2A expression in kidney cells reduces ENaC activity whereas augmenting K-Ras expression increases ENaC activity. These results indicate that the mechanism by which K-Ras is activated and identification of the signaling molecules that are activated by K-Ras are important steps in the understanding ENaC activity regulation. Activation and Regulation of K-Ras To be active, K-Ras must be methylated and changes in the rate of K-Ras methylation control the amount of membrane-associated K-Ras. Only the methylated, membrane-associated form binds GTP to increase its activity further. Thus, changes in the activity of the enzyme that methylates K-Ras, isoprenyl-cysteine-O-carboxy-methyl-transferase, alter K-Ras-induced ENaC activity. One way of regulating active, membrane-associated methyl transferases is to control the cytosolic

618

SECTION VII / NEPHROLOGY

concentrations of the end product of methylation, S-adenosylhomocysteine (SAH). SAH is produced when the endogenous methyl donor, S-adenosyl-methionine, transfers its methyl group to a target protein. SAH is a potent inhibitor of all methyl transferases. The primary determinant of the SAH concentration in cells is generally not the rate of SAH production by methyl transferases, but rather, the rate of their degradation by another enzyme, S-adenosylL-homocysteine hydrolase or SAH hydrolase. For this reason, the extent of protein methylation can be controlled by changing the activity of SAH hydrolase rather than altering the intrinsic activity of the methyl transferase. Increased activity of the methyl transferase (measured as a rise in Vmax) does occur when there is protein kinase C-mediated phosphorylation of the methyl transferase. In response to aldosterone, there is an increase in the activity of SAH hydrolase ensuring that the activity of the methyl transferase is not inhibited by the accumulation of its enzymatic product, SAH. How the SAH hydrolase is activated in response to aldosterone is unclear, though additional transcription and translation are not necessary. The Mechanism for K-Ras Activation of ENaC Although K-Ras2A is necessary for ENaC activation, the mechanism by which K-Ras activates ENaC is not clear. The simplest mechanism would be a direct interaction of ENaC with K-Ras. It is likely, however, that other signaling elements lie between Ras and ENaC (any such element would have to be closely associated with ENaC in cell-free patches because it is possible to activate ENaC via Ras methylation in such patches). The traditional effector for Ras is Raf kinase and notably, Raf activation could activate the mitogen-activated protein kinase pathway, which in turn can activate other genes. Activation of the mitogen-activated protein kinase pathway, however, actually inhibits ENaC activity so K-Ras activation of Raf kinase does not appear to be responsible for K-Ras-mediated activation of ENaC. Another pathway leading to direct activation of ENaC is K-Ras2A-mediated activation of inositol lipid kinases that produce phosphatidylinositol bisphosphate (PIP2) and phosphatidylinositol trisphosphate (PIP3). For example, Ras proteins are known to activate several lipid kinases including phosphatidylinositol (PI)-3-kinase and PIP-5-kinase. Application of PIP2 or PIP3 to the cytosolic surface of excised cell patches strongly activates ENaC via an increase in the probability of an open channel. Regulation of Na Channels by Inositol Lipids PIP2 and PIP3 are ubiquitous components of eukaryotic membrane phospholipids. The traditional view of phosphoinositide metabolism suggests that PI is sequentially phosphorylated by PI-4-kinase and PI-5-kinase to produce PI-4,5-bisphosphate or PIP2. Once formed, PIP2 can be phosphorylated to produce PI-3,4,5-triphosphate or PIP3 via PI-3-kinase or it can be hydrolyzed by phospholipase C to produce 1,4,5-inositol trisphosphate (IP3) and diacylglycerol. It was previously assumed that the sole purpose of PIP2 production was to act as a precursor for IP3 whhereas the remaining diacylglycerol would activate protein kinase C. It is clear that PIP2 is a signaling molecule by itself and can directly activate other effector and signaling proteins. In particular, PIP2 can bind to ion channels and modify their activity. The best-studied interactions in this case are between PIP2 and K channels. PIP2 binds the C-terminal domains of these K channels to increase channel activity; PIP2 is absolutely required for the gating of these K+ channels. The binding of the anionic PIP2 to K+ channels appears to involve, but is not limited to, electrostatic interactions. PIP2 also modulates the activity of ENaC through similar interactions with hydrophobic and positively charged residues in the cytosolic tails of the β- and γ-subunits.

PHOSPHOINOSITIDE METABOLISM IS IMPORTANT FOR ENaC ACTIVITY 4,5-PIP2 is just one of the inositol phospholipids that are produced by different inositol lipid kinases. PI-3-kinase is present in kidney cells. Inhibition of this enzyme with a specific PI-3-kinase inhibitor, LY-294002, blocks the basal rate and insulinstimulated rate of Na+ transport. The products of PI-3-kinase include PI-3-P (PIP), PI-3,4-P (PIP2), and PI-3,4,5-P (PIP3). These observations imply that phosphoinositides are involved in regulation of ENaC activity. THE MOLECULAR BASIS FOR ENaC REGULATION BY ALDOSTERONE The mineralocorticoid, aldosterone, is the primary hormonal regulator of total body sodium balance. Increases in the circulating levels of the steroid specifically increase Na+ reabsorption at the apical membrane by a two-phase reaction: an initial phase that increases transport four- to sixfold in the first 2–6 h followed by a late phase at 12–48 h, which increases transport another three to fourfold. The mechanisms for the early and late phases appear to be different. Aldosterone, like other steroid hormones, enters target cells (in the case of the kidney, the major targets are the principal cells of the collecting duct) and binds to cytosolic, mineralocorticoid receptor complexes. In mammalian principal cells, mineralocorticoid receptors are not activated by circulating glucocorticoids, because the enzyme, 11-β-hydroxysteroid dehydrogenase, changes the glucocorticoid structure to a form that does not bind aldosterone receptors. The enzyme does not change the aldosterone structure. After binding aldosterone, there is a conformational change in the aldosterone receptor. In its new conformation, it acts as a DNA binding protein and interacts with mineralocorticoid response elements. Binding to the mineralocorticoid response elements leads to changes in gene expression. For example, an increase in Na+ transport can be measured within 1 h of exposure to aldosterone, by a mechanism that depends on gene transcription and translation. The induced gene products are referred to as aldosterone-induced proteins. Aldosterone and increased intracellular sodium also promote expression of the sodium pump during long-term exposure. Many attempts have been made to identify the aldosteroneinduced proteins. The relationship between the genes identified and the mechanism by which aldosterone increases Na+ transport is often unclear. Originally, it was hypothesized that aldosterone induces ENaC synthesis and insertion into the cell membrane. Although the evidence from electrophysiological measurements remains somewhat controversial, the available biochemical evidence suggests that the number of sodium channel proteins in the apical membrane does not change after aldosterone treatment (at least in the first 2–4 h when the increase in sodium transport is most dramatic). The evidence suggests instead that the early effects of aldosterone are indirect and that the proteins that are synthesized are modulatory proteins that convert poorly transporting ENaC in the apical membrane into channels that readily transport sodium. The three homologous ENaC subunit proteins form a heteromultimeric channel protein apparently made up of two α-, one β-, and one γ-subunit (although there is some disagreement about the exact stoichiometry). The most obvious link between aldosterone and increased ENaC activity would be a post-translational modification of the channels leading to an increase in the probability that the channels are open and hence, transporting sodium. A second hypothesis is that most or all of the ENaC channel subunits are present but dispersed in the apical membrane and aldosterone promotes the assembly of a channel complex, which transports sodium.

CHAPTER 59 / HYPERTENSION AND SODIUM CHANNEL TURNOVER

619

Figure 59-3 Schematic diagram of the short-term action of aldosterone. Aldosterone produces an increase in K-Ras2A protein (A), but K-ras protein is inactive unless post-translationally modified by prenylation and methylation. The rate of methylation is usually a rate limiting step in K-Ras activation; therefore, aldosterone activation of the isoprenylcysteine-O-methyl transferase is critical for complete activation of K-Ras (B). In contrast to K-Ras, however, aldosterone does not increase expression of methyl transferase. Rather, aldosterone activates protein kinase C and protein kinase C phosphorylates and activates methyl transferase. At the same time, aldosterone increases the activity of SAH hydrolase. This is critical for normal methyl transferase because one end product of the methylation reaction is SAH, which is a potent inhibitor of methyl transferases. Methylation can only proceed if SAH is hydrolyzed to homocysteine. Once K-Ras is methylated, the protein associates with the membrane and can be activated by a guanine nucleotide exchange factor: K-Ras is activated by aldosterone-induced increases in the Ras guanine nucleotide exchange factor, Sos. K-Ras activation leads to the activation of the inositol lipid kinase, 4-PIP-5-Kinase (C) that produces PIP2, which directly activates ENaC (D). PIP2 is also metabolized by a second aldosterone-induced kinase, PI-3-Kinase (E) to produce PIP3. PIP3 can also directly activate ENaC, but also initiates the late phase of the aldosterone response (F). Other events initiated by aldosterone involve heterotrimeric G proteins that are necessary for ENaC activation by a poorly understood mechanism (G).

The mechanism for the long-term effect of aldosterone that increases sodium transport is also unclear. There is evidence that the total cellular amount of ENaC subunits may increase but whether this is owing to an increase in translation is unknown because changes in ENaC mRNA levels vary in different tissues. The longterm rise in ENaC activity could be due to an increase in translation, an increase in trafficking of subunits to the membrane, post-translational modifications of channels in the membrane, or less degradation of membrane channels. THE MOLECULAR MECHANISM FOR THE SHORT-TERM EFFECTS OF ALDOSTERONE ON SODIUM TRANSPORT One possibility for the activation of sodium channels with little or no change in the number of channels is a post-translational modification that changes the open probability of channels already present in the surface membrane. The problem is that most of the usual post-translational modifications like phosphorylation seem to have no effect on channel activity or even decrease its activity. Other types of posttranslational modifications, however, do appear in response to aldosterone and can alter sodium channel open probability (Fig. 59-3).

Aldosterone-Induced Methylation Activates ENaC The amount of sodium transport that can be measured in apical membrane vesicles obtained from kidney cells is markedly enhanced by prior application of agents that methylate membrane proteins. Application of aldosterone stimulates the methylation of membrane proteins whereas inhibition of methylation eliminates the effects of aldosterone. The isoprenyl-cysteine-O-carboxy-methyltransferase responsible for K-Ras2A methylation is not an aldosterone-induced protein, but the intrinsic activity of the enzyme does increase in response to aldosterone (apparently via protein kinase C phosphorylation). In addition, the activity of the other enzyme intimately involved in protein methylation, SAH hydrolase, is increased by aldosterone. Taken together, it is clear that aldosterone-induced methylation modulates ENaC activity. Some investigators have suggested that the target for methylation is an ENaC subunit, but the primary amino acid sequence of the subunits makes it difficult to accept this view. Almost exclusively, reversible methylation of signaling proteins occurs only at a restrictive consensus sequence: the so-called “CAAX” box. Even if there

620

SECTION VII / NEPHROLOGY

Figure 59-4 Schematic diagram of the long-term action of aldosterone. In the absence of aldosterone, there is little PIP2 in the membrane so ENaC is inactive. In addition, Nedd4 ubiquitinates ENaC and promotes rapid internalization and degradation (A). In the presence of aldosterone, all the initial short-term events depicted in Fig. 59-3 occur, starting at (B) and leading to an increase in PIP2 and immediate activation of preexisting ENaC. The late aldosterone responses are initiated when PI-3-kinase is activated by K-Ras to metabolize PIP2 to PIP3 (C). PIP3 activates PDK1/2 (D), which phosphorylates SGK, which has already been strongly induced by aldosterone (E). SGK phosphorylates Nedd4 (F). Phosphorylation of Nedd4 reduces its affinity for ENaC subunits and, thereby, reduces ENaC subunit ubiquitin conjugation. In the absence of ubiquitin conjugation, ENaC subunits are not internalized and degraded and accumulate in the surface membrane (G) in which they contribute to the aldosterone-induced increases in sodium transport.

were significant C-terminal proteolytic trimming of ENaC, there are no cysteine residues that would meet the criteria for methylation. There must be another target for aldosterone methylation. K-Ras2A: An Aldosterone-Induced Protein and Target for Aldosterone-Mediated Methylation Proteins differentially expressed by aldosterone include one isoform of K-Ras, K-Ras2A. Methylation of K-Ras is required for its activation and Ras methylation is induced by aldosterone. K-Ras is methylated by an isoprenylcysteine methyl transferase. This enzyme requires the presence of an isoprenyl group conjugated to the cysteine of the CAAX box. Therefore, it is not surprising that inhibition of prenylation blocks the action of aldosterone. K-Ras2A and Aldosterone Activate PI Lipid Kinases One of the genes induced by aldosterone and by K-Ras2A is PIP-5-kinase. An increase in PIP-5-kinase also leads to an increase in membrane PIP2. Besides the 5-kinase, PI-3-kinase is activated by aldosterone whereas inhibition of PI-3-K activity blocks aldosterone-induced increases in Na+ transport. Thus, the aldosterone-induced shortterm increase in ENaC activity is due to an increase in anionic lipids that can directly activate existing ENaC channels. THE MECHANISM FOR THE LONG-TERM EFFECTS OF ALDOSTERONE IN A6 CELLS The observation that PI-3kinase is required for aldosterone activation at times longer than 4 h suggests an alternative mechanism is active following the initial

PIP2-mediated increase in open probability of ENaC. The product of PI-3-kinase, PIP3, activates proteins that have PIP3-binding domains (also known as pleckstrin homology or PH domains). Notable among these proteins are the PI-dependent kinases, PDK1/2. Because aldosterone increases PIP3, aldosterone activates PDK1/2. Activation of PDK1/2 is interesting because one of the kinase’s phosphorylation targets is another kinase, serum and glucocorticoid-dependent kinase (SGK). SGK is coded by an aldosteroneinduced gene and the long-term effects of aldosterone require activation of SGK by a phosphorylation reaction. The kinase activity of SGK targets a variety of proteins including the ubiquitin ligase, neural-precursor-cell-expressed-developmentally downregulated protein (Nedd4). Nedd4 is intimately related to degradation of ENaC proteins in the surface membrane. A Specific Ubiquitin Ligase, Nedd4, is Responsible for Ubiquitin Conjugation and Subsequent Degradation of ENaC in the Membrane For many proteins including ENaC, ubiquitin conjugation is a prerequisite for membrane internalization and subsequent degradation. In kidney cells, membrane-associated ENaC α- and β-subunits are ubiquitin conjugated (whether the γ-subunit is also conjugated to ubiquitin has not been determined) and the conjugates can be degraded by proteasomes. Ubiquitinconjugation of ENaC subunits requires a specific ubiquitin ligase, Nedd4. Nedd4 was found to have the same localization pattern as

CHAPTER 59 / HYPERTENSION AND SODIUM CHANNEL TURNOVER

ENaC in principal cells present in the cortex and outer medulla of the collecting duct. Nedd4 apparently directly associates with ENaC to produce ubiquitin conjugation. Nedd4 can regulate the number of functional ENaC in the surface membrane because blocking Nedd4 synthesis with anti-sense oligonucleotides led to an increase in the trans-epithelial current. These results imply that Nedd4 is the ubiquitin ligase responsible for ENaC ubiquitination in native renal cells. ENaC Degradation in Renal Epithelial Cells In kidney cells, proteasome inhibition increases the total amount of all ENaC subunits. The half-lives of the β- and γ-subunits in the total cellular pool is relatively short (1–3 h) but rises severalfold when proteasome activity is inhibited. Proteasome inhibition also increases amiloride-sensitive trans-epithelial current measured in kidney cells. This response is associated with an increase in the apical density of sodium channels as measured by patch clamp methods. There also is an increase in the number of ENaC subunits in the cell surface and an increase in the ubiquitin conjugation of ENaC proteins present in the surface membrane. The half-lives of membrane-associated (i.e., biotinylated β- and γ-ENaC) subunit increase markedly when proteasome activity is inhibited. The change is from 3.7 h for β-subunit or 7.5 h for γ-subunit to greater than 24 h. Because the half-life of α-subunits at the surface of native cells is at least 42 h, it is not clear if proteasome inhibition will also increase the half-life of α-subunits. Aldosterone Regulation of ENaC Degradation Rate in Renal Cells Aldosterone increases the number of ENaC subunits on the apical surface of cultured kidney cells even in the absence of any significant change in mRNA levels for any of the subunits. This implies that aldosterone can alter ENaC degradation. Because aldosterone increases the amount and activity of SGK, and because SGK can phosphorylate Nedd4, aldosterone could impair degradation of ENaC by the proteasome because a phosphorylated Nedd4 would have a lower affinity for binding to ENaC subunits. Figure 59-4 provides a schematic view of a possible mechanism for the long-term action of aldosterone.

CONCLUSION Aldosterone, the hormone primarily responsible for altering epithelial Na transport, apparently uses at least two mechanisms to regulate ENaC activity. The initial rapid response to aldosterone involves the production of PIP2 (and PIP3). These effectors rapidly increase the activity of ENaCs that are already in the surface membrane. Aldosterone also increases the number of functional channels in the surface membrane, at least in part, by reducing the degradation rate of ENaC and allowing an increase in the surface membrane pool of functional ENaC.

ACKNOWLEDGMENTS This work was supported by DHHS 1R37 DK037963, 1R01 HL071621, and 1P50 AA-013757 to DCE.

SELECTED REFERENCES Al-Baldawi NF, Eaton DC. Aldosterone activates isoprenyl methyltransferase in A6 epithelia by serine phosphorylation. FASEB J 2001;15: A432, abstract. Al-Baldawi NF, Stockand JD, Al Khalili OK, Yue G, Eaton DC. Aldosterone induces Ras methylation in A6 epithelia. Am J Physiol Cell Physiol 2000;279(2):C429–C439.

621

Becchetti A, Kemendy AE, Stockand JD, Sariban-Sohraby S, Eaton DC. Methylation increases the open probability of the epithelial sodium channel in A6 epithelia. J Biol Chem 2000;275(22):16,550–16,559. Canessa CM, Horisberger JD, Rossier BC. Epithelial sodium channel related to proteins involved in neurodegeneration. Nature 1993;361:467–470. Canessa CM, Schild L, Buell G, et al. Amiloride-sensitive epithelial Na+ channel is made of three homologous subunits. Nature 1994;367: 463–467. Corvol P, Persu A, Gimenez-Roqueplo AP, Jeunemaitre X. Seven lessons from two candidate genes in human essential hypertension: angiotensinogen and epithelial sodium channel. Hypertension 1999; 33(6):1324–1331. Goulet CC, Volk KA, Adams CM, Prince LS, Stokes JB, Snyder PM. Inhibition of the epithelial Na+ channel by interaction of Nedd4 with a PY motif deleted in Liddle’s syndrome. J Biol Chem 1998;273(45): 30,012–30,017. Grunder S, Zagato L, Yagil C, Yagil Y, Sassard J, Rossier BC. Polymorphisms in the carboxy-terminus of the epithelial sodium channel in rat models for hypertension. J Hypertens 1997;15(2):173–179. Hummler E. Epithelial sodium channel, salt intake, and hypertension. Curr Hypertens Rep 2003;5(1):11–18. Hummler E, Horisberger JD. Genetic disorders of membrane transport. V. The epithelial sodium channel and its implication in human diseases. Am J Physiol 1999;276:G567–G571. Kamynina E, Staub O. Concerted action of ENaC, Nedd4-2, and Sgk1 in transepithelial Na(+) transport. Am J Physiol Renal Physiol 2002; 283(3):F377–F387. Lingueglia E, Voilley N, Waldmann R, Lazdunski M, Barbry P. Expression cloning of an epithelial amiloride-sensitive Na+ channel. A new channel type with homologies to Caenorhabditis elegans degenerins. FEBS Lett 1993;318:95–99. Malik B, Schlanger L, Al Khalili O, Bao HF, Yue G, Price SR, et al. ENaC degradation in A6 cells by the ubiquitin-proteosome proteolytic pathway. J Biol Chem 2001;276(16):12,903–12,910. Matsubara M. Genetic determination of human essential hypertension. Tohoku J Exp Med 2000;192(1):19–33. Pearce D. The role of SGK1 in hormone-regulated sodium transport. Trends Endocrinol Metab 2001;12(8):341–347. Pratt JH, Ambrosius WT, Agarwal R, Eckert GJ, Newman S. Racial difference in the activity of the amiloride-sensitive epithelial sodium channel. Hypertension 2002;40(6):903–908. Rossier BC, Pradervand S, Schild L, Hummler E. Epithelial sodium channel and the control of sodium balance: interaction between genetic and environmental factors. Annu Rev Physiol 2002;64:877–897. Schafer JA. Abnormal regulation of ENaC: syndromes of salt retention and salt wasting by the collecting duct. Am J Physiol Renal Physiol 2002;283(2):F221–F235. Schild L. The ENaC channel as the primary determinant of two human diseases: Liddle syndrome and pseudohypoaldosteronism. Nephrologie 1996;17(7):395–400. Staub O, Abriel H, Plant P, Ishikawa T, Kanelis V, Saleki R, et al. Regulation of the epithelial Na+ channel by Nedd4 and ubiquitination. Kidney Int 2000;57(3): 809–815. Stockand JD, Al-Baldawi NF, Al Khalili OK, Worrell RT, Eaton DC. S-adenosyl-L-homocysteine hydrolase regulates aldosterone-induced Na+ transport. J Biol Chem 1999;274(6):3842–3850. Stockand JD, Edinger RS, Al-Baldawi NF, et al. Isoprenylcysteine-Ocarboxyl methyltransferase regulates aldosterone-sensitive Na+ reabsorption. J Biol Chem 1999;274(38):26,912–26,916. Stockand JD, Edinger RS, Eaton DC, Johnson JP. Toward understanding the role of methylation in aldosterone-sensitive Na(+) transport. News Physiol Sci 2000;15:161–165. Warnock DG. Aldosterone-related genetic effects in hypertension. Curr Hypertens Rep 2000;2(3):295–301. Yue G, Malik B, Yue G, Eaton DC. Phosphatidylinositol 4,5-bisphosphate (PIP2) stimulates epithelial sodium channel activity in A6 cells. J Biol Chem 2002;277(14):11,965–11,969.

60 Nephrogenic Diabetes Insipidus Water and Urea Transport

JEFF M. SANDS AND DANIEL G. BICHET SUMMARY Nephrogenic diabetes insipidus (NDI) is a rare disorder in which the kidney is unresponsive to the water-retaining action of vasopressin (antidiuretic hormone). Congenital NDI results from a genetic mutation in the V2-receptor, AQP2, or other transport proteins, such as the UT-B urea transporter, involved in generating a hypertonic renal medulla. Acquired NDI most commonly results from lithium therapy, but can also result from prolonged hypercalcemia, protein malnutrition, hypokalemia, and following the release of unilateral or bilateral ureteral obstruction. The current therapeutic options for congenital NDI are limited and only partially beneficial in reducing the excessive urine output. Key Words: Adenylyl cyclase; antidiuretic hormone; aquaporins (water channels); countercurrent exchange; countercurrent multiplication; diabetes insipidus; kidney medulla; lithium therapy; renal collecting duct; urea transporters; urine concentrating mechanism; vasopressin receptor.

INTRODUCTION Patients with diabetes insipidus (DI) produce large quantities of dilute urine. DI can be “central,” owing to failure of the posterior pituitary to make or secrete vasopressin (arginine vasopressin [AVP], also called antidiuretic hormone), or “nephrogenic,” from failure of the kidney to respond to vasopressin. Nephrogenic DI (NDI) has many causes that fall into two general categories, failure of the collecting duct to respond to vasopressin and reabsorb water, and failure of the medulla to generate an osmotic gradient. Water is reabsorbed by principal cells in the collecting duct (Fig. 60-1). Vasopressin binds to V2-receptors (V2R) in the basolateral membrane of principal cells. The V2R is a seven transmembrane-spanning, G protein-coupled receptor. Activation of this receptor results in the generation of the second messenger cyclic adenosine 5´-monophosphate (AMP), activation of protein kinase A, and stimulation of the insertion of aquaporin-2 (AQP2) water channels into the apical membrane. AQP2 exists as homotetramers in the apical membrane, and at least one of the four AQP2 molecules needs to be phosphorylated at serine-256 for proper trafficking and regulation of water reabsorption. Water From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

molecules enter the principal cell via AQP2 and exit via the related water channels AQP3 and AQP4 in the basolateral membrane. When the vasopressin stimulus is removed, AQP2 is removed from the apical membrane by endocytosis. These AQP2 molecules are recycled into new endosomes and wait for the next stimulus by vasopressin. Any abnormality in any of the steps involved in collecting duct water reabsorption results in NDI. In addition to needing sufficient vasopressin and a normally functioning collecting duct, the medulla must generate an osmotic gradient for water reabsorption to occur (Fig. 60-2). In the outer medulla, NaCl is actively reabsorbed in the thick ascending limb of the loop of Henle by the Na+-K+-2Cl– cotransporter NKCC2/BSC1. Active NaCl reabsorption is critical for establishing a hypertonic medullary interstitium. Inhibiting this process with a loop diuretic results in isosmotic urine. In the inner medulla, NaCl is passively reabsorbed in the thin ascending limb, driven by the chemical gradients for urea and NaCl. The passive mechanism hypothesis requires that the concentration of urea in the inner medullary interstitium exceed that found in the lumen of the thin ascending limb. The urea concentration gradient permits the NaCl concentration in the interstitium to be lower than in the lumen of the thin ascending limb, thereby establishing a gradient for passive NaCl reabsorption in the absence of an osmotic gradient. Any abnormality in any of the steps involved in generating a hypertonic medulla results in NDI.

CONGENITAL NDI About 90% of patients with congenital NDI are males with X-linked recessive inheritance (MIM 304800) who have mutations in the arginine vasopressin 2 receptor (AVPR2) gene that codes for the V2R. The gene is located in chromosome region Xq28. In less than 10% of the families that have been studied, congenital NDI has an autosomal-recessive or autosomal-dominant mode of inheritance (MIM 222000 and 125800). In many of these families, mutations have been identified in the AQP2 gene, which is located in chromosome region 12q13 and codes for the vasopressin-sensitive AQP2 water channel. The clinical characteristics of NDI include hypernatremia, hyperthermia, mental retardation, and repeated episodes of dehydration in early infancy. Mental retardation, a probable consequence of repeated episodes of dehydration, was prevalent in 70–90% of the patients reported in the original studies.

622

CHAPTER 60 / NEPHROGENIC DIABETES INSIPIDUS

623

Figure 60-1 Schematic representation of the effect of vasopressin (AVP) to increase water permeability in the principal cells of the collecting duct. Note that Na+ reabsorption, through the epithelial Na channel is not represented. Vasopressin is bound to the V2-receptor (a G protein-linked receptor) on the basolateral membrane. The basic process of G protein-coupled receptor signaling consists of 3 steps: a hepta-helical receptor that detects a ligand (in this case, AVP) in the extracellular milieu, a G protein that dissociates into α-subunits bound to GTP and βδ-subunits after interaction with the ligand-bound receptor, and an effector (in this case, adenylyl cyclase) that interacts with dissociated G protein subunits to generate small-molecule second messengers. AVP activates adenylyl cyclase increasing the intracellular concentration of cyclic AMP (cAMP). The topology of adenylyl cyclase is characterized by two tandem repeats of six hydrophobic transmembrane domains separated by a large cytoplasmic loop and terminates in a large intracellular tail. Generation of cAMP follows receptor-linked activation of the heteromeric G protein (Gs) and interaction of the free Gs-chain with the adenylyl cyclase catalyst. Protein kinase A is the target of the generated cAMP. Cytoplasmic vesicles carrying the water channel proteins (represented as homotetrameric complexes) are fused to the luminal membrane in response to vasopressin, thereby increasing the water permeability of this membrane. Microtubules and actin filaments are necessary for vesicle movement toward the membrane. The mechanisms underlying docking and fusion of AQP2 bearing vesicles are not known. The detection of the small GTP-binding protein Rab3a, synaptobrevin 2, and syntaxin 4 in principal cells suggests that these proteins are involved in AQP2 trafficking. When vasopressin is not available, water channels are retrieved by an endocytic process, and water permeability returns to its original low rate. AQP3 and AQP4 water channels are expressed on the basolateral membrane.

However, data suggest that mental retardation is less prevalent and may have been overestimated in NDI patients. Early recognition and the subsequent treatment of congenital NDI with an abundant intake of water allows a normal lifespan with normal physical and mental development. Two characteristics suggestive of X-linked NDI are the familial occurrence and the confinement of mental retardation to male patients. It is then tempting to assume that the family described by McIlraith and discussed by Reeves and Andreoli was affected by congenital NDI. Lacombe and Weil described an autosomaldominant type of transmission without any associated mental retardation. The descendants of the family originally described by Weil were later found to have autosomal-dominant central (neurogenic) DI, a well-characterized entity secondary to mutations in the prepro-vasopressin-neurophysin II gene, MIM 192340. Patients with autosomal-dominant central (neurogenic) DI retain some limited capacity to secrete vasopressin during severe dehydration and the polyuric–polydipsic symptoms usually appear after the first year of life when the ability of the infants to seek water is more likely to be understood by adults. Other inherited disorders with mild, moderate, or severe inabilility to concentrate

urine include Bartter’s syndrome (MIM 601678) and cystinosis (MIM 219800).

NDI OWING TO V2-RECEPTOR MUTATIONS The majority (approx 90%) of patients with congenital NDI have the X-linked form; affected male patients do not concentrate their urine after the administration of vasopressin. Because recessive X-linked disease is rare, females are unlikely to be affected, but heterozygous females exhibit variable degrees of polyuria and polydipsia because of skewed X-chromosome inactivation. In Quebec, the incidence of this disease was estimated to be approx 8.8 per million male live births for the 10-yr period 1988–1997. A founder effect for a particular AVPR2 mutation in Irish families of Scottish descent resulted in an increased prevalence of X-linked NDI in their descendants who emigrated to North America in 1761 on the Ship Hopewell. The incidence in Nova Scotia and New Brunswick has been estimated at 58 per million male live births for the 10-yr period 1988–1997. The W71X mutation was identified as the cause of NDI in the extended “Hopewell” kindred and in families in the Canadian Maritime provinces. The W71X mutations of these patients are probably identical by descent,

624

SECTION VII / NEPHROLOGY

Figure 60-2 Diagram showing the location of the major transport proteins involved in the urine concentrating mechanism in the outer and inner medulla. UT, urea transporter; AQP, aquaporin; NKCC/BSC, Na+-K+-2Cl– cotransporter; ROMK, renal outer medullary K channel; ClC-K1, chloride channel; NCC/TSC, Na-Cl cotransporter.

although a common ancestor has not been identified in all cases. Among patients with X-linked NDI in North America, the W71X mutation is more common than any other AVPR2 mutation. To date, 183 putative disease-causing AVPR2 mutations have been identified in 284 NDI families (Fig. 60-3; additional information is available from the NDI Mutation Database: http://www. medicine.mcgill.ca/nephros/). Half of the mutations are missense mutations: frameshift mutations owing to nucleotide deletions or insertions (25%); nonsense mutations (10%); large deletions (10%); inframe deletions or insertions (4%); splice-site mutations; and one complex mutation account for the remainder. Mutations have been identified in every domain of the AVPR2 receptor, but on a per nucleotide basis, about twice as many mutations occur in transmembrane domains compared with the extracellular or intracellular domains. Private mutations, recurrent mutations, and mechanisms of mutagenesis have been previously identified. The 10 recurrent mutations (D85N, V88M, R113W, Y128S, R137H, S167L, R181C, R202C, A294P, and S315R) were found in 35 ancestrally independent families. The occurrence of the same mutation on a different haplotype was considered evidence of recurrent mutation. In addition, the most frequent mutations — D85N, V88N, R113W, R137H, S167L, R181C, and R202C — occurred at potential mutational hot spots (a G-to-T or G-to-A nucleotide substitution occurred at a CpG dinucleotide). When studied in vitro, most AVPR2 mutations lead to receptors that are trapped intracellularly and unable to reach the plasma membrane. A few mutant receptors reach the cell surface but are unable to bind vasopressin or to properly trigger an intracellular cyclic AMP signal. AVPR2 mutations are thus inducing a defect in intracellular protein transport, an increasingly recognized anomaly of hereditary disease-causing mutations. Of clinical interest, nonpeptide V2R antagonists were found to facilitate the folding of mutant AVPR2 receptors and to increase urine

osmolality in patients bearing the 62-64, R137H, and W164S mutations.

NDI OWING TO AQP2 WATER CHANNEL MUTATIONS On the basis of dDAVP (desmopressin) infusion studies and phenotypic characteristics of males and females affected with NDI, a non-X-linked form of NDI with a postreceptor (postcyclic AMP) defect was suggested. A patient who presented shortly after birth with typical features of NDI but who exhibited normal coagulation and normal fibrinolytic and vasodilatory responses to dDAVP was shown to be a compound heterozygote for two missense mutations (R187C and S217P) in the AQP2 gene (Fig. 60-4). To date, 32 putative disease-causing AQP2 mutations have been identified in 40 NDI families. By type of mutation, there are 69% missense, 19% frameshift owing to small nucleotide deletions or insertions, 6% nonsense, and 6% splice-site mutations (additional information is available in the NDI Mutation Database at http://www.medicine.mcgill.ca/nephros/.) Reminiscent of expression studies done with AVPR2 proteins, misrouting of AQP2 mutant proteins has been shown to be the major cause underlying autosomal-recessive NDI. In contrast to the AQP2 mutations in autosomal-recessive NDI, which are located throughout the gene, the dominant mutations are predicted to affect the carboxyl terminus of AQP2. The autosomal-dominant family described by Ohzeki et al. was found to be heterozygous for the 721G mutation. The dominant AQP2-E258K mutation was retained in the Golgi apparatus, which differs from AQP2 mutants in recessive NDI that are retained in the endoplasmic reticulum. The dominant action of the AQP2-E258K mutation can be explained by the formation of heterotetramers of mutant and wild-type AQP2 that are impaired in their routing after oligomerization. Hetero-oligomerization of AQP2-727G with wild-type

CHAPTER 60 / NEPHROGENIC DIABETES INSIPIDUS

625

Figure 60-3 Schematic representation of the V2-receptor and identification of 183 putative disease-causing AVPR2 mutations. Predicted amino acids are given as the one-letter code. Solid symbols indicate missense or nonsense mutations, a number indicates more than one mutation in the same codon. The names of the mutations and extracellular, transmembrane, and cytoplasmic domains were defined according to recommended nomenclature. The common names of the mutations are listed by type. Eighty-nine missense, 18 nonsense mutations, 45 frameshift, 7 inframe deletions or insertions, four splice-site, as well as 19 large deletions, and one complex mutation have been identified. (Reproduced with permission from Fujiwara TM, Bichet DG. Molecular biology of hereditary diabetes insipidus. J Am Soc Nephrol 2005;16:2836–2846.)

AQP2 and the consequent mistargeting of this complex to late endosomes/lysosomes results in the absence of AQP2 in the apical membrane, which can explain the dominant phenotype.

ROLE OF AQP2 IN ACQUIRED NDI The acquired form of NDI is much more common, especially in adult medicine, than the congenital form of NDI, but it is rarely severe. The ability to elaborate hypertonic urine is usually preserved, in spite of the impairment of the maximal concentrating ability of the nephrons. Polyuria and polydipsia are therefore moderate (3–4 L/d). Lithium administration has become the most frequent cause of acquired NDI. A review reported this abnormality in at least 54% of 1105 unselected patients on chronic lithium therapy. Nineteen percent of these patients had polyuria, as defined by a 24-h urine output exceeding 3 L. Lithium inhibits adenylyl cyclase in a number of cell types including renal epithelia. The concentration of lithium in the urine of patients on wellcontrolled lithium therapy (i.e., 10–40 mmol/L) is sufficient to exert this effect. Measurements of adenylyl cyclase activity in membranes isolated from a cultured pig kidney cell line revealed

that lithium in the concentration range of 10 mmol/L interfered with the hormone-stimulated guanyl nucleotide regulatory unit (Gs). The effect of chronic lithium therapy on the expression of AQP2 has been studied in rat kidney membranes prepared from the inner medulla. Lithium caused a marked downregulation that was only partially reversed by cessation of therapy, thirsting, or dDAVP treatment, consistent with clinical observations of slow recovery from lithium-induced urinary concentrating defects. Hypokalemia, hypercalciuria, low-protein diets, and the release of bilateral ureteral obstruction have also been shown to downregulate AQP2 protein expression in rat kidney medulla. In patients on long-term lithium therapy, amiloride has been proposed to prevent the uptake of lithium into the collecting ducts, thus preventing the inhibitory effect of intracellular lithium on water transport.

NDI OWING TO UT-B UREA TRANSPORTER MUTATIONS Urea is a small but highly polar molecule that undergoes extensive hydrogen-bonding in aqueous solution. Despite the perception that urea is freely permeable across all cell membranes, its

626

SECTION VII / NEPHROLOGY

Figure 60-4 (A) Schematic representation of the aquaporin-2 (AQP2) protein and identification of 32 putative disease causing AQP2 mutations. A monomer is represented with six transmembrane helices. The location of the protein kinase A phosphorylation site (Pa) is indicated. The extracellular, transmembrane, and cytoplasmic domains are defined according to Deen et al. (1994). Solid symbols indicate the location of the mutations. The common names of the mutations are listed by type. Twenty-five missense, two nonsense, six frameshift, as well as two splice-site mutation have been identified. (B) Schematic representation of the 3D structure of AQP2 by analogy to the X-ray crystallographic structure of AQP1. The asterisk indicates the position in which the molecular pseudo-twofold symmetry is strongest. (Reproduced with permission from Fujiwara TM, Bichet DG. Molecular biology of hereditary diabetes insipidus. J Am Soc Nephrol 2005;16:2836–2846.)

permeability across artificial lipid bilayers is low, as would be expected for a polar molecule. However, urea’s permeability is not zero, and given sufficient time, it diffuses across cell membranes and achieves equilibrium in the steady state. In the kidney, red blood cells (RBCs) traverse the vasa recta, and tubule fluid traverses the collecting duct much too rapidly to permit urea concentrations to achieve equilibrium solely by passive diffusion. Rapid urea transport and the high urea permeability of RBC and terminal inner medullary collecting duct cell membranes were the initial evidence suggesting the presence of specific urea transporter proteins. The existence of urea transport proteins has been confirmed by the cloning of two urea transporter genes and several cDNA isoforms. In addition to RBCs and kidney, urea transporters are expressed in liver, testis, heart, and brain, and in some of these tissues, their abundance is altered in uremic rats. UT-B is the RBC-facilitated urea transporter and was cloned from a human erythropoietic cell line. The UT-B gene (SLC14A1) gene is located on chromosome 18q12. It is about 30-kb long, contains 11 exons, and the coding region begins in exon four and extends through exon 11. Interestingly, the UT-B protein is also the Kidd (or Jk) antigen. The Kidd antigen is a minor blood group antigen, and several mutations involving exons six and seven of

the UT-B/Kidd antigen (SCL14A1) gene have been reported. People lacking Kidd antigen have a form of congenital NDI as they are unable to concentrate their urine greater than 800 mOsm/kg H2O, even following overnight water deprivation and exogenous vasopressin administration. These people lack UT-B-mediated urea transport in their RBCs and presumably also lack it in their descending vasa recta. Although no biopsy data are available to prove this assumption in humans, a UT-B knock-out mouse does lack urea transporter in both its RBCs and descending vasa rectae, and has impaired urea recycling and urine concentrating ability. Individuals lacking UT-B have congenital NDI because UT-B is necessary for efficient countercurrent exchange between the ascending and descending vasa recta, and any decrease in countercurrent exchange reduces urine concentrating ability.

ACQUIRED NDI OWING TO UT-A UREA TRANSPORTER DOWNREGULATION The UT-A gene SLC14A2 is located on chromosome 18, adjacent to the gene for UT-B. It is about 68 kb in length and contains 20 exons. The human gene is organized in a manner that is similar to the rat and mouse genes (Slc14a2) and is an atypical gene because it has two promoter elements, promoter I, which is

CHAPTER 60 / NEPHROGENIC DIABETES INSIPIDUS

upstream of exon 1 and drives transcription of isoforms UT-A1, UT-A3, UT-A4, UT-A5, and UT-A6. and promoter II, which is located within intron 12 and drives transcription of isoform UTA2. To date, no mutations in any of the UT-A isoforms have been reported in people. However, single nucleotide polymorphisms in UT-A2 are associated with variation in blood pressure in men, but not in women. UT-A1 is the best studied and largest UT-A protein and is expressed in the apical membrane of the inner medullary collecting duct. UT-A1 is stimulated by cyclic AMP when expressed in Xenopus oocytes. UT-A3 is also expressed in the inner medullary collecting duct and stimulated by cAMP. A UT-A1/UT-A3 double knock-out mouse has a severe impairment in urine concentrating ability. Despite its name, UT-A2 was the first cloned urea transporter cDNA. UT-A2 is expressed in thin descending limbs and is not stimulated by cyclic AMP analogs. A UT-A2 knock-out mouse has a mild impairment in urine concentrating ability. Lithium causes a marked reduction in AQP2 protein abundance. Lithium also causes a marked reduction in UT-A1 and UTB protein abundances and interferes with vasopressin’s ability to phosphorylate UT-A1. In inner medullary collecting ducts from normal rats, vasopressin rapidly increases UT-A1 phosphorylation and urea transport. However, in inner medullary collecting ducts from lithium-fed rats, vasopressin does not increase UT-A1 phosphorylation. Thus, lithium causes NDI by decreasing the abundance of several critical transport proteins involved in the urine concentrating mechanism, AQP2, UT-A1, and UT-B. The reduction in AQP2 decreases water reabsorption across the collecting duct apical membrane in response to vasopressin, thereby reducing transepithelial water reabsorption. The reduction in UTA1 and phosphorylated UT-A1 decreases urea reabsorption across the inner medullary collecting duct in response to vasopressin, thereby reducing inner medullary interstitial urea accumulation, which reduces urine concentrating ability. The reduction in UT-B decreases urea recycling and the efficiency of countercurrent exchange, which reduces urine concentrating ability. Unfortunately, if lithium-induced NDI is not diagnosed quickly and lithium therapy stopped, the acquired NDI often becomes irreversible.

ACKNOWLEDGMENTS This work was supported by National Institutes of Health Grants R01-DK41707, R01-DK63657, and P01-DK61521 (to JMS) and by the Canadian Institutes of Health Research Grant MT-8126 and the Kidney Foundation of Canada (to DGB). Address for correspondence: Dr. Jeff M. Sands, Emory University School of Medicine, Renal Division, WMRB Room 338, 1639 Pierce Drive, NE, Atlanta, GA 30322 USA. Phone: 404-727-2525; FAX: 404-727-3425; E-mail: [email protected].

SELECTED REFERENCES Aridor M, Balch WE. Integration of endoplasmic reticulum signaling in health and disease. Nat Med 1999;5:745–751. Arthus M-F, Lonergan M, Crumley MJ, et al. Report of 33 novel AVPR2 mutations and analysis of 117 families with X-linked nephrogenic diabetes insipidus. J Am Soc Nephrol 2000;11:1044–1054. Bagnasco SM. Gene structure of urea transporters. Am J Physiol Renal Physiol 2003;284:F3–F10. Bagnasco SM, Peng T, Janech MG, Karakashian A, Sands JM. Cloning and characterization of the human urea transporter UT-A1 and mapping of the human Slc14a2 gene. Am J Physiol Renal Physiol 2001;281:F400–F406.

627

Bagnasco SM, Peng T, Nakayama Y, Sands JM. Differential expression of individual UT-A urea transporter isoforms in rat kidney. J Am Soc Nephrol 2000;11:1980–1986. Bernier V, Lagace M, Lonergan M, Arthus MF, Bichet DG, Bouvier M. Functional rescue of the constitutively internalized V2 vasopressin receptor mutant R137H by the pharmacological chaperone action of SR49059. Mol Endocrinol 2004;18:2074–2084. Bichet DG, Arthus M-F, Lonergan M, et al. X-linked nephrogenic diabetes insipidus mutations in North America and the Hopewell hypothesis. J Clin Invest 1993;92:1262–1268. Bichet DG, Birnbaumer M, Lonergan M, et al. Nature and recurrence of AVPR2 mutations in X-linked nephrogenic diabetes insipidus. Am J Hum Genet 1994;55:278–286. Bichet DG, Fujiwara TM. Nephrogenic diabetes insipidus. In: Scriver CR, Beaudet AL, Sly WS, Vallee D, Childs B, Kinzler KW et al., eds. The Metabolic and Molecular Bases of Inherited Disease. New York: McGraw-Hill, 2001; pp. 4181–4204. Bode HH, Crawford JD. Nephrogenic diabetes insipidus in North America: the Hopewell hypothesis. N Engl J Med 1969; 280:750–754. Brown D. The ins and outs of aquaporin-2 trafficking. Am J Physiol Renal Physiol 2003;284:F893–F901. Crawford JD, Bode HH. Disorders of the posterior pituitary in children. In: Gardner LI, ed. Endocrine and Genetic Diseases of Childhood and Adolescence. Philadelphia: W.B. Saunders, 1975; pp. 126–158. Deen PMT, Verdijk MAJ, Knoers NVAM, et al. Requirement of human renal water channel aquaporin-2 for vasopressin-dependent concentration of urine. Science 1994;264:92–95. Deen PMT, Croes H, van Aubel RAMH, Ginsel LA, van Os CH. Water channels encoded by mutant aquaporin-2 genes in nephrogenic diabetes insipidus are impaired in their cellular routing. J Clin Invest 1995;95:2291–2296. Fenton RA, Chou CL, Stewart GS, Smith CP, Knepper MA. Urinary concentrating defect in mice with selective deletion of phloretin-sensitive urea transporters in the renal collecting duct. Proc Natl Acad Sci USA 2004;101:7469–7474. Fenton RA, Stewart GS, Carpenter B, et al. Characterization of the mouse urea transporters UT-A1 and UT-A2. Am J Physiol Renal Physiol 2002;283:F817–F825. Fujiwara TM, Bichet DG. Molecular biology of hereditary diabetes insipidus. J Am Soc Nephrol 2005;16:2836–2846. Galluci E, Micelli S, Lippe C. Non-electrolyte permeability across thin lipid membranes. Arch Int Physiol Biochim 1971;79:881–887. Hoekstra JA, van Lieburg AF, Monnens LA, Hulstijn-Dirkmaat GM, Knoers VV. Cognitive and psychosocial functioning of patients with congenital nephrogenic diabetes insipidus. Am J Med Genet 1996;61:81–88. Holtzman EJ, Kolakowski LF, O’Brien D, Crawford JD, Ausiello DA. A null mutation in the vasopressin V2 receptor gene (AVPR2) associated with nephrogenic diabetes insipidus in the Hopewell kindred. Hum Mol Genet 1993;2:1201–1204. Ito M, Yu RN, Jameson JL. Mutant vasopressin precursors that cause autosomal dominant neurohypophyseal diabetes insipidus retain dimerization and impair the secretion of wild-type proteins. J Biol Chem 1999;274:9029–9037. Kamsteeg EJ, Wormhoudt TA, Rijss JP, van Os CH, Deen PM. An impaired routing of wild-type aquaporin-2 after tetramerization with an aquaporin-2 mutant explains dominant nephrogenic diabetes insipidus. EMBO J 1999;18:2394–2400. Karakashian A, Timmer RT, Klein JD, Gunn RB, Sands JM, Bagnasco SM. Cloning and characterization of two new mRNA isoforms of the rat renal urea transporter: UT-A3 and UT-A4. J Am Soc Nephrol 1999;10: 230–237. Kim Y-H, Kim D-U, Han K-H, et al. Expression of urea transporters in the developing rat kidney. Am J Physiol Renal Physiol 2002;282:F530– F540. Kimoto Y, Constantinou CE. Effects of [1-desamino-8-D-arginine]vasopressin and papaverine on rabbit renal pelvis. Eur J Pharmacol 1990;175:359–362. Klein JD, Gunn RB, Roberts BR, Sands JM. Down-regulation of urea transporters in the renal inner medulla of lithium-fed rats. Kidney Int 2002;61:995–1002.

628

SECTION VII / NEPHROLOGY

Klein JD, Sands JM, Qian L, Wang X, Yang B. Upregulation of urea transporter UT-A2 and water channels AQP2 and AQP3 in mice lacking urea transporter UT-B. J Am Soc Nephrol 2004;15:1161–1167. Kokko JP, Rector FC. Countercurrent multiplication system without active transport in inner medulla. Kidney Int 1972;2:214–223. Kuwahara M, Iwai K, Ooeda T, et al. Three families with autosomal dominant nephrogenic diabetes insipidus caused by aquaporin-2 mutations in the C-terminus. Am J Hum Genet 2001;69:738–748. Kuznetsov G, Nigam SK. Folding of secretory and membrane proteins. N Engl J Med 1998;339:1688–1695. Lucien N, Sidoux-Walter F, Olives B, et al. Characterization of the gene encoding the human Kidd blood group urea transporter protein: evidence for splice site mutations in Jknull individuals. J Biol Chem 1998;273:12,973–12,980. Lucien N, Sidoux-Walter F, Roudier N, et al. Antigenic and functional properties of the human red blood cell urea transporter hUT-B1. J Biol Chem 2002;277:34,101–34,108. Marr N, Bichet DG, Lonergan M, et al. Heteroligomerization of an aquaporin-2 mutant with wild-type aquaporin-2 and their misrouting to late endosomes/lysosomes explains dominant nephrogenic diabetes insipidus. Hum Mol Genet 2002;11:779–789. McIlraith CH. Notes on some cases of diabetes insipidus with marked family and hereditary tendencies. Lancet 1892;2:767, 768. McKusick VA. Online Mendelian Inheritance in Man OMIM, (TM). McKusick-Nathans Institute for Genetic Medicine, Johns Hopkins University (Baltimore, MD) and National Center for Biotechnology Information, National Library of Medicine (Bethesda, MD), 2000. World Wide Web URL: http://www.ncbi.nlm.nih.gov/omim/. Morello J-P, Bichet DG. Nephrogenic diabetes insipidus. Annu Rev Physiol 2001;63:607–630. Mulders SB, Knoers NVAM, van Lieburg AF, et al. New mutations in the AQP2 gene in nephrogenic diabetes insipidus resulting in functional but misrouted water channels. J Am Soc Nephrol 1997;8:242–248. Mulders SM, Bichet DG, Rijss JPL, et al. An aquaporin-2 water channel mutant which causes autosomal dominant nephrogenic diabetes insipidus is retained in the golgi complex. J Clin Invest 1998;102:57–66. Nakayama Y, Naruse M, Karakashian A, Peng T, Sands JM, Bagnasco SM. Cloning of the rat Slc14a2 gene and genomic organization of the UT-A urea transporter. Biochem Biophys Acta 2001;1518:19–26. Niaudet P, Dechaux M, Leroy D, Broyer M. Nephrogenic diabetes insipidus in children. In: Czernichow P, Robinson AG, eds. Frontiers of Hormone Research. Basel: Karger, 1985; pp. 224–231. Nielsen S, Frokiaer J, Marples D, Kwon ED, Agre P, Knepper M. Aquaporins in the kidney: from molecules to medicine. Physiol Rev 2002;82:205–244. Ohzeki T, Igarashi T, Okamoto A. Familial cases of congenital nephrogenic diabetes insipidus type II: remarkable increment of urinary adenosine 3′,5′-monophosphate in response to antidiuretic hormone. J Pediatr 1984;104:593–595. Okusa MD, Crystal LJT. Clinical manifestations and management of acute lithium intoxication. Am J Med 1994;97:383–389. Olives B, Mattei M-G, Huet M, et al. Kidd blood group and urea transport function of human erythrocytes are carried by the same protein. J Biol Chem 1995;270:15,607–15,610. Olives B, Neau P, Bailly P, et al. Cloning and functional expression of a urea transporter from human bone marrow cells. J Biol Chem 1994;269:31,649–31,652. Pallone TL, Turner MR, Edwards A, Jamison RL. Countercurrent exchange in the renal medulla. Am J Physiol Regul Integr Comp Physiol 2003;284:R1153–R1175. Ranade K, Wu KD, Hwu CM, et al. Genetic variation in the human urea transporter-2 is associated with variation in blood pressure. Hum Mol Genet 2001;10:2157–2164. Reeves WB, Andreoli TE. Nephrogenic diabetes insipidus. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The Metabolic and Molecular Bases of Inherited Disease. New York: McGraw-Hill, 1995; pp. 3045–3071.

Rittig R, Robertson GL, Siggaard C, et al. Identification of 13 new mutations in the vasopressin-neurophysin II gene in 17 kindreds with familial autosomal dominant neurohypophyseal diabetes insipidus. Am J Hum Genet 1996;58:107–117. Sands JM. Molecular approaches to urea transporters. J Am Soc Nephrol 2002;13:2795–2806. Sands JM. Molecular mechanisms of urea transport. J Membr Biol 2003;191:149–163. Sands JM. Urea transporters. Annu Rev Physiol 2003;65:543–566. Sands JM. Renal urea transporters. Curr Opin Nephrol Hypertens 2004;13:525–532. Sands JM, Gargus JJ, Fröhlich O, Gunn RB, Kokko JP. Urinary concentrating ability in patients with Jk(a-b-) blood type who lack carriermediated urea transport. J Am Soc Nephrol 1992;2:1689–1696. Sands JM, Layton HE. Urine concentrating mechanism and its regulation. In: Seldin DW, Giebisch G, eds. The Kidney: Physiology and Pathophysiology. Philadelphia: Lippincott, Williams and Wilkins, 2000; pp. 1175–1216. Sands JM, Nonoguchi H, Knepper MA. Vasopressin effects on urea and H20 transport in inner medullary collecting duct subsegments. Am J Physiol Renal Physiol 1987;253:F823–F832. Sidoux-Walter F, Lucien N, Nissinen R, et al. Molecular heterogeneity of the Jknull phenotype: expression analysis of the Jk(S291P) mutation found in Finns. Blood 2000;96:1566–1573. Stephenson JL. Concentration of urine in a central core model of the renal counterflow system. Kidney Int 1972;2:85–94. Tamarappoo BK, Verkman AS. Defective aquaporin-2 trafficking in nephrogenic diabetes insipidus and correction by chemical chaperones. J Clin Invest 1998;101:2257–2267. Timmer RT, Klein JD, Bagnasco SM, et al. Localization of the urea transporter UT-B protein in human and rat erythrocytes and tissues. Am J Physiol Cell Physiol 2001;281:C1318–C1325. Timmer RT, Sands JM. Lithium intoxication. J Am Soc Nephrol 1999;10:666–674. Uchida S, Sohara E, Rai T, Ikawa M, Okabe M, Sasaki S. Impaired urea accumulation in the inner medulla of mice lacking the urea transporter UT-A2. Mol Cell Biol 2005;25:7357–7363. van Lieburg AF, Knoers NVAM, Monnens LAH. Clinical presentation and follow-up of 30 patients with congenital nephrogenic diabetes insipidus. J Am Soc Nephrol 1999;10:1958–1964. van Os CH, Deen PM. Aquaporin-2 water channel mutations causing nephrogenic diabetes insipidus. Proc Assoc Am Physicians 1998;110: 395–400. Waring AG, Kajdi L, Tappan V. Congenital defect of water metabolism. Am J Dis Child 1945;69:323–325. Weil A. Ueber die hereditare form des diabetes insipidus. Arch Klin Med 1908;93:180–290. Weil A. Ueber die hereditare form des diabetes insipidus. Virchows Arch Path Anat Physiol Klin Med 1884;95:70–95. Wieth JO, Funder J, Gunn RB, Brahm J. Passive transport pathways for chloride and urea through the red cell membrane. In: Bolis K, Bloch K, Luria SE, Lynen F, eds. Comparative Biochemistry and Physiology of Transport. Amsterdam: Elsevier/North-Holland, 1974; pp. 317–337. Williams RM, Henry C. Nephrogenic diabetes insipidus transmitted by females and appearing during infancy in males. Ann Intern Med 1947;27:84–95. Yang B, Bankir L, Gillespie A, Epstein CJ, Verkman AS. Urea-selective concentrating defect in transgenic mice lacking urea transporter UT-B. J Biol Chem 2002;277:10,633–10,637. You G, Smith CP, Kanai Y, Lee W-S, Stelzner M, Hediger MA. Cloning and characterization of the vasopressin-regulated urea transporter. Nature 1993;365:844–847. Zhang C, Sands JM, Klein JD. Vasopressin rapidly increases the phosphorylation of the UT-A1 urea transporter activity in rat IMCDs through PKA. Am J Physiol Renal Physiol 2002;282:F85–F90.

61 Glomerulonephritis and Smad Signaling HUI Y. LAN AND RICHARD J. JOHNSON SUMMARY Transforming growth factor-β and its signaling pathway, Smads, is the major mechanism mediating fibrosis. However, the involvement of other signaling pathways and the interplay among these pathways in fibrosis remain largely unclear. In this chapter, we focus on the current understanding of the molecular basis of transforming growth factor-β signaling and its crosstalk pathways in renal fibrosis, which may be also applicable to a wide arrange of other disease conditions involving fibrosis. In addition, results from current studies also suggest that Smad signaling may be a final common pathway of fibrosis and targeting the Smad pathway may provide a new therapeutic strategy for tissue and/or organ fibrosis. Key Words: Diabetes; fibrosis; gene therapy; hypertension; kidney disease; TGF-β; signaling pathways; Smads.

INTRODUCTION Fibrosis is the final common pathway leading to end-stage disease in a variety of tissues and organs including heart, liver, lung, skin, and kidney, regardless of the initial causes of damage. The common pathological feature of fibrosis is the accumulation of extracellular matrix (ECM) in tissues. ECM is made from collagen types I, III, and IV, fibronectin, and laminin and this response is mediated by multiple factors, involving several signaling pathways. Reports have identified transforming growth factor (TGF)-β and its signaling pathway, Smads, as one of the major mechanisms mediating fibrosis. The involvement of other signaling pathways and the interplay among different pathways to produce fibrosis remain largely unclear. This chapter focuses on the understanding of the molecular basis of TGF-β signaling and its crosstalk pathways in producing fibrosis in the kidney. These pathways may also be applicable to a wide range of other conditions including pulmonary fibrosis, hepatic fibrosis and cirrhosis, and cardiovascular sclerosis.

TGF-β AND RENAL FIBROSIS TGF-β, a multifunctional cytokine with fibrogenic properties, has long been considered a key mediator in the pathogenesis of renal fibrosis. Increased TGF-β expression is a feature of many human and experimental renal diseases, including diabetic and

From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

hypertensive nephropathy. TGF-β stimulates ECM deposition by increasing the synthesis of ECM proteins and acting to inhibit their degradation. In addition, TGF-β mediates renal fibrosis through the transformation of tubular epithelial cells to ECM-producing myofibroblasts. The important role of TGF-β in producing renal fibrosis is demonstrated by the finding that fibrosis can be induced by the overexpression of TGF-β1 within the normal rat kidney. This fibrosis can be prevented or ameliorated by blocking TGF-β activity with a neutralizing TGF-β antibody, decorin, or by antisense strategies. In a number of experimental kidney disease models TGF-β plays a central role in renal fibrosis.

TGF-β SIGNALING The discovery that Smad proteins are the intracellular mediators and regulators of TGF-β signaling has provided important insights into the specific mechanisms by which TGF-β mediates the fibrogenic response. As shown in Fig. 61-1, TGF-β exerts its biological effects by signaling through the type-I and -II serine/ threonine kinase receptors, TβRI and TβRII. TGF-β binds to receptor II, which then recruits and phosphorylates TβRI. The activated TβRI directly initiates signals to downstream intracellular substrates, receptor-regulated Smads (R-Smads) including Smad2 and -3. Activated R-Smads interact with the common partner, Smad4 (Co-Smad), to form a heteroligomer and these complexes are translocated into the nucleus to regulate target gene expression and initiate the fibrogenic process. Activation of the TGF-β signaling pathway can also lead to the expression of inhibitory Smads (I-Smads) including Smad6 and -7. In mammals, Smad2 and -3 are TGF-β/activin-specific R-Smads, whereas Smad1/5/8 are the bone morphogenetic protein (BMP)-specific R-Smads. Smad4 is the only one that functions as a Co-Smad. Smad6 (preferential inhibitor of the signaling processes initiated by BMP) and Smad7 (a potent inhibitor of signaling initiated by TGF-β or BMP) both act as inhibitory factors (I-Smads) and function as antagonists of R-Smads and Co-Smad signaling. It has been shown that Smad7 can act by specifically inhibiting Smad2 and -3 phosphorylation. The inhibition results from blocking their access to TβRI or by causing degradation of TβRI. Thus, overexpression of Smad7 may be a strategy that will terminate TGF-β signaling, thereby blocking TGF-β-mediated fibrosis. The structure of the Smad family is highly conserved. Smad2 and -3 are the two R-Smads that are activated by TβRI. They are structurally very similar, with 91% of their amino acid sequences identical. Both contain an N-terminal “mad homology 1” domain,

629

630

SECTION VII / NEPHROLOGY

Figure 61-1 TGF-β signaling and fibrosis. Binding of TGF-β to receptor II causes phosphorylation of TβRI that results in phosphorylation of receptor-associated Smads (R-Smads), Smad2 and -3. Activated R-Smads form heteroligomers with the common partner Smad4 to translocate into the nucleus and regulate target gene expression including ECM genes and inhibitory Smad6 and -7. These I-Smads inhibit Smad2 and -3 phosphorylation by blocking their access to TβRI or by causing degradation of TβRI, thereby blocking TGF-β-mediated fibrosis.

which has DNA-binding activity, and a C-terminal mad homology 2 (MH2) domain that is responsible for translocation of the Smad complex into the nucleus in which it regulates transcription of target genes. TβRI-mediated phosphorylation of the C-terminal sequence of R-Smads, SSXS, leads to Smad2 and -3 activation. Smad4 lacks the SSXS sequence and therefore cannot be phosphorylated by the TβRI. The interaction between R-Smads (Smad2/3) and Co-Smad (Smad4) is primarily mediated by the mad homology 2 domain. Once within the nucleus, the Smads can function as transcriptional transactivators or coactivators, and may also directly bind to DNA, resulting in activation of the target genes that are the next step in producing fibrosis.

TGF-β SIGNALING IN RENAL FIBROSIS It is clear that TGF-β mediates tissue fibrosis through Smad signaling pathway. Many ECM genes contain activator protein-1 binding sites in their regulatory regions; their induction by TGF-β has been shown to be Smad3 dependent.This conclusion is further supported by the finding that mice that are null for Smad3 are protected against radiation-induced fibrosis on the skin. In the context of fibrosis in the kidney, both Smad2 and -3 are strongly activated in human diabetic kidney (Fig. 61-2) or in the rat following obstruction of the kidney (Fig. 61-3). Activation of Smad2 and -3 within the kidney, therefore, can contribute to the development of glomerulosclerosis, tubulointerstitial fibrosis, and vascular sclerosis (Figs. 61-2 and 61-3), indicating that this TGF-β signaling system is a critical pathway leading to tissue fibrosis. In vitro, it has been shown that TGF-β-induced collagen matrix production by mesangial cells and tubular epithelial cells is mediated by Smad2 and -3. In addition, glucose-mediated collagen matrix production by mesangial

cells, tubular epithelial cells, endothelial cells, or vascular smooth muscle cells is also dependent on the activation of Smad2 and -3. Activation of Smads is required for TGF-β-induced, tubular epithelial cell–myofibroblast transition, an important process for tubulointerstitial fibrosis. The functional role of Smad2 and -3 in initiating fibrosis is delineated by results demonstrating that blockade of Smad2/3 activation by overexpression of Smad7 results in abrogation of bleomycin-induced lung fibrosis in mice and in fibrosis following obstructive kidney disease in rats. All of these reports provide strong evidence supporting a critical role for TGF-β/Smad signaling in the development of tissue scarring. The precise role of individual Smads and the interplay among them in initiating the fibrotic process remains largely unknown. Although both Smad2 and -3 are structurally very similar, their modes of actions and the phenotypes of Smad2 and -3 are distinct. Smad3 regulates gene activity directly by binding to DNA, whereas Smad2 activates transcription by binding to other transcription factors that, in turn, can bind DNA and modulate gene activity. The distinct activities of Smad2 and -3 are highlighted by the finding that Smad2–/– mice are embryonic lethal, whereas Smad3–/– mice are viable but have impaired immunity. Similarly, different genes are regulated by different Smads or combinations of Smads in vitro. For example, in mouse embryo fibroblasts, the induction of matrix metalloproteinase-2 by TGF-β is selectively dependent on Smad2, whereas the induction of c-fos is dependent on Smad3, and both Smad2 and -3 are required for inducing the plasminogen activator inhibitor-1. Interestingly, the absence of Smad3 has no effect on TGF-β-induced inhibition of mammary epithelial cell proliferation, but without Smad3 there is decreased cell apoptosis but no compensatory changes in the expression or

CHAPTER 61 / GLOMERULONEPHRITIS AND SMAD SIGNALING

631

Figure 61-2 Smad2 and -3 activation in diabetic and hypertensive nephropathy. In the normal human kidney, activated Smad2 (A) and -3 (B) cells (black nuclear location) are rare (arrows). In contrast, in diabetic (C) and hypertensive (D) nephropathy, activation (black nuclear location) of Smad2 (C) and -3 (D) are prominent, contributing to glomerulosclerosis, tubulointerstitial fibrosis, and vascular sclerosis. (g, glomerulus; a, artery)

activation of Smad2. These findings suggest that epithelial cell Smad3 is not required for the influence of TGF-β on proliferation, but Smad3 contributes in a nonredundant manner to the induction of apoptosis. Interestingly, TGF-β-induced vascular endothelial growth factor expression by renal tubular epithelial cells depends on Smad3, but not Smad2, whereas TGF-β-induced thrombospondin-1 expression is stimulated by Smad2, but not Smad3. Although both Smad2 and -3 are required for TGF-β-induced tubular mesenchymal transition, Smad2 regulates a loss of an epithelial gene E-cadherin, but Smad3 contributes to de novo expression of a mesenchymal gene, α-SMA. Thus, Smad2 and -3 seem to work in a nonredundant manner, but both may be required in certain pathophysiological processes.

TGF-β-INDEPENDENT SMAD SIGNALING IN RENAL AND VASCULAR FIBROSIS Smad2 and -3 can also be activated by a non-TGF-β-dependent pathway. Advanced glycation end products (AGEs), the molecules that are key mediators of diabetic complications, can also activate Smads directly and independently of TGF-β. For example, AGEs can induce a rapid activation of Smad2 and -3 in tubular epithelial cells, mesangial cells, or vascular smooth muscle cells; activation

is evident at 5 min and peaks by 30 min. This rapid activation of Smads occurs in the absence of TGF-β and a neutralizing TGF-β antibody will not block AGE-induced Smad activation. Importantly, the finding that AGEs, but not TGF-β, can activate Smads in both TβRI and TβRII mutant cells supports the presence of a TGF-β-independent Smad signaling pathway as a mediator of renal and vascular complications of diabetes. The mechanism by which AGEs activate Smad directly and independently of TGF-β has been uncovered; AGEs signal through the receptor for AGEs (RAGE) to induce Smad activation rapidly via the extracellular signal-regulated kinase (ERK)/p38 mitogen-activated protein (MAP)-Smad crosstalk pathway. For example, AGE-induced activation of Smad2/3 can be blocked by a neutralizing anti-RAGE antibody or by specific MAP kinase (MAPK) inhibitors of ERK1/2 (PD98059) and p38 (SB203580). AGEs are able to activate Smad2/3 after exposure for 24 h via the TGF-β-dependent mechanism. Thus, AGEs could mediate diabetic complications directly via the MAPK-Smad signaling pathway or indirectly through the classic TGF-β-Smad signaling pathway. Interestingly, substantial inhibition of AGE-induced Smad activation and collagen production by inhibitors of ERK/p38 MAPK or, to a lesser extent, by an anti-TGF-β antibody indicates that the MAPK-Smad

632

SECTION VII / NEPHROLOGY

Figure 61-3 Overexpression of Smad7 inhibits Smad2 and -3 activation and renal fibrosis in a rat obstructive nephropathy. The left ureter was ligated and the left kidney was transfected with either Smad7 (treated animals) or empty vectors (control animals) in a mixture of Optison followed immediately by ultrasound as described. Smad7 transgene expression was induced by doxycycline in the drinking water for 7 d. In the normal rat kidney, phosphorylated Smad2 (pSmad2 nuclear location) are rare (A) with minor collagen I accumulation in tubulointerstitium (B). In contrast, marked activation of pSmad2 is found in tubulointerstitium (C), contributing to tubulointerstitial fibrosis (D). Significantly, kidney treated with Smad7 substantially prevents Smad2 activation (E) and tubulointerstitial fibrosis (F).

signaling crosstalk pathway is a key mechanism involved in the pathogenesis of AGE-mediated diabetic scarring. These findings also indicate that the Smad crosstalk pathway may play a critical role in development of the end-stage organ or in the tissue scarring characteristic of diabetes. Viewed in this fashion, anti-TGF-β therapy may not be sufficient to block diabetic complications. The importance of MAPK-Smad signaling and crosstalk in initiating fibrosis has been demonstrated by finding a relationship between angiotensin (Ang)-II and Smads in the development of kidney and cardiovascular fibrosis. Preliminary studies demonstrate that, like AGEs, Ang-II-mediated collagen matrix production by both renal and vascular cells is Smad-dependent. Ang-II signals through its type 1 receptor to activate Smad2 and -3 within 5 min by a direct mechanism that is independent of TGF-β but dependent on ERK/p38 and MAPK. Ang-II is also able to activate Smad2 and -3 at 24 h, by a mechanism that is TGF-β dependent. Thus, activation of Smads via MAPK-Smad crosstalk pathways may be critical for developing diabetic- or hypertension-induced kidney and vascular sclerosis as shown in Fig. 61-2. In fact, increased Smad2 and -3 activation has been demonstrated in the scarred tissue that results following myocardial infarction;

this activation was attenuated by Ang-II signals through its type1 receptor antagonists. In an obstructive kidney of rats, Ang-II has been shown to play a role in generating renal fibrosis and blockade of Smad2 and -3 activation by the overexpression of Smad7 substantially inhibits renal fibrosis. Direct evidence for a role of Smads in Ang-II-mediated fibrosis comes from the observation that mice null for Smad3 do not develop Ang-II-induced cardiovascular cell or kidney scarring. Together, these studies constitute strong evidence that there is crosstalk between Ang-II (or AGEs) and TGF-β signaling pathways that result in cardiovascular and renal scarring. These data also demonstrate the importance of a functional interaction between the MAPK and Smad signaling pathways in the process of fibrosis. Smad signaling may be a common pathway for the pathogenesis of tissue scarring, as shown in Fig. 61-4. Mechanisms whereby Smads act as signal integrators to form the crosstalk pathways among fibrogenic molecules have been suggested from a number of studies. This is a complicated area however, because activation of Smads can be regulated by other signaling pathways, such as the MAPK pathway, the Janusactivated kinase/signal transducer and activator of transcription

CHAPTER 61 / GLOMERULONEPHRITIS AND SMAD SIGNALING

633

Figure 61-4 Smad signaling as a common pathway of fibrosis. Although TGF-β signals through TβRs to activate Smads to mediate fibrosis, both AGEs and Ang-II signal through their individual receptors to activate Smads via the MAP kinases (MAPK)-Smad signaling crosstalk pathway, in addition to the classic TGF-β-dependent Smad signaling pathway. Overexpression of the inhibitory Smad7 blocks Smad signaling and fibrosis generally.

pathway, and the nuclear factor-κB pathway. Of these pathways, the interaction between MAPK and Smad signaling pathways is the best delineated. It has been shown that signals derived from growth factor receptors with tyrosine kinase activities are capable of modulating Smad-dependent effects. Presently, the mechanism involves activation of a kinase downstream of mitogen-activated protein kinase (MEK)-1, the upstream activator of the ERK/ MAPK pathway and results in the phosphorylation of Smad2. Phosphorylation of Smad1, -2, and -3 by Ras-activated ERK1/2 in its linker region will inhibit nuclear translocation of these transcription factors. In contrast, the intracellular kinase, mitogenactivated protein kinase kinase (MEKK)-1, can participate in Smad2-dependent transcriptional events. MEKK-1 is also capable of selectively activating Smad2-dependent transcriptional activity in cultured endothelial cells by a mechanism that is independent of TβRI-mediated responses. These responses do not require the presence of the C-terminal SSXS motif of Smad2, (the site of TβRI-mediated phosphorylation) and are positively regulated by the MEK1-ERK pathway. Activation of Smad2 by MEKK-1 results in enhanced Smad2–Smad4 interactions, nuclear localization of the Smad2–Smad4 complex, and the stimulation of Smad transcriptional coactivator interactions. Notably, overexpression of Smad7 can inhibit the MEKK-1-mediated stimulation of Smad2 transcriptional activity. For example, an inhibitor of the ERK1/2 pathway, U0126, or inhibitors of the p38 MAPK pathway, SB203580 and SKF86002, will block TGF-β-induced, aggrecan gene expression during chondrocyte differentiation. Other evidence for crosstalk between the ERK/MAPK pathway

and the Smad pathways is that both pathways are required for TGF-β1-induced, furin gene transactivation. A variety of other kinases have been implicated in TGF-β signaling; jun N-terminal kinase (JNK) is rapidly activated by TGFβ in a Smad-independent manner and JNK can phosphorylate Smad3 outside of the SSXS motif. Smad3 phosphorylation by JNK will facilitate, both Smad3 activation by the TGF-β receptor complex, and its nuclear accumulation. In fact, JNK cooperates with the Smad signaling to induce Smad7 transcription via the activator protein-1 element, and hence inhibits ERK activity. These results indicate that inhibitory Smad7 is negatively regulated by ERK, but positively regulated by JNK. Thus, there is an interdependent relationship between the MAPK and Smad pathways in TGF-β-mediated transcription. Overall, signaling by TGF-β-like factors is regulated in both positive and negative pathways, and is tightly controlled temporally and spatially through multiple mechanisms at the extracellular, membrane, cytoplasmic, and nuclear levels. Positive regulation might prove to be critical for amplification of signaling by TGF-β and involve activation of R-Smads 2/3, whereas negative regulation may play an important role in restriction and termination of TGF-β signaling via activation of I-Smad7.

SMADS AS A CENTRAL PATHWAY OF FIBROSIS AND AS A NOVEL THERAPEUTIC TARGET FOR TISSUE SCARRING As shown in Fig. 61-4, Smad signaling may act as a central pathway that leads to fibrosis regardless of the initial pathogenic

634

SECTION VII / NEPHROLOGY

cause. Because Smad7 can act as a negative regulator of Smad signaling, overexpression of Smad7 might inhibit Smad2 and -3 activation to terminate Smad signaling and block Smad-mediated ECM production. In vitro, overexpression of Smad7 will block the fibrogenic effects of TGF-β in tubular epithelial or mesangial cells of the kidney. In vivo, overexpression of Smad7 using a gene transfer technique can inhibit Smad2 and -3 activation in bleomycininduced TGF-β-dependent lung fibrosis or in a model of obstructive nephropathy in rats. A novel, effective, and inducible gene therapy that specifically blocks TGF-β/Smad signaling and fibrosis in obstructive nephropathy in rats, has been accomplished by transferring a doxycycline-regulated Smad7 using an ultrasound-microbubble-mediated system. This technique substantially increases the transfection rate and transgene expression by up to 1000-fold compared with the naked DNA transfer strategy. Using this technique, investigators transfected a doxycycline-regulated Smad7 gene into the kidney and induced the Smad7 transgene expression to a therapeutic level by controlling the doses of doxycycline in the drinking water. In the rat obstructive nephropathy model (Fig. 61-3), progressive renal fibrosis that is routinely induced by both TGF-β and Ang-II, is blocked by the expression of the doxycycline-induced Smad7 transgene. There also is complete inhibition of Smad2 and -3 activation plus complete inhibition of myofibroblast proliferation and collagen matrix production. In the rat diabetic nephropathy model, renal injury is mediated by multiple factors including AGEs, high glucose, TGF-β and/or Ang II. Overexpression of Smad7, however, results in complete inhibition of Smad2 and -3 activation and collagen matrix accumulation in both glomeruli and tubulointerstitium. Thus, Smad signaling could be a central pathway to organ fibrosis. Targeting Smad signaling could provide a therapeutic strategy to prevent or combat the scarring resulting in an end-stage organ. Overexpression of Smad7 is also able to inhibit immune and inflammatory responses. Ultrasound-microbubble-mediated, doxycycline-induced Smad7 transgene expression results in substantial inhibition of inflammation in the kidney including suppression of inflammatory cytokines (IL-1, tumor necrosis factor-α), adhesion molecules (intercellular adhesion molecule1and vascular cell adhesion molecule-1), chemotactic molecule (osteopontin), and macrophage and T-cell accumulation in the obstructed kidney model in rats. These changes are associated with inhibition of nuclear factor (NF)-κB activation and indicate that Smad7 may have a unique role in initiating antifibrosis and antiinflammatory and immunosuppressive effects. Inhibition of Smad2 and -3 by Smad7 may terminate fibrosis, but the ability of Smad7 to inhibit NF-κB activation may be a key mechanism by which Smad7 could act to resolve renal inflammation. It is possible that overexpression of Smad7 leads to an increase in IkBa, the endogenes inhibitor of NF-κB activity; the mechanism for Smad7 effects would involve inhibition of IkBa degradation resulting in inactivation of the survival factor, NF-κB. This response would result in cell apoptosis and suppression of NFκB-driven inflammation. It should be noted that extensive and uncontrollable overexpression of Smad7 could cause massive cell death through apoptosis, consistent with reports that Smad7 is an inducer of cell apoptosis. This response may be associated with inhibition of a survival factor NF-κB plus the activation of the JNK pathway. Consequently, it will be critical to control the degree of Smad7 transgene expression in order to maintain a physiological balance

among the TGF-β, NF-κB, and JNK signaling crosstalk pathways when using a strategy involving overexpression of Smad7. The degree of Smad7 transgene expression could be controlled by varying the concentrations of doxycycline in vitro or in vivo if the inducible gene is used. Both the in vitro and in vivo results imply that inducible Smad7 gene therapy using the ultrasoundmicrobubble technique might provide a novel, safe, and effective therapeutic strategy for controlling or preventing both inflammatory and fibrotic diseases.

CONCLUSION The discovery of TGF-β signaling through Smads has markedly improved the understanding of the molecular mechanisms that lead to tissue scarring. Identifying the activity of individual Smads may lead to improved understanding of the specific role for each Smad in the pathophysiologic process of fibrosis, inflammation, and even cancer. Identification of specific Smad-signaling crosstalk pathways will enable recognition of the molecular mechanisms and the complicated processes in the pathogenesis of diseases. In organ fibrosis, Smad signaling probably represents a common pathway of tissue scarring. In this case, TGF-β binds directly to its receptor, resulting in activation of Smad2 and -3 to induce extracelllular matrix production, whereas AGEs, Ang-II, and other growth factors/molecules could bind to their individual receptors to activate Smads via the MAPK-Smad crosstalk pathway. These responses ultimately lead to tissue injury. In contrast, overexpression of Smad7 can terminate Smad signaling by blocking Smad2 and -3 activation. This response would inhibit the fibrotic effects of TGF-β, high glucose, AGEs, and Ang-II. These findings suggest that fibrosis is regulated positively by Smad2 and -3, but negatively by Smad7. The ability of Smad7 to block experimental models of renal and lung fibrosis indicates that targeting Smad signaling may provide a new therapeutic strategy for tissue and/or organ fibrosis.

SELECTED REFERENCES Border WA, Noble NA. Evidence that TGF-beta should be a therapeutic target in diabetic nephropathy. Kidney Int 1998;54:1390, 1391. Bottinger EP, Bitzer M. TGF-beta signaling in renal disease. J Am Soc Nephrol 2002;13:2600–2610. Chen R, Huang C, Morinelli TA, Trojanowska M, Paul RV. Blockade of the Effects of TGF-beta1 on Mesangial Cells by overexpression of Smad7. J Am Soc Nephrol 2002;13:887–893. Dixon IM, Hao J, Reid NL, Roth JC. Effect of chronic AT(1) receptor blockade on cardiac Smad overexpression in hereditary cardiomyopathic hamsters. Cardiovasc Res 2000;46:286–297. Flanders KC, Sullivan CD, Fujii M, et al. Mice lacking Smad3 are protected against cutaneous injury induced by ionizing radiation. Am J Pathol 2002;160:1057–1068. Funaba M, Zimmerman CM, Mathews LS. Modulation of Smad2-mediated signaling by extracellular signal-regulated kinase. J Biol Chem 2002;277:41,361–41,368. Isono M, Chen S, Won Hong S, Iglesias-de la Cruz MC, Ziyadeh FN. Smad pathway is activated in the diabetic mouse kidney and Smad3 mediates TGF-beta-induced fibronectin in mesangial cells. Biochem Biophys Res Commun 2002;296:1356–1365. Kavsak P, Rasmussen RK, Causing CG, et al. Smad7 binds to Smurf2 to form an E3 ubiquitin ligase that targets the TGF-beta receptor for degradation. Mol Cell 2000;6:1365–1375. Kretzschmar M, Doody J, Timokhina I, Massagué J. A mechanism of repression of TGFβ/Smad signaling by ongenic ras. Genes Dev 1999;13:804–816.

CHAPTER 61 / GLOMERULONEPHRITIS AND SMAD SIGNALING

Lan HY. Tubular epithelial-myofibroblast transdifferentiation mechanisms in proximal tubule cells. Curr Opin Nephrol Hypertens 2003;12:25–29. Lan HY, Mu W, Tomita N, et al. Inhibition of renal fibrosis by gene transfer of inducible Smad7 using ultrasound-microbubble system in rat UUO model. J Am Soc Nephrol 2003;14:1535–1548. Li JH, Huang XR, Zhu HJ, Johnson RJ, Lan HY. Role of TGF-β Signaling in extracellular matrix production under high glucose conditions. Kidney Int 2003;63:2010–2019. Li J, Huang XR, Zhu HJ, et al. Advanced glycation end products activate Smad signaling via TGF-β dependent and independent mechanisms: Implications for diabetic renal and vascular disease. FASEB J 2004; 18:176–178. Li JH, Zhu HJ, Huang XR, Lai KN, Johnson RJ, Lan HY: Smad7 inhibits fibrotic effect of TGF-β on renal tubular epithelial cells by blocking Smad2 activation. J Am Soc Nephrol 2002;13:1464–1472. Massague J, Chen YG. Controlling TGF-β signaling. Genes Dev 2000;14:627–644.

635

Mazars A, Lallemand F, Prunier C, et al. Evidence for a role of the JNK cascade in Smad7-mediated apoptosis. J Biol Chem 2002;276: 36,797–36,803. Piek E, Ju WJ, Heyer J, et al. Functional characterization of transforming growth factor β signaling in Smad2- and Smad3-deficient fibroblasts. J Biol Chem 2001;276:19,945–19,953. Roberts AB, Piek E, Bottinger EP, Ashcroft G, Mitchell JB, Flanders KC. Is Smad3 a major player in signal transduction pathways leading to fibrogenesis? Chest 2001;120:43S–47S. Terada Y, Hanada S, Nakao A, Kuwahara M, Sasaki S, Marumo F. Gene transfer of Smad7 using electroporation of adenovirus prevents renal fibrosis in post-obstructed kidney. Kidney Int 2002;61: S94–S98. Verrecchia F, Chu ML, Mauviel A. Identification of novel TGF-beta/Smad gene targets in dermal fibroblasts using a combined cDNA microarray/promoter transactivation approach. J Biol Chem 2001;276: 17,058–17,062.

62 Interstitial Nephritis CARLA ZOJA AND GIUSEPPE REMUZZI SUMMARY Chronic nephropathies with higly enhanced glomerular permeability to proteins are accompanied by tubulointerstitial inflammation and scarring and time progression to renal function deterioration. Proteins filtered through the glomerular capillary in excessive amount have intrinsic renal toxicity possibly linked to the subsequent process of proximal tubular reabsorption. Protein overloading of proximal tubular cells regulates transcription of nuclear factor-κB-dependent and nuclear factorκB-independent genes. This forms endothelin-1, chemokines, and cytokines that are secreted into the renal interstitium and incite inflammatory and fibrogenic reaction. Autocrine pathways of activation of tubular epithelial cells contribute to interstitial injury and fibrosis. Key Words: Chemokines; complement; interstitial inflammation; progressive nephropathy; proteinuria; proximal tubular cells; renal scarring; transforming growth factor-β; tubulointerstitial damage; ultrafiltered plasma proteins.

INTRODUCTION Progression to end-stage renal failure is the final common pathway of many forms of glomerular disease, independent of the type of initial insult. Progressive glomerulopathies have persistently high levels of urinary protein excretion and tubulointerstitial lesions in common. There is evidence that the severity of tubular interstitial damage more strongly correlates with the risk of renal failure in comparison with the severity of glomerular lesions. Among the cellular mechanisms that may determine progression of kidney damage, regardless of etiology, excess proteins filtered from the glomerulus because of altered glomerular permeability, may initiate interstitial inflammatory reactions by stimulating the tubular synthesis of exaggerated amounts of vasoactive and inflammatory substances. These substances are released into the renal interstitium (Fig. 62-1). This chapter discusses the tubular cell-dependent pathways causing interstitial inflammation and fibrosis that are triggered by enhanced protein trafficking through the glomerulus to cause renal scarring and loss of function.

From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

CHANGES IN GLOMERULAR HEMODYNAMICS INITIATE KEY EVENTS IN PROGRESSIVE NEPHROPATHY Among the several theories about the pathophysiology of progressive nephropathies, Brenner et al. offered the most consistent unifying hypothesis. They suggested that the surviving nephrons undergo hemodynamic responses that serve to maintain glomerular filtration. These responses cause structural injury and, hence, a self-perpetuating cycle. A rat model of renal ablation was used to clarify how this adaptive response results in pathological mechanisms and nephron loss. With a reduction of renal mass, remnant intact nephrons in rats undergo hypertrophy and a lowering of arteriolar resistance leading to increased glomerular plasma flow. These changes result from a differential decrease in vessel tone, which is less pronounced in the afferent in comparison with the efferent arterioles. An adaptive increase in glomerular capillary hydraulic pressure results, yielding more filtrate per nephron. These changes become detrimental over months and years. Therapies that attenuate these adaptive responses also limit the decline in glomerular filtration rate and the structural damage. Angiotensin converting enzyme inhibitors (ACEi) reduce intraglomerular capillary pressure more than other antihypertensive agents; they also protect the kidneys of rats with renal mass reduction or experimental diabetes from progressive loss of kidney function. Glomerular hypertension leads to progressive renal injury in the long term because it enlarges the radii of the pores perforating the glomerular membrane; the mechanism at least partly involves angiotensin II. The enlargement of the pores impairs the size-selective function of the glomerular barrier and increases protein ultrafiltration.

ULTRAFILTERED PROTEINS AND TUBULOINTERSTITIAL INJURY It had been suspected that proteins abnormally filtered through the glomerular capillary were more than a marker of the severity of renal lesions. These proteins could exhibit renal toxicity and contribute to progressive renal damage. Sequential analysis of renal biopsies of rats with adriamycin nephrosis or age-related proteinuria showed that the extent of glomerular lesions correlated with tubulointerstitial damage, which was preceded by the presence of protein reabsorption droplets in the cytoplasm of proximal tubular cells. These droplets were associated with focal breaks along the tubular basement membrane plus extravasation of the

636

CHAPTER 62 / INTERSTITIAL NEPHRITIS

637

Figure 62-1 Schematic representation of the events that lead to tubulointerstitial inflammation and renal scarring in proteinuric nephropathy.

tubular cell contents into the renal interstitium. Implications from these studies are that protein reabsorption activates proximal tubular cells to promote interstitial infiltration of inflammatory cells and ultimately, renal fibrosis. The most convincing experimental evidence that proteins leaking into the urinary space trigger tubulointerstitial inflammation was provided by models of “overload proteinuria.” This disorder follows repeated injections of albumin leading to an increase in glomerular barrier permeability and massive proteinuria. There are tubular changes consisting of heavy infiltration of macrophages and T-lymphocytes into the renal interstitium. Other evidence that protein reabsorption by proximal tubular cells participates in the development of interstitial inflammation derives from an analysis of the time-course and sites of protein accumulation and interstitial cellular infiltration in two models of proteinuric nephropathies. In rats with 5/6 nephrectomy, albumin, and immunoglobulin (IgG) are accumulated in proximal tubular cells and precede the interstitial infiltration of major histocompatibility complex II-positive cells and macrophages. The infiltrates develop at or near tubules that contain deposits of intracellular IgG. These patterns are also found in the model of passive Heymann nephritis. The conclusion is that the interstitial inflammatory reaction develops at the sites of protein overload, regardless of the type of glomerular injury. In both models there also is a link between excess plasma protein reabsorption in the proximal tubule and the expression of osteopontin, a cytokine that attracts mononuclear cells into the renal interstitium. Osteopontin was detected with IgG and the sites of adjacent infiltrates, suggesting that osteopontin is a mediator of a proximal tubule-dependent inflammatory pathway occurring because of tubular cell protein reabsorption. The primitive amphibian, the axolotl, has no glomerulus so protein uptake by tubular cells can be studied in the absence of glomerular damage. Injection of proteins into the peritoneal cavity of the axolotl causes selective uptake and transient storage of proteins in tubular epithelial cells of nephrons. Protein-loaded tubules

undergo luminal dilatation with deposition of protein sludge plus massive, diffuse accumulation of droplets containing proteins and lipids in the proximal tubular epithelial cells. Progressive focal accumulation of fibrous tissue occurs around the protein-storing tubules. There also is fibronectin and transforming growth factor (TGF)-β in the tubular epithelial cells and interstitial cells.

TUBULAR HANDLING OF PROTEINS Proteins are reabsorbed by proximal tubules via processes that involve binding to the apical pole of these cells followed by internalization and lysosomal degradation of the protein to amino acids and small peptides. The initial step involves receptor-mediated endocytosis in clathrin-coated pits at the base of the microvilli. The coated pits are formed by a process involving clathrin adaptor proteins and other intracellular proteins yielding to endocytic vesicles called endosomes. Proton pumps present in the interior of these vescicles cause acidification and dissociation of ligands from the receptors. The receptors recycle to the luminal surface but the protein ligands are degraded by lysosomal enzymes. Two endocytic receptors, megalin, and cubilin, have a major role in proximal tubule reabsorption of proteins. Megalin is a large, transmembrane glycoprotein, whereas cubilin is a peripherally attached glycoprotein. The two receptors interact to mediate reabsorption of filtered proteins, including carrier proteins that transport several vitamins or lipids. Megalin and cubilin are also coexpressed in the small intestine, the visceral yolk sac and the placental cytotrophoblast, and megalin is present in glomerular podocytes. Dysfunction of megalin or cubilin results in tubular proteinuria and changes in vitamin metabolism because of defective proximal tubular reabsorption of carrier proteins. Megalin is a 600-kDa transmembrane glycoprotein belonging to the LDL receptor family, it is the most abundant endocytic receptor in the proximal tubule epithelium. It acts as a receptor for different protein ligands including albumin, insulin, prolactin, and vitaminbinding proteins. Megalin’s endocytotic function is regulated by

638

SECTION VII / NEPHROLOGY

heterotrimeric G protein-internalization signals (Gαi3, GAIP, and GIPC) that interact with the cytoplasmic tail of megalin. The tail portion of the molecule also contains Src-homology domains, PSD95/Discs-larget20-1 (PDZ) domains and protein kinase phosphorylation sites, suggesting that megalin is involved in signal transduction. In fact, the mitogenic effect of albumin in proximal tubular cells seems to be mediated by activation of phosphatidylinositol 3-kinase and subsequent activation of p70 ribosomal protein S6 kinase. Cubilin is a 460-kD peripheral membrane protein that binds albumin, transferrin, IgG light chains, and receptor-associated proteins. Cubilin has no transmembrane domain, suggesting that megalin mediates the endocytosis and intracellular trafficking of cubilin. There is evidence that the cubilin ligands, transferrin, and apolipoprotein A-I/high density lipoprotein, are not reabsorbed if anti-megalin antibodies or megalin anti-sense oligonucleotides are present to block megalin action. In megalin-deficient mice, transferrin accumulates on the luminal membrane of the proximal tubule cells but it is not internalized. In patients, reduced renal expression of megalin occurs in Dent’s disease when it expresses proteinuria and defective tubular re-uptake of filtered proteins. Disruption of ClC-5, the renal chloride channel, impairs proximal tubule endocytosis and causes a decrease in megalin expression in the proximal tubule. A decrease in megalin expression has been also described in a rat model of autosomal-dominant polycystic kidney disease.

EFFECTOR MECHANISMS MEDIATING PROTEIN OVERLOAD INJURY Highlights of specific mechanisms that link excessive traffic of plasma proteins with interstitial injury have been identified using polarized proximal tubular cells. Protein overloading of cultured proximal tubular cells activates the transcription of genes encoding vasoactive, inflammatory, and fibrogenic molecules. For example, when high concentrations of proteins (delipidated or lipid-enriched albumin, IgG, and transferrin) are added to cultured cells, the synthesis of endothelin-1, a vasoconstrictor peptide involved in progressive renal injury is increased. Endothelin-1 stimulates renal cell proliferation and extracellular matrix production and is chemotactic for monocytes. The monocyte chemoattractant protein-1 (MCP-1) and regulated upon activation, normal T-cell expressed and secreted (RANTES), two chemokines that have potent chemotactic properties for monocytes/macrophages, and T-lymphocytes, are upregulated by albumin and other plasma proteins. Albumin exposure also induces the production of interleukin (IL)-8, a chemokine of the C-X-C family. IL-8 has potent chemotactic activity for neutrophils and lymphocytes. In analysis of kidney biopsies from patients with heavy proteinuria, IL-8 mRNA expression was found to be localized to proximal tubules; nonnephrotic subjects did not have excess tubular IL-8 mRNA. Tubular epithelial cells organized as a continuous polarized cell layer can serve to separate the apical and basolateral cellular compartments. Cultured proximal tubular cells grown on filters in bicameral systems become polarized, develop microvilli on the apical membrane and tight junctions in the basolateral domain. Apical exposure of these cells to albumin leads to increased secretion of endothelin-1 and chemokines mainly toward the basolateral compartment of the cell. Consequently, the model could reveal mechanisms relevant to the tubulointerstitial inflammatory response and structural injury that is observed in response to proteinuria.

Fractalkine is a cell-membrane anchored chemokine that acts as an adhesive molecule to promote adhesion of mononuclear cells expressing the specific receptor, CX3CR1. Fractalkine has a CX3C-containing chemokine module on the top of a mucin-like stalk that is tethered to a transmembrane domain and has an intracytoplasmic tail. In addition, fractalkine may be cleaved to form a soluble chemo-attractant. Besides vascular endothelial cells and epithelial cells from lung and intestine, renal proximal tubular epithelial cells express fractalkine when stimulated by tumor necrosis factor. There is strong expression of fractalkine mRNA by tubules that are adjacent to inflammatory cell infiltrates in kidney biopsies from patients with acute renal allograft rejection. Fractalkine is overexpressed in cultured proximal tubular cells when they are exposed to albumin, as well as in a murine model of protein overload proteinuria. For example, fractalkine mRNA levels in the kidney after protein overload were more than twice as high as control animal kidneys. Treatment of mice with an antibody against the CX3CR1 fractalkine receptor was found to reduce the accumulation of monocytes/macrophages in the renal interstitium. Fractalkine might contribute in proteinuric condition by directing mononuclear cells into the peritubular interstitium presumably by enhancing adhesion to favor interstitial inflammation and progression of renal insufficiency.

IN VITRO EVIDENCE FOR THE TOXICITY OF PROTEINS Molecular mechanisms by which proteinuria can activate chemokine genes include nuclear transcription factors, such as NF-κB. The NF-κB/Rel family includes homodimeric or heterodimeric complexes composed of p50, p52, p65, c-rRel, and RelB proteins. The prototype, NF-κB, is composed of p50–p65 subunits. In resting cells, NF-κB exists in the cytoplasm as an inactive form bound to the protein inhibitor of κB (IkBα). Upon cell activation by cytokines, viruses, or oxidants, IkBα is phosphorylated by the IkB kinase complex and subsequently ubiquitinated and degraded. This process releases NF-κB, which is translocated into the nucleus for binding to specific DNA motifs in gene promoters. Cultured proximal tubular cells exposed to albumin develop a dose-dependent increase in NF-κB activation followed by the upregulation of RANTES, MCP-1, and IL-8; these responses were abrogated by NF-κB inhibitors. Moreover, when IkBα was induced in cells, exposure to albumin suppressed the expression of fractalkine mRNA indicating that NF-κB has a role in the activation of chemokine genes. Reactive oxygen species can also serve as second messengers and activate NF-κB in response to albumin exposure. Both albumin and IgG elicit a rapid and sustained generation of hydrogen peroxide (H2O2) in human proximal tubular cells. Treatment with the antioxidant scavengers or metal chelators, dimethyl-thiourea, or pyrrolidine dithiocarbamate, can prevent H2O2 production and abolishes the enhancement of NF-κB activity that follows exposure of the cells to either protein. Treatment of tubular cells with exogenous H2O2 yielded a subunit pattern of NF-κB that is similar to that observed after protein challenge. In other cellular systems, oxidants are generated following protein kinase C (PKC) activation; activated PKC translocates from the cytoplasm to cell membrane and mediates reactive oxygen species production and NF-κB activation. In protein overloaded tubular cells, inhibitors of PKC prevented H2O2 generation and inhibited NF-κB-DNA binding activity. Oxidant scavengers or PKC inhibitors in turn, can prevent

CHAPTER 62 / INTERSTITIAL NEPHRITIS

639

MCP-1 and IL-8 gene upregulation induced by albumin, indicating that PKC and oxygen radicals generate critical signals for the expression of NF-κB dependent genes when tubular cells are exposed to protein overload.

IN VIVO EVIDENCE FOR PROTEIN TOXICITY Evidence that proteinuria can activate transcription factors and expression of chemokines in vivo is available based upon results from experimental or human progressive nephropathies. In rats with protein-overload proteinuria, there is interstitial inflammation and tubular upregulation of MCP-1 and osteopontin. NF-κB activity is also increased in tubular epithelial cells. In rats with 5/6 nephrectomy, increased urinary protein excretion is associated with an increase in NF-κB activity in the remnant kidneys. There is strong nuclear staining for the p50 subunit of NFκB in proximal tubular cells and in the sparse cells of the interstitium, only weak staining is found in the kidneys of control animals. MCP-1 mRNA upregulation preceded the accumulation of monocytes/macrophages and T-lymphocytes in the remnant kidney interstitium, suggesting that mononuclear cell recruitment may be the result of MCP-1-dependent mechanisms. In other models of proteinuric nephropathies, MCP-1 overexpression in the kidney precedes or coincides with the infiltration of mononuclear cells into the renal interstitium. Finally, administering a neutralizing anti-MCP-1 antibody to rats with tubulointerstitial nephritis was found to decrease macrophage infiltration, suggesting the possibility that MCP-1 is functionally important in eliciting an inflammatory response in the kidney interstitium. If the interstitial inflammatory reaction is a consequence of the filtration and proximal tubular reabsorption of proteins, then limiting this enhanced protein traffic should also limit the biological effect of excessive tubular protein reabsorption and slow renal disease progression. The best means of testing this relationship in experimental animals or humans is by the use of ACEi. ACEi in rats with remnant kidneys was found to reduce urinary protein excretion, to suppress NF-κB DNA-binding activity almost completely and to reduce MCP-1 mRNA expression. The accumulation of mononuclear cells in the renal interstitium was also greatly limited. In an immune model of passive Heymann nephritis, a progressive increase in proteinuria is associated with a remarkable increase in renal NF-κB activity; this response was reversed by the early administration of an ACEi. The decrease in NF-κB activation was associated with downregulation of MCP-1 expression and a reduction in interstitial inflammation. In line with the possibility that NF-κB activation has a role in tubulointerstitial injury in proteinuric rats, rats with adriamycininduced nephropathy exhibited suppression of NF-κB and marked reduction of interstitial monocyte infiltration during chronic treatment with the NF-κB inhibitor, pyrrolidine dithiocarbamate. Likewise, when a recombinant adenovirus vector expressing the truncated form of IkBα that lacked the phosphorylation sites essential for the activation of NF-κB, was injected into renal arteries of rats with proteinuria, NF-κB activation in tubular cells and interstitial infiltration of mononuclear cells was prevented. Interstitial edema and fibrosis were also suppressed. Gene transfer of mutant IkB also suppressed the expression of profibrogenic TGF-β and matrix fibronectin. These results suggest the possibility that gene therapy targeting NF-κB could interrupt the process of tubulointerstitial injury.

Figure 62-2 Section of remnant kidney tissue of rat at 30 d after renal mass reduction, double stained for C3 and MHC class II. Peritubular accumulation of MHC-II-positive cells preferentially occurs in the vicinity of tubular sites of C3 deposition and of intracellular accumulation of C3 in proximal tubular cells.

Analysis of renal biopsy specimens from patients with severe proteinuria revealed NF-κB activation in tubular epithelial cells. The degree of activation was correlated with the magnitude of proteinuria. There also was a concomitant upregulation of proinflammatory chemokines, MCP-1, RANTES, and osteopontin in tubular epithelial cells. Patients with a progressive disease had more intense expression.

THE PIVOTAL ROLE OF COMPLEMENT Among secondary processes that lead to interstitial damage in proteinuric conditions, the activation of complement proteins in the proximal tubule excites proinflammatory responses. Complement factors filtered at the glomerulus of rats with protein overload proteinuria, renal mass ablation or aminonucleoside nephrosis can form deposits (C3, C5b-9) along the tubular lumen and within proximal tubular cells. Complement-depletion provided direct evidence for the deleterious effects of urinary complement components on the tubulointerstitium. C5-b9 can mediate the interstitial inflammation and fibrosis accompanying chronic proteinuric nephropathy. In rats with a severe form of the nephrotic syndrome, a congenital absence of C6 was found to reduce the degree of interstitial inflammation and improve renal function dramatically. In these C6-deficient rats, subtotal nephrectomy had markedly reduced evidence of tubulointerstitial injury and loss of renal function. Consequently, treatments directed at reducing the C5b-9 attack in tubular cells may slow disease progression and facilitate the recovery of renal function. The process of protein accumulation in proximal tubular cells is followed by interstitial infiltration of mononuclear cells that concentrate almost exclusively in regions containing C3 positive proximal tubules (Fig. 62-2). Treatment with ACEi prevents proteinuria and limits tubular accumulation of C3 and IgG as well as interstitial inflammation. C3 is an essential component of both the classical and alternative pathways of complement activation. In vitro, proximal tubular cells exposed to human serum activated to develop C5b-9 membrane attack complex (MAC) neoantigen complexes on the tubular cell surface. These events cause marked cytoskeleton alterations including disruption of the network of actin stress fibers,

640

SECTION VII / NEPHROLOGY

Figure 62-3 Cellular mechanisms of injury via exaggerated plasma protein uptake by proximal tubular cells, resulting in interstitial inflammation and fibrosis.

formation of blebs, and cytolysis. There also is production of superoxide anions, H2O2, and the synthesis of proinflammatory cytokines such as IL-6 and tumor necrosis factor-α. Besides activating complement, proximal tubular epithelial cells exposed to human serum can synthesize complement components including C3. In the kidney, complement is indeed synthesized and detectable at peritubular sites, suggesting that locally synthesized complement participates in tubulointerstitial damage. Incubation of proximal tubular cells with transferrin led to overexpression of C3 mRNA and to both apical and basolateral C3 secretion. Thus, proximal tubule cells exposed to serum can synthesize and release complement components as well as activate them.

TGF-β1 is the most important cytokine leading to renal fibrogenesis; it induces epithelial mesenchymal transformation of tubular epithelial cells. High concentrations of albumin upregulate TGF-β1 production in proximal tubular cells in culture. Other mediators of the fibrogenic response include platelet-derived growth factor and endothelin-1, which can activate α-SMA gene expression. Finally, growth factors such as hepatocyte growth factor and TGF-β1 itself can contribute to the induction of fibrosis in vivo. In diabetic nephropathy, hepatocyte growth factor, and TGF-β were detected in proximal tubular fluid; upregulation of the expression of collagen type I and II as well as fibronectin occurs in interstitial myofibroblasts.

PROFIBROGENIC SIGNALING FROM PROXIMAL TUBULAR CELLS LOADED WITH FILTERED PROTEINS

CONCLUSION AND FUTURE PROSPECTS

In experimental models of progressive renal disease, both the interstitial accumulation of myofibroblasts and phenotypic changes in tubular epithelial cells can contribute to the fibrogenic reaction. First, proinflammatory activation of tubular cells fosters the local recruitment of macrophages and lymphocytes. These cells release TGF-β, platelet-derived growth factor and other cytokines that may transform interstitial cells into myofibroblasts. Proximal tubular epithelial cells can interact with interstitial fibroblasts directly to promote fibrogenesis via the paracrine release of profibrogenic molecules. In vivo evidence that both the inflammatory, cell-dependent pathway and the tubular, paracrine pathway are activated is found in remnant kidneys of rats. When there is proteinuria, cells expressing the myofibroblast-associated marker α-smooth muscle actin (α-SMA), accumulate. Subsequently, focal staining for α-SMA occurs in proximal tubules, supporting the possibility that the same pathway may play a role in the epithelial–mesenchymal transformation or transdifferentiation of proximal tubular cells. Treatment of rats with the remnant kidney lesion using ACEi inhibits excess protein accumulation and interstitial inflammatory cell infiltration, and abrogates abnormal TGF-β1 gene expression in tubular cells and myofibroblast accumulation.

The tubulointerstitial damage that follows an injury to the glomerulus results from a complex interaction of biologically active factors that cause renal inflammation and scarring (Fig. 62-3). Experimental results have established that filtered proteins exert toxic response owing to overreabsorption by proximal tubular cells. The result of this toxicity is the release of mediators into the renal interstitium. Chemokines attract mononuclear cells and TGF-β can stimulate interstitial fibroblasts to deposit extracellular matrix components. Macrophages that accumulate at tubular sites of enhanced protein uptake can synthesize TGF-β and hence will induce tubular epithelial cells to acquire a mesenchymal phenotype. Interstitial inflammation and fibrosis and ultimately progression of kidney damage can be limited by ACEi because they strengthen the glomerular barrier to proteins and limit proteinuria. Experimental and clinical data strongly suggest that remission of this process is achievable in some patients with chronic renal disease. Unfortunately, the delay between starting treatment and remission still leads to progression of many patients to end stage renal disease. For this reason, multi-drug strategies that simultaneously interrupt secondary pathways of progressive renal disease are needed to maximize renoprotection for the majority of patients with proteinuric nephropathies. For example, anti-TGF-β antibody plus chronic ACE inhibition

CHAPTER 62 / INTERSTITIAL NEPHRITIS

arrest proteinuria and renal injury in experimental diabetic nephropathy. This sort of strategy could lead to remission of the disease in patients who do not completely respond to ACE inhibitors.

ACKNOWLEDGMENT The authors would like to thank Dr. Antonella Piccinelli for his invaluable help in preparing the manuscript.

SELECTED REFERENCES Abbate M, Zoja C, Corna D, Capitanio M, Bertani T, Remuzzi G. In progressive nephropathies, overload of tubular cells with filtered proteins translates glomerular permeability dysfunction into cellular signals of interstitial inflammation. J Am Soc Nephrol 1998;9: 1213–1224. Abbate M, Zoja C, Rottoli D, Corna D, Tomasoni S, Remuzzi G. Proximal tubular cells promote fibrogenesis by TGF-beta1-mediated induction of peritubular myofibroblasts. Kidney Int 2002;61:2066–2077. Abbate M, Zoja C, Rottoli D, et al. Antiproteinuric therapy while preventing the abnormal protein traffic in proximal tubule abrogates protein and complement-dependent interstitial inflammation in experimental renal disease. J Am Soc Nephrol 1999;10:804–813. Anderson S, Rennke HG, Brenner BM. Therapeutic advantage of converting enzyme inhibitors in arresting progressive renal disease associated with systemic hypertension in the rat. J Clin Invest 1986;77: 1993–2000. Baldwin AS. The transcription factor NF-kB and human disease. J Clin Invest 2001;107:3–11. Bazan JF, Bacon KB, Hardiman G, et al. A new class of membrane-bound chemokine with a CX3C motif. Nature 1997;385:640–644. Bertani T, Cutillo F, Zoja C, Broggini M, Remuzzi G. Tubulo-interstitial lesions mediate renal damage in adriamycin glomerulopathy. Kidney Int 1986;30:488–496. Bertani T, Zoja C, Abbate M, Rossini M, Remuzzi G. Age-related nephropathy and proteinuria in rats with intact kidneys exposed to diets with different protein content. Lab Invest 1989;60:196–204. Brenner BM, Meyer TW, Hostetter TH. Dietary protein intake and the progressive nature of kidney disease: the role of hemodynamically mediated glomerular injury in the pathogenesis of progressive glomerular sclerosis in aging, renal ablation, and intrinsic renal disease. N Engl J Med 1982;307:652–659. Brunskill NJ. Mechanisms of albumin uptake by proximal tubular cells. Am J Kidney Dis 2001;37:S17–S20. Chakravorty SI, Cockwell P, Girdlestone J, Brooks CJ, Savage COS. Fractalkine expression on human renal tubular epithelial cells: potential role in mononuclear cell adhesion. Clin Exp Immunol 2002;129:150–159. Christensen EI, Gburek J. Protein reabsorption in renal proximal tubule function and dysfunction in kidney pathophysiology. Pediatr Nephrol 2004;19:714-721. David S, Biancone L, Caserta C, Bussolati B, Cambi V, Camussi G. Alternative pathway complement activation induces proinflammatory activity in human proximal tubular epithelial cells. Nephrol Dial Transplant 1997;12:51–56. Donadelli R, Abbate M, Zanchi C, et al. Protein traffic activates NF-kB gene signaling and promotes MCP-1-dependent interstitial inflammation. Am J Kidney Dis 2000;36:1226–1241. Donadelli R, Zanchi C, Morigi M, et al. Protein overload induces fractalkine upregulation in proximal tubular cells through nuclear factor kappaB- and p38 mitogen-activated protein kinase-dependent pathways. J Am Soc Nephrol 2003;14:2436–2446. Eddy AA. Role of cellular infiltrates in response to proteinuria. Am J Kidney Dis 2001;37(Suppl 2):S25–S29. Eddy AA. Interstitial nephritis induced by protein-overload proteinuria. Am J Pathol 1989;135:719–733. Eddy AA, Giachelli CM. Renal expression of genes that promote interstitial inflammation and fibrosis in rats with protein-overload proteinuria. Kidney Int 1995;47:1546–1557.

641

Gomez-Garre D, Largo R, Tejera N, Fortes J, Manzarbeitia F, Egido J. Activation of nuclear factor kB in tubular epithelial cells of rats with intense proteinuria: role of angiotensin II and endothelin-1. Hypertension 2001;37:1171–1178. Gross ML, Hanke W, Koch A, Ziebart H, Amann K, Ritz E. Intraperitoneal protein injection in the axolotl: the amphibian kidney as a novel model to study tubulointerstitial activation. Kidney Int 2002;62:51–59. Jinde K, Nikolic-Paterson DJ, Huang XR, et al. Tubular phenotypic change in progressive tubulointerstitial fibrosis in human glomerulonephritis. Am J Kidney Dis 2001;38:761–769. Johnson DW, Saunders HJ, Baxter RC, Field MJ, Pollock CA. Paracrine stimulation of human renal fibroblasts by proximal tubule cells. Kidney Int 1998;54:747–757. Lou X, Mcquistan T, Orlando RA, Farquhar MG. GAIP, GIPC and Gai3 are concentrated in endocytic compartments of proximal tubule cells: putative role in regulating megalin’s function. J Am Soc Nephrol 2002;13:918–927. Mezzano SA, Barria M, Droguett MA, et al. Tubular NF-kB and AP-1 activation in human proteinuric renal disease. Kidney Int 2001;60: 1366–1377. Morigi M, Macconi D, Zoja C, et al. Protein overload-induced NF-kB activation in proximal tubular cells requires H2O2 through a PKCdependent pathway. J Am Soc Nephrol 2002;13:1179–1189. Nangaku M, Pippin J, Couser WG. C6 mediates chronic progression of tubulointerstitial damage in rats with remnant kidneys. J Am Soc Nephrol 2002;13:928–936. Nomura A, Morita Y, Maruyama S, et al. Role of complement in acute tubulointerstitial injury of rats with aminonucleoside nephrosis. Am J Pathol 1997;151:539–547. Rangan GK, Wang Y, Tay Y-C, Harris DCH. Inhibition of nuclear factorkB activation reduces cortical tubulointerstitial injury in proteinuric rats. Kidney Int 1999;56:118–134. Rangan GK, Pippin JW, Coombes JD, Couser WG. C5b-9 does not mediate chronic tubulointerstitial disease in the absence of proteinuria. Kidney Int 2005;67:492–503. Remuzzi G, Bertani T. Is glomerulosclerosis a consequence of altered glomerular permeability to macromolecules? Kidney Int 1990;38: 384–394. Remuzzi G, Bertani T. Pathophysiology of progressive nephropathies. N Engl J Med 1998;339:1448–1456. Remuzzi G. Nephropathic nature of proteinuria. Curr Opin Nephrol Hypertens 1999;8:655–663. Ruggenenti P, Schieppati A, Remuzzi G. Progression, remission, regression of chronic renal diseases. Lancet 2001;357:1601–1608. Takase O, Hirahashi J, Takayanagi A, et al. Gene transfer of truncated IkBa prevents tubulointerstitial injury. Kidney Int 2003;63: 501–513. Tang S, Leung JCK, Abe K, Wah Chan K, Neng Lai K. Albumin stimulates interleukin-8 expression in proximal tubular epithelial cells in vitro and in vivo. J Clin Invest 2003;111:515–527. Tang S, Sheerin NS, Zhou W, Brown Z, Sacks SH. Apical proteins stimulate complement synthesis by cultured human proximal tubular epithelial cells. J Am Soc Nephrol 1999;10:69–76. Tang WW, Ulich TR, Lacey DL, et al. Platelet-derived growth factor-BB induces renal tubulointerstitial myofibroblast formation and tubulointerstitial fibrosis. Am J Pathol 1996;148:1169–1180. Verroust P, Birn H, Nielsen R, Kozyraki R, Christensen EI. The tandem endocytic receptors megalin and cubilin are important proteins in renal pathology. Kidney Int 2002;62:745–756. Wang S-N, LaPage J, Hirschberg R. Role of glomerular ultrafiltration of growth factors in progressive interstitial fibrosis in diabetic nephropathy. Kidney Int 2000;57:1002–1014. Wang Y, Chen J, Chen L, Tay YC, Rangan GK, Harris DC. Induction of monocyte chemoattractant protein-1 in proximal tubule cells by urinary protein. J Am Soc Nephrol 1997;8:1537–1545. Wang Y, Rangan GK, Tay Y-C, Harris DCH. Induction of monocyte chemoattractant protein-1 by albumin is mediated by nuclear factor kB in proximal tubule cells. J Am Soc Nephrol 1999;10: 1204–1213.

642

SECTION VII / NEPHROLOGY

Yard BA, Chorianopoulos E, Herr D, van der Woude FJ. Regulation of endothelin-1 and transforming growth factor-beta1 production in cultured proximal tubular cells by albumin and heparan sulphate glycosaminoglycans. Nephrol Dial Transplant 2001;16:1769– 1775. Zatz R, Dunn BR, Meyer TW, Anderson S, Rennke HG, Brenner BM. Prevention of diabetic glomerulopathy by pharmacological amelioration of glomerular capillary hypertension. J Clin Invest 1986;77: 1925–1930. Zeisberg M, Strutz F, Muller GA. Renal fibrosis: an update. Curr Opin Nephrol Hypertens 2001;10:315–320.

Zhou W, Marsh JE, Sacks SH. Intrarenal synthesis of complement. Kidney Int 2001;59:1227–1235. Zoja C, Donadelli R, Colleoni S, et al. Protein overload stimulates RANTES production by proximal tubular cells depending on NF-κB activation. Kidney Int 1998;53:1608–1615. Zoja C, Morigi M, Figliuzzi M, et al. Proximal tubular cell synthesis and secretion of endothelin-1 on challenge with albumin and other proteins. Am J Kidney Dis 1995;26:934–941. Zoja C, Benigni A, Remuzzi G. Cellular responses to protein overload: key event in renal disease progression. Curr Opin Nephrol Hypertens 2004;13:31–37.

63 The Pathophysiology of Acute Renal Failure DIDIER PORTILLA, GUR P. KAUSHAL, ALEXEI G. BASNAKIAN, AND SUDHIR V. SHAH SUMMARY

ROLE OF THE VASCULAR ENDOTHELIUM

Acute renal failure (ARF) is a syndrome that can be defined as an abrupt decrease in renal function sufficient to result in retention of nitrogenous waste in the body. ARF can result from a decrease of renal blood flow, intrinsic renal parenchymal diseases, or obstruction of urine flow. There has been little progress in preventing and treating ARF, but with a better understanding of cell injury mechanisms and by carefully defining populations at risk for developing ARF, treatment strategies could be developed. Key Words: ARF; GFR; inflammatory cells; kidney; mitochondria; sublethal cell injury.

INTRODUCTION Acute renal failure (ARF) is a syndrome that can be defined as an abrupt decrease in renal function sufficient to result in retention of nitrogenous waste (e.g., blood urea nitrogen and creatinine) in the body. ARF can result from a decrease of renal blood flow (prerenal azotemia), intrinsic renal parenchymal diseases (renal azotemia), or obstruction of urine flow (postrenal azotemia). The most common intrinsic renal disease that leads to ARF is referred to as acute tubular necrosis (used interchangeably with ARF in this chapter), a clinical syndrome in which there is an abrupt and sustained decline in glomerular filtration rate occurring within minutes to days following an acute ischemic or nephrotoxic insult. Its clinical recognition requires exclusion of prerenal and postrenal causes of azotemia, followed by exclusion of other causes of intrinsic ARF (e.g., glomerulonephritis, acute interstitial nephritis, or vasculitis). Ischemic ARF is associated with a mortality rate of about 50% in hospitalized patients and 70–80% in patients in intensive care units. This rate has essentially remained constant over the last 50 yr. Although the high mortality rate is due in part to the presence of comorbid conditions, there is evidence that ARF is responsible for the high mortality rate. The most common pathophysiological mechanism involved in the development of ARF is the presence of a complex interplay between altered renal hemodynamics, inflammation, and tubular dysfunction (Fig. 63-1). Advances have improved the understanding of the mechanisms involved in the development of ARF.

From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

Blood flow to the kidney accounts for 25% of the cardiac output, the majority going to the kidney cortex with very small amounts going into the medullary portion of the kidney. Measurement of PO2 in the medulla of normal kidneys reveals a relatively hypoxic region (PO2 values as low as 5–15 mmHg). The critically low PO2 coupled with high energy demands of the medullary portion of the kidney results in increased susceptibility of the cortico-medullary portion of the kidney to injury. Following an ischemic insult, total renal blood flow is reduced to 40–50% of normal value during reperfusion both in animal models and in human ischemic ARF. This “no-reflow” phenomenon has been observed after ischemia in other organs as well. Using intravital videomicroscopy of renal microcirculation during the immediate reperfusion period, investigators demonstrated profound and sustained deceleration of blood flow yielding a temporary loss of capillary patency and shifts in blood flow through postischemic capillaries from orthograde to the retrograde direction. These patterns of capillary blood flow in the postischemic kidney may explain the mechanisms of no-reflow phenomenon. Other evidence of abnormal “vascular reactivity” in the ischemic kidney includes a profound loss of acetylcholine-induced vasodilatation of the ischemic renal vasculature, inhibition of vasorelaxation in response to stimuli that generate endothelium-derived relaxing factor, and reduced production of nitric oxide in response to bradykinin. The potential role of endothelial cells in producing ischemia/ reperfusion injury was suggested by the finding that endothelial cells undergo swelling to narrow the lumen. Other support for the ischemia/reperfusion pathophysiologic cause of ARF includes the significant improvement in the peritubular capillary blood flow in the postischemic period in animals that had been transplanted with human umbilical vein endothelial cells and human embryonic kidney cells which stably expressed human endothelial or type-III NO synthase (eNOS). The observation that endothelial cells can rescue the function of an ischemic organ illustrates the principle of cell–cell interaction in which the dysfunction of one cell type affects the functions of other cells. Furthermore, implantation of cells expressing the functional eNOS (even though the cells are not endothelial cells per se) can ameliorate renal dysfunction following acute ischemia, also implicating eNOS in the pathophysiology of ARF.

643

644

SECTION VII / NEPHROLOGY

Figure 63-1 Complex interplay between altered renal hemodynamics, inflammation, and tubular dysfunction.

Study results provide further support for the role of microvascular endothelium in the pathophysiology of ischemic ARF. They demonstrated high circulating levels of von Willebrand factor (a marker for endothelial cell injury) plus increase in F-actin aggregates found in the basolateral aspects of renal microvascular endothelial cells within the first 24 h of ischemic injury. In addition, there were structural changes in vascular endothelial-cadherin at endothelial cell–cell junctions as early as 2 h, which were accompanied by a marked increase in microvascular permeability, resulting in interstitial edema. These observations implicate an important role of endothelial cells and could result in new therapeutic modalities for treatment of ARF.

PATHOLOGICAL ROLE OF INFLAMMATORY CELLS Studies in experimental acute tubular necrosis previously suggested that inflammatory cells did not play an important role in renal tubular injury. Newer studies, however, support a role for inflammatory cells in ARF. Histological evaluation has shown that the accumulation of neutrophils occurs within the first 4 h and is maximal at 24–48 h of ischemia/reperfusion injury to the kidney. Most of the neutrophil accumulation occurs in the outer medulla and in the cortex. Leukocyte depletion achieved by antineutrophil serum protected the kidney, establishing a cause–effect relationship between neutrophil accumulation and tissue injury during ischemia/reperfusion. Studies using pharmacological agents support the idea that decreased neutrophil infiltration leads to prevention of ischemia/reperfusion injury. For example, administration of α-melanocyte-stimulating hormone has been associated with a

reduction of neutrophil infiltration and kidney injury after ischemic/reperfusion. The adherence of leukocytes to the vascular endothelium is an early critical process resulting in the outer medullary congestion present in ischemic ARF. Leukocytes increase endothelial permeability leading to erythrocyte aggregation and edema as well as obstruction of the vasa recta. Neutrophils also release reactive oxygen species, proteases, elastases, myeloperoxidase, and other enzymes that damage tissue. In addition, increases in leukotriene B4 and platelet-activating factor produced by monocytes or platelets can increase vascular permeability and upregulate the expression of adhesion molecules to promote further inflammation. The adhesion of neutrophils to endothelial cells occurs through a complex series of events involving the presence of several classes of adhesion molecules including selectins, mucin and other selectin ligands and integrins, plus members of the Ig superfamily. CD11/ CD18 integrins and intracellular adhesion molecule (ICAM)-1 are also important molecules participating in the pathogenesis of ischemic renal injury. Protection from ischemic/reperfusion injury occurs in animals that are given antibodies to CD11a, CD11b, and ICAM-1 or antisense oligonucleotides to ICAM-1; ICAM-1-deficient mice are also protected. Activation of adenosine A2 receptors on endothelial cells reduces adherence factors and, ultimately, neutrophil accumulation. Hence neutrophil adherence to endothelial cells plays an important role in the pathogenesis of ischemic reperfusion injury. Besides neutrophils, other inflammatory cells participate in ischemia/reperfusion injury. Biopsies of human kidneys with

CHAPTER 63 / THE PATHOPHYSIOLOGY OF ARF

acute tubular necrosis reveal the presence of lymphocytes, but whether these cells directly participate in organ injury or whether macrophages and T cells appear only in the recovery phase of ARF is controversial. Indirect evidence for a role for T lymphocytes in ischemic ARF includes studies showing that lymphocyterelated cytokines are upregulated in the postischemic kidney. Moreover, CD4/CD8 knockout mice are protected from ischemia/ reperfusion injury. In addition, administration of a selectin ligand resulted in attenuated postischemic renal dysfunction and decreased mortality. In contrast to these reports, mice that are deficient in the recombination-activating gene, RAG-1, have no T- or B-cells and do not produce immunoglobulins or T-cell proteins, but still experience tubular necrosis and neutrophil infiltration when subjected to ischemia-reperfusion. Taken together, the studies in humans or rodent models of ARF support the idea that endothelial cell injury is accompanied by increased leukocyte and perhaps T-cell-endothelial cell interactions that compromise blood flow to the corticomedullary junction.

MECHANISMS OF TUBULAR CELL INJURY SUBLETHAL CELL INJURY The apical brush border of the proximal tubule is a site of early morphological change, with patchy loss of apical microvilli that occurs within minutes of acute renal ischemia. Subsequently, there is complete loss of the apical microvilli that are shed into the tubular lumen. These same morphological alterations of the proximal tubule brush border occur in the postischemic renal allografts of patients. Alterations in the actin cytoskeleton are dramatic. In in vivo as well as in vitro models, there is a disassociation of the heterodimeric capping protein from the actin cytoskeleton during renal ischemia. Dephosphorylation of ezrin and actin depolimerizing factor have been observed in hypoxic tubules and could account for the loss of the cytoskeletal support of the microvillar surface membrane with subsequent microvillar collapse following acute ischemic injury. In addition to the microvillar actin disruption, there is loss of the cortical actin cytoskeleton response to ischemia in vivo or ATP depletion in vitro. Investigators also found conversion of G-actin to F-actin and aggregation of F-actin molecules in the cytoplasm of the cell; redistribution of ankyrin and fodrin, two proteins that anchor NaK-ATPase; redistribution of NaK-ATPase to the apical domain; and inactivation of RhoGTPase with disruption of the tight junction complex during ATP depletion in vitro. Following ATF depletion, plasma membrane protrusions or blebs containing cytosol and endoplasmic reticulum are formed, reflecting disruption of volume regulation mechanisms and cytoskeleton organization. The multiple alterations in renal epithelial cell cytoskeletal structure and its surface membrane polarity may contribute to the decrease in the glomerular filtration rate of ARF. Increased paracellular permeability could be the result of changes in tight junctions whereas loss of the integrity of the epithelial cell layer from sloughing of viable cells may contribute to back-leak of the glomerular filtrate. Finally, the loss of cell polarity and the redistribution of β1 integrin impairs cell adhesion, resulting in clumping of sloughed and nonsloughed cells within the tubular lumen, leading to tubular obstruction. LETHAL INJURY: APOPTOTIC AND NECROTIC PATHWAYS In ARF induced by ischemia, endotoxemia, nephrotoxins, or other causes, renal tubular epithelial cells die during a catastrophic breakdown of regulated cellular homeostasis known as necrosis. In freshly isolated proximal tubules subjected to

645

hypoxia/reoxygenation, investigators described the formation of pathological plasma membrane pores, or so-called “death channels,” these pores can be blocked by glycine at physiological levels. Late in hypoxic stress, an irreversible state of proximal tubule cell injury develops, signified by lysosomal disruption, bleb coalescence and subsequent bleb rupture, followed by loss of membrane permeability. The abrupt failure of the plasma membrane permeability barrier releases cytosolic contents and a collapse of all of the plasmalemmal and electrical and ionic gradients. In this form of cell death, cells swell and there is loss of plasma membrane integrity and rupture of cells releasing a variety of inflammatory mediators. The modern era of cell death began with the landmark publication by Kerr, Wyllie, and Currie in which they coined the term apoptosis and made a distinction between necrosis and apoptosis based on morphological criteria. In apoptosis, cells shrink, lose their microvilli and cell junctions, and explode into a series of membrane-bound, condensed apoptotic bodies. These bodies are phagocytosed by adjacent viable cells with minimal leakage of the contents of the dead cell, thereby invoking no inflammation. Because apoptosis does not evoke inflammation, it is an important mechanism in embryogenesis and normal tissue turnover. One of the major advances in understanding cell death has been the recognition that the pathways traditionally associated with apoptosis can be critical in determining the type of cell injury associated with necrosis; the same insult may result in mild injury resulting in apoptosis or severe injury, producing necrosis. Thus, the pathway is determined by both the nature and severity of insults. It appears likely that the pathways that lead to apoptotic or necrotic cell death are activated almost simultaneously but there are some common pathways. The important role that apoptosis plays in renal function during ischemic ARF was highlighted by an infusion of GTP before inducing ischemic reperfusion injury. There was substantial functional protection and amelioration of apoptosis but not necrosis. Administration of guanosine also prevented the expression of p53, whereas a p53 inhibitor, pifithrin, selectively modified apoptosis and ameliorated renal function if it was given within 14 h of the initial insult. These results suggest that amelioration of apoptosis rather than necrosis will account for the major mechanism that protects renal function during ischemic ARF.

DNA FRAGMENTATION AND ENDONUCLEASES IN RENAL TUBULAR EPITHELIAL CELL INJURY Several reports have demonstrated chromatin condensation, the morphological hallmark of apoptosis, in models of ARF including ischemia/reperfusion injury. However, much of the evidence for the activation of apoptotic mechanisms in cell injury is based on the demonstration of endonuclease activation resulting in oligonucleosome-length DNA fragmentation (approx 200 bp) as one of the biochemical hallmarks of apoptosis. For example, DNA fragmentation in the kidney cortex, or the typical ladder pattern of DNA, has been detected within 12 h after beginning reperfusion. Key questions are whether endonuclease activation is related to cell death and whether an attempt to halt this process would prevent cell death. The 200 bp ladder, commonly used because of its simplicity, actually measures very late events and only approx 40 double-strand DNA breaks per cell have been shown to be lethal. The role of endonucleases in kidney epithelial cell injury is linked to the biochemical and cellular mechanisms of oxidant injury. For example, hydrogen peroxide in the presence of iron contained in

646

SECTION VII / NEPHROLOGY

DNA yields site-specific generation of hydroxyl radicals that damage DNA. In hydrogen peroxide-induced injury, endonuclease activation is an early event leading to DNA fragmentation, and endonuclease inhibitors can prevent hydrogen peroxideinduced DNA strand breaks, DNA fragmentation, and cell death. Endonuclease activation may be important in reactive oxygen mechanisms leading to tubular cell injury. Subsequently, it was demonstrated that hypoxia/reoxygenation injury of freshly isolated rat proximal tubules results in DNA strand breaks and nuclear DNA fragmentation that precedes cell death. This injury increased DNA degrading activity and this new activity was found in molecules with an apparent molecular mass of 15 kDa. Endonuclease inhibitors protected against DNA damage induced by hypoxia/reoxygenation injury and provided significant protection against cell death. There also is a role of endonuclease in a model of chemical hypoxia. Taken together, endonuclease activation resulting in DNA damage and cell death is an integral part of hypoxia/reoxygenation injury. Despite unequivocal evidence of endonuclease activation, the morphologic features of apoptosis including chromatin condensation were not observed. This result is consistent with studies indicating that chromatin condensation and DNA fragmentation may be triggered through separate metabolic pathways. Cells with morphological features of either apoptosis or necrosis will have evidence of endonuclease activity. ENDONUCLEASES IN THE KIDNEY Limited information is available regarding the identity of the endonuclease(s) responsible for DNA fragmentation in the kidney. Evidence for two major endonucleases exists in kidneys and kidney cells, a 15 kDa endonuclease just discussed and a 30–34 kDa DNase I-like endonuclease. The 30 kDa is mainly a cytoplasmic enzyme, whereas the 15 kDa endonuclease is located in the nuclei. Because the 30 kDa enzyme is similar to a DNase I by its biochemical characteristics, the role of DNase I in renal injury is being examined. The activity of 30 kDa endonuclease is increased during ischemia/reperfusion injury in the rat kidney and at least in vitro renal tubular epithelial cells that had been transfected with a phosphorothioated DNase I antisense construct were protected against DNA damage and cell death.

CASPASES IN RENAL TUBULAR EPITHELIAL CELL INJURY A specific class of proteases, the “caspases” (cysteine aspartatespecific proteases), were found to be involved in apoptosis in genetic studies of the nematode Caenorrhabditis elegans. Caspases are a family of structurally related cysteine proteases that play a central role in the execution of apoptosis. When cells are subjected to a pro-apoptotic stimulus, the procaspases are proteolytically processed to the active forms; at least fourteen caspases encoded by distinct genes have been cloned and sequenced in mammals, caspase-2, -8, -9, and -10 have large prodomains and initiate the activation of downstream caspases. Caspase-3, -6, and -7 contain smaller domains that have been identified as effecter or executioner caspases. The executioner caspases are the major active caspases detected in apoptotic cells and are widely regarded as mediators of apoptosis by cleaving and inactivating intracellular proteins that are essential for cell survival and proliferation. The specificity of downstream executioner caspases to cleave cellular proteins is unique. For example, following activation, caspase-3 primarily recognizes DEVD or DMQD tetrapeptide sequences whereas caspase-6 recognizes and cleaves the VEID tetrapeptide sequence after the aspartate residue in the target proteins.

There is increasing evidence for a role for caspases in hypoxic renal tubular cell injury. Exposure of freshly isolated renal tubular epithelial cells to hypoxia results in caspase activation, cell membrane damage, and necrotic cell death. A pan-caspase inhibitor attenuates the hypoxia-induced increase in caspase activity in these cells and provides protection against hypoxia-induced cell membrane damage, as determined by release of lactic dehydrogenase from the cells. Chemical hypoxia induced by treatment with antimycin A increases caspase activity that precedes DNA damage and cell death. Caspase inhibitors prevent hypoxia-induced DNA fragmentation as measured either by agarose gel electrophoresis or by in situ labeling of cell nuclei. Partial ATP depletion results in marked increase in activation of caspase-8 and the caspase inhibitors provide marked protection against cell death. Activation of caspase-3 by hypoxia or ATP depletion was accompanied by the translocation of Bax, a member of the Bcl-2 family, from the cytosol to the mitochondria and the release of cytochrome-c from mitochondria to the cytosol. There is differential regulation of caspase activities in kidneys subjected to ischemia/reperfusion injury. Caspase-3 activity was significantly increased in the rat and murine models of renal ischemia/reperfusion injury whereas administration of the pancaspase inhibitor, Z-VAD-FMK, at the time of reperfusion significantly prevented caspase-1 and -3 activities and provided marked protection against ischemic ARF. A rat model of ischemia/reperfusion injury indicated that prolonged ischemia induces proapoptotic mechanisms, including increases in the Bax/Bcl-2 ratio, caspase-3 expression, poly(ADP-ribose) polymerase cleavage, DNA fragmentation, and apoptotic cell number in renal proximal and distal tubules. Besides caspase-3, the proinflammatory caspase-1 is involved; it participates in the proteolytic cleavage of the precursor forms of the proinflammatory cytokines, IL-I β and -18, resulting in the active forms of these cytokines. Because caspase-1 mediated formation of IL-1β and -18 is associated with inflammation in renal ischemia/reperfusion, caspase-1 may play an important role in this form of injury. Results from caspase-1–/– mice have been inconsistent.

MITOCHONDRIA: CENTRAL COORDINATORS OF CELL DEATH Mitochondria can play a key role in the decision of cells to undergo apoptosis or necrosis by regulating the apoptotic process (Fig. 63-2). When mitochondrial ATP formation by oxidative phosphorylation ceases, anoxic and ischemic injury result. In the case of the proximal tubule, glycolytic capacity is limited and the lack of hexokinase further limits the cell’s potential to generate ATP from sources other than fatty acids. Uncoupled respiration has been postulated as a potential mechanism contributing to cell death because it represents a more severe form of metabolic disruption to mitochondria than the inhibition of ATP production. Uncoupling causes collapse of pH and electrical gradients across the mitochondrial inner membrane and activates the mitochondrial F1F0 ATPase. Indirect evidence for the presence of uncoupling during ischemic ARF includes the presence of an increased expression of uncoupling protein-2 detected in a microarray analysis of ischemic kidney cortex tissue. Profound inhibition of fatty acid oxidation and increased accumulation of long-chain fatty acids and long-chain acylcarnitines have been demonstrated in ischemic ARF in vivo and in freshly isolated tubules responding to hypoxic injury. Accumulation of long-chain fatty acid metabolites is likely to be toxic and contribute to mitochondrial dysfunction during ischemic ARF.

CHAPTER 63 / THE PATHOPHYSIOLOGY OF ARF

647

Figure 63-2 Receptor-mediated and mitochondrial-dependent pathways of cell death. TRADD, TNF receptor-associated death domain protein; ICAD, inhibitors of caspase activity DNase; CAD, caspase activity DNase; AIF, apoptosis-inducing factor.

Evidence suggests that mitochondria act as central coordinators of the downstream execution phase of cell death in the apoptotic pathway. Several proapoptotic signal transduction and damage pathways converge on mitochondria to induce mitochondrial membrane permeabilization; this response is under the control of Bcl-2-related proteins. The inner membrane is characterized by a transmembrane potential ∆ψm, which is generated through the activity of proton pumps of the respiratory chain. When cells are induced to die and the outer mitochondrial membrane becomes permeable to proteins, ∆ψm, dissipates, resulting in the leakage of toxic mitochondrial intermembrane proteins into the cytoplasm. These cytochromes orchestrate the degradation phase of cell death. There is conflicting evidence about the formation of the permeability transition pore. It is made up of a complex of proteins including the adenine nucleotide translocator protein in the inner mitochondrial membrane, the voltage-dependent anion channel in the outer membrane, the cyclosporin A binding protein, cyclophyllin D, in the mitochondrial matrix, and the pro-apoptotic protein, Bax. Although permeability transition pores are assumed to be a normal constituent of mitochondrial membranes, induction of the mitochondrial permeability typically requires damage to the mitochondrial membrane, attack from oxygen radicals, and

protein disulfide cross-linking agents. Evidence for a mitochondrial permeability transition pore is found in freshly isolated proximal tubule cells that are subjected to hypoxia/reoxygenation injury. Two relatively well-characterized cell death pathways activate the downstream executioner, caspase-3. One is receptor mediated and the other is mitochondrial dependent. The receptor-dependent pathway is initiated by activation of cell death receptors (Fas or tumor necrosis factor-α) leading to activation of procaspase-8, which in turn cleaves procaspase-3 to activate caspase-3. The mitochondrial-dependent pathway is triggered by cytochrome-c release from the mitochondria. Surprisingly, the holo-cytochrome-c (but not the apo-cytochrome-c) is required for the activation of caspase-3 in a cell-free system. Cytochrome-c, once present in the cytosol, drives the assembly of a high-molecular-weight, caspase-activating complex termed the apoptosome. Cytochrome-c binds to apoptotic protease-activating factor-1 (Apaf-1), within its C-terminal region rich in WD motifs. Initially believed to be required only transiently for caspase-9 activation, the Apaf-1/caspase-9 complex is thought to represent the active form of caspase-9. Thus, Apaf-1 must be viewed not simply as a caspase-9 activator, but rather as an essential regulatory subunit of a caspase-9 holoenzyme. This holoenzyme—often referred to as the apoptosome—is a very large complex and it could contain additional proteins.

648

SECTION VII / NEPHROLOGY

Figure 63-3 Potential pathways for cellular injury. CAD, caspase activity DNase.

A new protein with the dual name Smac/DIABLO is also released from mitochondria and promotes caspase activation by associating with the apoptosome whereas inhibiting inhibitors of apoptosis proteins (IAPs), the naturally occurring inhibitor of caspase-3. Besides these pathways, there is a mitochondrial protein named apoptosis-inducing factor (AIF) that induces apoptosis independently of caspases. Another mitochcondrial factor translocating to the nucleus has also been isolated: endonuclease G (Endo G). After reaching the cytosol, Endo G translocates toward the nucleus in which it generates oligonucleosomal DNA fragmentation, even in the presence of caspase inhibitors. Endo G can catalyse both high-molecular-weight DNA cleavage and oligonucleosomal DNA breakdown in a sequential fashion; it also cooperates with exonuclease and DNase I to facilitate DNA processing. On induction of apoptosis pathways, mitochondria release several potentially lethal proteins that either participate in caspase activation (cytochrome-c, Smac/DIABLO) and/or induce cell death in a caspase-independent fashion (AIF, Endo G). It is difficult to weigh the relative contribution of each of these factors to apoptosis (Fig. 63-3). REGULATORY MECHANISMS In living cells, mitochondrial changes are predominantly prevented by antiapoptotic members of the Bcl-2 family of proteins. Bcl-2 was first discovered as a proto-oncogene in follicular B-cell lymphoma. Subsequently, it was identified as a mammalian homolog to the apoptosis repressor, ced-9, in Caenorhabditis elegans. Since then, at least 19 Bcl-2 family members have been identified in mammalian cells and these members possess at least one of four conserved motifs known as Bcl-2 homology domains (BH1–BH4). The Bcl-2 family members can be subdivided into three categories according to their function and structure: antiapoptotic members, such as Bcl-2, Bcl-XL, Bcl-2, Mcl-1, and A1 (Bfl-1); proapoptotic molecules, such as Bax, Bak, and Bok (Mtd); and the BH3-only proteins, Bid, Bad, and Bim (these are called BH3-only proteins because of 4 Bcl-2 homology regions, they share only the third). Studies from several laboratories have recognized the phosphatidylinositor-3 kinase pathway, PI-3 kinase/Akt phosphorylation, as one of the signaling pathways that blocks apoptosis and promotes cell survival in response to diverse apoptotic stimuli in

different cell types. Akt (also known as protein kinase B) was originally identified as the cellular homolog of the transforming oncogene of the Akt8 retrovirus. Akt is a serine/threonine kinase and is one of the downstream targets of PI-3 kinase because it phosphorylates Akt to activate it. Several pathways downstream of PI-3/Akt phosphorylation have been proposed to account for survival of the cell. One of the well-studied molecules that mediates cell survival by Akt phosphorylation is the proapoptotic Bcl-2 family member, Bad. Bad has the ability to interact and bind directly to antiapoptotic Bcl-2 and Bcl-XL and block their survival function. Phosphorylated Akt can directly phosphorylate Bad both in vitro and in vivo and may render Bad incapable of binding to Bcl-XL. This response would restore the antiapoptotic function of Bcl-2. Sequestering phosphorylated Bad by 14-3-3 proteins may also participate in the Akt survival pathway by making Bad unavailable to bind to Bcl-2 or prevent it from damaging the mitochondria. Evidence has been presented that PI-3 kinase-mediated Akt phosphorylation is associated with Bad phosphorylation and suppression of caspase-9 and caspase-3 activation in cisplatin-induced injury to renal tubular epithelial cells. Similar results on the activation of caspase-3 and caspase-9 were obtained by inhibiting PI-3 kinase activity following hypoxia-induced injury to renal tubular epithelial cells. One study has shown that Akt can also phosphorylate human caspase-9 resulting in reduction of caspase-9 activity. Based on these studies, the inhibition of Akt phosphorylation and Bad phosphorylation in cisplatin-induced injury contributes to enhanced activation of mitochondrial-dependent caspase-3 and caspase-9 but not receptor-mediated activation of caspase-8 or proinflammatory caspase-1.

CHANGES IN GENE EXPRESSION IN ISCHEMIC ARF During the course and induction of ARF many intrarenal alterations occur including changes in the microvascular endothelium, changes in cell polarity, increased levels of inflammatory molecules, expression of adhesion molecules, changes in cellular metabolism, and changes in the patterns of apoptosis and necrosis. DNA microarray technology has been applied to search for other responses. The goal is to analyze the expression of genes during ischemic/reperfusion injury in order to uncover new mechanisms of pathogenesis. Using this technique, changes in the expression of 18 genes were identified during ischemia/reperfusion of the rat kidney and these responses were confirmed by RT-PCR analysis. There were increased mRNA levels of potential “protective genes” such as clusterin, fibronectin, and hemoxygenase-1, similar to previously described changes. There was decreased expression of uromodulin (Tamm-Horsfall protein), an abundant protein exclusively localized to the thick ascending limb of Henle, also previously reported. Analysis of the gene array uncovered novel responses including increased expression of a disintegrin domain and metalloproteinase (ADAM-2), which mediates cell–cell communication. ADAM-2 likely plays a role in repair after ischemic injury. In addition, there were increased mRNA levels of thymosin B4, a member of the family of actin-monomer sequestering proteins that can enhance wound healing, suggesting that this molecule plays a role in repair of the cell cytoskeleton and in the repolarization of renal tubular epithelial cells after ischemic ARF. Increased expression of uncoupled protein 2, a gene involved in mitochondrial hydrogen peroxide generation and in mitochondrial biogenesis, was also found. Finally, the gene array data revealed downregulation of PPAR-γ during ischemic reperfusion injury.

CHAPTER 63 / THE PATHOPHYSIOLOGY OF ARF

PPAR-γ belongs to the family of nuclear receptors that serves as a transcription factor and regulates several physiologic pathways associated with lipid metabolism and inflammation. Several fatty acid oxidation target genes of PPAR-α have been identified that are downregulated by ischemic ARF. That another PPAR target gene such as stromyelisin is also downregulated underscores the potential role that PPARs might play in cell repair following ischemic injury. The cDNA microarray technique has also been used to study the responses to ischemic injury in a mouse model of ARF. Hierarchical clustering revealed several temporal patterns of gene expression. Early changes (at 3 h) during ischemic reperfusion injury included increased expression of transcription factors such as Egr-1 and c-fos, repair genes such as heat shock protein 70, growth factors such as hepatocyte growth factor, and Bcl-2. Genes that were upregulated at later periods (12–24 h) of reflow included heparin-binding epidermal growth factor, insulin-like growth factor, transforming growth factor, the cell-cycle protein p21, and cyto-protective factors such as heme-oxygenase 1 and α crystalline. The array confirmed downregulation of the sodium hydrogen exchanger-3, α albumin, and members of the cytochrome-p450 family after 24 h of reflow. Regarding apoptosis, there was persistent increased expression of FADD, DAXX, p53, and Bad at 3, 12, and 24 h of reperfusion. The changes in proapoptotic mRNA transcripts were confirmed at the protein level by immunohistochemistry demonstrating staining for FADD, Bak, Bad, and DAXX localized predominantly to the distal tubule. Changes in gene expression likely represent a multifaceted response to an ischemic insult that includes cell death, dedifferentiation of viable cells, proliferation, differentiation, and restitution of a normal epithelium.

CONCLUSION The study of cell injury has emerged as one of most intensively investigated areas of research in life sciences. These studies have led to a better understanding of the complexity of the cell death process including the intricacies of the various paths with their complex, interweaving redundancies. There has been little progress in preventing and treating ARF, but with a better understanding of cell injury mechanisms and by carefully defining populations at risk for developing ARF, treatment strategies could be developed. In addition, understanding the pathophysiologic mechanism may allow identification of biomarkers, which may be utilized for early treatment and/or prevention of acute renal failure.

SELECTED REFERENCES Brodsky SV, Yamamoto T, Tada T, et al. Endothelial dysfunction in ischemic acute renal failure: rescue by transplanted endothelial cells. Am J Physiol Renal Physiol 2002;282:F1140–F1149. Datta SR, Brunet A, Greenberg ME. Cellular survival: a play in three akts. Genes Dev 1999;13:2905–2927. Earnshaw WC, Martins LM, Kaufmann SH. Mammalian caspases: structure, activation, substrates, and functions during apoptosis. Annu Rev Biochem 1999;68:383–424.

649

Hengartner MO. The biochemistry of apoptosis. Nature 2000;407(6805): 770–776. Kelly KJ, Plotkin Z, Dagher P. Guanosine supplementation reduces apoptosis and protects renal function in the setting of ischemic injury. J Clin Invest 2001;9:1291–1298. Kelly KJ, Plotkin Z, Vulgamott SL, Dagher P. P53 mediates the apoptic response to GTP depletion after renal ischemia-reperfusion: protective role of a p53 inhibitor. J Am Soc Nephrol 2003;14:128–138. Kerr JFR, Wyllie AH, Currie AR. Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer 1972;26:239–257. Lemasters JJ, Nieminen AL, Qian T, et al. The mitochondrial permeability transition in cell death: a common mechanism in necrosis, apoptosis and autophagy. Biochim Biophys Acta 1998;1366:177–196. Okusa, MD. The inflammatory cascade in acute ischemic renal failure. Nephron 2002;90:133–138. Portilla D, Dai G, Peters J, Gonzalez FJ, Crew M, Proia AD. Etoximirinduced PPARα-modulated enzymes protect during acute renal failure. Am J Physiol Renal Physiol 2000;278:F667–F675. Rabb H, Daniels F, O’Donnell M, et al. Pathophysiological role of T lymphocytes in renal ischemia-reperfusion injury. Am J Physiol 2000;279: F525–F531. Ravagnan L, Roumier T, Kroemer G. Mitochondria, the killer organelles and their weapons. J Cell Physiol 2002;192:131–137. Saikumar P, Dong Z, Weinberg JM, Venkatachalam MA. Mechanisms of cell death in hypoxia/reoxygenation injury. Review. Oncogene 1998; 17(25):3341–3349. Strasser A, O’Connor L, Dixit VM. Apoptosis signaling. Annu Rev Biochem 2000;69:217–245. Supavekin S, Zhang W, Kucherlapati R, Kaskel FJ, Moore LC, Devarajan P. Differential gene expression following early renal ischemia/reperfusion. Kidney Int 2003;63:1714–1724. Sutton TA, Fisher CJ, Molitoris BA. Microvascular endothelial injury and dysfunction during ischemic acute renal failure. Kidney Int 2002;62: 1539–1549. Sutton TA, Mang HE, Campos SB, Sandoval RM, Yoder MC, Molitoris BA. Injury of the renal microvascular endothelium alters barrier function following ischemia. Am J Physiol Renal Physiol 2003;10: F191–F198. Sutton TA, Molitoris BA. Mechanisms of cellular injury in ischemic acute renal failure. Semin Nephrol 1998;18(5):490–497. Thornberry NA, Lazebnik Y. Caspases: enemies within. Science 1998;281:1312–1316. Ueda N, Walker PD, Hsu S-M, Shah SV. Activation of a 15-kDa endonuclease in hypoxia/reoxygenation injury without morphologic features of apoptosis. Proc Natl Acad Sci USA 1995;92:7202–7206. Weinberg JM, Venkatachalam MA, Roeser NF, Nissim I. Mitochondrial dysfunction during hypoxia/reoxygenation and its correction by anaerobic metabolism of citric acid cycle intermediates. Proc Natl Acad Sci USA 2000;97(6):2826–2831. Wolf BB, Green DR. Suicidal tendencies: apoptotic cell death by caspase family proteinases. J Biol Chem 1999;274:20,049–20,052. Yamamoto T, Tada T, Brodsky SV, et al. Intravital videomicroscopy of peritubular capillaries in renal ischemia. Am J Physiol Renal Physiol 2002;282:F1150–F1155. Yoshida T, Kurella M, Beato F, et al. Monitoring changes in gene expression in renal ischemia-reperfusion in the rat. Kidney Int 2002;61:1646–1654. Zimmerman KC, Green DR. How cells die: apoptosis pathways. J Allergy Clin Immunol 2001;108:S99–S103.

64 Loss of Lean Body Mass in Uremia S. RUSS PRICE AND WILLIAM E. MITCH SUMMARY The pathophysiological consequences of kidney failure and other catabolic conditions such as diabetes, burn injury, cancer, sepsis, muscle denervation and starvation, include loss of protein stores. This chapter will describe the impressive concordance between studies performed with cultured muscle cells and studies involving experimental animals and patients suggests that there is strong evolutionary pressure to maintain these “stress”-related responses. Understanding the mechanisms leading to activation of transcription could lead to improved methods of preventing the loss of muscle protein. Key Words: Body mass; catabolic signals; glucocorticoids; ubiquitin-proteasome; UbC.

INTRODUCTION The pathophysiological consequences of kidney failure and other catabolic conditions (e.g., diabetes, burn injury, cancer, sepsis, muscle denervation, starvation, and so on) include loss of protein stores. Available information suggests a common pathway is responsible for the loss of protein in all of these conditions. Fundamental questions are why the rate of protein loss varies so widely and what signals or mechanisms can explain why protein mass declines. In healthy adults, muscle proteins turn over at a high rate, the protein equivalent of approx 1–1.5 kg of muscle mass per day. This rate of turnover in muscle is necessary to replace proteins that are incorrectly translated, that become damaged or modified, or that require rapid degradation because they regulate critical cellular functions. However, because the rate of turnover is so high, even a small but persistent decrease in protein synthesis or an increase in protein degradation or both of these changes occurring simultaneously results in substantial muscle atrophy. This is important to clinical medicine because loss of lean body mass is associated with increased morbidity and mortality.

ACTIVATION OF THE UBIQUITIN-PROTEASOME SYSTEM IN MUSCLE In most conditions associated with loss of muscle mass, a decrease in protein synthesis is not the principal defect causing loss of muscle protein. Instead, there is overwhelming evidence that protein degradation is accelerated. Regarding specific proteolytic From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

pathways that degrade proteins, in muscle, like most tissues, there are several systems. The major systems are lysosomal proteases (e.g., cathepsins), calcium-activated proteases (e.g., calpains), and proteasome-mediated proteolysis (including protein degradation involving ATP and ubiquitin). In experimental animal models of conditions causing muscle atrophy, studies directed at identifying the specific proteolytic pathways that are activated to degrade muscle protein have repeatedly demonstrated involvement of the ATPdependent ubiquitin-proteasome pathway. Protein substrates of this proteolytic system must be marked for degradation. Thus, the substrates undergo a series of reactions that covalently link ubiquitin to form a polyubiquitin chain, which serves to target proteins for degradation in the proteasome. Ubiquitin is a small protein member of the heat-shock protein family that is present in all cells. Figure 64-1 summarizes the general ubiquitin conjugation and degradation process. Subunits in the 19S regulatory “cap” complexes of the 26S proteasome recognize the polyubiquitin chain and in an ATPdependent process, unfold the substrate protein. It is then inserted into the proteasome where other subunits remove and perhaps disassemble the polyubiquitin chains for recycling of ubiquitin. In the central chamber of the barrel-shaped proteasome, there are protease activities that utilize the hydroxyl group of threonine to perform nucleophilic attacks on the peptide bonds. The proteolytic sites recognize different amino acid motifs in the substrate protein; hence, there is trypsin-like activity, chymotrypsin-like activity, and so on. The result is a breakdown of the substrate protein into short peptide fragments that are released into the cytoplasm. Subsequently, cytoplasmic endo- and exo-peptidases complete the degradation process. Thus, the ubiquitin-proteasome system is a tightly controlled mechanism for degrading proteins; the substrate protein must first be marked by a ubiquitin chain through a complex series of conjugation reactions and the protein is degraded only after removal of the ubiquitin chain and insertion into the proteasome. A remarkable property of this system is its ability to recognize the different proteins that are destined for degradation. This is owing to the specificity of the myriad of combinations of E2 ubiquitin conjugating enzymes and E3 ubiquitin-protein ligases. The mechanisms by which these enzymes achieve specificity are under intensive investigation. One mechanism of muscle proteolysis has been shown to utilize a specific E3 enzyme, E3α. This enzyme recognizes proteins with distinct amino terminal amino acids and is known as the N-end rule pathway. However, even this process is complex because many proteins are initially modified to change the amino terminal amino acid and hence, the rate of degradation.

650

CHAPTER 64 / LOSS OF LEAN BODY MASS IN UREMIA

651

Figure 64-1 General scheme of protein degradation by the ubiquitin-proteasome system. Substrates that are degraded by the ubiquitin-proteasome system first undergo covalent addition of a polyubiquitin chain by a series of reactions that involve (1) ATP-dependent activation of ubiquitin by an E1 ubiquitin activating enzyme; (2) transfer of activated ubiquitin to the ε-NH2 group of a lysine residue in the substrate and subsequent polyubiquitin chain formation by combinations of E2 ubiquitin conjugating enzymes and E3 ubiquitin protein ligase complexes (E3s can be monomeric or multimeric complexes). Substrate specificity is conferred by the various combinations of E2s and E3s; (3) recognition of the polyubiquitin chain by subunits of the 19S regulatory complex, which attaches to the end of a 20S core proteasome to form the 26S proteasome; and (4) unfolding, insertion and degradation of the substrate by the 26S proteasome (a process that requires ATP) to yield small peptides that are further degraded by peptidases. Ubiquitin chains are released from the protein and recycled.

Evidence that the ubiquitin-proteasome system is activated to break down muscle protein in kidney failure and other catabolic conditions comes from studies involving measurements of rates of protein degradation in isolated muscles incubated with inhibitors of the different proteolytic pathways. For example, muscles isolated from rats with chronic renal failure were incubated with inhibitors of calcium-dependent and lysosomal proteases and the rate of protein degradation was measured. With these inhibitors, there was a small and parallel decrease in the rates of protein degradation in muscles of both the chronically uremic rats and the sham-operated, pair-fed rats. However, these inhibitors did not block the accelerated rate of proteolysis stimulated by kidney failure. In contrast, when an inhibitor of the proteasome (i.e., MG132) was added or when ATP generation was blocked, the accelerated rate of protein degradation was eliminated and this occurred even in the presence of inhibitors of the calcium-activated and lysosomal proteases. Notably, the rate of protein degradation in muscles of the control rats was also suppressed to a low rate, indicating that the ubiquitin-proteasome system is responsible for the degradation of the bulk of protein in muscle. Additional evidence that the ubiquitin-proteasome system is activated in muscle of rats with various wasting conditions includes: 1. Increased conjugation of ubiquitin to muscle proteins. 2. Increased levels of the proteins that are components of the system.

3. Increased levels of mRNAs that encode ubiquitin and other component proteins of the system. 4. Return of these responses to control levels or lower when the condition causing accelerated muscle proteolysis is attenuated. Two genes encoding ubiquitin protein ligases (i.e., E3s) have been identified in experiments based on microarray or differential display analyses of muscles undergoing atrophy. One of these proteins, muscle ring finger 1 (MuRF1 or atrogin-1), is a ring finger protein that interacts with titin, a large myofibrillar protein that has several functions. The other newly identified protein, atrogin-1 (also called muscle atrophy F-box [MAFbx]), belongs to the F-box (SCF) family of proteins that includes components of the multi-subunit skpcullin-F box E3 ligases. Consistent with the role of the ubiquitin-proteasome system as the principal mechanism degrading muscle protein, the mRNA for atrogin-1 is specifically expressed in skeletal and cardiac muscle and its expression is increased >10-fold in muscles undergoing atrophy. Moreover, overexpression of atrogin-1 in C2C12 muscle cells resulted in thinner myotubes, reminiscent of muscle atrophy whereas muscles of mice with null alleles for either MuRF1/atrogin-1 do undergo atrophy in response to muscle denervation, but the degree of muscle wasting was attentuated in comparison with that in wild-type mice. Given the complexity of the pool of muscle proteins, it is likely that multiple ubiquitin conjugation systems are involved when the proteolytic process is accelerated.

652

SECTION VII / NEPHROLOGY

The mechanisms leading to activation of the ubiquitin-proteasome system in conditions leading to muscle atrophy are poorly understood. For example, the “rate-limiting” step in muscle proteolysis has not been identified. Potentially important steps include: 1. Availability of muscle protein substrates because actomyosin and myofibrils are not degraded by the system, so the substrates actin and myosin must be liberated before they are degraded by the ubiquitin-proteasome system. 2. Recognition and polyubiquitination of substrate proteins by combinations of E2 and E3 enzymes acting in concert. 3. The proteolytic capacity of the cell related to changes in the levels of the many components of the system. There are clues or insights into how catabolic signals activate this proteolytic system. One potential clue is the frequent observation that mRNAs encoding various components of the system (e.g., ubiquitin, ubiquitinconjugating enzymes, subunits of the 26S proteasome) are increased in each model of different conditions causing accelerated loss of muscle protein. This finding has been extended to patients with catabolic conditions (e.g., head trauma and sepsis) and includes patients with chronic renal failure. These subjects not only have evidence of muscle protein losses but also have higher levels of pathway mRNAs in muscle. At least in chronic kidney failure, removal of the catabolic stimulus, metabolic acidosis, was associated with a coordinated decrease in mRNA level for ubiquitin and muscle proteolysis suggesting the two responses are linked. Specifically, patients with end-stage renal disease receiving chronic ambulatory peritoneal dialysis were treated to remove the metabolic acidosis associated with kidney failure over 1 mo. This treatment yielded an increase in body weight and a simultaneous reduction in the level of ubiquitin mRNA in their muscle. Several extracellular signals could accelerate proteolysis through the ubiquitin-proteasome system in muscle. In kidney failure, metabolic acidosis, and diabetes (or possibly insulin resistance), cytokines are potential signals acting directly or indirectly to stimulate muscle proteolysis and increase pathway mRNAs. Besides these mechanisms, glucocorticoids play a role because they can regulate the ubiquitin-proteasome system in muscle but their role varies depending on their level in animals. For example, when glucocorticoids are present in the high physiologic range, they are necessary for the activation of proteolysis plus the increase in pathway mRNAs in muscle, but their role is permissive because glucocorticoids require the presence of an additional catabolic signal (e.g., acidosis, cytokines, or a low level of insulin). In contrast, if rats are given pharmacologic doses of glucocorticoids, muscle atrophy does occur and levels of mRNAs encoding components of the ubiquitin-proteasome system in muscle are higher, at least transiently. The clinical relevance of glucocorticoids derives from reports that catabolic conditions upregulate the glucocorticoid signaling system in muscle; the number of glucocorticoid receptors is increased in muscle of rats with experimental sepsis following cecal ligation and puncture. Moreover, the proteolytic response in muscles was blocked by giving septic rats the glucocorticoid receptor antagonist, RU486, suggesting that a positive feedback system is activated in the muscle. In addition to glucocorticoids, factors that stimulate expression of the components of the ubiquitin-proteasome pathway and enhance muscle protein degradation could include a decrease in muscle cell pH (or some other ion change linked to a fall in extracellular pH) or

impairment of insulin signaling. There is evidence in kidney cells that acidification can activate gene transcription via specific signaling pathways. The impact of intracellular acidification in stimulating the ubiquitin-proteasome pathway was investigated by examining chronically uremic rats with the same characteristics as rats expressing accelerated muscle proteolysis and higher levels of mRNAs for the components of the ubiquitin-proteasome system in muscle. Using nuclear magnetic resonance (NMR), pH measurements were taken in the intact muscle of these rats and the pair-fed, sham-operated control rats. It was neither subnormal nor was there a delay in recovery of cell pH when the muscle was stimulated to tetany in order to drop muscle cell pH acutely. Although there could be a change in pH in some compartment of muscle that was not measured, these results do not support a direct role for changes in pH causing stimulation of the ubiquitin-proteasome pathway or transcription of genes. It has been shown that both acidosis and kidney failure can impair insulin signaling. The latter is important because an acute decrease in insulin markedly stimulates protein degradation and the expression of mRNAs encoding components of the ubiquitin-proteasome pathway in muscle. Furthermore, studying the signals that activate muscle protein loss in acute diabetes shows that stimulation of the ubiquitin-proteasome pathway, including the increase in muscle levels of mRNAs for components of the system, was not blocked when the development of metabolic acidosis was prevented by feeding acutely diabetic rats NaHCO3. The stimulation of the ubiquitinproteasome system in acutely diabetic rats, like those with metabolic acidosis, required glucocorticoids to activate the system.

CATABOLIC SIGNALS INITIATE A PROGRAM OF TRANSCRIPTIONAL RESPONSES IN MUSCLE Whether greater amounts of ubiquitin-proteasome pathway mRNAs arise from increased message stabilization or transcriptional activation was revealed from evaluation of nuclear run-off assays using nuclei isolated from the hindquarter muscle of rats with acute diabetes or chronic renal failure. In both cases, there was increased transcription of the ubiquitin and the proteasome C3 subunit genes in comparison with rates measured in muscles of pair-fed control rats. In L6 muscle cells, dexamethasone, a potent glucocorticoid, was found to increase the transcription of the ubiquitin (UbC) and proteasome C3 subunit genes, leading to increased levels of these proteins. It has also been reported that there is no difference in the rate of disappearance of ubiquitin mRNA when extensor digitorum longus muscles from septic and sham-operated control rats are incubated in the presence of actinomycin to inhibit transcription. This finding indicates that the increase in mRNA is not attributable to message stabilization or post-transcriptional events. Thus, muscle atrophying conditions activate a program of specific transcriptional responses in muscle. Genes encoding several components of the ubiquitin-proteasome system in rodents and humans have been characterized (e.g., proteasome C3 and C5 subunits, UbC). Surveys of their promoter regions reveal cis-acting elements with the potential for responding to inflammatory or stress-related signals. It has also been reported that CCAAT enhancer binding proteins, (C/EBP)-β and -δ, and activator protein-1, are activated in muscle of rats with sepsis; these are the same transcription factors that mediate inflammatory responses. The role of glucocorticoids in the activation of these factors was examined by giving septic rats an oral dose of RU486, a glucocorticoid receptor antagonist. RU486 blocked the

CHAPTER 64 / LOSS OF LEAN BODY MASS IN UREMIA

653

Figure 64-2 Mechanism of proteasome C3 subunit transactivation by glucocorticoids. (A) In muscle, NF-κB is activated by proteasome-mediated degradation of IκB, a cytosolic inhibitor protein. NF-κB then translocates to the nucleus in which it binds to sites in the C3 subunit promoter and suppresses transcription. (B) Glucocorticoids prevent the proteasome-mediated degradation of IκB, thus inhibiting NF-κB nuclear translocation. Sequestration of NF-κB in the cytosol results in stimulation of proteasome C3 subunit expression.

activation of C/EBP-β and -δ but did not block the activation of activator protein-1. These findings suggest that stress-related mediators play a part in activating the program of transcriptional responses in muscle occurring in response to cachectic signals. NF-κB IS A NEGATIVE REGULATOR OF THE PROTEASOME C3 SUBUNIT EXPRESSION Glucocorticoids are potent regulators of gene expression and can influence transcription by multiple mechanisms. Investigations of the role of glucocorticoids in the transcriptional responses of the ubiquitin-proteasome system in muscle included studying the influence of glucocorticoids on the regulation of the proteasome C3 subunit in L6 muscle cells. Inspection of the human C3 proteasome subunit promoter region did not identify a classic glucocorticoid response element despite the fact that glucocorticoids-induced expression of this subunit in L6 muscle cells. However, there was a region of the C3 subunit promoter between –400 and –256 bp (+1 = transcription start site) that was essential for stimulation by dexamethasone. This region contains several nuclear factor (NF)-κB-like binding sites, notable because glucocorticoid receptors and NF-κB can exert mutually antagonistic effects to repress or activate gene transcription. Further examination of a binding site at –322/–313 using mobility shift assays revealed that incubating cells to dexamethasone for

only 20 min decreased the formation of a DNA–NF-κB complex. This antagonism of NF-κB binding to the promoter region persisted for at least 24 h. Because glucocorticoids act to stimulate the transcription of the proteasome C3 subunit, these findings indicate that NF-κB must act as a repressor of C3 subunit transcription (Fig. 64-2A). Thus, glucocorticoids act to stimulate transcription by antagonizing the transrepression by NF-κB of the C3 subunit of the proteasome. This mechanism was confirmed by demonstrating that overexpressing NF-κB subunits or activating cellular production of NF-κB with cytokines reduced transcription of the C3 proteasome subunit. In contrast, blocking NF-κB activity stimulated transcription of this gene. How do glucocorticoids antagonize NF-κB transactivation? No evidence exists for a direct interaction between the glucocorticoid receptor and NF-κB subunits nor did dexamethasone increase the transcription of inhibitor of κB (IκB). However, glucocorticoids increased the amount of IκB protein in the cytosol with a concurrent increase in the amount of cytosolic NF-κB p65 subunit, leading to the conclusion that glucocorticoids enhance proteasome C3 subunit transcription by causing a sequestration of inactive NF-κB in the cytosol (Fig. 64-2B). Subsequent studies of muscle from septic and control rats recapitulated the findings with L6 muscle

654

SECTION VII / NEPHROLOGY

cells and confirmed that NF-κB serves as a suppressor of proteasome C3 subunit transcription in muscle of intact animals. GLUCOCORTICOIDS INCREASE UBC TRANSCRIPTION BY A SP1-DEPENDENT MECHANISM Three independent genes (i.e., UbA, UbB, and UbC) encode protein products that are processed to yield “free” ubiquitin protein. Among these genes, UbC mRNA has typically increased the most in response to conditions causing muscle atrophy. The human UbC promoter region has been characterized, including a number of potential binding sites for factors that could regulate transcription. By cloning the rat UbC promoter, conservation of a number of the potential response elements was found, including multiple sites for NF-κB, other inflammatory class transcription factors, and specificity factor 1 (Sp1). As in the case of the proteasome C3 subunit, however, there was a conspicuous absence of consensus glucocorticoid response elements in the rat or human UbC promoter even though dexamethasone stimulated transcription of the UbC gene in L6 muscle cells. These findings prompted investigation of the cis-acting elements that confers glucocorticoid responsiveness. Initially, in vivo genomic footprinting assays were performed in L6 muscle cells to identify regions of the UbC promoter that undergo structural changes in response to dexamethasone. Surprisingly, these bases were located in a potential Sp1 binding site. This location suggested that dexamethasone induces a change in the binding of Sp1 and results of promoter deletion and mutational analysis coupled with mobility shift assays did show that dexamethasone increases the in vitro binding of Sp1 to a UbC-specific DNA probe. The importance of Sp1 in mediating the transactivation of the UbC promoter was confirmed when an inhibitor of Sp1 binding, mithramycin, was shown to prevent the dexamethasone-induced increase in UbC gene activity. Mithramycin also decreased the levels of free ubiquitin protein and ubiquitin conjugated to endogenous muscle cell proteins. Thus, it is clear that a mechanism for increased UbC expression by dexamethasone involves Sp1. Sp1 has traditionally been considered a ubiquitous transcription factor involved in the basal transcription of many genes. To determine how glucocorticoids increase Sp1 binding to the UbC promoter, the total cellular content of Sp1 was measured and the distribution of Sp1 between the cytosol and nucleus was examined. Dexamethasone did not change the amount or distribution of Sp1 in L6 cells, suggesting that glucocorticoids do not simply augment the level of Sp1 in the nucleus. There are reports that Sp1 binding activity is modulated by phosphorylation. Moreover, activation of the p42/44 mitogen-activated protein (MAP) kinase (i.e., extracellular signal regulated kinase [ERK] 1/2) signaling pathway has been linked to increased Sp1-mediated transcription and glucocorticoids can stimulate ERK activity. Examination of the role of MAP kinase/ERK kinase (MEK)1/2, the kinases upstream of ERK, using a MEK1/2 inhibitor, U0126 revealed that treating L6 muscle cells with U0126 blocked the induction of UbC-driven luciferase activity by dexamethasone. In contrast, transfecting L6 cells to express constitutively active MEK1 mimicked dexamethasone by inducing UbC transcription. These findings demonstrate that glucocorticoids stimulate UbC transcription by a mechanism involving both Sp1 and MEK1. A potential signaling scheme for this response is shown in Fig. 64-3.

CONCLUSIONS Studies involving a variety of models of pathological states consistently indicate that catabolic conditions causing loss of muscle

Figure 64-3 Potential mechanism of Ubiquitin C (UbC) transactivation by glucocorticoids. Glucocorticoids increase Sp1 binding to the UbC promoter. MEK1/2 is involved in the transactivation process but the mechanism is unknown. In the potential scheme shown, glucocorticoids bind to their receptor, which indirectly stimulates MEK1/2, which activates ERK1/2, resulting in the phosphorylation of Sp1. Phosphorylation of Sp1 increases its ability to bind to the UbC promoter, thus enhancing its transcription.

protein induce a concomitant program of transcriptional responses that result in higher levels of the mRNAs that encode components of the ubiquitin-proteasome system. The transactivation mechanisms for many of these conditions have not been defined and are likely to be complex. The mechanism by which glucocorticoids induce the UbC and proteasome C3 subunit genes differ substantially. In the case of the proteasome C3 subunit, glucocorticoids act by antagonizing NF-κB, a suppressor of subunit transcription. Conversely, glucocorticoids induce UbC by increasing the binding of Sp1 and the mechanism requires components of a MAP kinase pathway linked to cell growth and survival. Although the importance of the increase in mRNAs encoding components of the ubiquitin-proteasome system in terms of proteolysis is debated, the majority of evidence indicates that catabolic signals that accelerate protein degradation via the ubiquitin-proteasome pathway initiate a program of selective transcriptional responses in muscle. The impressive concordance between studies performed with cultured muscle cells and studies involving experimental animals and patients suggests that there is strong evolutionary pressure to maintain these “stress”-related responses. Understanding the mechanisms leading to activation of transcription could lead to improved methods of preventing the loss of muscle protein.

ACKNOWLEDGMENT This work was supported by the National Institutes of Health through grants DK50740, DK63658, and DK37175.

SELECTED REFERENCES Alvestrand A. Carbohydrate and insulin metabolism in renal failure. Kidney Int Suppl 1997;62:S48–S52. Bailey JL, England BK, Long RC Jr, Weissman J, Mitch WE. Experimental acidemia and muscle cell pH in chronic acidosis and renal failure. Am J Physiol 1995;269:C706–C712. Bailey JL, Wang X, England BK, Price SR, Ding X, Mitch WE. The acidosis of chronic renal failure activates muscle proteolysis in rats by augmenting transcription of genes encoding proteins of the ATP-dependent, ubiquitin-proteasome pathway. J Clin Invest 1996;97:1447–1453. Bodine SC, Latres E, Baumhueter S, et al. Identification of ubiquitin ligases required for skeletal muscle atrophy. Science 2001;294:1704–1708.

CHAPTER 64 / LOSS OF LEAN BODY MASS IN UREMIA

Cecchin F, Ittoop O, Sinha MK, Caro JF. Insulin resistance in uremia: insulin receptor kinase activity in liver and muscle from chronic uremic rats. Am J Physiol 1988;254:E394–E401. DeFronzo RA, Alvestrand A, Smith D, Hendler R. Insulin resistance in uremia. J Clin Invest 1981;67:563–568. Du J, Mitch WE, Wang X, Price SR. Glucocorticoids induce proteasome C3 subunit expression in L6 muscle cells by opposing the suppression of its transcription by NF-κB. J Biol Chem 2000;275:19,661–19,666. Glickman MH, Ciechanover A. The ubiquitin-proteasome proteolytic pathway: destruction for the sake of construction. Physiol Rev 2002;82(2): 373–428. Gomes MD, Lecker SH, Jagoe RT, Navon A, Goldberg AL. Atrogin-1, a muscle-specific F-box protein highly expressed during muscle atrophy. Proc Natl Acad Sci USA 2001;98(25):14,440–14,445. Hasselgren PO. Glucocorticoids and muscle catabolism. Curr Opin Clin Nutr Metab Care 1999;2:201–205. Jagoe RT, Goldberg AL. What do we really know about the ubiquitinproteasome pathway in muscle atrophy? Curr Opin Clin Nutr Metab Care 2001;4(3):183–190. Jagoe RT, Lecker SH, Gomes M, Goldberg AL. Patterns of gene expression in atrophying skeletal muscles: response to food deprivation. FASEB J 2002;16(13):1697–1712. Karin M, Chang L. AP-1-glucocorticoid receptor crosstalk taken to a higher level. J Endocrinol 2001;169(3):447–451. Marinovic AC, Zheng B, Mitch WE, Price SR. Ubiquitin (UbC) expression in muscle cells is increased by glucocorticoids through a mechanism involving Sp1 and MEK1. J Biol Chem 2002;277(19):16,673–16,681. May RC, Kelly RA, Mitch WE. Metabolic acidosis stimulates protein degradation in rat muscle by a glucocorticoid-dependent mechanism. J Clin Invest 1986;77:614–621. McKay LI, Cidlowski JA. Cross-talk between nuclear factor-κB and steroid hormone receptors: mechanisms of mutual antagonism. Mol Endocrinol 1998;12(1):45–56.

655

Mehrotra R, Kopple JD. Nutritional management of maintenance dialysis patients: why aren’t we doing better? Annu Rev Nutr 2001;21:343–379. Mitch WE, Bailey JL, Wang X, Jurkovitz C, Newby D, Price SR. Evaluation of signals activating ubiquitin-proteasome proteolysis in a model of muscle wasting. Am J Physiol 1999;276:C1132–C1138. Mitch WE, Goldberg AL. Mechanisms of muscle wasting: The role of the ubiquitin-proteasome pathway. N Engl J Med 1996; 335(25):1897–1905. Mitch WE, Price SR. Mechanisms activated by kidney disease and the loss of muscle mass. Am J Kidney Dis 2001;38(6):1337–1342. Nenoi M, Mita K, Ichimura S, et al. Hetergeneous structure of the polyubiquitin gene UbC of Hela S3 cells. Gene 1996;175:179–185. Penner CG, Gang G, Wray C, Fischer JE, Hasselgren PO. The transcription factors NF-kappaB and AP-1 are differentially regulated in skeletal muscle during sepsis. Biochem Biophys Res Commun 2001;281(5): 1331–1336. Pickering WP, Price SR, Bircher G, Marinovic AC, Mitch WE, Walls J. Nutrition in CAPD: serum bicarbonate and the ubiquitin-proteasome system in muscle. Kidney Int 2002;61(4):1286–1292. Price SR, Bailey JL, Wang X, et al. Muscle wasting in insulinopenic rats results from activation of the ATP-dependent, ubiquitin proteasome proteolytic pathway by a mechanism including gene transcription. J Clin Invest 1996;98:1703–1708. Samson SL, Wong NC. Role of Sp1 in insulin regulation of gene expression. J Mol Endocrinol 2002;29(3):265–279. Solomon V, Goldberg AL. Importance of the ATP-ubiquitin-proteasome pathway in the degradation of soluble and myofibrillar proteins in rabbit muscle extracts. J Biol Chem 1996;271(43):26,690–26,697. Varshavsky A. The N-end rule and regulation of apoptosis. Nat Cell Biol 2003;5(5):373–376. Wiborg O, Pedersen MS, Wind A, Berglund LE, Marcker KA, Vuust J. The human ubiquitin multigene family: some genes contain multiple directly repeated ubiquitin coding sequences. EMBO J 1985;4: 755–759.

65 Mechanisms of Renal Allograft Rejection DANIEL R. GOLDSTEIN, ANIRBAN BOSE, AND FADI G. LAKKIS SUMMARY The immune response to donor antigens leads to the rejection of transplanted organs. The earliest event in this response is the activation of the innate immune system. Innate activation culminates in the migration of antigen presenting cells to the recipient’s secondary lymphoid tissues where they trigger the adaptive immune response. The principal events of this response are the activation of T and B lymphocytes, which generate effector cells that inflict immediate damage. Adaptive immunity also produces long-lived memory lymphocytes that cause either acute or chronic rejection. Here, we will summarize the cellular and molecular mechanisms of the innate and adaptive responses that lead to graft rejection. Key Words: Adaptive immunity; antigen presenting cells; B cells; cytokines; innate immunity; lymphocytes; memory; natural killer cells; rejection; T cells; transplantation.

INTRODUCTION Transplanting an organ from one member of a species into a nonidentical member of the same species results in a donor-specific immune response referred to as alloimmunity. If left untreated, alloimmune responses lead to the rejection of the transplanted organ (the allograft). The central mediator of the alloimmune response is the T lymphocyte. On activation by donor antigens (alloantigens) presented on antigen-presenting cells (APCs), allospecific T lymphocytes proliferate and differentiate into effector cells or provide help to other cells that inflict damage on the allograft. This chapter provides an overview of the cellular and molecular mechanisms underlying the immune response that leads to allograft rejection.

OVERVIEW OF THE ALLOIMMUNE RESPONSE The alloimmune response, like any other immune response to a foreign antigen, occurs in two broad phases: the innate and adaptive phases (Fig. 65-1). In the innate phase, inflammation in the transplanted organ activates APCs and induces their maturation and migration to secondary lymphoid organs (the spleen and lymph nodes). The adaptive immune response is initiated when APCs activate antigen-specific T cells within secondary lymphoid organs leading to effector cell generation and their migration to the allograft where they mediate rejection. The majority of effector T cells eventually undergoes apoptosis and the few that survive From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

become long-lived memory T cells that endanger the survival of a subsequent organ transplant. Before describing the molecular and cellular mechanisms of the innate and adaptive immune responses to a transplanted organ, it is important to define “alloantigen.” The principal non-self-antigens responsible for initiating alloimmune responses are the major histocompatibility complex (MHC) proteins, known as human leukocyte antigens (HLAs) in man. MHC proteins are encoded by a family of highly polymorphic genes and are divided into two classes. MHC class-I antigens are expressed on the vast majority of nucleated cells whereas MHC class-II antigens are expressed primarily on APCs such as dendritic cells (DCs), macrophages, and B-lymphocytes. MHC class-II expression is also induced on activated T cells and, in humans, on endothelial cells. The physiologic function of MHC molecules is to bind antigenic peptides and present them to T lymphocytes. In fact, the T-cell receptor for antigen (TCR) does not interact with whole antigens but instead recognizes small antigenic peptides bound to the grooves of MHC molecules present on APCs. In general, exogenous antigens (e.g., circulating foreign proteins) are taken up by APCs, degraded within endocytic vesicles, and presented by class-II molecules, whereas endogenous antigens (e.g., viral proteins produced within the cell cytosol) are presented by class-I molecules. The highly polymorphic MHC ensures that humans mount immune responses against a wide array of pathogens. Unfortunately, the same polymorphism forms the basis of tissue incompatibility between members of the same species. T cells of one individual recognize the non-self-MHC present on another individual’s tissues as foreign, leading to the phenomenon of transplant rejection. In addition to MHC, nonMHC polymorphic molecules, known as minor histocompatibility antigens, also initiate alloimmune responses. Single or multiple minor histocompatibility differences between the donor and the recipient lead to the rejection of MHC-matched organs. A classic example of minor histocompatibility antigens is the HY antigen expressed on male but not female cells.

THE INNATE IMMUNE RESPONSE TO A TRANSPLANTED ORGAN CELLULAR AND MOLECULAR COMPONENTS OF INNATE IMMUNITY The innate immune system acts as “the first line of defense” against foreign antigens, whether an allograft or an invading microbial pathogen. The cellular components of this phase of the immune system consist of APCs such as DCs and macrophages, natural killer (NK) cells, and neutrophils. These cells reside in

656

CHAPTER 65 / MECHANISMS OF RENAL ALLOGRAFT REJECTION

657

Figure 65-1 Innate and adaptive immune mechanisms lead to allograft rejection. Ischemia-reperfusion injury and inflammation in a newly transplanted organ activates APC, which then migrate to secondary lymphoid organs (the spleen and lymph nodes). In the secondary lymphoid organs, APC activate antigen-specific T cells, which then differentiate into effector cells that migrate to the allograft and cause its rejection.

tissues and are activated, in an antigen-nonspecific manner, by local inflammation that is induced by infections and other noxious stimuli. A set of innate immune receptors, toll-like receptors (TLRs), plays a central role in innate immunity. These receptors are germline-encoded, transmembrane, and intracellular receptors that are critical for the detection of microbial pathogen products, often referred to as pathogen-associated molecular patterns (PAMPS). PAMPS include bacterial lipopolysaccharide, lipoproteins, peptidoglycans, flagellin, and unmethylated cytosine-guanine (cPG) nucleic acids. TLRs are a form of pathogen recognition receptors that are expressed by APCs. To date, at least 10 TLRs have been identified with individual specificity for different microbial products. Once ligated by PAMPS, TLRs initiate a signaling pathway via the universal signal adaptor protein, MyD88, which induces the translocation of nuclear factor-κB and the expression of proinflammatory cytokines (tumor necrosis factor-α, IL-1, 6, 8, and 12). These cytokines induce DCs to mature, migrate to the draining secondary lymphoid tissues, and express surface proteins essential for T-cell costimulation (such as CD40, CD80, and CD86). In secondary lymphoid organs, DCs activate naïve T cells, thus initiating the adaptive phase of the immune response. INNATE IMMUNITY IN TRANSPLANTATION Evidence supports the concept that TLRs can be activated by pathogenunrelated, endogenous signals such as heat-shock proteins, heparan sulfate, surfactant, and the contents of necrotic cells.

Therefore, it is possible that TLRs may play a role in initiating the immune response to a transplanted organ. Consistent with this hypothesis, investigators have demonstrated that rejection of minor mismatched (HY antigen-mismatched) skin allografts is critically dependent on signaling via MyD88, the adaptor protein required for TLR signaling. Specifically, they demonstrated that female mice that had no MyD88 were not able to reject male allografts from MyD88-deficient donors. Furthermore, the inability to reject these allografts was owing to reduced accumulation of mature DCs in draining lymph nodes and impaired expansion of antigraft-reactive T cells. Thus, the initiation of the alloimmune response to minor mismatched transplant antigens is dependent on innate immune mechanisms that trigger the maturation and migration of APCs. It is not clear whether the rejection of fully MHCmismatched allografts also depends on TLR signaling via Myd88. In addition, the nature of the ligands that stimulates TLRs in the transplant setting is not known. The role of other cellular components of the innate immune response (monocyte/macrophages, neutrophils and NK cells) has also been investigated in murine transplantation models. None of these cellular components has been found to be indispensable for the rejection of either allografts or xenografts. However, studies have shown that graft infiltration with these cell types precedes T-cell infiltration, consistent with the idea that they may orchestrate effector T-cell migration and function. For example, NK cells activated

658

SECTION VII / NEPHROLOGY

by non-self-antigens secrete interferon (IFN)-γ, which stimulates local production of chemokines that recruit APCs and effector T cells to the site of inflammation. Soluble components of innate immunity have also been investigated in the rejection of vascularized allografts. The complement system, a set of plasma proteins used to attack extracellular pathogens, has been shown to be critical for the rejection of renal allografts in mice. Specifically, donor allografts that were deficient in C3, a critical protein for complement activation, underwent significantly delayed allograft rejection compared to wild-type donor allografts. If the recipients were C3-deficient, wild-type allografts were rejected in a normal tempo demonstrating that local expression of complement within the graft is important for the rejection process. However, it is not clear whether complement activation plays a role in initiating the immune response (e.g., by activating and mobilizing APCs) or in the effector phase of rejection (e.g., by enhancing the recruitment of effector T cells). The complement cascade is also critical in hyperacute rejection of vascularized allografts and xenografts. In this case, recipients have preformed antibodies against donor transplant antigens. A classic example is the hyperacute rejection of ABO-incompatible allografts, whereby preformed antibodies deposit on the donor endothelium and initiate the complement cascade leading to platelet activation, thrombin deposition, and hemorrhagic infarction of the organ.

THE ADAPTIVE IMMUNE RESPONSE TO A TRANSPLANTED ORGAN The first stage in the adaptive phase of the alloimmune response is the activation of alloreactive T cells by APCs. This process occurs within the organized structures of secondary lymphoid organs (the spleen, lymph nodes, and mucosal lymphoid tissues). On activation, antigen-specific T cells begin to expand in a logarithmic fashion (the expansion phase) and acquire effector functions that allow them, with the help of B-lymphocytes and other mononuclear cells, to eliminate the foreign antigen (the effector phase). T-cell expansion, however, does not continue indefinitely but quickly halts as autoregulatory mechanisms to ensure that most effector T cells generated during the immune response are eliminated by apoptosis (the regulatory phase). The few T cells that survive the death phase become long-lived memory T cells that confer life-long protection against the foreign antigen (the memory phase). There are cellular and molecular mechanisms underlying the stages of the adaptive immune response to a transplanted organ. THE EXPANSION PHASE OF THE ALLOIMMUNE RESPONSE Allorecognition Foreign transplant antigens (alloantigens) are recognized by and activate recipient T lymphocytes via two pathways (Fig. 65-2). In the direct pathway, the recipient’s T lymphocytes are triggered by alloantigens expressed on donor APCs, commonly referred to as passenger leukocytes, which migrate out of the transplanted organ to the host’s secondary lymphoid tissues. These alloantigens are intact allo-MHC molecules complexed to endogenous peptides derived from either histocompatibility antigens or from conserved proteins common to all members of the same species (self-peptides). It is the polymporphic MHC in the latter case that confers “foreignness” to the MHC-peptide complex and leads to the activation of alloreactive T cells. In the indirect pathway, alloantigens are shed by donor cells, taken up and processed by recipient APCs, and presented to recipient T lymphocytes in the groove of self-MHC molecules. Although the direct

Figure 65-2 The direct and indirect pathways of allorecognition. Both pathways contribute to the initiation of the alloimmune response. However, the direct pathway leads to a more robust immune response by activating a larger number of T cells.

pathway appears to mount a more vigorous immune response, either pathway of allorecognition is sufficient for graft rejection. The prevailing view had been that T-cell activation occurs within the transplanted organ and does not require the presence of secondary lymphoid organs. However, studies have challenged this point. Investigators found that mice that lacked all secondary lymphoid organs (splenectomized alymphoplastic mice) did not reject fully vascularized heart allografts but ignored them as if they were self. Using different mutant strains of mice (splenectomized, lymphtoxin-α (LTα–/–), and lymphotoxin-β receptor (LTβR–/–)-deficient mice), other investigators found that allograft rejection was severely delayed in the absence of secondary lymphoid organs although it did eventually occur. The discrepancy between these two studies may be owing to the differences in the mutations present in the recipient mice. Both the LTα–/– and the LTβR–/– mice have accumulations of CD4 T cells that may be already activated and thus not require the presence of secondary lymphoid organs to initiate allograft rejection. Nonetheless, both studies demonstrate that secondary lymphoid organs are important for initiating the adapting alloimmune response. Homing of activated APCs to the spleen and draining lymph nodes following transplantation is guided by the chemokine receptor CCR7 and its ligand, secondary lymphoid chemokine. SLC is expressed on the endothelium of the afferent lymphatic system, including high endothelial venules in lymph nodes, where it directs the entry of both activated APCs and naïve T cells. APC–T cell interaction within the T cells’ zones of secondary lymphoid organs then initiates the alloimmune response. T Lymphocyte Activation The first event in APC–T cell interaction that leads to T-cell activation is the binding of the TCR to the MHC–peptide complex on the APC (Fig. 65-3). The ability of the TCR to trigger intracellular signaling in T lymphocytes is dependent on the TCR 2 (TCRz) chain and on noncovalent interactions with the membrane molecule CD3 (Fig. 65-3). The intracellular signaling pathways initiated by the TCRz chain and CD3 include protein kinases, phospholipases, and calcineurin, which is a calcium-dependent protein phosphatase. The clinically important immunosuppressive agents, cyclosporine and tacrolimus, exert their immunosuppressive effects by inhibiting calcineurin

CHAPTER 65 / MECHANISMS OF RENAL ALLOGRAFT REJECTION

Figure 65-3 T cell activation. Recognition of the foreign MHC–peptide complex by the TCR is the first step in T-cell activation, often referred to as signal 1. The TCR is dependent on the accessory molecule CD3 to deliver intracellular signals that are necessary but not sufficient for triggering T cell proliferation. Additional stimulation, often referred to as signal 2, is required for T-cell proliferation and is delivered by costimulatory molecules present on T cells. Costimulatory molecules are engaged by counter-receptors expressed on activated APCs.

action. The CD4 and CD8 molecules, expressed on the surface of distinct T cell types, often referred to as CD4 helper (Th) and CD8 cytotoxic T lymphocytes (CTL), bind to nonpolymorphic regions of MHC class-II and -I molecules, respectively, and enhance signaling via the TCR-CD3 complex (Fig. 65-3). TCR-CD3 interaction with the foreign MHC-peptide complex, however, is not sufficient for T-cell proliferation and differentiation into helper or cytotoxic cells. Additional help has to be provided by costimulatory molecules expressed on T lymphocytes (Fig. 65-3). The ligands for costimulatory molecules are present on professional APCs. These costimulatory pathways contribute to the formation of a tight “immunological synapse” between APCs and T lymphocytes. The synapse guarantees that TCR engagement by the MHC-peptide complex is sustained long enough to induce T-cell proliferation. Others have argued that costimulatory molecules provide intracellular signals that are distinct from those delivered by the TCR. Although many costimulatory molecules have been identified, the ones that are well studied and proven to be critical for the induction of alloimmunity are CD28 and the CD40 ligand (CD40L or CD154). During APC-T cell interaction, B7-1 (CD80) and B7-2 (CD86) on APCs bind to CD28 on the surface of resting T lymphocytes. CD28 crosslinking facilitates T-cell activation by recruiting membrane and intracellular kinase-rich raft microdomains to the site of TCR engagement, resulting in enhanced and sustained TCR-CD3 signaling followed by increased IL-2 production and T-cell proliferation. Proof that the CD28-B7 costimulatory pathway is critical for allograft rejection derives from studies showing markedly prolonged allograft survival in experimental animals treated with CTLA4Ig, a recombinant fusion protein that binds B7-1 and B7-2 with high affinity and prevents their interaction with CD28. Unlike CD28, the costimulatory molecule CD40L is expressed only on activated T lymphocytes. Its counter-receptor, CD40, is a glycoprotein present on the surface of professional APCs and endothelial cells. Studies indicate that the crosslinking of CD40 by CD40L mediates the immunologic functions of this costimulatory pathway. First, CD40 engagement on DCs upregulates B7-1 and B7-2 expression and induces IL-12 production. These activation steps enhance the ability of DCs to induce the proliferation of CD4+ T cells and the differentiation of CD8+ T cells

659

into CTL. Second, CD40 ligation on B-lymphocytes by T cells is essential for the generation of adaptive humoral immune responses characterized by immunoglobulin isotype switching from IgM to IgG. Third, CD40 ligation on monocytes or macrophages by T cells augments their pro-inflammatory activities, and CD40 engagement on endothelial cells by T cells upregulates cell adhesion molecules, which are essential for the migration of leukocytes into sites of inflammation. That the CD40L–CD40 costimulatory pathway plays a crucial role in allograft rejection is demonstrated by studies showing that monoclonal antibodies, which block CD40L–CD40 interaction, significantly prolong allograft survival in rodents and nonhuman primates. CD28 and CD40L are not the only costimulatory molecules present on T lymphocytes. Other membrane proteins such as 4-1BB, ICOS, OX40L, and CD70L provide costimulation for T cells, and murine studies indicate that allograft survival can be prolonged by blocking the binding of these molecules to their ligands on APCs. Cytokines Cytokines produced by APCs and activated alloreactive T lymphocytes act in an autocrine or paracrine fashion to enhance APC functions, T-cell proliferation, and T-cell differentiation into effector cells. The role of cytokines in allograft rejection has been largely inferred from their in vitro functions and from in vivo experiments in which cytokine expression is correlated with acute rejection. These studies however have led to contradictory results owing to the complexity of cytokine cascades triggered after transplantation. In vivo experiments using cytokine gene-knockout mice and cytokine-neutralizing antibodies point to two main conclusions: (1) cytokine actions are highly redundant; and (2) a single cytokine may have multiple actions and thus, influence the activation, effector, and regulatory phases of the alloimmune response. The redundancy phenomenon is supported by data showing that cytokine gene-knockout mice that lack IL-2, IL-4, or IFN-γ reject allografts vigorously, despite the fact these cytokines have indispensable immunostimulatory functions in vitro. One important exception to cytokine redundancy is that IFN-γ is necessary for the acute rejection of MHC class-II-incompatible allografts. This unique requirement for IFN-γ in the rejection process can be attributed to its essential role in inducing or upregulating MHC class-II expression on APCs and endothelial cells. The observation that a single cytokine could have multiple and sometimes opposing actions derives from experiments showing that immunostimulatory cytokines such as IFN-γ and IL-2 also have essential immunoregulatory roles in vivo. For example, IL-2 is a T-cell mitogen that also programs T lymphocytes for apoptosis. The absence of IL-2 in vivo does not lead to immunodeficiency but, instead, results in the accumulation of activated T lymphocytes, autoimmunity, and failure to accept allografts longterm. THE EFFECTOR PHASE OF THE ALLOIMMUNE RESPONSE The effector pathways that lead to allograft rejection involve a multiplicity of mechanisms, classified here according to the principal cell type that is responsible for tissue destruction. Cytotoxic T Cells CD8+ CTL recognize their targets in a MHC-restricted, antigen-specific manner. The end-result of the interaction between a CTL and an allograft is the apoptosis of target cells. Two pathways of apoptosis have been described. In the first, activated CD8+ T cells synthesize perforin and granzymes that are then stored in intracellular granules. This transformation is assisted by IFN-γ and IL-2 secreted by CD4+ Th cells. On contact with alloantigen presented in the context of MHC class-I molecules, the granules release their contents, and perforin forms pores in the target cell membrane. In addition to their ability to cause

660

SECTION VII / NEPHROLOGY

osmotic lysis of target cells, perforin-induced pores allow granzymes, a family of proteases released simultaneously with perforin from CTL granules, to enter the cytoplasm where they cause the release of cytochrome c from the mitochondria and trigger target cell apoptosis. In the second pathway, CTL induce target cell apoptosis via the FasL-Fas pathway. Fas, a member of the tumor necrosis factor (TNF) family of death receptors, is constitutively expressed on most cells whereas FasL is expressed on T cells following their activation. The binding of FasL on the CTL to Fas on the target cell leads to apoptosis via caspase-dependent pathways. Although perforin, granzymes, and FasL appear to be necessary components of the killing machinery in vitro, their role in acute allograft rejection in vivo is not obligatory. Mice that lack either perforin or FasL reject allografts at the same rate as their wildtype counterparts. The nonobligatory roles of these cytotoxic molecules point to the redundancy of the system. This may occur because of the presence of other cytotoxic molecules, for example TNF-related apoptosis inducing ligand (TRAIL), or because other cell types mediate allograft rejection in the absence of CTL-mediated cytotoxicity. Monocytes and Macrophages Monocytes that infiltrate an organ transplant differentiate into macrophages that participate in graft rejection. They form an integral part of the infiltrate that characterizes delayed type hypersensitivity (DTH) responses. DTH responses are orchestrated by CD4+ T cells. Activated CD4+ T lymphocytes stimulate macrophages via the CD40L-CD40 pathway and by secreting IFN-γ and lymphotoxin. Macrophages then release a host of mediators that include cytokines (TNF, IL-1, 6, 10, 12, and 15), chemokines, reactive oxygen species, nitric oxide, proteolytic enzymes, and extracellular matrix proteins that lead to fibrosis. Nitric oxide can be cytotoxic by itself at sufficiently high local concentrations. TNF induces expression of adhesion molecules on endothelial cells, enhances intravascular thrombosis, and activates neutrophils, eosinophils, and macrophages. Like TNF, IL-1 mediates local tissue inflammation. IL-6 stimulates the growth of activated B-cells at a late stage in their differentiation into plasma cells. IL-12 induces IFN-γ production by NK and T cells and enhances the generation of CTLs. IL-15 serves as a mitogen to NK and T cells, particularly CD8+ memory T cells. Natural Killer Cells NK cells are a lymphocyte subset, which expresses neither T- nor B-lymphocyte markers. They are activated by IL-2, IL-15 and type-1 IFNs. They release IFNγ on stimulation by IL-12. IFN-γ in turn induces chemokine secretion by cells of the graft, particularly endothelial cells, leading to the recruitment of more activated T cells. Inflammatory chemokines shown to play a significant role in allograft rejection include CXCR3 ligands (IP-10, MIG, and I-TAC) and to some extent CCR5 ligands (the MIP-1 family of chemokines). Like CTLs, NK cells kill target cells by releasing perforins and granzymes. In addition, they secrete TNF, which induces apoptosis of cells expressing the TNF receptor. NK cells recognize their target cells in two different ways. First, by expressing low affinity receptor for the constant (Fc) portion of IgG, they bind to and kill antibody-coated target cells (antibody-dependent cell-mediated cytotoxicity or ADCC). Second, NK cells also express receptors that recognize MHC class-I molecules complexed to self peptides. Many of these receptors, however, are inhibitory such that NK cells are shut down when they engage self MHC proteins. This implies that NK cells react to non-self by recognizing the absence of self-peptides in the grooves of MHC class-I molecules. Although NK cell depletion does not significantly alter acute rejection, they may play a subtle role in chronic rejection.

B Cells and Alloantibodies Antibodies directed against donor HLA molecules (alloantibodies) pose the greatest threat to a transplanted organ. Alloantibody generation is generally the consequence of sensitization. Sensitization occurs in patients who have been previously exposed to human HLA proteins in the form of transfusions, pregnancy or prior organ transplants. B cells that recognize specific HLA molecules via their B-cell receptor for antigen (membrane-bound immunoglobulins), internalize the HLA-antibody complex, cleave it into allopeptides, and present it in the context of MHC class-II molecules to CD4+ T cells. The production of IL-2, IL-4, and IL-5 by activated CD4+ T cells and T–B interactions via CD40–CD40L leads to B-cell maturation, proliferation, isotype switching, and the production of harmful alloantibodies. The most dramatic example of antibody-mediated allograft damage is hyperacute rejection, initiated by preformed donor-specific antibodies present in the recipient at the time of transplantation. Preformed antibodies that cause hyperacute rejection include anti-ABO and anti-HLA IgG antibodies, particularly those targeted against class-I HLA. Complement and coagulation cascades are activated leading to widespread endothelial damage, intravascular thrombosis, and graft necrosis. Alloantibodies cause tissue damage by activating the complement cascade or by mediating ADCC. In nonsensitized recipients, humoral rejection has been described as a cause of both acute and chronic rejection, but their relative importance is not known. B-cell-deficient mice, for example, are capable of vigorous acute allograft rejection. Chronic transplant arteriosclerosis, on the other hand, does not develop in B-cell-deficient mice, suggesting that alloantibodies play an important role in chronic rejection. In humans, humoral renal allograft rejection is diagnosed by staining for the complement component C4d in biopsy specimens. Eosinophils Eosinophils are recruited into the allograft by the cytokines IL-4, 5, and 13, which are produced by Th2-type CD4+ T cells. Th2-mediated cellular rejection can occur in experimental models in the absence of CD8+ or CD4+ Th1 cells and is characterized by an eosinophilic infiltrate. The clinical significance of this finding is not known, but hypereosinophilia has been reported to precede acute rejection, and activated eosinophils and IL-5 expression have been observed in acutely rejecting allografts. THE REGULATORY PHASE OF THE ALLOIMMUNE RESPONSE Expansion of antigen-specific lymphocytes and their differentiation into effectors in response to foreign antigen is followed by rapid clonal contraction. Deletion of the majority of activated/effector lymphocytes ensures that collateral damage to the host does not occur during immune responses. Elimination of the antigen that initially triggered lymphocyte stimulation is the ultimate means by which primary immune responses are terminated. Several down-regulatory mechanisms, however, participate in cases of large or persistent antigen loads such as a transplanted organ. These mechanisms include cell surface molecules and cytokines, which regulate proliferation and survival of activated T lymphocytes. There is also evidence that regulatory T cells that suppress immune responses may exist. Activation-Induced Cell Death Apoptosis of activated and effector T cells is perhaps the most conspicuous means by which immune responses are regulated. Repeated stimulation of T lymphocytes by antigen results in the apoptosis of activated T cells instead of their continued proliferation. This phenomenon is known as antigen- or activation-induced cell death (AICD). AICD is in large part mediated by interaction between the cell surface

661

CHAPTER 65 / MECHANISMS OF RENAL ALLOGRAFT REJECTION

molecule Fas (CD95), a “death” receptor, and its ligand (FasL). Antigenic stimulation upregulates Fas expression and induces de novo FasL expression on T lymphocytes. FasL-Fas binding then leads to apoptosis of activated T lymphocytes. Other membrane molecules that trigger T cell apoptosis include members of the TNF receptor (TNFR) family such as CD30, TNF receptor-related apopotosis-mediating protein, and the TRAIL receptor (DR4). Among cytokines that act on T cells, IL-2 is essential for AICD, a function not shared by other T cell mitogenic cytokines such as IL-4, 7, 9, and 15, which also engage the common cytokine receptor γ-chain (γc). Although IL-2 is a potent mitogen to resting CD4+ and CD8+ T lymphocytes, it programs these cells to undergo apoptosis when restimulated with antigen. It does so by upregulating death-transducing receptors such as Fas and TNFR and by downregulating death-inhibiting molecules such as FLIP and γc. IL-2deficient mice are not capable of long-term allograft acceptance because of impaired alloantigen-induced apoptosis of activated T cells. Similarly, allograft acceptance cannot be achieved in IFNγ-deficient mice because of exaggerated T-cell expansion in response to alloantigen. It is still debated whether IFN-γ regulates immune responses by promoting alloantigen-induced apoptosis of activated T cells, by limiting their proliferation, or both. The cytotoxic molecule perforin also contributes to the apoptosis of activated T cells by effecting the suicide or fratricide of both CD8+ and CD4+ T cells. Perforin gene-knockout mice have dysregulated immune responses and are unable to accept allografts even with potent immunosuppression. Inhibition of T-Cell Proliferation and Differentiation Another mechanism by which immune responses are regulated is the inhibition of T-cell proliferation. The T-cell surface molecule CTL antigen-4 (CTLA4 or CD152) has emerged as an important inhibitor of activated T-cell proliferation. CTLA4 is homologous to CD28 and shares the same ligands, B7-1 and B7-2. Unlike CD28, CTLA4 is expressed on the plasma membrane only on activated T cells, has a much higher affinity to B7 proteins, and delivers a negative signal that turns off the proliferation of activated T cells. Thus, early in the immune response B7-1 and B7-2 interact with CD28 receptors on naïve T cells to effect costimulation, whereas later the high-affinity interaction between B7 proteins and CTLA4 on activated T cells leads to inhibition of cell cycling. The CTLA4B7-negative feedback pathway is essential for the optimal induction of peripheral tolerance to protein antigens and to transplanted organs. A regulatory cytokine of paramount importance, TGFβ, is a potent inhibitor of activated T-cell differentiation into effector cells. TGFβ has been shown to downregulate alloimmune responses in several murine transplantation models. Suppressor T Cells There has been an explosive resurgence of the concept that regulatory/suppressor T cells that modulate immune responses exist. The most commonly described suppressor phenotype is the CD4+ T cell constitutively expressing CD25+ on its surface, often referred to as Treg. Other regulatory populations have also been described, including Tr1 cells that seem to suppress immune responses by secreting TGF-β and IL-10. In addition, CD8+ regulatory/suppressor T cells may exist. CD4+CD25+ Treg cells appear to mediate their suppressive effects by direct cell–cell interaction but the exact mechanism by which they inhibit the proliferation of other T cells is not known. An increasing body of evidence suggests that Treg have a role in maintaining self-tolerance, neonatal tolerance, and long-term allograft acceptance. A transcriptional regulator, Foxp3, which is uniquely associated with Treg has been identified. Genetic deletion of Foxp3 leads to absence of Treg

Table 65-1 Advantages of Memory T Cells Over Naïve T Cells Characteristic Survival

Migration Response and effector function

Naïve T cell Months to years Dependent on TCR interaction with MHC/antigen Secondary lymphoid tissues Requires stimulation with antigen

In vivo protection Long-lived but against foreign relatively inefficient antigen

Memory T cell Years to lifetime Not dependent on TCR interaction with MHC/antigen Both lymphoid and nonlymphoid tissues Requires restimulation with antigen but is faster and larger in magnitude than naïve T cells Long-lived and efficient

AICD, activation-induced cell death; MHC, major histocompatibility complex; TCR, T-cell receptor for antigen.

and massive expansion of autoreactive, pathogenic T cells. The generation of suppressor T-cell clones may be the result of interaction of naïve T cells with immature DCs or the elusive plasmacytoid DC, DC2. THE MEMORY PHASE OF THE ALLOIMMUNE RESPONSE A cardinal feature of the adaptive immune response is the generation of memory lymphocytes. Although the vast majority of effector T cells undergoes apoptosis as the primary immune response dissipates, few lymphocytes survive to become long-lived memory T cells. Memory T cells that recognize microbial antigens provide the organism with long-lasting protection against potentially fatal infections. Conversely, memory T cells that recognize donor alloantigens jeopardize the survival of life-saving organ transplants by mediating both acute and chronic rejection. Memory T cells have several inherent advantages over their naïve counterparts that endow them with the ability to clear a foreign antigen, whether a microbial pathogen or an allograft, in a vigorous fashion (Table 65-1). The first advantage is that their response to a foreign antigen (recall response) is greater in magnitude and faster than the naïve T-cell response. Memory T cells generate a considerable number of effector T cells, capable of cytokine secretion cytolytic activity, or both, within hours of antigenic restimulation, whereas naïve T cells generate a smaller number of effectors at a much slower pace (days). Second, memory T cells have a survival advantage over their naïve counterparts. Antigen-specific memory T-cell populations persist for years to lifetime in humans and their survival appears to be antigen- and MHC-independent. Mature naïve T-cell populations also persist for a relatively long period of time (months to few years in humans) but their survival is dependent on constant, low-grade stimulation with MHC–self-peptide complexes. Third, the circulation of naïve T cells is restricted to secondary lymphoid tissues, the site where they are activated by foreign antigens presented by APC; memory T cells circulate through both secondary lymphoid tissues and peripheral nonlymphoid tissues. Unlike naïve T cells, memory T cells can directly encounter foreign antigen and mount a productive immune response within nonlymphoid tissues. The migratory advantage of memory T cells, therefore, allows them to detect and eliminate a foreign intruder substantially before it reaches secondary lymphoid tissues. The activation, survival, and migration advantages of memory T cells

662

SECTION VII / NEPHROLOGY

confer the organism with enhanced protection against foreign antigens long after the primary immune response has dissipated. The same advantages of memory T cells that make them efficient at eliminating microbial pathogens enable them to rapidly reject foreign tissues. Unlike inbred mice, outbred animals including humans, harbor a significant number of memory T cells that are alloreactive. These memory T cells generally arise if an individual is exposed to alloantigens via pregnancy, blood transfusion, or a previous organ transplant. However, alloreactive memory T cells also exist in individuals who have never been exposed to foreign tissues. It is proposed that such alloreactivity in the T-cell memory pool is caused largely by cross-reactivity; repeated exposure to viral and bacterial antigens over time leads to the development of memory T cells that also recognize alloantigens. This phenomenon, commonly referred to as heterologous immunity, has been demonstrated in murine infection models, and is supported by experimental findings in humans. Alloreactive memory T cells, whether generated by previous exposure to alloantigens or microbial infection, can cause acute and chronic rejection and interfere with the induction of immunologic tolerance (donor-specific unresponsiveness).

CONCLUSIONS Despite advances in understanding the cellular and molecular mechanisms of allograft acceptance, long-term survival of transplanted organs remains an important clinical challenge. Moreover, transplantation tolerance and allograft acceptance in the absence of continuous immunosuppression have been elusive clinical goals. The reason for these seemingly insurmountable challenges stems from the properties of the alloimmune response. These properties include the large number of alloreactive T cells present in a given individual, the limitations of immunoregulatory mechanisms responsible for turning off the alloimmune responses, and

the fact that immune responses to foreign antigens, by virtue of evolutionary design, are destined to generate immunologic memory. The latter property is perhaps the most formidable obstacle to long-term allograft acceptance and transplantation tolerance. Therefore, an important but unmet challenge in clinical transplantation is whether selective deletion or suppression of alloreactive memory T cells can be accomplished without globally compromising the host’s immune system.

SELECTED REFERENCES Adams A, Williams A, Jones T, et al. Heterologous immunity provides a potent barrier to transplantation tolerance. J Clin Invest 2003;111: 1887–1895. Chalasani G, Lakkis FG. Immunologic ignorance of organ allografts. Curr Opin Organ Transplant 2001;6:83–88. Dai Z, Lakkis FG. The role of cytokines, CTLA-4, and costimulation in transplant tolerance and rejection. Curr Opin Immunol 1999;11:504–508. Goldstein D, Tesar B, Akira S, Lakkis F. Critical role of the Toll-like receptor signal adaptor protein Myd88 in acute allograft rejection. J Clin Invest 2003;111:1571–1578. Kaech SM, Wherry EJ, Ahmed R. Effector and memory T cell differentiation: Implications for vaccine development. Nat Rev Immunol 2002; 2:251–262. Lakkis F. Where is the alloimmune response initiated? Am J Transplant 2003;3:241–242. Lakkis FG. Role of cytokines in transplantation tolerance: lessons learned from gene-knockout mice. J Am Soc Nephrol 1998;9:2361–2367. Lakkis FG, Arakelov A, Konieczny BT, Inoue Y. Immunologic ‘ignorance’ of vascularized organ transplants in the absence of secondary lymphoid tissue. Nat Med 2000;6:686–688. Salama A, Remuzzi G, Harmon W, Sayegh M. Challenges to achieving clinical transplantation tolerance. J Clin Invest 2001;108:943–948. Sayegh MH, Turka LA. The role of T-cell costimulatory activation pathways in transplant rejection. N Engl J Med 1998;338:1813–1821. Womer KL, Velle JP, Sayegh MH. Chronic allograft dysfunction: mechanisms and new approaches to therapy. Semin Nephrol 2000;20:126–147.

MUSCULOSKELETAL VIII SECTION EDITOR:

DANIEL J. GARRY

Abbreviations VIII. MUSCULOSKELETAL ANC-1 bHLH CaM CaMK CDK CKI CPK CsA DGC ECM ES FGF FKHR FKRP FSHD GSK-3β HDAC5 HGF/SF HLH Id IGF1 IGF2 IGF-2 IGFs IRS LGMD

LOH MEF2 MEF2 MGF mTOR NFAT NMJ nNOS NRF-1 PABP2 PCR PGC-1

Nuclear-anchoring protein in C. elegans basic helix–loop–helix calmodulin Ca2+/CaM-dependent protein kinase cyclin-dependent protein kinase cyclin-dependent protein kinase inhibitor creatine phosphokinase cyclosporin dystrophin glycoprotein complex extracellular matrix embryonic stem fibroblast growth factor fork head homolog rhabdomyosarcoma fukutin-related protein facioscapulohumeral muscular dystrophy glycogen synthase kinase-3β histone deacetylase 5 hepatocyte growth factor/scatter factor helix–loop–helix inhibitor of differentiation insulin growth factor 1 insulin growth factor 2 insulin-like growth factor II insulin-like growth factors intergroup Rhabdomyosarcoma Study limb-girdle muscular dystrophy

PI-3K PKC POMGnT1 POMT1 PPAR-γ Rb SEPN1 SERCA Shh SP SR T TGF-β

664

loss of heterozygosity muscle-enriched factor 2 myocyte enhancer binding factor-2 mechano growth factor mammalian target of rapamycin nuclear factor of activated T cells neuromuscular junction neuronal nitric oxide synthase nuclear respiratory factor poly(A) binding protein 2 polymerase chain reaction peroxisome proliferator-activated receptor γ coactivator-1 phosphoinositide 3-kinase protein kinase C O-mannose β-1,2-N-acetylglucosaminyltransferase O-mannosyltransferase peroxisome proliferator-activated receptor γ retinoblastoma protein selenoprotein N sarcoplasmic reticulum calcium ATPase pump sonic hedgehog side population sarcoplasmic reticulum transverse transforming growth factor-β

66 Muscle Development and Differentiation ERIC N. OLSON SUMMARY Skeletal muscle is formed through the differentiation and fusion of myoblasts into a multinucleated syncytium. Advances in molecular biology techniques and the generation of transgenic mouse models have contributed to the identification of the molecular pathways involved in skeletal muscle growth and differentiation. During embryogenesis, the specification and development of the myogenic lineage is determined by the spatiotemporal expression of a family of muscle-specific transcription factors. Understanding the molecular mechanism controlling skeletal muscle development may lead to new insights into therapeutic strategies that may alleviate some of the pathological conditions associated with muscular diseases. Key Words: bHLH proteins; growth factors; MEF2 proteins; MRF4; Myf5; myoblast; MyoD; myogenin; myogenesis; skeletal muscle; somites.

INTRODUCTION Vertebrate species contain dozens of different skeletal muscles, each with unique positions, sizes, shapes, contractile properties, and patterns of innervation. Dramatic progress has been made toward understanding the genetic pathways that control the formation and patterning of skeletal muscle during embryogenesis. The discovery of several families of genes that act at different steps in the developmental pathway leading to skeletal muscle formation has provided a framework for understanding the complexity of this tissue and will undoubtedly yield insight into the causes of several neuromuscular disorders, as well as provide the potential for therapeutic intervention into such diseases.

REGULATION OF MUSCLE DIFFERENTIATION IN VITRO Much of the progress in understanding muscle development is attributable to the fact that many of the events associated with muscle differentiation in vivo can be recapitulated in tissue culture. Immortalized muscle cell lines or skeletal myoblasts isolated from embryos proliferate in vitro in the presence of fetal bovine serum and other mitogens such as fibroblast growth factor. Myoblasts are committed to the myogenic lineage, but they do not express muscle structural genes until they are forced to exit the cell cycle. Myoblast differentiation is accompanied by irreversible withdrawal From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

from the cell cycle, fusion to form multinucleate myotubes, and expression of a large array of muscle-specific genes whose products are required for the specialized contractile, excitable, and metabolic properties of the differentiated muscle fiber. Several studies suggested the existence of myogenic regulatory factors with the potential to activate muscle gene expression. Experiments in which heterokaryons were formed between skeletal muscle cells and various nonmuscle cells demonstrated that muscle genes became activated in nonmuscle cell nuclei and suggested the existence of trans-acting factors in muscle cells, which could activate muscle gene expression. Genetic evidence for such myogenic factors was provided by experiments in which the fibroblast cell line 10T1/2 was shown to form skeletal muscle following treatment with the demethylating agent 5′-azacytidine. The frequency of conversion of 10T1/2 cells to muscle (up to 50%) suggested that one gene was responsible for establishing the myogenic phenotype following its activation by demethylation. This hypothesis was confirmed by DNA transfection experiments in which genomic DNA from myoblasts or 5′-azacytidine-treated 10T1/2 cells was shown to convert 10T1/2 cells to myoblasts with a frequency consistent with a single regulatory gene.

THE MYOD FAMILY OF MUSCLE-SPECIFIC TRANSCRIPTION FACTORS The first myogenic regulatory gene to be cloned was MyoD, which was isolated by cloning mRNAs that were expressed in myoblasts, but not 10T1/2 fibroblasts. When a cDNA clone encoding MyoD was expressed ectopically in fibroblasts, it was found to activate skeletal muscle gene expression (Fig. 66-1). Subsequently three related genes, myogenin, Myf5, and MRF4 (also called Myf6 and herculin) were discovered. These genes, which are referred to as the MyoD family, are expressed exclusively in skeletal muscle; there is no detectable expression in cardiac or smooth muscle, even though the three muscle cell types express many of the same muscle-specific genes. Members of the MyoD family exhibit distinct expression patterns during differentiation of muscle cells in culture. Most established muscle cell lines express either MyoD or Myf5 when they are proliferating as undifferentiated myoblasts. When induced to differentiate by exposure to growth factor-deficient medium, myogenin is rapidly upregulated immediately before activation of muscle structural genes. MyoD and Myf5 also continue to be expressed in differentiated myotubes. MRF4 is not expressed until late in the differentiation program.

665

666

SECTION VIII / MUSCULOSKELETAL

Figure 66-2 Diagrammatic representation of the four myogenic bHLH proteins. A linear representation of each myogenic regulatory factor is shown. The basic (+++) and HLH regions are indicated. The number of amino acids in each protein is shown at the end of each box.

genes. These autoregulatory interactions have been proposed as a mechanism whereby these factors amplify and maintain their expression and thereby commit cells to a myogenic fate.

THE BASIC HELIX–LOOP–HELIX FAMILY OF TRANSCRIPTION FACTORS

Figure 66-1 Activation of MHC expression in fibroblasts expressing exogenous MyoD. 10T1/2 fibroblasts were transfected with a MyoD expression vector. Cells were then stained using antibodies specific for MyoD (red) and myosin heavy chain (green). MyoD can be seen within the nuclei of transfected cells and myosin is expressed only in the cells that express MyoD. Figure provided by S. Tapscott, Fred Hutchinson Cancer Research Center. (Please see color insert.)

Members of the MyoD family can induce skeletal muscle gene expression when they are introduced into a wide variety of cell types including fibroblasts, melanoblasts, chondroblasts, smooth and cardiac muscle cells, and osteoblasts. In some cell types, these factors activate the entire myogenic program, resulting in formation of multinucleate myotubes and expression of the full array of muscle-specific genes, whereas in other cell types, only a subset of muscle genes is expressed. In some cell types, such as adipocytes, forced expression of the myogenic factors represses the endogenous program of cell-specific gene expression, whereas in other cell types the skeletal muscle program is coexpressed with the endogenous program. There are also cell types, such as hepatocytes and several types of transformed cells, in which the myogenic factors are unable to activate muscle gene expression. The failure of the myogenic factors to activate muscle gene expression in these cell types suggests that they lack certain cofactors required by the myogenic factors, that they contain inhibitors of muscle gene expression, or both. In addition to activating the expression of muscle structural genes, members of the MyoD family regulate each others’ expression. For example, if MyoD is introduced into fibroblasts, it activates the endogenous MyoD gene as well as the myogenin and MRF4

Members of the MyoD family share about 80% amino acid identity within a 70-amino acid segment that encompasses a region rich in basic amino acids, followed by a region postulated to adopt a helix–loop–helix (HLH) conformation in which two amphipathic α-helices are separated by an intervening loop (Fig. 66-2). Related, but more divergent, basic-HLH (bHLH) motifs are found in members of a superfamily of proteins that regulate cell proliferation and differentiation in species ranging from yeast to humans. Among these are members of the c-myc family of oncoproteins and the products of several Drosophila genes that regulate embryonic cell fates, including the achaete-scute gene products, which regulate neurogenesis, and twist, which regulates mesoderm formation. The HLH motif mediates protein dimerization and brings together the basic regions of bHLH proteins to form a bipartite DNA binding domain that recognizes the consensus sequence CANNTG (N = any nucleotide), known as an E-box (Fig. 66-3). This DNA sequence motif is found in the control regions of numerous muscle-specific genes, in which it binds the myogenic factors and mediates transcriptional activation. The myogenic bHLH factors homodimerize inefficiently, but they readily form heterodimers with members of a family of ubiquitous bHLH proteins, known as E-box-binding proteins. There are also HLH proteins that lack basic regions and inhibit the activity of bHLH proteins by forming heterodimers incapable of binding DNA. Among this class of HLH proteins are members of the inhibitor of differentiation (Id) family. Mutagenesis studies have revealed several functional domains in the myogenic bHLH proteins (Fig. 66-4). These factors contain transcription activation domains in their amino- and carboxyl termini that are important for efficient activation of muscle-specific transcription. The basic regions of the myogenic factors play a dual role in the control of muscle gene expression by mediating DNA binding and by conferring muscle specificity to transcriptional activation. Fine mapping of individual amino acids in the DNA-binding domains of the myogenic factors has shown that two clusters of basic amino acids are required for binding to the E-box consensus sequence (Fig. 66-4). In addition, two adjacent

CHAPTER 66 / MUSCLE DEVELOPMENT AND DIFFERENTIATION

667

Figure 66-4 Schematic representation of the myogenin protein. Myogenin is a 224-amino acid protein with the bHLH motif near the center. The 12 amino acids within the basic region are required for sequence-specific DNA binding in conjunction with bHLH motif. The alanine–threonine in the center of the basic region is conserved in and specific to all members of the MyoD family and are important for activation of muscle-specific transcription. The threonine is also a PKC phosphorylation site, which inhibits DNA binding when phosphorylated. S, S/T, serine- and serine/threonine-rich regions, respectively.

GROWTH FACTOR CONTROL OF MYOGENESIS Figure 66-3 Schematic representation of a MyoD/E12 heterodimer is shown. The basic region and helix-1 form a contiguous α-helix separated from helix-2 by an intervening loop. Each basic region recognizes half of the palindromic sequence CANNTG.

amino acids, alanine–threonine, in the center of the DNA-binding domains, are required for activation of muscle gene expression. These residues are conserved in the basic regions of all known myogenic bHLH proteins in species ranging from Drosophila and sea urchins to humans, and they are absent from the basic regions of the more than 40 other bHLH proteins described. The specificity and conservation of these residues suggest that they represent part of an ancient mechanism for activation of muscle gene expression. If the alanine–threonine residues in the basic region of a myogenic bHLH protein are replaced with the amino acids found at that position in the basic regions of E proteins, the resulting mutant protein will bind DNA normally, but is unable to activate muscle gene expression. Conversely, if these residues in addition to a third amino acid between the basic region and helix-1 of the myogenic factors are introduced in the corresponding positions in E proteins, they confer on the E proteins the ability to activate muscle gene expression. The exact mechanism whereby amino acids in the basic region direct muscle-specific transcription is unclear. However, one possibility is that these amino acids induce an allosteric change in the myogenic factors following DNA binding, which might expose a protein interface that can interact with accessory factors required for the activation of muscle gene expression. In this regard, the crystal structure of a MyoD homodimer binding to an E-box DNA sequence has been deduced. The structure of the complex confirms that the basic region and first helix of the HLH region form a continuous α-helix. Alanine–threonine in the MyoD basic region lie within the major groove of the DNA binding site and are, therefore, inaccessible to other proteins. However, the interaction of these residues with the DNA appears to induce a conformational change in the protein, which may influence the proteins with which it can interact.

As in many cell types, differentiation of myoblasts is coupled to withdrawal from the cell cycle. Whereas many cell types can re-enter the cell cycle when stimulated with mitogens, skeletal muscle cells lose the ability to reinitiate DNA synthesis when they differentiate and, therefore, become irreversibly committed to the postmitotic state. Because myoblasts express MyoD or Myf5 (these factors are unable to induce muscle gene expression when cells are exposed to growth factors, however), there must be posttranslational mechanisms that inhibit their activities. There appear to be multiple mechanisms whereby growth factors prevent myogenic bHLH proteins from activating muscle-specific genes. For example, Id proteins are expressed at high levels in proliferating undifferentiated myoblasts and are downregulated when myoblasts are induced to differentiate. Because Id proteins dimerize preferentially with E proteins, they sequester the dimerization partners for myogenic bHLH proteins in myoblasts. The downregulation of Id has been proposed to release E proteins so that they can dimerize with myogenic bHLH proteins and activate muscle-specific genes. Consistent with this model, forced expression of Id from an expression plasmid is sufficient to prevent myoblast differentiation under conditions that would normally promote differentiation. In the embryo, Id expression is also downregulated in regions of skeletal muscle differentiation. The nuclear oncogene products Fos and Jun, which are upregulated by mitogens, can also block muscle differentiation by interfering with the activities of myogenic bHLH proteins. The inhibitory activities of Fos and Jun appear to be mediated by competition with the myogenic factors for interaction with a common cofactor required for myogenesis and by direct interactions with the myogenic factors. Members of the MyoD family have also been shown to be targets for several protein kinases. Protein kinase C (PKC), which is activated by growth factors, can substitute for growth factors and block myoblast differentiation. The threonine residue in the center of the DNA-binding domains of the myogenic factors, which mediates muscle-specific gene activation, has been shown to be efficiently phosphorylated by PKC in vitro and in vivo. In the case of myogenin, this phosphorylation prevents DNA

668

SECTION VIII / MUSCULOSKELETAL

Figure 66-5 Model depicting the interaction between the cell cycle machinery and the myogenic regulatory pathway. MyoD or Myf5 are expressed in myoblasts and induce the expression of the CDK inhibitor p21 in the absence of growth factors. p21 induces growth arrest, which results in myogenin expression. Myogenin in turn activates muscle differentiation. Growth factor signals induce cyclin D1, which activates CDK4, resulting in phosphorylation of the Rb pocket, as well as inhibition of MyoD function. Phosphorylated Rb is unable to bind E2F and is, therefore, inactive as a growth suppressor. In the absence of growth factor signals, Rb is dephosphorylated and binds to E2F. E2F normally activates expression of genes required for cell proliferation, but it is in activated when bound to Rb. Rb, retinoblastoma protein.

binding, presumably because it introduces a negative charge into the center of the DNA-binding domain, resulting in electrostatic repulsion from the DNA. Fibroblast growth factor can induce phosphorylation of this threonine in cultured cells, resulting in repression of myogenin’s ability to activate muscle gene expression. The ability of this threonine to serve as a target for growth factor-dependent phosphorylation pathways that block myogenesis demonstrates that this single amino acid plays a dual role in the control of muscle development; it is structurally important for transcriptional activation, and it serves as a target for growth factor signal transduction pathways under control of PKC. Type-β transforming growth factor (TGF)-β also inhibits muscle differentiation by blocking the activities of the myogenic factors. In myoblasts exposed to TGF-β, myogenic bHLH proteins are expressed, and they can bind to their target DNA sequences associated with muscle-specific genes but are unable to activate expression of those genes. It has been proposed that TGF-β interferes with the expression or activity of a cofactor required by the myogenic factors to activate myogenesis. The linkage of muscle differentiation to cell cycle withdrawal suggests that the cell cycle machinery interfaces with the mechanism for muscle gene activation. The retinoblastoma protein (Rb) has been shown to be an important regulator of cell cycle progression. Rb is a nuclear phosphoprotein that is phosphorylated in a cell cycle-dependent manner. In cycling cells, Rb is phosphorylated in a specific domain known as the “pocket,” whereas in growtharrested cells Rb is dephosphorylated. When quiescent cells are stimulated to re-enter S phase, Rb becomes phosphorylated during the G1/S phase transition before initiation of DNA synthesis. The pocket of Rb interacts with a variety of nuclear proteins to control cell cycle progression. Among these is the transcription factor E2F, which activates the expression of several genes involved in cell proliferation. Binding of E2F to dephosphorylated Rb inactivates E2F and results in a block to proliferation (Fig. 66-5). Conversely, phosphorylation of Rb releases E2F so that it can activate cell growth. The oncoproteins encoded by several DNA tumor viruses, such as SV40 large T antigen and adenovirus E1A, also bind to the pocket of Rb, which prevents interaction between Rb and E2F, resulting in uncontrolled proliferation. These oncogenes are potent inhibitors of myogenesis and when expressed in terminally differentiated myotubes can lead to reinitiation of DNA synthesis. MyoD has also been shown to bind to the pocket of dephosphorylated Rb. This interaction requires the basic region and first helix of MyoD and is dependent on the alanine–threonine residues

in the basic region that are required for activation of muscle gene expression. Because MyoD and E2F bind the same region of Rb, they must interact with separate molecules of Rb. Exactly how MyoD binds DNA and Rb simultaneously is unclear. The role of Rb in the MyoD–DNA complex also remains to be determined. Rb is upregulated during myogenesis and is required for irreversible exit of muscle cells from the cell cycle. Myoblasts lacking Rb are able to form myotubes and express muscle-specific genes, but they do not irreversibly exit the cell cycle and can resynthesize DNA in response to mitogenic stimulation. The pocket of Rb is phosphorylated by cyclin-dependent protein kinases (CDKs). The CDKs are expressed constitutively throughout the cell cycle but are activated by association with regulatory proteins called cyclins. Cyclin D1 is induced during the G1 to S phase transition and activates CDK4, resulting in phosphorylation of Rb. Forced expression of cyclin D prevents activation of muscle gene expression by MyoD, presumably because it leads to inactivation of Rb, although a more direct role of the cyclin D1/CDK4 complex in phosphorylating MyoD has not been excluded. There is also a family of CDK inhibitors (CKIs) that act as inhibitors of cell growth by blocking CDK activity. One of these CKIs, called p21/WAF1/CIP1, is upregulated during myogenesis and is likely to play a role in locking cells out of the cell cycle. When CKI is expressed in proliferating myoblasts, it can inhibit growth and induce muscle differentiation even in the presence of high concentrations of mitogens (Fig. 66-5); p21 is also expressed in differentiating muscle cells during embryogenesis. Upregulation of p21 expression is also observed in several differentiated cell types in the embryo, suggesting it may be part of a common mechanism linking cell-cycle withdrawal to activation of cell differentiation.

EMBRYONIC ORIGINS OF SKELETAL MUSCLE The formation of skeletal muscle during embryogenesis involves the commitment of mesodermal stem cells to the skeletal muscle lineage and the subsequent differentiation of myoblasts to form multinucleate myotubes, which mature into skeletal muscle fibers. Skeletal muscle is derived from the somites, which form in a rostral-to-caudal gradient by segmentation of the paraxial mesoderm along the neural tube (Fig. 66-6). Newly formed somites appear as an epithelial sphere, which subsequently becomes compartmentalized to form the dermamyotome and the sclerotome. The sclerotome gives rise to the ribs and vertebrae. The

CHAPTER 66 / MUSCLE DEVELOPMENT AND DIFFERENTIATION

669

Figure 66-6 Diagrammatic representation of somite maturation. Schematic representation of an early chick embryo is shown on the left. Somites form in a rostral-to-caudal progression by segmentation of the segmental plate mesoderm. Right: cross-sections of the embryo at the indicated level. (A) Paraxial mesodermal cells in the segmental plates have not yet formed somites. (B) Newly formed somites appear as epithelial spheres. (C) Immature somites have become compartmentalized to form the dermamyotome and sclerotome. (D) Mature somites have formed the dermatome, myotome, and sclerotome. (Please see color insert.)

dermamyotome gives rise to the myotome and dermatome, which form axial muscles of the back and the dermis, respectively. Myf5 is the first member of the MyoD family to be expressed during mouse embryogenesis, appearing in the somite at embryonic d 8 (E8). Myogenin transcripts appear in the myotome by E8.5 and MRF4 and MyoD are expressed at E9 and E10.5, respectively. Muscle cells in the limbs arise from cells that emigrate from the ventrolateral edge of the dermamyotome. These migrating cells are committed to the myogenic lineage, but they do not express the myogenic bHLH factors until they reach the limbs. The homeobox gene Pax3 is expressed in the migrating myogenic precursor cells. Evidence that Pax3 is involved in migration or commitment of these cells to the myogenic lineage has been provided by analysis of a mouse mutant called Splotch. The Splotch mutation maps to the Pax3 gene and results in the absence of limb muscle. Pax3 may mediate activation of MyoD and Myf5 in response to muscleinducing signals from overlaying ectoderm. Interestingly, a chromosomal translocation resulting in structural rearrangement of the human Pax3 gene has been associated with alveolar rhabdomyosarcoma, a malignant tumor of skeletal muscle (see Chapter 71). Whereas limb muscles are derived from myogenic precursor cells that migrate from the ventrolateral edge of the dermamyotome, the axial muscles of the back are derived from the myotomal compartment of the somites. The neural tube and notochord have been shown to play important roles in somite maturation by serving as a source for signaling molecules that induce patterning of the somites. In the absence of neural tube, the myotome fails to form. In contrast, the formation of limb muscles is independent of the neural tube or notochord.

Relatively little is known about the developmental underlying formation of head skeletal muscles. Head muscles are derived from multiple cell lineages, including prechordal mesoderm anterior to the first somites and paraxial mesodermal precursors that migrate into the branchial arches. Two related bHLH proteins, called MyoR and capsulin, are transiently expressed in migratory paraxial mesodermal cells in the branchial arches. The expression of these bHLH genes precedes that of the MyoD family in head muscles. Several of the myogenic bHLH genes have been analyzed for cis-acting DNA sequences that direct expression in the early somites and subsequently during skeletal muscle formation. The best characterized of these genes is myogenin, which is controlled by DNA sequences in the proximal promoter region. When these sequences are linked to the β-galactosidase gene as a marker and are introduced into transgenic mice, they direct the expression of β-galactosidase in the exact pattern as the endogenous myogenin gene (Fig. 66-7). Within the myogenin gene control region is an E-box that binds the myogenic bHLH proteins with high affinity and a binding site for the myocyte enhancer-binding factor-2 (MEF2) family of transcription factors. Together, these two families of regulators control the expression of myogenin in the somites and limb buds.

A GENETIC PATHWAY FOR MUSCLE DEVELOPMENT REVEALED BY GENE TARGETING IN TRANSGENIC MICE To determine the specific roles of the myogenic bHLH genes in muscle development in vivo, these genes have been deleted from the genome of transgenic mice by homologous recombination.

670

SECTION VIII / MUSCULOSKELETAL

Figure 66-7 Expression pattern of a myogenin-lacZ transgene in an 11.5-d mouse embryo. The myogenin promoter was linked to a β-galactosidase reporter gene and introduced into transgenic mice. The expression of β-galactosidase activity was detected by a colorimetric assay and is shown in blue. The expression pattern of the transgene is identical to that of the endogenous gene. (Please see color insert.)

Remarkably, mice lacking MyoD are fully viable and show no obvious muscle abnormalities. The only apparent consequence of MyoD inactivation is a twofold increase in expression of Myf5. Paradoxically, mice homozygous for a null mutation of Myf5 also develop normal skeletal muscle, but die at birth because of the absence of ribs, which results in an inability to breathe. It is unclear why Myf5, which is expressed exclusively in skeletal muscle cells, affects the formation of ribs, which are derived from the somite. One possibility is that Myf5 controls the expression of growth factors or extracellular matrix molecules by muscle cells, which are involved in regulating the behavior of adjacent rib precursor cells in the somite. When both MyoD and Myf5 are inactivated in the same animal, there is a complete absence of skeletal muscle and there are no detectable myoblasts. These findings suggest that MyoD and Myf5 play overlapping roles in the generation of myoblasts. This type of genetic redundancy could occur if MyoD and Myf5 were expressed in the same cells or if they were expressed in separate cells, either of which could support normal muscle development if the other cell population were eliminated. When lacZ expression is driven by the Myf5 locus in Myf5 null embryos, muscle progenitor cells migrate aberrantly. Thus, Myf5 protein appears to be necessary for cells to respond correctly to positional cues and to adopt a myogenic fate. The phenotype of mice lacking myogenin suggests that myogenin acts in a genetic pathway downstream of MyoD and Myf5. In myogenin-null mice, myoblasts are normally specified and positioned, but there is a

near-complete absence of differentiated skeletal muscle fibers. The undifferentiated myoblasts that populate the presumptive muscle forming regions of myogenin-null mice express MyoD and Myf5, indicating that these myogenic factors do not require myogenin for their expression and suggesting that they cannot direct the formation of normal skeletal muscle in the absence of myogenin. During normal muscle development, myogenin is downregulated at birth as MRF4 becomes upregulated to high levels. MRF4 is not expressed above background levels in myogenin-null mice, which suggests that it lies downstream of myogenin in the myogenic pathway. MRF4-null mice have normal skeletal muscle and are fully viable. However, myogenin is overexpressed in these mice, which suggests that it may compensate for the lack of MRF4. This upregulation of myogenin in the absence of MRF4 also suggests that MRF4 is required for the downregulation of myogenin that normally occurs in postnatal skeletal muscle. Together, the results of gene knockout experiments suggest that muscle cell determination and differentiation are controlled by the type of genetic cascade shown in Fig. 66-8. The bHLH genes MyoR and capsulin have also been shown to be essential for development of specific facial skeletal muscles. In mice lacking these two genes, the muscles of mastication are missing and members of the MyoD family are not expressed. The absence of specific head muscles, as well as markers of the corresponding myogenic lineages, in MyoR/capsulin double mutant mice resembles the effect of MyoD/Myf5 double mutations on all skeletal muscles. This phenotype differs from that of myogenin mutant mice, in which myoblasts express myogenic bHLH genes but are unable to differentiate. This phenotype demonstrates that MyoR and capsulin redundantly regulate an initial step as specification of a subset of facial skeletal muscles. In the absence of these factors, myogenic bHLH genes are not switched on, and the cells from these lineages undergo programmed cell death.

MEF2 FACTORS AND MYOGENESIS The majority of skeletal muscle genes contains E-boxes in their control regions and are therefore probably directly activated by the myogenic bHLH factors. However, there are also skeletal muscle genes that lack E-boxes. Many of these genes are controlled by the myogenic bHLH factors through a cascade of events in which the myogenic factors induce the expression of intermediate regulators, which in turn activate muscle-specific genes. MEF2 proteins are among these intermediate regulators. MEF2 factors are unable to activate muscle gene expression alone, but they potentiate the transcriptional activity of bHLH proteins. This potentiation appears to be mediated by direct interactions between the DNAbinding domain of these different types of transcription factors. The MEF2 factors belong to the MADS box family of transcription factors, named for MCM1, which regulates mating type-specific genes in yeast, agamous and deficiens, which act as homeotic genes to control flower development in plants, and serum response factor, which controls several serum-inducible and muscle-specific genes. There are four MEF2 genes in vertebrates: MEF2A, -B, -C, and -D (Fig. 66-9). The proteins encoded by these genes are nearly identical within the MADS domain, which is located at their amino termini. The MADS domain mediates homo- and heterodimerization and DNA binding to the sequence CTA(A/T)4TAG, which is found in the control regions of nearly every skeletal as well as cardiac muscle gene. Immediately adjacent to the MADS domain is a region known as the MEF2 domain, in which the

CHAPTER 66 / MUSCLE DEVELOPMENT AND DIFFERENTIATION

671

Figure 66-8 The genetic pathway for skeletal muscle development. According to this model, either MyoD or Myf5 become expressed when mesodermal precursor cells from the somite become committed to the myogenic lineage. Myoblast-specific genes are regulated by MyoD and Myf5. When the extracellular concentration of growth factors is reduced, MyoD or Myf5 activate the expression of myogenin, which in turn activates myotube-specific genes. Myogenin is normally downregulated when MRF4 is upregulated after birth. In MRF4-null mice, myogenin continues to be expressed at a high level, suggesting that MRF4 is normally required for its downregulation.

Figure 66-9 Diagrammatic representation of the proteins encoded by the four MEF2 genes. A linear representation of each MEF2 factor is shown. The MADS and MEF2 domains are indicated and alternative exons are shown.

MEF2 factors also share extensive homology. The function of the MEF2 domain has not been determined. The carboxyl termini of the MEF2 factors are required for transcriptional activation and are subject to complex patterns of alternative splicing, which generates enormous complexity of this family. In addition to regulating muscle structural genes, MEF2 factors have been shown to regulate the myogenic bHLH genes. The promoters of the myogenin, MyoD, and MRF4 genes have been shown to contain MEF2 sites that are essential for transcription of these genes in vivo and in vitro. During mouse embryogenesis, the MEF2 factors are expressed in the somite myotome after Myf5 and myogenin, which suggests that they do not play a role in the initial activation of these genes, but they are likely to be required for amplification and maintenance of the expression of these genes once their expression has been initiated.

MEF2 expression can be induced in nonmuscle cells by forced expression of myogenic bHLH proteins, which suggests that the MEF2 genes lie in a genetic pathway downstream of the myogenic bHLH genes. However, forced expression of MEF2 in fibroblasts has been shown to induce the expression of the myogenic bHLH genes and to activate muscle differentiation. This suggests that MEF2 factors and myogenic bHLH proteins function within a complex regulatory network in which members from either family of factors can autoregulate one another and crossregulate the other family members (Fig. 66-10). In this way, these two families of regulators reinforce one another’s expression and drive cells toward differentiation. The four MEF2 genes are also expressed at the earliest stages of heart development, as well as in smooth muscle cells, suggesting that MEF2 factors may regulate muscle gene expression in multiple muscle cell types. The functions of MEF2 have been analyzed genetically in Drosophila, which contains only a single MEF2 gene, called D-mef2, which encodes a protein that is virtually identical to the vertebrate MEF2 proteins in the MADS and MEF2 domains. The Drosophila MEF2 protein also binds the same DNA sequence as the mammalian factors and can activate transcription through the MEF2 site in mammalian cells. Like its vertebrate relatives, D-mef2 is expressed specifically in myogenic precursors and their descendants during embryogenesis. Drosophila embryos lacking D-mef2 contain myoblasts that are normally specified and positioned, but these cells are unable to differentiate. The severe muscle defect in these mutant flies suggests that MEF2 provides a function required for activation of muscle gene expression in multiple muscle cell types and suggests that the genetic functions of MEF2 have been conserved because flies and humans evolved from a common ancestor over 600 million years ago.

RELATIONSHIP OF SKELETAL MUSCLE TO OTHER TISSUES There are several reasons to expect that networks of cell type-specific bHLH proteins may control cell determination and differentiation in cell types other than skeletal muscle. The presence of E-boxes in the control regions of numerous tissue-specific genes suggests that these genes are controlled by specific bHLH proteins. The ubiquitous expression of E proteins also suggests the existence of cell type-specific dimerization partners for

672

SECTION VIII / MUSCULOSKELETAL

Figure 66-10 Activation of muscle gene expression by myogenic bHLH and MEF2 factors. Members of the MyoD family positively autoregulate their own expression and directly activate transcription of muscle-specific genes containing E-boxes in their control regions. MEF2 factors are induced in response to myogenic bHLH proteins and can in turn activate the expression of muscle-specific genes that lack E-boxes in their control regions. MEF2 proteins also bind the control regions of several myogenic bHLH genes and, therefore, participate in a positive feedback loop that amplifies and maintains the myogenic program. Myogenic bHLH proteins are shown in black, designated M.

these proteins. Id proteins are also expressed in many types of undifferentiated cells. Indeed, several cell type-specific bHLH proteins have been identified in Drosophila and vertebrate species. Further investigations into their mechanisms of action may reveal fundamental mechanisms for the control of tissuespecific gene expression during development.

SELECTED REFERENCES Benezra R, Davis RL, Lockshon D, Turner DL, Weintraub H. The protein Id: a negative regulator of helix-loop-helix DNA binding proteins. Cell 1990;61:49–59. Black BL, Olson EN. Transcriptional control of muscle development by myocyte enhancer factor-2 (MEF2) proteins. Annu REv Cell Dev Biol 1998;14:167–196. Blais A, Tsikitis M, Acosta-Alvear D, Sharan R, Kluger Y, Dynlacht BD. An initial blueprint for myogenic differentiation. Genes Dev 2005;19:553–569. Bour BA, O’Brien MA, Lockwood WL, et al. Drosophila MEF2, a transcription factor that is essential for myogenesis. Genes Dev 1995; 9:730–741. Braun T, Rudnicki MA, Arnold HH, Jaenisch R. Targeted inactivation of the muscle regulatory gene Myf-5 results in abnormal rib development and perinatal death. Cell 1992;71:369–382. Brennan TJ, Chakraborty T, Olson EN. Mutagenesis of the myogenin basic region identifies an ancient protein motif critical for activation of myogenesis. Proc Natl Acad Sci USA 1991;88:5675–5679. Buchberger A, Nomokonova N, Arnold HH. Myf5 expression in somites and limb buds of mouse embryos is controlled by two distinct distal enhancer activities. Development 2003;130:3297–3307. Buckingham M, Bajard L, Chang T, et al. The formation of skeletal muscle: from somite to limb. J Anat 2003;202:59–68. Cheng T-C, Wallace M, Merlie JP, Olson EN. Separable regulatory elements govern myogenin transcription in embryonic somites and limb buds. Science 1993;261:215–218.

Cserjesi P, Olson EN. Myogenin induces muscle-specific enhancer binding factor MEF-2 independently of other muscle-specific gene products. Mol Cell Biol 1991;11:4854–4862. Davis RL, Cheng P-F, Lassar AB, Weintraub H. The MyoD DNA-binding domain contains a recognition code for muscle-specific gene activation. Cell 1990;60:733–746. Davis RL, Weintraub H. Acquisition of myogenic specificity by replacement of three amino acid residues from MyoD into E12. Science 1992;256:1027–1030. Davis RL, Weintraub H, Lassar AB. Expression of a single transfected cDNA converts fibroblasts to myoblasts. Cell 1987;51:987–1000. Delgado I, Huang X, Jones S, et al. Dynamic gene expression during the onset of myoblast differentiation in vitro. Genomics 2003;82: 109–121. Goldhamer DJ, Brunk BP, Faerman A, King A, Shani M, Emerson CP, Jr. Embryonic activation of the myoD gene is regulated by a highly conserved distal control element. Development 1995;121:637–649. Gu W, Schneider JW, Condorelli G, Kaushal S, Mahdavi VJ, NadalGinard B. Interaction of myogenic factors and the retinoblastinoma protein mediates muscle cell commitment and differentiation. Cell 1993;72:309–324. Hadchouel J, Carvajal JJ, Daubas P, et al. Analysis of a key regulatory region upstream of the Myf5 gene reveals multiple phases of myogenesis, orchestrated at each site by a combination of elements dispersed throughout the locus. Development 2003;130:3415–3126. Halevy O, Novitch BG, Spicer DB, et al. Correlation of terminal cell cycle arrest of skeletal muscle with induction of p21 by MyoD. Science 1995;267:1018–1021. Hasty P, Bradley A, Morris JH, et al. Muscle deficiency and neonatal death in mice with a targeted mutation in the myogenin gene. Nature 1993;364:501–506. Kislinger T, Gramolini AO, Pan Y, Rahman K, MacLennan DH, Emili A. Proteome dynamics during C2C12 myoblast differentiation. Mol Cell Proteomics 2005;7:887–901. Konieczny SF, Emerson CP, Jr. 5-azacytidine induction of stable mesodermal stem cell lineages from 10T1/2 cells: evidence for regulatory genes controlling determination. Cell 1984;38:791–800. Lassar AB, Paterson BM, Weintraub H. Transfection of DNA locus that mediates the conversion of 10T1/2 fibroblasts to myoblasts. Cell 1986;47:649–656. Ledent V, Vervoort M. The basic helix-loop-helix protein family: comparative genomics and phylogenetic analysis. Genome Res 2001;11:754–770. Li L, Zhou J, James G, Heller-Harrison R, Czech M, Olson EN. FGF inactivates myogenic helix-loop-helix proteins through phosphorylation of a conserved protein kinase C site in their DNA binding domains. Cell 1992;71:1181–1194. Lu JR, Bassel-Duby R, Hawkins A, et al. Control of facial muscle development by MyoR and capsulin. Science 2002;298:2378–2381. Ma PC, Rould MA, Weintraub H, Pabo CO. Crystal structure of MyoD bHLH domain-DNA complex: perspectives on DNA recognition and implications for transcriptional activation. Cell 1994;77:451–459. Mal A, Harter ML. MyoD is functionally linked to the silencing of a muscle-specific regulatory gene prior to skeletal myogenesis. Proc Natl Acad Sci USA 2003;100:1735–1739. McKinsey TA, Zhang CL, Olson EN. Signaling chromatin to make muscle. Curr Opin Cell Biol 2002;14:763–772. Molkentin JD, Olson EN. Defining the regulatory networks for muscle development. Curr Opin Genet Dev 1996;6:445–453. Murre C, McCaw PS, Vaessin H, et al. Interactions between heterologous helix-loop-helix proteins generate complexes that bind specifically to a common DNA sequence. Cell 1989;58:537–544. Nabeshima Y, Hanaoka L, Hayasaka M, et al. Myogenin gene disruption results in perinatal lethality because of severe muscle defect. Nature 1993;364:532–535. Nevins JR. E2F: a link between the Rb tumor suppressor protein and viral oncoproteins. Science 1992;258:424–429. Olson EN. The MyoD family, a paradigm for development? Genes Dev 1990;4:1454–1461. Parker MH, Seale P, Rudnicki MA. Looking back to the embryo: defining transcriptional networks in adult myogenesis. Nat Rev Genet 2003;4:497–507.

CHAPTER 66 / MUSCLE DEVELOPMENT AND DIFFERENTIATION

Pownall ME, Gustafsson MK, Emerson CP, Jr. Myogenic regulatory factors and the specification of muscle progenitors in vertebrate embryos. Annu Rev Cell Dev Biol 2002;18:747–783. Puri PL, Sartorelli V. Regulation of muscle regulatory factors by DNAbinding, interacting proteins, and post-transcriptional modifications. J Cell Physiol 2000;185:155–173. Rao SS, Chu C, Kohtz DS. Ectopic expression of cyclin D1 prevents activation of gene transcription by myogenic basic helix-loop-helix regulators. Mol Cell Biol 1994;14:5259–5267. Relaix F, Rocancourt D, Mansouri A, Buckingham M. A Pax3/Pax7-dependent population of skeletal muscle progenitor cells. Nature 2005;435:948–953. Rudnicki MA, Schnegelsberg PNJ, Stead RH, Braun T, Arnold HH, Jaenisch R. MyoD or Myf-5 is required for the formation of skeletal muscle. Cell 1993;75:1351–1359. Schneider JW, Gu W, Zhu L, Mahdavi V, Nadal-Ginard B. Reversal of terminal differentiation mediated by p107 in Rb-/-muscle cells. Science 1994;264:1467–1471.

673

Shapiro DN, Sublett JE, Li B, Downing JR, Naeve CW. Fusion of PAX3 to a member of the forkhead family of transcription factors in human alveolar rhabdomyosarcoma. Cancer Res 1993;53: 5108–5112. Tapscott SJ, Davis RL, Thayer MJ, Cheng P-F, Weintraub H, Lassar AB. MyoD1: a nuclear phosphoprotein required in a myc homology region to convert fibroblasts to myoblasts. Science 1988;242: 405–411. Tapscott SJ. The circuitry of a master switch: Myod and the regulation of skeletal muscle gene transcription. Development 2005;132(12): 2685–2695. Weintraub H. The MyoD family and myogenesis: redundancy, networks and thresholds. Cell 1993;75:1241–1244. Weintraub H, Tapscott SJ, Davis RL, et al. Activation of muscle specific genes in pigment, nerve, fat, liver and fibroblast cell lines by forced expression of MyoD. Proc Natl Acad Sci USA 1989;86: 5434–5438.

67 Skeletal Muscle Structure and Function ELIZABETH M. MCNALLY, KAREN A. LAPIDOS, AND MATTHEW T. WHEELER SUMMARY This chapter summarizes the basic organization of skeletal muscle. Skeletal muscle is responsible for all voluntary movement, and its unique organization is optimized for this function. The sarcomere is the unit of muscle contraction and the sarcomere is linked to the plasma membrane through the Z band. Skeletal muscle is also unique in that mature myofibers are a multinucleate syncytium. Efficient delivery of calcium is necessary for coordinated muscle contraction. Neuronal control is also a key contributor to muscle function. Mutations in the genes encoding many “structural” proteins of muscle lead to muscle weakness and degeneration. Key Words: Actin; dystrophin glycoprotein complex; myosin; neuromuscular junction; nitric oxide synthase; sarcolemma; sarcomere; sarcoplasmic reticulum; titin; Z bands.

INTRODUCTION Skeletal muscle is one of the most highly organized structures in the biological world. The importance of responsive movement to the survival of the organism as a whole has contributed to the evolution of highly ordered and near crystalline array of proteins found in the mature skeletal myocytes or myofiber. Muscle is a unique tissue in that it arises from the fusion of mononuclear myoblasts to form a multinucleate syncytium. The syncytial nature of the mature myotube ensures a coordinated response to neural input. Muscle fibers are modified for fast twitch activity or slow tetanic activity. This variation derives from the specific needs of the fibers and is reinforced by fiber type-specific innervation. The major structural components of muscle relate to their function and include: 1. The sarcomere, made up of actin, myosin, and associated proteins. These proteins directly convert chemical energy in the form of ATP to movement and force using molecular motors and a scaffold of networked thin filaments. 2. The membrane that overlies individual myofibers interacts directly with the underlying cortical cytoskeleton, the Z band, and the extracellular matrix to help transmit force. 3. Interwoven through the myofibers is a membrane-bound structure for the efficient delivery of calcium that regulates the timing of muscle contraction.

From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

4. Myocyte nuclei are typically found in the periphery of the myocyte syncytium. Like the plasma membrane, underlying the nuclear membrane is a scaffold of proteins with both structural and signaling roles. Each of these areas has come into focus with the identification of genetic mutations that alter function and produce muscle disease.

ARCHITECTURE OF MUSCLE AND THE MYOFIBER Individual muscle cells or myofibers may span the length of an entire muscle. At each end is a myotendinous junction, a unique structure similar to focal adhesions found in nonmuscle cells. Myotendinous junctions transmit force longitudinally along the long axis of the muscle to result in movement. In the resting or uncontracted state, myofibers exert minimal but definitive force because of the opposing nature of many muscle groups. Sarcomeres, the units of muscle contraction, exist in register along the long axis of the myofiber. Myofibrils are made up of contractile proteins and can be isolated from muscle by removing the overlying plasma membrane. Groups of myofibers are encased in an external sheath of dense connective tissue called epimysium. Emanating from the epimysial sheath are septa of perimysia, and surrounding each individual myofiber is an additional layer of connective tissue called the endomysium. The nuclei are found in the periphery of individual myofibers adjacent to the sarcolemma. Staining for basement membrane proteins such as laminin can clearly distinguish the position of myofiber nuclei from other nuclei that are present in muscle, such as satellite cells. Satellite cells are under the basal lamina, by definition, and are stem cells in muscle with the ability to self-renew and to differentiate into myoblasts. Once a satellite cell has differentiated into a myoblast, it can fuse to a myofiber thereby contributing to myofiber growth. Immediately after regeneration, myofiber nuclei are found in the center, as opposed to the periphery, of myofibers. Central nucleation is thought to arise from myoblast fusion followed by synthesis of sarcomeric proteins to result in cytoplasmic structure surrounding the nucleus. Studies in syncytial cells of the model system Caenorhabditis elegans have identified proteins that appear essential for proper nuclear localization. The nuclear-anchoring protein in C. elegans, called ANC-1, is an enormously large protein that contains an actin-binding domain and spectrin repeats. Mutations in ANC-1 produce disorganized nuclei in syncytial cells in which normally the nuclei are ordered. Orthologous genes and proteins to ANC-1 are found in the human and mouse, but it is unknown whether these proteins have a similar

674

CHAPTER 67 / SKELETAL MUSCLE STRUCTURE AND FUNCTION

function in mammalian cells and specifically whether they have a similar role in the syncytial mammalian muscle fiber. The connective tissue that surrounds each myofiber is contiguous with the myotendinous junction and ultimately the tendons themselves. Proteins important for attachment to the intracellular matrix are highly concentrated in the myotendinous junction. Integrins are heterodimeric proteins expressed in nearly all cell types. Integrins bind to two major types of proteins in the extracellular matrix, fibronectin, and laminin. In muscle, the most highly expressed integrin isoforms are the laminin receptor α7β1 and fibronectin receptor α5β1. Integrin α7β1 is the major integrin expressed in mature muscle and is highly concentrated at the myotendinous junction in addition to being found at the plasma membrane of muscle. Muscle contraction produces the shortening of individual myofibers, and the connective tissue that surrounds each myofiber is important for force transmission. The register of thin and thick filaments is foreshortened and, thus, redundancy is present in sarcolemma surrounding each myofiber. Because not every myofiber extends the entire length of the muscle, the sheaths that surround each fiber necessarily must function in force transmission. Moreover, the myotendinous junctions transmit force of the entire muscle directly on the tendons and underlying bone. THE SARCOMERE, THE UNIT OF CONTRACTION H&E staining of longitudinal myofibers shows cross striations that are perpendicular to the long axis of the myofiber (Fig. 67-1). Alternating light and dark bands comprise the cross-striations. The darker bands, or A bands, are so named for their anisotropic properties in that they are birefringent in polarized light. The light bands, called I bands or isotropic, do not polarize light. The Z band, (or Z disk, Z line), is a thin dark band that bisects each I band. Thin filaments are found in the I band, and the Z band is made up of actin binding proteins. Overlying the Z bands at the sarcolemma are specialized structures called costameres. Costameric patterning on the surface of myofibers reflects the underlying Z band structures and provides attachments from the Z band to the surface plasma membrane. In the center of the A band is a lighter area, the H band, which represents thick filaments without any overlapping thin filaments (Fig. 67-1). Myosin and actin, the components of thick and thin filaments were among the first proteins to be purified and help contribute to the “domain” theory of protein function. Myosin is a hexamer with two heavy chains (220 kDa each) and four light chains (20–25 kDa each). The amino terminus of myosin is globular in nature and the two pairs of light chains bind at the junction between the globular head and the carboxy-terminal rod region. Within the head domain are the sequences required for ATP hydrolysis and actin binding. Proteolysis was used to separate the head region of myosin from the rod region and, using this approach, it was found that the rod region self-assembles with the periodicity of the intact myosin hexamer (143 Å). That is, the 850 amino acids that consist of the head region represent the enzymatically active region of the molecule, whereas the carboxy-terminal 1200 amino acids have an α-helical coiled nature responsible for macromolecular assembly. Thick filaments assemble bidirectionally, so that heads from myosin orient in opposite directions. A typical native thick filament assembles to a length not greater than 1–1.5 µm. Myosin binding protein C, a thick filament binding protein, may help regulate thick filament assembly or may help stabilize thick filaments once they are formed. The role of myosin binding protein C is especially important in the cardiomyocyte in

675

Figure 67-1 Shown are longitudinal (A) and transverse (B) crosssections of human skeletal muscle. In the longitudinal section, the cross striations are visible. Each myofiber is a single cell that contains many nuclei located at the periphery of each myofiber. In transverse crosssections each fiber is uniform in size, and fibers are tightly packed and encased in connective tissue (seen running from the upper left to the lower right of the section). Note the peripheral position of myonuclei. (C) shows an electron micrograph of a sarcomere, the unit of contraction. The Z band anchors the thin filaments. The thin filaments interdigitate with the myosin containing thick filaments. The A band represents overlapping thin and thick filaments. The I band is made up of only thin filaments that contain actin, tropomyosin, and the troponins.

676

SECTION VIII / MUSCULOSKELETAL

Figure 67-2 (A) Shows relaxation (left diagram) and contraction (right diagram) of the sarcomere. In the presence of calcium, a conformation change occurs in the thin filament that allows actin and myosin to interact. Through the myosin power stroke, the overlap between thick and thin filaments increases, and the sarcomere and muscle shorten. (B) Shows the myosin heads (gray and black) interacting with actin containing thin filaments. (C) Shows a cross-section of thin filaments, further highlighted in D. Thin filament regulatory proteins, tropomyosin and the troponins, regulate the actomyosin interface. Large proteins such as nebulin and titin anchor thick filaments. The middle panel of C shows a cross-section through a myosin thick filament surrounded by thin filaments. The right panel in C shows a cross-section through a myosin thick filament demonstrating the tight packing of myosin rod regions. Myosin is made up of two heavy chains and two each of regulatory and essential light chains. Hexamers of myosin tightly pack to form the thick filament.

which mutations in the cardiac form of myosin binding protein C lead to hypertrophic cardiomyopathy. Thin filaments are made up of several different proteins, but filamentous actin is the major constituent. Monomeric, or G actin, is 43 kDa. Modulated by ionic strength and actin binding proteins, actin assembles to form a right-handed double helical filament (F actin). The pitch of the actin filament is 36 nm with seven actin monomers per strand. Along the grooves of the actin double helix are the thin filament regulatory proteins, troponins T, C, and I and tropomyosin. Tropomyosin is a rod-like molecule, and troponin T directly interacts with tropomyosin. Troponin C directly binds calcium providing calcium sensitivity to the actomyosin interaction. From these biochemical observations and the structural observations of muscle, the sliding filament theory was proposed. In this model, the heads of myosin protrude from the thick filaments (Fig. 67-2). Protruding myosin heads directly interact with actin thin filaments to increase the overlap of thick and thin filaments, reducing the I band and effectively shortening the sarcomere and the muscle fiber as a whole. Force is transmitted parallel to the long axis of muscle. As the Z band anchors thin filaments, force is also to some degree transmitted perpendicular to the long axis of myofibers. The Z band also attaches to the plasma membrane at specialized structures called costameres. Further support for the sliding filament

model was found by examining the appearance of isolated thick filaments where protruding myosin heads could be seen. In the sliding filament model, the “rowing” movement of myosin protruding heads must occur within the myosin head or neck to be effective in moving the thick filaments along thin filaments. Although electron microscopy of muscle provided anatomic support for the sliding filament theory, demonstrating movement within the myosin head or neck region has been difficult. A revolution in understanding myosin and actin function occurred with the development of in vitro motility assays that required only purified actin and myosin to visualize the movement of fluorescently labeled actin filaments along myosin fixed to a surface. These assays have provided data on the domains of myosin required for ATP hydrolysis and actin binding, and have allowed the identification of domains required for movement and force production. The crystal structure of the myosin head and structures of myosin heads with actin have been used to identify varying conformational states of the actomyosin interface and have provided support for the sliding filament theory. One central question in myosin biology is to determine the length of the step that a myosin head can make along the length of actin filament for a hydrolysis of ATP (Fig. 67-3). This calculation, referred to as myosin step size, has been estimated from several

CHAPTER 67 / SKELETAL MUSCLE STRUCTURE AND FUNCTION

Figure 67-3 Shown is the ATPase cycle of the actomyosin interaction. After hydrolyzing ATP to ADP plus Pi, myosin undergoes a conformational change. The interaction of the myosin head with actin occurs. After release of Pi, myosin is thought to undergo an additional conformational change that moves actin along the actin thin filament. The distance that myosin can move along the actin thin filament per cycle of ATP hydrolysis is referred to as the myosin step size.

approaches. The estimates range from 5 to 15 nm, and this is in the context of a myosin that is approx 20 nm in length itself. There are at least 16 types of myosin molecules defined using electronic database searching and biochemical characterization. Myosins fall into two broad classes. The first, myosin I, are specialized motors for intracellular transport. These myosins have a head region similar to the myosin found in muscle, but have modified tail regions specialized for the cargo and vesicle attachment. The neck regions of myosin are also specialized in that this region may bind light chains or calmodulin to regulate the function of myosin. Muscle myosin, referred to as myosin II, represents the second class. These myosins have the specialized rod region that can form thick filaments. Thus, type-II myosins function as many motors together on a rod whereas type-I myosins may function as single heads or as single molecules with two heads. For example, myosin V has two heads and can take an enormous step of 36 nm. This large step can be accomplished while the second myosin head remains attached. The processivity of this myosin appears to differ from muscle myosin, and this is further supported by kinetic studies that show a slow release of adenosine 5′-diphosphate (ADP) to aid the processive nature of myosin V. Muscle myosin appears to have smaller step size, and a much faster release of ADP. Some estimates calculated the step size of muscle myosin to be as small as 5 nm. Other estimates, which argue that the arrangement of actin filaments influences measurements of myosin head step size, have calculated step sizes as large as 15–30 nm. These investigators reason that to achieve this larger step size, myosin actually takes “substeps” in its interactions with actin. Kinetic studies have not necessarily supported the presence of multiple small “substeps” although further experimentation is required to fully understand the interaction of myosin with actin and the molecular basis of biological movement. Current structural studies focus on the neck region of myosin, as this appears to have regions capable of serving the lever function for movement. Comparisons of evolutionarily and functionally diverse myosins have revealed myosin VI as a molecule that contains a distinct lever region. In vitro studies of this myosin showed surprisingly that it moves in an opposite orientation relative to all other characterized myosins. Generally myosins move

677

toward the barbed end of actin filaments. In contrast, myosin VI moves toward the pointed end, supporting the idea that not all myosin molecules have the identical unidirectional movement. To date, myosin VI is the only myosin described capable of reverse movement. ANCHORING THE SARCOMERE, PASSIVE FORCE, ELASTIC RECOIL, AND FORCE TRANSMISSION Titin is giant protein that spans the length of the sarcomere providing attachment from the sarcomere to the Z band and also to the membrane. Titin is alternatively spliced to produce a variety of protein forms that are several megadaltons in size. As an elongated structure, titin can span nearly 1 µm, close to the length of a sarcomere. Titin has a highly repetitive structure thought to impart elastic recoil properties of muscle. Actin-containing thin filaments are anchored to the Z band. A number of Z band-associated proteins have been identified that are important for skeletal muscle and cardiac muscle function. Titin has attachments along the Z band and can interact with proteins such as telethonin (T cap) and subunits of the calcium-activated protease, calpain. Mutations in titin, telethonin, and calpain 3 can lead to muscular dystrophy. Myotilin is a smaller protein that contains several immunoglobulin domains. Genetic mutations in myotilin also are associated with muscular dystrophy. The precise mechanism of how genetic defects in these proteins lead to muscle degeneration is not known, but their convergence on the Z band of muscle implies the importance of this muscular structure. ANCHORING Z BANDS TO THE PLASMA MEMBRANE Z bands anchor the sarcomere to the plasma membrane. Concentrated over the Z band at the plasma membrane is the dystrophin glycoprotein complex (DGC). The specialized connections of the Z bands to the plasma membrane are referred to as costameres and can be visualized on longitudinal sections of muscle. Dystrophin, the protein product of the Duchenne muscular dystrophy locus, is a 427 kDa protein with an actin-binding domain found at its amino terminus. Within its elongated rod region are 24 spectrin repeats that are interrupted by hinge sites. Hinge sites are defined by breaks in the typical spectrin repeats and are thought to provide flexibility. Spectrin repeats are independently folding domains of roughly 100 amino acids that have a triple-helical bundled structure. Each spectrin repeat has a limited degree of flexibility. Spectrin repeats are named for the prototypical protein spectrin, a protein component of the erythrocyte membrane. Genetic defects in spectrin lead to instability of the erythrocyte membrane and to hemolytic anemia. Similarly, absence or reduction in the dystrophin component of the plasma membrane leads to instability of the muscle membrane producing muscle degeneration and muscular dystrophy. In skeletal muscle, regeneration can occur, but typically in Duchenne muscular dystrophy, regeneration is insufficient to match the ongoing degeneration and therefore, muscle wasting and weakness ensue. The Dystrophin Glycoprotein Complex and its Role in Membrane Stability The DGC is a macromolecular structure that can be purified from the membrane fraction of muscle (Fig. 67-4). The DGC has cytoplasmic and transmembrane components. A major transmembrane component is the laminin receptor dystroglycan. Dystroglycan is made up of an α- and β-subunit produced by proteolytic cleavage of a single polypeptide. Dystroglycan is broadly expressed, is found in most tissue types and has an essential role in development where it organizes laminin in the extracellular matrix. In addition, in nonmuscle cells dystroglycan serves as a receptor for a number of pathogenic

678

SECTION VIII / MUSCULOSKELETAL

Figure 67-4 Shown is a schematic representation of the dystrophin glycoprotein complex and the integrin complex, two membrane-associated complexes in muscle that serve to connect the extracellular matrix to the membrane and the underlying cytoskeleton. Dystrophin attaches to dystroglycan. The sarcoglycan complex may stabilize the interaction of α and β dystroglycan. The integrin complex is concentrated at Z bands and at the myotendinous junction.

agents. In its intracellular portion, dystroglycan can directly bind to dystrophin. Dystroglycan can also bind Grb2, a protein implicated in Ras. In the central and peripheral nervous systems, α-dystroglycan is 120 kDa, whereas in skeletal muscle, the larger subunit of dystroglycan is 156 kDa, and this difference relates to differential glycosylation. The sarcoglycan complex is a muscle specific subcomplex within the DGC, and mutations in the genes encoding α-, β-, γand δ-sarcoglycan lead to autosomal-recessive limb girdle muscular dystrophy. The precise role of the sarcoglycan complex is not fully understood, but like the entire DGC, the sarcoglycan complex likely has both mechanical and signaling roles for the maintenance and normal function of skeletal muscle membranes. Sarcospan is a four membrane-spanning tetraspanin protein. Other tetraspanin proteins have been demonstrated to interact with integrins. Mice engineered with a null mutation of sarcospan do not develop muscular dystrophy. A number of other tetraspanin proteins are expressed in muscle and compensatory upregulation of other tetraspanins may limit the expression of a phenotype from the loss of sarcospan. Mice have been engineered with null mutations in the sarcoglycan proteins and these mice recapitulate the human phenotypes. The mdx mouse is a spontaneously arising mutant that has a nonsense mutation 23 of the dystrophin gene, and, thus, makes no full-length dystrophin. These mice display an identical histopathology to what is seen in Duchenne muscular dystrophy, but do not develop overt muscle weakness, are fertile and have a nearly normal lifespan. Mouse models of dystrophin and sarcoglycan deficiency have proved useful for determining a DGC function in maintaining membrane stability in muscle. The cytoplasmic face of the DGC includes a number of intracellular proteins including neuronal nitric oxide synthase (NOS or nNOS1), dystrobrevins, and the syntrophins. The syntrophins are PDZ domain containing proteins that can bind directly to nNOS. The dystrobrevins have homology to the carboxy-terminal region of dystrophin and bind directly to dystrophin and each other. The syntrophins bind directly to dystrophin and to dystrobrevin, thus forming a complex that anchors nNOS. Consistent with this, mutations in dystrophin displace nNOS from the plasma membrane. nNOS can also bind to caveolin, a small membrane-bound protein important for the formation of caveolae structures.

Dystrophin, dystroglycan, and laminin are thought to form a mechanical link that stabilizes the plasma membrane through its interaction with the extracellular matrix through the considerable deformation and stress faced through repeated muscle contraction. Mutations that reduce dystrophin result in destabilization of the remainder of the DGC including dystroglycan and sarcoglycan. The loss of the DGC is thought to render the membrane susceptible to contraction-induced damage. Supporting this, muscle lacking dystrophin displays a significant increase in membrane disruption and damage when subjected to eccentric contraction, resulting in a reduction in maximal force production. A common molecular feature between sarcoglycan mutant muscle and dystrophin mutant muscle is the loss of the sarcoglycan complex. Interestingly, in γ-sarcoglycan mutant muscle the sarcoglycan complex is largely destabilized but dystrophin, dystroglycan, and laminin are present. In similar eccentric contraction protocols, γsarcoglycan muscle does not demonstrate contraction-induced damage supporting a nonmechanical role for the sarcoglycan complex. On the cytoplasmic face, muscle lacking for nNOS or the syntrophins does not develop muscular dystrophy. This may relate to compensatory upregulation of related proteins or may indicate that these subunits participate in alternative intracellular functions. Mice engineered to lack α-dystrobrevin develop a mild muscular dystrophy phenotype although the properties of membrane instability may be different than what is seen with sarcoglycan or dystrophin mutations. At its amino terminus, dystrophin has a classic actin-binding site similar to that seen in α-actinin and other actin-binding proteins. Given the localization of dystrophin at the plasma membrane and especially its concentration overlying Z bands, dystrophin is thought to preferentially interact with γ-actin that forms part of the cytoskeletal network that is under the sarcolemma. Supporting this, when the plasma membrane is mechanically peeled from the underlying myofibrils, the under surface of the plasma membrane has a costameric arrangement of γ-actin (Fig. 67-5). Moreover, when dystrophin is mutant, the pattern of γ-actin is disrupted and is no longer associated with the plasma membrane, remaining instead affiliated with the myofibrillar structure including Z bands. Therefore, dystrophin participates in a mechanically strong linkage that includes γ-actin.

CHAPTER 67 / SKELETAL MUSCLE STRUCTURE AND FUNCTION

679

Figure 67-5 Shown is a longitudinal section of mouse muscle immunostained with an antibody for δ-sarcoglycan. The pattern of staining that runs perpendicular to the long axis of the myofiber represents the costameric pattern overlying Z bands. The dystrophin glycoprotein complex is concentrated more highly at costameres.

Dystrophin associates with actin not only through the aminoterminal actin-binding domain, but also through a side-to-side interaction with actin filaments along its rod. Utrophin is a chromosome-6 encoded homolog of dystrophin that is highly expressed in developing muscle and expressed at lower levels in mature muscle. The overall structural similarity between utrophin and dystrophin is substantiated as utrophin can substitute for dystrophin in the mdx mouse, a mouse model of muscular dystrophy. Utrophin is also concentrated at neuromuscular junctions (NMJs). The NMJ is a specialization for the organized delivery of neural input to muscle. Utrophin mutant mice have a reduction in the number of folds normally found in the NMJ and concomitant decrease in NMJ function. Moreover, mice doubly mutant for dystrophin and utrophin have a more severe phenotype indicating overlapping roles of dystrophin and utrophin.

NEURONAL CONTROL Skeletal muscle requires neuronal input from motor neurons and from muscle spindles, receptors required for spatiotemporal proprioception. This input contributes directly to the development and relaxation of muscle groups for coordinated movement. Neural input also provides feedback on a longer time-scale to regulate gene expression and modify fiber typing for the requirements of repeated use. Hypertrophy and fiber recruitment occur in response to training. Muscle fibers fall into two major categories. Type-I fibers are slow twitch muscle with aerobic metabolism. In contrast, type-II fibers are modified for fast twitch function and thereby depend on anaerobic metabolism utilizing glycogenolytic pathways. The program of gene expression varies considerably between these two types of fibers and most human skeletal muscles contain a mixture of fiber types. Some fibers have intermediate properties and certain fiber types may be more susceptible to certain disease processes. Slow and fast motorneurons provide input to type-I and type-II fibers, respectively. Denervation of muscle leads to atrophy. Gene

expression profiling has identified genes and proteins that are uniquely upregulated in atrophic states. Regulation of muscle mass occurs during development and also is constantly regulated in adult states. Myostatin, a member of the type-β transforming growth factor family of proteins, is a negative regulator of muscle mass. Mice engineered that completely lack myostatin have muscle mass nearly 200% of normal. Adult mice treated with inhibitors of myostatin function can increase by muscle mass 20–30%. The ability to inhibit the negative regulation of muscle mass may prove useful in disease states. Myostatin is thought to act directly on muscle itself through paracrine mechanisms although further study is required to fully understand its scope of action. The NMJ is a highly redundant structure to allow rapid transmission of neuronal input to the muscle. Acetylcholine receptors are the major receptor of this input. Molecules such as agrin participate in acetylcholine receptor clustering and regulation of receptor function. The intracellular surface kinases, such as MuSK, regulate acetylcholine function. The acetylcholine receptor is important in disease states such as myasthenia gravis where defects in receptor function can be induced through immune mediated or inherited mechanisms. Neural input leads to membrane depolarization that, in turn, must lead to coordinated muscle membrane depolarization and intracellular calcium release. Depolarization of the membrane is directly propagated through membrane invaginations known as transverse (T) tubules. As T tubules contact the sarcoplasmic reticulum (SR) directly, this allows conformational changes in the calcium release channels of the SR.

REGULATION OF MUSCLE CONTRACTION BY CALCIUM For coordinated activation of actomyosin interaction, intracellular calcium is tightly regulated. Such excitation contraction coupling occurs at regions where the intracellular sarcoplasmic reticulum, an extension of the endoplasmic reticulum, forms

680

SECTION VIII / MUSCULOSKELETAL

junctions with the sarcolemma. Tubular invaginations of the sarcolemma, T tubules, form a junction with two terminal cisternae regions of the SR forming a triad junction. Within the SR and the T tubule system, there are a number of proteins that regulate calcium sequestration and calcium release. Among these are the intracellular ryanodine receptor, a large >500 kDa protein that forms a tetramer and the dihydropyridine receptor, a voltage gated L-type calcium channel. Together, these proteins are important for the regulation of depolarization-induced calcium release as well as calcium-induced calcium release. Thus, these proteins are essential for coupling excitation and contraction through the release of calcium. Uptake after contraction is equally essential to maintaining tight control on the actomyosin interaction. For this role, the sarcoplasmic reticulum calcium ATPase pump (SERCA) is required. Triadin, junctin, mitsugumin, and junctophilin are proteins that participate in the formation of the triad junction facilitating the interaction of the T-tubular dihydropyridine receptor and the ryanodine receptor. Calsequestrin is a luminal protein that directly binds calcium. Together these proteins each participate in the regulation of the triad junction and play a role in excitation contraction coupling in skeletal muscle. Highlighting the importance of calcium regulation to disease, mutations in the genes encoding both the ryanodine receptor and SERCA lead to muscle dysfunction and disease. Mutations in the ryanodine receptor lead to malignant hyperthermia in which a sudden increase in intracellular calcium can occur leading to hypercontraction. In Brody’s disease, mutations in SERCA lead to abnormal relaxation characterized by painless muscle cramping and exercise-induced impairment of muscle relaxation. Brody patients frequently show an inability to relax skeletal muscle following contracture.

MYOCYTE NUCLEI As in other eukaryotic cells, the nuclear membrane is made up of two distinct lipid bilayers, the inner and outer nuclear membranes. Facing the nucleus and associated with the inner nuclear membrane is a network of proteins that scaffolds nuclear contents. Lamins A and C are produced from a single gene and form higher order filamentous structures. As in most other terminally differentiated cells, lamins A and C are intermediate filament proteins that provide support to the mature myofiber inner nuclear membrane. Mutations in the gene encoding lamin A/C are a common cause of muscular dystrophy. Recent studies suggest that lamin A/C is important for the mechanical properties of myofiber nuclei. In addition, mutations in the gene encoding the nuclear membrane protein emerin cause an X-linked form of muscular dystrophy. Like lamin A/C, emerin is broadly expressed, yet mutations in this gene lead to a tissue specific phenotype. The identification of lamin A/C and emerin, and their role in muscle degeneration highlights the importance of nucleoskeleton.

CONCLUSIONS The structure of skeletal muscle is unique and provides the machinery necessary for controlled movement. The cell biological aspects of skeletal muscle have emphasized both structural and nonstructural roles for a variety of proteins in the sarcomere, the Z band, the plasma membrane, and the nuclear membrane. The importance of these proteins to normal skeletal muscle function has been highlighted with the identification of a large number of genetic mutations that lead to skeletal muscle dysfunction and

disease. In general, many of these genetic defects have not targeted muscle development, as such mutations may prove too lethal for the organism’s survival. However, muscle developmental paradigms are increasingly important with the characterization of pluripotent stem cells in muscle and the role of satellite cells for muscle regeneration. These topics are covered in Chapters 66 and 68.

SELECTED REFERENCES Ahn AH, Kunkel LM. The structural and functional diversity of dystrophin. Nat Genet 1993;3:283–291. Barral JM, Epstein HF. Protein machines and self assembly in muscle organization. Bioessays 1999;21:813–823. Bodine SC, Latres E, Baumhueter S, et al. Identification of ubiquitin ligases required for skeletal muscle atrophy. Science 2001;294: 1704–1708. Bonne G, Carrier L, Richard P, Hainque B, Schwartz K. Familial hypertrophic cardiomyopathy: from mutations to functional defects. Circ Res 1998;83:580–593. Clark KA, McElhinny AS, Beckerle MC, Gregorio CC. Striated muscle cytoarchitecture: an intricate web of form and function. Annu Rev Cell Dev Biol 2002;18:637–706. Durbeej M, Campbell KP. Muscular dystrophies involving the dystrophinglycoprotein complex: an overview of current mouse models. Curr Opin Genet Dev 2002;12:349–361. Emery AE. Emery-Dreifuss muscular dystrophy—a 40 year retrospective. Neuromuscul Disord 2000;10:228–232. Ervasti JM. Costameres: The Achilles’ heel of herculean muscle. J Biol Chem 2003;278:13,591–13,594. Faulkner G, Lanfranchi G, Valle G. Telethonin and other new proteins of the Z-disc of skeletal muscle. IUBMB Life 2001;51:275–282. Franzini-Armstrong C, Protasi F. Ryanodine receptors of striated muscles: A complex channel capable of multiple interactions. Physiol Rev 1997;77:699–729. Gomes MD, Lecker SH, Jagoe RT, Navon A, Goldberg AL. Atrogin-1, a muscle-specific F-box protein highly expressed during muscle atrophy. Proc Natl Acad Sci USA 2001;98:14,440–14,445. Gregorio CC, Antin PB. To the heart of myofibril assembly. Trends Cell Biol 2000;10:355–362. Hack AA, Cordier L, Shoturma DI, Lam MY, Sweeney HL, McNally EM. Muscle degeneration without mechanical injury in sarcoglycan deficiency. Proc Natl Acad Sci USA 1999;96:10,723–10,728. Hack AA, Groh ME, McNally EM. Sarcoglycans in muscular dystrophy. Microsc Res Tech 2000;48:167–180. Henry MD, Campbell KP. Dystroglycan inside and out. Curr Opin Cell Biol 1999;11:602–607. Houdusse A, Sweeney HL. Myosin motors: missing structures and hidden springs. Curr Opin Struct Biol 2001;11:182–194. Huxley AF. Cross-bridge action: present views, prospects, and unknowns. J Biomech 2000;33:1189–1195. Kron SJ, Spudich JA. Fluorescent actin filaments move on myosin fixed to a glass surface. Proc Natl Acad Sci USA 1986;83:6272– 6276. Lee SJ, McPherron AC. Myostatin and the control of skeletal muscle mass. Curr Opin Genet Dev 1999;9:604–607. Leong P, MacLennan DH. Complex interactions between skeletal muscle ryanodine receptor and dihydropyridine receptor proteins. Biochem Cell Biol 1998;76:681–694. Ma J, Pan Z. Junctional membrane structure and store operated calcium entry in muscle cells. Front Biosci 2003;8:D242–D255. MacLennan DH, Abu-Abed M, Kang C. Structure-function relationships in Ca(2+) cycling proteins. J Mol Cell Cardiol 2002;34:897–918. Mayer U. Integrins: redundant or important players in skeletal muscle? J Biol Chem 2003;278:14,587–14,590. Muller FU, Kirchhefer U, Begrow F, Reinke U, Neumann J, Schmitz W. Junctional sarcoplasmic reticulum transmembrane proteins in the heart. Basic Res Cardiol 2002;97:I52–I55. Pardo JV, Siliciano JD, Craig SW. A vinculin-containing cortical lattice in skeletal muscle: transverse lattice elements (“costameres”) mark sites

CHAPTER 67 / SKELETAL MUSCLE STRUCTURE AND FUNCTION

of attachment between myofibrils and sarcolemma. Proc Natl Acad Sci USA 1983;80:1008–1012. Petrof BJ, Shrager JB, Stedman HH, Kelly AM, Sweeney HL. Dystrophin protects the sarcolemma from stresses developed during muscle contraction. Proc Natl Acad Sci USA 1993;90:3710–3714. Rafael JA, Brown SC. Dystrophin and utrophin: genetic analyses of their role in skeletal muscle. Microsc Res Tech 2000;48:155–166. Rando TA. The dystrophin-glycoprotein complex, cellular signaling, and the regulation of cell survival in the muscular dystrophies. Muscle Nerve 2001;24:1575–1594. Rock RS, Rice SE, Wells AL, Purcell TJ, Spudich JA, Sweeney HL. Myosin VI is a processive motor with a large step size. Proc Natl Acad Sci USA 2001;98:13,655–13,659. Rybakova IN, Patel JR, Ervasti JM. The dystrophin complex forms a mechanically strong link between the sarcolemma and costameric actin. J Cell Biol 2000;150:1209–1214. Spudich JA. The myosin swinging cross-bridge model. Nat Rev Mol Cell Biol 2001;2:387–392.

681

Starr DA, Han M. ANChors away: an actin based mechanism of nuclear positioning. J Cell Sci 2003;116:211–216. Starr DA, Han M. Role of ANC-1 in tethering nuclei to the actin cytoskeleton. Science 2002;298:406–409. Trinick J, Tskhovrebova L. Titin: a molecular control freak. Trends Cell Biol 1999;9:377–380. Walker ML, Burgess SA, Sellers JR, et al. Two-headed binding of a processive myosin to F-actin. Nature 2000;405:804–807. Wells AL, Lin AW, Chen LQ, et al. Myosin VI is an actin-based motor that moves backwards. Nature 1999;401:505–508. Whittemore LA, Song K, Li X, et al. Inhibition of myostatin in adult mice increases skeletal muscle mass and strength. Biochem Biophys Res Commun 2003;300:965–971. Worman HJ, Courvalin JC. How do mutations in lamins A and C cause disease? J Clin Invest 2004;113:349–351. Yanagida T, Kitamura K, Tanaka H, Hikikoshi Iwane A, Esaki S. Single molecule analysis of the actomyosin motor. Curr Opin Cell Biol 2000;12:20–25.

68 Stem Cells and Muscle Regeneration CINDY M. MARTIN, THOMAS J. HAWKE, AND DANIEL J. GARRY SUMMARY Somatic (adult) stem cell populations are resident in postnatal tissues such as skeletal muscle and function in the maintenance and regeneration of the respective tissues. Use of transgenic and emerging technologies will enhance understanding of the regulation of the somatic stem cell populations and regenerative mechanisms in normal, aging and myopathic states. Furthermore, these studies may promote the use of adult stem cells for cell replacement therapies or for use as vehicles for gene replacement therapy. Key Words: Aging; embryonic stem (ES); skeletal muscle; somatic; stem cell; regeneration.

INTRODUCTION The ancient Greeks recognized the potential of tissue regeneration in the legend of Prometheus. As punishment for attempting to steal the secret of fire from the Gods, Prometheus was chained to the side of Mount Caucasus and sentenced to a life of eternal torture as vultures would feed on his liver during the daytime and at night his liver would undergo repair and regeneration. This ability for tissue regeneration remains an area of intense interest and research. Titans or champions of regeneration include the hydra, zebrafish, and the newt, which have the capability of regenerating entire limbs, heart, jaw, retina, and tail. Although a number of common pathways and networks are shared between urodeles (i.e., newts) and mammals, the regenerative capacity of humans and rodents is more limited and dependent on resident stem cell populations, a permissive niche or milieu and molecular networks to promote repair and regeneration of adult tissues such as skeletal muscle.

EMBRYONIC STEM CELLS: PROTOTYPIC STEM CELL POPULATION Perhaps the most intensely studied stem cell population is the embryonic stem (ES) cell. ES cells are derived from the embryonic blastocyst and are pluripotent. The ES cells have a distinct molecular signature as they express Esg1, Utf1, Oct3/4, Nanog, Rex3, Tex20, and a number of additional factors that promote their unlimited proliferative capacity and multipotentiation. In addition, the ES cells are routinely cultured in the presence of feeder cells (growth arrested or mitotically inactive fibroblasts) to provide growth factors and chemokines, such as leukemia inhibitory factor, which maintain From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

the ES cell population in an undifferentiated and proliferative state. The extensive analyses of murine ES cells ultimately resulted in the isolation of human ES cells. A number of studies have been undertaken to further characterize the culture conditions, proliferative capacity, and the potential of human ES cells to contribute to alternative lineages using in vitro conditions. For example, although human ES cells grow more slowly (double every 36 h) than mouse ES cells (double every 12 h), the human and mouse ES cells are similar in their expression of high levels of telomerase, and their capability of self-renewal and multilineage differentiation in vitro (e.g., muscle, blood, fat, and so on) (Fig. 68-1). In contrast to the ES population, the somatic (adult) stem cell populations are more limited in their proliferative capacity and plasticity.

SOMATIC STEM CELLS FOR THE MAINTENANCE AND REPAIR OF ADULT TISSUES Tissue-specific adult (somatic) stem cells are important in the growth, maintenance, and regeneration of postnatal and adult tissues. Nearly all adult tissues contain a resident stem cell population including the hematopoietic stem cells and stromal cells (bone marrow), neural stem cells (subventricular zone and hippocampus of the brain), epithelial stem cells (intestinal crypts), oval stem cells (liver), and satellite cells (skeletal muscle) (Fig. 68-1). Myogenic progenitor cells (satellite cells) are resident in skeletal muscle as small mononuclear cells that occupy a peripheral location with respect to the adjacent multinucleated myotube (Fig. 68-2). The satellite cells can be identified using electron microscopy as they occupy a sublaminar (beneath the basal lamina) position with respect to the adjacent myofiber (Fig. 68-2). In neonatal skeletal muscle, approx 30% of the nuclei are satellite cells that fuse to the growing myofibers and contribute to postnatal growth of skeletal muscle. The satellite cells in adult skeletal muscle are reduced in number, representing approx 2–5% of the nuclei. The quiescent satellite cells are arrested at an early stage of the myogenic program such that they do not express any of the myogenic basic helix-loophelix (bHLH) regulatory factors of the MyoD family (Myf5, MyoD, Myogenin, and Mrf4). Limited molecular markers for this cell population include Foxk1, m-cadherin, c-met, syndecan-4, Pax7, and CD34. Foxk1 and Pax7 are transcription factors that have important functional roles in the regulation of satellite cells. Using a gene disruption strategy (i.e., gene knockout), mice lacking the forkhead/ winged helix transcription factor, Foxk1, have a severe growth deficit and impaired muscle regeneration. The impairment in muscle regeneration results from decreased numbers of satellite cells

682

CHAPTER 68 / STEM CELLS AND MUSCLE REGENERATION

683

Figure 68-1 Properties of embryonic and adult (somatic) stem cell populations. Schematic representation of similarities and differences of an adult stem cell population such as satellite cells which are resident in adult skeletal muscle and embryonic stem cells.

due to perturbed cell-cycle regulation in Foxk1-deficient satellite cells. The perturbed cell-cycle progression is because of the dysregulation of the cyclin-dependent kinase inhibitor, p21, which is a putative downstream target for Foxk1. Similarly, mice lacking the paired box transcription factor, Pax7 have an absence of satellite cells in postnatal skeletal muscle. These results suggest that Pax7 is an essential transcription factor for the specification of satellite cells from an undefined precursor cell population. Furthermore, the relative normal skeletal muscle architecture in the absence of satellite cells further suggested the presence of other somatic stem cell populations in Pax7-deficient skeletal muscle.

SIDE POPULATION CELLS ARE ENRICHED IN REPOPULATING CELLS Dual wavelength flow cytometry and a DNA-binding dye (Hoechst 33342) have been used to identify side population (SP) cells that are enriched in repopulating cells (i.e., stem cells) (Fig. 68-3). These SP cells are resident in adult tissues (i.e., skeletal muscle, bone marrow, heart, liver, brain, and so on) and are able to efflux the Hoechst dye because of the multidrug resistance protein (Abcg2), which is a member of the ATP binding cassette transporter family. SP cell populations are rare but are capable of multipotentiation as they can adopt alternative fates when placed in permissive environments (Fig. 68-3). For example, SP cells isolated from the bone marrow of normal adult mice were able to reconstitute the lethally irradiated mdx (a mouse model that lacks dystrophin and is a model of muscular dystrophy) bone marrow; later these cells were recruited from the bone marrow to participate in muscle repair and regeneration. Similarly, SP cells isolated from adult skeletal muscle were able to reconstitute the irradiated bone marrow. These studies provide supportive evidence that SP

cells have stem cell capabilities and are able to participate in tissue repair when placed in a permissive environment (Fig. 68-3).

SKELETAL MUSCLE INJURY AND REGENERATION Adult skeletal muscle has the remarkable capacity for complete repair following severe injury. For example, a chemical-induced muscle injury that destroys more than 70–80% of the muscle will be regenerated with restoration of the anatomical architecture within a 2-wk period. The regenerative capacity of skeletal muscle due to the presence of resident stem cell populations (satellite and SP cells and other potential progenitor cell populations). These resident stem cell pools are quiescent in unperturbed skeletal muscle (Fig. 68-4). In response to a severe injury, the satellite cells are activated and express the bHLH transcription factor, MyoD. These activated satellite cells migrate to sites of injury and enter the cell cycle. The activated satellite cells have an increased proliferative capacity (capable of an estimated 80–150 doublings) and undergo asymmetric cellular division to re-establish a pool of quiescent stem cells. This property of self-renewal is characteristic of stem cell populations and has been shown to involve the Notch signaling pathway. Approximately 5–7 d following injury, the proliferating satellite cells withdraw from the cell cycle and differentiate to form centronucleated myotubes (Fig. 68-4). Over the following 1-wk period the regenerated myotubes increase in size because of fusion of myoblasts and ultimately restore the architecture of the adult skeletal muscle. These regenerative responses require an orchestrated interactive response between the inflammatory system, growth factors, extracellular matrix (ECM), vascularization, and the stem cell population resident in skeletal muscle. During the initial phase following a severe muscle injury, a prominent neutrophillic infiltrate is observed resulting in the phagocytosis of necrotic myofibers

684

SECTION VIII / MUSCULOSKELETAL

Figure 68-2 Satellite cells are resident in adult skeletal muscle. (A) Using transmission electron microscopic techniques, satellite cells are identified as small mononuclear cells that occupy a peripheral location regarding the larger multinucleated myofiber. (B) Schematic figure of A emphasizing that the mononucleated satellite cell occupies a sublaminar position (beneath the basal lamina marked with an arrow). The arrowheads mark the plasmalemma. (C) Satellite cells (arrowhead) are observed on the surface of an isolated myofiber using scanning electron microscopy. (D) High magnification of the myofiber in C revealing a satellite cell that is active and undergoing chemotaxis. SC, satellite cell; n, myonuclei; and MF, myofiber.

Figure 68-3 SP cells are enriched in repopulating cells. In addition to the satellite cells, adult skeletal muscle has a second population of cells that participate in repair of muscle. (A) Using Hoechst 33342 dye and dual wavelength flow cytometry, a cell SP (see gated cell population) is identified based on the ability to efflux Hoechst dye in contrast to the main population. (B) SP cells isolated from adult skeletal muscle have been shown to contribute to alternative lineages (i.e., hematopoietic) when placed in a permissive environment.

CHAPTER 68 / STEM CELLS AND MUSCLE REGENERATION

685

Figure 68-4 Myogenic progenitor cells participate in skeletal muscle regeneration. (A) Schematic of one population of progenitor cells (i.e., satellite cells) that are resident in adult skeletal muscle. The quiescent satellite cell occupies a peripheral position with respect to the larger multinucleated myofiber. Following injury the satellite cell is activated (within 2–6 h) and proliferates (d 1–4). These proliferating satellite cells withdraw from the cell cycle and form differentiated central nucleated myofibers. Fusion of additional satellite cells results in an increase in the size of the myofibers and ultimately the nuclei assume a peripheral position in the mature myofiber. Satellite cells are capable of self-renewal to replenish the satellite cell pool. (B) Histological analysis of skeletal muscle 1 d following injury reveals increased inflammation (neutrophils and macrophages) and edema. (C) Two days following injury there is a marked hypercellular response representing, in part, increased numbers of myogenic progenitor cell populations. (D) Restoration of the skeletal muscle architecture is largely complete within 10–14 d. Newly regenerated central nucleated myofibers (arrowhead) are observed at this time period.

(Fig. 68-4). At later stages, macrophages are also associated with the inflammatory response. These inflammatory cell populations liberate a number of interleukins, chemokines, and growth factors that further modulate the stem cell populations and the ECM (Fig. 68-4). A prominent remodeling of the ECM is observed early during the injury/repair process and continues to mature during the final stages of regeneration. The ECM has a number of important functional roles including the regulation of morphogens as well as the establishment of a supportive matrix for the regenerating tissue. Growth factors such as insulin growth factors insulin-like growth factor (IGF)-1 and IGF-2, fibroblast growth factor and transforming growth factor (TGF)-β in concert with cytokines (interleukin-4 and interleukin-13) regulate myogenesis and promote regeneration. Studies have identified the cytokine interleukin-4 as a molecular signal that regulates myoblast fusion with myotubes. Similarly, TGF-β is a multifunctional factor that regulates fibroblasts, ECM maturation, and cellular migration to promote the repair process. Furthermore, the extracellular proteoglycans such as biglycan may bind factors such as TGF-β and thereby modulate the presentation or the sequestration of these factors and impact

the remodeling of the muscle and extracellular matrix. This repair process ultimately results in restoration of the skeletal muscle architecture with a lack of any significant fibrosis (i.e., scar). Although this regenerative response is highly efficient, it is not an infinite process as ultimately there is an exhaustion of satellite cells in pathological states such as muscular dystrophy resulting in a failure of regeneration.

DECREASED REGENERATIVE CAPACITY WITH AGING As age progresses beyond the sixth decade, there is a loss of muscle mass (also known as sarcopenia), a decrease in the size of myofibers and a shift in myofiber composition (i.e., loss of fast twitch glycolytic myofibers), which collectively results in decreased muscle strength. In addition to the loss of muscle strength, aging skeletal muscle is further limited in its ability to undergo efficient regeneration. This impairment in regeneration is multifactorial and highly complex. Although aging muscles are more susceptible to contraction-induced damage, the molecular changes associated with senescent skeletal muscle as well as

686

SECTION VIII / MUSCULOSKELETAL

Figure 68-5 Decreased satellite cell activation/proliferation capacity in aging skeletal myofibers. (A) Activated satellite cells migrate off of the isolated myofibers obtained from young adult mice and proliferate (arrowheads) in culture. (B) Decreased activation/proliferation of satellite cells are observed in myofibers isolated from aged/senescent mice. Inset is a high magnification of the myofiber showing satellite cells that do not become activated and remain associated with the myofiber isolated from aging skeletal muscle.

the host environment further contribute to the impaired regenerative response. Aging skeletal muscle has increased levels of stress proteins (heat shock factors), DNA damaged inducible genes, and oxidative stress inducible genes. The aging immunological system results in decreased phagocytosis of neutrophils and macrophages. In addition, decreased levels of chemokines, hormones (vitamin D, IGF-1, steroid or sex hormones, and so on) and growth factors further contribute to the impaired immunological response. In response to injury of aging skeletal muscle, there is increased fibrosis and adipogenesis resulting in replacement of the myotubes with fat, which negatively impacts the regenerative response and muscle function. Advancing age is associated with decreased absolute numbers of myogenic progenitor cells, which have a decreased proliferative capacity, in part resulting from impaired activation of the progenitor cells to re-enter the cell cycle (Fig. 68-5). Studies using transgenic technologies to overexpress IGF-1 resulted in improved strength, increased satellite cellular proliferation, and improved regeneration of injured, aging skeletal muscle. These transgenic studies suggest that therapeutic intervention using replacement strategies (i.e., cells or hormones) may be effective in the restoration of muscle strength and regeneration with progressing age. In addition, many of the defects associated with aging such as muscle atrophy, decreased absolute numbers of myogenic progenitor cells and decreased growth factors may be related to neuromuscular defects. The neuromuscular junction is a source of stimulation (electrical-contractile coupling) and permissive factors resulting in increased numbers of myogenic progenitor cells at this junction. Perturbation of the neuromuscular junction associated with aging results in the loss of these permissive/supportive factors and physiological stimuli that further contribute to the limitations associated with aging.

PARALLEL PATHWAYS FOR THE DEVELOPMENT AND REGENERATION OF SKELETAL MUSCLE Skeletal myogenesis is a highly coordinated program thought to be initiated in the somite that persists to ultimately contribute to established muscle groups during embryogenesis. Elegant studies utilizing gene disruption strategies have identified bHLH transcription factors including members of the MyoD family that function

in the specification (i.e., Myf5 and MyoD) and differentiation/ maturation (i.e., Myogenin and Mrf4) of developing myofibers. Biochemical and transgenic studies have further demonstrated that these MyoD family members (i.e., Myf5, MyoD, Myogenin, and Mrf4) interact in combination with other transcription factors (such as the myocyte enhancer factor 2 family members) and cell-cycle regulatory proteins to coordinate myogenesis. A number of shared genetic pathways are observed in developing and regenerating skeletal muscle including the MyoD, MADS box, and T-box transcription factor families. Although a number of parallel pathways are observed, repair/regeneration of skeletal muscle is not simply a recapitulation of ontogeny or embryogenesis. For example, a notable difference is observed in the Pax7 null mice that had relatively normal skeletal muscle development but severely impaired muscle regeneration. Furthermore, the signals and cues from adjacent lineages in the developing somite from the neural tube, notochord and overlying ectoderm are markedly different compared to the adult regenerating skeletal muscle which has adjacent ECM, vasculature, and inflammatory/immunological modulatory influences. These shared and distinct genetic pathways during muscle development and regeneration will be further defined using emerging technologies to comprehensively examine the molecular candidates that regulate skeletal muscle regeneration.

SELECTED REFERENCES Asakura A, Seale P, Girgis-Gabardo A, Rudnicki MA. Myogenic specification of side population cells in skeletal muscle. J Cell Biol 2002; 159:123–134. Brockes JP, Kumar A. Plasticity and reprogramming of differentiated cells in amphibian regeneration. Nat Rev Mol Cell Biol 2002;3: 566–574. Chakravarthy MV, Davis BS, Booth FW. IGF-1 restores satellite cell proliferative potential in immobilized old skeletal muscle. J Appl Physiol 2000;89:1365–1379. Conboy IM, Rando TA. The regulation of notch signaling controls satellite cell activation ad cell fate determination in ostnatal myogenesis. Dev Cell 2002;3:397–409. Cornelison DDW, Filla MS, Stanley HM, Rapraeger AC, Olwin BB. Syndecan-3 and syndecan-4 specifically mark skeletal muscle satellite cells and are implicated in satellite cell maintenance and muscle regeneration. Dev Biol 2001;239:79–94.

CHAPTER 68 / STEM CELLS AND MUSCLE REGENERATION

Cornelison DDW, Wold BJ. Single cell analysis of regulatory gene expression in quiescent and activated mouse skeletal muscle satellite cells. Dev Biol 1997;191:270–283. Fuchs E, Segre JA. Stem cells: a new lease on life. Cell 2000;100:143–155. Garry DJ, Meeson AP, Elterman J, et al. Myogenic stem cell function is impaired in mice lacking the forkhead/winged helix protein MNF. Proc Natl Acad Sci USA 2000;97:5416–5421. Goetsch SC, Hawke TJ, Gallardo TD, Richardson JA, Garry DJ. Transcriptional profiling and regulation of the extracellular matrix during muscle regeneration. Physiol Genomics 2003;14:261–271. Goodell MA, Rosenzweig M, Kim H, et al. Dye efflux studies suggest that hematopoietic stem cells expressing low or undetectable levels of CD34 antigen exist in multiple species. Nat Med 1997;3:1337–1345. Gussoni E, Soneoka Y, Strickland CD, et al. Dystrophin expression in the mdx mouse restored by stem cell transplantation. Nature 1999;401: 390–394. Hawke TJ, Garry DJ. Myogenic satellite cells: physiology to molecular biology. J Appl Physiol 2001;91:534–551. Hawke TJ, Jiang N, Garry DJ. Absence of p21 rescues myogenic progenitor cell proliferative and regenerative capacity in Foxk1 null mice. J Biol Chem 2003;278:4015–4020. Horsley V, Jansen KM, Mills ST, Pavlath GK. IL-4 acts as a myoblast recruitment factor during mammalian muscle growth. Cell 2003;113: 483–494. Ivanova NB, Dimos JT, Schaniel C, Hackney JA, Moore KA, Lemischka I. A stem cell molecular signature. Science 2002;298:601–604. Jackson KA, Mi T, Goodell MA. Hematopoietic potential of stem cells isolated from murine skeletal muscle. Proc Natl Acad Sci USA 1999;96:14,482–14,486.

687

McGann CJ, Odelberg SJ, Keating MT. Mammalian myotube dedifferentiation induced by newt regneration extract. Proc Natl Acad Sci USA 2001;98:13,699–13,704. Musaro A, McCullagh K, Paul A, et al. Localized Igf-1 transgene expression sustains hypertrophy and regeneration in senescent skeletal muscle. Nat Genet 2001;27:195–200. Odorico JS, Kaufman DS, Thomson JA. Multilineage differentiation from human embryonic stem cell lines. Stem Cells 2001;19:193–204. Polesskaya A, Seale P, Rudnicki MA. Wnt signaling induces the myogenic specification of resident CD45+ adult stem cells during muscle regeneration. Cell 2003;113:841–852. Ramalho-Santos M, Yoon S, Matsuzaki Y, Mulligan RC, Melton DA. “Stemness”: transcriptional profiling of embryonic and adult stem cells. Science 2002;298:597–600. Seale P, Sabourin LA, Girgis-Gabardo A, Mansouri A, Gruss P, Rudnicki MA. Pax7 is required for the specification of myogenic satellite cells. Cell 2000;102:777–786. Tanaka EM. Regeneration: if they can do it, why can’t we? Cell 2003;113: 559–562. Weindruch R, Kayo T, Lee C, et al. Microarray profiling of gene expression in aging and its alteration by caloric restriction in mice. J Nutr 2001;131:918–923. Weissman IL. Stem cells: units of development, units of regeneration and units in evolution. Cell 2000;100:157–168. Zhou S, Schuetz JD, Bunting KD, et al. The ABC transporter Bcrp1/Abcg2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat Med 2001;7:1028–1034.

69 Skeletal Muscle Hypertrophy and Response to Training ZHEN YAN AND R. SANDERS WILLIAMS SUMMARY Skeletal muscle of adult mammals retains remarkable plasticity. Hypertrophy is evoked by resistance exercise through increased satellite cell activity and activation of the phosphoinositide 3-kinase pathways involving insulin-like growth factor signaling. Skeletal muscle undergoes fast-to-slow fiber type transformation along with enhanced mitochondrial biogenesis and angiogenesis following endurance exercise. Activation of the Ca2+/calmodulin-dependent kinase and phosphatase signaling pathways and exercise-induced expression of peroxisome proliferator-activated receptor-γ coactivator1α play important roles in neuromuscular activity-dependent induction and maintenance of slow muscle gene expression. Key Words: Angiogenesis; calcineurin; Ca2+/calmodulindependent protein kinase; exercise; fiber type transformation; gene regulation; glycogen synthase kinase; hepatocyte growth factor/scatter factor; histone deacetylase; hypertrophy; insulin-like growth factor; mammalian target of rapamycin; mitochondrial biogenesis; myocyte enhancer factor-2; nuclear factor of activated T cells; peroxisome proliferator-activated receptor-γ coactivator1α; phosphoinositide 3-kinase; satellite cell; signal transduction; skeletal muscle.

INTRODUCTION Skeletal muscles of adult mammalian species, including humans, are the source of power for locomotion and other daily activities essential for survival. Loss of skeletal muscle contractile function is a major cause of falling, morbidity, and mortality, especially in elderly populations. More importantly, skeletal muscles collectively influence total body metabolism of glucose, fat, and protein; abnormalities of these functions are associated with a variety of common diseases. Physical inactivity, a decreased use of the musculoskeletal system, is associated with a modern lifestyle and contributes to an epidemic emergence of modern chronic diseases, such as coronary heart disease, hypertension, obesity, type-II diabetes, certain types of cancer, depression, osteoporosis, and sarcopenia. As a nonpharmacological, efficient, and economical therapy, regular exercise has been shown to have significant positive effects on human health and longevity. From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

Skeletal muscle of adult mammals, including humans, retains remarkable plasticity such that it regenerates after injury and adapts to changes in functional demand. Depending on the type and duration of exercise training, skeletal muscle exhibits changes in the size and the structural/functional characteristics of the myofiber population. Extensive research has provided much of the understanding of the phenotypic nature and the molecular mechanisms of these adaptations. A better understanding of the cellular and molecular mechanisms of skeletal muscle adaptation will provide information to guide the use of regular exercise for preventing and treating various diseases, and to facilitate the discovery of new drug targets and the design of new therapeutics to enhance human health. This chapter focuses on findings in signal transduction and gene regulation in skeletal muscle adaptation.

SATELLITE CELL ACTIVITY IN SKELETAL MUSCLE HYPERTROPHY Animal tissue size and shape are, to a certain degree, genetically programmed. However, many of the differentiated, fully functional cells demonstrate changes in size and shape in response to systemic and local stimuli. This adaptability becomes a prerequisite for the survival of animal species. Skeletal muscle hypertrophy (an increase in the volume of individual myofibers) is one of the pronounced responses to high-intensity, weight-bearing exercise interventions, so-called resistance exercise. Experimentally, adult skeletal muscle hypertrophy can be induced by stretch, removal of synergistic muscles, eccentric contraction, or resistance exercise with variations in phenotypic characteristics. Adult skeletal muscle hypertrophy is mediated mainly by intrinsic mechanisms. Responses in the pre-existing mature myofibers and activity of myogenic stem cells, so called satellite cells, are of importance to skeletal muscle hypertrophy. Quiescent satellite cells are located between the basal lamina and the sarcolemma of mature myofibers. On activation, satellite cells reenter the cell cycle, proliferate, and then exit the cell cycle either to renew the quiescent satellite cell pool or to differentiate into mature myofibers. Satellite cell activity is clearly essential to skeletal muscle hypertrophy in vivo, as sterilization of satellite cell activity by irradiation abolishes adult skeletal muscle hypertrophy. To an extent, the myofibers can only expand with the incorporation of new nuclei, because a constant ratio of nuclei to cytoplasmic volume (nuclear domain) likely is maintained throughout all hypertrophic responses. Therefore,

688

CHAPTER 69 / SKELETAL MUSCLE HYPERTROPHY AND RESPONSE TO TRAINING

Figure 69-1 Satellite cell activity in adult skeletal muscle hypertrophy. Hypertrophic stimuli induce IGF-1 expression, as well as NO production, which results in subsequent release of HGS/SC. HGS/SC activates quiescent satellite cells, promotes proliferation, and inhibits differentiation. IGF-1, conversely, promotes satellite cell proliferation and differentiation, depending on the stage of the satellite cells.

skeletal muscle hypertrophy is dependent on activation, proliferation, differentiation, and fusion of satellite cells with the existing myofibers. Disruption or impairment of any of these processes is likely to block or attenuate hypertrophy. As an initial step in the multistep process, the activation of satellite cells plays an essential role in skeletal muscle hypertrophy. But the initial signal, timing, and character of activation are still largely unknown. Although many growth factors and cytokines have been shown to have the ability to modulate the proliferative activity of satellite cells, factors such as fibroblast growth factors, insulin-like growth factors (IGFs), platelet-derived growth factor-BB, transforming growth factor β1 and β2, and epidermal growth factor do not stimulate quiescent satellite cells to enter the cell cycle. Colocalization of hepatocyte growth factor/scatter factor (HGF/SF) with its receptor c-met is thought to be the earliest indicator of satellite cell activation. It has been shown that HGF/SF is sufficient and necessary for satellite cell activation in vivo and in vitro and plays a dual role in regulating skeletal muscle satellite cell activity, by promoting cell proliferation and inhibiting differentiation. Evidence indicates that satellite cell activation is achieved through HGF/SF release mediated by rapid local production of nitric oxide, which plays an important role in compensatory hypertrophy. The overall process of satellite cell activity in skeletal muscle hypertrophy is illustrated in Fig. 69-1.

IGF-1 AND SKELETAL MUSCLE HYPERTROPHY Extensive research efforts have also been directed at many other growth factors with potential roles in regulating satellite cell proliferation and differentiation, among which IGF-1 has received the most attention. IGF-1 mRNA expression is induced through either autocrine or paracrine mechanisms in skeletal muscle undergoing compensatory hypertrophy. Many studies demonstrated a similarly close association of induced IGF-1 mRNA and protein expression with skeletal muscle hypertrophy in vivo. Of particular interest is that both IGF-1 and a muscle-specific splice variant of IGF-1, mechano growth factor, are markedly upregulated in response to increased workload with different time-courses, suggesting that IGF-1 and mechano growth factor play distinct biological roles

689

Figure 69-2 Calcineurin- and PI-3K/Akt-dependent pathways for hypertrophic signaling in adult skeletal myofiber. Hypertrophic stimuli activate Ca2+/CaM-dependent calcineurin directly or indirectly through IGF and its receptor, IGF-1 receptor. Activated calcineurin dephosphorylates NFAT cells and promotes its nuclear translocation and transcriptional activation of the hypertrophic genes. Hypertrophic stimuli also activate PI-3K and Akt, resulting in inhibition of GSK-3 and maintenance of NFAT nuclear distribution. Increased protein synthesis that accompanies skeletal muscle hypertrophy is regulated by mammalian target of rapamycin through Akt activation. Mammalian target of rapamycin stimulates initiation of protein translation by the activation of p70s6k, which phosphorylates the ribosomal protein S6, and the inactivation of 4E-BP1 protein, an inhibitor of the translation initiation factor eIF4E.

during skeletal muscle hypertrophy. IGF-1, when overexpressed in skeletal muscle in transgenic mice or administrated locally by long-term infusion, is sufficient to induce hypertrophy and to counter the decline in muscle mass of dystrophic or aged mice. Consistent with the idea that different IGF-1 isoforms play distinct biological roles was the finding that muscle-specific IGF-1 overexpression alone was not sufficient to prevent unloading-induced atrophy. It is generally believed that IGF-1 induces muscle hypertrophy via a combination of enhancement of satellite cell activity and stimulation of protein synthesis in differentiated myofibers, as demonstrated from in vitro and in vivo observations.

THE CALCINEURIN PATHWAY IGF-1 overexpression or treatment in cultured muscle cells results in myotube hypertrophy through a mechanism involving calcineurin-dependent dephosphorylation and nuclear translocation of nuclear factor of activated T cells (NFAT) (Fig. 69-2). The role of this pathway in physiological models of skeletal muscle hypertrophy in vivo has been less certain and the use of pharmacological inhibitor of calcineurin, cyclosporin (CsA), has given different results in different experimental settings. An argument against the role of calcineurin/NFAT pathway in skeletal muscle hypertrophy is based mainly on the findings that muscle-specific overexpression of constitutively active calcineurin in vivo is sufficient to induce slow muscle gene expression without inducing hypertrophy, and calcineurin activity does not increase during hypertrophy. By comparison, a role for calcineurin-NFAT signaling in cardiac hypertrophy has been established by multiple gain-of-function and

690

SECTION VIII / MUSCULOSKELETAL

loss-of-function experiments. Further research is required to determine whether the calcineurin/NFAT pathway is involved in transducing signals for skeletal muscle hypertrophy.

THE PHOSPHOINOSITIDE 3-KINASE/AKT PATHWAY Research has clearly demonstrated the importance of the phosphoinositide 3-kinase (PI-3K)/Akt signaling pathway in the control of myocyte size in the heart. Accumulating evidence supports a similar role of this pathway in skeletal muscle hypertrophy (Fig. 69-2). A functional role for Akt in control of adult skeletal myofiber size has been assessed in vivo following transfection of constitutively active form of Akt by direct plasmid injection. There are at least two distinct downstream pathways, both of which are essential for skeletal muscle hypertrophy. First, myofibers transfected with constitutively active Akt showed a significant increase in fiber cross-sectional area in a rapamycin-sensitive manner, acting through the downstream target, mammalian target of rapamycin, and phosphorylation of p70s6k and 4E-BP1. Second, glycogen synthase kinase (GSK)-3β, a downstream target protein of Akt, is phosphorylated during compensatory hypertrophy and has been confirmed to be a negative regulator of skeletal muscle hypertrophy, inhibition of which leads to myotube hypertrophy in vitro. One of the possible downstream targets of GSK-3β is NFAT, and decreased GSK-3β would be predicted to maintain NFAT in the nucleus thereby promoting NFAT-dependent transcription.

MUSCLE CONTRACTILE ACTIVITY AND FIBER-TYPE SPECIALIZATION Adult mammalian skeletal muscle fibers are heterogeneously specialized as a result of embryonic and postnatal development, varying in contractile properties, metabolic capacities, ultrastructures, and susceptibility to fatigue. Based on the expression of the predominant isoform of myosin heavy chain protein, adult myofibers can be characterized as type I, IIa, IIx, or IIb. Even though terminally differentiated and incapable of proliferation, adult myofibers in rodents remain plastic such that alteration in contractile load, hormonal shifts, or systemic diseases can induce stable, long-term adaptations. Endurance exercise promotes transformation of IIb to IIa myofibers, an increase in mitochondrial density, and an increase in capillary density. A central role of motor nerve activity in determining skeletal muscle fiber-type composition has been revealed by cross-innervation and electric stimulation studies, which demonstrated complete and reversible transformation of preexisting myofibers by changing patterns of neuronal firing. There has been intense interest for several decades in elucidating the mechanism responsible for the reversible fiber-type transformation. Studying the signaling and molecular mechanisms became possible only with advances in molecular genetics.

THE CALCINEURIN PATHWAY IN FIBER-TYPE SPECIALIZATION Fiber-type specific gene expression in adult skeletal muscles is controlled by signals transduced from the external milieu, which include hormonal, mechanical, and electrical influences, to certain sets of genes in the nuclei of the myofibers. Consistent with a longhypothesized model, exercise activates multiple signal transduction pathways and transcription factors resulting in genetic reprogramming and ultimate phenotypic changes. Calcineurin/NFAT signaling has been proposed to link neuromuscular activity to expression

Figure 69-3 Multiple signaling pathways involved in cooperative activation of slow oxidative gene expression in skeletal myocytes. Tonic neuromuscular activity induces low amplitude, sustained elevation of intracellular Ca2+ and its binding to CaM. Ca2+/CaM dependent activation of calcineurin promotes NFAT-cells nuclear translocation, whereas Ca2+ signal promotes the maintenance of NFAT nuclear distribution by phosphorylation-mediated deactivation of GSK-3β. Ca2+/CaM signal activates Ca2+/CaM-dependent protein kinase and promotes PPAR-γ coactivator-1 expression, facilitating muscle-enriched factor 2 and NRF function in maintenance and regulation of the slow oxidative genes.

of genes characteristic of slow oxidative myofibers (Fig. 69-3). Supporting this central hypothesis were the findings that treatment with the calcineurin inhibitor, cyclosporin A, blocks nuclear localization of NFAT and slow muscle gene expression, and overexpression of constitutively active calcineurin in fast-twitch skeletal muscle increases the percentage of slow myofibers. Other studies suggest different mechanisms of slow muscle gene expression. The positive findings suggest a molecular mechanism in which signals from neuromuscular activity are translated by Ca2+/calmodulin (CaM) to protein kinases and phosphatases that modulate the functions of transcription factors controlling mitochondrial biogenesis and fastto-slow fiber-type switching. More specifically, tonic neuromuscular activity induces a low-amplitude sustained elevation of intracellular Ca2+ concentration and the binding to the intracellular Ca2+ sensor protein CaM. Ca2+-bound CaM activates Ca2+/CaM dependent serine/threonine phosphatase, calcineurin, which dephosphorylates NFAT and myocyte enhancer factor-2 (MEF-2) and promotes their transactivation activities on the target genes cooperatively (Fig. 69-3). This notion is supported by findings that nuclear translocation and distribution of NFAT are specifically dependent on slow neuromuscular activity. This calcineurin-dependent pathway is regulated in the skeletal muscle by means of a negative feedback mechanism conferred by calcineurin-dependent expression of an endogenous calcineurin inhibitor, modular calcineurin inhibitory protein. In addition, endurance exercise has been shown to deactivate GSK-3α and GSK-3β in rats in an Akt-independent manner (Fig. 69-3). Because GSK-3-mediated phosphorylation of NFAT promotes its export from the nucleus, thereby antagonizing NFAT-dependent transcription, it is suggested that exercise-induced GSK-3β phosphorylation plays a synergistic role with calcineurin in promoting the expression of the slow muscle genetic program. Identification of the upstream regulators responsible for the phosphorylation of GSK-3 in skeletal muscle during exercise is anticipated.

CHAPTER 69 / SKELETAL MUSCLE HYPERTROPHY AND RESPONSE TO TRAINING

MITOCHONDRIAL BIOGENESIS AND SLOW MUSCLE GENE EXPRESSION Findings in peroxisome proliferator-activated receptor (PPAR)-γ coactivator-1α (PGC-1α) gene regulation and function have led to the consideration of PGC-1α as a key regulator of important features of skeletal muscle adaptation. PGC-1α is a transcriptional coactivator cloned originally by a yeast two-hybrid screen from a differentiated brown fat cell line using PPAR-γ as bait. PGC-1α mRNA and protein are highly expressed in slow, oxidative fibers compared to the fast, glycolytic fibers, consistent with the function of a gene involved in fiber-type specialization. Several different models have suggested the functional importance of PGC-1α in striated muscles. A possible association between polymorphism of the PGC-1α gene and type-II diabetes has also been suggested. PGC-1α protein interacts with many nuclear regulatory proteins, such as thyroid hormone receptor, estrogen receptor α, PPAR-γ, nuclear respiratory factor-1 (NRF-1) and CREB-binding protein. PGC-1α also stimulates the expression of multiple effector genes during various biological processes. The most striking property of PGC-1α function in skeletal muscle is that its overexpression in cultured myoblasts induces mitochondrial biogenesis, and overexpression in transgenic mice causes a remarkable fast-to-slow fiber-type switching and enhanced mitochondrial biogenesis. These findings suggest the possibility that increased expression of PGC-1α in skeletal muscle plays a key role in endurance exercise-induced skeletal muscle adaptation by promoting mitochondrial biogenesis and other genetic and biochemical events. The profound effect of PGC-1α overexpression probably results from the enhancement of the transactivation activities of regulatory proteins, such as NRF-1 and MEF-2, as well as the induction of the genes in encoding these transcription factors. The importance of PGC-1α function in promoting the acquisition of slow-twitch muscle contractile and metabolic properties also is suggested by the observation that endurance exercise induces PGC-1α mRNA expression in the recruited skeletal muscle through transcriptional control. PGC-1α protein expression along with that of NRF-1 and NRF-2 are also increased in exercised skeletal muscle. Endurance exercise is likely to induce PGC-1α gene expression through a combination of transcriptional activation of the PGC-1α gene and increased protein stability. Studies to elucidate the signal transduction pathways and the molecular mechanism for exerciseinduced PGC-1α gene expression in skeletal muscle will yield critically important information.

THE CA2+/CAM-DEPENDENT PROTEIN KINASE PATHWAY Ca2+/CaM, acting through calcineurin and Ca2+/CaM-dependent protein kinase (CaMK) pathways, stimulates MEF-2 transactivation activity and downstream target genes, such PGC-1α (Fig. 69-3). Relatively little is known about the exact CaMK pathway(s) in exercise-induced skeletal muscle adaptation. A molecular mechanism for CaMK-induced activation of MEF-2 activity has been elucidated in cardiac myocytes. Phosphorylation of histone deacetylase 5 results in its binding with the intracellular chaperone protein 14-3-3 and promotes its nuclear export leading to derepression of MEF-2 activity. This mechanism plays a role in maintenance of mitochondrial biogenesis in the heart. However, the endogenous kinase that is responsible for transducing the signal and catalyzing the dephosphorylation of HDAC remains unknown. Furthermore, it is not known whether the same regulatory mechanism accounts for the adaptation in skeletal muscle in response to exercise training.

691

The effects of activated CaMKIV in skeletal muscle in inducing PGC-1α gene expression, enhancing mitochondrial biogenesis, and promoting myosin isoform switching in a transgenic mouse model have been described. In this model, a constitutively active form of CaMKIV was expressed in fast-twitch myofibers under the control of the muscle creatine kinase promoter. A significant increase in percentage of slow-twitch fibers in normally fast-twitch muscles and a marked increase in subsarcolemmal mitochondrial density were confirmed along with increases in various genetic and biochemical markers for a slow oxidative muscle phenotype, including a profound induction of PGC-1α mRNA. Another study reported that energy deprivation-induced mitochondrial biogenesis is dependent on an activation of the AMP-activated protein kinase pathway, which was associated with increased CaMKIV protein and PGC-1α mRNA expression. These data suggest a role for CaMK signaling in the acquisition of a slow-twitch muscle phenotype in adults, but do not identify the specific isoforms of CaMK that are relevant to this response.

CONCLUSIONS Cell size and specialized characteristics of myofibers in adult skeletal muscles are controlled by multiple signaling pathways. Ca2+-dependent activation of calcineurin-NFAT signaling and exercise-induced induction of PGC-1α are both involved in the activity- dependent induction and maintenance of fiber-type-specific gene expression. Satellite cell activity and intracellular signaling activities involving the PI-3K/Akt/mammalian target of rapamycin and PI-3K/ Akt/GSK-3 pathways in mature myofibers appear to mediate skeletal muscle hypertrophy in response to hypertrophic stimuli, such as resistance exercise. The future challenge is to define relationships between these and other signaling pathways that are responsible for neuromuscular activity-mediated adaptations. Such knowledge may foster the development of novel therapeutics and preventive technologies to improve human fitness and health.

SELECTED REFERENCES Adams GR, McCue SA. Localized infusion of IGF-I results in skeletal muscle hypertrophy in rats. J Appl Physiol 1998;84:1716–1722. Baar K, Wende AR, Jones TE, et al. Adaptations of skeletal muscle to exercise: rapid increase in the transcriptional coactivator PGC-1. FASEB J 2002;16:1879–1886. Barton ER, Morris L, Musaro A, Rosenthal N, Sweeney HL. Muscle-specific expression of insulin-like growth factor I counters muscle decline in mdx mice. J Cell Biol 2002;157:137–148. Beals CR, Sheridan CM, Turck CW, Gardner P, Crabtree GR. Nuclear export of NF-ATc enhanced by glycogen synthase kinase-3. Science 1997;275:1930–1934. Bodine SC, Stitt TN, Gonzalez M, et al. Akt/mTOR pathway is a crucial regulator of skeletal muscle hypertrophy and can prevent muscle atrophy in vivo. Nat Cell Biol 2001;3:1014–1019. Chakravarthy MV, Abraha TW, Schwartz RJ, Fiorotto ML, Booth FW. Insulin-like growth factor-I extends in vitro replicative life span of skeletal muscle satellite cells by enhancing G1/S cell cycle progression via the activation of phosphatidylinositol 3′-kinase/Akt signaling pathway. J Biol Chem 2000;275:35,942–35,952. Chin ER, Olson EN, Richardson JA, et al. A calcineurin-dependent transcriptional pathway controls skeletal muscle fiber type. Genes Dev 1998;12:2499–2509. Coleman ME, DeMayo F, Yin KC, et al. Myogenic vector expression of insulin-like growth factor I stimulates muscle cell differentiation and myofiber hypertrophy in transgenic mice. J Biol Chem 1995;270: 12,109–12,116.

692

SECTION VIII / MUSCULOSKELETAL

Czubryt MP, McAnally J, Fishman GI, Olson EN. Regulation of peroxisome proliferator-activated receptor gamma coactivator 1 alpha (PGC-1 alpha) and mitochondrial function by MEF2 and HDAC5. Proc Natl Acad Sci USA 2003;100:1711–1716. Delling U, Tureckova J, Lim HW, De Windt LJ, Rotwein P, Molkentin JD. A calcineurin-NFATc3-dependent pathway regulates skeletal muscle differentiation and slow myosin heavy-chain expression. Mol Cell Biol 2000;20:6600–6611. DeVol DL, Rotwein P, Sadow JL, Novakofski J, Bechtel PJ. Activation of insulin-like growth factor gene expression during work-induced skeletal muscle growth. Am J Physiol 1990;259:E89–E95. Dunn SE, Chin ER, Michel RN. Matching of calcineurin activity to upstream effectors is critical for skeletal muscle fiber growth. J Cell Biol 2000;151:663–672. Lin J, Wu H, Tarr PT, et al. Transcriptional co-activator PGC-1 alpha drives the formation of slow-twitch muscle fibres. Nature 2002;418:797–801. Liu Y, Cseresnyes Z, Randall WR, Schneider MF. Activity-dependent nuclear translocation and intranuclear distribution of NFATc in adult skeletal muscle fibers. J Cell Biol 2001;155:27–39. Lu J, McKinsey TA, Nicol RL, Olson EN. Signal-dependent activation of the MEF2 transcription factor by dissociation from histone deacetylases. Proc Natl Acad Sci USA 2000;97:4070–4075. Markuns JF, Wojtaszewski JF, Goodyear LJ. Insulin and exercise decrease glycogen synthase kinase-3 activity by different mechanisms in rat skeletal muscle. J Biol Chem 1999;274:24,896–24,900. McKinsey TA, Zhang CL, Lu J, Olson EN. Signal-dependent nuclear export of a histone deacetylase regulates muscle differentiation. Nature 2000;408:106–111. Michael LF, Wu Z, Cheatham RB, et al. Restoration of insulin-sensitive glucose transporter (GLUT4) gene expression in muscle cells by the transcriptional coactivator PGC-1. Proc Natl Acad Sci USA 2001;98: 3820–3825. Molkentin JD, Lu JR, Antos CL, et al. A calcineurin-dependent transcriptional pathway for cardiac hypertrophy. Cell 1998;93:215–228. Musaro A, McCullagh KJ, Naya FJ, Olson EN, Rosenthal N. IGF-1 induces skeletal myocyte hypertrophy through calcineurin in association with GATA-2 and NF-ATc1. Nature 1999;400:581–585. Naya FJ, Mercer B, Shelton J, Richardson JA, Williams RS, Olson EN. Stimulation of slow skeletal muscle fiber gene expression by calcineurin in vivo. J Biol Chem 2000;275:4545–4548. Pallafacchina G, Calabria E, Serrano AL, Kalhovde JM, Schiaffino S. A protein kinase B-dependent and rapamycin-sensitive pathway controls skeletal muscle growth but not fiber type specification. Proc Natl Acad Sci USA 2002;99:9213–9218. Pilegaard H, Saltin B, Neufer PD. Exercise induces transient transcriptional activation of the PGC-1alpha gene in human skeletal muscle. J Physiol 2003;546:851–858. Puigserver P, Rhee J, Lin J, et al. Cytokine stimulation of energy expenditure through p38 MAP kinase activation of PPARgamma coactivator-1. Mol Cell 2001;8:971–982.

Puigserver P, Wu Z, Park CW, Graves R, Wright M, Spiegelman BM. A cold-inducible coactivator of nuclear receptors linked to adaptive thermogenesis. Cell 1998;92:829–839. Rommel C, Bodine SC, Clarke BA, et al. Mediation of IGF-1-induced skeletal myotube hypertrophy by PI(3)K/Akt/mTOR and PI(3)K/Akt/GSK3 pathways. Nat Cell Biol 2001;3:1009–1013. Rosenthal SM, Cheng ZQ. Opposing early and late effects of insulin-like growth factor I on differentiation and the cell cycle regulatory retinoblastoma protein in skeletal myoblasts. Proc Natl Acad Sci USA 1995;92:10,307–10,311. Rothermel B, Vega RB, Yang J, Wu H, Bassel-Duby R, Williams RS. A protein encoded within the down syndrome critical region is enriched in striated muscles and inhibits calcineurin signaling. J Biol Chem 2000;275:8719–8725. Schiaffino S, Reggiani C. Molecular diversity of myofibrillar proteins: Gene regulation and functional significance. Physiol Rev 1996;76:371–423. Semsarian C, Wu MJ, Ju YK, et al. Skeletal muscle hypertrophy is mediated by a Ca2+-dependent calcineurin signalling pathway. Nature 1999;400:576–581. Serrano AL, Murgia M, Pallafacchina G, et al. Calcineurin controls nerve activity-dependent specification of slow skeletal muscle fibers but not muscle growth. Proc Natl Acad Sci USA 2001;98:13,108–13,113. Shioi T, McMullen JR, Kang PM, et al. Akt/protein kinase B promotes organ growth in transgenic mice. Mol Cell Biol 2002;22:2799–2809. Swoap SJ, Hunter RB, Stevenson EJ, et al. The calcineurin-NFAT pathway and muscle fiber-type gene expression. Am J Physiol Cell Physiol 2000;279:C915–C924. Taigen T, De Windt LJ, Lim HW, Molkentin JD. Targeted inhibition of calcineurin prevents agonist-induced cardiomyocyte hypertrophy. Proc Natl Acad Sci USA 2000;97:1196–1201. Tatsumi R, Anderson JE, Nevoret CJ, Halevy O, Allen RE. HGF/SF is present in normal adult skeletal muscle and is capable of activating satellite cells. Dev Biol 1998;194:114–128. Tatsumi R, Hattori A, Ikeuchi Y, Anderson JE, Allen RE. Release of hepatocyte growth factor from mechanically stretched skeletal muscle satellite cells and role of pH and nitric oxide. Mol Biol Cell 2002;13:2909–2918. Wu H, Kanatous SB, Thurmond FA, et al. Regulation of mitochondrial biogenesis in skeletal muscle by CaMK. Science 2002;296:349–352. Wu H, Rothermel B, Kanatous S, et al. Activation of MEF2 by muscle activity is mediated through a calcineurin-dependent pathway. EMBO J 2001;20:6414–6423. Wu Z, Puigserver P, Andersson U, et al. Mechanisms controlling mitochondrial biogenesis and respiration through the thermogenic coactivator PGC-1. Cell 1999;98:115–124. Yang S, Alnaqeeb M, Simpson H, Goldspink G. Cloning and characterization of an IGF-1 isoform expressed in skeletal muscle subjected to stretch. J Muscle Res Cell Motil 1996;17:487–495. Zong H, Ren JM, Young LH, et al. AMP kinase is required for mitochondrial biogenesis in skeletal muscle in response to chronic energy deprivation. Proc Natl Acad Sci USA 2002;99:15,983–15,987.

70 Muscular Dystrophies Mechanisms

PETER B. KANG AND LOUIS M. KUNKEL SUMMARY Since the cloning of the gene responsible for Duchenne and Becker muscular dystrophy two decades ago, a number of other genes have been associated with various forms of muscular dystrophy. The protein products of these genes display a diversity of cellular localizations and functions, suggesting a complex pathophysiology within this class of disorders. Two recently identified functions of selected proteins include membrane repair and glycosylation, providing important insights into the disease process. A definitive cure remains elusive, but supportive therapies have become increasingly sophisticated, improving the quality of life and lengthening the life expectancy for many affected individuals. Key Words: Becker; calpain; congenital muscular dystrophy; duchenne; dysferlin; dystroglycan; dystrophin; facioscapulohumeral muscular dystrophy; limb-girdle; merosin; muscular dystrophy; sarcoglycan.

INTRODUCTION The muscular dystrophies are a group of inherited primary diseases of muscle that are characterized clinically by progressive, chronic weakness, and pathologically by degeneration of muscle fibers with necrosis and connective tissue infiltration. This definition distinguishes muscular dystrophies from other primary diseases of muscle, including the congenital myopathies and inflammatory myopathies, although there are some areas of overlap. The clinical course of muscular dystrophy varies widely between specific diseases from the rapidly fatal Walker–Warburg disease to the milder Becker muscular dystrophy. There are four major categories of muscular dystrophy: dystrophinopathies, limb-girdle muscular dystrophies (LGMD), congenital muscular dystrophies, and the syndromic muscular dystrophies (Table 70-1). These categories are largely based on the clinical characteristics of the diseases and do not reflect the differences in molecular and biochemical origins. An international collaborative effort in the mid-1980s led to positional cloning of dystrophin, the gene on the X-chromosome responsible for Duchenne muscular dystrophy. Since then, the genes that cause more than a dozen other types of muscular dystrophy have been described (summarized at http://www.dmd.nl/). These proteins, with other associated muscle From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

proteins that have not been shown to play a primary role in human disease, have been extensively studied (Fig. 70-1). The majority of the proteins produced by these genes appear primarily to be structural in function and are linked to each other in the vicinity of the sarcolemma. Some of these proteins may also play a role in cellular signaling and repair. Very few are enzymes.

DYSTROPHINOPATHIES Duchenne muscular dystrophy is caused by mutations in the DMD gene on the X-chromosome. It is the most common muscular dystrophy, occurring in 1 in 3300 live male births. The high incidence is because of the large size of the gene, which has 79 exons and produces a 427-kDa protein called dystrophin. The majority of mutations are deletions that cluster in exons 2–20 and 44–53, areas in which the intronic sequences are unusually long, approaching 200 bp. Polymerase chain reaction amplification techniques concentrating on these exons identify about two-thirds of affected patients. The remainders have point mutations, deletions in other regions, and duplications. Becker muscular dystrophy, a milder variant, is also associated with mutations in DMD. Shifting or preservation of the reading frame is more important than the size of the deletion in determining whether an affected child will have the Duchenne vs Becker phenotype; most children with Duchenne muscular dystrophy have a frameshift mutation, whereas those with Becker muscular dystrophy typically have a preserved translational reading frame. The dystrophin protein is expressed in a number of organs besides muscle. Brain, Purkinje cells, and muscle produce the fulllength 427 kDa protein. Retina, Schwann cells, glial cells, and kidney produce shorter isoforms. In skeletal muscle, dystrophin is located in the subsarcolemmal region and has four major domains: amino terminus, rod domain, cysteine-rich domain, and carboxy terminus. The amino terminus binds to actin, whereas the carboxy terminus binds to a protein complex that extends into the extracellular matrix. Dystrophin’s main role appears to be structural, although there is evidence that it, with its associated proteins, may be involved in more dynamic functions such as regulating the perfusion and repair of muscle. Clinically, Duchenne muscular dystrophy typically presents with gait abnormalities between the ages of 3–5 yr, although in retrospect parents may recall symptoms as early as 1 yr of age. The gait abnormalities may include toe walking, frequent falling

693

694

SECTION VIII / MUSCULOSKELETAL

Table 70-1 Muscular Dystrophies Disorder

Gene

Protein location

Function

Dystrophinopathies Duchenne MD Becker MD

DMD DMD

Subsarcolemma Subsarcolemma

Structural Structural

Dominant limb-girdle muscular dystrophies 1A TTID 1B LMNA 1C CAV3 1D ? (6q23) 1E ? (7q) 1F ? (5q31)

Sarcomere Nucleus Sarcolemma Unknown Unknown Unknown

Structural Structural Structural Unknown Unknown Unknown

Recessive limb-girdle muscular dystrophies 2A CAPN3 2B DYSF 2C SGCG 2D SGCA 2E SGCB 2F SGCD 2G TCAP 2H TRIM32 2I FKRP 2J TTN 2K POMT1

Cytoplasm Sarcolemma Sarcolemma Sarcolemma Sarcolemma Sarcolemma Sarcomere Unknown ER Sarcomere Golgi complex

Protease Unknown Structural Structural Structural Structural Structural Ligase? Glycosylation Structural Glycosylation

Congenital muscular dystrophies Ullrich disease COL6A1, A2, A3 Rigid spine SEPN1 Merosin-negative LAMA2 MDC1C FKRP MDC1D LARGE Fukuyama FCMD Muscle-eye brain POMGnT1 Walker-Warburg POMT1 FCMD FKRP

Extracellular ER Extracellular ER Golgi complex Golgi complex Golgi complex Golgi complex Golgi complex ER

Structural Unknown Structural Glycosylation Glycosyltransferase Glycosylation Glycosylation Glycosylation Glycosylation Glycosylation

Syndromic muscular dystrophies Emery-Dreifuss MD EMD LMNA Facioscapulohumeral ? (4q35) Oculopharyngeal MD PABP2

Nucleus Nucleus Unknown Nucleus

Unknown Structural Unknown Unknown

ER, endoplasmic reticulum; ?, gene not identified.

and tripping, waddling gait, and delayed walking. Examination is notable for weakness of the proximal muscles, most marked in the hip girdle in the early stages and more widespread with disease progression. Reflexes are initially normal or reduced and become absent in the later stages. Becker muscular dystrophy has a similar but milder presentation with a typical onset in later childhood or adolescence and a slower progression of weakness. In suspected cases of Duchenne or Becker muscular dystrophy, initial laboratory evaluation should include measurement of muscle enzymes in the serum: creatine phosphokinase (CPK), aldolase, alanine aminotransferase, aspartate transaminase, and lactate dehydrogenase. The first two are most specific for muscle; the others are also produced in the liver. These enzymes are elevated in both diseases, but more markedly in Duchenne muscular dystrophy. The diagnosis may then be confirmed in about two-thirds of cases by polymerase chain reaction amplification of

the most commonly deleted exons on blood leukocytes. When this test is negative, a muscle biopsy with antibody staining for dystrophin confirms or excludes the diagnosis. Routine histochemical staining of muscle tissue reveals muscle necrosis and inflammation with replacement of muscle fibers by connective tissue. On antibody staining, dystrophin is virtually absent in Duchenne muscular dystrophy and present in abnormal size or amount in Becker muscular dystrophy. The need for about one-third of boys with Duchenne muscular dystrophy to undergo muscle biopsy, an invasive procedure that requires general anesthesia in children, is not ideal. Robust DNA sequencing of all 79 exons in children suspected of having Duchenne muscular dystrophy should enable more cases to be confirmed without muscle biopsy. The course is progressive with Duchenne muscular dystrophy patients often surviving into early adulthood and Becker muscular dystrophy patients surviving well into adulthood. The most common causes of death are respiratory and cardiac complications.

CHAPTER 70 / MUSCULAR DYSTROPHIES

695

Figure 70-1 Schematic diagram of muscle fiber architecture, including dystrophin-associated protein complex and other proteins involved in the muscular dystrophies. Rounded rectangles indicate discrete cellular regions or compartments, single lines indicate physical associations between proteins, double lines indicate phospholipid membranes. FKRP, fukutin-related protein; POMT1, O-mannosyltransferase 1; POMGnT1, O-mannose β-1,2-N-acetylglucosaminyltransferase.

LIMB-GIRDLE MUSCULAR DYSTROPHIES Clinically and histologically, the LGMDs are similar to the dystrophinopathies, with a progressive course of predominantly proximal weakness and signs of inflammation, necrosis, and connective tissue replacement on biopsy. The severity ranges widely from a mild course similar to Becker muscular dystrophy to a more rapidly progressive one that resembles Duchenne muscular dystrophy. Cardiac complications appear to be common in types 1B, 1D, 2G, and 2I, but the overall frequency is unclear because of the small number of patients reported for a number of subtypes. Intelligence is usually normal. The dystrophinopathies may be regarded as an X-linked recessive category of LGMD, but for historical reasons are usually classified separately. The genes that cause certain LGMDs (1B, 1C, 2B, and 2I) have also been shown to cause other diseases. Certain individual mutations in caveolin and dysferlin are associated with multiple phenotypes, suggesting the involvement of modifying factors. AUTOSOMAL-DOMINANT LGMD (TYPE 1) Biochemically, the protein products of the identified autosomal-dominant LGMDs do not appear to have much in common. The common thread that runs through these disorders is the inheritance pattern and the generally mild course of disease compared with the autosomalrecessive LGMDs. Type 1 is much rarer than type 2, comprising less than 10% of cases. LGMD type 1A is caused by mutations in the myotilin gene, whose protein product localizes to the Z-disk of the sarcomere. Myotilin’s exact function is unknown, but the protein appears to play a structural role in the Z-disk. Mutations lead to abnormalities in the Z-disk, which may appear under light microscopy as rodshaped structures similar to the rods found in nemaline myopathy. Patients with LGMD type 1A have, in addition to muscle weakness, a distinctive dysarthric pattern of speech. The age of onset is usually in early adulthood. In the initial stages, the weakness is

mostly proximal, but slowly spreads to involve distal muscles as the disease progresses. CPK levels are significantly elevated. Mutations in the lamin A/C gene cause LGMD 1B, Emery– Dreifuss muscular dystrophy, cardiomyopathy, and lipodystrophy. A missense mutation in the lamin A portion of the gene has been identified as a cause of Hutchinson–Gilford progeria. The two protein products, lamins A and C, are produced by alternative splicing and localize to the nuclear lamina. The mechanism of disease is unclear, but may be related to impaired interactions with other nuclear proteins or dysfunctional filament assembly. Cardiac complication rates are high in both LGMD 1B and Emery–Dreifuss muscular dystrophy. The two diseases are distinguished clinically by the absence of early contractures and a proximal pattern of weakness in LGMD 1B. Mutations in the gene encoding caveolin 3 cause LGMD 1C, rippling muscle disease, distal myopathy, and hyper-CKemia. The same mutation in the caveolin gene has been shown to cause rippling muscle disease with distal myopathy in some individuals, and rippling muscle disease with LGMD in others. Caveolins are the proteins found in caveolae, which are 50–100 nm invaginations of the plasma membrane found in most cell types, including muscle. Caveolae are involved in membrane trafficking, sorting, transport, and signal transduction. Caveolin 3, a muscle-specific caveolin protein, forms a hairpin loop within the sarcolemma, enabling both the amino- and carboxy-terminal ends to face the cytoplasm. The protein coprecipitates with dystrophin, indicating a link to the dystrophin-associated protein complex. This connection is further supported by the development of dystrophin deficiency in mice when caveolin 3 is overexpressed. In mouse models, a deficiency of caveolin 3 leads to a deficiency of caveolae and abnormalities in the T-tubule system. Onset of proximal muscle weakness in LGMD 1C is typically at 5 yr, with calf hypertrophy, Gowers’ sign, and occasional reports of exercise intolerance. CPK levels are significantly elevated.

696

SECTION VIII / MUSCULOSKELETAL

In LGMD 1D–1F, the genetic loci have been mapped, but the exact genes have not been identified. Genetic linkage has identified the locus responsible for LGMD 1D to a 3-cM region of 6q23. Two nearby laminins have been excluded as causes, but their proximity suggests that a previously unknown laminin may be the gene responsible. The most prominent symptoms are cardiac with dilated cardiomyopathy and cardiac conduction defects the most common manifestations. Sudden death may occur. Onset is typically in early adulthood, sometimes in late adolescence. Mild proximal muscle weakness frequently occurs, and may precede or follow the initial cardiac symptoms. Serum CPK levels are mildly elevated. The phenotype resembles that of LGMD 1B. The genetic region of interest in LGMD 1E appears to be a 9-cM interval on 7q, although the odds ratio for this region is not ideal. Additional kindreds must be analyzed to confirm this finding and pinpoint the responsible gene. Clinical features in affected individuals include the classic pattern of proximal muscle weakness. Vocal cord and pharyngeal weakness with autosomal-dominant distal myopathy has been localized to a 12-cM interval on 5q31. Although it is a distal myopathy with little evidence of inflammation on muscle biopsy, it is classified as LGMD 1F. The onset of symptoms is typically in middle adulthood. Clinical manifestations include voice change, finger extensor weakness, peroneal weakness, shoulder girdle weakness, and dysphagia. Serum CPK levels may be mildly elevated. AUTOSOMAL-RECESSIVE LGMD (TYPE 2) The autosomalrecessive LGMDs (type 2) are more common than the dominant forms, and generally have a more severe clinical course. The gene encoding calpain 3 is mutated in LGMD 2A. Calpain 3 is a muscle-specific calcium-activated proteolytic enzyme, the first nonstructural protein demonstrated to cause LGMD. Calpain 3 has been shown to cleave filamin 2, but the exact pathogenesis of disease remains obscure; the enzyme is quickly degraded, and is, thus, difficult to study. In human muscle biopsy tissue and mouse models, calpain-3 deficiency is associated with myonuclear apoptosis and abnormal expression of the transcriptional factor nuclear factor-κB and its inhibitor (IκBα). Injection of Evans blue dye, which cannot cross intact membranes, into the mouse model revealed disruption of the sarcolemma, a phenomenon seen commonly in other muscular dystrophies. It is not clear whether the apoptotic cascade or membrane disruption is the more primary event. LGMD 2A (calpainopathy), thought to be the most common form of LGMD, is characterized by progressive, predominantly proximal weakness, without cardiac or facial involvement. The severity varies widely and does not correlate consistently with the level of expression of calpain 3 in muscle. Mutations in the gene encoding dysferlin cause LGMD 2B, Miyoshi myopathy, and distal myopathy with anterior tibial onset. Deletion of 1 bp causes both Miyoshi myopathy and distal myopathy with tibial onset in different kindreds, whereas missense mutations cause Miyoshi myopathy and LGMD 2B in different individuals within the same family. The gene is large, consisting of 55 exons and a 6243 bp open reading frame, and is expressed mainly in skeletal and cardiac muscles. The dysferlin protein plays an important role in the calcium-mediated process of membrane resealing, which is crucial for the stability of the muscle membrane. LGMD 2B has a proximal pattern of weakness; Miyoshi myopathy tends to involve more distal muscles, particularly gastrocnemius. Difficulty standing on the toes is a common initial complaint of patients with Miyoshi myopathy. Onset is typically in childhood or

early adulthood. CPK levels are markedly elevated in all the phenotypes. The course is milder than in the other recessive LGMDs; only about 10% of patients lose the ability to ambulate. The four sarcoglycanopathies are LGMDs 2C–2F and are linked to mutations in the genes encoding γ-, α-, β-, and δ-sarcoglycans, respectively. A fifth sarcoglycan (ε) has been identified. Mutations in ε-sarcoglycan cause a dystonia rather than a muscular dystrophy. The sarcoglycans form a complex that spans the sarcolemma. Because they are physically linked, a mutation in the gene encoding one sarcoglycan often causes a secondary deficiency in one or more of the others, especially when primary mutations in β- and δ-sarcoglycan are present. This frequently makes it difficult to diagnose a sarcoglycanopathy based entirely on antibody staining of muscle tissue, even though antibodies to all four disease-causing sarcoglycans are available. Of the four sarcoglycanopathies, α-sarcoglycanopathy frequently has a mild phenotype; all reported cases of δ-sarcoglycanopathy have been severe. The telethonin protein localizes to the Z-disk of the sarcomere and is expressed in skeletal and cardiac muscles. Mutations in its gene cause LGMD 2G. Its specific function is unknown, but it interacts with titin, mutations of which cause LGMD 2J. Patients with LGMD 2G typically present in late childhood or adolescence with proximal weakness, sometimes accompanied by distal weakness that may resemble the phenotype of Miyoshi myopathy. A significant minority of patients lose the ability to ambulate independently in the third or fourth decade. Cardiac complications are common. LGMD 2H is caused by mutations in the gene encoding TRIM 32, an E3 ubiquitin ligase expressed in skeletal and cardiac muscle. A deficiency of the enzyme could presumably cause the accumulation of its target protein, but histological examination has not revealed any unusual deposits in muscle fibers. The clinical disease is mild and slowly progressive. Mutations in the gene encoding fukutin-related protein (FKRP) cause both LGMD 2I and congenital muscular dystrophy 1C. In contrast to the genotype–phenotype correlations with dysferlin, different mutations in FKRP correspond to the two different phenotypes, suggesting that the mutations themselves are the primary determinants of the disease manifestations. Patients with LGMD 2I frequently have a C826A (Leu276Ileu) mutation. In more severely affected patients, the patterns of protein expression may resemble that of the more severe forms of congenital muscular dystrophy, with secondary deficiencies of laminin-α2 and α-dystroglycan observed on muscle tissue immunohistochemistry, Western blots, or both. In the more severe form of LGMD 2I, children develop hypotonia, motor delays, and muscle hypertrophy (which sometimes causes macroglossia), usually in the first 2 yr of life. Clinically, they resemble patients with Duchenne muscular dystrophy, with cardiac complications and loss of ambulation in early adolescence. The milder form of LGMD 2I has been described in a Tunisian family, having a later onset and a slower progression, with preservation of ambulation in nearly all cases. CPK levels are markedly elevated in both forms. A sarcomeric protein named titin has been demonstrated to be mutated in LGMD 2J. These patients have proximal weakness with elevated CPK levels. Immunohistochemistry of muscle tissue reveals a secondary deficiency of calpain 3. A novel form of LGMD, LGMD2K, has recently been described in a small number of Turkish patients. Affected individuals tend to have mild proximal muscle weakness, mild muscle hypertrophy,

CHAPTER 70 / MUSCULAR DYSTROPHIES

and marked elevations in serum CK levels. Onset is between 1 and 3 yr, although this range may be expanded as further patients are identified. Patients remain ambulatory at least until late adolescence. Unusual features include microcephaly and mental retardation. However, brain imaging so far has not revealed any structural abnormalities. Histologically, there is a reduction of α-dystroglycan staining in muscle tissue. A mutation in O-mannosyltransferase-1 (POMT1) has been identified in this cohort (A200P). The genetic causes of LGMD have most likely not been fully described. Other genes responsible for this syndrome may yet be identified.

CONGENITAL MUSCULAR DYSTROPHIES Congenital muscular dystrophies generally share with the other muscular dystrophies the characteristic findings on routine muscle histology. As the name implies, patients typically present at birth with hypotonia and weakness. There are three broad categories of congenital muscular dystrophies, which may be distinguished by the immunohistochemical staining pattern of laminin-α2 on muscle biopsy and by the clinical course. The mild laminin-α2 (merosin)-positive congenital muscular dystrophies are associated with normal MRI findings and normal cognitive development. Patients with the moderately severe congenital muscular dystrophies typically have white matter changes on MRI and normal or slightly impaired cognitive development. This category includes the most common single variant, laminin-α2 (merosin)-negative congenital muscular dystrophy. The severe congenital muscular dystrophies are accompanied by significant structural brain lesions and mental retardation, and often have a rapidly progressive course. Immunohistochemical staining typically demonstrates reduced but not absent laminin-α2. MILD CONGENITAL MUSCULAR DYSTROPHIES The laminin-α2 (merosin)-positive congenital muscular dystrophies are characterized by normal immunohistochemical staining of laminin-α2 on muscle biopsy. It is rare to have structural brain lesions, and intelligence is normal. Many cases of laminin-α2positive congenital muscular dystrophy are not further classified, but the genetic origins of two distinct congenital muscular dystrophy syndromes, Ullrich disease, and rigid spine syndrome, have been identified. Mutations in the collagen-VI α 1, 2, and 3 genes cause some cases of Ullrich disease. These findings correspond to the deficiencies of collagen VI demonstrated on immunohistochemistry of muscle tissue from affected patients. Collagen VI is widely expressed in all connective tissues and anchors collagen I/III fibrils to basement membranes. In contrast, collagen IV expression is upregulated, perhaps as a compensatory mechanism. Patients with Ullrich disease present in infancy with marked hypotonia, motor delays, and muscle atrophy. The distinguishing clinical features of the disease include distal joint hyperextensibility and proximal joint contractures. Some are able to walk for several years, but all patients eventually are unable to walk. Intelligence is usually normal. Rigid spine syndrome is caused by a mutation in the gene encoding selenoprotein N (SEPN1), which incorporates selenium in the form of selenocysteine. The function of this protein remains obscure, but selenium deficiency is known to cause cardiomyopathy in humans and muscular dystrophy in livestock. Patients affected with rigid spine syndrome generally have a mild degree of weakness, retaining the ability to walk and having little or no elevation in

697

the level of CPK. However, they tend to develop respiratory insufficiency requiring ventilation, perhaps related to selective involvement of the diaphragm. Multiminicore disease has also been linked to mutations in SEPN1. Most children with laminin-α2 (merosin)-positive congenital muscular dystrophy have a mild course and do not have the phenotype of either Ullrich disease or rigid spine syndrome. These patients bear further study; other genetic loci may be identified in the future. MODERATE CONGENITAL MUSCULAR DYSTROPHIES Primary mutations in the laminin-α2 (merosin) gene cause laminin-α2-negative congenital muscular dystrophy. Laminin-α2 is an extracellular matrix protein that is associated with α-dystroglycan. A deficiency of laminin-α2 on immunohistochemistry of muscle tissue does not necessarily imply a primary laminin-α2negative congenital muscular dystrophy—this finding only indicates a protein deficiency, which may be caused by either a mutation in laminin-α2 or may be secondary to a mutation in an associated protein, as occurs in the severe forms of congenital muscular dystrophy and LGMD 2I. Many of the published cases of laminin-α2-negative congenital muscular dystrophy do not include mutation analysis. Thus, laminin-α2-negative congenital muscular dystrophy as a disease entity is in flux. In the classic, severe form of primary laminin-α2-negative congenital muscular dystrophy, affected children present in the first 6 mo of life with hypotonia, muscle weakness, and contractures. The course is moderately severe; most patients never ambulate independently. Serum CPK levels are mildly to moderately elevated. Although these patients have diffuse white matter lesions evident on MRI, gray matter is generally preserved, and intelligence is typically normal, although mental retardation has been reported in some cases. Other less commonly encountered features include seizures, cardiac complications, and neuronal migration defects. MDC1C with mutations in FKRP also spare cognitive function, although the muscular defect is severe and survival past the first decade may only be possible with assisted ventilation. SEVERE CONGENITAL MUSCULAR DYSTROPHIES The three classic severe congenital muscular dystrophies, Fukuyama congenital muscular dystrophy, muscle-eye-brain disease, and Walker–Warburg syndrome, share in common a more severe course than the other congenital muscular dystrophies, brain abnormalities in both white and gray matter, and mental retardation. The genetic origins of all three have been described, and the protein products all localize to the Golgi apparatus, which is consistent with their putative role in protein glycosylation. Protein glycosylation appears to play a role in cell adhesion, growth, and differentiation. Muscle tissue from affected patients has secondary deficiencies in laminin-α2 and α-dystroglycan on immunohistochemistry. Mutations in FCMD, the gene encoding fukutin on 9q31, cause Fukuyama congenital muscular dystrophy, and have also been associated with the Walker–Warburg phenotype. The most common molecular defect is a 3-kb retrotransposal insertion in the 3′ untranslated region of the gene. Some patients also have point mutations in the coding region. Both defects lead to an absence of protein expression in affected patients. The exact function of fukutin is unknown, although a deficiency of highly glycosylated α-dystroglycan has been noted in the muscle tissue of these patients, suggesting that fukutin may play a role in glycosylation. Patients with Fukuyama congenital muscular dystrophy have severe hypotonia and weakness, and almost never walk

698

SECTION VIII / MUSCULOSKELETAL

independently. Cortical and cerebellar dysgenesis (especially micropolygyria), hydrocephalus, and white matter abnormalities are typical. Mental retardation and seizures are common. Patients may survive to early adulthood. Mutations in the gene encoding O-mannose β-1,2-N-acetylglucosaminyltransferase (POMGnT1) have been demonstrated in patients with muscle-eye-brain disease. This enzyme, a Golgi glycosyltransferase, catalyzes the transfer of N-acetylglucosamine from UDP-GlcNAc to O-mannosyl glycoproteins, which is the second step in the biosynthesis of O-mannosylglycan, a component of α-dystroglycan. Children with muscle-eye-brain disease have severe hypotonia and weakness, and most are, thus, unable to walk. The eye manifestations include myopia, glaucoma, optic disk pallor, and retinal hypoplasia. MRI commonly reveals lissencephaly, flattened brainstem, cerebellar hypoplasia, and white matter abnormalities. Mental retardation is universal and epilepsy is common. Patients usually survive into early adulthood. The gene encoding POMT1 has been demonstrated to be mutated in individuals with Walker–Warburg syndrome from at least six different families. A handful of patients with mutations in FCMD and FKRP also fit the traditional clinical picture of Walker–Warburg syndrome. POMT1, like fukutin and POMGnT1, plays a role in glycosylation, as indicated by the absence of glycosylation of α-dystroglycan in the muscle tissue of affected patients. The amount of α-dystroglycan is also reduced in these muscles. POMT1 is ubiquitously expressed, but peak levels are found in fetal brain, skeletal and cardiac muscle, and adult testis, corresponding to several of the organs most affected. Walker–Warburg syndrome is the most severe of all muscular dystrophies, with patients rarely surviving infancy. The mothers of affected patients often have a history of miscarriages. These children are born with severe hypotonia and weakness. Brain malformations include cobblestone lissencephaly, agenesis of the corpus callosum, cerebellar hypoplasia, hydrocephalus, and extensive white matter abnormalities. Buphthalmos, glaucoma, and other eye abnormalities are present, as are testicular defects. Serum CPK levels are moderately elevated.

SYNDROMIC MUSCULAR DYSTROPHIES The syndromic muscular dystrophies do not share genetic or clinical features beyond those common to all the muscular dystrophies. Each one, however, has a distinct clinical presentation, which has traditionally enabled physicians to make the diagnosis before the genetic etiologies were identified. EMERY–DREIFUSS MUSCULAR DYSTROPHY Defects in two distinct genes have been found to cause Emery–Dreifuss muscular dystrophy: EMD, located on Xq28, and LMNA. Mutations in LMNA are inherited in an autosomal-dominant pattern, and also cause LGMD 1B. EMD is ubiquitously expressed, but is most prevalent in skeletal muscle, heart, colon, testis, ovary, and pancreas. The clinical presentation of Emery–Dreifuss muscular dystrophy is striking. The three characteristic findings are muscle weakness and atrophy in a humeroperoneal distribution, heart block, and contractures involving the posterior neck, elbows, and heel cords. The contractures are an early finding. Intelligence is preserved. FACIOSCAPULOHUMERAL MUSCULAR DYSTROPHY The genetics and pathogenesis of facioscapulohumeral muscular dystrophy (FSHD) remain largely a mystery. In patients with FSHD, deletions in a D4Z4 tandem repeat sequence on 4q35 have

been identified. Each D4Z4 sequence is 3.3-kb long and contains two homeobox domains. Affected individuals have fewer than 11 D4Z4 repeats. The length of the deletion correlates with the severity of disease, and is stably inherited. However, this sequence does not contain a known coding region. The most likely mechanism appears to be position effect variegation, in which a number of genes upstream of the D4Z4 region on 4q35 are upregulated in the disease state. In normal individuals, a multiprotein complex consisting of Yin Yang1, high-mobility box 2, and nucleolin binds to an element within D4Z4 and mediates transcriptional repression of those other genes, including adenine nucleotide transporter 1 and FSHD region genes 1 and 2. When too few copies of D4Z4 exist, the baseline repression of these genes is disrupted, leading to their overexpression and potentially explaining the molecular basis of FSHD. Patients with FSHD have a distinct pattern of muscle wasting and weakness. As the disease name suggests, the muscles most affected are in the face, shoulder girdle, and humerus. Pelvic girdle, forearm, and peroneal muscles are also affected later in the course. Intelligence is preserved. Sensorineural hearing loss and retinal telangiectasias may occur, requiring periodic hearing tests and ophthalmological evaluations. The majority of patients remain ambulatory. Variants of FSHD that have been linked to the same genetic locus are scapuloperoneal and scapulohumeral dystrophies. OCULOPHARYGEAL MUSCULAR DYSTROPHY Oculopharyngeal muscular dystrophy is the only muscular dystrophy caused by a triplet repeat expansion. Although myotonic dystrophy is also caused by a triplet repeat expansion and involves muscle wasting and weakness, most muscle samples from these patients do not appear dystrophic on muscle biopsy and it is usually classified as a myotonic disorder. The gene linked to oculopharyngeal muscular dystrophy is poly(A)-binding protein 2, which in control subjects contains a (GCG)6 repeat but in affected patients contains a (GCG)8–13 expansion. The expansions are stably inherited and have low copy numbers, in contrast to other triplet repeat expansion diseases. The typical pattern of inheritance is autosomal-dominant, but the (GCG)7 expansion on one allele may modify (worsen) the phenotype typically caused by a higher copy number on the other allele. The (GCG)7 expansion, when present on both alleles, may behave as an autosomal-recessive disease trait. Patients with oculopharyngeal muscular dystrophy typically present in late adulthood with dysphagia, ptosis, and proximal muscle weakness. The characteristic pathological findings are nuclear filament inclusions in muscle fibers.

CONCLUSION Since the cloning of the gene encoding dystrophin, 20 other genes causing various forms of muscular dystrophy have been discovered, elucidating the genetic origins of many muscle diseases. Although these genes are in some ways heterogeneous, some interesting patterns emerge. Many are physically connected, either directly or indirectly, to dystrophin, including a large sarcolemmal and subsarcolemmal dystrophin-associated protein complex that appears to stabilize the sarcolemmal membrane and connect the internal cytoskeleton with the extracellular matrix. Many of the dystrophinopathies and LGMDs are linked to components of this complex, and ones that are not may have indirect connections yet to be described. Dysferlin is the first of what may be a number of proteins involved in the process of membrane resealing, which is critical for

CHAPTER 70 / MUSCULAR DYSTROPHIES

maintaining the integrity of the muscle membrane. The congenital muscular dystrophies are more severe and generally related to primary or secondary defects in the extracellular matrix, specifically laminin-α2 (merosin) and the glycosylation of α-dystroglycan. The genotype–phenotype correlations raise interesting questions for further investigation. Several genes, LMNA, DYSF, and FKRP, cause markedly different phenotypes in different patients, raising the possibility that modifier genes may help determine the actual clinical presentation. The molecular basis of FSHD may be an extreme example of this. Conversely, Emery–Dreifuss muscular dystrophy may be caused by mutations on either of two genes, which are on entirely different chromosomes. Further study of the pathogenesis of these diseases may yield insights into the mechanisms of transcription regulation, as well as potential therapeutic approaches.

SELECTED REFERENCES Balci B, Uyanik G, Dincer P, et al. An autosomal recessive limb girdle muscular dystrophy (LGMD2) with mild mental retardation is allelic to Walker-Warburg syndrome (WWS) caused by a mutation in the POMT1 gene. Neuromuscul Disord 2005;15:172–275. Bansal D, Miyake K, Vogel SS, et al. Defective membrane repair in dysferlin-deficient muscular dystrophy. Nature 2003;423:168–172. Bashir R, Britton S, Strachan T, et al. A gene related to Caenorhabditis elegans spermatogenesis factor fer-1 is mutated in limb-girdle muscular dystrophy type 2B. Nat Genet 1998;20:37–42. Beltrán-Valero de Bernabé D, Currier S, Steinbrecher A, et al. Mutations in the O-mannosyltransferase gene POMT1 give rise to the severe neuronal migration disorder Walker-Warburg syndrome. Am J Hum Genet 2002;71:1033–1043. Bione S, Maestrini E, Rivella S, et al. Identification of a novel X-linked gene responsible for Emery-Dreifuss muscular dystrophy. Nat Genet 1994;8:323–327. Bonne G, Di Barletta MR, Varnous S, et al. Mutations in the gene encoding lamin A/C cause autosomal dominant Emery-Dreifuss muscular dystrophy. Nat Genet 1999;21:285–288. Brais B, Bouchard JP, Xie YG, et al. Short GCG expansions in the PABP2 gene cause oculopharyngeal muscular dystrophy. Nat Genet 1998;18:164–167. Brockington M, Yuva Y, Prandini P, et al. Mutations in the fukutin-related protein gene (FKRP) identify limb girdle muscular dystrophy 2I as a milder allelic variant of congenital muscular dystrophy MDC1C. Hum Mol Genet 2001;10:2851–2859. Burghes AH, Logan C, Hu X, Belfall B, Worton RG, Ray PN. A cDNA clone from the Duchenne/Becker muscular dystrophy gene. Nature 1987;328:434–437. Bushby KMD, Beckmann JS. The 105th ENMC sponsored workshop: Pathogenesis in the non-sarcoglycan limb-girdle muscular dystrophies, Naarden, 12–14 April 2002. Neuromuscul Disord 2003;13:80–90. Darras BT, Menache CC, Kunkel LM. Dystrophinopathies. In: Jones HR Jr, De Vivo DC, Darras BT, eds. Neuromuscular Disorders of Infancy, Childhood, and Adolescence: A Clinician’s Approach. Boston: Butterworth-Heinemann, 2003; pp. 649–699. De Sandre-Giovannoli A, Bernard R, Cau P, et al. Lamin A truncation in Hutchinson-Gilford progeria. Science 2003;300:2055–2058. Demir E, Sabatelli P, Allamand V, et al. Mutations in COL6A3 cause severe and mild phenotypes of Ullrich congenital muscular dystrophy. Am J Hum Genet 2002;70:1446–1458. Eriksson M, Brown WT, Gordon LB, et al. Recurrent de novo point mutations in lamin A cause Hutchinson-Gilford progeria syndrome. Nature 2003;423:293–298.

699

Feit H, Silbergleit A, Schneider LB, et al. Vocal cord and pharyngeal weakness with autosomal dominant distal myopathy: clinical description and gene localization to 5q31. Am J Hum Genet 1998;63:1732–1742. Fisher J, Upadhyaya M. Molecular genetics of facioscapulohumeral muscular dystrophy (FSHD). Neuromuscul Disord 1997;7:55–62. Frosk P, Weiler T, Nylen E, et al. Limb-girdle muscular dystrophy type 2H associated with mutation in TRIM32, a putative E3-ubiquitin-ligase gene. Am J Hum Genet 2002;70:663–672. Gabellini D, Green MR, Tupler R. Inappropriate gene activation in FSHD: a repressor complex binds a chromosomal repeat deleted in dystrophic muscle. Cell 2002;110:339–348. Hauser MA, Horrigan SK, Salmikangas P, et al. Myotilin is mutated in limb girdle muscular dystrophy 1A. Hum Mol Genet 2000;9:2141–2147. Helbling-Leclerc A, Zhang X, Topaloglu H, et al. Mutations in the laminin α2-chain gene (LAMA2) cause merosin-deficient congenital muscular dystrophy. Nat Genet 1995;11:216–218. Higuchi I, Shiraishi T, Hashiguchi T, et al. Frameshift mutation in the collagen VI gene causes Ullrich’s disease. Ann Neurol 2001;50:261–265. Jones KJ, Morgan G, Johnston H, et al. The expanding phenotype of laminin-α2 chain (merosin) abnormalities: case series and review. J Med Genet 2001;38:649–657. Kobayashi K, Nakahori Y, Miyake M, et al. An ancient retrotransposal insertion causes Fukuyama-type congenital muscular dystrophy. Nature 1998;394:388–392. Lim LE, Campbell KP. The sarcoglycan complex in limb-girdle muscular dystrophy. Curr Opin Neurol 1998;11:443–452. Liu J, Aoki M, Illa I, et al. Dysferlin, a novel skeletal muscle gene, is mutated in Miyoshi myopathy and limb girdle muscular dystrophy. Nat Genet 1998;20:31–36. Messina DN, Speer MC, Percak-Vance MA, McNally EM. Linkage of familial dilated cardiomyopathy with conduction defect and muscular dystrophy to chromosome 6q23. Am J Hum Genet 1997;61:909–917. Minetti C, Sotgia F, Bruno C, et al. Mutations in the caveolin-3 gene cause autosomal dominant limb-girdle muscular dystrophy. Nat Genet 1998;18:365–368. Moghadaszadeh B, Petit N, Jaillaird C, et al. Mutations in SEPN1 cause congenital muscular dystrophy with spinal rigidity and restrictive respiratory syndrome. Nat Genet 2001;29:17, 18. Monaco AP, Neve RL, Colletti-Feener C, Bertelson CJ, Kurnit DM, Kunkel LM. Isolation of candidate cDNAs for portions of the Duchenne muscular dystrophy gene. Nature 1986;323:646–650. Moreira ES, Wiltshire TJ, Faulkner G, et al. Limb-girdle muscular dystrophy type 2G is caused by mutations in the gene encoding the sarcomeric protein telethonin. Nat Genet 2000;24:163–166. Muntoni F, Guicheney P. 85th ENMC international workshop on congenital muscular dystrophy: 6th international CMD workshop: 1st workshop of the myo-cluster project ‘GENRE’: 27–28 October 2000, Naarden, the Netherlands. Neuromuscul Disord 2002;12:69–78. Lennon NJ, Kho A, Backsai BJ, et al. Dysferlin interacts with annexins A1 and A2 and mediates sarcolemmal wound-healing. J Biol Chem 2003;278:50,466–50,473. Richard I, Broux O, Allamand V, et al. Mutations in the proteolytic enzyme calpain 3 cause limb-girdle muscular dystrophy type 2A. Cell 1995;81:27–40. Speer MC, Vance JM, Grubber JM, et al. Identification of a new autosomal dominant limb-girdle muscular dystrophy on chromosome 7. Am J Hum Genet 1999;64:556–562. Yoshida A, Kobayashi K, Manya H, et al. Muscular dystrophy and neuronal migration disorder caused by mutations in a glycosyltransferase, POMGnT1. Dev Cell 2001;1:717–724. Zatz M, Vainzof M, Passos-Bueno MR. Limb-girdle muscular dystrophy: One gene with different phenotypes, one phenotype with different genes. Curr Opin Neurol 2000;13:511–517.

71 Rhabdomyosarcomas STEPHEN J. TAPSCOTT SUMMARY Significant progress has been made in understanding the molecular biology of rhabdomyosarcomas and in improving therapeutic regimens for the treatment of these cancers. Alveolar rhabdomyosarcomas are associated with translocations that create a novel fusion protein of either PAX3 or PAX7 with the FKHR transcription factors, whereas, embryonoal rhabdomayosarcomas are associated with a loss of heterozygosity at 11p15.5. In addition, animal models suggest that the combination of factors that induce myogenesis with an abrogation of tumor suppressors such as p53 and Rb can lead to the formation of rhabdomyosarcomas. Historically, treatment stratification has been based on histological criteria, however, the increasing knowledge of distinct molecular phenotypes of rhabdomyosarcomas will likely inform treatment strategies in the future. Key Words: Alveolar; embryonal; FKHR; muscle differentiation; PAX3; PAX7; rhabdomyosarcoma; tumor suppressors.

INTRODUCTION Rhabdomyosarcomas are the most common soft tissue sarcomas in childhood, yet they are still a relatively rare tumor, with an incidence in the United States of 1.3–4.5/million children/yr. The histological classification of rhabdomyosarcoma has historically relied on the expression of genes associated with skeletal muscle in some of the tumor cells, giving a subset of cells the appearance of skeletal muscle cells or myoblasts. Several histological subcategories of rhabdomyosarcomas have also been recognized. Embryonal rhabdomyosarcomas occur most frequently during the first 3 yr after birth and account for about 50–60% of rhabdomyosarcomas. Additional distinctions can be made within the group of embryonal rhabdomyosarcomas, for example, there are the botyroid and spindle cell variants. Alveolar rhabdomyosarcomas are distinguished from embryonal rhabdomyosarcomas histologically by characteristic open spaces in pathology sections, reminiscent of alveoli in the lungs. Alveolar rhabdomyosarcomas account for approx 20% of rhabdomyosarcomas and have a bimodal incidence distribution with peaks at approx ages 3 and 15 yr. A third group of pleomorphic or primitive rhabdomyosarcomas is not easily distinguished from other small round cell tumors of childhood, such as neuroblastoma, Ewing’s sarcoma, primitive neuroectodermal tumors, and non-Hodgkin’s lymphoma based on From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

standard histology. For these primitive tumors, classification as rhabdomyosarcomas can be confirmed using antibodies or other molecular probes to genes characteristically expressed in skeletal muscle, such as desmin, or muscle creatine kinase.

MOLECULAR GENETICS OF ALVEOLAR RHABDOMYOSARCOMAS Cytogenetic studies identified multiple abnormalities in cells from rhabdomyosarcomas; however, alveolar rhabdomyosarcomas had a characteristic translocation between chromosome 2 and 13, t(2;13)(q35;q14). Fine mapping of the breakpoint identified PAX3 as a candidate gene on chromosome 2, leading to the demonstration that the rearrangement occurred within the PAX3 gene. Ultimately a novel chimeric transcript was identified, encoding the amino-terminal portion of the PAX3 protein from chromosome 2 and the carboxy-terminal portion of a gene on chromosome 13, with homology to the fork head family of genes. PAX3 is a transcription factor that is expressed in the neural tube, neural crest derivatives, and in the muscle precursor cells that migrate from the somites into the limb bud. Dominant or semidominant mutations of PAX3 in mice, the splotch mutant, and humans, the Waartenberg syndrome, are noted for defects in neural crest and neural tube development. Embryos homozygous for some alleles of splotch have deficient limb muscle formation that can be attributed to the inadequate generation or migration of limb muscle precursor cells. This may partly be the basis for limb deformities in human variations of the Waartenberg syndrome that are homozygous for PAX3 mutations. The t(2;13) in alveolar rhabdomyosarcomas creates a novel gene with the transcription start site and 5′ coding region of the PAX3 gene and the 3′ region of the fork head homolog rhabdomyosarcoma (FKHR) gene, a member of the fork head family of transcription factors, so-called because the prototype gene of the family was described as a mutation in Drosophila melanogaster that produced two spiked head structures (see Chapter 3). Additional vertebrate members of the fork head family include the hapatocyte nuclear factors (HNF-3α, HNF-3β, HNF-3γ) and brain factor-1. The fusion protein includes two DNA-binding domains in the carboxy-terminal half of the PAX3 protein, a paired box DNA-binding domain and a homeodomain, but disrupts the DNA-binding region of the FKHR protein. The carboxy-terminal portion of FKHR maintained in the fusion protein is rich in prolines and acidic amino acids, attributes consistent with an acidic activation domain, and transfection experiments demonstrate that the fusion protein is more potent than PAX3

700

CHAPTER 71 / RHABDOMYOSARCOMAS

as a transcriptional activator of reporter constructs driven by PAX3 binding sites. Further, DNA-binding studies show that the fusion protein binds to PAX3 binding sites, but that it has a mildly altered DNA-binding affinity compared to PAX3. Therefore, the PAX3FKHR fusion protein may have oncogenic activity based on enhanced activation of genes regulated by PAX3 or through altered binding affinity to specific DNA sequences. Although PAX3 is apparently expressed in the early migrating muscle progenitor cells, its role in development is not fully understood. Genetic experiments in mice indicate that PAX3 plays a critical role in specifying cells to become myoblasts, in part by regulating the expression of the myogenic basic helix–loop–helix (bHLH) protein MyoD. Although PAX3 plays a positive role in specifying myoblast fate, PAX3 expression inhibits myogenesis in cultured muscle cell lines and inhibits the ability of MyoD to activate the myogenic program in fibroblasts. The PAX3-FKHR fusion protein is a more potent inhibitor of myogenesis and MyoD activity in this system. Based on these observations and the developmental expression of PAX3 in the migrating muscle precursor cells, but not in skeletal muscle, it is possible that the normal function of PAX3 is to initiate and maintain an early phase of myogenesis and the fusion protein either enhances this activity or loses an element necessary to suppress this activity, resulting in an arrested state of development. Although much less frequent than t(2;13), a t(1;13)(p36;q14) translocation has been described in some alveolar rhabdomyosarcomas. This translocation creates a fusion protein between PAX7 on chromosome 1 and FKHR that is structurally similar to the PAX3-FKHR fusion protein. PAX7 is also expressed in muscle precursor cells and is capable of binding to PAX3 sites either as a homodimer or as a heterodimer with PAX3. Genetic studies in mice revealed that PAX7 is necessary for the formation of muscle satellite cells, the cell population that regenerates skeletal muscle in the adult. Therefore, it appears that PAX3 and PAX7 both have central roles in specifying cells to become skeletal muscle. It is likely that each of the fusion proteins associated with alveolar rhabdomyosarcomas, PAX3-FKHR, and PAX7-FKHR, maintains some ability to initiate the skeletal muscle program but cannot sustain normal differentiation. The role of the FKHR protein is largely unknown, but it promotes myoblast fusion during terminal differentiation and dominant negative mutants of FKHR can prevent fusion. Therefore, the truncated FKHR fusion protein might block specific aspects of the program of terminal differentiation.

MOLECULAR GENETICS OF EMBRYONAL RHABDOMYOSARCOMAS Most rhabdomyosarcomas are sporadic; however, the appearance of rhabdomyosarcomas in families with the Li-Fraumeni syndrome or the Beckwith-Wiedemann syndrome has provided valuable insights into the genetics of this tumor (see Chapter 7). Beckwith-Wiedemann syndrome (see Chapter 116) is associated with growth excess and a high incidence of tumors, including Wilms’ tumor, hepatoblastoma, adrenal carcinoma, and rhabdomyosarcoma. Because of the association of Wilms’ tumor with a locus on chromosome 11p13, this region was analyzed for loss of heterozygosity (LOH) in both rhabdomyosarcomas and in the BeckwithWiedemann syndrome. LOH of chromosome 11 was detected in a large portion of rhabdomyosarcomas, but the common region of LOH mapped to 11p15.5 similar to the mapping of the region of LOH for the Beckwith-Wiedemann syndrome. A preferential

701

retention of the paternal allele in both embryonal rhabdomyosarcomas and in the Beckwith-Wiedemann syndrome has led to the suggestion that LOH alters the level of gene expression maintained by an imprinted allele, either decreasing expression of a tumor suppressor gene by loss of the nonimprinted allele or increasing expression of a growth-promoting gene by duplication of the nonimprinted allele. The presence of a tumor growth suppressor gene at 11p15 has been supported by transfer of subchromosomal fragments to rhabdomyosarcoma cells. Fragments of chromosome 11 that contain 11p15 suppress growth and xenograft tumor formation in embryonal rhabdomyosarcomas, although similar experiments have not been performed in alveolar rhabdomyosarcomas, nor has parental origin of the suppressing allele been analyzed. The gene relevant to rhabdomyosarcoma formation at 11p15 has not been clearly identified, although some candidates have been studied. Insulin-like growth factor II (IGF-2) is a maternally imprinted gene that maps to 11p15, transcriptionally silent on the maternal allele and expressed from the paternal allele. Elevated levels of IGF-2 resulting from paternal disomy associated with LOH may give a growth advantage to cells and have been postulated as the cause of the somatic hypertrophy in the BeckwithWiedemann syndrome. IGF-2 has been shown to act as an autocrine growth factor in some rhabdomyosarcomas and loss of imprinting, i.e., expression of IGF-2 from both alleles has been demonstrated in alveolar rhabdomyosarcomas and embryonal rhabdomyosarcomas that do not have LOH at 11p15. The growthpromoting activity of increased IGF-2 levels may play a critical role in the generation of rhabdomyosarcomas; however, this model would not explain the somatic cell genetic experiments that implicate a tumor suppressor gene at 11p15. A second candidate gene that maps to 11p15, H19, is unusual because the H19 transcript does not contain a conserved open reading frame and may function as a regulatory polyadenylated RNA. H19 is imprinted in the opposite manner to IGF-2, expressed from the maternal allele and silent on the paternal allele; therefore, loss of the maternal allele decreases H19 expression. Transfection of H19 into rhabdomyosarcoma cell lines suppresses both growth and tumorigenesis, making it a candidate for the 11p15 tumor suppressor. Implicating imprinting as a critical factor in the LOH at 11p15 means that the transcription of numerous genes in the region might contribute to the transformed phenotype, and both H19 and IGF-2 may contribute to the generation of rhabdomyosarcomas. Many additional genes have been mapped to the 11p15.5 region, including the cyclin-dependent kinase inhibitor p57KIP2, the nucleosome assembly protein hNAP2, and the retinoblastomaassociated protein RbAp48. Mutations of the p57 gene have been reported in Beckwith-Wiedemann syndrome and cause developmental abnormalities in mice that are similar to BeckwithWiedemann syndrome, indicating that p57 has a critical role in this syndrome. It remains to be determined whether altered expression of p57 or other genes in this region contributes to the generation of rhabdomyosarcomas.

RHABDOMYOSARCOMAS EXPRESS MYOD OR RELATED MYOGENIC FACTORS Expression of the myogenic regulatory protein, MyoD, or one of the related members of this group of myogenic bHLH proteins, which consists of MyoD, Myf5, Myogenin, and MRF4, can be

702

SECTION VIII / MUSCULOSKELETAL

used to categorize less differentiated tumors as rhabdomyosarcomas, because almost all rhabdomyosarcomas express one or more of this group of proteins. The myogenic bHLH proteins are transcription factors that orchestrate the expression of genes that characterize skeletal muscle. Gene knockout experiments demonstrate that these genes are necessary for skeletal myogenesis, and the ability to convert nonmyogenic cells to skeletal muscle by expressing one of the myogenic bHLH proteins indicates that in many cell types, they are sufficient to mediate differentiation. In addition to activating expression of muscle structural genes, MyoD has been shown to mediate withdrawal from the cell cycle and induction of p21 expression. The role of these proteins in normal development is discussed in Chapter 92. Given the ability of the myogenic proteins to activate transcription of skeletal muscle genes, the expression of MyoD in rhabdomyosarcomas can account for the low level of muscle differentiation seen in the tumor. In the tumors that have been analyzed, however, most of the cells in the tumor express MyoD, but do not terminally differentiate into postmitotic muscle cells, indicating that the program of muscle differentiation mediated by the MyoD family of transcription factors has been partially blocked. Analysis of the myogenic bHLH proteins in several tumors has demonstrated that they are not mutant, yet they are inefficient activators of muscle gene expression. Several molecular mechanisms have been identified that can prevent MyoD and related myogenic bHLH transcription factors from functioning. Relevant to the mutations that can cause rhabdomyosarcomas, normal Rb and p53 are necessary for muscle gene expression by MyoD, and amplified MDM2 can prevent MyoD transcriptional activity. RB, P53, AND ONCOGENES Just as the association of rhabdomyosarcomas with the Beckwith-Wiedemann syndrome focused attention on 11p15, the appearance of rhabdomyosarcomas in families with Li-Fraumeni syndrome suggested that p53 inactivation may be important for tumor formation. Analysis of p53 abnormalities, without respect to alveolar or embryonal histological types, indicated that approximately half of the tumors had abrogated normal p53 function. When histological subtype is considered, it appears that the frequency of p53 mutations is higher in the embryonal rhabdomyosarcomas. Differences between embryonal and alveolar rhabdomyosarcomas are further suggested by a higher frequency of N-myc amplification in alveolar rhabdomyosarcomas, whereas mutation in N-ras and K-ras appear to be more frequent in embryonal rhabdomyosarcomas. Mouse models have indicated several different regulatory pathways that can contribute to the formation of rhabdomyosarcomas. Mice with a mutation that inactivates the p16INK4a and the p14ARF transcripts, two different transcripts from the same locus that, respectively, regulate pRB and p53 function, mainly develop lymphomas and fibrosarcomas. However, when combined with a transgene that expresses hepatocyte growth factor/scatter factor, the ligand for the c-MET receptor expressed on satellite cells and myoblasts, the INK4a/ARF mutation induces a high rate of rhabdomyosarcomas in mice. In this model, blocking both p53 and pRb activity in the presence of a constitutive myogenic growth factor was sufficient to generate rhabdomyosarcomas. This is similar to the finding that amplified MDM2 can contribute to rhabdomysarcomas by inhibiting the activity of both p53 and Rb when expressed in myoblasts. There are likely many different molecular pathways that can contribute to rhabdomyosarcoma

formation, as evidenced by the formation of rhabdomysarcomas in mice deficient in the Patched receptor that inhibits signaling by Sonic Hedgehog. Sonic Hedgehog is necessary to specify skeletal muscle cells in the developing embryo and acts, at least in part, by regulating the expression of the myogenic bHLH transcription factor Myf5, similar to the role of PAX3 in regulating the expression of the related myogenic transcription factor MyoD. One theme that emerges from these different models of rhabdomyosarcomas is the constitutive or aberrant activity of factors that induce myogenesis coupled with conditions that prevent normal terminal differentiation.

CONCLUSIONS Therapy of rhabdomyosarcomas combines chemotherapy, radiation therapy, and surgery. The Intergroup Rhabdomyosarcoma Studies (IRS) utilized a risk-based design that increased therapeutic intensity in groups with marginally poorer prognosis. In addition to tumor staging based on invasiveness, involvement of lymph nodes, metastasis, and degree of surgical excision, tumor histology is a significant variable in predicting response to treatment. In IRS-II, the 5-yr survival rate was 95% for the favorable category of botyroid, 67% for embryonal, and approx 50% for alveolar and undifferentiated. IRS-III includes therapy intensification of the groups with the less favorable prognosis and has achieved survival rates for alveolar tumors comparable to embryonal ones. One challenge remaining is to determine if survival can be improved and therapy-related complications minimized by using the available molecular markers to further refine risk-based therapy. An analysis of patients with metastatic alveolar rhabdomyosarcoma enrolled in the IRS-IV study revealed a more favorable outcome for those with a PAX7-FKHR translocation than those with a PAX3-FKHR translocation, whereas the type of translocation did not appear related to outcome in nonmetastatic disease. The availability of molecular methods to accurately identify fusion transcripts, gene amplifications, and gene mutations might permit the identification of other factors associated with response to therapy and survival. Ideally further understanding of the molecular biology of rhabdomyosarcomas eventually will lead to rationally designed therapies. Perhaps the development of drugs that inhibit PAX3 expression or agents that degrade the PAX3-FKHR fusion transcript would be useful adjuvants to therapy.

SELECTED REFERENCES Barr FG, Galili N, Holick J, Biegel JA, Rovera G, Emanuel BS. Rearrangement of the PAX3 paired box gene in paediatric solid tumor alveolar rhabdomyosarcoma. Nat Genet 1993;3:113–117. Best LG, Hoekstra RE. Wiedemann-Beckwith syndrome: autosomal dominant inheritance in a family. Am J Med Genet 1981;9:291–299. Bober E, Franz T, Arnold HH, Gruss P, Tremblay P. Pax-3 is required for the development of limb muscles: a possible role for the migration of dermomyotomal muscle progenitor cells. Development 1994;120: 603–612. Clark J, Rocques PJ, Braun T, et al. Expression of members of the myf gene family in human rhabdomyosarcomas. Br J Cancer 1991;64: 1039–1042. Crist W, Gehan EA, Ragab AH, et al. The third intergroup rhabdomyosarcoma study. J Clin Oncol 1995;13:610–630. Davis RJ, D’Cruz CM, Lovell MA, Biegel JA, Barr FG. Fusion of PAX7 to FKHR by the variant t(1;13)(p36;q14) translocation in alveolar rhabdomyosarcoma. Cancer Res 1994;54:2869–2872.

CHAPTER 71 / RHABDOMYOSARCOMAS

Douglas EC, Valentine M, Etcubanas E, et al. A specific chromosomal abnormality in rhabdomyosarcomas. Cytogenet Cell Genet 1987;45: 148–155. Epstein JA, Lam P, Jepeal L, Maas RL, Shapiro DN. Pax3 inhibits myogenic differentiation of cultured myoblast cells. J Biol Chem 1995; 270:11,719–11,722. Felix CA, Kappel CC, Mitsudomi T, et al. Frequency and diversity of p53 mutations in childhood rhabdomyosarcoma. Cancer Res 1992;52: 2243–2247. Fredericks WJ, Galili N, Mukhopadhyay S, et al. The PAX3-FKHR fusion protein created by the t(2;13) translocation in alveolar rhabdomyosarcomas is a more potent transcriptional activator that PAX3. Mol Cell Biol 1995;15:1522–1535. Galili N, Davis RJ, Fredericks WJ, et al. Fusion of a fork head domain gene to PAX3 in the solid tumor alveolar rhabdomyosarcoma. Nat Genet 1993;5:230–235. Hao Y, Crenshaw T, Moulton T, Newcomb E, Tycko B. Tumour-suppressor activity of H19 RNA. Nature 1993;365:764–767. Koi M, Johnson LA, Kalikin LM, Little PF, Nagamura Y, Feinberg AP. Tumor cell growth arrest caused by subchromosomal transferable DNA fragments from chromosome 11. Science 1993;260:361–364. Koufos A, Hansen MF, Copeland NG, Jenkins NA, Lampkin BC, Cavenee WK. Loss of heterozygosity in three embryonal tumours suggests a common pathogenetic mechanism. Nature 1985;316: 330–334. Loh WE, Scrable HJ, Livanos E, et al. Human chromosome 11 contains two different growth suppressor genes for embryonal rhabdomyosarcoma. Proc Natl Acad Sci USA 1992;89:1755–1759.

703

Mulligan LM, Matlashewski GJ, Scrable HJ, Cavenee WK. Mechanisms of p53 loss in human sarcomas. Proc Natl Acad Sci USA 1990;87:5863–5867. Scrable H, CaveneeW, Ghavimi F, Lovell M, Morgan K, Sapienza C. A model for embryonal rhabdomyosarcoma tumorigenesis that involves genome imprinting. Proc Natl Acad Sci USA 1989;86:7480–7484. Scrable H, Witte D, Shimada H, et al. Molecular differential pathology of rhabdomyosarcoma. Genes Chromosomes Cancer 1989;1:23–35. Seale P, Sabourin LA, Girgis-Gabardo A, Mansouri A, Gruss P, Rudnicki MA. Pax7 is required for the specification of myogenic satellite cells. Cell 2000;102:777–786. Sharp R, Recia JA, Jhappan C, et al. Synergism between INK4a/ARF inactivation and aberrant HGF/SF signaling in rhabdomyosarcomagenesis. Nat Med 2002;8:1276–1280. Sorensen PHB, Lynch JC, Qualman SJ, et al. PAX3-FKHR and PAX7FKHR gene fusions are prognostic indicators in alveolar rhabdomyosarcoma: a report from the Children’s Oncology Group. J Clin Oncol 2002;20:2672–2679. Stratton MR, Fisher C, Gusterson BA, Cooper CS. Detection of point mutations in N-ras and K-ras genes of human embryonal rhabdomyosarcomas using oligonucleotide proges and the polymerase chain reaction. Cancer Res 1989;49:6324–6327. Tapscott SJ, Thayer MJ, Weintraub H. Deficiency in rhabdomyosarcomas of a factor required for MyoD activity and myogenesis. Science 1993;259:1450–1453. Zhan S, Shapiro DN, Helman LJ. Activation of an imprinted allele of the insulin-like growth factor II gene implicated in rhabdomyosarcoma. J Clin Invest 1994;94:445–448.

IX

ONCOLOGY SECTION EDITOR:

W. STRATFORD MAY, JR.

Abbreviations IX. ONCOLOGY 17-AAG 2-5A ABL ADH ALCL ALK ALL AMACR AML α-MSH AO AOA AP-2 APaf-1 APC APL APO-3 AR ARF ART AT ATM ATRA BCL-2 BCR bFGF BH bHLH BID BIR BL BM BMSC BPH CAD C-ALCL CAMs CBF CCND1 CD CDK c-FLIP CIITA CI CIMP CIN

17-Allyamino-17-demethoxygeldanamycin 2′-5′-oligoadenylate Abelson atypical ductal hyperplasia anaplastic large-cell lymphoma anaplastic lymphoma kinase acute lymphoblastic leukemia alpha-methylacyl coenzyme A racemase acute myeloid leukemia α-melanocyte stimulating hormone anaplastic oligodendroglioma mixed anaplastic oligoastrocytoma activator protein-2 apoptosis protease activating factor adenomatous polyposis coli acute promyelocytic leukemia apoptosis antigen 3 androgen receptor alternative reading frame antiretroviral therapy adenine-thymine ataxia-telangectasia mutated all-trans-retinoic acid B-cell leukemia/lymphoma gene breakpoint cluster region basic fibroblast growth factor Bcl2 homology basic helix loop helix BH3 interacting domain death agonist baculovirus inhibitor of apoptosis repeat Burkitt’s lymphoma/leukemia bone marrow bone marrow stromal cell benign prostatic hyperplasia caspase-activated DNase cutaneous anaplastic large-cell lymphoma cell adhesion molecules core binding factor cyclin D-1 cluster designation cyclin-dependent kinase c-FLICE-like inhibitory protein class II transactivator chromosomal instability CpG island methylator phenotype cervical intraepithelial neoplasia

CIS CKI CKI CLL CML CMP CTL Cyt c DAPK DCIS Dex DISC DLCL DNMT DR EB EBNA EBV ECM EGF EGFR ER ERE ERK ET FAB FAK FAP FAS.L FGF FGF-2 FHIT FISH FLD FN FPD/AML FTI G0 G2 GBM GC GM-CSF GMP GPCR GRO GRP GSK3β

705

carcinoma in situ CDK-inhibitor cyclin-dependent kinase inhibitor chronic lymphocytic leukemia chronic myelogenous leukemia common myeloid progenitor cytolytic T-cell cytochrome c death-associated protein kinase ductal carcinoma in situ dexamethasone death-induced signaling complex diffuse large-cell lymphoma DNA methyltransferases death receptor end-binding protein Epstein-Barr virus nuclear antigen Epstein-Barr virus extracellular matrix epidermal growth factor epidermal growth factor receptor estrogen receptor estrogen response element extracellular signal-regulated kinase endothelin French–American–British focal adhesion kinase familial adenomatous polyposis Fas ligand fibroblast growth factor fibroblast growth factor-2 fragile histidine triad fluorescence in situ hybridization flexible loop domain fibronectin familial platelet defect with propensity to develop AML farnesyl transferase inhibitor non-proliferative state Gap2 glioblastoma multiforme germinal center granulocyte-macrophage colony-stimulating factor granulocyte-monocyte progenitor G protein-coupled receptor growth-regulated protein gastrin-releasing peptide glycogen synthase-3β kinase

706 GST H HAT HCV HDAC HGF HHV-8 HIF HIV-1 HNPCC HPC HPC1 HPV HSC Hsp90 HTLV IAP ICC ICE IFN IGF IGF-1 IGH IgH IgL IKK IL ITAM ITD Jak JMML JPS k-NN KS LC LCIS LDH-A Lef LMO LMP LOH LTR M MALT MAPK MC1-R MCHR MCL MCR MDR MDS MGSA MGUS MHC II Min MIP MIP-1α Mitf MLL MM

SECTION IX / ONCOLOGY

glutathione S-transferase helicobacter histone acetyl transferase hepatitis C virus histone deacetylase hepatocyte growth factor human herpesvirus 8 hypoxia-inducible factor-1 HIV type 1 hereditary nonpolyposis colorectal cancers hereditary prostate cancer hereditary prostate cancer 1 human papilloma virus hematopoietic stem cell heat shock protein 90 human T-cell leukemia viruses inhibitors of apoptosis invasive cervical carcinoma interleukin-Iβ-converting enzyme interferon insulin-like growth factor insulin-like growth factor-1 immunoglobulin H immunoglobulin heavy chain immunoglobulin light chain IκB kinase interleukin immunoreceptor-tyrosine-based activation motif internal tandem duplications Janus tyrosine kinase juvenile myelomonocytic leukemia juvenile polyposis syndrome nearest neighbor prediction model Kaposi’s sarcoma light chain lobular carcinoma in situ lactate dehydrogenase A lymphoid enhancer factor cysteine-rich LIM-domain only latent membrane protein loss of heterozygosity long terminal repeat mitosis (when used in cell cycle discussions) mucosa-associated lymphoid tissue mitogen-activated protein kinase melacortin-1 receptor melanocyte-stimulating hormone receptor mantle cell lymphoma mutation cluster region multidrug resistance myelodysplastic syndrome melanocyte growth stimulatory activity monoclonal gammopathy of unknown significance major histocompatibility class II multiple intestinal neoplasia macrophage inflammatory protein macrophage inflammatory protein-1 α microphthalmia-associated transcription factor mixed lineage leukemia multiple myeloma

MMP MMR MOA MRD MSI MSR1 MTC NE NF NFAT NFκB NHL NK NNB NNK NPC NPM NS5A NSCLC OMM OPG PAH PAI-1 PAR-1 PCD PCL PDGF PDK Ph PI3K PIA PIN PJS PKB PKC PLC-γ PML PR pRB PSA PTEN PTHrP PTK PTP RANKL RAR RARα Rb RGP RNASEL ROS RT RTK S SCC SCF SCL SCLC

matrix metalloproteinase mismatch repair mixed oligoastrocytoma minimal residual disease microsatellite instability macrophage scavenger receptor 1 gene maternal-to-child neuroendocrine nuclear factor nuclear factor of activated T-cells nuclear factor κB non-Hodgkin’s lymphoma natural killer non neoplastic brain 4-(methylnitrosanino)-1-(3-pyridyl)-1-butanone nasopharyngeal carcinoma nucleophosmin nonstructural protein 5A non-small cell lung cancer outer mitochondrial membrane osteoprotegerin polycyclic aromatic hydrocarbon type 1 plasminogen inhibitor protease-activated receptor 1 programmed cell death plasma cell leukemia platelet-derived growth factor phosphoinsitol-dependent protein kinase Philadelphia chromosome phosphoinositide 3-kinase proliferative inflammatory atrophy prostatic intraepithelial neoplasia Peutz-Jeghers syndrome protein kinase B protein kinase C phospholipase C-γ promyelocytic leukemia progesterone receptor retinoblastoma gene product prostate-specific antigen phosphatase and tensin homolog deleted on chromosome 10 parathyroid hormone-related protein protein tyrosine kinases protein tyrosine phosphatase receptor activator of nuclear factor-κB ligand retinoic acid receptor retinoic acid receptor α retinoblastoma protein or susceptibility gene radial growth phase RNase L gene reactive oxygen species reverse transcriptase receptor tyrosine kinase synthesis (when used in cell-cycle discussions) squamous cell carcinoma stem cell factor stem cell leukemic protein small-cell lung cancer

707

CHAPTER 72 / APOPTOSIS

SERM SH2 sIg siRNA SL-IC SLL SRC Stat T-ALL Tcf TCR TDLU TDT TGFα TGFβ TGFβRII TIF2

selective ER modulator src-homology domain 2 surface immunoglobulins small-interfering RNA SCID leukemia-initiating cells small lymphocytic lymphoma w/e signal transducer and activators of transcription T-cell acute lymphoblastic leukemia T-cell factor T-cell receptors terminal duct lobular unit terminal deoxynucleotidyl-transferase transforming growth factor α transforming growth factor-β transforming growth factor-β receptor II transcriptional intermediary factor 2

TIMP-1 T-LBL TM TNF TNFR TRAIL TRRAP TSG uPA uPAR UPR UV VE VEGF VGP WHO

tissue inhibitor of metallproteinases T-lymphoblastic lymphoma transmembrane tumor necrosis factor tumor necrosis factor receptor TNF-related apoptosis-inducing ligand transactivating/transformation domain-associated protein tumor suppressor gene urokinase-type 1 plasminogen activator urokinase-type plasminogen activator receptor unfolded protein response ultraviolet vascular endothelial vascular endothelial growth factor vertical growth phase World Health Organization

SOLID TUMORS

72 Apoptosis W. STRATFORD MAY, JR. AND XINGMING DENG SUMMARY Cell growth represents the net effect of cell proliferation balanced by attrition from terminal differentiation and/or apoptotic cell death. Apoptosis is a genetically and biochemically regulated process known as programmed cell death. Successful execution of apoptosis is required for normal embryonic development, regulation of the immune response, control of cell numbers, the process of aging and senescence and the cell’s ability to compensate for otherwise damaging stresses. This chapter focuses on mechanisms responsible for regulating apoptosis in the context of cancer and other pathology. Key Words: Anoikis; apoptosis; autophagy; BCL-2; caspase; cell growth; cytolytic T-cells; natural killer; necrosis.

INTRODUCTION Cell growth represents the net effect of cell proliferation balanced by attrition from terminal differentiation and/or apoptotic cell death. Apoptosis is a genetically and biochemically regulated process known as programmed cell death. Successful execution of apoptosis is required for normal embryonic development, regulation of the immune response, control of cell numbers (e.g., in rapidly turning over cells in the GI tract, bone marrow, skin, and uterus), the process of aging and senescence and the cell’s ability to compensate for otherwise damaging stresses (e.g., nutrient deprivation, or toxicant drug exposure and DNA damage). Indeed, dysregulation of apoptosis (i.e., too little or too much) can lead to disease pathology including developmental defects, chronic degenerative, and autoimmune diseases and malignancy. Specifically with respect to malignancy, disabling apoptosis affords the cancer cell not only a selective growth advantage but also resistance to chemoradiation treatment. This feature not only helps to ensure survival when the cancer cell finds itself in an otherwise inhospitable microenvironment, for example following metastases, but also when evading attack by the body’s immune system. This chapter focuses on mechanisms responsible for regulating apoptosis in the context of cancer and other pathology.

CELL DEATH NECROSIS AND APOPTOSIS Cell death can result by two distinct processes, necrosis and apoptosis (Table 72-1). Necrosis is a functionally passive process, which occurs when cells are exposed From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

to extreme physical and chemical environmental imbalances (e.g., hyperthermia, hypoxia) or as a result of forceful physical trauma (e.g., blunt trauma) that directly damages membranes and tissues. Such damage leads to an osmotic-type shock that results in the rapid influx of water and extracellular ions through damaged membranes including those of the organelles, which produces cell swelling and rupture with extrusion of cytoplasmic contents, including lysosomal enzymes, into the microenvironment. This triggers a secondary, local inflammatory response that may lead to the destruction of undamaged bystander cells. By contrast, apoptosis is an energy-dependent, active process that is highly regulated and not easily triggered. However, when activated its biochemical consequences lead to the cell’s proteolytic destruction from within, primarily through the action of a specific set of proteases, the caspases. This type of cell suicide results from the destruction of key regulatory and structural proteins necessary for the maintenance of cell integrity and viability. The classic morphological features of apoptosis, which contrast with those of necrosis, include cell shrinkage, membrane blebbing, partitioning of nuclear and cytoplasm components into membrane bound apoptotic bodies (containing nuclear-ribosomal or organellar remnants), and specific internucleosomal degradation of cellular DNA, which is the classic molecular “hallmark” of this process. Because the apoptotic cell undergoes volume loss associated with nuclear and cytoplasmic condensation, this process has also been called “shrinkage necrosis.” Ultimately the process leads to classic DNA degradation that yields a regular “ladder pattern” of fragmented genetic material that can be diagnostically recognized following gel electrophoresis of DNA as regularly separated, 180-bp fragments. Such DNA “laddering” is caused by the action of a specific class of endonucleases, the caspase-activated DNases (CADs), which are primarily activated by caspases but can also apparently be activated in a caspase-independent manner consistent with a caspaseindependent form of apoptosis. Caspase-independent apoptosis is also initiated by leakage of mitochondrial proteins, including endonuclease G and apoptosis-inducing factor. However, rather than activate caspases these apoptogenic factors are translocated to the nucleus in which they can act to mediate characteristic DNA degradation. Therefore, unlike necrosis, apoptosis is a quiet process with respect to a lack of inflammation and damage to surrounding cells. Instead, the neighbor cells are actively recruited into this morbid process by specific “eat me” phospholipid signals displayed on the surface of the dying cell. Neighbor cells respond by engulfing and phagocytosing the corpse, resulting in its elimination without inflammatory fanfare or collateral injury.

709

710

SECTION IX / ONCOLOGY

Table 72-1 Types of Cell Death Necrosis “Explosive” but passive process Cell swelling and lysis DNA intact Inflammation Anoikis Apoptosis following loss of cell–cell contact Autophagy Golgi, mitochondrial, endoplasmic reticulum membrane degradation Autophagic vacuoles formed May “blend” with apoptosis Apoptosis Cell directed “suicide” Volume loss Membrane blebbing Chromatin condensation DNA fragmentation Apoptotic bodies Noninflammatory process Neighborhood cell engulfment

ANOIKIS AND AUTOPHAGY Two other types of apoptoticlike cell death occur. Anoikis occurs in adherent cells (e.g., fibroblasts and some epithelial cells) that have lost their cell–cell or cell–matrix contact necessary for providing survival signals via integrin binding and “inside-out” survival signaling. Following loss of contact with its environmental substratum, the detached cell undergoes apoptosis. By contrast, autophagy was initially described in yeast and later found to operate in mammalian cells. Autophagy is a process whereby yeast cells are forced to engulf their own organelles in response to amino acid/nutrient starvation. Such recycling occurs in a unique “double-membrane” vesicle, the autophagosome, which surrounds and degrades organellar proteins in an attempt to replenish the depleted intracellular pool of amino acids/nutrients. This provides the affected cell with a means to survive, at least short term, until a more hospitable microenvironment becomes available. However, if uncompensated, by improved nutritional conditions, ongoing autophagy apparently will evolve into cell death indistinguishable from apoptosis. CYTOLYTIC T CELLS AND NATURAL KILLER CELLS INDUCE APOPTOSIS Cytolytic T-cells (CTLs) and natural killer (NK) lymphocytes utilize a granule-mediated exocytosis pathway to destroy their target, including virally-infected and malignant cells, by inducing apoptosis. For example, after targeting a cell for immune destruction, CTLs and NK cells release their cytotoxic granules that contain perforin and serine proteases known as granzymes. Perforin, a pore forming protein, permeates the membrane of target cell to facilitate the entrance of the injected granzymes. Once inside the cell, granzyme B, for example, can cleave and activate procaspase-3, BH3 interacting domain (BID) death agonist and inhibitor of the CAD nuclease to initiate apoptosis. Other granzymes (i.e., granzymes A and C) can apparently induce a caspase-independent type of cell death.

PHYSIOLOGICAL RELEVANCE Apoptosis is essential for the regulation of normal embryonic development as well as the maintenance of cellular and tissue homoeostasis in the adult organism (Table 72-2). Unbalancing

Table 72-2 Physiological Relevance of Apoptosis Embryonic development Regulation of immune system Immune cell development Cytolytic T-cell killing (e.g., immune surveillance, viral infections) Terminating active immune response Tight regulation/renewal of cell numbers (e.g., bone marrow, gastrointestinal, uterus, skin) Compensatory response to cell stress Intrinsic pathway (e.g., Growth factor removal, external radiation therapy, chemotherapy) Extrinsic pathway (e.g., FAS.Ag, TNF-R activation) Senescence/aging

Table 72-3 Apoptosis and Pathology Insufficient (too little) Cancer Developmental defects Excessive (too much) Neurodegenerative Autoimmune Cardiovascular Developmental defects AIDS (T4 cells)

apoptosis by either inappropriate activation or a failure to activate apoptosis pathways can lead to either a shortening or lengthening, respectively, of cell survival. This can result in various pathological consequences, including developmental defects, chronic degenerative and autoimmune disorders, and tumor development (Table 72-3). For example, accelerated or excessive apoptosis has been causally linked to autoimmune disorders including rheumatoid arthritis, systemic lupus erythematosus, as well as the chronic neurodegenerative disorders including Alzheimer’s, Parkinson’s, and Huntington’s disease (see BCL-2 Family Members Regulate Apoptosis Pathway). In addition, primary or secondary immunodeficiency diseases (e.g., HIV), infertility, as well as congenital developmental defects (e.g., limb bud malformation) may result. Furthermore, failure to execute apoptosis following appropriate cues, including developmental or injurious stress, is a characteristic feature of cancer cells that facilitates their prolonged survival and drug-resistant phenotype. APOPTOSIS PATHWAYS Accounting for its ancient importance, apoptosis is conserved in metazoans from worms to man. Apoptosis is triggered in response to developmental cues or environmental stresses (e.g., viral infections, toxicant treatment, and so on) that activate a compensatory elimination response. In addition, the mitochondrion has emerged as the principle organelle that assumes the functional role of master “sensor” organ that responds to cell stress. Indeed, it is the mitochondria that act as a clearinghouse for stress injuries that trigger the intrinsic pathway. As such, this organelle pulls “double duty” for the cell by functioning as the cell’s major source of energy production and its potential death arsenal. Although the mitochondrion functions during normal, unperturbed cell growth to generate lifesustaining chemical energy in the form of ATP that results from

CHAPTER 72 / APOPTOSIS

711

Figure 72-1 Apoptotic pathways. The extrinsic pathway is activated by binding of a death ligand (e.g., TNF-α, FAS.L, TRAIL) to its receptor, which sequentially triggers activation of the initiator caspase 8 followed by an executioner/effector caspase. The intrinsic pathway is activated following a death stimulus resulting from chemotherapy, irradiation or growth factor withdrawal that induces mitochondrial dysfunction characterized by leakage of apoptogenic factors, including cytochrome C, that lead to activation of initiator caspase 9 and effector caspase. Executioner caspases degrade specific substrates to bring about cell death. Cross talk between the extrinsic and intrinsic pathways can occur when caspase 8 cleaves cytosolic BID to truncated BID that is able to translocate to the mitochondria and trigger the intrinsic pathway. This serves to amplify the extrinsic pathway.

electron transfer and oxidative phosphorylation, it also sequesters and can release on cell injury factors that act as potent proapoptotic regulators. Perhaps surprisingly, such apoptogenic factors include the electron carrier cytochrome-C that not only serves as a major transducer of cellular energy but also as the key activator of procaspases when released into the cytosol. It is, therefore, critical to maintain mitochondrial integrity in order to protect cell survival. Apoptotic cell death proceeds through the highly regulated and coordinated signaling of two main pathways, the extrinsic and intrinsic pathways (Fig. 72-1). Although the net result of activation of either pathway will lead to cellular suicide via caspase activation, the regulatory components involved in the two pathways and their mechanism(s) of regulation are both varied and overlapping. EXTRINSIC APOPTOTIC PATHWAY The extrinsic pathway is triggered by the binding of either soluble cytokines (tumor necrosis factor [TNF]-α, TNF-related apoptosis-inducing ligand [TRAIL]) or cell-bound death ligands (FAS.L) to their cognate surface receptors on target cells (see Fig. 72-1). TNF-α, and the FAS.L represent immune mediators produced by inflammatory

cells that are necessary for an appropriate immune response because they work with the effector immune cells that facilitate the removal of infected, neoplastic, aged or otherwise unwanted cells. Interestingly, the same cytokines also work to limit the immune attack by inducing the CTL to undergo apoptosis once its target is destroyed. INTRINSIC APOPTOTIC PATHWAY By contrast, the intrinsic pathway is activated following either damage to an intracellular organelle (e.g., endoplasmic reticulum), DNA or the cytoskeleton following toxicant treatment or from the loss of growth factor–mediated survival signals. Mechanistically, the mitochondrion is able to “sense” cellular stress induced by damage to DNA, intracellular organelles (e.g., endoplasmic reticulum, Golgi, liposomes) or the cytoskeleton. This damage/stress results in the generation of an activated B-cell lymphoma (BCL)-2 homology (BH)-3-only proapoptotic member, (e.g., BID, BIM, and BAD), that translocates to the mitochondria to facilitate a BAX (or BAK) conformational change with oligomerization to form the “death pore” in the outer mitochondrial membrane (OMM). This

712

SECTION IX / ONCOLOGY

Figure 72-2 Cell signaling and apoptosis pathways are integrated. A schematic demonstrating interactions between growth and apoptotic signaling. Arrows (→) indicate stimulation and crosses (⊥) indicate inhibitory effects. The DISC and apoptosome platforms that enable activation of initiator caspases in extrinsic and intrinsic pathways, respectively, are indicated by light shading. Refer to text and other figures for details. Bcl2, B-cell lymphoma gene 2; GF, growth factor.

leads to leakage of intramembranous contents including cytochrome C and other apoptogenic factors that mediate activation of caspases. In this manner, the intrinsic pathway functions as the cell’s compensatory response to overwhelming, nonrepairable damage and/or the prolonged depletion or loss of survival signals. The result is inducing its own death while sparing the undamaged neighbor cells the same fate. To regulate mitochondrial integrity, the BCL-2 gene family has evolved as the principle guardians of the mitochondrion (Fig. 72-2). Interestingly, some of the members of this large gene family are antiapoptotic and function to maintain mitochondrial integrity by inhibiting the proapoptotic members that function to facilitate the release of apoptogenic mitochondrial intramembranous contents. Formation of a mitochondrial “death pore” with leakage of apoptogenic factors requires one of two proapoptotic members, BAX or BAK (see Caspase Proteases Mediate Apoptosis). Thus, although either BAX or BAK null mice are more resistant to toxicant treatment than wild type mice, BAX and BAK double knockout mice are highly resistant. The antiapoptotic BCL-2 family members of which BCL-2 is the prototype function to block mitochondrial dysfunction by binding to and inhibiting pore formation by BH3-only proapoptotic proteins interacting with BAX or BAK. In addition, BCL-2 can also function to block Ca2+ release from the endoplasmic reticulum that

can trigger apoptosis but the mechanism is not clear. Thus, the direct addition of proapoptotic BAX and BAK to isolated mitochondria induces cytochrome-C leakage in vitro, whereas overexpression of BCL-2 (or BclXL, Mcl1, and so on) can block BAX or BAK induced cytochrome C release.

CASPASE PROTEASES MEDIATE APOPTOSIS The intrinsic and extrinsic proteolytic doom that is created by apoptosis is mediated by a unique class of substrate-specific cysteine proteases known as caspases (i.e., c-terminal aspartate site cleaving proteases; see Fig. 72-1). Caspases are synthesized as inactive proenzymes that reside in the cytosol and can be activated by autoproteolysis. The identification and characterization of the caspases began with studies of one of the C. elegans death genes, Ced-3. Subsequent studies in mammalian cells led to the identification of the first human caspase, IL-Iβ-converting enzyme, now called Caspase 1. Iβ-converting enzyme is activated by autoproteolysis of its C terminus at a specific aspartate residue and the resulting heterodimer then specifically cleaves pro-IL-1β into the active, proinflammatory cytokine, IL-1β. More than 16 caspases have been identified, which cleave their substrates, which are either critical regulatory or structural proteins necessary for cell integrity. For example, cleavage of the nuclear lamins, PARP and DFF will lead to disruption of the

CHAPTER 72 / APOPTOSIS

713

Figure 72-3 BCL-2 family. Anti- and proapoptotic BCL-2 members are shown. BCL-2 homology (BH) regions transmembrane domain and alpha helical structures are indicated.

nuclear envelope with classic DNA “ladder” degradation, whereas cleavage of α-Fodrin, MEKK, or FAK will affect growth signals that regulate and maintain cell shape. By degrading such substrates, caspases mediate intracellular liquefaction. TWO CLASSES OF CASPASES Caspases are divided into initiators (e.g., caspases 8 and 10) and effectors (e.g., caspases 3, 6, and 7) based on their role in either initiating or executing apoptosis. In addition, different caspases have preferred aspartate cleavage sites, represented by a minimal putative tetrapeptide sequence (e.g., IETD for caspase 3; VETD for caspase 8), which is found in specific substrates. Activated caspases cleave their substrates on the C-terminal side of such aspartate residues. Once activated, the effector caspases target and degrade/inactivate key regulatory and structural polypeptides necessary for the maintenance of cell integrity. In addition to proteolysis of selected substrates and to ensure inactivation of the dying cell’s genetic capabilities, effector caspase activation also results in activation of specific endonucleases, the CADs, which degrade cellular DNA to produce the “hallmark” molecular signature of apoptosis, DNA laddering. CASPASE ACTIVATION Caspases are synthesized and remain in their inactive, proenzyme state during unstressed cell growth. However, developmental cues or noxious, toxicant stresses can trigger their activation. Procaspases are made up of an N-terminal

prodomain (long in length in case of initiator caspases and short for effector caspases) that is cleaved to yield a p20 and a p10 kDa domain molecule that heterodimerize. Autocleavage of the procaspases occurs at a specific internal aspartate site to yield the p20/p10 dimer that then forms a tetramer, which represents the mature form of the activated protease (Fig. 72-3). Initially, an initiator procaspase becomes activated, which cleaves and activates a downstream effector caspase. This serial activation is referred to as the “caspase cascade.” Initiator caspase activation occurs on one of two different bioplatforms, the death-inducing signaling complex (DISC) in the extrinsic and the apoptosome in the intrinsic apoptotic pathway. Death-mediating ligands including TNF-α, FAS ligand or TRAIL bind to their cognate surface receptors and induce aggregation of receptor-associated molecules. This creates an intracellular “platform,” called the DISC, for the assembly of specific adapter and regulatory molecules (i.e., TRADD or FADD) necessary to achieve activation of procaspase 8 (Fig. 72-4). The DISC functions like a sponge to increase the local concentration of procaspase 8 or 10 and regulatory molecules and facilitate autocleavage and activation of the initiator caspases (i.e., induced proximity model). Once an initiator caspase is activated, it targets and cleaves cytosolic procaspase 3. Activated effector

714

SECTION IX / ONCOLOGY

Figure 72-4 Apoptosome and endoplasmic reticulum. Dysregulation of the mitochondria or endoplasmic reticulum can lead to activation of the intrinsic apoptotic pathway. Bcl2, B-cell lymphoma gene 2; BH3, BCL-2 homology 3.

caspase 3 then proceeds to wreak intracellular proteolytic doom. Receptor-mediated activation of apoptosis occurs independently of the mitochondria in some cells called type-I cells (e.g., Thymocytes). Alternately, caspase 8 activation may lead to “recruitment” of the intrinsic/mitochondrial pathway in cells referred to as type-II cells (e.g., hepatocytes). This “recruitment” serves to amplify the extrinsic pathway and results from caspase 8 being able to directly activate caspase 3 and also being able to target and cleave the cytosolic BH-3 only proapoptotic molecule, BID, into its truncated, activated form, tBID. Activated tBID then is able to translocate from the cytosol to the mitochondrion where it binds and disrupts the interaction between BCL-2 (or other antiapoptotic members like BclXL) and directly activates BAX (or BAK) (see Fig. 72-2). As a result, BAX (or BAK) becomes “functionally” released from inhibition and undergoes a structural conformational change that apparently facilitates its integral insertion into the OMM via its transmembrane domain. Insertion and oligomerization create the “death pore” through which the intramembranous contents are leaked into the cytosol to trigger activation of initiator procaspase 9. For the intrinsic pathway, initiator and effector caspases are activated following a nonreceptor mediated stress event (e.g., toxicant treatment or withdrawal of growth factor survival signals) that results in mitochondrial dysfunction with release of cytochrome C and other apoptogenic factors. Release induces formation of the apoptosome complex that consists of apoptosis protease activating factor (APaf-1), cytochrome C, dATP, and procaspase 9. Following release from mitochondria, cytochrome-C, and Apaf-1 bind, allowing procaspase 9 to also bind Apaf-1 via its caspase activation and recruitment domain, a feature of all caspases. The “mature” apoptosome then facilitates the efficient autocleavage of procaspase 9, which in turn activates cytosolic procaspase 3 in a postmitochondrial mechanism.

INHIBITION OF CASPASES As a safeguard against chance activation of caspases, normal cellular (and some viral) proteins known as inhibitors of apoptosis (IAP; e.g., including cFLIP, cIAP, and survivin) or the baculovirus p35 protein can inhibit caspases. Interestingly, IAP activity was initially recognized in a baculovirus gene product, p35, which inhibits the proteolytic activity of caspases following viral infection of insect cells. Caspase inhibition allows the virus to maximally replicate by overcoming this host-antiviral mechanism. Later, mammalian cells were also found to express a related activity, cIAP. The IAP contain common tandem repeats known as baculovirus inhibitory repeats that enable their binding to and inactivation of caspases (see Fig. 72-2). Cellular IAP may function by either directly inhibiting caspases or by targeting the caspase for ubiquitination and degradation by the proteosome. However, to ensure success of activated caspases, IAP activity can also be inhibited by mitochondrial factors also released along with apoptogenic factors, including SMAC/DIABLO and OMI/HtrA2. These inhibitors function to either directly block cIAP from binding to the activation site of a caspase or by competing with the cIAP once bound to an activated caspase. Because SMAC/DIABLO and OMI/HtrA2 are also sequestered in the mitochondria and released along with cytochrome-C (and other caspase activators), they help to ensure a feedforward functioning of the death pathway once activated. DEATH RECEPTORS MAY ALSO BE ASSOCIATED WITH SURVIVAL SIGNALING In addition to TNF-α and FAS.L, death ligand-induced apoptosis may also be triggered by TRAIL and apoptosis antigen 3 binding to their receptors, DR 4/5 or DR3. TRAIL or apoptosis antigen 3 binding also sequentially induces caspases 8 and 3 activation and can activate BID to “recruit” the intrinsic death pathway. Because some cancer cells are reported to maintain sensitivity to TRAIL-induced killing, exploitation of this vulnerability as a potential cancer treatment is being studied.

CHAPTER 72 / APOPTOSIS

715

Figure 72-5 Apoptosis signaling through extrinsic death receptors. Binding of the death ligand TNF-α (or FAS.L or TRAIL) to the TNF receptor leads to the formation of the DISC. Made up of adapter molecules, TRADD or FADD bind to the death domain of the receptor and bind to procaspase 8 via its death effector domain (DED-DED). This leads to activation of initiator caspase 8 and triggering of the extrinsic pathway. In some cells, TNF-α binding to its receptor, TNF-R1, and TRADD may recruit RIP, RAIDD and TRAF2, which can result in NF-κβ activation and cell survival signaling that results in upregulation of NF-κβ genes including IAP and cFLIP that inhibit caspases.

However, although FAS.L and TRAIL or APO3L signaling induce apoptosis, TNF-α binding does not always induce cell death. Under certain circumstances, TNF-α can promote cell growth. The mechanism(s) by which TNF-α decides on death or growth appears to result from a cell-type specific type of differential signaling. Thus, while binding of the TRADD and FADD adapter protein to the intracytoplasmic portion of the TNF receptor will induce caspase 8 activation, TRADD binding to RIP and RAIDD may affect other regulators including TRAF2, that can stimulate nuclear factor (NF)κβ and inhibitor of NF-κβ (Iκβ) kinase. Thus when activated, Iκβ kinase phosphorylates and releases Iκβ, inhibitor of NF-κβ, resulting in its transcriptional activation. This induces NF-κβ-responsive genes, including cFLIP and other IAP, whose activities help prevent procaspase 8 from being activated and results in blocking the TNFα death signal (Fig. 72-5). However, when the NF-κβ signal falters or becomes weak, the preassembled TRADD-procaspase 8 complex can now activate caspase 8 to engage the extrinsic pathway. Such differential signaling by TNF-α indicates the delicate balance between life and death in certain cells. It also suggests a novel antitumor strategy whereby blocking TNF-α-mediated NF-κβ survival signaling in certain tumors could trip the activation of the extrinsic death pathway. When combined with the effects of chemo- or radiation therapy, this could potentially enhance their cytotoxicity.

BCL-2 FAMILY MEMBERS REGULATE APOPTOSIS PATHWAY BCL-2 is the prototype and founding member of this large family of regulators of apoptosis (see Fig. 72-3). BCL-2 was discovered as the product of translocation of chromosome 14 and 18, t (14; 18) that characteristically occurs in up to 85% of human follicular BCLs. The resulting chromosomal fusion conjoins BCL-2 and the proximal immunoglobulin promoter, which functionally results in overexpression of BCL-2 in B cells. However, unlike other oncogenes, BCL-2 does not promote cell proliferation, rather it cooperates with oncogenes by enhancing cell survival. This unique property was discovered when it was demonstrated that forced expression of BCL-2 in IL-3 factor-dependent cells significantly delayed cell death following growth factor deprivation or on treatment with cytotoxic agents or irradiation. Because BCL-2 is primarily localized to mitochondria, with some expression in the endoplasmic reticulum and nuclear envelope, this highlighted mitochondria in the cell death process. The discovery of BCL-2 also sparked a new field of biomedical research in the mid 1980s and has led to the identification of a large family of structurally related BCL-2 proteins with both antiapoptotic and proapoptotic functions. For example, BCL-2, BclXL, Mcl, and Al contain all four BH domains and are

716

SECTION IX / ONCOLOGY

antiapoptotic and suppress cell death. Other members, including BAX, BAK, BAD, BID, and BIM that contain BH 1-3 or BH-3 only domains are proapoptotic and can induce cell death (see Figs. 72-2 and 72-3). Consistent with its oncogenic role in tumorigenesis, mice whose B cells overexpress the BCL-2 transgene develop a polyclonal follicular hyperplasia, and some even develop a monoclonal BCL that recapitulates the human lymphoproliferative disease. By contrast, BCL-2 null mice display early and massive apoptosis of their lymphocytes as well as other developmental defects including renal insufficiency and a deficiency in pigmentation of the melanocyte. Interestingly and of particular relevance to tumor biology, in addition to follicular lymphomas, many or even most human cancers including breast, lung, colon, prostate, melanoma, and leukemia also express high levels of BCL-2 (or other antiapoptotic members such as BclXL or Mcll) albeit not as the result of a t(14;18) chromosomal translocation. Interestingly, in addition to functioning to suppress apoptosis, BCL-2 has two other properties: it can retard cell-cycle entry from G1 and has an antioxidant function such that its expression in cells is associated with decreased reactive oxygen species (ROS) levels. ANTI- AND PROAPOPTOTIC FUNCTION OF BCL-2 MEMBERS BCL-2 family members are classified based on their sharing of conserved BH domains, BH1-4 (see Fig. 72-3). Although antiapoptotic members contain all four BH domains, proapoptotic members contain either BH1-3 or a BH-3-only domain. The BH-3 domain is necessary for the proapoptotic function. Indeed, removing BCL-2’s BH4 domain and leaving its BH3 domain unopposed, can convert BCL-2 into a proapoptotic BH1-3 containing molecule (like BAX or BAK). Solution and crystal structure of BclXL and BCL-2 have revealed that the BH1, 2, and 3 domains fold to create a hydrophobic pocket that will accommodate the BH-3 domain of another proapoptotic family member such as BAX (or BAK), which results in the effective neutralization of its killing potential. Furthermore, it is the BH3-domain that apparently must be so shielded because synthetic BH-3-only containing peptides, or small organic molecules that can block BCL-2 (or BclXL) and BAX (or BAK) heterodimerization, can also bring about apoptosis. This appears to result in the exposure of the α-helical BH-3 domain of BAX (or BAK) with subsequent integral insertion into the OMM to form the “death pore.” Although BAX (or BAK) is required for mitochondrial disruption and apoptosis, simply overexpressing BAX or BAK in most cells is not sufficient to trigger apoptosis without a concurrent stress signal. This indicates that some type of posttranscriptional regulation is generally required for their activation. BH-3 only proapoptotic members include BID, BAD, BIM, NOXA, and PUMA. When activated following stress, these proapoptotic members act as messengers to convey the death signal generated by a damaged organelle or DNA to the mitochondria. Functionally, when BAX (or BAK) becomes “activated” as a result of a conformational change in its structure triggered either by stress or binding to a BH-3-only containing molecule like tBID, BAX (or BAK) undergoes insertion and oligomerization within the OMM to create a “death” pore-like gateway through which mitochondrial contents, including cytochrome C and other apoptogenic factors, may leak to trigger caspase activation. Mechanistically, an accepted model holds that in viable, unstressed cells, BAX exists primarily in the cytosol or is only peripherally associated in monomeric form but not yet integrated into the OMM. Following a death stimulus, BAX undergoes a conformational change leading to exposure of its transmembrane domain that facilitates insertion into the OMM in which it

also may bind a BH3-only member like tBID to facilitate the oligomerization that produces pore formation. Because of the similarity between the α-helical structures of bacterial pore-forming toxic proteins and BAX, one general model holds that oligomerized BAX/BAK may form death pores large enough to release cytochrome C. This is considered a viable hypothesis because BAX forms pores in artificial lipid membranes that are capable of releasing cytochrome C. Another model implicates the well described mitochondrial permeability transition pore that forms between the inner and OMMs as a result of complexing between the voltage dependent anion channel, the adenine nucleotide transporter and cyclophilin D. This channel will leak ions and dissipate ion gradients and result in the collapse of the mitochondrial membrane potential (∆Ψm) that leads to mitochondrial swelling and rupture, with release of cytochrome-C along and apoptogenic activators. However, because cytochrome-C release precedes both permeability transition pore formation and loss of ∆Ψm that occurs during apoptosis, this is a less attractive mechanism to explain how cytochrome C is released and caspases are activated in apoptosis. REGULATION OF BCL-2 MEMBERS BY PHOSPHORYLATION Antiapoptotic BCL-2 (and MCL-1 and BclXL) as well as proapoptotic BAD are regulated by phosphorylation (Fig. 72-6). For BCL-2, serum and growth factors such as IL-3 mediate its phosphorylation on serine 70, a site located within the flexible loop domain (FLD) that positively regulates its antiapoptotic function. The “regulatory” FLD domain lies between BH4 and BH3 regions in BCL-2 (see Fig. 72-4). Phosphorylation can be mediated by several growth factor-activated protein kinases, including the MAP kinases, ERK 1 and 2 and JNK1 and 2, as well as PKC and the CDC2 kinase. Multisite phosphorylation of BCL-2 in the FLD at serine 70, threonine 69 and serine 87 can also occur when cells are treated with microtubule disrupting agents such as paclitaxel and vincristine. As revealed by phosphomimetic or nonphosphorylatable BCL-2 mutants, phosphorylation at any of these sites can significantly enhance its antiapoptotic function. Mechanistically, phosphorylation appears to stabilize the BCL-2/BAX interaction as well as protect BCL-2 from degradation that can occur during apoptosis. The majority of BCL-2 expressed in the cell is localized in the mitochondria with some in the endoplasmic reticulum and nuclear envelope. Interestingly, the majority of cellular ROS are generated (at least in nonphagocytic cells) in the mitochondria. In addition, although the mechanism is not clear, BCL-2 phosphorylation also regulates its antioxidant and cell cycle entry functions. That is, expression of wt and phosphomimetic but not nonphosphorylatable BCL-2 mutants in cells results in a significantly reduced level of cellular ROS and slower cell growth. Furthermore, slower growth results from BCL-2’s ability to retard cell cycle entry from G1 and is closely associated with upregulation of the cyclin inhibitor p27 and downregulation of CdK2 activity. Increasing ROS levels in such slowly cycling cells by adding low concentrations of hydrogen peroxide directly to them can reverse the effect, suggesting that BCL-2’s antioxidant function may play a role in regulating cellcycle entry. This apparent functional linkage between BCL-2’s antiapoptotic, cell-cycle retardation and antioxidant properties seem to underscore BCL-2’s central role in managing cell stress and apoptosis. That is, cells can generally withstand the effects of an apoptotic-inducing stress better when they are not cycling and ROS levels are lower, as compared to cells undergoing DNA replication and/or cell mitotic division when ROS levels are elevated.

CHAPTER 72 / APOPTOSIS

717

Figure 72-6 Regulation of BCL-2 and BAD by phosphorylation. Growth factors such as IL-3 can activate protein kinases to directly phosphorylate BCL-2 and BAD. BCL-2 phosphorylation blocks BAX oligomerzation and “death pore” formation mediated by tBID that leads to cytochrome C release. BCL-2 also can block cleavage of full length p21 BAX to p18 BAX by a cathepsin-like protease. Phosphorylated BAD is inactive and sequestered in the cytosol. Following a death signal, BAD becomes dephosphorylated and can translocate to the mitochondria and bind and inhibit BCL-2’s antiapoptotic function. BCL-2, B-cell lymphoma gene 2.

Whether other antiapoptotic BCL-2 members also may regulate the cell cycle and possess antioxidant effects is not clear. BAX is the first proapoptotic member of the BCL-2 family to be identified as the heterodimeric binding partner of BCL-2. One popular model holds that BCL-2 binds and neutralizes BAX’s proapoptotic activity. The ratio of anti- to proapoptotic activity blocking binding of BH3-only proapoptotic molecules (i.e., BCL2/BAX) is thought to act like a survival “biorheostat” and set the threshold of cell susceptibility to activation of the intrinsic pathway. Growth factors cannot only provide survival signals to activate antiapoptotic and suppress proapoptotic members but also can transcriptionally increase expression of BclXL (and BCL-2 and Mcl1) to facilitate survival. By contrast, stress-induced activation of p53 can induce upregulation of p53 responsive genes including the proapoptotic molecules BAX/PUMA and NOXA that can induce cell death. This outcome alters the BCL-2/BAX ratio such that increased BCL-2 (or BclXL) or reduced BAX or (BAK) levels favor survival whereas reduced BCL-2 (or BclXL) or increased BAX or (BAK) expression favor increased sensitivity to cytotoxic treatments. In the former case, tumor cells would potentially gain both a survival advantage and a drug-resistance phenotype advantage that could provide a selective growth advantage. As mentioned, regulating the function of BH-3-only proapoptotic members can occur by several mechanisms including phosphorylation and sequestration away from the mitochondria or proteolytic activation with translocation to the mitochondria. For example, IL-3 addition to factor dependent cells induces multisite phosphorylation of BAD, which then bind to protein 14-3-3 in the cytosol away from the mitochondria. However, following growth factor withdrawal, a BAD PP2A-like phosphatase is activated that mediates dephosphorylation, which interrupts its binding to 14-3-3. That leads to BAD’s translocation to mitochondria in which it can directly bind BCL-2

(or BclXL) to disrupt the BCL-2:BAX interaction and lead to BAX activation. This causes “death pore” formation and mitochondrial dysfunction with activation of the intrinsic pathway as discussed. Another BH-3-only molecule, BIM, can also be sequestered from the mitochondria by localizing to the myotubule-associated dynin light chain (LC8) that is part of the motor complex. Following drug treatment, the BIM-LC8 complex is released and BIM is translocated to the mitochondria in which it can now bind and inhibit BCL-2 (or BclXL) interaction with to gain access to BAX (or BAK). Thus, activation and translocation of BH3-only proapoptotic molecules to mitochondria represents a common regulatory theme for BH-3 only proapoptotic molecules. Although antiapoptotic BCL-2 and BclXL members potentially function to bind and neutralize BAX (or BAK), their activity can be overcome by a barrage of such activated BH-3-only molecules. This has prompted investigators to design novel agents that might target and block the BCL-2/ BAX interaction. Various strategies including antisense technology (e.g., BCL-2 antisense) and design and synthesis of cell permeable small molecule inhibitors of BCL-2 (or BclXL) that bind to the BH-3 domain are being developed as potential cancer treatments. Additionally, stabilizing the BCL-2/ BAX interaction or inhibiting caspase activation may potentially preserve the vulnerable tissues in degenerative diseases where the intrinsic pathway is overactive. p53 AND APOPTOSIS Independent of its transcriptional role in mediating cell-cycle arrest and apoptosis, p53 apparently can also stimulate apoptosis when localized in the cytosol. This can occur by directly binding to BCL-2 (or BCL-2XL) to promote the disassociation of BCL-2/BAX, which results in enhanced BAX proapoptotic activity (see Fig. 72-2). p53 not only resides in the nucleus where it functions as a transactivator but also surprisingly may accumulate in the cytosol. Even when deficient in

718

SECTION IX / ONCOLOGY

transactivation function, p53 can directly bind BCL-2 (or BCL-2XL) and lead to BAX activation and engagement of the intrinsic apoptotic program. Although the mechanism is not clear, p53 may facilitate binding of a BH-3 only proapoptotic molecule to BclXL (or BCL-2). Thus, the p53 tumor suppressor may pull “double duty” in regulating cellular responses to stress: first, by functioning in the nucleus as a transcriptioned regulator of genes involved in cell-cycle arrest, and DNA repair and apoptosis, second by working also in the cytosol as a proapoptotic regulator of the intrinsic death pathway.

APOPTOSIS AND DISEASE GROWTH FACTOR SIGNALING AND APOPTOTIC PATHWAYS IN MALIGNANT DISEASE Tumorigenesis is a multistep process resulting from a spontaneous or hereditary mutation or epigenetic alteration of cellular protoonocogenes or tumor suppressor genes. Mutations confer the transformed phenotype of the cancer cell that results from the cell having acquired at least two of the following properties: 1. 2. 3. 4.

Replicate in the absence of growth signals. Overcome growth-inhibitory signals. Avoid the host immune surveillance. Maintain an adequate oxygen and nutrient supply necessary for invasion and metastases to unfamiliar microenvironments. 5. Avoid apoptotic cell death. However, without thwarting apoptosis, the tumor cell would be unable to survive to exhibit its abnormal growth. Furthermore, because malignant cells can both multiply and survive under the most inhospitable of metabolic and microenvironmental conditions (e.g., hypoxia, nutrient depletion) and must be able to successfully defend themselves against or avoid the host antitumor immune response, it is not surprising that mutational or epigenetic alterations of these same genes may affect both proliferation and cell survival. In addition, because chemotherapy and irradiation mediate toxicity by triggering the intrinsic apoptotic pathway, blocking or inhibiting apoptosis in cancer cells may not only cooperate in oncogenicity but also confer a drug resistant phenotype. This linkage seems teleological because it results in conservation of energy and functional simplicity. However, with such interdependence comes the potential risk that mutation of a common component may substantially affect both processes. Indeed, as discussed, either inactivating mutations or downregulation of proapoptotic genes, including BAX, have been identified in some cancers, which would result in a weakening of the cell’s ability to efficiently undergo apoptosis following chemoradiation therapy. In addition, increased expression or an activating mutation of an antiapoptotic gene product like BCL-2 or Mcl-1 may potentially facilitate the development of the malignant phenotype by enhancing cell survival. Interestingly, in addition to lymphomas, BCL-2 is reported to be over/highly expressed in many other cancers, indicating that altered expression of BCL-2 or other antiapoptotic BCL-2 members is a common target in malignancies. Other antiapoptotic regulators may be involved in tumor cells. For example, the cIAP, including cFLIP, are overexposed in some cancers and implicated in the resistance of tumor cells to chemotherapy-induced activation of caspases. Already well described mutations in critical oncogenes such as RAS, AKT, Myc, PI3K, NF-κβ, and MDM2, clearly lead

to enhanced mitogenesis and to enhanced survival as a result of the downregulation of the intrinsic death pathway. Furthermore, inactivating mutations of tumor suppressor genes including p53, p19ARF, ATM, ChK2, and Rb also prevent cell-cycle inhibition and promote survival of cancer cells. Of particular note, p53 apparently can play a dual role in regulating apoptosis. Although an inactivating mutation of p53 not only inhibits its transcriptional activity and upregulation of proapoptotic genes like BAX, NOXA, PUMA, it may also block its newly described function to directly bind and prevent BclXL (or BCL-2) from neutralizing BAX’s proapoptotic activity. Moreover, mutations in transcriptional silencing of some of the positive regulators of the extrinsic (e.g., FAS Antigen/CD95, TRAIL-R1/R2, or proCaspase 8) or intrinsic (e.g., Apaf-1) apoptotic pathways have been reported in various cancers. Therefore, therapeutic targeting with the intent of inhibiting these proapoptotic regulators has the potential for enhancing cell death in cancer cells. This concept represents an active field of research among oncologists and has the potential to add a therapeutic approach to the anticancer armamentarium. Finally, patients with a heterozygous mutation in the FAS antigen or those with initiator caspase-10 deficiency may develop an autoimmune lymphoproliferative syndrome characterized by ineffective lymphocyte apoptosis. In addition, mutation of caspase-8, another initiator caspase, can result in autoimmunity characterized by a developmental immunodeficiency resulting from defects in T-, B-, and NK-cell activation. APOPTOSIS IN NONMALIGNANT DISEASES Neurodegenerative disorders, including Alzheimer’s, Parkinson’s, Huntington’s, and the prion disease, scrapie, represent diseases in which inappropriately enhanced apoptosis has been observed and may be causative. For example, in Alzheimer’s and Huntington’s disease, failure of proper folding of proteins within the endoplasmic reticulum that results from defects in chaperone protein partner binding, appears to lead to a misfolded or “unfolded” protein. This constitutes a form of organelle stress and stimulates the unfolded protein response, which is a normal, compensatory response leading to degradation of the misfolded products in an attempt to restore proper “protein homeostasis.” However, when overwhelmed by too many “misfolded” proteins, the normal unfolded protein response may itself lead to cell damage of the endoplasmic reticulum and trigger endoplasmic reticulum-initiated apoptosis. For example, in Alzheimer’s disease, neurotoxicity is associated with an abnormal accumulation of amyloid-β-peptides owing to a failure of appropriate cleavage of the normal amyloid precursor protein. This buildup results in local endoplasmic reticulum-lipid peroxidation, leads to increased Ca2+ release from the endoplasmic reticulum, and triggers activation of caspase-12, leading to mitochondrial dysregulation and neuronal cell death by activation of the intrinsic pathway. In Huntington’s disease, a mutant Huntingtin protein fails to function normally in the formation of a clathrin complex involved in normal vesicular transport during neurotransmitter release from neurons. Malfunction of this protein appears to result in the activation of both the initiator caspase-8 and effector caspase-3 and is linked to the neurotoxic properties of mutant Huntingtin protein cleavage products. Finally, for at least one form of Parkinson’s disease, the autosomalrecessive juvenile Parkinson’s disease, protein homeostasis in the endoplasmic reticulum may be disrupted in a different way by utilizing the process of endoplasmic reticulum-associated protein degradation of ubiquitinated proteins by the proteosome pathway.

CHAPTER 72 / APOPTOSIS

Activation of this pathway may result in the death of dopaminergic neurons characteristic of Parkinson’s disease. Neuronal cell death may also play a role in another neurodegenerative disorder, the prion disease scrapie. Although cellular prion protein, PrPc, is ubiquitously expressed on the cell membrane as a glycosylphosphotidyl-inositol-linked glycoprotein, when misassembled into an abnormal conformer, PrPsc, it is associated with prion-mediated apoptosis. Thus, although apoptosis appears to play a critical role in neurodegenerative diseases, whether it is causative or merely an effect of the disease is not clear. Nonetheless, blocking activation of caspases and apoptosis in such diseased tissues is an active strategy that, if successful, may at least slow the degenerative process.

ACKNOWLEDGMENT We would like to thank Janice Taylor for her excellent administrative support in the preparation of this manuscript.

SELECTED REFERENCES Alnemri ES, Livingston DJ, Nicholson DW, et al. Human ICE/CED-3 protease nomenclature. Cell 1996;87:171. Ashkenazi A, Dixit VM. Death receptors signaling and modulation. Science 1998;281:1305–1308. Bakhshi A, Jensen JP, Goldman P, et al. Cloning the chromosomal breakpoint of t(14:18) human lymphomas: clustering around JH on chromosome 14 and hear a transcriptional unit on 18. Cell 1985;41: 899–906. Boise LH, Gonzalez-Garcia M, Postema CE, et al. Bcl-x, a bcl-2-related gene that functions as a dominant regulator of apoptotic cell death. Cell 1993;74:597–608. Bratton SB, Cohen GM. Apoptotic death sensor: an organelle’s alter ego. Trends Pharmacol Sci 2001;22:306–315. Cory S, Adams JM. The Bcl2 family: regulators of the cellular life-ordeath switch. Nat Rev Cancer 2002;2:647–656. Cory S, Huang DCS, Adama JM. The Bcl2 family: roles in cell survival and oncogenesis. Oncogene 2003;22:8590–8607. Deng X, Gao F, Flagg T, May WS. Mono- and multi-site phosphorylation enhance Bcl2’s anti-apoptotic and inhibition of cell cycle entry functions. Proc Natl Acad Sci USA 2004;101:153–158. Du C, Fang M, Li Y, Li L, Wang X. Smac, a mitochondrial protein that promotes cytochrome c-dependent caspase activation by eliminating IAP inhibition. Cell 2000;102:33–42. Enari M, Sakahira H, Yokoyama H, Okawa K, Iwamatsu A, Nagata S. A caspase-activated DNase that degrades DNA during apoptosis, and its inhibitor ICAD. Nature 1998;391:43–50. Ferri KF, Kroemer G. Organelle-specific initiation of cell death pathways. Nat Cell Biol 2001;3:255–263. Goldstein JC, Waterhouse NJ, Juin P, Evan GI, Green DR. The coordinate release of cytochrome c during apoptosis is rapid, complete and kinetically invariant. Nat Cell Biol 2000;2:156–162.

719

Hengartner MO. Programmed cell death in the nematode C. elegans. Recent Prog Horm Res 1999;54:213–222. Heusel JW, Wesselschmidt RL, Shresta S, Russel JH, Ley TJ. Cytotoxic lymphocytes require granzyme B for the rapid induction of DNA fragmentation and apoptosis in allogeneic target cells. Cell 1994;76: 977–987. Hockenbery D, Nunez G, Milliman C, Schreiber RD, Korsmeyer SJ. Bcl-2 is an inner mitochondrial membrane protein that blocks programmed cell death. Nature 1990;348:334–336. Hsu H, Xiong J, Goeddel DV. The TNF receptor 1-associated protein TRADD signals cell death and NF-kappa B activation. Cell 1995;81: 495–504. Huang DC, Strasser A. BH3-only proteins-essential initiators of apoptotic cell death. Cell 2000;103:839–842. Kerr JF, Wyllie AH, Currie AR. Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer 1972; 26:239–257. Kornblau SM, Vu HT, Ruvolo P, et al. Bax and PKCα modulate the prognostic impact of Bcl2 expression in acute myelogenous leukemia. Clin Cancer Res 2000;6:1401–1409. Levine AJ. P53, the cellular gatekeeper for growth and division. Cell 1997;88:323–331. Levine B, Klionsky DJ. Development by self-digestion: molecular mechanisms and biological functions of autophagy. Dev Cell 2004;6: 463–477. Meier P, Finch A, Evan G. Apoptosis in development. Nature 2000;407: 796–801. Miyashita T, Reed JC. Tumor suppressor p53 is a direct transcriptional activator of the human bax gene. Cell 1995;80:293–299. Nicholson DW. Caspase structure, proteolytic substrates and function during apoptotic cell death. Cell Death Differ 1999;6:1028–1042. Salvesen GS, Dixit VM. Caspases: intracellular signaling by proteolysis. Cell 1997;91:443–446. Takeshige K, Baba M, Tsuboi S, Noda T, Ohsumi Y. Autophagy in yeast demonstrated with proteinase-deficient mutants and conditions for its induction. J Cell Biol 1992;119:301–311. Thornberry NA, Lazebnik Y. Caspases: enemies within. Science 1998; 281:1312–1316. Vaux DL, Cory S, Adams JM. Bcl-2 gene promotes haemopoietic cell survival and cooperates with c-myc to immortalize pre-B cells. Nature 1988;335:440–442. Vogelstein B, Lane D, Levine AJ. Surfing the p53 network. Nature 2000; 408;307–310. Wang CY, Mayo MW, Baldwin AS. TNF- and cancer therapy-induced apoptosis: potentiation by inhibition of NF-κB. Science 1996;274: 777–784. Wei MC, Zong WX, Cheng EH, et al. Proapoptotic BAX and BAK: a requisite gateway to mitochondrial dysfunction and death. Science 2001;292:727–730. Zha J, Harada HK, Yang E, Jokel J, Korsemeyer SJ. Serine phosphorylation of death agonist BAD in response to survival factor results in binding to 14-3-3 not BclXL. Cell 1996;87:619–628. Zhang H, Huang Q, Ke N, et al. Drosophilia proapoptotic Bcl2/BAX homologue reveals evolutionary conservation of cell death mechanisms. J Biol Chem 2000;275:27,303–27,306.

73 Colorectal Cancer SATYA NARAYAN SUMMARY Colorectal cancer is the third most common death-causing disease in the developed countries. It arises after a series of mutations in various tumor suppressor and proto-oncogenes, each of which are accompanied by specific alterations and pathological conditions. Recent advances have contributed a great deal of understanding of the molecular basis of events that lead to colorectal tumorigenesis. Mutation in the adenomatous polyposis coli (APC) gene is considered to be one of the earliest events in the development of colon cancer. The familial adenomatous polyposis and hereditary nonpolyposis colorectal cancer (HNPCC) are the most commonly inherited colorectal cancers. Familial adenomatous polyposis and HNPCC develop owing to mutations in APC and DNA mismatch repair (MMR) genes, respectively. The main functions of APC are known to regulate β-catenin protein levels through Wnt-signaling pathway, involved in cell migration and cell–cell adhesion, and chromosomal stability. Mutations in the APC gene disrupt the normal functioning of these pathways and thus are involved in the development of colon cancer. Colorectal cancer also develops owing to alterations in the transforming growth factor-β-signaling pathway. Mutations in the pro-apoptotic gene Bax is found in many colon tumors from HNPCC patients. A summary of the molecular genetics of the development of colorectal cancer is briefly discussed here. Key Words: Adenomatous polyposis coli; chromosomal instability; colorectal cancer; familial adenomatous polyposis; hereditary nonpolyposis colorectal cancers; heterozygosity; microsatellite instability; mismatch repair; mutation cluster region; transforming growth factor-β.

INTRODUCTION Prognosis of colorectal cancer depends on the tumor stage at the time of diagnosis with surgery being the most effective treatment. Colorectal cancers develop through a series of histologically distinct stages from “adenoma to carcinoma.” The temporal order in which mutations occur in different genes relates to the progression through the histological stages of cancer from adenoma to carcinoma. Genes that are mutated at different stages of colorectal cancer development include tumor suppressor genes, proto-oncogenes, DNA repair genes, growth factors and their receptor genes, cell-cycle checkpoint genes, and apoptosis-related genes (Fig. 73-1). It is thought From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

that mutations in one of these genes set the stage for initiation and transformation of the normal colonic epithelial cells. Further accumulation of mutations in other genes then contributes to the progression of cancer through the adenoma-carcinoma-metastasis stages. During accumulation of genetic changes, a complex signaling network is established among inactivated and activated cellular pathways. Many cells bearing defective signaling pathways may go through programmed cell death or apoptosis and be removed from the normal population of cells; however, one of the target cells can go through the selection process and survive among other cells by overruling cell-cycle checkpoints and abrogating apoptosis pathways. After clonal expansion, the genetically modified single cell becomes a full-grown tumor. Many colon cancer syndromes have been characterized based on their phenotypic, histological, and genetic changes. Among them, the most common and highly studied colon cancer syndromes are familial adenomatous polyposis (FAP) and hereditary nonpolyposis colorectal cancers (HNPCC), which are caused by mutations in the adenomatous polyposis coli (APC) and mismatch repair (MMR) genes, respectively. Other colon cancer syndromes include PeutzJeghers syndrome, juvenile polyposis syndrome, hereditary mixed polyposis syndrome, and Cowden’s syndrome. These syndromes contain hamartomatous polyps and are inherited in an autosomaldominant fashion. Mutations in serine-threonine kinase II, STKII, gene, Sma and Mad-related protein 4/deleted in pancreatic carcinoma 4, Smad4/DPC4, gene, and phosphatase and tensin homolog deleted on chromosome 10 (pten) gene, and pten gene are linked with Peutz-Jeghers syndrome, juvenile polyposis syndrome, and Cowden’s syndromes, respectively. The inherited syndromes account for only 3–5% of all colon cancers, and the rest are the somatic colon cancers in which both alleles of the tumor suppressor genes are inactivated somatically. This chapter mainly discusses the functions of apc and mmr genes.

APC GENE MUTATIONS Mutations in the APC gene on chromosome 5q21 locus are considered one of the earliest events in the initiation and progression of colorectal cancer. In FAP patients, allelic mutation of the APC gene followed by a loss of heterozygosity is a common feature. Notably, mutations in the APC gene also are found in 60–80% of sporadic colorectal cancers and adenomas. Patients with APC mutations are prone to hundreds to thousands of colorectal adenomas and early onset carcinoma. FAP patients also are prone to small intestinal adenomas (and carcinomas), intra-abdominal

720

CHAPTER 73 / COLORECTAL CANCER

721

Figure 73-1 Model for genetic alterations in the development of colorectal cancer. Based on genetic analysis, at least two pathways are characterized in detail, which lead to colon cancer development. One pathway initiates with mutations in the adenomatous polyposis coli gene and CI followed by mutations in K-ras, deleted in colorectal cancer and p53 genes. The second pathway is initiated by the mutations in the MMR genes and MSI followed by mutations in TGFβIIR, Bax, Tcf-4, insulin-like growth factor 2 receptor, and E2F4 genes. Other pathways are less characterized but a high degree of overlap is expected among them. At least seven gene mutations are needed to develop a normal epithelial cell into carcinoma. However, a cluster of gene mutations is observed in carcinoma and metastatic tumors. Bcl-2, B-cell leukemia/lymphoma 2; CTNNB1, β-catenin; GTBP, G/T mismatch-binding protein; MLH1, mutL homolog1; MSH2, mutS homolog2; PMS, postmeiotic segregation increased; RER, replication error (+).

desmoids and osteomas (Gardner’s syndrome), congenital hypertrophy of retinal pigment epithelium, fundic gland polyps in the stomach, pancreas and thyroid, dental abnormalities, and epidermal cysts. APC is expressed constitutively within the normal colonic epithelium. The APC gene product is a 310-kDa-homodimeric protein localized in both the cytoplasm and the nucleus. The APC gene is inducible, which is transcriptionally upregulated by p53 in response to DNA damage. p53 is a negative regulator of normal cell growth and division. Although mutation in the p53 gene is also widely present in colorectal cancers, its consequence on the expression of the wild-type or mutant APC gene is not clear. In the ApcMin/+ (Min, multiple intestinal neoplasia) mice model, an increased multiplicity and invasiveness of intestinal adenomas is often associated with deficiency for p53. Also, the occurrence of desmoid fibromas in these mice was strongly enhanced by p53 deficiency. The structure of the APC protein with different protein interaction domains is given in Fig. 73-2. At the N-terminal site, the APC protein contains oligomerization and armadillo repeatbinding domains; at the C-terminal site there are EB1 and tumor suppressor protein DLG-binding domains. APC protein also contains three 15-amino acids and seven 20-amino acids repeat regions in which the latter is involved in the negative regulation of β-catenin protein levels in cells. APC functions as a nuclearcytoplasmic shuttling protein and as a β-catenin chaperone. There are three APC nuclear export signals located in the 20-amino acid repeat region of 3, 4, and 7. The first 20-amino acid repeat in the APC gene is located at the 5′-end of the mutation cluster region (MCR) and often produces truncated proteins. An association exists between severe polyposis phenotype and germline mutations in the MCR. Selective pressure for an MCR mutant has been proposed based on the germline mutation in FAP. These findings suggest that a truncation mutation in MCR is necessary for APC

to lose β-catenin-binding and nuclear localization signals. These mutations produce loss of function and tumor progression. APC REGULATION OF β-CATENIN LEVELS THROUGH Wnt-SIGNALING APC acts as a negative regulator of β-catenin signaling in the transformation of colonic epithelial cells and in melanoma progression. The role of the Wingless/Wnt signaling pathway has been described in Drosophila, Xenopus, and in vertebrates. This pathway is important in organ development, cellular proliferation, morphology, motility, and the fate of embryonic cells. In a simple model shown in Fig. 73-3, Wingless/Wnt signaling regulates the assembly of a complex consisting of axin (and its homologs Axinl and conductin), APC, β-catenin, and glycogen synthase-3β kinase (GSK3β). Axin (Axil/conductin) binds to form a complex with APC, β-catenin, and GSK3β to promote β-catenin phosphorylation and subsequent binding with slim (β-TrCP), which mediates its ubiquitination and degradation in the proteasome. GSK3β regulates this process by phosphorylating β-catenin, APC, and Axin (Axil/conductin) complex. Activation of the Wingless/ Wnt signaling pathway inhibits GSK3β and stabilizes β-catenin. Mutations of β-catenin or truncation of APC, which occur both in colon cancer and melanoma cells, increases the stability and transcriptional activity of β-catenin. The stabilized pool of β-catenin associates with members of the T-cell factor (Tcf)–lymphoid enhancer factor (Lef) family of transcription factors. There are four known members of the Tcf and Lef family in mammals, one of which, the human Tcf4 gene, is expressed specifically in colon cancer cells. The β-catenin/Tcf4 complex regulates the proto-oncogene and cell-cycle regulator c-myc, the G1/S-regulating cyclin D1, the gene encoding the matrix-degrading metalloproteinase, matrysin, the AP-1 transcription factors c-jun and fra-1; and the urokinase-type plasminogen activator (uPA) receptor gene. From this discussion, it is clear that the inactivation of APC causes activation of β-catenin, which results in the constitutive activation of Tcf/Lef response genes.

722

SECTION IX / ONCOLOGY

Figure 73-2 Structural features of the adenomatous polyposis coli protein. Most of the mutations in APC occur in the mutator cluster region and create truncated proteins. The truncated proteins contain APC-stimulated guanine nucleotide exchange factor and β-catenin-binding sites in the armadillo repeat domain but lose the β-catenin regulatory activity that is located in the 20-amino acids repeat domain. Somatic mutations are selected more frequently in FAP patients with germline mutations outside of the MCR. aa, amino acid; EB1, end-binding protein 1; hDLG, human disks large; KAP, kinesin superfamily associated protein; NES, nuclear export signal; NLS, nuclear localization signal; PP2-B56α, protein phosphatase 2 B56α-subunit.

In many cases colon tumors carrying mutations in the APC gene also carried increased levels of c-Myc, a known factor for cellular proliferation. A direct link has been established between APC gene mutation, β-catenin activation, and c-myc gene upregulation in colon cancer development. The increased expression of c-Myc through Wnt-signaling pathway upregulates the expression of Cdk4 gene, whose product is responsible for cell-cycle regulation in G1 phase. The c-myc gene encodes a transcription factor of helix-loop-helix leucine zipper family that binds as a heterodimer with Max to E boxes (CACGTG) on target promoters, for example Cdk4 gene, and activates its expression. Max can also interact with Mad and Mxi1 and downregulate c-Myc target gene expression. It has been suggested that the increased levels of Cdk4 protein can phosphorylate pRB. The E2F/DP transcription factor then dissociates from the hyperphosphorylated pRB, which actively transcribes genes involved in cell-cycle progression through G1 phase. It is also suggested that the increased levels of Cdk4 may sequester cellcycle kinase inhibitor p21, p27, and p16. This sequestration may account for the ability of c-Myc overexpression to substitute for p16 deficiency as noted in mouse fibroblast transformation. Thus there is a link among APC gene mutation, β-catenin stabilization, c-myc gene activation, and Cdk4/cyclin D1/pRB/p16 pathway in colorectal cancer development. APC INVOLVEMENT IN ACTIN CYTOSKELETAL INTEGRITY, CELL–CELL ADHESION, AND CELL MIGRATION Actin cytoskeletal integrity is necessary to maintain the shape and adherence junctions of cells. The imbalance in actin cytoskeletal integrity can cause disturbance in cell–cell adhesion and cell migration. The role of APC in actin cytoskeletal maintenance is predicted through its interaction with β-catenin. β-catenin establishes a link between APC and actin by providing a bridge to β-catenin. In Drosophila, mutations in E-APC affect the organization of adherence junctions. Another link of APC with actin is shown through its interaction

with PDZ domain of DLG protein. Because APC colocalizes with DLG at the lateral cytoplasm in rat colon epithelial cells, the APC–DLG complex may participate in regulation of cell-cycle progression. Mutant APC lacking the S/TXV motif for DLG-binding exhibits weaker cell-cycle blocking activity at Go/G1 phase than the intact APC. Interaction of APC with β-catenin and the members of the cadherin family of proteins have been implicated in cell–cell adhesion. The C-terminal domain of E-cadherin interacts with β- and γ-catenin, which associate with α-catenin and form an E-cadherin complex with actin cytoskeleton. This complex maintains the stable cell–cell adhesion. APC becomes a part of the cell–cell adhesion complex linked with E-cadherin, because it directly binds with β-catenin, γ-catenin, and actin filament. The tyrosine phosphorylation of β-catenin by epidermal growth factor, hepatocyte growth factor, and c-Met receptors is important in modulating cadherincatenin complexes from membrane bound form to free cytosolic form. The phosphorylation of β-catenin at tyrosine residue, which is blocked by tyrosine phosphatase Pez, is involved in epithelial cell migration. If the Wnt-pathway and the epidermal growth factor receptor or c-Met receptors pathway are activated at the same time, then the degradation of β-catenin can be inhibited and it may translocate to the nucleus, bind to the Lef–Tcf transcription factor, and downregulate the transcription of E-cadherin gene, CDH1, expression. These complex interactions may finally result in the reduction in E-cadherin-mediated cell–cell adhesion and proliferation of cells. Another important role for APC is in cell migration. Colonic epithelial cells, derived from a committed stem cell, divide in the lower two-thirds of the crypts and migrate rapidly to the surface to form a single layer. During migration, they differentiate into absorptive, secretory, Paneth, and endocrine cells. The function of a wild-type APC is necessary in maintaining the direction of

CHAPTER 73 / COLORECTAL CANCER

723

Figure 73-3 A model for the Wnt-signaling pathway. (A) downregulation of β-catenin transactivation activity in normal colonic epithelial cells. β-Catenin remains in a complex of Axin/Axil/conductin, adenomatous polyposis coli, GSK3β and casein kinase 1 or 2 (CK1 or 2). In the absence of Wnt-signaling, GSK3β and CK1 or 2 kinases become active and phosphorylate β-catenin at serine and threonine residues in the N-terminal domain. Axin and APC promote phosphorylation of β-catenin by acting as a scaffold protein and bringing together enzyme(s) and substrate(s). The phosphorylated β-catenin then binds with F-box protein β-transducin repeat-containing protein of the Skp1-Cullin-F-box complex of ubiquitin ligases and undergoes proteasomal degradation. Even though Tcf–Lef transcription factor without β-catenin may bind to DNA in the absence of β-catenin, the repressors and corepressors such as carboxy-terminal-binding protein, CREB-binding protein, Groucho, and LDL receptorrelated protein bind with Tcf–Lef and repress c-myc or cyclin D1 gene expression to control cell-cycle progression. Some other known genes that are regulated by β-catenin/Tcf–Lef pathway include Cyclin D1, CDH1, Tcf-1, c-jun, Fra-1, PPARd, Gastrin, uPAR, MMP7, Conductin, CD44, Id2, Siamois, Xbra, Twin, and Ubx. (B) The role of mutations in the APC or β-catenin protein in the regulation of β-catenin level and its transactivation property in colon cancer cells. The mutant β-catenin escapes its degradation through the Wnt-pathway and becomes stabilized in the cytoplasm. The stabilized level of β-catenin then heterodimerizes with Tcf–Lef transcription factor and moves into the nucleus, where it actively transcribes cell-cycle-related genes causing cellular proliferation. The binding of β-catenin with Tcf–Lef inhibits the binding of carboxyterminal-binding protein, CREB-binding protein, Groucho or LDL receptor-related protein and potentiates its transcriptional activity.

upward movement of these cells along the crypt-villus axis. Loss of wild-type APC functions resulting from loss of expression or mutations affects cell migration. Instead of migrating upward toward the gut lumen, these cells migrate aberrantly or less efficiently toward the crypt base in which they accumulate and form polyps. In time, they become aneuploid because of defects in chromosome segregation as well as acquire β-catenin stabilization and activation of genes for cell proliferation. The mechanisms by which APC might be involved in cell migration can be understood by its association with the kinesin superfamily associated protein KAP3 that has been established in cell–cell adhesion and migration. APC, mediated by KAP3, interacts with kinesin motor proteins that transport it as well as β-catenin along the microtubules to the growing ends of the cytoskeleton protruding into motile cell membranes. At the tip of microtubule, APC interacts with the endbinding protein, EB1, and protein tyrosine phosphatase (PTP)-BL. PTP-BL modulates the steady state levels of tyrosine phosphorylations of APC-associated proteins such as β-catenin and GSK3β.

In fact, GSK3β kinase activity has been implicated in microtubule dynamics. The mechanism by which mutated APC might play a role in the migration of colorectal tumor cells is predicted through its interaction with Rac-specific guanine nucleotide exchange factor ASEF. APC binds with ASEF and controls its activity. ASEF is activated in colorectal cancer cells containing truncated APC. Active ASEF decreases E-cadherin-mediated cell–cell adhesion and promotes cell migration. Thus, the dynamic association of APC, EB1, ASEF, catenins, EGFR or c-Met receptor, PTP-BL, and E-cadherin proteins at cell–cell adherence junctions and microtubule ends play an important role in cell–cell communication, cell migration, and carcinogenesis.

GENETIC INSTABILITY IN COLON CANCER PROGRESSION Both sporadic and hereditary colorectal cancers exhibit a defined set of biological and genetic cell heterogeneity through a series of molecular events. Two specific pathologically distinct

724

SECTION IX / ONCOLOGY

Figure 73-4 CI in cells carrying mutations in adenomatous polyposis coli gene. (A) A model for the interaction of APC with plus-end of microtubule through end-binding protein and with kinetochore of chromosome through budding uninhibited by benzimidazole 1 in normal colonic epithelial cells. APC can also bind microtubules directly via the C-terminal basic domain. (B) A disruption in the interaction between spindle microtubules and kinetochores because of expression of truncated form of APC in colon cancer cells.

genetic pathways for colorectal cancer have been identified—the chromosomal instability (CI) pathway and the microsatellite instability (MSI) pathway. The CI and MSI are associated with two major inherited syndromes, FAP and HNPCC, respectively. The MSI leads to a 1000-fold increase in the rate of subtle DNA changes, whereas CI enhances the rate at which gross chromosomal changes occur during cell division such as chromosome breaks, duplication, rearrangements, and deletions. Findings describing these two pathways are discussed below. APC GENE MUTATION ASSOCIATED WITH CI Aneuploidy, the abnormal number of chromosomes both quantitatively and qualitatively, is a common characteristic of colon cancer cells. It is thought that the aneuploidy in colon cancer cells arises during mitosis through a defective cell division leading to CI. CI is a common feature of approx 85% colorectal cancers, and it has been detected in the smallest adenoma, suggesting that CI may occur at very early stages of colorectal cancer development. The mechanism(s) by which CI is generated in colon cancer cells is largely unclear. Because the main feature of CI is aneuploidy, it has been suggested that it may arise because of structural changes to the chromosomes and abnormal mitosis. In this scenario, the wild-type APC is involved in maintaining the proper connection of microtubules with chromosomes. APC performs a bridging function between microtubules and chromosomes. APC binds at the plus end of the microtubule through EB1, stretches it to the chromosomes, and inserts them into kinetochores after binding with benzimidazole 1 (Bub1). APC colocalizes to kinetochores and forms complexes with Bub1 and Bub3, the two mitotic checkpoint

proteins. The successful complex formation facilitates proper growth of spindle formation and helps in maintaining euploidy (Fig. 73-4A). Once the APC gene is mutated, the truncated APC protein loses its ability to bind with Bub1 and it is unable to properly maintain the attachment of microtubules at kinetochores, resulting in defect in segregation of chromosomes (Fig. 73-4B). Another observation noted with CI was that after blocking apoptosis in either ApcMin/+ or Apc1638T ES cells, the number of aberrant chromosomes were much greater than in control ES cells that were unable to undergo apoptosis. From these findings, a multiplehit hypothesis of colorectal cancer development has been suggested. The chromosome segregation defect in colon cancer cells with mutated APC gene could lie dormant until an additional genetic-hit suppresses the mitotic checkpoint or the apoptosis of defective cells. In this regard, the proper functioning of the hSecurin, a protein necessary for the completion of the anaphase portion of mitosis, is critical to maintain euploidy, because the loss of hSecurin is often associated with the loss of chromosomes at a high frequency. Furthermore, telomere dysfunction promotes CI that triggered early carcinogenesis in the ApcMin/+ Terc-/- mouse models, whereas telomerase activation restored genomic stability to a level permissive for tumor progression, suggesting that early and transient telomere dysfunction is a major mechanism underlying CI of human cancers. Thus, multiple factors may influence colorectal cancer development from normal epithelial cells to adenoma to carcinoma stages. MMR GENE MUTATIONS ASSOCIATED WITH MSI MSI is characterized by the size variation of microsatellites in tumor DNA as compared to matching normal DNA because of defects in

CHAPTER 73 / COLORECTAL CANCER

725

Figure 73-5 TGF-β signaling in normal colonic epithelium and cancer cells. (A) The functional pathway of TGF-β signaling in normal colonic epithelial cells. TGF-β ligand-binding with TGF-β receptor II recruits TGFβRI into a tetrameric receptor complex resulting in transphosphorylation and activation of TGFβRI. After phosphorylation, the TGFβRI becomes an active kinase that phosphorylates Smad2 and Smad3. Smad anchor for receptor activation (SARA), a scaffolding protein, facilitates interaction between Smad2, Smad3, and TGFβRI. Phosphorylated Smad2 and Smad3 allow the formation of homo- and heterodimerization complex, including Smad4. The Smad complex then translocates into nucleus, binds with DNA, and stimulates the expression of target genes including p15 (inhibitor of cell-cycle kinase that controls cell cycle into G1 phase) and PAI-1 (inhibitor of protease 1 that degrades extracellular matrix proteins during metastasis). (B) The abnormal pathway of TGF-β signaling in colon cancer cells. Many colon cancer cells with MSI because of defective MMR activity induce mutations in TGFβRII gene. Often these are frameshift mutations that insert or delete one or two adenine bases located within a 10 bp polyadenine repeat region (base pairs 709–718, codons 125–128; referred to as BAT-RII) of TGFβRII gene. These mutations encode TGFβRII proteins that are truncated between 129 and 161 amino acids of the cytoplasmic domain, which causes functional inactivation of these proteins. Thus, the loss of TGF-β signaling may abolish cell-cycle control and induce metastasis of colon cancer cells by inhibiting p15 and PAI-1 gene expression, respectively. hMLH1, human mutL homolog1.

the MMR system. MMR system is critical for the maintenance of genomic stability. MMR increases the fidelity of DNA replication by identifying and excising single-base mismatches and insertion– deletion loops that may arise during DNA replication. Thus, the MMR system serves a DNA damage surveillance function by preventing incorrect base pairing or avoiding insertion–deletion loops by slippage of DNA polymerase. Once cells compromise with these functions, it may lead to accumulation of mutations resulting in the initiation of cancer. The MMR genes are involved in one of the most prevalent cancer syndromes in humans known as HNPCC or Lynch syndrome. HNPCC is characterized as an autosomal-dominant inherited disease with 85–90% gene penetrance for early onset of colorectal carcinoma. HNPCC accounts for approx 5–7% of cases of colorectal carcinoma. The molecular diagnosis of HNPCC is based on determining MMR genes for germline mutations. There are at least six different proteins required for the complete MMR system: hMSH2, hMLH1, hPMS1, hPMS2, hMSH3, and hMSH6 (GTBP). hMSH2 forms a heterodimer with either hMSH6 or hMSH3 (depending on the type of lesion to be repaired) and binds to the mismatch site. The

complex of hMSH2 with hMSH6 is called hMutSα, and it is required for the correction of single-base mispairs. The complex of hMSH2 with hMSH3 is called hMutSβ, and it is required for the correction of insertion–deletion loops. Subsequently, a heterodimer of hMLH1 and hPMS2 proteins are recruited by hMutSα or hMutSβ proteins to the mismatch recognition complex, which with other proteins are involved in excision, resynthesis, and ligation of DNA. Mutations in the hMLH1 and hPMS2 have been found in approx 90% of HNPCC cases. Mutations in other MMR genes have been less frequent in HNPCC patients. In many sporadic colon cancers, hypermethylation of the hMLH1 gene promoter resulting in its transcriptional silencing has been observed more than mutations. Both mutations and methylation of hMLH1 gene have been linked with MSI playing a causal role in the initiation of colorectal cancer. The CpG island methylator phenotype pathway has been suggested as a novel mutator pathway that predisposes to colonic tumorigenesis. In many cases of the sporadic colon cancers, the MSI can be induced by CpG island methylator phenotype, followed by hMLH1 gene’s promoter methylation, loss of hMLH1 gene expression, and resultant MMR deficiency.

726

SECTION IX / ONCOLOGY

Mutation rate in cells with MMR deficiency are 100- to 1000-fold greater than in normal cells. There are many targets of MMR gene inactivation; however, the precise stage of tumorigenesis in which mutation of the wild-type MMR gene occurs is not clear. There are four well-studied gene targets of MMR, whose mutations in the microsatellite region have been found in colorectal cancers carrying a mutated hMLH1 gene. These gene targets are the transforming growth factor-β receptor-II (TGF-IIβ), insulin-like growth factor II receptor, antiapoptotic gene Bax, Tcf-4, and the cell-cycle regulator E2F4. Altered TGF-β Signaling in MSI Cells In colon cancers, TGF-β signaling potently inhibits the growth of normal epithelial cells; because the tumor cells are frequently resistant to TGF-β, they cause preneoplastic lesions, increase motility and spread cancer. The structural basis for TGF-β resistance in colon cancers is defined resulting from somatic mutations that inactivate TGF-β receptor II (TGFβRII). In human colon cancer cell lines with high rates of MSI, mutations in the TGFβRII gene were found. These are primarily frameshift mutations that add or delete one or two adenine bases within or from a 10 bp polyadenine repeat in the cysteine-rich coding region (codons 125–128) of the TGFβRII gene. These mutations produce truncated proteins that lack the cytoplasmic domain. As much as 90% of the colorectal cancers with MSI have a mutated TGFβRII gene. TGFβIIR responses are connected with Smads, tumor suppressor gene products, which help to initiate TGFβ-mediated gene transcription (Fig. 73-5). The transcriptional regulation of type-1 plasminogen inhibitor (PAI-1) and cyclin-dependent kinase inhibitor p15 genes are controlled by TGFβ signaling. PAI-1 is the primary inhibitor of tissue-type plasminogen activator and uPA. TGFβ inhibits cell proliferation by inducing a G1-phase cell-cycle arrest acting through increased expression of p15. Thus, the loss of TGFβIIR or Smad4 can abolish TGFβ-signaling and advocate cell proliferation and development of colorectal cancer. Mutations in Proapoptotic Gene BAX in MSI Cells Of tumors with MSI, approx 50% are found with mutations in the Bax gene. These mutations produce frameshift mutations within a coding region of eight nucleotide stretch of guanine residues, G8. Bax heterodimerizes with antiapoptotic protein Bcl2 and induces apoptosis. In the presence of apoptotic signals, BH3 domain proteins tBid are activated and bound with Bax. The interaction of tBid with Bax brings conformational changes in Bax, and this promotes its translocation to the mitochondria. Bax oligomerizes and inserts into the outer mitochondrial membrane in which it forms channels to release cytochrome c (Cyt c) into the cytoplasm. The released Cyt c forms a complex called apoptosome with Apaf-1, dATP, and procaspase 9. The apoptosome recruits and processes procaspase 9 to form a holoenzyme complex, which in turn recruits and activates the effector caspases leading to apoptosis. Thus, the expression of mutated Bax protein may fail to release Cyt c and increase the Bax-Bcl2 ratio resulting in the escape from apoptosis and inducing initiation of colorectal carcinogenesis.

CONCLUSIONS Development of colorectal cancer is a complex, multistep process in which several gene defects coordinate in genotypic and phenotypic outcome. Mutations in many tumor suppressor and protooncogenes in the development of sporadic and hereditary colorectal cancers are well established; however, their precise role in this process is unclear. For example, it is established that mutations

in APC gene may be necessary for the early onset of FAP. Mutations in the APC gene perhaps set a stage for mutations in other genes such as K-ras, deleted in colorectal cancer, and p53. However, the mechanism(s) by which APC gene mutations may contribute to the accumulation of mutations in the genes associated with the colon cancer progression remain unclear. Mutations in the MMR genes can be linked to an increased rate of mutations; however, MMR negative cells develop resistance to apoptosis rather than accumulation of mutations. Thus, MMR gene mutations and their role in MSI and apoptosis need further investigation in context with apoptosis and cell-cycle-related genes. These studies will provide a better understanding of colorectal cancer development and its intervention by genetic or chemotherapeutic means.

SELECTED REFERENCES Ben-Ze’ev A, Geiger B. Differential molecular interactions of betacatenin and plakoglobin in adhesion, signaling and cancer. Curr Opin Cell Biol 1998;10:629–639. Bright-Thomas RM, Hargest R. APC, β-catenin and hTcf-4; an unholy trinity in the genesis of colorectal cancer. Eur J Surg Oncol 2003;29:107–117. Buermeyer AB, Deschenes SM, Baker SM, Liskay RM. Mammalian DNA mismatch repair. Annu Rev Genet 1999;33:533–564. Cain K, Bratton SB, Cohen GM. The Apaf-1 apoptosome: a large caspaseactivating complex. Biochimie 2002;84(2–3):203–214. Dikovskaya D, Zumbrunn J, Penman GA, Nathke IS. The adenomatous polyposis coli protein: in the limelight out at the edge. Trends Cell Biol 2001;11:378–384. Erdmann KS, Kuhlmann J, Lessmann V, et al. The adenomatous polyposis coli-protein (APC) interacts with the protein tyrosine phosphatase PTP-BL via an alternatively spliced PDZ domain. Oncogene 2000;19: 3894–3901. Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell 1990;61:759–767. Fodde R, Kuipers J, Rosenberg C, et al. Mutations in the APC tumour suppressor gene cause chromosomal instability. Nat Cell Biol 2001;3: 433–438. Hannon GJ, Beach D. p15INK4B is a potential effector of TGF-betainduced cell cycle arrest. Nature 1994;371:257–261. Henderson BR. Nuclear-cytoplasmic shuttling of APC regulates β-catenin subcellular localization and turnover. Nat Cell Biol 2000;2:653–660. Hermeking H, Rago C, Schuhmacher M, et al. Identification of CDK4 as a target of c-MYC. Proc Natl Acad Sci USA 2000;97:2229–2234. Huber O, Bierkanp C, Kemler R. Cadherins and catenins in development. Curr Opin Cell Biol 1998;8:685–691. Ilyas M, Straub J, Tomlinson IP, Bodmer WF. Genetic pathways in colorectal and other cancers. Eur J Cancer 1999;35:1986–2002. Jaiswal AS, Narayan S. p53-dependent transcriptional regulation of the APC promoter in colon cancer cells treated with DNA-alkylating agents. J Biol Chem 2001;276:18,193–18,199. Jallepalli PV, Waizenegger IC, Bunz F, et al. Securin is required for chromosomal stability in human cells. Securin is required for chromosomal stability in human cells. Cell 2001;105:445–457. Kawasaki Y, Sato R, Akiyama T. Mutated APC and ASEF are involved in the migration of colorectal tumour cells. Nat Cell Biol 2003;5:211–215. Kawasaki Y, Senda T, Ishidate T, et al. ASEF, a link between the tumor suppressor APC and G-protein signaling. Science 2000;289:1194–1197. Korinek V, Barker N, Morin PJ, et al. Constitutive transcriptional activation by a β-catenin-Tcf complex in APC–/– colon carcinoma. Science 1997;275:1784–1787. Lamlum H, Ilyas M, Rowan A, et al. The type of somatic mutation at APC in familial adenomatous polyposis is determined by the site of the germline mutation: a new facet to Knudson’s ‘two-hit’ hypothesis. Nat Med 1999;5:1071–1075. Lynch HT, de la Chapelle A. Genetic susceptibility to non-polyposis colorectal cancer. J Med Genet 1999;36:801–818. Markowitz S, Wang J, Myeroff L, et al. Inactivation of the type II TGFbeta receptor in colon cancer cells with microsatellite instability. Science 1995;268:1336–1338.

CHAPTER 73 / COLORECTAL CANCER

Morin PJ, Sparks AB, Korinek V, et al. Activation of β- catenin-Tcf signaling in colon cancer by mutations in β-catenin or APC. Science 1997;275:1787–1790. Moss SF, Liu TC, Petrotos A, Hsu TM, Gold LI, Holt PR. Inward growth of colonic adenomatous polyps. Gastroenterology 1996;111:1425–1432. Narayan S, Jaiswal AS. Activation of adenomatous polyposis coli (APC) gene expression by the DNA-alkylating agent N-methyl-N′-nitro-Nnitrosoguanidine requires p53. J Biol Chem 1997;272:30,619–30,622. Narayan S, Roy D. Role of APC and DNA mismatch repair genes in the development of colorectal cancers. Mol Cancer 2003;2:42–56. Kaplan KB, Burds AA, Swedlow JR, Bekir SS, Sorger PK, Nathke IS. A role for the adenomatous polyposis coli protein in chromosome segregation. Nat Cell Biol 2001;3:429–432. Neufeld KL, White RL. Nuclear localizations of the adenomatous polyposis coli protein. Proc Natl Acad Sci USA 1997;94:3034–3039. Palombo F, Iaccarino I, Nakajima E, Ikejima M, Shimada T, Jiricny J. hMutSbeta, a heterodimer of hMSH2 and hMSH3, binds to insertion/deletion loops in DNA. Curr Biol 1996;6:1181–1184. Polakis P. Wnt signaling and cancer. Genes Dev 2000;14:1837–1851. Rampino N, Yamamoto H, Ionov Y, et al. Somatic frameshift mutations in the BAX gene in colon cancers of the microsatellite mutator phenotype. Science 1997;275:967–969.

727

Rosin-Arbesfeld R, Townsley F, Bienz M. The APC tumour suppressor has a nuclear export function. Nature 2000;406:1009–1012. Souza RF, Appel R, Yin J, et al. Microsatellite instability in the insulinlike growth factor II receptor gene in gastrointestinal tumours. Nat Genet 1996;14:255–257. Souza RF, Yin J, Smolinski KN, et al. Frequent mutation of the E2F-4 cell cycle gene in primary human gastrointestinal tumors. Cancer Res 1997;57:2350–1353. Veigl ML, Kasturi L, Olechnowicz J, et al. Biallelic inactivation of hMLH1 by epigenetic gene silencing, a novel mechanism causing human MSI cancers. Proc Natl Acad Sci USA 1998;95:8698–8702. Wadham C, Gamble JR, Vadas MA, Khew-Goodall Y. The protein tyrosine phosphatase Pez is a major phosphatase of adherens junctions and dephosphorylates beta-catenin. Mol Biol Cell 2003;14:2520– 2529. Wijnhoven BP, Dinjens WN, Pignatelli M. E-cadherin-catenin cell–cell adhesion complex and human cancer. Br J Surg 2000;87:992–1005. Zhang F, White RL, Neufeld KL. Phosphorylation near nuclear localization signal regulates nuclear import of adenomatous polyposis coli protein. Proc Natl Acad Sci USA 2000;97:12,577–12,582. Zhang L, Yu J, Park BH, Kinzler KW, Vogelstein B. Role of BAX in the apoptotic response to anticancer agents. Science 2000;290:989–992.

74 Breast Cancer YI HUANG AND NANCY E. DAVIDSON SUMMARY Breast cancer is the most common malignancy among Western women. Like other human cancers, breast cancer results from the accumulation of a series of genetic and/or epigenetic changes to genes with diverse functions, ultimately evolving to a malignant state. Recent scientific advances have offered many new approaches to identify and target a growing number of these genetic and epigenetic alterations in breast cancer. It is expected that this improved understanding of breast cancer biology will enhance risk assessment, prevention, diagnosis, and treatment of this disease. Key Words: Breast cancer; breast cancer susceptibility genes; ductal carcinoma in situ (DCIS); endocrine therapies; epigenetic alterations; growth factor receptors; hormone regulation; hormone resistance; invasive breast cancer; oncogenes; steroid hormone receptors; tumor suppressor genes.

INTRODUCTION Breast cancer is the most common malignancy and a leading cause of mortality in women in the Western world. More than 200,000 American women were diagnosed with breast cancer in 2004. The incidence of breast cancer varies with multiple factors including gender, age, ethnicity, family history, reproductive factors, socioeconomic class, exogenous and endogenous hormones, radiation exposure, genetic susceptibility, and so on. Like other human cancers, breast cancer is recognized as a genetic disease in that it is thought to result from a progressive accumulation of genetic changes. Initiation of breast cancer results from uncontrolled cell proliferation and/or aberrant programmed cell death as a consequence of cumulative alterations of tumor suppressor gene and/or protooncogene expression. Epigenetic changes such as DNA methylation or chromatin modeling can also contribute to modified gene expression. These changes can modulate expression of a variety of critical genes with diverse functions. Previously, the only biological factor of clinical import in breast cancer was steroid receptor expression because it has been an important predictive factor of response to endocrine therapy. But over the last two decades, enormous advances have been made in the understanding of breast cancer at the molecular level. This understanding has revealed a large number of new targets that may play a role as risk, prognostic, or predictive factors and/or aid in the development of new effective therapies of breast cancer. From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

BREAST CANCER PATHOLOGY AND DEVELOPMENT The natural history of breast cancer involves a sequential progression of epithelial changes. Epithelial hyperplasia may be one of the first steps in breast carcinoma development. Hyperplasia without atypia (also termed usual or typical hyperplasia) is characterized by myoepithelial and epithelial growth, resulting in increased cellularity of the terminal duct lobular unit. Atypical ductal hyperplasia is associated with an increased risk of breast cancer progression and could represent the next step toward breast carcinoma. Ductal carcinoma in situ (DCIS) is a malignant proliferation of a subgroup of epithelial cells without invasion through basement membrane into stroma. DCIS is thought to be a precursor of invasive ductal carcinoma in some cases. Different histological patterns of DCIS have been described and differences in biological behavior have been ascribed to these patterns. For example, comedo DCIS is generally a high-grade lesion characterized by clusters of large epithelial cells with central necrosis, reflecting cell death because of the deprivation of essential metabolites, and is thought to be associated with poorer outcome. The noncomedo DCIS, like cribriform and micropapillary DCIS, are usually low grade with a relatively low mitotic rate. Histological grade is judged to be a more important determinant of outcome than the particular pattern of DCIS. In some cases, extended growth of DCIS into a breast lobule makes it difficult to distinguish DCIS from lobular carcinoma in situ (LCIS), a neoplastic proliferation of epithelial cells in the terminal duct lobular unit. In contrast to DCIS, cells of LCIS are usually regular with small, rounded nuclei and defined border. LCIS is felt to be a marker for the propensity to develop invasive breast cancer in the future. Accumulated genetic changes combined with other factors may lead to the development of invasive carcinoma. Infiltrating duct carcinomas are the most common type of invasive breast cancer, accounting for 60–80% of all breast cancers. The second most common type of invasive breast cancer (approx 10% of breast cancers) is invasive lobular carcinoma. Classic lobular carcinoma cells are small with rounded nuclei and dense chromatin. Generally, lobular carcinoma has slightly better overall survival rate than ductal tumors. Medullary carcinoma, comprising 5–7% of breast malignancies, is a circumscribed entity with extensive infiltration of lymphocytes and plasma cells. Other types of uncommon invasive breast carcinomas include tubular, adenoid cystic, mucinous, papillary, and metaplastic carcinomas. A hallmark of all invasive breast carcinomas is disruption of the basement membrane, which suggests that cancer cells

728

CHAPTER 74 / BREAST CANCER

729

Figure 74-1 Mechanisms of ligand-dependent activation of ERs. Binding of agonist to ER complex can induce ER-conformational change, which results in the disassociation of chaperones (hsp90 or hsp70) and ER dimerization. Coactivators of ER such as steroid receptor coactivator-1 and CREB-binding protein can form a complex with receptor dimers, which in turn binds to EREs of target genes. The coactivator complex activates gene transcription via interaction with basal transcription factors and stimulation of histone (His) acetylation (Acet.). ER corepressors N-Cor and silencing mediator of retinoid and thyroid receptors can suppress transcription of ER target genes by interacting with ER dimers and forming a complex with histone deacetylase. Ligand-bound ER may also activate genes without EREs (“nonclassic” activation) via the interaction of ER with other transcription factors such as activating protein-1 (AP-1) NF-κB, or SP1, which subsequently activates target genes that do not contain EREs but encompass specific binding sites for AP-1, NF-κB, or SP1. ER-mediated gene expression usually promotes the breast cell proliferation that may ultimately lead to breast tumorigenesis. Selective ER modulators can selectively inhibit the ligand–receptor interaction, thereby preventing estrogen molecules from binding to these receptors. CBP, CREB-binding protein; HSP, heat shock protein; N-CoR, nuclear receptor corepressor.

may have acquired the ability to metastasize to distant sites. Metastatic tumors can become fatal when they disrupt the function of vital organs.

HORMONE REGULATION ESTROGEN RECEPTOR Estrogen receptors (ER) are members of the steroid receptor superfamily. Two isoforms of ER (ERα and ERβ) have been identified. Although ERα and ERβ have only 30% overall sequence similarity, their DNA-binding and ligand-binding domains are highly homologous, suggesting that the two receptors likely share similar ligands and DNA-binding activity. Most of the knowledge about ER activity and function in breast cancer has been obtained from studies on ERα; much less is known about ERβ. The ER is a nuclear transcription factor that regulates the expression of a number of genes involved in regulation of cell proliferation and differentiation. Binding of a ligand to ER results in a ligand–ER complex that subsequently induces an ER-conformational change, disassociation of chaperones such as hsp90 or hsp70, and receptor dimerization (Fig. 74-1). Activated ER dimers can bind to the estrogen response elements (EREs) of target genes and regulate their transcription. Some nuclear proteins interact with ER and function as coactivators or corepressors of ER. Identified coactivators include SRC-1, SRC-2, SRC-3 (AIB-1), CREB-binding protein/p300, TRIP-1, TIF-2, and others. N-Cor and silencing mediator of retinoid and thyroid receptors are corepressors that can interact with ER and inhibit ER-activated transcription. Coactivators may either directly acetylate histones or recruit other molecules containing such activity, whereas corepressors have the opposite function by forming complexes with histone deacetylases (HDACs). Ligandbound ER may also activate genes without EREs in their promoter region (“nonclassic” activation). In that situation, ligand binding to ER leads to its interaction with other transcription factors such as AP-1, nuclear factor (NF)-κB, or SP1, which subsequently activate

target genes that do not contain EREs but rather encompass specific binding sites for AP-1, NF-κB, or SP1. Finally, some signaling in the estrogen pathway may take place through the membrane. Because of the pivotal role that ER plays in breast cancer progression, development of specific agents targeting ER or its ligands has become an important strategy for breast cancer treatment. Applied endocrine therapies include depletion of the ligand, estrogen, (by oophorectomy or luteinizing hormone releasing hormone analogs in premenopausal women or aromatase inhibitors in postmenopausal women), steroidal antiestrogens that destroy ER such as fulvestrant (also known as ICI 182,780), and selective ER modulators (SERMs) like tamoxifen. SERMs selectively stimulate or inhibit the ER in different target tissues. A SERM may interfere with ER activity in breast cells but activate the ER in other target tissues like uterus, bone, and liver. The most widely used SERM is tamoxifen, which can selectively inhibit the ligand–ER interaction in breast, thereby preventing estrogen molecules from binding to these receptors. But unlike estrogen, the binding of tamoxifen to ER does not cause the ER to acquire the altered conformation that allows it to bind properly to coactivators. As a result, ER target genes that stimulate cell proliferation cannot be activated. Most studies show that early-stage breast cancer patients with ERexpressing tumors benefit from tamoxifen treatment with longer disease-free and overall survival. However, tamoxifen has no obvious impact on the outcome of ER-poor tumors. Thus assessment of ER status in breast cancer patients has been of major utility to predict the potential response to hormonal therapy. Unfortunately de novo or acquired hormone resistance is a feature of some breast cancers. Approximately 30% of breast cancers lack ER-gene expression (ER-negative). Tumors lacking ER protein are usually associated with higher growth rate, poorer differentiation, and worse clinical outcome. Thus, ER expression has been a possible prognostic factor for early breast cancer patients. Genetic changes that could account for the loss of ER in breast cancer include

730

SECTION IX / ONCOLOGY

deletions, insertions, point mutations, or rearrangements of ER gene. However, these alterations do not occur frequently enough to explain the loss of ER expression in a substantial fraction of breast cancers. As discussed later, inactivation of ER-gene expression may result from hypermethylation of the ER-CpG island in promoter region or other epigenetic mechanisms. Support for this possibility comes from in vitro studies showing re-expression of the ER gene in ERnegative human breast cancer cell lines by using inhibitors of DNA methyltransferase or HDAC. In addition, studies using sensitive PCR-based assays have shown that CpG island methylation is a common feature of primary breast cancers that lack ER-protein expression by immunohistochemistry and ligand-binding assays. A number of mechanisms may account for hormone resistance in the presence of ER: a shift to dependence on other growth regulatory pathways, inappropriate balance of coactivators and corepressors, or other defects in the ER-transcriptional pathway. PROGESTERONE RECEPTOR The progesterone receptor (PR) gene also belongs to the nuclear receptor superfamily. Two isoforms of PR, PRA, and PRB, arise from two different transcription initiation sites within the same gene. Like ER, PR can also bind to ligand and target DNA to function as a transcription factor. About half of ER-positive tumors are reported to express both ER and PR genes (ER-positive/PR-positive) and these tumors are more differentiated and more responsive to hormonal therapy. About 25–30% of primary breast tumors are negative for both ER and PR. These tumors are generally associated with poorer differentiation and worse clinical outcome than ER-positive/PR-positive tumors (Table 74-1). Clinical trials have confirmed the value of PR gene expression in predicting the response of breast tumors to hormonal therapy. Elevated PR levels significantly correlate with increased response to tamoxifen in patients with ER-positive metastatic breast cancer. In ER-negative/PR-positive tumors, response to hormonal therapy is commonly observed, although responses are not as frequent as with tumors expressing both ER and PR (Table 74-1). As the ER gene is a critical regulator of PR expression, loss of ER or abrogation of the ER pathway is one mechanism for loss of PR expression. As with ER, the inactivation of PR gene in some breast tumors may also be a consequence of genetic changes or epigenetic modifications like CpG island methylation in the PR promoter region. The mechanism of methylation is of potential interest in accounting for the differential loss of PR in ER-positive/PR-negative phenotype.

ONCOGENES AND GROWTH FACTOR RECEPTORS The identification of alterations to a number of protooncogenes and growth factor receptors in breast carcinoma has highlighted a number of signal transduction cascades that have important roles in normal mammary growth and mammary tumorigenesis. THE HER FAMILY The members of the erb-B/epidermal growth factor receptor (EGFR)/class I family of receptor tyrosine kinases include HER1 (EGFR/c-erb-B1), HER2 (c-erb-B2/neu), HER3 (c-erb-B3), and HER4 (c-erb-B4) and share significant sequence homology to each other. HER2/neu has emerged as one of the most important oncogenes in breast cancer. The molecular mechanisms by which overexpression of HER2/neu induces cell transformation are not completely understood but may involve the formation of heterodimers of HER2/neu with other members of HER family and subsequent activation of several signaling pathways leading to cell proliferation. Although various ligands with selectivity for one of the HER receptors have been recognized, no

Table 74-1 Frequency of ER and PR Expression in Breast Cancer and Response to Hormonal Therapy Tumor phenotype

Frequency of phenotype

Response rate to hormonal therapy

ER+/PR+ ER+/PR– ER–/PR+ ER–/PR–

40% 30% Less than 5% 25–30%

75–80% 20–30% 40–50% Less than 10%

Adapted from Lapidus RG, Nass SJ, Davidson NE, 1998 with permission.

ligand has been identified that directly binds to HER2/neu. The major HER2/neu activated signaling pathways are summarized in Fig. 74-2. Several downstream substrates such as the adaptor molecules, Grb2, and/or Shc, can complex with activated HER dimer via src homology-2 domains. Interaction between the guaninenucleotide exchange factor, son of sevenless, and Shc/grb2 catalyzes the activation of the ras-GTP complex that subsequently activates the serine-threonine kinase Raf-1 and mitogen-activated protein kinase (MAPK) pathways. HER2/HER3 heterodimer can phosphorylate and activate the p85 unit of phosphoinositide 3-kinase (PI3K), an important enzyme for generation of an intracellular second message. MAPK and PI3K activation by HER2/neu leads to enhanced cell proliferation through the activation of a number of nuclear targets, including c-fos, c-jun, c-myc, cyclin D1, and so on. The direct activation of phospholipase C-γ (PLC-γ) by HER2/neu also appears to mediate the transforming potential of HER2/neu although the exact role of PLC-γ is not clear. The induced activity of these pathways may reflect the most important cellular effects of HER2/neu receptor activation in breast cancer cells. Amplification of the HER2/neu gene occurs in approx 20–30% breast carcinomas. Since the first report in 1987 of the correlation of HER2/neu gene amplification with poorer clinical outcome in a multivariate analysis in node-positive breast cancer patients, a number of prognostic studies have given mixed results on this finding. Others have addressed the possibility that HER2 overexpression correlates with absence of ER expression or tamoxifen resistance in an ER-expressing tumor. A possible mechanism for the latter is that HER2/neu overexpression can upregulate MAPK. MAPK then phosphorylates ER, leading to ligand-independent activation of ER and loss of the inhibitory effect of tamoxifen on ER-mediated transcription. These findings imply a role for HER2/neu overexpression as a potential predictive marker for hormone therapy and suggest that the possible utility of combinations of anti-HER2/neu and endocrine therapy in clinical practice. Increasing attention has been given to the HER2/neu receptor as a direct therapeutic target in breast cancer overexpressing HER2/neu. One such approach, use of the anti-HER2 monoclonal antibody, trastuzamab, is effective in the treatment of HER2-positive metastatic breast cancer. Trastuzamab can inhibit both PI3K and MAPK activation induced by HER2/neu and arrest cells at G1 phase. Preclinical and clinical studies suggest that the combination of trastuzamab with certain chemotherapeutic agents such as paclitaxel may significantly improve the therapeutic efficacy over chemotherapy alone. Other studies have focused on the role of the EGFR/HER1 gene in breast cancer. HER1 overexpression is associated with loss of ER expression and, like HER2/neu, HER1 expression may

CHAPTER 74 / BREAST CANCER

731

Figure 74-2 The major signal transduction pathways activated by HER2/neu heterodimers. Ligand induction of HER2/neu-containing dimerization induces phosphorylation on specific tyrosine residues in the C-terminus of the dimer. Downstream signaling substrates grb2 and/or Shc can complex with activated receptor tyrosine kinases via src homology-2 domains. Interaction between SOS and Shc/grb2 catalyzes the activation of ras-GTP complex. Subsequent activation of Raf-1 and MAP kinase pathways by ras leads to cell proliferation and breast tumor progression through activation of a number of nuclear targets, including c-fos, c-jun, myc, cyclin D1, E2F, and so on. Another pathway is the PI3K, whose activation results in increased pro-survival signals. In addition, PLC-γ mediates the transforming potential of HER2 via uncertain pathways. CDK, cyclin-dependent kinase; HER, Hercgulin; MEK, MAPK/ERK kinase; Rb, retinoblastoma protein; P, phosphate; PI3K, phosphoinositide 3-kinase; SOS, son of sevenless.

also be associated with endocrine resistance. Attempts to target HER1 by the small molecule tyrosine kinase inhibitor, gefitinib, have shown modest effects in heavily pretreated advanced breast cancer patients. Combined approaches with trastuzamab, endocrine therapy, or cytotoxics are under evaluation. C-MYC The proto-oncogene c-myc belongs to the myc gene family that also includes B-myc, L-myc, N-myc, and S-myc. c-myc is a nuclear transcription factor that, when dimerized with Max, binds to a specific DNA sequence in the promoter region of target genes that are involved in regulating cell proliferation and apoptosis. Cell transformation may occur when c-myc is aberrantly expressed or genetically altered. c-myc amplification has been found in approx 20% of primary human breast cancers and occurs more frequently in tumors that lack ER and/or PR expression. The transformation of normal cells by c-myc generally requires the coactivation and/or amplification of other important oncogenes like HER2/neu, cyclin D1 gene (ccnd1), and ras, and so on. The tumor suppressor BRCA1 can also physically interact with c-myc and repress c-myc-mediated transcription. However, loss of p53 in breast cells seems to have negligible effect on c-myc-induced mammary carcinogenesis. Although

several studies have indicated a relationship between c-myc gene amplification and poor prognosis, other studies did not find such a correlation. This may reflect complex and differential effects of c-myc on different proteins and signaling pathways that may ultimately have divergent, even opposite functions. RAS The Ras proto-oncogene family, including the N-, H- and K-Ras genes, encodes small GTP binding proteins that regulate many cellular processes, including proliferation, differentiation, cytoskeleton organization, and membrane trafficking. The Ras protein interacts with a variety of factors, including Raf, PI3kinase, and Ral-GDS and can activate several important proproliferation pathways like MEK/Erk, Akt/PKB, and so on. The role of Ras genes in human breast carcinogenesis is not completely clear. Although alteration of Ras gene expression has been reported to be associated with aggressive features of breast carcinoma, the incidence of significant Ras oncogenic mutations in human breast cancer is rare. Increased Ras activity in breast cancer cells has also been reported to occur through HER2/neu signaling mediation. Expression of activated Ras in ER-positive breast cancer cells results in increased estrogen-independent growth of cells in vitro. This might be caused by Ras-induced Erk activity and the

732

SECTION IX / ONCOLOGY

activation of Erk-regulated ER coactivator, AIB1. Ras signaling is a potential target in breast cancer treatment as Ras protein processing farnesyltransferase inhibitors inhibit growth of breast tumor cells in vitro and are under evaluation in advanced breast cancer clinical trials. CYCLIN D1 Cyclin D1 protein is encoded by CCND1 gene and was first implicated in tumorigenesis following localization to chromosome 11q13, a region that is commonly amplified in human breast cancer. Cyclin D1, complexed with either cyclindependent kinase (CDK)4 or CDK6, controls cell cycle progression in the G1 phase by enabling the enzymes to phosphorylate retinoblastoma protein (Rb), a pivotal factor in cell-cycle regulation (see Fig. 74-2). Knockout studies have shown the important role of cyclin D1 in normal mammary gland physiology. Overexpression of CCND1 gene has been found in approx 50% of human mammary carcinomas. Preliminary studies on primary breast cancers indicate that lobular carcinomas universally overexpress CCND1, whereas overexpression in ductal carcinomas is confined to ERpositive cases.

TUMOR SUPPRESSOR GENES AND BREAST CANCER SUSCEPTIBILITY GENES Although activation of oncogenes is accepted as one essential mechanism to transform normal breast cells, it is clear that inactivation or alteration of tumor suppressor genes may also play an important role in breast cancer progression. Dozens of tumor suppressor genes have been identified and the investigation into the function of tumor suppressor genes in breast cancer has provided invaluable information. p53 The most commonly mutated tumor suppressor gene in human tumors is p53. Wild-type p53 functions to maintain the genomic stability and has been called “guardian of the genome.” When DNA damage occurs, p53 acts as a cell-cycle checkpoint to arrest the cells in the G1 phase for damage repair or to eliminate the damaged cells by triggering apoptotic pathways. Several independent pathways activate p53 under certain circumstances, including an ataxia-telangectasia mutated/human homolog of the Rad53 (Chk2)-dependent pathway activated by DNA doublestrand breaks, an INK4 and p14ARF dependent pathway, and a pathway activated by ultraviolet light or cytotoxic antitumor agents. Increased protein levels and DNA binding affinity of p53 may be caused by reduced mouse double minute 2-dependent proteolytic degradation. Involvement of p53 in breast cancer development was first reported when germline mutations in this gene were revealed to be responsible for Li–Fraumeni syndrome, a hereditary cancer syndrome that includes elevated risk of breast cancer. This clinical finding implies an important role for p53 in mammary carcinogenesis. However, comprehensive meta-analysis showed that only approx 20% of primary breast cancers express mutant p53 protein. Although loss of heterozygosity (LOH) in the p53 gene is a common phenomenon in primary breast cancer, the majority of cases with LOH retain one wild-type p53 allele. Several studies indicate that p53 mutations are rare in low-grade DCIS, but more frequent in high-grade DCIS. An increased rate of p53 mutations has been observed in cancers with germline BRCA1/BRCA2 mutations, suggesting that the distribution and type of p53 changes may be affected by BRCA1/BRCA2 status in breast cancer. Moreover, certain types of breast carcinoma like typical medullary carcinoma are associated with higher frequencies of p53 mutation.

Many studies support the prognostic significance of p53 mutations in breast cancer. There is evidence that tumors with missense p53 mutations affecting amino acids critical for DNA binding are associated with very aggressive phenotype and a short survival, whereas null mutations or other kinds of missense mutation display an intermediate clinical phenotype. Preclinical studies also suggest that analysis of p53 mutation status can provide valuable information to predict the response to DNA damaging chemotherapeutic drugs in breast cancer patients. It is hoped that molecular analysis of specific components in p53 signaling pathway may ultimately provide greater impact on the diagnosis, prognostication, and selection of optimal treatment for breast cancer patients. RETINOBLASTOMA GENE The Rb gene, the classic example of a tumor suppressor gene, is located on chromosomal band 13q14 and encodes for a 105 kDa protein. Hyperphosphorylation of Rb by cyclin-dependent kinases allows cells to progress from G1 to S-phase. Unphosphorylated Rb restricts cell-cycle progression in G1 by interacting with various proteins including the growthstimulatory protein, E2F. Loss of Rb activity has been implicated in breast cancer progression. Structural abnormalities of the Rb gene including chromosomal loss and mutations have been reported in approx 20–30% of breast cancers. Other mechanisms like methylation of the Rb promoter may also result in down-regulation of Rb gene expression. Estrogen may not only lead to the phosphorylation of Rb but also increase Rb mRNA and protein levels in estrogen-dependent breast cancer cells, suggesting that regulation of Rb by estrogen may contribute to breast tumorigenesis. BRCA1/BRCA2 It has been estimated that 5–10% of all breast cancers are linked to hereditary genetic predisposition. In the early 1990s, two breast cancer susceptibility genes, BRCA1 and BRCA2, were identified. Germline mutation of either of these two genes followed by somatic inactivation of the remaining wild type allele may lead to the cancer phenotype that accounts for some inherited breast cancer cases. Although BRCA mutations are rare, women who carry these mutations have a very high lifetime risk of developing breast cancer. Clinical features that imply a genetic alteration of BRCA1 and BRCA2 include bilateral breast cancer, breast cancer development at an unusually young age, and ovarian cancer. It is not clear why germline mutations of BRCA1 and BRCA2 genes, which are transmitted in an autosomal-dominant fashion, predispose particularly to breast or ovarian cancer. It is possible that other factors or exposures like reproductive factors and hormonal influences play a role. The BRCA1 and BRCA2 genes are located on chromosome 17q21 and 13q12-q13, respectively. BRCA1/2 proteins are thought to function as tumor suppressors and are involved in repair of DNA damage and maintenance of chromosome integrity. The understanding of the role of BRCA1 and BRCA2 in DNA repair is summarized in Fig. 74-3. When DNA damage occurs, BRCA1 is phosphorylated by ataxia-telangiectasia (ATM) mutated protein kinase. Phosphorylated BRCA1 in turn activates DNA repair machinery through formation of a complex with BRCA2, BRCA1-associated RING domain, Rad51 and/or Rad50, which are key proteins for mitotic recombination and DNA break repair. Loss of BRCA1 or BRCA2 function in breast cells leads to chromosomal instability and triggers checkpoint activation. If the checkpoint function is intact, critical checkpoint genes like p53 and p21 may be activated and induce cell-cycle arrest. However, mutation of BRCA genes frequently results in an accelerated accumulation of secondary genetic changes. When such secondary genetic changes occur to critical checkpoint genes like p53, the checkpoint cannot be activated and

CHAPTER 74 / BREAST CANCER

733

Figure 74-3 Function of breast cancer susceptibility genes BRCA1 and BRCA2 in DNA repair. In response to DNA damage, BRCA1 is phosphorylated by ataxia-telangectasia mutated protein kinase. Phosphorylated BRCA1 activates DNA damage repair by recombination with BRCA2, BRCA1-associated RING domain, Rad50 and/or Rad51. The complex might be involved in DNA-damage cell-cycle checkpoint. Mutations in BRCA1 and/or BRCA2 may cause a complex recombination defect in response to DNA damage and would lead to chromosomal instability. If the checkpoint function is intact, important checkpoint genes such as p53 and p21 can be activated and induce cell-cycle arrest. When such a checkpoint has already been disabled (e.g., p53 mutation), the chromosomal errors may be tolerated and lead to uncontrolled neoplastic progression.

uncontrolled cell growth may result. In support of this, an increased frequency and accelerated onset of breast tumors accompanying p53 inactivation has been observed in BRCA1 knockout mice. The identification of BRCA1 and BRCA2 makes predictive genetic testing possible for individuals in families where a diseaseassociated mutation is identified. However, BRCA1 and BRCA2 mutations are found in a limited proportion (30% of NSCLCs, particularly adenocarcinomas. In contrast, ras mutations have never been detected in SCLC. The presence of a K-ras mutation seems to be an unfavorable prognostic factor and may be associated with tumor progression of NSCLC. This is probably, at least in part, because cells harboring Ras oncogenes become more resistant to radiotherapy and chemotherapy, suggesting that an abnormal Ras signaling may lead to certain cellular changes that are required for developing a resistance to treatment. Given the relevance of ras oncogenes in lung cancer, the activated Ras proteins and/or their downstream signaling components have become attractive targets for therapeutic intervention. The major downstream targets activated by oncogenic Ras include the Raf-1/MEK (ERK-activating kinase)/ERK (extracellular signalregulated kinase) pathway and the Raf-independent pathways such as the JNK/stress-activated protein kinase pathway. Activation of the Raf-1/MEK/ERK pathway is frequently associated with transformation of primary cells by Ras. However, the Raf-independent pathways seem particularly important for transformation of epithelial cells. This is important because all carcinomas including lung cancer are derived from their epithelial origin. Studies have pinpointed the importance of the JNK pathway in the growth control of lung cancer cells. The JNK/cJun/AP-1 pathway plays an essential role in mediating oncogenic Ras transforming function in lung cancer cells, and a significant growth-promoting role of JNK has also been demonstrated in human lung cancer cells. A major target of the JNK pathway is the AP-1 transcriptional activator, which is made up of homo- or heterodimers of the Jun and Fos family proteins. AP-1 activity is abundantly present in NSCLC cells but not in SCLC cells. The induction of AP-1 activity is associated with a transition from SCLC to NSCLC phenotype in an in vitro SCLC progression model. These data suggest that AP-1 is an important mediator of oncogenic Ras signaling and plays a role during lung cancer progression. It is hoped that a detailed understanding of the Rafindependent signal transduction pathways will serve to identify novel targets for therapeutic intervention. The Myc Family Members of the myc gene family, c-myc, N-myc, and L-myc, are sequence-specific DNA-binding transcription factors that play important roles in cell cycle regulation, apoptosis, and differentiation. Myc proteins heterodimerize with other cellular proteins such as Max. The Myc–Max heterodimers bind to specific DNA sequences, leading to transcriptional activation of the targeted genes. The general mechanism of myc oncogene activation is gene amplification or genetic alterations that lead to increased transcription, which results in protein overexpression. Amplification of myc family genes is more frequently observed in SCLC than NSCLC, and has been associated with an adverse effect in survival of SCLC. High levels of c-myc mRNA have been

737

detected in 58% of primary NSCLCs in comparison to nontumor lung epithelium, suggesting that myc gene expression may significantly affect the development of NSCLCs. Moreover, studies have suggested that L-myc polymorphism may play a role in the progression of NSCLC and in determining the different susceptibility to lung cancer in smokers. Bcl-2 The Bcl-2 family of proteins are key regulators in programmed cell death, or apoptosis. The Bcl-2 family consists of proapoptotic and antiapoptotic subfamilies. The interactions between the pro- and antiapoptotic Bcl-2 members help determine cell fate during apoptosis. Bcl-2 is an antiapoptotic member that forms heterodimers with several proapoptotic Bcl-2 members such as Bax and Bak. The mechanisms by which Bcl-2 prevents apoptosis are not fully understood, but stabilizing the mitochondrial membrane and preventing the release of the proapoptotic proteins are possibilities. Bcl-2 is expressed in the early stage of bronchial preneoplasia and is related to smoke exposure. Bcl-2 expression is more common in SCLCs (75%) than NSCLCs, whereas SCCs show a higher level of Bcl-2 expression than adenocarcinomas. There is no clear conclusion regarding the Bcl-2 status and a survival benefit for patients. However, an inverse correlation between Bcl-2 and Bax expression in low grade (high in Bax) and high grade (high in Bcl-2) of NE lung tumors suggests that Bcl-2 may be linked to the disease progression. TUMOR SUPPRESSOR GENES p53 and p14/ARF The p53 gene, located at chromosome 17p13.1, encodes a transcription factor that plays important roles in cell cycle control, DNA damage response, genome stability, and apoptosis. In response to DNA damage, p53 is phosphorylated and functions as a sequence-specific DNA-binding transcription factor that regulates the expression of several downstream genes including p21/Cip1/WAF1, MDM2, GADD45, Bax, and cyclin G, thereby leading to cell cycle arrest or apoptosis. Most p53 mutations identified in human cancers including lung cancer are missense and occur frequently in the DNA-binding domain. Many of these mutations completely abrogate the DNA binding and transactivation function of p53. Thus, transactivation of gene expression by p53 is thought to significantly influence its tumor suppressing activity. Inactivation of p53 in lung cancer mainly occurs through point mutations or by small chromosomal deletions of the region containing the p53 gene. p53 mutations are present in approx 50% of NSCLCs and more than 90% of SCLCs. A significantly higher frequency of p53 mutations was observed in smokers compared with nonsmokers. Loss of p53 function is viewed as an early event in lung tumorigenesis. This is supported by the presence of p53 mutations and/or the accumulation of the p53 protein in preneoplastic lesions, such as bronchial epithelial dysplasia. p14/ARF, encoded by the p16/INK4A locus via an alternative reading frame, is one of the major upstream regulators in the p53 growth-control pathway. As an inhibitor of HDM2 (Mdm2 in mice) that reduces p53 levels through proteasome-dependent degradation, p14/ARF prevents p53 degradation by binding to HDM2 and sequestering it in nucleoli, resulting in p53 activation/stabilization. Thus, dysregulation of p14/ARF in cancer provides an additional means of disrupting the p53 growth-control pathway. p14/ARF mutations were found in 19–37% of NSCLCs; the absence of p14/ARF protein was observed in 65% of SCLCs. However, mutations in the p53 gene and loss of p14/ARF expression

738

SECTION IX / ONCOLOGY

are not always mutually exclusive suggesting the tumor suppressing function of p14/ARF may be p53 independent. p16/INK4A and Rb p16/INK4A and retinoblastoma (Rb) are key components of the cell cycle pathway of p16/INK4Acyclin D1-CDK4-Rb, which plays an essential role in controlling G1 to S phase transition of the cell cycle. p16/INK4A is a member of a family of cyclin-dependent kinase (CDK) inhibitory proteins and functions by inhibiting the kinase activity of the CDK4/cyclin D1 complex. The cyclin D1-associated CDK4 and CDK6 can phosphorylate Rb, which inactivates Rb’s ability to bind to a number of cellular proteins, including the E2F family of transcription factors, thereby releasing cells from the growth inhibitory stage (G0/G1) and promoting G1/S transition. Inactivation of p16/INK4A has been observed in 30–70% of NSCLCs, whereas Rb inactivation is much more common in SCLCs (approx 90%) than in NSCLCs (15–30%). These findings suggest that the p16/INK4A-cyclin D1-CDK4-Rb pathway is inactivated predominantly through p16/ARF4A inactivation in NSCLC, whereas Rb inactivation appears to be the preferred mechanism in SCLC. Homozygous deletion and epigenetic alterations of promoter hypermethylation are the most common mechanisms for p16/INK4A inactivation. Deletion, nonsense mutations, and splicing abnormalities that lead to truncated (mutant) Rb proteins are responsible for alterations of the Rb gene. A reciprocal relationship between Rb inactivation and p16/INK4A expression has been observed in primary lung cancers, suggesting the inactivation of p16/INK4A and Rb are mutually exclusive. Tumor Suppressor Genes at Chromosome 3p Homozygous deletions in specific chromosomal regions may imply the presence of putative TSGs. Chromosome 3p contains several distinct deletion regions including 3p25-26, 3p21-22, and 3p14, which are frequently deleted in lung cancer. Homozygous deletions at 3p14.2 in lung cancer led to the discovery of the fragile histidine triad (FHIT) gene that encodes the protein diadenosine 5′,5′-P1, P3-triphosphate (Ap3A) hydrolase. When introduced into cancer cells with a FHIT gene alteration, FHIT inhibits tumor growth through the induction of apoptosis and/or cell cycle arrest. This links the tumor-suppressor activity of FHIT to its proapoptotic function. Abnormal transcripts of FHIT were found in 40% of NSCLCs and 80% of SCLCs. In addition, approx 50% of primary lung tumors fail to express the FHIT protein, mainly owing to tumor-acquired promoter methylation. LOH of the FHIT gene is more common in smokers (80%) than in nonsmokers (approx 22%), indicating that the FHIT gene is a target for tobacco carcinogens and may occur early in lung tumorigenesis. The RASSF1 (the Ras association domain family 1) gene, located at the lung tumor suppressor locus 3p21.3, consists of two major alternative transcripts, RASSF1A and RASSF1C. RASSF1A is epigenetically inactivated in >90% of SCLCs and 40% of NSCLCs. Reexpression of RASSF1A reduced the growth of human lung cancer cells in vitro and in vivo, supporting a role for RASSF1A as a TSG. RASSF1A binds Ras in a GTP-dependent manner and may serve as a novel Ras effector that mediates the apoptotic effects of oncogenic Ras. RASSF1A inactivation and Kras activation are mutually exclusive events in the development of certain carcinomas. This observation further pinpoints the function of RASSF1A as a negative effector of Ras. Importantly, a significant correlation between RASSF1A promoter methylation and impaired lung cancer patient survival was reported, and RASSF1A silencing was correlated with several parameters of poor prognosis

and advanced tumor stage (e.g., poor differentiation, aggressiveness, and invasion). Thus, RASSF1A may serve as a useful marker for the prognosis of lung cancer patients. In addition to RASSF1A, several candidate TSGs including FUS1, SEMA3B, 101F6, and NPRL2 are all located within 100 kb of RASSF1A in the 3p21.3 region. Data indicate that LOH at 3p, in particular the 3p21.3 region, is one of the earliest changes occurring in smoking-damaged, but histologically normal lung epithelium, suggesting that alterations in the expression of these putative TSGs are a critical early step in the development of lung cancer.

CHROMOSOME ABNORMALITIES Numerous chromosomal abnormalities including common regions of chromosomal loss and LOH have been found in lung cancer. The chromosomal arms with the most frequent LOH are 1p, 3p, 4p, 4q, 5q, 8p, 9p (p16/INK4A), 9q, 10p, 10q, 13q (Rb), 15q, 17p (p53), 18q, 19p, Xp, and Xq. SCLC and NSCLC show distinct regions of frequent LOH, suggesting that SCLC and NSCLC frequently undergo different genetic alterations. Additionally, a much higher frequency of widespread chromosomal abnormalities has been found in lung adenocarcinoma from smokers than such tumors arising in nonsmokers. This observation supports the idea that lung cancers in nonsmokers arise through genetic alterations distinct from the common events in tumors from smokers. The most consistent chromosomal abnormality in lung cancer has been the allele loss of chromosome 3p (3p12-26), found in more than 90% of SCLCs and more than 80% of NSCLCs. Several distinct regions on 3p show frequent allele loss or independent homozygous deletions in lung cancer, which represent strong evidence for the presence of potential TSGs. A number of putative TSGs including RASSF1A and FHIT have been cloned from these 3p deletion regions. It should be noted that 3p allele loss also occurs in the normal epithelium of smokers without lung cancer as well as preneoplastic lesions of the lung. A progressive increase in the frequency and the size of 3p allele loss was correlated with the increasing severity of histopathological preneoplastic/preinvasive changes. These observations suggest that 3p allele loss is an early event involved in the pathogenesis of lung cancer.

EPIGENETIC CHANGES (ABERRANT PROMOTER METHYLATION) Epigenetic alterations in general and gene silencing events in particular are viewed as fundamental aspects of tumorigenesis. Tumor cells generally exhibit a global DNA hypomethylation and regional promoter hypermethylation that is associated with gene silencing. Aberrant methylation of CpG islands in promoter regions is one of the major mechanisms for silencing of TSGs in cancer cells. Methylation serves as an alternative to the genetic loss of a TSG function by deletion or mutation. TSGs that are frequently methylated in primary lung tumors include adenomatous polyposis coli, retinoic acid receptor beta, CDH13 (H-cadherin), FHIT, RASSF1A, p16/INK4A, death-associated protein kinase (DAPK), and others. The profile of methylated genes in NE tumors (SCLC and carcinoids) was very different from that of NSCLC, and significant differences in the methylation patterns also exist between the two major types of NSCLC (adenocarcinoma and SCC), indicating that distinct epigenetic alterations may contribute to the development of different subtypes of lung cancer.

CHAPTER 75 / LUNG CANCER

A significantly reduced overall, and/or disease-free survival has been linked to the promoter hypermethylation of DAPK and RASSF1A. Methylated DNA sequences can be detected in primary tumors, circulating in serum DNA from lung cancer patients, in sputum samples before the onset of invasive lung cancer, in preneoplastic lesions of lung carcinomas, as well as in the smokingdamaged bronchial epithelium from cancer-free heavy smokers. Thus, an assessment of promoter hypermethylation of certain TSGs may be useful biomarkers for early diagnosis and risk assessment of lung cancer.

TELOMERASE Telomerase is a specialized ribonucleoprotein polymerase that adds TTAGGG repeats at the ends of vertebrate chromosomal DNA called telomeres. Telomeres are important in maintaining stability of the chromosome, which undergo progressive shortening with cell division through replication-dependent sequence loss at DNA termini. Telomerase is thought to compensate for the loss of telomeric repeats and is associated with the acquisition of the immortal phenotype. Telomerase activity is detected in many types of human cancers. Almost all SCLCs and approx 80% of NSCLCs are telomerase positive. Telomerase activity has been directly correlated with malignant and metastatic phenotypes of human lung cancer. Telomerase RNA expression was present in 20% of hyperplasia, 53% of dysplasia, and 100% of microinvasion and carcinoma in situ. Lower levels of telomerase activity were detected in preneoplastic lung cancer lesions compared with invasive cancer. These findings indicate that dysregulation of telomerase expression increases along with lung tumor progression.

GROWTH FACTORS/RECEPTORS Many growth factors or neuropeptides and their receptors are expressed and/or mutated in human lung cancer cells or in the adjacent stroma. This results in multiple autocrine and/or paracrine circuits that activate the positive growth signaling pathways and promote tumor cell growth. Neuropeptides and their specific G protein-coupled receptors (GPCR) play a key role in the regulation of SCLC growth, whereas peptide growth factors and receptor tyrosine kinases (RTKs) are more important for stimulating NSCLC growth. In the context of the multistage evolution of cancer, these autocrine and/or paracrine mitogens may play a role as tumor promoters in the early stages of lung tumorigenesis and/or later as growth stimulators in the unrestrained growth of the fully developed lung tumor. NEUROPEPTIDE GROWTH FACTORS Neuropeptide growth factors include gastrin-releasing peptide (GRP) and bombesin-like peptides that have been extensively studied as neuropeptide mitogens in SCLC. Expression of GRP was demonstrated in 20–60% of SCLCs but less frequently in NSCLCs by immunohistochemical analysis. Neutralizing antibodies against GRP/bombesin and bombesin antagonists inhibit both in vitro and in vivo growth of SCLC cells. In addition, a high frequency (>50%) of coexpression of gastrin and its receptor CCK-B in SCLCs suggests that the neuropeptide autocrine signaling is a prominent driving force for SCLC proliferation. GRP and bombesin-like peptides function as autocrine growth factors, which signal through the GPCRs. GRP and bombesin-like peptides bind to at least three GPCRs, including the GRP receptor, the bombesin receptor, and the neuromedin B receptor, which lead to the activation of multiple intracellular protein kinase pathways

739

that ultimately converge on the regulation of some important cell cycle proteins. In particular, activation of protein kinase C (PKC) and the mitogen-activated protein kinase cascade play important roles in regulating neuropeptide autocrine growth signaling in SCLC cells. RECEPTOR TYROSINE KINASES RTKs are key components of signaling pathways that regulate cell growth, differentiation, and survival. RTKs consist of an extracellular-, transmembrane-, and intracytoplasmic-domain and are activated through appropriate ligand binding. Ligand binding of the receptor results in dimerization and activation of tyrosine kinase activity through autophosphorylation of the intracellular tyrosine kinase domain of the receptor itself. The phosphotyrosine residues of activated RTKs play a crucial role in transducing extracellular signals to intracellular signaling molecules. Oncogenic activation of RTKs occurs through deletion, point mutations, as well as overexpression, which lead to persistent activation of the receptor-mediated signaling pathways independent of ligand binding, resulting in uncontrolled cell growth. A number of RTKs are overexpressed or mutated in human lung cancer cells. These RTKs are attractive targets for therapeutic intervention against lung cancer. Epidermal Growth Factor Receptor Epidermal growth factor receptor (EGFR, also called erBB1) and heregulin receptor (HER2/neu) belong to the type-I EGFR family that regulates epithelial proliferation and differentiation. The EGFR is activated by binding the extracellular domain to one of several ligands, including EGF (betacellulin), transforming growth factor-α (epiregulin), and HB-EGF (amphiregulin). NSCLC tumors have been demonstrated to synthesize transforming growth factor-α and HB-EGF, and these growth factors seem to form an autocrine feedback loop with EGFR and, as a result, play an important role in the tumorigenesis. The EGFR expression varies according to histological subtypes. Overexpression of EGFR was found frequently in 60–80% of NSCLCs but rarely in SCLCs. Overexpression of EGFR is one of the earliest and most consistent abnormalities in bronchial epithelium of high-risk smokers. It is present at the stage of basal cell hyperplasia and persists through squamous metaplasia, dysplasia, and carcinoma in situ. The prognostic significance of EGFR overexpression in NSCLC is unclear. However, patients with strong EGFR-expressing tumors tended to have shorter survival than the patients with slight- or nonexpressing tumors. HER2/neu is highly expressed in approx 30% of NSCLC, particularly adenocarcinomas. High levels of HER2/neu expression are associated with a shortened survival time and with an intrinsic chemoresistance to three commonly used chemotherapeutic drugs (doxorubicin, etoposide, and cisplatin) in NSCLC. Coexpression of EGFR and HER2/neu also occurs in NSCLC tumors. HER2/neu and EGFR coexpression appears to have an additive effect on survival. High HER2/neu expression and high EGFR/HER2 coexpression gave unfavorable prognosis, indicating that HER2/neu and EGFR play a critical role in the biological behavior of NSCLCs. Proto-Oncogene c-Kit c-Kit RTK is a class III receptor similar to platelet-derived growth factor receptor. c-Kit and its natural ligand, stem cell factor, form an autocrine loop that is implicated in the development of SCLC. Coexpression of c-Kit and stem cell factor was reported in up to 70% of SCLC tumor specimens and cell lines. Inhibition of c-Kit activation by several tyrosine kinase inhibitors led to growth suppression and cell death in SCLC cell lines.

740

SECTION IX / ONCOLOGY

Several additional growth factor/RTK autocrine loops that have been implicated in the growth regulation of lung cancer include insulin-like growth factors and their receptor IGF-Rs as well as hepatocyte growth factor and its receptor c-Met.

DYSREGULATION OF APOPTOSIS Apoptosis, or programmed cell death, is a highly specific and regulated process that plays an important role in the development and homeostasis of multicellular organisms as well as tumorigenesis. There is an increasing evidence that the control of apoptosis is disrupted in many tumor cells. The dysregulated apoptosis is thought to contribute to enhanced tumor progression and metastasis. Furthermore, it has become evident that resistance to apoptosis is one potential mechanism whereby tumor cells escape from chemotherapy-induced cytotoxicity, leading to cell survival. Development of drug resistance (intrinsic or de novo) in cancer cells may be because of their selective alterations in the expression and/or function of genes important to the apoptotic response. These changes may provide a selective advantage for tumor cells, thereby rendering them resistant to chemotherapy. CASPASES-8 Apoptotic cell death is controlled by the activation of caspases, a family of aspartate-specific cysteine proteases. Caspases are normally expressed as latent zymogens and are activated by proteolytic cleavage at the onset of apoptosis. There are two well-characterized caspase-activating cascades that regulate apoptosis: one is mediated by cell surface death receptors and the other is regulated by the mitochondria. Chemotherapeutic drug-induced apoptosis is, in general, regulated via the mitochondrial pathway. Caspase-8 is a so-called initiator caspase and plays an essential role in the death receptor-mediated apoptosis. However, studies indicate that activation of caspase-8 is also involved in druginduced apoptosis. Caspase-8 mRNA expression is absent in most (79%) of SCLC cell lines, but retained in all NSCLC cell lines tested. Loss of caspase-8 gene expression is correlated with the absence of the protein. In addition, the absence of caspase-8 expression was also found in a subset of SCLC tumors but in none of NSCLC tumors. These observations suggest that inhibition of caspase-8 activity represents one potential mechanism by which lung tumor cells, especially SCLC cells, develop resistance to drug-induced apoptosis. SERINE/THREONINE KINASES Akt/PKB The serine/threonine kinase Akt (or protein kinase B) is the cellular homolog of a retroviral oncogene product, v-Akt. Akt is commonly activated in response to growth factor stimulation and plays a key role in cancer progression by stimulating cell proliferation and inhibiting apoptosis. Constitutively activated Akt can contribute to tumorigenesis in vivo in a variety of tissues. The antiapoptotic function of Akt is mediated by phosphorylating various proteins involved in apoptosis, including proapoptotic Bcl-2 family member Bad, the forkhead transcription factor, and glycogen synthase kinase 3. Akt activation might inhibit apoptosis by promoting the increased expression of survival molecules or the degradation of proapoptotic molecules. Constitutive Akt activity was found in premalignant and malignant human bronchial epithelial cells, but not in nonmalignant human bronchial epithelial cells. Furthermore, Akt is activated in primary human lung epithelial cells in vitro and in vivo by exposure to nicotine and to a tobacco-specific carcinogen, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone. Nicotine activation of Akt also promotes the survival of human lung epithelial cells by

suppressing apoptosis. The fact that phosphorylated Akt (an activated form) was detectable in 4-(methylnitrosamino)-1-(3pyridyl)-1-butanone-induced murine lung tumors and in human lung cancers derived from smokers suggests that nicotine activation of Akt plays a role in lung tumorigenesis, and sustained Akt activation might be necessary for tumor maintenance. Thus, Akt and the PI3K/Akt pathway might be important molecular targets for lung cancer prevention and treatment. Protein Kinase C PKC is a family of at least 11 structurally related serine/threonine protein kinases that play crucial roles in transducing signals that regulate diversified biological functions, including proliferation, differentiation, and apoptosis. PKC is known to play a role in the regulation of two survival pathways, the PI3K/Akt and MEK/ERK pathways, and has been implicated in regulating drug resistance in cancer cells. Alterations in expression and/or activity of PKCs have been linked to the drug resistant phenotype in lung cancer cell lines and primary lung tumors. Studies suggest that the antiapoptotic effect of specific PKC isoforms, in particular the novel PKC isoforms (PKC-ε and PKC-δ), may contribute to chemoresistance in lung cancer cells. Forced expression of PKC-ε in SCLC cells conferred chemoresistance by effectively blocking drug-induced apoptosis, which was accompanied by the inhibition of cytochrome-c release from mitochondria and caspase activation, indicating that dysregulation of the mitochondrial pathway may be involved in PKC-ε-mediated survival. In contrast, inhibition of PKC-δ function with rottlerin, a PKC-δ specific inhibitor, or by a dominant negative mutant, increased apoptosis and dramatically sensitized NSCLC cells to chemotherapeutic drugs. These findings suggest that specific PKC isoforms may be targeted for the novel therapeutic intervention in order to overcome drug resistance in lung cancer. Death-Associated Protein Kinase The DAPK is a family of calmodulin-regulated serine/threonine protein kinases that functions as a positive mediator of programmed cell death and is ubiquitously expressed in various tissues. A unique autoinhibiting mechanism is thought to keep these death-promoting kinases silent in healthy cells and ensures their activation only in response to apoptotic signals. DAPK is involved in both death receptor and mitochondrial pathways of apoptosis. DAPK is located on chromosome 9q34.1, a region of frequent allelic loss in 50–64% of lung cancer, and is inactivated in a large fraction of lung cancer via aberrant promoter methylation. Promoter methylation of the DAPK gene in NSCLCs is associated with poor prognosis and an advanced pathological stage, but not with exposure to tobacco or asbestos, suggesting that DAPK may be important in the progression of NSCLC from early to late stage disease.

CONCLUSION It is becoming clear that the development of lung cancer and its progression is the result of accumulations of multiple and complex genetic defects. Modern molecular and genetic techniques have identified numerous genes and proteins whose expression is altered in lung cancer and/or preneoplastic lesions. These molecular abnormalities have provided novel possibilities for translational research into the diagnosis, prognosis, and the treatment of the disease. Certainly, knowledge of the fundamental cellular and molecular biology of lung cancer will significantly advance with the information from the human genome project and the use of genome-wide techniques such as microarrays and proteinomics. One major challenge is to successfully translate knowledge into clinical practice.

CHAPTER 75 / LUNG CANCER

SELECTED REFERENCES Alberg AJ, Samet JM. Epidemiology of lung cancer. Chest 2003;123: 21S–49S. Bost F, McKay R, Dean N, Mercola D. The JUN kinase/stress-activated protein kinase pathway is required for epidermal growth factor stimulation of growth of human A549 lung carcinoma cells. J Biol Chem 1997;272:33,422–33,429. Bost F, McKay R, Potapova O, et al. The Jun kinase 2 isoform is preferentially required for epidermal growth factor-induced transformation of human A549 lung carcinoma cells. Mol Cell Biol 1999;19:1938–1949. Brambilla E, Gazzeri S, Lantuejoul S, et al. p53 mutant immunophenotype and deregulation of p53 transcription pathway (Bcl2, Bax, and Waf1) in precursor bronchial lesions of lung cancer. Clin Cancer Res 1998;4:1609–1618. Brambilla E, Negoescu A, Gazzeri S, et al. Apoptosis-related factors p53, Bcl2, and Bax in neuroendocrine lung tumors. Am J Pathol 1996;149: 1941–1952. Broers JL, Viallet J, Jensen SM, et al. Expression of c-myc in progenitor cells of the bronchopulmonary epithelium and in a large number of non-small cell lung cancers. Am J Respir Cell Mol Biol 1993;9:33–43. Budihardjo I, Oliver H, Lutter M, et al. Biochemical pathways of caspase activation during apoptosis. Annu Rev Cell Dev Biol 1999;15: 269–290. Chan DC, Gera L, Stewart JM, et al. Bradykinin antagonist dimer, CU201, inhibits the growth of human lung cancer cell lines in vitro and in vivo and produces synergistic growth inhibition in combination with other antitumor agents. Clin Cancer Res 2002;8:1280–1287. Cho JY, Kim JH, Lee YH, et al. Correlation between K-ras gene mutation and prognosis of patients with nonsmall cell lung carcinoma. Cancer 1997;79:462–467. Chun KH, Kosmeder JW II, Sun S, et al. Effects of deguelin on the phosphatidylinositol 3-kinase/Akt pathway and apoptosis in premalignant human bronchial epithelial cells. J Natl Cancer Inst 2003;95: 291–302. Clark AS, West KA, Blumberg PM, Dennis PA. Altered protein kinase C (PKC) isoforms in non-small cell lung cancer cells: PKCδ promotes cellular survival and chemotherapeutic resistance. Cancer Res 2003;63:780–786. Colgin LM, Reddel RR. Telomere maintenance mechanisms and cellular immortalization. Curr Opin Genet Dev 1999;9:97–103. Cory S, Adams JM. The bcl2 family: regulators of the cellular life-ordeath switch. Nat Rev Cancer 2002;2:647–656. Crowell JA, Steele VE. AKT and the phosphatidylinositol 3-kinase/AKT pathway: important molecular targets for lung cancer prevention and treatment. J Natl Cancer Inst 2003;95:252, 253. Dammann R, Li C, Yoon JH, et al. Epigenetic inactivation of a RAS association domain family protein from the lung tumour suppressor locus 3p21.3. Nat Genet 2000;25:315–319. Dammann R, Schagdarsurengin U, Strunnikova M, et al. Epigenetic inactivation of the Ras-association domain family 1 (RASSF1A) gene and its function in human carcinogenesis. Histol Histopathol 2003;18:665–677. Dempsey EC, Newton AC, Mochly-Rosen D, et al. Protein kinase C isozymes and the regulation of diverse cell responses. Am J Physiol Lung Cell Mol Physiol 2000;279:L429–L438. Ding L, Wang H, Lang W, Xiao L. Protein kinase C-epsilon promotes survival of lung cancer cells by suppressing apoptosis through dysregulation of the mitochondrial caspase pathway. J Biol Chem 2002; 277:35,305–35,313. Dong X, Mao L. Molecular biology of human lung cancer. In: Weitberg AB, ed. Cancer of the Lung, Totowa: Humana Press, 2002; pp. 103–127. Evan GI, Vousden KH. Proliferation, cell cycle and apoptosis in cancer. Nature 2001;411:342–348. Facchini FM, Spiro SG. Chemotherapy in small-cell lung cancer. In: Brambilla C, Brambilla E, eds. Lung Biology in Health and Disease: Lung Tumors, vol. 124. New York: Marcel Dekker, 1999; pp. 611–630. Ferreira CG, Span SW, Peters GJ, et al. Chemotherapy triggers apoptosis in a caspase-8-dependent and mitochondria-controlled manner in the

741

non-small cell lung cancer cell line NCI-H460. Cancer Res 2000;60:7133–7141. Fontanini G, De Laurentiis M, Vignati S, et al. Evaluation of epidermal growth factor-related growth factors and receptors and of neoangiogenesis in completely resected stage I–IIIA non-small cell lung cancer: amphiregulin and microvessel count are independent prognostic indicators of survival. Clin Cancer Res 1998;4:241–249. Franklin WA, Veve R, Hirsch FR, et al. Epidermal growth factor receptor family in lung cancer and premalignancy. Semin Oncol 2002;29 (Suppl 4):3–14. Gazzeri S, Della Valle V, Chaussade L, et al. The human p19ARF protein encoded by the beta transcript of the p16INK4a gene is frequently lost in small cell lung cancer. Cancer Res 1998;58:3926–3931. Girard L, Zochbauer-Muller S, Virmani AK, et al. Genome-wide allelotyping of lung cancer identifies new regions of allelic loss, differences between small cell lung cancer and non-small cell lung cancer, and loci clustering. Cancer Res 2000;60:4894–4906. Grandori C, Eisenman RN. Myc target genes. Trends Biochem Sci 1997;22:177–181. Greenlee RT, Hill-Harmon MB, Murray T, Thun M. Cancer statistics, 2001. CA Cancer J Clin 2001;51:15–36. Grossi F, Loprevite M, Chiaramondia M, et al. Prognostic significance of K-ras, p53, bcl-2, PCNA, CD34 in radically resected non-small cell lung cancers. Eur J Cancer 2003;39:1242–1250. Hanaoka T, Nakayama J, Mukai J, et al. Association of smoking with apoptosis-regulated proteins (Bcl-2, bax and p53) in resected nonsmall-cell lung cancers. Int J Cancer 2001;91:267–269. Hecht SS. Tobacco smoke carcinogens and lung cancer. J Natl Cancer Inst 1999;91:1194–1210. Hibi K, Takahashi T, Yamakawa K, et al. Three distinct regions involved in 3p deletion in human lung cancer. Oncogene 1992;7:445–449. Hirsch FR, Scagliotti GV, Langer CJ, et al. Epidermal growth factor family of receptors in preneoplasia and lung cancer: perspectives for targeted therapies. Lung Cancer 2003;41(Suppl 1):S29–S42. Hofmann J. Modulation of protein kinase C in antitumor treatment. Rev Physiol Biochem Pharmacol 2001;142:1–96. Hung J, Kishimoto Y, Sugio K, et al. Allele-specific chromosome 3p deletions occur at an early stage in the pathogenesis of lung carcinoma. JAMA 1995;273:558–563. Husgafvel-Pursiainen K, Boffetta P, Kannio A, et al. p53 mutations and exposure to environmental tobacco smoke in a multicenter study on lung cancer. Cancer Res 2000;60:2906–2911. Jacobson DR. ras mutations in lung cancer. In: Brambilla C, Brambilla, E, eds. Lung Tumors. New York: Marcel Dekker, 1999; pp. 139–156. Ji L, Nishizaki M, Gao B, et al. Expression of several genes in the human chromosome 3p21.3 homozygous deletion region by an adenovirus vector results in tumor suppressor activities in vitro and in vivo. Cancer Res 2002;62:2715–2720. Jones PA, Baylin SB. The fundamental role of epigenetic events in cancer. Nat Rev Genet 2002;3:415–428. Kalomenidis I, Orphanidou D, Papamichalis G, et al. Combined expression of p53, Bcl-2, and p21WAF-1 proteins in lung cancer and premalignant lesions: association with clinical characteristics. Lung 2001; 179:265–278. Kashiwabara K, Oyama T, Sano T, et al. Correlation between methylation status of the p16/CDKN2 gene and the expression of p16 and Rb proteins in primary non-small cell lung cancers. Int J Cancer 1998;79: 215–220. Kelley MJ, Johnson BE. Molecular genetics of lung cancer. In: Carney DN, ed. Lung Cancer. Boston: Arnold, 1995; pp. 245–266. Khosravi-Far R, Campbell S, Rossman K, Der CJ. Increasing complexity of Ras signal transduction: involvement of Rho family proteins. Adv Cancer Res 1998;72:57–107. Kim DH, Nelson HH, Wiencke JK, et al. Promoter methylation of DAPkinase: Association with advanced stage in non-small cell lung cancer. Oncogene 2001;20:1765–1770. Krystal GW, Hines SJ, Organ CP. Autocrine growth of small cell lung cancer mediated by coexpression of c-kit and stem cell factor. Cancer Res 1996;56:370–376.

742

SECTION IX / ONCOLOGY

Kumimoto H, Hamajima N, Nishimoto Y, et al. L-myc genotype is associated with different susceptibility to lung cancer in smokers. Jpn J Cancer Res 2002;93:1–5. Levine AJ. p53, the cellular gatekeeper for growth and cell division. Cell 1997;88:323–331. Lowy DR, Willumsen BM. Function and regulation of ras. Annu Rev Biochem 1993;62:851–891. Mabry M, Nakagawa T, Nelkin B, et al. v-Ha-ras oncogene insertion: a model for tumor progression of human small cell lung cancer. Proc Natl Acad Sci U S A 1988;85:6523–6527. Mao L, Hruban RH, Boyle JO, et al. Detection of oncogene mutations in sputum precedes diagnosis of lung cancer. Cancer Res 1994;54: 1634–1637. Mills NE, Fishman C, Rom WN, et al. Increased prevalence of K-ras oncogene mutations in lung adenocarcinoma. Cancer Res 1995;55: 1444–1447. Milner AE, Palmer DH, Hodgkin EA, et al. Induction of apoptosis by chemotherapeutic drugs: the role of FADD in activation of caspase-8 and synergy with death receptor ligands in ovarian carcinoma cells. Cell Death Differ 2002;9:287–300. Minna JD, Roth JA, Gazdar AF. Focus on lung cancer. Cancer Cell 2002;1:49–52. Mitsudomi T, Viallet J, Mulshine J, et al. Mutations of ras genes distinguish a subset of non-small-cell lung cancer cell lines from small-cell lung cancer cell lines. Oncogene 1991:6:1353–1362. Nicholson SA, Okby NT, Khan MA, et al. Alterations of p14ARF, p53, and p73 genes involved in the E2F-1-mediated apoptotic pathways in non-small cell lung carcinoma. Cancer Res 2001;61:5636–5643. Rathi A, Kazuo Y, Onuki N, et al. Telomerase and lung cancer. In: Brambilla C, Brambilla E, eds. Lung Biology in Health and Disease: Lung Tumors, vol. 124. New York: Marcel Dekker, 1999; pp. 269–277. Risse-Hackl G, Adamkiewicz J, Wimmel A, Schuermann M. Transition from SCLC to NSCLC phenotype is accompanied by an increased TRE-binding activity and recruitment of specific AP-1 proteins. Oncogene 1998;16:3057–3068. Robles AI, Linke SP, Harris CC. The p53 network in lung carcinogenesis. Oncogene 2002;21:6898–6907. Roz L, Gramegna M, Ishii H, et al. Restoration of fragile histidine triad (FHIT) expression induces apoptosis and suppresses tumorigenicity in lung and cervical cancer cell lines. Proc Natl Acad Sci U S A 2002;99: 3615–3620. Rozengurt E. Autocrine loops, signal transduction, and cell cycle abnormalities in the molecular biology of lung cancer. Curr Opin Oncol 1999;11:116–122. Rusch V, Klimstra D, Venkatraman E, et al. Overexpression of the epidermal growth factor receptor and its ligand transforming growth factor α is frequent in resectable non-small cell lung cancer but does not predict tumor progression. Clin Cancer Res 1997;3:515–522. Sanchez-Cespedes M, Ahrendt SA, Piantadosi S, et al. Chromosomal alterations in lung adenocarcinoma from smokers and nonsmokers. Cancer Res 2001;61:1309–1313. Sattler M, Ralgia R. Molecular and cellular biology of small cell lung cancer. Semin Oncol 2003;30:57–71. Scheid MP, Woodgett JR. Pkb/akt: functional insights from genetic models. Nat Rev Mol Cell Biol 2001;2:760–768. Sethi T, Langdon S, Smyth J, Rozengurt E. Growth of small cell lung cancer cells: Stimulation by multiple neuropeptides and inhibition by broad spectrum antagonists in vitro and in vivo. Cancer Res 1992;52(Suppl):2737s–2742s.

Seufferlein T, Rozengurt E. Galanin, neurotensin, and phorbol esters rapidly stimulate activation of mitogen-activated protein kinase in small cell lung cancer cells. Cancer Res 1996;56:5758–5764. Sherr CJ. Cancer cell cycles. Science 1996;274:1672–1677. Shih CM, Kuo YY, Wang YC, et al. Association of L-myc polymorphism with lung cancer susceptibility and prognosis in relation to ageselected controls and stratified cases. Lung Cancer 2002;36:125–132. Shivapurkar N, Reddy J, Chaudhary PM, Gazdar AF. Apoptosis and lung cancer: a review. J Cell Biochem 2003;88:885–898. Shivapurkar N, Toyooka S, Eby MT, et al. Differential inactivation of caspase-8 in lung cancers. Cancer Biol Ther 2002;1:65–69. Shohat G, Shani G, Eisenstein M, Kimchi A. The DAP-kinase family of proteins: study of a novel group of calcium-regulated death-promoting kinases. Biochim Biophys Acta 2002;1600:45–50. Sidransky D, Hollstein M. Clinical implications of the p53 gene. Annu Rev Med 1996;47:285–301. Sklar M. Increased resistance to cis-diamminedichloroplatinum(II) in NIH 3T3 cells transformed by ras oncogenes. Cancer Res 1988;48:793–797. Sklar M. The ras oncogenes increase the intrinsic resistance of NIH 3T3 cells to ionizing radiation. Science 1988;239:645–647. Smit E, Postmus P. Chemotherapy of small cell lung cancer. In: Carney D, ed. Lung Cancer. Boston: Arnold, 1995; pp. 156–172. Sozzi G, Sard L, De Gregorio L, et al. Association between cigarette smoking and FHIT gene alterations in lung cancer. Cancer Res 1997;57:2121–2123. Tang X, Khuri FR, Lee JJ, et al. Hypermethylation of the death-associated protein (DAP) kinase promoter and aggressiveness in stage I nonsmall-cell lung cancer. J Natl Cancer Inst 2000;92:1511–1516. Toyooka S, Toyooka KO, Maruyama R, et al. DNA methylation profiles of lung tumors. Mol Cancer Ther 2001;1:61–67. Tsai CM, Chang KT, Wu LH, et al. Correlations between intrinsic chemoresistance and HER-2/neu gene expression, p53 gene mutations, and cell proliferation characteristics in non-small cell lung cancer cell lines. Cancer Res 1996;56:206–209. Vaux DL, Korsmeyer SJ. Cell death in development. Cell 1999;96:245–254. Vos MD, Ellis CA, Bell A, et al. Ras uses the novel tumor suppressor RASSF1 as an effector to mediate apoptosis. J Biol Chem 2000;275: 35,669–35,672. Weber JD, Jeffers JR, Rehg JE, et al. p53-independent functions of the p19(ARF) tumor suppressor. Genes Dev 2000;14:2358–2365. West KA, Brognard J, Clark AS, et al. Rapid Akt activation by nicotine and a tobacco carcinogen modulates the phenotype of normal human airway epithelial cells. J Clin Invest 2003;111:81–90. Wingo PA, Ries LA, Giovino GA, et al. Annual report to the nation on the status of cancer, 1973–1996, with a special section on lung cancer and tobacco smoking. J Natl Cancer Inst 1999;91:675–690. Wistuba II, Behrens C, Virmani AK, et al. High resolution chromosome 3p allelotyping of human lung cancer and preneoplastic/preinvasive bronchial epithelium reveals multiple, discontinuous sites of 3p allele loss and three regions of frequent breakpoints. Cancer Res 2000;60:1949–1960. Xiao L, Lang W. A dominant role for the c-JunNH2-terminal kinase in oncogenic Ras-induced morphologic transformation of human lung carcinoma cells. Cancer Res 2000;60:400–408. Zochbauer-Muller S, Fong KM, Maitra A. 5’ CpG island methylation of the FHIT gene is correlated with loss of gene expression in lung and breast cancer. Cancer Res 2001;61:3581–3585. Zochbauer-Muller S, Gazdar AF, Minna JD. Molecular pathogenesis of lung cancer. Annu Rev Physiol 2002;64:681–708.

76 Discoveries and Frontiers in Prostate Cancer Translational Sciences JONATHAN W. SIMONS SUMMARY With the application of molecular biology and biotechnologies to human prostate cancer research, new targets in the lethal phenotypes of prostate cancer have been rapidly ascertained. How many are pharmacologically amenable and can be blocked with clinical relevance remains to be discovered. For all the emergent genetic and protein target insights, perhaps the greatest challenges will be the participation of more men with prostate cancer in translational research clinical trials that evaluate the new science of therapeutics designed specifically for prostate cancer. Key Words: Androgen receptor; benign prostatic hyperplasia; glutathione S-transferase; hereditary prostate cancer; polycyclic aromatic hydrocarbons; proliferative inflammatory atrophy; prostate cancer; prostatic intraepithelial neoplasia.

OVERVIEW Prostate cancer is among the most common human neoplasias. In 1990, prostate cancer overtook lung cancer as the most common noncutaneous cancer diagnosed in US men. More than 30,000 US men were projected to die from prostate cancer in 2004; this is an estimated annual loss of 300,000 yr of life. Prostate cancer is truly a medical burden created from successful increases in Homo sapiens’ life expectancy in the 20th century; less than 1% of cases are diagnosed under the age of 40. Specific genetic alterations in pathogenesis are reviewed later, but a key factor in human prostate cancer biology is time. Time is required to acquire the specific genetic alterations in the prostate epithelium to confer an invasive metastatic phenotype for human prostate cancer. Despite the decades it takes to emerge, prostate carcinogenesis begins early. Undiagnosed, microscopic foci of prostate cancer have been identified in autopsy series in men younger than 30 yr. The latency of most prostate cancer suggests very important gene–environment interactions may influence later steps in the development of the lethal phenotypes of prostate cancer. Prostate cancer initiation—manifested by a histologically identifiable lesion—is frequent, being detected at autopsy series in nearly one-third of men over age 45. The majority of these microscopic adenocarcinomas do not progress to clinically detectable From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

tumors within the lifetime of these men. A critical question is: why not? Epigenetic and genetic factors that create clinically aggressive cancers from incidentally diagnosed ones are at the heart of investigation. The long latency period of prostate cancer affords a new opportunity for chemoprevention, and potentially dietary interventions as described later. Despite its incidence and mortality in the developed world, prostate cancer has historically lagged behind others in molecular understanding of pathogenesis. With a significant push from the activism of national nonprofit organizations of prostate cancer survivors in the United States, federal funding has increased over 500% from its 1993 baseline to 2000. By 2003, a concomitant increase in new research models and basic understanding has affected the field and defined treatment targets for the future. Described later are discoveries that have generated new opportunities for translational research to reduce the morbidity and mortality of prostate cancer. A special emphasis is placed on implications for therapy in chemoprevention and the development of systemic therapies for advanced disease.

NEW CONCEPTS IN CARCINOGENESIS AND CHEMOPREVENTION Prostate neoplasia develops in two different regions of the gland, with most lesions (approx 80%) found in the periphery that can be palpated, and most of the remaining cancers found in a periurethral region transition zone. Benign prostatic hyperplasia (BPH) originates in the transition zone of the prostate. Based primarily on this tissue difference in the incidence of benign growth and adenocarcioma formation, and the finding that stromal cell proliferation is typically a major component of BPH. BPH is not the precursor of prostate cancer. Rather, prostatic intraepithelial neoplasia (PIN) containing foci of dysplastic ductal and acinar cells is thought to be the precursor lesion of prostate cancer. PIN lesions have become critical for application of molecular probes to define the earliest changes in malignant transformation at the level of DNA and RNA alterations and protein expression. The incidence of prostate cancer shows clear age, racial, and geographic dependencies. Prostate cancer is relatively uncommon in Asian populations and prevalent in Scandinavian countries, and the highest incidence (and mortality) rates known are in African Americans, being nearly twice higher than in white Americans.

743

744

SECTION IX / ONCOLOGY

Figure 76-1 GST-π inactivation is the earliest known and essential step in prostate cancer carcinogenesis. Loss of “caretaker” function permits electrophilic attack and mutations to be introduced into prostate epithelial cell DNA. Chemoprevention strategies are oriented to restoring this protective function of GST-π.

Mortality rates vary significantly by country, ranging from the highest reported rates in Trinidad, to 23/100,000 in the United States, to less than one-fifth US rates in Japan. Significant focus has been placed on interactions of diet and lifestyle and genes critical to prostate cancer tumorigenesis in the developed world. There is also species specificity to human prostate cancer risk. Despite the presence of a prostate in all male mammals, there is no reported high risk for prostate cancer in any other aging mammal in veterinary medicine except for the Canis familiaris (dog). Human lineage departed from other higher primates eight million years ago. Homo sapiens emerged only about 200,000 yr ago, and only in the past 15,000 yr have humans and dogs experienced dietary changes that might affect prostate cancer carcinogenesis. With civilization and departure from agrarian to urban dietary habits, cooked, smoked, processed, and stored meats became predominant dietary components during this time for man and, via leftovers, for dog. Polycyclic aromatic hydrocarbons and other components of burned red meat in particular are implicated as potential prostate cancer carcinogens; these are created in charbroiling fish and meat. As Asians cultures do not char meats in cooking, the reduced incidence of prostate cancer in Asian men has been speculated to be because of lesser lifetime exposure to intraprostatic polycyclic aromatic hydrocarbons. This would also help account for the increasing rates of prostate cancer in Asian Americans who immigrated to the United States and Western diets one generation ago. Genetic alterations in prostate cancer tumorigenesis suggest a specific pattern for carcinogenesis in the prostate epithelium (Fig. 76-1). The most common genomic alteration in prostate

cancer—and the first identified molecular event in prostate cancer—is the inactivation of the glutathione S-transferase (GST)-π gene. GSTs are enzymes that catalyze conjugation reactions between glutathione and various reactive chemical species. GSTs detoxify carcinogens, acting as “caretaker” genes to prevent carcinogeninduced DNA mutations and cancer development. The π-family of GSTs are ubiquitous, and have a critical role in genomic DNA damage defense. For example, GST-π knockout mice carrying disrupted genes exhibit increased skin cancers on cutaneous treatment with 7,12-dimethylbenzanthracene. Somatic inactivation of GST-π by transcriptional silencing associated with de novo CpG island hypermethylation occurs in more than 90% of prostate cancers. The loss of GSTP1 in PIN and prostate cancer exposes the cell to genomic damage inflicted by heterocyclic amine carcinogens, such as those present in “well-done” meats, and perhaps more importantly by reactive oxygen species produced by inflammatory cells in prostatitis (discussed later). Thus a unique feature of the prostate epithelium compared with other human cancers is how early in malignant transformation it loses a defined enzyme in detoxifying carcinogens. Some high-grade PINs and early adenocarcinomas appear to arise from areas of proliferative inflammatory atrophy (PIA). PIA has been recognized as the lesion creating genetically damaged clones that form PIN. PIA has become a focus for carcinogenesis research in prostate cancer. Inflammation and other environmental factors may (1) lead to the destruction of prostate epithelial cells, (2) increase epithelial turnover, and (3) drive the increased prostate epithelial proliferation that occurs in PIA. Decreased G1/S checkpoint control protein p27(Kip1) has been found in PIA

CHAPTER 76 / DISCOVERIES AND FRONTIERS IN PROSTATE CANCER TRANSLATIONAL SCIENCES

lesions. Increased oxidant and electrophile stress in the setting of increased proliferation rates in PIA lesions may be the presumed source for expanding clones that have accumulated new DNA mutations. Epithelial cells in PIA lesions characteristically exhibit several molecular signs of stress, including induced expression of GSTP1, GSTA1, and COX-2. Loss of GSTP1 expression via de novo GSTP1 CpG island hypermethylation seems to demarcate the switch from PIA to clonal, GSTP1-inactivated neoplasia in PIN or prostate adenocarcinoma. A major effort in ultimate chemoprevention of prostate cancer will be to block the epigenetic processes that drive PIA into PIN foci.

TRANSLATIONAL RESEARCH IMPLICATIONS: CHEMOPREVENTION Significant implications for chemoprevention of prostate cancer have arisen from the finding that human prostate cancer is characterized by an early and near-universal loss of expression of the phase 2 enzyme GST-π. First, on prostate biopsies that are negative for adenocarcinoma, diagnostic loss of GST-π expression in high frequency of normal appearing cells may indicate epithelium at lifetime higher risk for developing prostate cancer. Second, investigators are pursuing a mechanism-based prostate cancer preventive strategy that involves induction of phase 2 enzymes within the prostate to compensate for the loss of GSTP1 expression via dietary supplementation. NAD[P]H: (quinoneacceptor) oxidoreductase/quinone reductase enzymatic activity can be measured after treating the human prostate cancer cell lines with known phase 2 enzyme-inducing agents from distinct chemical classes. The search for antioxidant micronutrients in tomatoes and cruciferous vegetables in high throughput screening is ongoing. Given the interest in Cox-2 in chemoprevention of bowel and other tumors, the role of prostate cancer cell lines do not express basal levels of Cox-2 protein. Analysis of multiple primary human prostate cancer specimens by immunohistochemistry indicates there is no consistent overexpression of COX-2 in established prostate cancer or high-grade PIN, as compared with adjacent normal prostate tissue. Positive staining is observed only in scattered cells ( T or CC > TT transitions, the latter representing the hallmark of UV-induced mutagenesis. Melanin affords protection from UV radiation. In dark-colored skin, melanocytes accomplish a highly organized presentation of melanosomes throughout the epidermis producing an effective protective screen from damaging UV. Melanin, because it effectively absorbs both UV photons and free radicals induced by UV irradiation, protects the skin from sun damage. There are two types of melanin, eumelanin found in dark skin and hair and pheomelanin found in red hair and freckled individuals. Eumelanin synthesis is upregulated by α-melanocyte-stimulating hormone (MSH) binding to the G protein-coupled melacortin-1 receptor (MC1-R). Loss-offunction mutations of MC1-R prevent eumelanin production and are associated with most red-hair phenotypes. Nonfunctional MC1-R has been linked to enhanced cytotoxic effects of UV, as well as an increased incidence of melanoma. So why are melanocytes themselves sensitive to UV? Melanocytes may in fact respond to UV light differently than other cell types. Melanocytes produce the antiapoptotic protein B-cell leukemia/lymphoma gene (BCL-2) at high levels, which might explain the melanocytes’ relatively high resistance to apoptosis subsequent to extensive UV-induced DNA damage. Both BCL-2 and its regulator microphthalmia-associated transcription factor (MITF) have been demonstrated to be critical to melanocyte survival; deficiency of either gene in mice leads to a depletion of melanocytes. The cellular decision to undergo apoptosis or cell

754

SECTION IX / ONCOLOGY

cycle arrest and DNA repair in melanocytes may also differ from other cell types. In fibroblasts the decision seems to be via dose dependent p53-dependent induction of p21 at low UVB doses, but Bax, not p21, is induced at higher UVB doses leading to apoptosis. The elevated levels of BCL-2 in melanocytes may allow mutant melanocytes the opportunity to survive a high dose of UVB. The survival of melanocytes may serve to maintain the integrity of the skin’s sole source of melanin, ensuring uninterrupted protection from subsequent UV irradiation, but risking propagating severely damaged melanocytes and melanoma development. MELANOMA SUSCEPTIBILITY GENES It is estimated that 10% of melanoma cases report a first or second degree relative with melanoma. Two genes that confer susceptibility to melanoma have been identified within the high-risk families, CDKN2A and CDK4. Of these melanoma families the CDKN2A locus, one of the earliest and most frequently encountered defects at chromosome 9p21, accounts for susceptibility in 25–40%, whereas mutations in the CDK4 gene have been documented in far fewer kindreds. Transgenic animals deficient in the CDKN2A (NK4a/ ARF locus) develop cutaneous melanoma within a quarter of the time taken for wild-type mice to develop similar tumors. The CDKN2A gene is unique in that alternative splicing of exon 1B can result in two different proteins being transcribed in two different reading frames. The protein encoded by mRNA lacking exon 1B is the cyclin-dependent kinase inhibitor 2A, called p16/INK4a, which is involved in the retinoblastoma pathway. The protein encoded by inclusion of exon 1B of CDNK2A is called p14/ARF (alternative reading frame) and is involved in the p53 pathway. The p16/INK4a protein functions as a negative regulator of CDK4 activation by competing with cyclin D1 for CDK4 binding. Inhibition of CDK4 activity prevents phosphorylation of the retinoblastoma gene product (pRB). Phosphorylation of pRB results in the dissociation of pRB and E2F transcription factors and thus enables transcription of E2F target genes and progression of cells from G1 to S phase. Thus, p16/INK4a is considered as a tumor suppressor because it controls cell-cycle progression via this critical pathway. Loss of p16/INK4a results in hyperphosphorylation of pRB and, in turn, the uncontrolled activation of E2F-mediated transcription, resulting in excessive cellular growth. p16/INK4a may also play a role at the G2/M transition point in the cell cycle and loss of this protein may result in a lack of DNA repair following UV exposure as cells continue through successive divisions. The p14/ARF protein appears to sequester Mdm2 in the nucleus so that it is not available to bind p53, an essential step in the nuclear export and proteosomal degradation of p53. Thus, a loss of p14/ARF would accelerate p53 degradation and reduce p53mediated apoptosis. There is evidence that p14/ARF may be an independent melanoma predisposition gene. There is one report of a germline mutation that exhibited melanoma without other tumors. In addition, there is an unequivocal report of a germline deletion of p16/INK4a in the absence of concomitant loss of either CDKN2A or CDKN2B in a case of melanoma with neurilemmoma in the thoracic wall. p14/ARF may also influence the penetrance of CDKN2A mutation in mice. p14/ARF haploinsufficiency in a p16/INK4a null background suggests cooperation between p16/INK4a and p14/ARF. Mutation in the gene encoding CDK4 itself, mapping to 12q14, has also been detected in three families predisposed toward melanoma development. Interestingly, the CDK4 mutations documented in the three families all occur at codon 24, which change

arginine to either cysteine or histidine. Arginine at that position seems critical for interaction with p16/INK4a. Thus these mutations abrogate the capacity of p16/INK4a to associate with CDK4, reinforcing that loss of this inhibitory interaction as integral to melanoma susceptibility. As mentioned, UV radiation increases the incidence of melanoma. It also increases the penetrance of known melanoma susceptibility genes such as CDKN2A. These increases may simply reflect an increase in mutational burden that would lead to a higher probability of accumulating sufficient additional mutations required for cancer to develop or a more direct mechanism may exist. For example, UV radiation induces p16/INK4A expression in human skin where it is involved in UV-induced G2 phase cellcycle arrest. Mutations in the MSH receptor may also predispose toward melanoma development. Interaction of α-MSH with MSH receptor results in signaling responses that potentiate the UV-induced increase in p16 expression. In vitro experiments also indicate increased binding of p21 with cyclin E/cdk2 and cyclin A/cdk2 as well as p27 with cyclin E/cdk2 in transformed melanoma cells following exposure to UVB. Thus, the activities of cyclindependent kinases and transition through the cell cycle are key processes influenced by UV radiation in melanoma. TUMOR INITIATION AND PROGRESSION Advances in genomic technologies have afforded the identification of many melanoma associated gene expression changes. The ability of transcription factors to control gene expression through gene activation or repression can enhance tumor formation and progression, and act to promote proliferation, migration, or to inhibit apoptosis. Oncogenes are derived from genes with a role in cellular homeostasis that have undergone point mutations, amplifications, or promoter translocations that result in constitutive activation or inactivation of key regulatory pathways. The most commonly activated oncogenes in melanoma are those related to the RAS/ mitogen-activated protein kinase (MAPK) signaling pathway, which conducts cellular signaling from growth factor receptors. Oncogenic events in melanoma typically include activating mutations in N-RAS. These mutations maintain the proteins in the GTP-bound state, thus constitutively active. In the mouse model, melanocyte-specific expression of mutated Ha-RAS was carried out in an INK4a null environment resulting in mice that develop high frequency of spontaneous cutaneous melanoma. The accumulating data derived from both human and murine models suggest that RAS activation contributes not only to growth advantage but also to tumor maintenance. B-RAF is mutationally activated by point mutations in up to 60% of examined primary melanoma cell lines. B-RAF acts downstream of Ras in the MAPK pathway transducing signals from Ras through MAPK/ERK kinase to MAPK. Constitutive active B-RAF results in activation of downstream effectors that contribute to cell proliferation and inhibition of apoptosis. However, high frequency B-RAF mutation has been demonstrated in nevi, which are benign melanocytic proliferations. These data suggest that active B-RAF alone may not be sufficient to drive the transformation, but mutation in B-RAF may represent an early, perhaps prerequisite, step in the initiation of melanoma. In contrast however, B-RAF mutations were reported in only 10% of the earliest RGP melanomas, suggesting that B-RAF mutations may correlate with tumor progression and not initiation. Downstream of the RAS/MAPK is Cyclin D1, an important regulator of the G1/S transition. Acral melanomas (those occurring

CHAPTER 77 / CUTANEOUS MELANOMA

in the hands and feet), unlike cutaneous melanomas, are characterized by gene amplifications, frequently exhibiting amplifications of Cyclin D1. As mentioned, Cyclin D1 complexes with CDK4 to mediate the phosphorylation of pRB. Intriguingly, interfering with Cyclin D1 expression in a murine xenograft model of melanoma led to apoptosis and tumor regression, highlighting the role of this pathway in tumor maintenance. Gain or loss of transcription factor gene function plays a major role in melanoma progression. Abnormal expression of transcription factors can in turn modulate several genes involved in tumor growth and metastasis. Alterations in activity/expression of transcription-related genes including activator protein (AP)-2, calcium/ cAMP response element binding (CREB) protein, ATF-1, ATF-2 SNAI, MITF, and nuclear factor (NF)-κB have all been implicated to be involved in melanoma progression. The role of AP-2 in progression of melanoma has received much attention and it is now accepted that AP-2 is a major transcriptional regulator of genes involved in melanoma progression. Highly metastatic melanoma cells do not express AP-2 and transfection of AP-2 into highly metastatic melanoma cells inhibits their growth and metastatic potential in nude mice. The transition of melanomas from radial to VGP is associated with the loss of c-KIT and the overexpression of melanoma cell adhesion molecule MCAM/MUC18. Interestingly, there appears to be a direct association between cKIT and AP-2 and an inverse correlation between AP-2 and MCAM/MUC18 and both these genes have AP-2 binding sites. AP-2 also regulates other genes involved in the progression of human melanoma such as E-cadherin, p21WAF, IGFR-1, matrix metalloproteinase (MMP)-2, and vascular endothelial growth factor (VEGF). CREB expression correlates directly with metastatic potential of murine melanoma cells, and ATF is not detected in normal melanocytes but is easily found in metastatic melanoma cells. CREB/ATF-1 can act either as positive or negative regulators of transcription. CREB factors are thought to contribute to the acquisition of the metastatic phenotype in human melanoma by upregulating MMP-2 and MCAM/MUC18 expression, both of which have CRE regions in their promoters. CREB/ATF-1 may also act as survival factors for human melanoma cells. A variety of additional transcription factors may also affect melanoma progression. Overexpression of SNAIL suppresses expression of E-cadherin, thus, significantly increasing the metastatic potential of melanoma cells. MITF plays an important role in the differentiation of melanocytes and other neural crest-derived cells by regulating tyrosinase, tyrosinase-related proteins, and dopachrome tautomerase, all involved in melanin synthesis. NF-κB regulates IL-8 and VEGF in melanoma cells and so plays a significant role in the progression of human melanoma. Hence, loss of AP-2 and overexpression or hyperactivity of CREB/ATF-1, AFT-2, SNAIL, MITF, and NF-κB in metastatic melanoma cells may work in concert to regulate several genes contributing to the development of the malignant phenotype. AUTOCRINE AND PARACRINE REGULATION BY CYTOKINES AND GROWTH FACTORS IN MELANOMA Malignant melanoma cells express different growth factors and cytokines and their receptors at different stages of tumor progression, which by autocrine and paracrine effects enable them to grow autonomously and confer competence to metastasis. Autocrine growth factors (fibroblast growth factor [bFGF], melanocyte growth stimulatory activity/growth regulated protein [MGSA/GRO], IL8

755

and sometimes IL-6, platelet-derived growth factor [PDGF]-A, IL-10) produced by melanoma cells stimulate proliferation of the producing cell itself whereas paracrine growth factors (e.g., PDGF, EGF, transforming growth factor [TGF]-β, IL-1, granulocytemacrophage colony-stimulating factor [GM-CSF], IGF-1, NGF, VEGF) modulate the microenvironment to benefit tumor growth and invasion. Paracrine effects include angiogenesis, stroma formation, activation of proteolytic enzymes, adhesion or motility and metastasis formation. Some growth factors have inhibitory effects on melanocytes and early lesions (IL-1, IL-6, TGF-β, oncostatin-M [OSM] tumor necrosis factor, and interferon) but not on advanced stage melanomas and in some cases they switch to an autocrine stimulator (IL-6, TGF-β). Understanding the involvement of different growth factors and cytokines in the molecular mechanism of melanoma progression may provide insight into new therapeutic approaches. Autonomous overexpression of growth factors and their receptors are key events in melanocytic transformation. FGF-2 is a prime example of this process, whereby a normally repressed gene is activated and signals via its receptor in an autocrine manner, stimulating tumor growth. Overexpression of TGF-α, TGF-β, IGF-1, IGF-2, EGF, PDGF, and MGSA/GRO and their receptors have also been found in melanoma analyses. A survey of expression patterns from melanoma tumors revealed that increased expression of WNT5A levels correlated with increased motility and invasiveness of the melanoma cells. In a series follow-up experiments, the levels of WNT5A were demonstrated to affect the motility characteristics of cells. The WNT5A effect may also be mediated through protein kinase C (PKC), which is thought to be associated with the cell’s cytoskeletal organization, and that the PKC pathway also appeared to be downstream of the Frizzled5 (WNT5A receptor). These data may have uncovered an unexplored pathway in melanoma progression.

MELANOMA CELL ADHESION AND CELL MIGRATION Progression of cutaneous melanoma is associated with alterations in the expression profiles of many types of cell adhesion molecules. Cadherins traverse the cell membrane and form calcium-dependent homotypic or heterotypic connections with identical proteins on neighboring cells and are connected to the actin cytoskeleton via α- and β- or γ-catenins. E-cadherin is the predominant cadherin expressed on keratinocytes and nontransformed melanocytes. During the transformation of a melanocyte to a melanoma cell, expression of E-cadherin falls and is replaced by N-cadherin. The switch from E-cadherin expression to mostly N-cadherin seems to be an important step in melanoma cell transformation. Indeed, the level of N-cadherin expression seems to correlate with stage of transformation of melanoma cells and with their migratory capability. Aggressive melanoma cell lines revert back to more quiescent cells following transfection with wild-type E-cadherin. Studies using melanoma cell lines have demonstrated an upregulation of the E-cadherin transcriptional repressor SNAIL is associated with reduced E-cadherin protein levels. Melanoma cells also express hepatocyte growth factor and its receptor c-Met, which act in an autocrine manner via PI3K and MAPK signaling to downregulate both E-cadherin and desmoglein, thus, dissociating keratinocytes from melanoma cells. Furthermore, the E2A gene product can also repress E-cadherin expression during the transition from an epithelioid to a mesenchymal phenotype observed in melanoma.

756

SECTION IX / ONCOLOGY

Not only are cadherin protein levels and localization important but their degree and nature of glycosylation may also be salient to melanoma progression. N-cadherin is heavily glycosylated and different types of glycosylation have been observed in melanoma cell lines from different stages of melanoma or different sites of metastasis. In vitro experiments have also shown that altering E-cadherin glycosylation can contribute to the suppression of metastasis in a murine melanoma cell line. Normally vascular endothelial (VE)-cadherin is expressed only on endothelial cells and on placental trophoblast, both of which are exposed to circulating blood. VE-cadherin is expressed on the surfaces of melanoma cells, and cells expressing it appear to line channels within the tumor resembling blood. VE-cadherin expression and the ability to form channels has been correlated with aggressive melanoma metastasis. Clearly, strategies targeting restoration of E-cadherin expression or abrogation of N-cadherin or VE-cadherin function could have great therapeutic value. Cadherins are linked to the actin cytoskeleton via catenins. β- and γ-catenins also act as modulators of transcription via the TCF/LEF-1 transcription complex. Thus, β-catenin is an essential component of the Wnt signaling pathway and its levels within the cytoplasm are tightly controlled. Phosphorylation of β-catenin by glycogen synthase kinase an essential step in its interaction with adenomatous polyposis coli and axin and subsequent degradation by ubiquitination and the 26s proteosome pathway. Mutations in β-catenin or adenomatous polyposis coli may result in increased cytoplasmic levels of β-catenin, which can then interact with TCF/LEF-1 in the nucleus and influence target genes that facilitate tumor survival and progression. Although mutations in the CTNNB1 gene that encodes β-catenin are rare in melanoma, increase in nuclear β-catenin is frequently observed during melanoma progression. It is thought that the increases in nuclear β-catenin are likely attributable to post-translational modifications. Interestingly, the presence of serine phosphorylation of β-catenin in malignant melanoma is an indication of a poor prognosis. LEF-1 transcripts are also increased in actively migrating melanoma cells perhaps because of increased levels of nuclear β-catenin. β-catenin-LEF-1-mediated regulation of MITF may also be pertinent to melanoma cell growth. Furthermore, transcripts of ICAT, a specific inhibitor of β-catenin-TCF signaling, in malignant melanoma are low or even absent. Interestingly, the ICAT gene is located at 1p36.2, one of the loci initially identified in loss of heterozygosity studies of familial melanoma. One of the hallmarks of the progression of melanoma tumors from the nonmetastatic RGP to the potentially metastatic VGP is the downregulation of α6β1 and αvβ1 integrins and increased expression of β3-integrins. Several groups have observed a correlation between β3-integrin levels and melanoma progression. Expression of the αIIbβ3-integrin appears to enhance melanoma tumor cell survival, promoting increased adherence and migration. In addition, levels of the vitronectin receptor αvβ3 correlate directly with both tumor thickness and progression in melanoma and are indicative of a poor prognosis. αvβ3-integrin is detected in all metastatic lesions and may prove an effective target for therapeutic intervention. Activation of αvβ3-initiates signals that promote metastasis through the phosphorylation of focal adhesion kinase and activation of SRC. Integrin-linked kinase is associated with β1- and β3-integrins and its activation results in increased expression of LEF-1 and its downstream transcriptional targets, which include the E-cadherin repressor SNAIL.

Melanoma cells also express several other adhesion molecules including MUC18/MCAM, ALCAM, and L1. Upregulation of MUC18/MCAM expression correlates with E-cadherin loss and metastasis from melanoma cell tumors in nude mice. Signaling via MUC18/MCAM can lead to focal adhesion kinase activation and downstream signaling. ALCAM has been shown to be expressed in metastasizing but not in nonmetastasizing melanoma cell tumors in nude mice by differential display. Migration of melanoma cells involves, and may require, the degradation and remodeling of the extracellular matrix. MMPs and their inhibitors and components of the plasminogen activation system play leading roles in this process. Melanoma cells increase their ability to bind active MMPs that can degrade matrices in their path. Increased levels of active MMP-2 correlate with melanoma progression. Moreover, expression of MT1-MMP associated with MMP-2 activation increases tumor growth and vascularization in melanoma cells. Poorly aggressive melanoma cells do not appear to synthesize MMP-9 but do produce its inhibitor, tissue inhibitor of metalloproteinases-1. As melanoma progresses, this balance is reversed and cells derived from primary tumors increase MMP-9 synthesis whereas their production of tissue inhibitor of metalloproteinases decreases. Alendronate, a bisphosphonate, inhibits several MMPs and can reduce invasion of a melanoma cell line in vitro. Components of the plasminogen activation system can also regulate cellular migration. Immunocytochemical studies indicate that both activators and inhibitors and urokinase-type plasminogen activator receptor are upregulated in primary cutaneous melanoma. Increased uPA and urokinase-type plasminogen activator receptor immunostaining may correlate with increased invasive potential, whereas the inhibitor PAI1 tends to be enriched at the periphery of the lesion in which keratinocytes may still have some level of communication and control over neighboring melanocytes. Levels of PAI1 in particular correlate with different stages of melanoma progression and may be used prognostically. Elevated levels of protease-activated receptor (PAR)-1, the thrombin receptor, are also detected in metastatic human melanoma cells lines. Cleavage of PAR-1 by thrombin activates signaling cascades that affect levels of growth factors, integrins and MMPs and promote invasion. AP-2, a repressor acting on the PAR-1 promoter in nonmetastatic melanoma cells, is downregulated in highly metastatic melanoma cells. Malignant melanoma cells also express tissue factor, which can activate thrombin, which may act as a growth factor in human melanoma cells. METASTASIS Transendothelial migration is an essential step in the metastasis of melanoma. It involves adhesion to and then migration (diapedesis) across the endothelial monolayer. Many factors that are upregulated in melanoma tumors play a role in melanocyte chemotaxis and adhesion to endothelium or communication between melanomal and endothelial cells in vitro. Endothelial cells secrete chemokines that induce melanoma cell chemotaxis. One of the most crucial of these is IL-8, which acts through the CXCR1 receptor on melanoma cells. There is also evidence that interactions between melanoma cells and polymorphonuclear cells play a role in the adhesion of both cell types to the endothelium before transendothelial migration. As melanoma cells adhere to the endothelium, levels of VE-cadherin and PECAM between adjacent endothelial cells are diminished allowing melanoma cells to send membrane protrusions between endothelial cells. Melanoma cells appear to adhere to endothelium

CHAPTER 77 / CUTANEOUS MELANOMA

via N-cadherin, αvβ3-integrin, cell adhesion molecule L1, and MCAM/ MUC18. Furthermore, several members of the tetraspanin superfamily (CD9, CD81, and CD151) localize to these heterotypic points of contact in two- and three-dimensional cocultures of melanoma cells and endothelial cells, and antibodies against CD9 inhibit melanoma cell transmigration.

TUMOR PERFUSION Tumor growth requires adequate perfusion of blood and an ever-increasing network of vessels to convey that blood. Microvascular density within tumors is prognostically significant in the staging and diagnosis of melanoma. In general, histological examinations of highly aggressive melanoma samples show a greater density of microvessels than nonaggressive tumors. Doppler studies also indicate that greater blood flow is associated with poorer outcome. Thus, angiogenesis is considered one of the key targets for therapeutic intervention. ANGIOGENESIS Melanoma cells synthesize a multitude of proteins, which are themselves angiogenic or cleave other proteins to produce factors that can influence angiogenesis. Levels of two well-characterized angiogenic cytokines, VEGF and FGF-2, are constitutively higher in transformed melanocytes than in normal melanocytes, and elevated serum levels of VEGF and FGF-2 are strongly correlated with poor overall prognosis of melanoma. Invasive melanomas also display higher levels of immunostaining for VEGF when compared with noninvasive tumors. Although VEGF is primarily angiogenic, FGF-2 also has an autocrine role in melanoma cell survival and proliferation; both factors are, therefore, required to maintain tumor growth. The three isoforms of MGSA/GRO are also known to stimulate angiogenesis, whereas experiments in mouse models have demonstrated significant roles for endothelial cell-selective adhesion molecule and IL-1β, IL-1α and the IL-1 receptor in the angiogenic response in melanoma cellderived tumors. Neovascularization in melanoma is also achieved by reducing key antiangiogenic molecules. Indeed, invasion of both melanoma cells and endothelial cells is abrogated by angiostatin treatment. VASCULOGENIC MIMICRY “Vasculogenic mimicry” is the phrase coined to describe the apparent formation of a network of channels within metastatic cutaneous lesions, which resemble the vasculogenic networks formed by differentiating endothelial cells in the embryo. This is a fascinating example of the plasticity of aggressive melanoma cells and constitutes an angiogenesisindependent pathway of facilitating tumor perfusion and preventing necrosis. Evidence of vasculogenic mimicry in tumors has a most unfavorable prognosis. These channels are not lined by endothelial cells, yet erythrocytes and plasma can be detected within them raising the possibility that some anastomoses may occur between both endothelia-lined and nonendothelia-lined “vessels.” They are formed by aggressive melanoma cells expressing VE-cadherin. Downregulation of VEcadherin abrogates the ability of melanoma cells to form them. Introduction of aggressive melanoma cells into an ischemic environment (the hindlimb of a nude mouse) results in the expression of VE-cadherin and other endothelial associated genes and cell fate determination molecules reminiscent of embryonic vasculogenesis. Melanoma cells capable of engaging in vasculogenic mimicry express several endothelial-associated growth factors and their receptors including EphA2 and CD34. Notch 4 is also highly expressed by malignant melanoma cells undergoing neovascularization, and

757

inhibiting tumor vasculogenic mimicry is relatively unaffected by endostatin and, thus, any therapy targeting tumor perfusion must also consider the characteristics of these nonendothelial lined channels. MMP-2 and MT1-MMP-dependent cleavage of the laminin5γ2 chain produced fragments that could induce vasculogenic mimicry in poorly aggressive tumor cells in vitro if grown on an aggressive tumor cell conditioned matrix. Interestingly, the application of chemically modified tetracyclins, which inhibit MMP activity, also inhibits generation of fragments of laminin-5γ2 chain and the induction of vasculogenic mimicry-associated genes in less aggressive cell lines. Clearly, the microenvironment plays a significant role in neovascularization and targeting factors in the environment of aggressive melanomas may be a useful therapeutic adjunct.

SELECTED REFERENCES Baldi A, Santini D, Battista T, et al. Expression of AP-2 transcription factor and of its downstream target genes c-kit, E-cadherin and p21 in human cutaneous melanoma. J Cell Biochem 2001;83(3):364–372. Bar-Eli M. Gene regulation in melanoma progression by the AP-2 transcription factor. Pigment Cell Res 2001;14(2):78–85. Bastian BC. Understanding the progression of melanocytic neoplasia using genomic analysis: from fields to cancer. Oncogene 2003;22(20): 3081–3086. Bennett DC. Human melanocyte senescence and melanoma susceptibility genes. Oncogene 2003;22(20):3063–3069. Bishop DT, Demenais F, Goldstein AM, et al. Geographical variation in the penetrance of CDKN2A mutations for melanoma. J Natl Cancer Inst 2002;94(12):894–903. Bittner M, Meltzer P, Chen Y, et al. Molecular classification of cutaneous malignant melanoma by gene expression profiling. Nature 2000;406 (6795):536–540. Chin L, Pomerantz J, Polsky D, et al. Cooperative effects of INK4a and ras in melanoma susceptibility in vivo. Genes Dev 1997;11(21):2822–2834. Ciolczyk-Wierzbicka D, Gil D, Hoja-Lukowicz D, Litynska A, Laidler P. Carbohydrate moieties of N-cadherin from human melanoma cell lines. Acta Biochim Pol 2002;49(4):991–998. Davies H, Bignell GR, Cox C, et al. Mutations of the BRAF gene in human cancer. Nature 2002;417(6892):949–954. Degen WG, van Kempen LC, Gijzen EG, et al. MEMD, a new cell adhesion molecule in metastasizing human melanoma cell lines, is identical to ALCAM (activated leukocyte cell adhesion molecule). Am J Pathol 1998;152(3):805–813. Demunter A, Libbrecht L, Degreef H, De Wolf-Peeters C, van den Oord JJ. Loss of membranous expression of beta-catenin is associated with tumor progression in cutaneous melanoma and rarely caused by exon 3 mutations. Mod Pathol 2002;15(4):454–461. Dong J, Phelps RG, Qiao R, et al. BRAF oncogenic mutations correlate with progression rather than initiation of human melanoma. Cancer Res 2003;63(14):3883–3885. Dupin E, Le Douarin NM. Development of melanocyte precursors from the vertebrate neural crest. Oncogene 2003;22(20):3016–3023. Dupin E, Real C, Glavieux-Pardanaud C, Vaigot P, Le Douarin NM. Reversal of developmental restrictions in neural crest lineages: Transition from Schwann cells to glial-melanocytic precursors in vitro. Proc Natl Acad Sci USA 2003;100(9):5229–5233. Ferrier CM, van Muijen GN, Ruiter DJ. Proteases in cutaneous melanoma. Ann Med 1998;30(5):431–442. Folberg R, Hendrix MJ, Maniotis AJ. Vasculogenic mimicry and tumor angiogenesis. Am J Pathol 2000;156(2):361–381. Gruss C, Herlyn M. Role of cadherins and matrixins in melanoma. Curr Opin Oncol 2001;13(2):117–123. Halaban R. The regulation of normal melanocyte proliferation. Pigment Cell Res 2000;13(1):4–14. Hayward NK. Genetics of melanoma predisposition. Oncogene 2003; 22(20):3053–3062. Heikkila P, Teronen O, Moilanen M. Bisphosphonates inhibit stromelysin1 (MMP-3), matrix metalloelastase (MMP-12), collagenase-3 (MMP-

758

SECTION IX / ONCOLOGY

13) and enamelysin (MMP-20), but not urokinase-type plasminogen activator, and diminish invasion and migration of human malignant and endothelial cell lines. Anticancer Drugs 2002;13(3):245–254. Hendrix MJ, Seftor EA, Hess AR, Seftor RE. Molecular plasticity of human melanoma cells. Oncogene 2003;22(20):3070–3075. Hendrix MJ, Seftor EA, Hess AR, Seftor RE. Vasculogenic mimicry and tumour-cell plasticity: lessons from melanoma. Nat Rev Cancer 2003; 3(6):411–421. Hendrix MJ, Seftor EA, Meltzer PS, et al. Expression and functional significance of VE-cadherin in aggressive human melanoma cells: role in vasculogenic mimicry. Proc Natl Acad Sci USA 2001;98(14): 8018–8023. Herlyn M, Berking C, Li G, Satyamoorthy K. Lessons from melanocyte development for understanding the biological events in naevus and melanoma formation. Melanoma Res 2000;10(4):303–312. Hersey P, Zhang XD. How melanoma cells evade trail-induced apoptosis. Nat Rev Cancer 2001;1(2):142–150. Hess AR, Seftor EA, Gardner LM, et al. Molecular regulation of tumor cell vasculogenic mimicry by tyrosine phosphorylation: role of epithelial cell kinase (Eck/EphA2). Cancer Res 2001;61(8):3250–3255. Hsu MY, Meier F, Herlyn M. Melanoma development and progression: a conspiracy between tumor and host. Differentiation 2002;70(9–10): 522–536. Hsu MY, Meier FE, Nesbit M, et al. E-cadherin expression in melanoma cells restores keratinocyte-mediated growth control and down-regulates expression of invasion-related adhesion receptors. Am J Pathol 2000; 156(5):1515–1525. Ichihashi M, Ueda M, Budiyanto A, et al. UV-induced skin damage. Toxicology 2003;189(1–2):21–39. Jean D, Gershenwald JE, Huang S, et al. Loss of AP-2 results in up-regulation of MCAM/MUC18 and an increase in tumor growth and metastasis of human melanoma cells. J Biol Chem 1998;273(26):16,501–16,508. Jhappan C, Noonan FP, Merlino G. Ultraviolet radiation and cutaneous malignant melanoma. Oncogene 2003;22(20):3099–3112. Johnson JP. Cell adhesion molecules in the development and progression of malignant melanoma. Cancer Metastasis Rev 1999;18(3):345–357. Kamada S, Shimono A, Shinto Y, et al. BCL-2 deficiency in mice leads to pleiotropic abnormalities: accelerated lymphoid cell death in thymus and spleen, polycystic kidney, hair hypopigmentation, and distorted small intestine. Cancer Res 1995;55(2):354–359. Kannan K, Sharpless NE, Xu J, O’Hagan RC, Bosenberg M, Chin L. Components of the Rb pathway are critical targets of UV mutagenesis in a murine melanoma model. Proc Natl Acad Sci USA 2003;100(3): 1221–1225. Krimpenfort P, Quon KC, Mooi WJ, Loonstra A, Berns A. Loss of p16Ink4a confers susceptibility to metastatic melanoma in mice. Nature 2001; 413(6851):83–86. Kurschat P, Mauch C. Mechanisms of metastasis. Clin Exp Dermatol 2000;25(6):482–489. Li G, Ho VC. p53-dependent DNA repair and apoptosis respond differently to high- and low-dose ultraviolet radiation. Br J Dermatol 1998;139(1):3–10. Li G, Satyamoorthy K, Herlyn M. N-cadherin-mediated intercellular interactions promote survival and migration of melanoma cells. Cancer Res 2001;61(9):3819–3825. Li G, Satyamoorthy K, Meier F, Berking C, Bogenrieder T, Herlyn M. Function and regulation of melanoma-stromal fibroblast interactions: when seeds meet soil. Oncogene 2003;22(20):3162–3171. Li G, Schaider H, Satyamoorthy K, Hanakawa Y, Hashimoto K, Herlyn M. Downregulation of E-cadherin and Desmoglein 1 by autocrine hepatocyte growth factor during melanoma development. Oncogene 2001;20(56):8125–8135. Longo N, Yanez-Mo M, Mittelbrunn M, et al. Regulatory role of tetraspanin CD9 in tumor-endothelial cell interaction during transendothelial invasion of melanoma cells. Blood 2001;98(13):3717–3726. Luca MR, Bar-Eli M. Molecular changes in human melanoma metastasis. Histol Histopathol 1998;13(4):1225–1231. MacDougall JR, Bani MR, Lin Y, Muschel RJ, Kerbel RS. ‘Proteolytic switching’: opposite patterns of regulation of gelatinase B and its inhibitor TIMP-1 during human melanoma progression and consequences of gelatinase B overexpression. Br J Cancer 1999;80(3–4): 504–512.

Maniotis AJ, Folberg R, Hess A, et al. Vascular channel formation by human melanoma cells in vivo and in vitro: vasculogenic mimicry. Am J Pathol 1999;155(3):739–752. McGary EC, Lev DC, Bar-Eli M. Cellular adhesion pathways and metastatic potential of human melanoma. Cancer Biol Ther 2002;1(5):459–465. McGill GG, Horstmann M, Widlund HR, et al. BCL-2 regulation by the melanocyte master regulator MITF modulates lineage survival and melanoma cell viability. Cell 2002;109(6):707–718. Meier F, Satyamoorthy K, Nesbit M, et al. Molecular events in melanoma development and progression. Front Biosci 1998;3:D1005–D1010. Mollaaghababa R, Pavan WJ. The importance of having your SOX on: role of SOX10 in the development of neural crest-derived melanocytes and glia. Oncogene 2003;22(20):3024–3034. Nakayama K, Negishi I, Negishi I, Kuida K, Sawa H, Loh DY. Targeted disruption of BCL-2 alpha beta in mice: Occurrence of gray hair, polycystic kidney disease, and lymphocytopenia. Proc Natl Acad Sci USA 1994;91(9):3700–3704. Nierodzik ML, Chen K, Takeshita K, et al. Protease-activated receptor 1 (PAR-1) is required and rate-limiting for thrombin-enhanced experimental pulmonary metastasis. Blood 1998;92(10):3694–3700. Nyormoi O, Bar-Eli M. Transcriptional regulation of metastasis-related genes in human melanoma. Clin Exp Metastasis 2003;20(3):251–263. Pavey S, Gabrielli B. Alpha-melanocyte stimulating hormone potentiates p16/CDKN2A expression in human skin after ultraviolet irradiation. Cancer Res 2002;62(3):875–880. Pavey S, Russell T, Gabrielli B. G2 phase cell cycle arrest in human skin following UV irradiation. Oncogene 2001;20(43):6103–6110. Perez-Moreno MA, Locascio A, Rodrigo I, et al. A new role for E12/E47 in the repression of E-cadherin expression and epithelial-mesenchymal transitions. J Biol Chem 2001;276(29):27,424–27,431. Pollock PM, Harper UL, Hansen KS, et al. High frequency of BRAF mutations in nevi. Nat Genet 2003;33(1):19–20. Pollock PM, Hayward N. Mutations in exon 3 of the beta-catenin gene are rare in melanoma cell lines. Melanoma Res 2002;12(2):183–186. Polsky D, Cordon-Cardo C. Oncogenes in melanoma. Oncogene 2003;22(20):3087–3091. Poser I, Dominguez D, de Herreros AG, Varnai A, Buettner R, Bosserhoff AK. Loss of E-cadherin expression in melanoma cells involves up-regulation of the transcriptional repressor Snail. J Biol Chem 2001;276(27):24, 661–24,666. Ramjeesingh R, Leung R, Siu CH. Interleukin-8 secreted by endothelial cells induces chemotaxis of melanoma cells through the chemokine receptor CXCR1. FASEB J 2003;17(10):1292–1294. Randerson-Moor JA, Harland M, Williams S, et al. A germline deletion of p14(ARF) but not CDKN2A in a melanoma-neural system tumour syndrome family. Hum Mol Genet 2001;10(1):55–62. Reifenberger J, Knobbe CB, Wolter M, et al. Molecular genetic analysis of malignant melanomas for aberrations of the WNT signaling pathway genes CTNNB1, APC, ICAT and BTRC. Int J Cancer 2002;100(5):549–556. Reinke V, Lozano G. Differential activation of p53 targets in cells treated with ultraviolet radiation that undergo both apoptosis and growth arrest. Radiat Res 1997;148(2):115–122. Rizos H, Puig S, Badenas C, et al. A melanoma-associated germline mutation in exon 1beta inactivates p14ARF. Oncogene 2001;20(39): 5543–5547. Sanders DS, Blessing K, Hassan GA, Bruton R, Marsden JR, Jankowski J. Alterations in cadherin and catenin expression during the biological progression of melanocytic tumours. Mol Pathol 1999;52(3):151–157. Sandig M, Voura EB, Kalnins VI, Siu CH. Role of cadherins in the transendothelial migration of melanoma cells in culture. Cell Motil Cytoskeleton 1997;38(4):351–364. Sauter ER, Yeo UC, von Stemm A, et al. Cyclin D1 is a candidate oncogene in cutaneous melanoma. Cancer Res 2002;62(11):3200–3206. Seftor EA, Meltzer PS, Schatteman GC, et al. Expression of multiple molecular phenotypes by aggressive melanoma tumor cells: role in vasculogenic mimicry. Crit Rev Oncol Hematol 2002;44(1):17–27. Seftor RE, Seftor EA, Hendrix MJ. Molecular role(s) for integrins in human melanoma invasion. Cancer Metastasis Rev 1999;18(3):359–375.

CHAPTER 77 / CUTANEOUS MELANOMA

Seftor RE, Seftor EA, Koshikawa N, et al. Cooperative interactions of laminin 5 gamma2 chain, matrix metalloproteinase-2, and membrane type-1-matrix/metalloproteinase are required for mimicry of embryonic vasculogenesis by aggressive melanoma. Cancer Res 2001;61(17): 6322–6327. Sharpless E, Chin L. The INK4a/ARF locus and melanoma. Oncogene 2003;22(20):3092–3098. Simonetti O, Lucarini G, Brancorsini D, et al. Immunohistochemical expression of vascular endothelial growth factor, matrix metalloproteinase 2, and matrix metalloproteinase 9 in cutaneous melanocytic lesions. Cancer 2002;95(9):1963–1970. Slattery MJ, Dong C. Neutrophils influence melanoma adhesion and migration under flow conditions. Int J Cancer 2003;106(5):713–722. Slominski A, Wortsman J, Carlson AJ, Matsuoka LY, Balch CM, Mihm MC. Malignant melanoma. Arch Pathol Lab Med 2001;125(10):1295–1306. Smalley KS. A pivotal role for ERK in the oncogenic behaviour of malignant melanoma? Int J Cancer 2003;104(5):527–532. Soufir N, Avril MF, Chompret A, et al. Prevalence of p16 and CDK4 germline mutations in 48 melanoma-prone families in France. The French Familial Melanoma Study Group. Hum Mol Genet 1998;7(2):209–216. Streit M, Detmar M. Angiogenesis, lymphangiogenesis, and melanoma metastasis. Oncogene 2003;22(20):3172–3179. Suzuki I, Im S, Tada A, et al. Participation of the melanocortin-1 receptor in the UV control of pigmentation. J Investig Dermatol Symp Proc 1999;4(1):29–34. Takeda K, Yasumoto K, Takada R, et al. Induction of melanocyte-specific microphthalmia-associated transcription factor by Wnt-3a. J Biol Chem 2000;275(19):14,013–14,016.

759

Tellez C, Bar-Eli M. Role and regulation of the thrombin receptor (PAR-1) in human melanoma. Oncogene 2003;22(20):3130–3137. Tucker MA, Goldstein AM. Melanoma etiology: where are we? Oncogene 2003;22(20):3042–3052. Voronov E, Shouval DS, Krelin Y, et al. IL-1 is required for tumor invasiveness and angiogenesis. Proc Natl Acad Sci USA 2003;100(5):2645– 2650. Voura EB, Sandig M, Siu CH. Cell–cell interactions during transendothelial migration of tumor cells. Microsc Res Tech 1998;43(3):265–275. Walker GJ, Hayward NK. Pathways to melanoma development: lessons from the mouse. J Invest Dermatol 2002;119(4):783–792. Weeraratna AT, Jiang Y, Hostetter G, et al. Wnt5a signaling directly affects cell motility and invasion of metastatic melanoma. Cancer Cell 2002; 1(3):279–288. Widlund HR, Horstmann MA, Price ER, et al. Beta-catenin-induced melanoma growth requires the downstream target Microphthalmiaassociated transcription factor. J Cell Biol 2002;158(6):1079–1087. Wu C, Dedhar S. Integrin-linked kinase (ILK) and its interactors: a new paradigm for the coupling of extracellular matrix to actin cytoskeleton and signaling complexes. J Cell Biol 2001;155(4):505–510. Yang FC, Merlino G, Chin L. Genetic dissection of melanoma pathways in the mouse. Semin Cancer Biol 2001;11(3):261–268. Yoshimura M, Ihara Y, Matsuzawa Y, Taniguchi N. Aberrant glycosylation of E-cadherin enhances cell–cell binding to suppress metastasis. J Biol Chem 1996;271(23):13,811–13,815. Zuo L, Weger J, Yang Q, et al. Germline mutations in the p16INK4a binding domain of CDK4 in familial melanoma. Nat Genet 1996;12(1): 97–99.

78 Applications of Gene Expression Profiling to the Study of Malignant Gliomas ANTHONY T. YACHNIS AND HENRY V. BAKER SUMMARY Malignant central nervous system tumors that originate from brain tissue or from the brain’s coverings cause significant morbidity and mortality. Tumors of the glioma group are the most common primary brain tumors in adults and the most important group of diffusely infiltrating gliomas are the astrocytomas and oligodendrogliomas. Most gliomas are incurable because they infiltrate brain tissue, making it difficult to completely remove the tumor without damaging normal structures. Because of this, recurrences of such tumors are common, despite standard therapeutic efforts. Key Words: Anaplastic astrocytoma; anaplastic oligodendroglioma; astrocytomas; brain tumor; diffuse astrocytoma; epidermal growth factor receptor; glioblastoma multiforme; glioma; mixed anaplastic oligoastrocytoma; mixed oligoastrocytoma; oligodendrogliomas.

INTRODUCTION Malignant central nervous system tumors that originate from brain tissue or from the brain’s coverings (i.e., primary tumors) cause significant morbidity and mortality. Each year 7–10 new cases of primary intracranial tumors are diagnosed per 100,000 population. Tumors of the glioma group are the most common primary brain tumors in adults and the most important group of diffusely infiltrating gliomas are the astrocytomas and oligodendrogliomas. The biological behavior of such tumors varies from slowly growing indolent lesions that may produce symptoms over many years to aggressive, rapidly growing neoplasms that can cause death within a year of diagnosis. Most gliomas are incurable because they infiltrate brain tissue, making it difficult to completely remove the tumor without damaging normal structures. Because of this, recurrences of such tumors are common, despite standard therapeutic efforts (usually radiation and/or chemotherapy). Histology is the gold standard by which gliomas are classified. The following subtypes are recognized by the World Health Organization (WHO): diffuse astrocytoma, WHO grade II; oligodendroglioma, WHO grade II; mixed oligoastrocytoma, WHO

From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

grade II; anaplastic oligodendroglioma, WHO grade III (AO); anaplastic astrocytoma, WHO grade III (AA); mixed anaplastic oligoastrocytoma, WHO grade III (MAOA); and glioblastoma multiforme, WHO grade IV (GBM). Histopathological grading has provided clinically useful information regarding prognosis for groups of patients with diagnoses of glioma. Figure 78-1 shows the mean life expectancy after diagnosis of patients with the above subtypes of glioma. However, variability in outcome for individual patients with the same pathological diagnosis is often observed clinically and this is a significant problem in managing patients with diffuse gliomas. The biological behavior of gliomas (and patient prognosis) correlates with certain microscopic characteristics (i.e., tumor grade) such as pleomorphism, mitoses, vascular endothelial proliferation, and tumor necrosis. However, histopathological diagnosis is highly subjective with significant interobserver variability depending on the pathologist’s interpretation of or weight given to each histological characteristic. Although assessment of cell proliferation indices (e.g., Ki67 labeling index by immunohistochemistry) for gliomas has enhanced the ability to predict tumor behavior (rapid growth vs slow growth), such analysis is most helpful in distinguishing low (grade II) from intermediate grade (III) tumors and does not distinguish prognoses for patients with high-grade gliomas (grade IV). Most importantly, outcomes for individual patients cannot be accurately predicted by any brain tumor grading system. Also, few diagnostic markers are available for use in distinguishing neoplastic from normal cells and there are no reliable markers that distinguish glioma subtypes or grades of malignancy.

MOLECULAR CHARACTERIZATION OF BRAIN TUMORS Molecular characterization of brain tumors has already proven useful in predicting response to therapy for a particular type of glioma, the AO (WHO grade III). In particular, subtypes of this tumor that have mutations (i.e., loss of heterozygosity) of chromosomes 1p and 19q have a better overall prognosis and a better response to chemotherapy with procarbazine, 1-(2-choloroethyl)3-cyclohexyl-1-nitrosurea, and vincristine. However, not all AOs have 1p/19q deletions and not all tumors with 1p/19q losses respond to chemotherapy. Also, the success of chemotherapy with

760

CHAPTER 78 / APPLICATIONS OF GENE EXPRESSION PROFILING TO THE STUDY OF MALIGNANT GLIOMAS

Figure 78-1 Life expectancy means after diagnosis of malignant glioma according to subtype. Oligodendroglioma, WHO grade II (O); diffuse astrocytoma, WHO grade II (A); mixed oligoastrocytoma, WHO grade II (MOA); anaplastic oligodendroglioma, WHO grade III (AO); mixed anaplastic oligoastrocytoma, WHO grade III (MAOA); anaplastic astrocytoma, WHO grade III (AA); and glioblastoma multiforme, WHO grade IV (GBM).

some AOs has created increased pressure among pathologists to make this diagnosis in any malignant glioma with histology even remotely suggestive of an oligodendroglial phenotype. Functional genomics using DNA microarray technology is a major research tool in biology, particularly in the study of cancer biology. DNA microarray technology allows simultaneous analysis of the expression patterns of thousands of genes at one time. Therefore, instead of analyzing the expression state of a few genes using immunohistochemical markers, microarray technology allows for the analysis of literally tens of thousands of genes. Identifying genes whose altered expression is highly correlated with the malignant state will allow for better diagnostics simply by increasing the number of markers that signify the disease state. Furthermore, identification of genes that are aberrantly expressed in cancer cells will bring these genes into sharp focus for future studies aimed at understanding their role in the disease process. Gene expression profiles obtained from DNA microarray experiments and analyzed by various forms of statistics have already provided proof of principle of the utility of gene expression profiling in the analysis of various forms of cancer including gliomas. Early investigations utilizing microarray technology demonstrated the promise and potential of this new technology as a major research tool for cancer biology and as a clinical tool for the diagnosis and prognosis of individual tumors. Unfortunately some early studies also served to illustrate potential pitfalls associated with improper experimental methodologies and inadequate or faulty computational analysis. Key to the analysis and interpretation of microarray data is the understanding that the parameter being measured, signal intensity of indirectly labeled probes, is many steps removed from the parameter being inferred, gene expression. A typical microarray experiment represents a largescale physiological study in which cells are isolated, RNA is harvested, and labeled representations of the harvested RNA are prepared, and then used in hybridization experiments to indirectly label the nucleic acid probes constituting the array. Figure 78-2 shows the hybridization signal intensities of two Affymetrix GeneChips, one hybridized with material derived from an AO and the other from a GBM. The signal intensity of the label at each probe on the array is taken as a measure of gene expression for the genes specified by the probes. It is also important to realize that the

761

inferred gene expression is not that of a single cell but rather that of a population of cells. In some cases the population of cells under investigation is made up of many different cell types, each of which may have varied expression profiles unique to themselves. Microarray assays are sensitive, albeit indirect, assays capable of measuring the genomic response to both intracellular and extracellular changes in the environment that occur on and after neoplastic transformation. However, they are also capable of measuring gene expression changes resulting from uncontrolled variables that may occur at any step in the wet laboratory workup of specimens. Gene expression differences in response to uncontrolled experimental variables encountered during microarray experiments may add noise to resulting dataset and in some cases may be greater than differences associated with the pathophysiological state of the specimens. The extreme sensitivity of the microarray assay to uncontrolled experimental variables necessitates that replicate observations are performed before general inferences can be drawn regarding the global reprogramming that occur on transformation, or gene expression differences that distinguish one type and stage of cancer from another. Despite the limitations of the assay, with care it is possible to use microarrays as a tool to begin to discern the dynamic changes within cells undergoing neoplastic transformation and to probe different grades and classes of tumors for differences in gene expression patterns that may have relevance to the pathobiology of the disease. Microarray studies have been used for the purposes of class discovery and class prediction. Class discovery studies aim to identify new subclasses of tumor based on differences in gene expression patterns. In class prediction studies the initial goal is to identify a classifier made up of a collection of genes whose expression pattern can be used to predict a class of tumor, in which class may refer to any of a number of phenotypes such as tumor diagnosis or predicting which tumors will respond to therapy. The classifier may be made up of a few to more than a thousand genes. Many studies start as class discovery exercises and mature into ones with class prediction goals. Class discovery studies typically utilize unsupervised methods of analysis; whereas class prediction studies require supervised methods of analysis. Unsupervised methods of analysis, including principal component analysis, hierarchical clustering, k-means clustering, and self-organizing maps, can be used as tools for class discovery. Multidimensional scaling, a form of principal component analysis, has been used to separate high-grade GBMs from lower-grade astrocytomas (grades II and III) and oligodendrogliomas. Similarly hierarchical clustering can be used to distinguish highand low-grade gliomas. Figure 78-3 shows an example of unsupervised hierarchical cluster analysis of 31 tumors and 3 non-neoplastic brain (NNB) tissue specimens; tumors of like diagnosis tend to cluster together based on the gene expression patterns observed. It is important to note that the clustering pattern was derived from the inherent gene expression profiles of the tissues and not based on prior knowledge of their diagnosis. In situations where standard methods of assigning class labels are incomplete or inadequate, unsupervised class discovery methods are especially powerful. One goal of unsupervised analyses is to identify similarities between specimens heretofore unrecognized. Unsupervised approaches to determine differences in gene expression profiles among disease states have limitations that can be circumvented by the use of supervised learning methods.

762

SECTION IX / ONCOLOGY

Figure 78-2 Apparent gene expression patterns of malignant gliomas detected by Affymetrix Hu95A GeneChips. (A) Scanned image of an Affymetrix Hu95A GeneChip hybridized with material derived from an anaplastic oligodendroglioma; (B) scanned image of an Affymetrix Hu95A GeneChip hybridized with material derived from a glioblastoma multiforme.

Supervised approaches include class information such as diagnosis, divergence in clinical course, and responsiveness to therapy in identifying gene expression differences between the specimens. Because supervised analyses employ statistical tests involving analysis of variance to identify genes differentially expressed between classes, it is necessary to have multiple specimens of each class under investigation. In many cases, supervised analyses are used for purposes of identifying a classifier of probes or genes that can be used for class prediction; for example, to diagnose and differentiate diseased from normal tissue. In this case the investigators’ goal is to identify genes whose collective expression patterns are predictive of the class labels. Once identified these gene expression patterns can be used to identify the nature of the specimen as normal or diseased using one or more of several prediction models. However, microarray experiments exemplify the “small n large p” trap. The number of probes on a typical microarray, tens of thousands, vastly exceeds the number of categories into which the arrays (tumors) can be classified. Thus, by chance alone therefore, it is likely that a few too many probes can be identified out of a typical dataset that seemingly can distinguish between the small numbers of class labels (tumor types) in a typical study. Cross-validation studies and Monte Carlo simulations may be employed to gauge the significance of the probes identified as predictors. As studies mature the best method of validating a classifier is not through cross-validation studies, but prospectively in which the classifier is trained on one set of tumors, used to predict the class of a new set of tumors that was not used during the training phase of the study. Many reports must be regarded as preliminary because of the small number of specimens from which the datasets were derived and the lack of cross-validation or prospective validation of the findings. Methods for the diagnosis of infiltrating gliomas utilize a variety of modalities including clinical history and radiographic findings,

in addition to histology and immunohistochemistry of biopsy specimens. Even among highly trained neuropathologists, the diagnosis of specific stages of glioma can be difficult and somewhat subjective. Care should be taken when supervised learning approaches are applied to diagnostically difficult gliomas because the supervised analysis can only be as good as the supervision applied. In an effort to increase precision and to reduce error resulting from misdiagnosis, at least two neuropathologists should review diagnostically difficult gliomas independently before they are included as specimens for supervised learning approaches. One of the overarching goals in the application of microarray technology to the study of malignant tumors is to derive a list of sentinel genes whose expression levels may be used to predict the disease’s clinical course and the tumors’ responsiveness to therapy. Several studies have demonstrated the ability of gene expression patterns to predict differences in clinical course. The first study to do so was reported for lymphomas. For brain malignancies, the first used gene expression profiles to predict the clinical course of embryonal tumors. In the case of gliomas, gene expression profiling has been applied to create a classifier for distinguishing GBMs from AOs. A 19 gene classifier was derived using 14 classic GBM specimens and 7 classic AO specimens. Using the classifier with a five nearest neighbor prediction model (5 k-NN) and leave-one-out-cross-validation, the classifier had an accuracy rate of 86%. When the classifier was applied prospectively to nonclassic GBMs and nonclassic AOs, which are diagnostically difficult based on traditional pathology methods, the classifier gave significantly better separation of the tumors based on survival curves than did classification based on traditional pathology. Accurate predictions of a tumor’s aggressiveness and responsiveness to therapy will aid physicians and patients in choosing the most appropriate treatment options. Supervised analyses are also being used to identify gene expression differences among individuals within a given diagnosis

CHAPTER 78 / APPLICATIONS OF GENE EXPRESSION PROFILING TO THE STUDY OF MALIGNANT GLIOMAS

Figure 78-3 Unsupervised hierarchical cluster analysis of infiltrating gliomas and several other selected specimens. The image shows the gene expression patterns of 4658 genes whose expression varied most across the dataset. The hierarchical clustering dendrogram on top of the figure is derived from the Pearson’s correlations and represents the similarity of the gene expression among the tissue samples. O, oligodendroglioma, WHO grade II; AO, anaplastic oligodendroglioma, WHO grade III; MAOA, mixed anaplastic oligoastrocytoma, WHO grade III; AA, anaplastic astrocytoma, WHO grade III; GBM, glioblastoma multiforme, WHO grade IV; Lymp, lymphoma; Epen, ependymoma; and NNB, non-neoplastic brain tissue.

that have differing clinical courses. The precise strategy and the training sets used in supervised analyses of this nature depend on

763

the specific characteristics of the dataset. For analyses of these types it is necessary to take into account obvious covariates such as sex and age. Once genes are identified whose expression levels correlate with clinical course within a given diagnosis of glioma, it will be necessary to validate the ability of those genes to act as a classifier capable of predicting clinical course and outcome. Supervised learning approaches have been used to identify differences between GBMs that differ in their epidermal growth factor receptor (EGFR) status. Overexpression of EGFR has been associated with primary GBMs, but is rarely observed in secondary GBMs that arise from lower grade gliomas. Ninety genes were identified that could be used as classifiers of EGFR status of the GBMs in the study. Stratifications of GBMs based on EGFR status did not correlate with significant differences in survival; however, the study was small and may not have been adequately powered to detect such differences. Multiclass classifiers can also be derived capable of differentiation among several grades of glioma and NNB tissue. Figure 78-4 shows the hierarchal cluster pattern of a 1071 feature classifier derived from 26 specimens capable of differentiating between AO, mixed anaplastic oligoastrocytoma, GBM, and NNB with an accuracy rate ranging between 80 and 88% when tested prospectively on 25 new tumor specimens using four prediction models including 1 k-NN, 3 k-NN, nearest centroid, and linear discriminant analysis. At the basic science level, gene expression analysis can provide insights into the pathobiology of the disease process. In comparisons between GBM and NNB tissue, there is increased expression of genes encoding members of several signal transduction pathways, cell cycle genes, genes encoding components of the basement membrane, collagen binding proteins in addition to inflammatory response genes among other families of genes; whereas in NNB tissue there is elevated expression, compared to GBM tumor tissue, of genes involved in conducting normal brain functions such as genes encoding neurofilaments, and components of synaptic vesicles among others. Although these results are not unexpected, finding genes with elevated expression in GBMs known to play a role in cell migration and tumor genesis serves to validate the approach. The information content of the datasets generated is tremendous and has yet to be exhausted. Future challenges include that many of the genes identified whose expression correlates with a disease state have unknown functions. What role, if any, do these genes play in the disease process? It is important to consider that finding a gene whose expression is correlated with disease state does not imply that the gene is involved with the pathophysiology of the disease state, rather the gene could be responding to the altered physiology of the cell and its environment in response to the disease state. Supervised analyses allow for the identification of genes whose expression are consistently and reproducibly altered between different tumor types, between patients with different survival times who receive the same diagnosis, between patients who respond to radiation therapy and those who do not, and between patients who respond to chemotherapy and those who do not. In addition to stratifying patients regarding clinical outcome, these studies will identify genes whose expression is altered in neoplastic cells. This information will allow greater insights into the malignant process and may suggest targets for chemotherapeutic interventions leading to cures.

764

SECTION IX / ONCOLOGY

Figure 78-4 1071 probe set signature of malignant gliomas. The figure shows the hierarchical clustering pattern of a 1071 gene glioma classifier across 51 tumor and non-neoplastic brain specimens. The dendrogram on top of the figure is derived from the Pearson’s correlations and represents the similarity of the gene expression among the tissue samples.

SELECTED REFERENCES Alizadeh AA, Eisen MB, Davis RE, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 2000;403:503–511. Berger MS, Deliganis AV, Dobbins J, Keles GE. The effect of extent of resection on recurrence in patients with low grade cerebral hemisphere gliomas. Cancer 1994;74:1784–1791. Burger PC. What is an oligodendroglioma? Brain Pathol 2002;12: 257–259. CBTRUS. Statistical Report: primary brain tumors in the United States, 1995–1999. Central Brain Tumor Registry of the United States, 2002. Coons SW, Johnson PC, Scheithauer BW, Yates AJ, Pearl DK. Improving diagnostic accuracy and interobserver concordance in the classification and grading of primary gliomas. Cancer 1997;79:1381–1393. Daumas-Duport C, Scheithauer B, O’Fallon J, Kelly P. Grading of astrocytomas. A simple and reproducible method. Cancer 1998;62: 2152–2165. Davis FG, Preston-Martin S. Epidemiology: incidence and survival. In: Bigner DD, McLendon RE, Bruner JM, eds. Russell and Rubinstein’s

Pathology of Tumors of the Nervous System. New York: Oxford University Press, 1998; pp. 5–45. Fuller GN, Hess KR, Rhee CH, et al. Molecular classification of human diffuse gliomas by multidimensional scaling analysis of gene expression profiles parallels morphology-based classification, correlates with survival, and reveals clinically-relevant novel glioma subsets. Brain Pathol 2002;12:108–116. Giannini C, Scheithauer BW, Weaver AL, et al. Oligodendrogliomas: reproducibility and prognostic value of histologic diagnosis and grading. J Neuropathol Exp Neurol 2002;60:248–262. Golub TR, Slonim DK, Tamayo P, et al. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 1999;286:531–537. Hess KR, Fuller GN, Rhee CH, Zhang W. Statistical pattern analysis of gene expression profiles for glioblastoma tissues and cell lines. Int J Mol Med 2001;8:183–188. Holter NS, Mitra M, Maritan A, Cieplak M, Banavar JR, Fedoroff NV. Fundamental patterns underlying gene expression profiles: simplicity from complexity. Proc Natl Acad Sci USA 2000;97: 8409–8414.

CHAPTER 78 / APPLICATIONS OF GENE EXPRESSION PROFILING TO THE STUDY OF MALIGNANT GLIOMAS

Hoshino T, Rodriguez LA, Cho KG, et al. Prognostic implications of the proliferative potential of low-grade astrocytomas. J Neurosurg 1988;69:839–842. Hsu DW, Louis DN, Efird JT, Hedley-Whyte ET. Use of MIB-1 (Ki-67) immunoreactivity in differentiating grade II and grade III gliomas. J Neuropathol Exp Neurol 1997;56:857–865. Ino Y, Betensky RA, Zlatescu MC, et al. Molecular subtypes of anaplastic oligodendroglioma: implications for patient management at diagnosis. Clin Cancer Res 2001;7:839–845. Kim S, Dougherty ER, Shmulevich L, et al. Identification of combination gene sets for glioma classification. Mol Cancer Ther 2002;1:1229–1236. Kleihues P, Louis DN, Scheithauer BW, et al. The WHO classification of tumors of the nervous system. J Neuropathol Exp Neurol 2002;61: 215–225. Lal A, Glazer CA, Martinson HM, et al. Mutant epidermal growth factor receptor up-regulates molecular effectors of tumor invasion. Cancer Res 2002;62:3335–3339. Ljubimova JY, Khazenzon NM, Chen Z, et al. Gene expression abnormalities in human glial tumors identified by gene array. Int J Oncol 2001;18:287–295. Louis DN, Edgerton S, Thor AD, Hedley-Whyte ET. Proliferating cell nuclear antigen and Ki-67 immunohistochemistry in brain tumors: a comparative study. Acta Neuropathol (Berl) 1991;81:675–679. Louis DN, Holland EC, Cairncross JG. Glioma classification: a molecular reappraisal. Am J Pathol 2001;159:779–786. Mischel PS, Nelson SF, Cloughesy TF. Molecular analysis of glioblastoma: pathway profiling and its implications for patient therapy. Cancer Biol Ther 2003;2:242–247. Mischel PS, Shai R, Shi T, et al. Identification of molecular subtypes of glioblastoma by gene expression profiling. Oncogene 2003;22: 2361–2373. Nutt CL, Mani DR, Betensky RA, et al. Gene expression-based classification of malignant gliomas correlates better with survival than histological classification. Cancer Res 2003;63:1602–1607.

765

Paleologos NA, Macdonald DR, Vick NA, Cairncross JG. Neoadjuvant procarbazine, CCNU, and vincristine for anaplastic and aggressive oligodendroglioma. Neurology 1999;53:1141–1143. Perou CM, Jeffrey SS, van de RM, et al. Distinctive gene expression patterns in human mammary epithelial cells and breast cancers. Proc Natl Acad Sci USA 1999;96:9212–9217. Perry A, Scheithauer BW, Stafford SL, Lohse CM, Wollan PC. “Malignancy” in meningiomas: a clinicopathologic study of 116 patients, with grading implications. Cancer 1999;85:2046–2056. Perry A, Stafford SL, Scheithauer BW, Suman VJ, Lohse CM. Meningioma grading: an analysis of histologic parameters. Am J Surg Pathol 1997;21:1455–1465. Perry A, Stafford SL, Scheithauer BW, Suman VJ, Lohse CM. The prognostic significance of MIB-1, p53, and DNA flow cytometry in completely resected primary meningiomas. Cancer 1998;82: 2262–2269. Pomeroy SL, Tamayo P, Gaasenbeek M, et al. Prediction of central nervous system embryonal tumour outcome based on gene expression. Nature 2002;415:436–442. Rickman DS, Bobek MP, Misek DE, et al. Distinctive molecular profiles of high-grade and low-grade gliomas based on oligonucleotide microarray analysis. Cancer Res 2001;61:6885–6891. Sallinen SL, Sallinen PK, Haapasalo HK, et al. Identification of differentially expressed genes in human gliomas by DNA microarray and tissue chip techniques. Cancer Res 2000;60:6617–6622. Scott JN, Rewcastle NB, Brasher PM, et al. Which glioblastoma multiforme patient will become a long-term survivor? A population-based study. Ann Neurol 1999;46:183–188. Shai R, Shi T, Kreman TJ, et al. Gene expression profiling identifies molecular subtypes of gliomas. Oncogene 2003;22:4918–4923. Smith JS, Perry A, Borell TJ, et al. Alterations of chromosome arms 1p and 19q as predictors of survival in oligodendrogliomas, astrocytomas, and mixed oligoastrocytomas. J Clin Oncol 2000;18: 636–645.

HEMATOLOGICAL MALIGNANCIES

79

Acute Myeloid Leukemias IRIS T. CHAN AND D. GARY GILLILAND

SUMMARY The acute myeloid leukemias (AML) are a heterogeneous group of malignant diseases of hematopoietic progenitor cells with a spectrum of molecular genetic abnormalities, clinical characteristics, and variable outcomes with available treatments. A virtual explosion has been seen in the identification and characterization of disease alleles in AML in cell culture and murine models of leukemia. These insights have provided a platform for development of molecularly targeted therapies of AML that may improve clinical outcome. Key Words: Acute myeloid leukemias; acute promyelocytic leukemia; core binding factor; hematopoietic stem cell; internal tandem duplications; RAS mutations.

INTRODUCTION The acute myeloid leukemias (AML) are a heterogeneous group of malignant diseases of hematopoietic progenitor cells with a spectrum of molecular genetic abnormalities, clinical characteristics, and variable outcomes with available treatments. For several decades, the cornerstone for treatment of AML has been empirically derived cytotoxic chemotherapy. Although this remains the foundation of treatment, advances have been made in the development of molecularly targeted therapies of AML based on an improved understanding of the genetic basis of the disease. Furthermore, emerging data indicate that AML can be viewed as a hierarchy, similar to that observed in the normal hematopoietic development, in which there is a critical population of leukemic stem cells that are necessary for continued growth and propagation of AML. These leukemic stem cells are likely to be the basis for relapsed disease after remission induction, and may perhaps be the best target for developing novel therapeutic approaches to disease.

EPIDEMIOLOGY AML is a rare disease, occurring at a frequency of about 1/100,000/yr in the general population. However, the age-specific incidence of AML rises linearly after age 40 yr with a median age of approx 65 yr. AML is rarely a heritable disease, but several congenital disorders predispose to the development of AML. These include Fanconi’s anemia, and Bloom, Down, Kostmann, and Diamond-Blackfan syndromes, respectively. Antecedent hematological disease, including the myeloproliferative syndromes, From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

myelodysplastic syndrome (MDS), and paroxysmal nocturnal hemoglobinuria, also increases the risk of AML development. Therapyrelated AML may develop as a consequence of exposure to chemotherapy, including alkylating agents and epipodophyllotoxins, and ionizing radiation. Finally, there are rare, inherited predispositions to AML, such as the familial platelet defect with propensity to develop AML (FPD/AML) syndrome.

CLASSIFICATION OF AML AND RELATIONSHIP TO PROGNOSIS Morphological criteria for the classification of AML were developed more than 25 yr ago, and are known as the French–American– British (FAB) classification. The classification has been modified by the World Health Organization to incorporate cytogenetics, molecular genetics, as well as morphological and immunophenotypic findings not previously described. In addition, a diagnosis previously required at least 30% bone marrow blasts, whereas AML is now diagnosed by World Health Organization criteria with a minimal blast percentage of 20% in blood or bone marrow. It is likely that the classification of AML will undergo further iterations as additional molecular abnormalities are identified and correlated with prognosis. The diagnosis of AML requires morphological evaluation of the peripheral blood smear, bone marrow aspirate and core biopsy, cytogenetics, molecular genetics, and immunophenotyping. Cytogenetics are now recognized to convey important prognostic information that is interrelated with the type of therapy. For example, patients whose leukemic cells have translocations t(8;21), t(16;16), or inv(16) that involve the heterodimeric hematopoietic transcription factor core binding factor (CBF) have a favorable outcome with induction chemotherapy and intensive postremission consolidation chemotherapy with high-dose cytosine arabinoside. Similarly, patients with acute promyelocytic leukemia (APL) associated with t(15;17) have an excellent prognosis when treated with a combination of all-trans-retinoic acid (ATRA) and intensive induction chemotherapy that includes an anthracycline. In contrast, abnormalities of chromosomes 5, 7, 11q23 or complex karyotypes, have a very poor outcome with available induction and postremission chemotherapy. Patients with a normal karyotype or with trisomy 8 have an intermediate prognosis.

DISEASE ALLELES IN AML Identification of mutant genes causally implicated in the pathogenesis of leukemia has been challenging, in part because leukemia

767

768

SECTION IX / ONCOLOGY

is a rare disease and most cases are sporadic. Thus, strategies that are informative in heritable diseases, such as generalized linkage analysis, are not applicable for elucidation of disease alleles in leukemia. There are several lines of evidence that AML is a clonal disorder that is the consequence of acquired somatic mutation that occurs in a hematopoietic progenitor. Among these are recurring cytogenetic abnormalities in leukemia cells, which have been valuable clues to the genomic localization of leukemia-associated alleles, as detailed later. However, there has also been a marked expansion of the number of disease alleles in AML that are not evident by conventional cytogenetic analysis. In all, more than 100 disease alleles have been identified, far exceeding the number of subtypes of AML recognized by other diagnostic criteria. Thus, one would expect that many of these mutations would target similar transcriptional or signal transduction targets. Mutations and gene rearrangements will be presented in a functional context that emphasizes the targeting of shared pathways of transformation. MUTATIONS THAT TARGET SIGNAL TRANSDUCTION INTERMEDIATES Several clues revealed that mutations that activate signal transduction pathways and confer proliferative and/or survival advantages to hematopoietic progenitors would be important in AML. First, rare cases of de novo AML are associated with t(9;22), resulting in expression of the BCR-ABL fusion, a constitutively activated tyrosine kinase. Similarly, cases of AML arising from chronic myelogenous leukemia (myeloid blast crisis) also harbor activating mutations in BCR-ABL. In addition, it had been demonstrated that activating mutations in RAS family members, primarily N-RAS and K-RAS, are present in AML, as well as in solid tumors. Also, FLT3-activating mutations have been identified in AML at a frequency of approx 30%, making FLT3 the most commonly mutated gene in AML. Collectively, mutations that constitutively activate signal transduction pathways have been identified in approx 50% of AML, and it is plausible that appropriate screens will identify similar mutations in the remainder. These mutations appear to form a complementation group, in that they do not occur together in the same AML patient. For example, FLT3 and RAS mutations, though frequent in leukemia as individual mutations, rarely occur together in the same patient. This observation suggests that these mutations subserve similar functions in AML in providing proliferation and survival signals to leukemia blasts. Because these mutations may occur with disease progression, it is possible that on those very rare occasions when both mutations are observed in a patient with AML, they have arisen independently in separate clones. MUTATIONS THAT CONSTITUTIVELY ACTIVATE RECEPTOR TYROSINE KINASES Activating mutations have been identified in FLT3 (approx 30%) and c-KIT (50% of the cases. These differences are in part resulting from differing molecular pathogenesis, which results in a higher proportion of tumors with de novo pan-drug resistance Key Words: Acute lymphoblastic leukemia; chemotherapy; chronic myelogenous leukemia; E2A-PBX1; Tel-AML1; tumor suppressor.

INTRODUCTION Acute lymphoblastic leukemia (ALL) is the most common malignancy among children and represents 80 and 10–20%, respectively, of childhood and adult acute leukemias. Advances in the understanding of childhood ALL have transformed the disease from a rapidly progressive and usually lethal disorder to a disease with an 80% cure rate. These improved outcomes are resulting from the development of multiagent, dose-intensive chemotherapy regimens, central nervous system (CNS) prophylaxis and improved supportive care. In parallel to the improved understanding of therapeutic approaches to ALL, knowledge of the pathogenesis of ALL has increased dramatically. But uncovering the molecular mechanisms behind this disease has not yet produced targeted therapies that specifically act on these elements, as has been the case, for example, with tyrosine kinase inhibition for chronic myelogenous leukemia (CML). Such novel treatment approaches are needed in particular in adult ALL, which although morphologically indistinguishable from pediatric disease, has a considerably poorer response to chemotherapy and From: Principles of Molecular Medicine, Second Edition Edited by: M. S. Runge and C. Patterson © Humana Press, Inc., Totowa, NJ

remains incurable in >50% of the cases. These differences in part result from differing molecular pathogenesis, which results in a higher proportion of tumors with de novo pan-drug resistance. ALL is characterized by the accumulation of malignant precursors of either B or T lymphocytes in bone marrow and/or peripheral blood. Although some maturation might occur in the periphery, the proliferating cells generally are arrested in early developmental stages within the marrow. The primitive cells crowd out the normal marrow elements, causing pancytopenia. Tumor infiltration frequently causes splenomegaly, hepatomegaly, and diffuse lymphadenopathy. Patients usually present with a few weeks of nonspecific symptoms such as fatigue, often resulting from anemia, or may have a flu-like syndrome. Fever, petechial rash, bleeding, or easy bruising owing to cytopenias may also prompt medical evaluation. Bone pain might also be present, especially in children. Leukostasis or leptomeningeal spread causing focal neurological deficits can occur at diagnosis, but is uncommon. Complete blood counts usually point to the diagnosis, showing pancytopenia of mature cells and often circulating blasts. However, the WBC is variable, and might be decreased, normal, or elevated. Marked elevation of leukocyte count with >100,000 cells/mm3 is not unusual. Lactate dehydrogenase is typically elevated and there may be evidence of disseminated intravascular coagulation or spontaneous tumor lysis with hyperuricemia, hyperkalemia, and hyperphosphatemia. A bone marrow aspirate and biopsy with immunophenotyping and cytogenetic analysis is used to confirm diagnosis. Lumbar puncture is mandatory to assess the presence of leukemic cells in the CNS, which, with the testes, can serve as a sanctuary site. Untreated ALL is rapidly fatal, with death usually occurring in weeks to a few months. Therapy consists of several dose-intense cycles of multiagent, noncross-resistant chemotherapy. Many classes of chemotherapeutic agents share activity in ALL including anthracyclines, steroids, alkylating agents, vinca alkaloids, topoisomerase II inhibitors, asparaginase, antimetabolites, and purine analogs. With current treatment strategies, approx 80% of children with ALL are cured. In adults, the rate is closer to 30%, virtually all of them who are younger than 60 yr. The goal of therapy is to induce complete remission as defined by the elimination of leukemic blasts by diagnostic surveillance techniques and restoration of normal peripheral blood counts. Because of advances in laboratory detection of minimal populations of leukemic cells, it is somewhat difficult to compare the success rates of remission induction therapies to those in the past, when morphological data was primarily used. This is because

776

CHAPTER 80 / ACUTE LYMPHOBLASTIC LEUKEMIA

the flow cytometric and molecular probing techniques in use are considerably more sensitive to minimal populations of tumor cells. Indeed, current assays are capable of detecting as few as one leukemic cell among 1 million normal cells. Regardless, historic data show excellent responses to induction chemotherapy, with approx 98% of children and 80% of adults entering a complete remission, as defined by 250 >23 20 13

8.0 7.0 6.8; 7.2 4.7 5.6 2.4

surface area (140 µm2) relative to volume (85 µm3) is diminished. This is owing either to the reduction in, or the lack of (in the most severe instances), a protein. The defect may be secondary. For example, spectrin reduction may stem from a primary deficiency in ankyrin. The red cell tends to acquire a spherical shape making it prone to premature sequestration by the spleen (see Fig. 85-2). Osmotic gradient ektacytometry shows that spherocytes are

832

SECTION X / HEMATOLOGY

Figure 85-2 Gallery of normal and abnormally shaped red cells. (A) Normal red cells. (B) Spherocytes. (C) Elliptocytes with some poikilocytes. (D) Ovalostomatocytes in southeast Asian ovalocytosis. (E) Stomatocytes (OHS). (F) Stomatocytes (DHS).

osmotically fragile, in keeping with the increase in the percentage of hyperdense cells. Evidence is accumulating that the spheroid shape is, at least in part, acquired during erythropoiesis and not exclusively through the loss of microparticles once the mature erythrocyte has been launched into the circulation. Noteworthy, reticulocytes are small in genetically determined HS, whereas they are of normal size in spherocytosis of immune origin. In HS, parvovirus infections produce a dramatic, yet reversible, drop of the red cell count, acute erythroid aplasia combining with chronic hemolysis. HS in newborn and infants is the subject of ongoing studies; the symptoms are distorted by the “sluggish” phase exhibited by erythropoiesis during the first months of life. Five genes can be basically responsible for HS (in the order of decreasing frequency): ANK1, SLC4A1, SPTB, EPB42, and SPTA1 (see Table 85-1). The inheritance pattern is dominant for mutations in the three first genes and recessive for those in the two last. De novo mutations are common in the ANK1 and the SPTB genes. Salient features have surfaced about these genes.

Mutations within the SLC4A1 gene may produce a nonhematological disease, that is, distal renal tubular acidosis. As a general rule, mutations yielding the dominantly inherited variety of distal renal tubular acidosis fail to produce HS and vice versa. Some remarkable mutations have been recorded, such as a mutation cancelling the initiation codon of spectrin β-chain (AUG→ GUG; β-spectrin Promissão). These are natural mutations that pinpointed the protein (or one of the proteins) that links band 3 complex (including band 3, ankyrin, spectrin, glycophorin A, and protein 4.2, among others) and the Rh complex (including, in particular, of the Rh polypeptides, Rh-associated glycoprotein [RhAG], CD47, glycophorin B, and Landsteiner–Wiener antigen [LW]) (see Fig. 85-1). Briefly, in patients devoid of protein 4.2 or of band 3, CD47 was secondarily missing. In addition, it appeared that the glycosylation of some proteins (RhAG, in particular) was increased when protein 4.2 was missing. It was surmised that protein 4.2 or, in a secondary manner, CD47, could chaperone some glycoproteins. Their absence would disturb their trafficking toward

CHAPTER 85 / DISORDERS OF THE RED CELL MEMBRANE

the membrane and alter their glycosylation. Mutations in some genes of the Rh complex lead to severe hemolytic anemia. Natural HS mutations in the mouse have been found in the Spna1 gene (chromosome 1), encoding spectrin α-chain (sph/sph, sph2BC/sph2BC, sphJ/sphJ mice), in the Sptb1 gene (chromosome 12), encoding spectrin β-chain (ja/ja mouse), and in the Ank1 gene (chromosome 8) encoding ankyrin (nb/nb mouse). Mouse Slc4a1 (chromosome 11) and Epb4.2 (chromosome 2) genes have been knocked out. One of the mice with inactivated Slc4a1 gene showed a hypercoagulable state. Orthologues of human genes have been found. In a remote biological context, these orthologues are likely to play different roles, and natural mutations will produce phenotypes distinct from the phenotypes in humans, most presumably. In the Zebra fish, mutation retsina alters the orthologue of human SLC4A1 gene and produces a dyserythropoiesis. Nevertheless, it is questionable whether this dyserythropoiesis is similar or even close to human CDA II. In humans, the mature red cell membrane shows specific biochemical stigmata, which have not been investigated in the Zebra fish (nucleated) erythrocytes, and involve the glycan moiety of band 3. Also, the human CDA II candidate gene, at least one of them, has been mapped, which excluded the SLC4A1 gene that is the polypeptide moiety of band 3. Still in the Zebra fish, mutation merlot/chablis involves the orthologue of human EPB41 gene and generates a severe congenital anemia. Mutation riesling modifies the orthologue of human SPTB gene, that encodes β-spectrin, and yields HS. There is a recurrent, weak SPTA1 allele, allele αLEPRA, which dramatically aggravates the expression of (rare) weak or null SPTA1 alleles that would happen to occur in trans. In allele αLEPRA, the c→t transition at position-99 of intron 30 results in the enhancement of an acceptor splice site at position-70, causing aberrant splicing, frameshift and premature termination of translation. In spite of this, 16% of the transcripts escape the abnormal splicing. Such a combination, the allele in trans of allele αLEPRA being a virtually null allele of the SPTA1 gene carrying a mutation in exon 51, has been encountered. The output of both abnormal SPTA1 genes collapsed to 8% of normal. These observations underscore the fact that HS, unlike HE as next discussed, is in general the result of some reduced or missing red cell skeletal protein, analogous to the situation found in the thalassemias among hemoglobinopathies. THE RHNULL CONDITION As an extension of spherocytosis, it is pertinent to consider the recessive Rhnull condition, in which there is a failure of expression of all Rhesus antigen proteins on the red cell membrane. Hemolysis is associated with abnormally shaped red cells ranging from spherocytes to stomatocytes. There is a mild cation leak. A total of six proteins (CE and D Rh polypeptides, RhAG, CD47, LW, GP8), achieving the Rh complex, are absent from the membrane. If any core protein of the assembly, such as RhAG, is mutated, then the entire complex is unstable and/or cannot be trafficked to the membrane. HEREDITARY ELLIPTOCYTOSIS AND ITS AGGRAVATED FORM, HEREDITARY POIKILOCYTOSIS Three genes may carry mutations causing HE and its aggravated form, HPP (see Fig. 85-2): the SPTA1, SPTB, and the EPB41 genes (see Table 85-1). Except for the latter case in which protein 4.1 is reduced and exceptionally missing, spectrin mutations are basically “qualitative mutations” that alter specific functions within spectrin α- or β-chains. (The absence of either chain of spectrin would be past HS and would be lethal.)

833

Mutations in SPTA1 and in SPTB genes are clustered in the 5′and the 3′-regions of these genes, respectively. The spectrin α and β chains are organized in an antiparallel fashion, such that the mutations tend to affect the N-terminal region of the former and the C-terminal region of the latter. In spectrin α-chain, however, the cluster of mutations is somewhat stretched, covering approximately one-third of the entire polypeptide. The most upstream mutation (mutation Lograno; T→G, exon 2; I24S) lies in the first repeating segment (spectrin chains are made up of 106 amino acid repeating segments over most of their length). The mutation furthest downstream (mutation Saint-Claude; T→G, in the 3′-acceptor splice site of intras 19 ninth repeating segment, immediately upstream of the SH3 domain. By contrast, mutations in the C-terminal region of the β-chain are concentrated in the last (17th) repeating segment. The previously mentioned mutations impede the head-to-head self-association process of spectrin dimers into tetramers or higher order oligomers. Mutations in the α-chain of spectrin that are situated more “inland” must act on oligomerization through illdefined long distance interactions. The whole structure of the membrane skeleton is “loosened” and HE/HPP ensues. Concerning SPTA1-related HE, a remarkable finding is that in virtually all kindreds, elliptocytosis may assume a mild to moderate presentation, depending on the intrinsic severity of the mutation, or a much more severe form (poikilocytosis in many instances). The aggravation takes place when a weak allele, coined allele αLELY, lies in trans to the allele responsible for elliptocytosis, or the αHE allele. Allele αLELY is so common, with a 0.15–0.30 frequency in all ethnic groups investigated, that any kindred with a αHE allele is likely to harbor allele αLELY as well, making αHE/allele αLELY compound heterozygosity a most predictable combination among family members. Allele αLELY is made up of three mutations that have always been found in complete linkage disequilibrium: (1) A→G in exon 40; L1857V, (2) –12 g→t in intron 45, and (3) –12 g→a in intron 46. The first change is a functionally neutral polymorphism. The third one may appear alone in otherwise normal SPTA1 alleles. The second mutation is the potentially deleterious alteration. It yields partial skipping of exon 45 (18 bp), corresponding to six amino acid residues, in exactly 50% of the transcripts. These six amino acid residues are essential for the onset of the side by side dimerization of an α-chain with its partner β-chain. It must be pointed out that the dimerization process takes place not far from the Cand the N-termini of spectrin α- and the β-chains, respectively, that is, at the end opposite to the self-association process. The absence of the exon 46-encoded six amino acids thus removes 50% of the α-chains issued from allele αLELY, the remaining αchains being expected to participate normally in the tetramerization process subsequently. This would appear to cause little harm, for α-chains are produced in large excess though the precise extent of this excess is not known with accuracy. Nonetheless, consider an αHE allele lying in trans to allele αLELY. In all likelihood, this dysfunctional αHE allele will have a normal output and a normal ability to bind β chains, the HE mutation standing in the 5′ region. As a result, it will fill the room left by the reduced set of α-chains issued from allele αLELY. More dysfunctional αHEβ dimers will thus be formed (rising from 50% to 66%) and the deficient selfassociation process will become more prominent. Eventually, the phenotype will be surprisingly aggravated (mimicking homozygosity). It often turns to poikilocytosis with torn red cells of every size and shape.

834

SECTION X / HEMATOLOGY

Some SPTA1 HE mutations are frequent in malaria-infested regions, suggesting a potential protective effect against the disease. This is the case, among others, of the 648ins3 insertion in exon 4 (additional Leu in repeating segment 2), in sub-Saharan countries. Interestingly, this mutation followed migrations so that it is sporadically found in North Africa and Southern Italy as well, but not apparently in Spain. One explanation is that the mutation was absent when the Straits of Gibraltar was crossed but was present when the Straits of Sicily was passed by the North Africans. Some odd situations may be encountered. For example, a case of HE was attributed to a double change in spectrin, both occurring close to the oligomerization site, but one (well known before) in the α-chain, and the other in the β-chain (β-chain Kuwaitino: C→A in exon 30; A2018D in repeat β17). The sphDem/sphDem mouse carries a mutation in the Spna1 gene (insertion of an intracisternal a particle element in intron 10 causing exon skipping and an in-frame deletion of 46 amino acids from repeat 5). It shows HE and thrombophilia. A mouse in which the EPB41 gene was knocked out has been obtained. The 4.1 (–) trait results from the absence of one haploid set of protein 4.1. It is symptomless. Very rare cases of totally missing protein 4.1 produce a rather severe poikilocytosis. Missing protein 4.1 results in the absence of glycophorin C and protein p55, which has hinted at a protein complex, now characterized. SOUTHEAST ASIAN OVALOCYTOSIS Southeast Asian ovalocytosis (SAO) is a dominantly inherited, asymptomatic trait that alters the red cell morphology and mechanical properties in the heterozygous state. Homozygosity for the SAO allele has never been encountered and is thought to be lethal. SAO is widespread among various ethnic groups in Papua New Guinea, the Philippines, and Indonesia, resulting from its protective effects against malaria, and has diffused round the Indian Ocean shore in the wake of migration trails. From there, a few carriers reached Europe. Smears display unmistakable ovalostomatocytes (see Fig. 85-2). The genetic alteration is a 27 nucleotide deletion in the SLC4A1 gene. As a consequence, amino acids 400–408, which stand at the very junction of the cytoplasmic and membrane domains of band 3 (see Fig. 85-1), are missing. Such a molecular defect is thought to be responsible for the cellular rigidity associated with ovalocytosis. The deletion is in linkage disequilibrium with the so-called “Memphis I” polymorphism (AAG→GAG; K56E).

GENETIC DISORDERS OF RED CELL MEMBRANE PERMEABILITY There are a series of further conditions in which the main pathology lies not in a defect in the mechanical strength of the membrane, but in its permeability, especially to the cations Na+ and K+, an abnormality that can compromise the osmotic integrity of the cell and cause premature lysis. These are collected under the title, “hereditary stomatocytoses and allied conditions”, reflecting the name given to the first pedigree, but subsequent experience has shown that there are leaky cases without stomatocytes, and that there are very stomatocytic cases without a cation leak. The morphological abnormality “stomatocytosis” (see Fig. 85-2) is almost certainly secondary to expansion of the inner leaflet of the lipid bilayer. It is not clear why these leaky cells should show this feature; it is not simply the result of the cation leak, for a leaky cell swells up into a sphere before bursting. Theoretically, stomatocytosis could occur either because of a gross abnormality in membrane lipid composition, or because there is an imbalance in the forces that push lipids between one

leaflet and the other, the best characterized of which is the aminophospholipid flippase. There are a series of forms of these diseases, which can be distinguished on a number of criteria, as follows: severity, both of the hemolysis and the magnitude of the cation leak; cell hydration, which can be increased or decreased, as can easily be shown by osmotic gradient ektacytometry; the effect of temperature on the passive leak; the presence of the 32 kDa membrane protein “stomatin;” and the presence of perinatal ascites in a pedigree. The way in which these criteria apply is clarified below in the discussion of these conditions’ different forms. OVERHYDRATED HEREDITARY STOMATOCYTOSIS AND THE STOMATIN PROTEIN OHS is the most severe of these conditions and the first to be described. The patients show obvious hemolytic anemia (hemoglobin 8–12 g/dL) with jaundice and gallstones with a catastrophic diffusional, Fick-type cation leak. There is a major molecular abnormality in these cells. A 32-kDa protein, now known as stomatin after this disease, is missing from the membrane. The human gene sequence is known but there does not seem to be any mutation in the gene in these patients. Mice in which the Epb72 gene (encoding stomatin) has been inactivated fail to show stomatocytosis. The protein is ubiquitously distributed, from Archaea to humans, and is found in many (but not all) human cells. In OHS, it seems that the protein is lost as the cell matures through the reticulocyte stage. Whether this happens because of a structural abnormality in another supporting protein (as in Rhnull syndrome seen above), or perhaps because of the cation leak, is unknown. The function of the stomatin protein is likewise unknown. It is associated with cholesterol + sphingomyelin-rich “rafts” in the membrane. DEHYDRATED HEREDITARY STOMATOCYTOSIS Also known as “hereditary xerocytosis,” DHS is a milder condition than OHS, and the cells show a much less marked cation leak. As the name implies, the cells show a high mean corpuscular hemoglobin concentration. It is by far the commonest of this group. It is identical to the condition first labeled, “hereditary hyperphosphatidylcholine anemia,” and typically phosphatidylcholine levels in the membrane are high. This condition maps to a locus on chromosome 16, but the responsible gene is not yet known. In a minority of pedigrees, DHS can be associated with a very unusual syndrome of perinatal ascites, in which the fetus shows major fluid accumulation, especially in the peritoneal cavity. This tendency gradually abates after delivery and is gone after 6 mo of life or less. Ascitic taps are commonly required, either in utero or after delivery. Some of the pedigrees show abnormal temperature effects in the passive leak, manifesting as pseudohyperkalemia, discussed later. Like OHS, DHS does not respond well to splenectomy, and after this procedure the patients show a tendency to thrombosis, which can be difficult to treat with anticoagulants and can be life threatening. CRYOHYDROCYTOSIS This condition shows a mild hemolytic state, only occasionally requiring transfusions, with some stomatocytes on the film. The red cells show a blatant temperature effect. If left in the refrigerator overnight, the cells lyse dramatically. Like the pseudohyperkalemia that is found in some of the other variants, this cold lysis is secondary to an abnormality in the temperature dependence of the passive leak to Na+ and K+, which is markedly enhanced at 0°C compared with normal (Fig. 85-3), although the Na+-K+ pump is effectively silenced at these temperatures. The cells also lose K+ and gain Na+, because of similar temperature effects at room temperature. There is a prelytic

CHAPTER 85 / DISORDERS OF THE RED CELL MEMBRANE

835

Figure 85-3 Different temperature variants among the hereditary stomatocytosis and familial pseudohyperkalaemia variants. K+ influx, in the presence of ouabain and bumetanide, was measured using Rb as a tracer. In all panels, normals, which show a min. at about 8°C, are shown by open circles. (A) shows cryohydrocytosis “Hemel” (filled circles) and familial pseudohyperkalemias (FP) “Cardiff” (filled squares). (B) shows the “flat”, “shallow slope” profiles: DHS “Blackburn” (filled squares) and FP “Edinburgh” (filled circles). (C) shows the “shoulder” profile: FP “Falkirk” (filled squares) and FP “Chiswick” (filled circles).

leak of Na+ and K+ (giving pseudohyperkalemia) before the cation abnormality becomes critical and the cell lyses, losing its entire K+ content to the plasma. FAMILIAL PSEUDOHYPERKALEMIA There are a series of hematologically trivial conditions that present via the chemical pathologists with unpredictably high plasma potassium concentrations, not associated with acidosis or renal failure or Addisonian stigmata. An artifact is suspected and the diagnosis of red cell pseudohyperkalemia can be revealed by a simple storage test on red cells at room temperature. It is a reflection of the major heterogeneity in all of these conditions that three variants can be distinguished from six unrelated pedigrees in the United Kingdom. The difference lies in the exact temperature dependence of the “leak” cation flux, most conveniently measured as the ouabain + bumetanide-resistant K+ influx. The results are shown in Fig. 85-3. The normal curve shows a minimum at about 8°C; the “Edinburgh” pedigree (filled circles) shows a flat, shallow slope pattern; the “Chiswick” pattern shows a min. at 30°C, with a max. at 12°C, followed by a further fall. In the Cardiff variant, the temperature profile shows a min. at 23°C. In all of these cases, whereas the leak shows these abnormal patterns, the NaK pump to which the leak is opposed always shows the same monotonic fall with temperature. At room temperature, the leak and pump are not balanced and the red cell loses K+ to the plasma, giving “pseudohyperkalemia.” In all of these FP cases, the leak at 37°C is not markedly abnormal. The molecular basis of the leak in these conditions is not understood. One pedigree (Edinburgh) maps to the chromosome 16 locus as DHS.

CONGENITAL DYSERYTHROPOIETIC ANEMIAS CONGENITAL DYSERYTHROPOIETIC ANEMIA, TYPE I CDA I is a rare genetic condition, and is the first CDA for which the responsible gene has been identified in humans. It includes dysplastic changes in the late erythroid precursors: internuclear chromatin bridges (Fig. 85-4), spongy heterochromatin, and invagination of the nuclear envelope carrying cytoplasmic

organelles into the nucleus. The bone marrow is hyperactive, but the output of red cells is decreased. CDA I may be associated with a series of dysmorphologies of the skeleton. The severe ineffective erythropoiesis leads to iron loading. Red cells that succeed in getting into the circulation show a number of altered shapes, such as elliptocytosis, dacryocytosis, and poikilocytosis. This indicates that the mature erythrocytes themselves are abnormal and account for the hyperhemolysis that aggravates the anemia of central origin. The red cell morphological changes are associated with a small but significant reduction of protein 4.1R, which is sufficient to alter the erythrocyte shape. It is part of the dyserythropoietic process, but must occur substantially downstream in the chain of events stemming from the primary genetic lesion. They provide an unknown yet simple and reliable aid to the diagnosis. The CDAN1 gene whose mutations are responsible for CDA I was mapped to chromosome 15 q15.1–15.3. Homozygosity mapping was used in a cluster of 45 highly inbred Israeli Bedouins living in the Negev Desert and constituting an isolate. The CDAN1 gene has been identified. It is made up of 28 exons and encodes a ubiquitously expressed mRNA of 4738 bases. The encoded protein has been termed codanin-1 and contains 1226 amino acids. It is expected to be O-glycosylated. The first 150 residues of codanin-1 display sequence similarities with collagens and contain two short segments showing a weak resemblance with microtubule-associated proteins, MAP1B (neuraxin) and synapsin. Putative mouse and Fugu orthologs of codanin-1 were identified. The functions of codanin-1 are yet to be discovered. In a different line, a treatment of CDA I has been found serendipitously. α-interferon, which was used to treat hepatitis C in a patient with CDA I, was also effective in controlling the CDA. Thereafter, a number of reports have confirmed the efficacy of α-interferon in CDA I, but the mechanism of action of interferon-α2b remains an enigma.

836

SECTION X / HEMATOLOGY

Figure 85-4 Eythroblasts in congenital dyserythropoietic anemia, types I and II (CDA I and CDA II). (A) CDA I. The most conspicuous aspect is the interchromatin bridge, which links two erythroblasts (arrow). (B) CDA II. The main feature is the numerous binucleated erythroblasts (arrows).

CONGENITAL DYSERYTHROPOIETIC ANEMIA, TYPE II CDA II is a rare genetic condition. It is characterized by the presence of a number of binucleated erythroblasts (see Fig. 85-4). CDA II, which is a purely hematological condition, displays major changes in the N-glycosylation of band 3. Such a stigmata is easy to show and is pathognomonic of CDA II. The CDAN2 gene has been mapped to 20q11.2, but not yet identified. Given the glycosylation abnormalities of band 3 and of membrane glycosphingolipids, too, it has been thought that some enzymes involved in the building-up of the glycan moiety were the primarily mutated proteins. Despite much effort and the exclusion of some genes, the culprit remains to be identified. As mentioned, it is likely that human CDA II is but loosely related to the CDA seen in the Zebrafish.

SELECTED REFERENCES Alloisio N, Dalla Venezia N, Rana A, et al. Evidence that red blood cell protein p55 may participate in the skeleton-membrane linkage that involves protein 4.1 and glycophorin C. Blood 1993;82:1323–1327. Alloisio N, Morlé L, Maréchal J, et al. Sp alpha V/41: a common spectrin polymorphism at the alpha IV-alpha V domain junction. Relevance to the expression level of hereditary elliptocytosis due to alpha-spectrin variants located in trans. J Clin Invest 1991;87:2169–2177.

Basu A, Carey P, Stewart G, Richmond S. Dehydrated hereditary stomatocytosis with transient perinatal ascites. Arch Dis Child Fetal Neonatal Ed 2003;88:F438, F439 Bosma PJ, Chowdhury JR, Bakker C, et al. The genetic basis of the reduced expression of bilirubin UDP-glucuronosyltransferase 1 in Gilbert’s syndrome. N Engl J Med 1995;333:1171–1175. Bruce LJ, Beckmann R, Ribeiro ML, et al. A band 3-based macrocomplex of integral and peripheral proteins in the RBC membrane. Blood 2003;101:4180–4188. Bruce LJ, Ghosh S, King MJ, et al. Absence of CD47 in protein 4.2-deficient hereditary spherocytosis in man: an interaction between the Rh complex and the band 3 complex. Blood 2002;100:1878–1885. Bruce LJ, Tanner MJ. Erythroid band 3 variants and disease. Baillieres Best Pract Res Clin Haematol 1999;12:637–654. Carella M, Stewart GW, Ajetunmobi JF, et al. Genomewide search for dehydrated hereditary stomatocytosis (hereditary xerocytosis): mapping of locus to chromosome 16 (q23-qter). Am J Hum Genet 1998; 63: 810–816. Cartron JP. Rh blood group system and molecular basis of Rh-deficiency. Baillieres Best Pract Res Clin Haematol 1999;12:655–689. Cartron JP. Rh-deficiency syndrome. Lancet 2001;358 Suppl 1:S57. Coles S, Chetty M, Ho M, et al. Two British families with variants on the “cryohydrocytosis” form of hereditary stomatocytosis. Br J Haematol 1999;105:1055–1065. Coles SE, Ho MM, Chetty MC, Nicolaou A, Stewart GW. A variant of hereditary stomatocytosis with marked pseudohyperkalaemia. Br J Haematol 1999;104:275–283. Da Costa L, Mohandas N, Sorette M, Grange MJ, Tchernia G, Cynober T. Temporal differences in membrane loss lead to distinct reticulocyte features in hereditary spherocytosis and in immune hemolytic anemia. Blood 2001;98:2894–2899. Dalla Venezia N, Wilmotte R, Morlé L, et al. An alpha-spectrin mutation responsible for hereditary elliptocytosis associated in cis with the alpha v/41 polymorphism. Hum Genet 1993;90:641–644. Delaunay J. Red cell membrane and erythropoiesis genetic defects. Hematol J 2003;4:225–232. Delaunay J, Iolascon A. The congenital dyserythropoietic anaemias. Baillieres Best Pract Res Clin Haematol 1999;12:691–705. Dgany O, Avidan N, Delaunay J, et al. Congenital dyserythropoietic anemia type I is caused by mutations in codanin-1. Am J Hum Genet 2002;71:1467–1474. Entazami M, Becker R, Mensen H, Marcinkowki M, Versmold H. Xerocytosis with concomitant intrauterine ascites: First description and therapeutic approach. Blood 1996;90:5392, 5393. Feder JN, Gnirke A, Thomas W, et al. A novel MHC class I-like gene is mutated in patients with hereditary haemochromatosis. Nat Genet 1996;13:399–408. Fournier CM, Nicolas G, Gallagher PG, Dhermy D, Grandchamp B, Lecomte MC. Spectrin St Claude, a splicing mutation of the human alpha-spectrin gene associated with severe poikilocytic anemia. Blood 1997;89:4584–4590. Fricke B, Argent A, Pizzey A, et al. The “stomatin” gene and protein in overhydrated hereditary stomatocytosis. Blood 2003;102:2268–2277. Gallagher PG, Forget BG. Hematologically important mutations: band 3 and protein 4.2 variants in hereditary spherocytosis. Blood Cells Mol Dis 1997;23:417–421. Gallagher PG, Forget BG. Hematologically important mutations: spectrin and ankyrin variants in hereditary spherocytosis. Blood Cells Mol Dis 1998;24:539–543. Gasparini P, Miraglia del Giudice E, Delaunay J, et al. Localization of the congenital dyserythropoietic anemia II locus to chromosome 20q11.2 by genomewide search. Am J Hum Genet 1997;61: 1112–1116. Gore D, Chetty M, Fisher J, Nicolaou A, Stewart G. Familial pseudohyperkalaemia Cardiff: a mild version of cryohydrocytosis. Br J Haematol 2002;117:212–214. Grootenboer S, Schischmanoff PO, Laurendeau I, et al. Pleiotropic syndrome of dehydrated hereditary stomatocytosis, pseudohyperkalemia, and perinatal edema maps to 16q23-q24. Blood 2000;96: 2599–2605.

CHAPTER 85 / DISORDERS OF THE RED CELL MEMBRANE

Haines P, Crawley C, Chetty M, et al. Familial pseudohyperkalaemia Chiswick: a novel congenital thermotropic variant of K and Na transport across the human red cell membrane. Br J Haematol 2001;112:469–474. Haines P, Crawley C, Chetty M, et al. Familial pseudohyperkalaemia Chiswick. Br J Haematol 2001;112(2):469–474. Haines P, Jarvis H, King S, et al. Two further British families with the “cryohydrocytosis” form of hereditary stomatocytosis. Br J Haematol 2001;113:932–937. Hassoun H, Wang Y, Vassiliadis J, et al. Targeted inactivation of murine band 3 (AE1) gene produces a hypercoagulable state causing widespread thrombosis in vivo. Blood 1998;92:1785–1792. Hoekstra D, van IJzendoorn SC. In search of lipid translocases and their biological functions. Dev Cell 2003;4:8, 9. Iolascon A, Miraglia del Giudice E, Perrotta S, et al. Exclusion of three candidate genes as determinants of congenital dyserythropoietic anemia type II (CDA-II). Blood 1997;90:4197–4200. Iolascon A, Stewart GW, Ajetunmobi JF, et al. Familial pseudohyperkalemia maps to the same locus as dehydrated hereditary stomatocytosis (hereditary xerocytosis). Blood 1999;93:3120–3123. Lande WM, Thiemann PW, Mentzer WM. Missing band 7 membrane protein in two patients with high Na, low K erythrocytes. J Clin Invest 1982;70:1273–1280. Lavabre-Bertrand T, Blanc P, Navarro R, et al. Alpha-Interferon therapy for congenital dyserythropoiesis type I. Br J Haematol 1995;89: 929–932. Lock SP, Sephton Smith R, Hardisty RM. Stomatocytosis: a hereditary haemolytic anomaly associated with haemolytic anaemia. Br J Haematol 1961;7:303–314. Lux S, Palek J. Disorders of the red cell membrane. In: Handin R, Lux S, Stossel T, eds. Blood: Principles and Practice of Hematology. Philadelphia, PA: JB Lippincott, 1995; pp.1701–1818. Maillet P, Alloisio N, Morlé L, Delaunay J. Spectrin mutations in hereditary elliptocytosis and hereditary spherocytosis. Hum Mutat 1996;8: 97–107. Maréchal J, Wilmotte R, Kanzaki A, et al. Ethnic distribution of allele alpha LELY, a low-expression allele of red-cell spectrin alpha-gene. Br J Haematol 1995;90:553–556. Miller DR, Rickles FR, Lichtman MA, La Celle PL, Bates J, Weed RI. A new variant of hereditary hemolytic anemia with stomatocytosis and erythrocyte cation abnormality. Blood 1971;38:184–203. Mouro-Chanteloup I, Delaunay J, Gane P, et al. Evidence that the red cell skeleton protein 4.2 interacts with the Rh membrane complex member CD47. Blood 2003;101:338–344. Parquet N, Devaux I, Boulanger L, et al. Identification of three novel spectrin alpha I/74 mutations in hereditary elliptocytosis: further support for a triple-stranded folding unit model of the spectrin heterodimer contact site. Blood 1994;84:303–308.

837

Paw BH, Davidson AJ, et al. Cell-specific mitotic defect and dyserythropoiesis associated with erythroid band 3 deficiency. Nat Genet 2003; 34:59–64. Peters LL, Jindel HK, Gwynn B, et al. Mild spherocytosis and altered red cell ion transport in protein 4. 2- null mice. J Clin Invest 1999;103:1527–1537. Peters LL, Shivdasani RA, Liu SC, et al. Anion exchanger 1 (band 3) is required to prevent erythrocyte membrane surface loss but not to form the membrane skeleton. Cell 1996;86:917–927. Sheetz M, Singer SJ. Biological membranes as bilayer couples: a molecular mechanism of drug-erythrocyte interactions. Proc Natl Acad Sci USA 1974;71:4457. Shi ZT, Afzal V, Coller B, et al. Protein 4.1R-deficient mice are viable but have erythroid membrane skeleton abnormalities. J Clin Invest 1999;103:331–340. Snyers L, Umlauf E, Prohaska R. Association of stomatin with lipid-protein complexes in the plasma membrane and the endocytic compartment. Eur J Cell Biol 1999;78:802–812. Southgate CD, Chishti AH, Mitchell B, Yi SJ, Palek J. Targeted disruption of the murine erythroid band 3 gene results in spherocytosis and severe haemolytic anaemia despite a normal membrane skeleton. Nat Genet 1996;14:227–230. Stewart GW, Ellory JC. A family with mild xerocytosis showing increased cation permeability at low temperatures. Clin Sci 1985;69:309–319. Stewart GW, O’Brien H, Morris SA, Owen JS, Lloyd JK, Amess JAL. Stomatocytosis, abnormal platelets and pseudo-homozygous hypercholesterolaemia. Eur J Haematol 1987;38:376–380. Stewart GW, Amess JAL, Eber S, et al. Thrombo-embolic disease after splenectomy for hereditary stomatocytosis. Br J Haematol 1996;93: 303–310. Tamary H, Shalmon L, Shalev H, et al. Localization of the gene for congenital dyserythropoietic anemia type I to a