Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression: A Selection of Readings for Health Services Providers [1 ed.] 012824013X, 9780128240137

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression ― A Selection of Readings for Health Service

376 42 5MB

English Pages 246 [231] Year 2021

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression: A Selection of Readings for Health Services Providers [1 ed.]
 012824013X, 9780128240137

Table of contents :
Front-Matt_2021_Starving-Cancer-Cells--Evidence-Based-Strategies-to-Slow-Can
Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression
Copyrigh_2021_Starving-Cancer-Cells--Evidence-Based-Strategies-to-Slow-Cance
Copyright
Forewor_2021_Starving-Cancer-Cells--Evidence-Based-Strategies-to-Slow-Cancer
Foreword
Prefac_2021_Starving-Cancer-Cells--Evidence-Based-Strategies-to-Slow-Cancer-
Preface
Acknowledgme_2021_Starving-Cancer-Cells--Evidence-Based-Strategies-to-Slow-C
Acknowledgments
Chapter-1---Intro_2021_Starving-Cancer-Cells--Evidence-Based-Strategies-to-S
1 . Introduction
1.1 Why this book?
1.2 Alternate lifestyles of a cell
1.3 Medical cancer treatment in the United States-now and then
1.4 The epidemiology of cancer
1.4.1 Breast cancer
1.4.2 Colorectal cancer
1.4.3 Prostate cancer
1.4.4 Pancreatic cancer
1.4.5 Lung cancer
1.4.6 Leukemias
1.4.7 Kidney and renal pelvis cancers
1.4.8 Glioblastoma
1.4.9 Head and neck cancers
1.4.10 Multiple myeloma
1.5 References
1.5.1 Weblinks
1.6 Suggested additional reading
Chapter-2---The-metabolic-theory-of_2021_Starving-Cancer-Cells--Evidence-Bas
2 . The metabolic theory of cancer and its clinical implications
2.1 The treatment implications of the Warburg discovery
2.2 Cell metabolism 101. A brief review
2.3 When mitochondrial genes and nuclear genes uncouple: trouble in paradise
2.4 The principal current metabolic strategies
2.5 References
2.6 Suggested further reading
Chapter-3---Tumor-starvation-b_2021_Starving-Cancer-Cells--Evidence-Based-St
3 . Tumor starvation by L-arginine deprivation
3.1 Introduction
3.1.1 References
3.2 Kremer JC and Van Tine BA. 2017. Therapeutic arginine starvation in ASS1-deficient cancers inhibits the Warburg effect. Mol ...
3.2.1 Abstract
3.2.2 References
3.3 Patil MD, Bhaumik J, Babykutty S, Banerjee UC, and Fukumura D. 2016. Arginine dependence of tumor cells: targeting a chink ...
3.3.1 Abstract
3.3.2 Introduction
3.3.3 Enzyme-mediated arginine deprivation: a potential anti-cancer approach
3.3.4 Arginine deiminase
3.3.5 PEGylated ADI
3.3.6 Tumor sensitivity toward ADI
3.3.7 Tumor resistance toward ADI
3.3.8 Anti-tumor mechanisms of ADI treatment
3.3.8.1 Role of autophagy and apoptosis in ADI-mediated arginine deprivation therapy
3.3.9 Inhibition of de novo protein synthesis by ADI-mediated arginine deprivation
3.3.10 Anti-angiogenic effects of ADI-mediated arginine deprivation
3.3.11 Arginase
3.3.12 PEGylated recombinant human arginase I
3.3.13 Anti-tumor mechanisms of arginase-mediated arginine deprivation
3.3.14 Role of autophagy in arginase-mediated arginine deprivation
3.3.15 Role of apoptosis in arginase-mediated arginine deprivation
3.3.16 Arginine decarboxylase
3.3.17 Concluding remarks
3.3.18 References
3.4 Zou S, Wang X, Liu P, Ke C, and Xu S. 2019. Arginine metabolism and deprivation in cancer therapy. Biomedicine and Pharmaco ...
3.4.1 Abstract
3.4.2 Introduction
3.4.3 Metabolism of arginine
3.4.3.1 Endogenous synthesis of arginine
3.4.3.2 Arginine degradation
3.4.3.3 Regulation of arginine metabolism
3.4.3.4 Function of arginine in health and disease
3.4.4 Arginine and cancer therapy
3.4.4.1 Enzymatic agents on arginine depletion
3.4.4.2 Functions of arginine metabolites in cancer development
3.4.4.2.1 Polyamines
3.4.4.2.2 Nitric oxide
3.4.4.2.3 Agmatine
3.4.4.2.4 Glutamine and proline
3.4.5 Preclinical studies on arginine depletion in cancer treatment
3.4.6 Clinical studies on arginine depletion in cancer treatment
3.4.7 Drug resistance in arginine deprivation therapeutics
3.4.8 Conclusions
3.4.9 References
3.5 Riess C, Shokraie F, Classen CF, Kreikemeyer B, Fiedler T, Junghanss C, and Maletzki C. 2018. Arginine-depleting enzymes - ...
3.5.1 Synopsis
3.6 Burki TK. 2016. Arginine deprivation for ASS1-deficient mesothelioma. Lancet Oncology, Oct 1; 17(10): e423. DOI: https://do ...
3.6.1 Abstract
3.7 Qiu F, Chen YR, Liu X, Chu CY, Shen LJ, Xu J, Gaur S, Forman HJ, Zhang H, Zheng S, Yen Y, Huang J, Kung HJ, and Ann DK. 201 ...
3.7.1 Abstract
3.8 Kim RH, Bold RJ, and Kung H-J. 2009. ADI, autophagy and apoptosis: Metabolic stress as a therapeutic option for prostate ca ...
3.8.1 Abstract
3.9 Wheatley DN, Campbell E, Lai PBS, and Cheng PNM. 2005. A rational approach to the systemic treatment of cancer involving me ...
3.9.1 Synopsis
Chapter-4---L-canavanine-depri_2021_Starving-Cancer-Cells--Evidence-Based-St
4 . L-canavanine deprives tumors of L-arginine
4.1 Introduction
4.2 Rosenthal GA and Nkomo P. 2000. The natural abundance of L-canavanine, an active anticancer agent in alfalfa, medicago sati ...
4.2.1 Abstract
4.3 Rosenthal GA. 1977. The biological effects and mode of action of L-canavanine, a structural analogue of L-arginine. The Qua ...
4.3.1 Introduction
4.4 Jacobi B, Stroeher L, Leuchtner N, Echchannaoui H, Alexander Desuki A, Kuerzer L, Habermeier A, Antunes E, Amann E, John Bo ...
4.4.1 Abstract
4.4.2 Methods
4.4.3 Results
4.4.4 Conclusion
4.5 Windschmitt J, Jacobi B, Bülbül Y, Sester L, Tappe J, Hiebel C, Behl C, Theobald M, and Munder M. 2018. Arginine ...
4.5.1 Introduction
4.5.2 Methods
4.5.3 Results
4.5.4 Conclusions
4.6 Rosenthal GA. 1998. L-Canavanine: A potential chemotherapeutic agent for human pancreatic cancer. (Ph.D. Thesis, 1997, Univ ...
4.6.1 Abstract
4.6.2 Introduction
4.6.3 Biochemical basis for canavanine antimetabolic properties
4.6.4 Article of human diet
4.6.5 Canavanine antineoplastic activity
4.6.6 Canavanine cytotoxic effect on human pancreatic cells
4.6.7 Combination therapy
4.6.8 Canavanine accumulative toxicity
4.6.9 Canavanine derivatives as chemotherapeutic agents
4.6.10 Conclusions
4.6.11 References
4.7 Swaffar DS, Ang CY, Desai PB, and Rosenthal GA. 1994. Inhibition of the growth of human pancreatic cancer by the arginine a ...
4.7.1 Summary
4.8 Ding Y, Matsukawa Y, OhtaniFujita N, Kato D, Dao S, Fujii T, Naito Y, Yoshikawa T, Sakai T, and Rosenthal GA. 1999. Growth ...
4.8.1 Abstract
4.9a Nurcahyanti ADR and Wink M. 2017. L-Canavanine potentiates cytotoxicity of chemotherapeutic drugs in human breast cancer ce ...
4.9a.1 Abstract
Objectives
Method
Results
Conclusion
Reference
4.10a Vynnytska-Myronovska B, Bobak Y, Garbe Y, Dittfeld C, Stasyk O, and Kunz-Schughart LA. 2012. Single amino acid arginine sta ...
4.10a.1 Abstract
4.11 Cautionary note
Chapter-5---Glucose-deprivatio_2021_Starving-Cancer-Cells--Evidence-Based-St
5 . Glucose deprivation and fasting strategies
5.1 Introduction
5.1.1 Reference
5.2 Muti P, Quattrin T, Grant BJB, Krogh V, Micheli A, Schünemann HJ, Ram M, Freudenheim, JL, Sieri S, Trevisan M, and Ber ...
5.2.1 Abstract
5.3 Ye H, Adane B, Khan N, Alexeev E, Nusbacher N, Minhajuddin M, Stevens BM, Winters AC, Lin X, Ashton JM, Purev E, Xing L, Po ...
5.3.1 Summary
5.3.2 Introduction
5.3.3 Results
5.3.3.1 Leukemia induces IR and reduces serum insulin level
5.3.3.2 Adipose-derived IGFBP1 induces the development of IR in leukemia
5.3.3.3 Modulation of IGFBP1 mediates leukemia growth in vivo
5.3.3.4 Loss of active GLP-1 and serotonin contributes to inhibition of insulin secretion in leukemia pathogenesis
5.3.4 Leukemia-associated microbiota facilitates disease progression
5.3.5 Microbiota-derived short-chain fatty acids impede leukemia progression and are reduced in leukemic mice
5.3.6 Restoring systemic glucose metabolism provides survival benefits
5.3.7 Human leukemia induces an insulin-resistant phenotype
5.3.8 Discussion
5.3.9 References
5.4 Grasmann G, Smolle E, Olschewski H, and Leithner K. 2019. Gluconeogenesis in cancer cells - Repurposing of a starvation-ind ...
5.4.1 Abstract
5.5 Sun S, Sun Y, Rong X, and Bai L. 2019. High glucose promotes breast cancer proliferation and metastasis by impairing angiot ...
5.5.1 Abstract
5.5.2 Introduction
5.5.3 Method and materials
5.5.3.1 Cell culture
5.5.3.2 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay
5.5.4 Cell counting
5.5.5 Transwell assay
5.5.5.1 Soft agar assay
5.5.5.2 Western blot
5.5.5.3 Real-time polymerase chain reaction
5.5.5.4 Statistical analysis
5.5.6 Results
5.5.6.3 High glucose inhibits AGT expression
5.5.6.4 AGT inhibits breast cancer proliferation and metastasis
5.5.6.5 Overexpression of AGT inhibits the influence of high glucose on cell proliferation and metastasis
5.5.7 Discussion
5.5.8 Conclusion
Abbreviations
5.5.9 References
5.6 Han L, Ma Q, Li J, Liu H, Li W, Ma G, Xu Q, Zhou S, and Wu E. 2011. High glucose promotes pancreatic cancer cell proliferat ...
5.6.1 Abstract
5.7 VanHook AM. 2018. Starving cancer cells to death. Science Signaling, Aug; 11(542): eaau9719. DOI: https://doi.org/10.1126/s ...
5.7.1 Summary
5.7.2 References
5.8 Yu S, Chen Z, Zeng X, Chen X, and Gu Z. 2019. Advances in nanomedicine for cancer starvation therapy. Theranostics, Oct; 9( ...
5.8.1 Abstract
5.9 Simone BA, Champ CE, Rosenberg AL, Berger AC, Monti DA, Dicker AP, and Simone NL. 2013. Selectively starving cancer cells t ...
5.9.1 Abstract
5.10 de Groot S, Hanno Pijl H, van der Hoeven JJM, and Kroep JR. 2019. Effects of short-term fasting on cancer treatment. Journa ...
5.10.1 Abstract
5.11 Lee C and Longo VD. 2011. Fasting vs dietary restriction in cellular protection and cancer treatment: from model organisms ...
5.11.1 Abstract
5.12 Brandhorst S and Longo VD. 2016. Fasting and caloric restriction in cancer prevention and treatment. Recent Results in Canc ...
5.12.1 Introduction
5.13 Hou Y, Zhou M, Xie JM, Chao P, Feng Q, and Wu J. 2017. High glucose levels promote the proliferation of breast cancer cells ...
5.13.1 Summary
5.14 Naveed S, Aslam M, and Ahmad A. 2014. Starvation based differential chemotherapy: A novel Approach for cancer treatment. Om ...
5.14.1 Summary
Chapter-6---Tumor-starvation-by-de_2021_Starving-Cancer-Cells--Evidence-Base
6 . Tumor starvation by deprivation of glutamine and aspartate
6.1 Glutamine and aspartate
6.2 Choi Y-K and Park K-G. 2018. Targeting glutamine metabolism for cancer treatment. Biomolecules and Therapeutics (Seoul), Ja ...
6.2.1 Synopsis
6.3 Jiang J, Srivastava S, and Zhang J. 2019. Starve cancer cells of glutamine: Break the spell or make a hungry monster? Cance ...
6.3.1 Abstract
6.3.2 Introduction
6.3.3 Glutamine, a versatile biosynthetic substrate
6.3.4 Non-biosynthetic role of glutamine
6.3.5 Glutamine starvation: An experimental condition or pathophysiological stress?
6.3.6 Influence of other amino acids on glutamine starvation
6.3.6.1 Asparagine
6.3.6.2 Aspartate and arginine
6.3.6.3 Cystine
6.3.7 What is the critical limiting metabolite during glutamine starvation?
6.3.8 Key variants impacting the definition of critical limiting metabolite
6.3.9 Complexity of glutamine starvation in tumors in vivo
6.3.10 Therapeutic implication
6.3.11 Conclusions
6.3.12 References
6.4 Garcia-Bermudez J, Baudrier L, La K, Zhu XG, Fidelin J, Sviderskiy VO, Papagiannakopoulos T, Molina H, Snuderl M, Lewis CA, ...
6.4.1 Abstract
6.5 Sullivan LB, Luengo A, Danai LV, Bush LN, Diehl FF, Hosios AM, Lau AN, Elmiligy S, Malstrom S, Lewis CA, Vander Heiden MG. ...
6.5.1 Abstract
6.6 Xie G, Zhou B, Zhao A, Qiu Y, Zhao X, Garmire L, Shvetsov YB, Yu H, Yen Y, and Jia W. 2015. Lowered circulating aspartate i ...
6.6.1 Abstract
Index_2021_Starving-Cancer-Cells--Evidence-Based-Strategies-to-Slow-Cancer-P
Index
A
B
C
D
E
F
G
H
I
K
L
M
N
O
P
R
S
T
U
V
W

Citation preview

STARVING CANCER CELLS: EVIDENCE-BASED STRATEGIES TO SLOW CANCER PROGRESSION A Selection of Readings for Health Services Providers ROBERT FRIED Emeritus professor, Doctoral Faculty, Behavioral Neuroscience, City University of New York (CUNY), New York, NY, United States

RICHARD M. CARLTON Integrative physician, Port Washington, NY, United States

DENNIS A. FRIED Assistant Professor, Department of Epidemiology, Rutgers School of Public Health, Newark, NJ; Health Science Specialist, U.S. Department of Veterans Affairs, War-Related Illness and Injury Study Center, East Orange, NJ, United States

Academic Press is an imprint of Elsevier 125 London Wall, London EC2Y 5AS, United Kingdom 525 B Street, Suite 1650, San Diego, CA 92101, United States 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom Copyright © 2021 Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier. com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN: 978-0-12-824013-7 For information on all Academic Press publications visit our website at https://www.elsevier.com/books-and-journals Cover image: Used with permission from Scientific Animations Inc. Publisher: Stacy Masucci Acquisitions Editor: Rafael Teixeira Editorial Project Manager: Kristi Anderson Production Project Manager: Punithavathy Govindaradjane Cover Designer: Christian Bilbow

Typeset by TNQ Technologies

Foreword I am honored to write the foreword to this important book by Drs. Fried, Carlton, and Fried titled, “Starving Cancer Cells.” Cancer persists as a major cause of morbidity and mortality throughout the world. While extensive genetic heterogeneity exists among the tumor cells within the various cancers, most if not all cancer cells express abnormalities in the number, structure, and function of their mitochondria. These abnormalities cause tumor cells to rely more heavily on fermentation than on oxidative respiration for synthesizing ATP. As most of the energy in a cell is needed to maintain the activity of membrane ionic pumps, global cellular dysfunction and ultimately organ and systems failure will arise if energy flow to the pumps is disrupted. Hence, chemical energy by itself is the central issue for cell viability. Without ATP no cell can survive. Cancer cells obtain their energy for survival and growth largely through the fermentation of the sugar glucose and the amino acid glutamine. The dependence on glucose and glutamine fermentation for ATP synthesis underlies the ability of tumor cells to live in hypoxic environments and to resist death from conventional radiation, chemo-, and immunotherapies. When exposed to oxygen, tumor cells produce excessive levels of reactive oxygen species (ROS) that are carcinogenic and mutagenic. The ROS not only enhance damage to respiration and cause a greater dependency on fermentation but also produce genomic instability and the vast numbers of somatic mutations found in most cancers. In other words, the nuclear gene mutations seen in tumors arise as downstream effects of respiratory dysfunction. The protracted transition from respiration to fermentation can arise from any number of risk factors including advancing age, oncogenic viruses, intermittent hypoxia, systemic inflammation, environmental carcinogens, and rare inherited mutations. It is also interesting that cancer is very rare in chimpanzees despite sharing almost 99% gene and protein sequence identity with humans. Breast cancer kills over 41,000 American women each year in the United States, but has never been documented in a female chimpanzee. Unlike chimpanzees, humans have deviated dramatically from ancestral diets and life styles. Hence, it is diet and life style issues that put us at risk for cancer. In recognizing cancer as a mitochondrial metabolic disease, nontoxic therapeutic strategies become a logical alternative to current standards xiii

xiv

Foreword

of care, which are often toxic, overly expensive, and only marginally effective. Only those cells with a flexible genome will be capable of surviving abrupt changes in metabolic landscape. The adaptation to environmental extremes is conserved within the genome according to the ecological instability theory of Rick Potts, Director of Human Origins at the Smithsonian Institution. Eons of radical changes in the external physical environment have provided normal cells in the human body with an exceptional ability for metabolic adaptation. The notion that tumor cells are more versatile and adaptable than nontumor cells is illogical in the context of evolutionary biology. The multiple genomic defects in cancer cells would limit their flexibility thus making them less adaptable to metabolic stress and therefore vulnerable to elimination through principles of metabolic starvation. Ketogenic metabolic therapy, using calorie-restricted high fat, low carbohydrate ketogenic diets, is a therapeutic strategy that can be effective for starving cancer cells. This approach shifts the body away from using glucose to using ketone bodies and fats for energy, thus starving cancer cells of their prime fermentable fuel. The dependency of tumor cells on glucose has become known as the Warburg effect, in recognition of Otto Warburg who first described cancer as a disease of mitochondrial energy metabolism. As ketone bodies and fats are nonfermentable, these fuels are accessible only to normal cells, which contain normal mitochondrial function. We found that ketogenic metabolic therapy used with drugs that also target glutamine can be effective in managing advanced-stage brain cancer. The key is to simultaneously restrict availability of both glucose and glutamine. As most if not all cancers depend on glucose and glutamine fermentation for survival, this therapeutic starvation strategy should be effective in managing most if not all cancers. In their book, Drs. Fried, Carlton, and Fried review well-referenced published studies on metabolic starvation strategies for the potential management and elimination of tumor cells. The information in this book is valuable, as it highlights the vulnerability of tumor cells to metabolic stress. The first two chapters provide important background information on the rationale for how starving cancer cells of growth metabolites is a logical therapeutic strategy for management. The authors review studies in Chapters 3 and 4 on how arginine deprivation, using drugs and diets, results in a significant shift in tumor cell metabolic phenotype thus inhibiting the Warburg effect. They highlight the importance of combinatorial approaches in arginine targeting to achieve maximal therapeutic benefit. Chapter 5 focuses on intermittent fasting and therapeutic strategies for targeting glucose

Foreword

xv

availability to tumor cells. It is well known that water-only fasting has had success in cancer management. The information in this chapter should be viewed together with the many YouTube videos from Mr. Guy Tenenbaum who discusses with Dr. Eric Berg how he resolved his advanced prostate cancer using intermittent fasting. Chapter 6 addresses the importance of glutamine and aspartate deprivation for cancer management. Although glutamine is a nonessential amino acid, glutamine can become an important growth fuel for many cancers. Strategies to target glutamine must be done carefully, however, as glutamine is critical for the urea cycle, the immune system, and gut health. The authors discuss several studies on the challenges encountered in targeting glutamine for cancer management. While some tumor cells might adapt to glutamine starvation, it is less likely that they can adapt under conditions where the availability of both glucose and glutamine is limited. Hence, therapeutic strategies that can simultaneously restrict glucose and glutamine availability should be effective in managing most cancers regardless of cell or tissue origin. The authors have done an excellent job in reviewing information on the success and challenges in attempting to starve tumor cells of essential metabolites needed for their survival. Much of the information is distilled down from detailed studies published in scientific journals that might be inaccessible to many physicians and to cancer patients. As the current treatment strategies for cancer management have done little in reducing the annual death rates for decades, it is now time to consider alternative and complementary approaches to management based on the underlying metabolic origin of the disease. Starving Cancer Cells takes a bold step in this new direction and can help practicing physicians and their cancer patients make more informed decisions on treatment options. Thomas N. Seyfried, PhD Biology Department, Boston College, Chestnut Hill, MA, Unites States Author: Cancer as a Metabolic Disease: On the Origin, Management, and Prevention of Cancer. Hoboken: John Wiley & Sons, 2012.

Preface The information about evidence-based cancer deprivation strategies that has been published in medical journals seems to be largely unknown to most medical clinical cancer and related nutrition and other health sciences professionals. Yet, there is a growing amount of it. Of course, 20 years is not a long time in the history of science, and most of these publications appeared within the past 20 years. We would love to be able to present all of them here but there are so many of them now that this is not possible. Yet, it is important for those interested in this aspect of cancer and its treatment to see at least some of them in full. And so, we facilitate that by selecting and representing here those that seem most representative of the complexity and thoroughness of the “science” of cancer cell starvation. Beginning with Chapter 3. Tumor Starvation by L-Arginine Deprivation, each section features one or more full-length journal articles and a number of abstracts. This by no means implies that those only abstracted here are in any way less meritorious. Also, you will find more citations in connection with those deprivation substances that have received the most scrutiny. To help you to follow up on research or clinical findings, we converted the references in the full-length journal articles from sundry journal or any other styles to a more or less AMA style. Also, as has been our practice in past publications, we list all authors of publications wherever possible. We believe that author abbreviations are a generally arbitrary convention intended to save journal space and do not serve the science community. We hold that all authors are equally important and they should be cited. In addition, we name journals in full. The references should be cited in a way that makes it easy for readers to find them should they wish to do so. We have edited journal within-text reference citations to cite at least three authors to facilitate retrieval, and we also use that same citation format in our chapters and section prologues. Wherever possible, we also provide the Digital Object Identifier (DOI) of publications simplifying online reference retrieval. In most cases, the DOI is all that is needed to find a reference. Unfortunately, a small number of journal publications do not have a DOI. However, in some cases, there is an NIH public access PMC or PMCID number.

xvii

xviii

Preface

Also, we have adopted a numbering system that supersedes withinjournal section number (where those appear) so as to facilitate your access to the information. Finally, some of you may wonder why this collection of readings was not assembled by experts in the field of oncology. We wondered also why they did not do it. It is monumental science and its story should be told. So, we took it upon ourselves to do it in their stead with the hope that we do it justice. Robert Fried, PhD, New York City, NY Richard M. Carlton, MD, Port Washington, NY Dennis A. Fried, PhD, East Brunswick, NJ

Acknowledgments We wish to express our sincere appreciation to Ms. Stacy Masucci, Publisher, Translational Medicine, ELSEVIER, for her enthusiastic support of this project, as well Mr. Rafael Teixeira, Elsevier Reference books for Cancer Research/Oncology, our acquisition editor, for helping this project along. Finally, we are especially grateful to Ms. Kristi Anderson, Senior Editorial Project Manager, ELSEVIER/Academic Press, for her unwavering professional (and good humored) help with the production of this book. We also wish to thank the following journal and book sourcesdin alphabetical orderdfor permission to reproduce text materials in whole or in part, with figures and tables, or abstracts: Agents in Medicinal Chemistry, American Cancer Society, Annals of Surgery, Autophagy, Biochimica et Biophysica Acta. Reviews on Cancer, Biomedicine and Pharmacotherapy, Biomolecules and Therapeutics (Seoul), Bioscience Reports, Blood, Breast Cancer (Dove Med Press), Cancer (Basel), Cancer Cell, Cancer Epidemiology, Biomarkers and Prevention. Cancer Research, Cellular Physiology and Biochemistry, Frontiers in Cell and Developmental Biology, Future Oncology, Gene Therapy and Molecular Biology, International Journal of Cancer, JAMA Oncology, Japanese Journal of Cancer Research, Journal of Experimental and Clinical Cancer Research, Lancet. Oncology, Metabolism Clinical and Experimental, Metabolism in Cancer. Recent Results in Cancer Research, Molecular Cellular Oncology, National Geographic, Nature Cell Biology, Oman Medical Journal, Oncogene, Oncotarget, Pharmacological Biology, PLoS One, Science Signaling, The Quarterly Review of Biology, Theranostics.

xix

CHAPTER 1

Introduction

Figure 1.1 Micro-CT image showing a tumor in an ancient human toe bone. (Photo by P. Randolph-Quinney, Northumbria University; Weblink 1. Reprinted with permission.)

The illustration (Figure 1.1) is fossil evidence that cancer has been part of the human experience from prehistory. The word “cancer” is credited to Hippocrates (460e370 BC), who used the term carcinos, meaning crab, to describe nonulcer-forming and ulcer-forming tumors. Celsus (28e50 BC) translated the Greek term into cancer, the Latin word for crab, while Galen (130e200 AD) used the word oncos (Greek word for swelling) to describe tumors. Although the crab analogy of Hippocrates and Celsus is still used to describe malignant tumors, Galen’s term is now used as a part of the name for cancer specialists, “oncologists.”

1.1 Why this book? It is generally the case that subspecialties in medical science publish their research findings in journals that fairly narrowly represent research and clinical applications in that specialty. These journals therefore constitute a subcommunity of scientific research publications that serve a very parochial readership of investigators and clinicians that rarely reaches outside that sphere. It can be said that these scientists talk to each other, but rarely to any Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression ISBN 978-0-12-824013-7 https://doi.org/10.1016/B978-0-12-824013-7.00002-X

© 2021 Elsevier Inc. All rights reserved.

1

2

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

“outsiders.” These researchers and clinicians, expert in this field, communicate their astonishing findings to each other but typically not to the science community at large. Many medical professionals whose patients could benefit from implementation of tumor starvation strategies are unlikely to be subscribers to journals of that subcommunity that feature findings that relate to tumor cell auxotrophy and starvation strategies. Here are just a dozen examples of these journals: • Advances in Medical Sciences (2019) • Biochimica et Biophysica Acta - Reviews on Cancer (2019) • Journal of Enzyme Inhibition and Medical Chemistry (2000) • Blood (2015) • Cancer Research (2006) • Genetic Engineering and Biotechnology News (2019) • Journal of Hematology and Oncology (2010) • Nature Cell Biology (2018) • Nutrition and Metabolism (2010) • Oncogene (2016) • Pharmacological Biology (2000) • Science Signaling (2014) Even the title of the journal Nutrition and Metabolism does not hint at tumor “starvation.” Furthermore, most professionals have neither the time nor the training to launch extensive electronic library or other databases of research findings to learn about only possibly helpful information that is neither presently “mainstream” nor part of their treatment protocols. This is the case even though tumor starvation strategy findings are published in not only conventional but actually prestige medical journals. Our aim is to correct this situation and to open a window, as it were, to this information for the broader health sciences community who may be totally unaware that it exists. This book aims to provide a sample of this research in the hope that it may whet the appetite of practitioners for a more focused look at the possibilities it unfolds: • It is intended to be informative, but by no means comprehensive. • It collects representative research publications and categorizes them according to major subthemes. • It is intended also for readers with a basic knowledge of “science,” but who are not experts in cancers.

Introduction

3

• It intends to shed an early light on a relatively new approach to our understanding of the cancer cell idiosyncratic metabolic dysfunction, and on evidence-based new treatment strategies derived from that understanding. • It is intended to inform healthcare providers who treat cancer patients about nutritional adjunctive guidelines that can support conventional treatment. The title was chosen to make it clear to the potential reader that this is a book about research and clinical findings in the conventional medicine framework, published in conventional medical journals, and not a book about yet another “quick fix,” a novel complementary treatment, or about anecdotal observations.

1.2 Alternate lifestyles of a cell All cells in complex multicellular organisms require oxygen and nutrients to create the energy that powers their life functions. Recent findings point to a shift to anaerobic metabolism in cancer cells. Indeed, we are witnessing a revolution in understanding the fundamental nature of cancer(s) and, therefore, in the strategies devised to combat it. This revolution centers on the perception that a distinguishing feature of cancer is abnormal cell metabolism. As the pioneers Seyfried and Shelton put it: Emerging evidence indicates that impaired cellular energy metabolism is the defining characteristic of nearly all cancers regardless of cellular or tissue origin. In contrast to normal cells, which derive most of their usable energy from oxidative phosphorylation, most cancer cells become heavily dependent on substrate level phosphorylation to meet energy demands. Evidence is reviewed supporting a general hypothesis that genomic instability and essentially all hallmarks of cancer, including aerobic glycolysis (Warburg effect), can be linked to impaired mitochondrial function and energy metabolism [1].

Indeed, many of the publications show that amino acid auxotrophy targets mitochondrial function. In other words, while most cells derive energy from mitochondrial oxidative metabolism, distortion of the metabolism of cancer cells causes them in many cases to rely on a less effective form, anaerobic glycolysis, where glucose is converted to lactate in the presence of oxygen (but without using oxygen). In consequence of this shift, their basic nutrient needs changes. This metabolic shift is known as the “Warburg effect” [2]. Why this shift? It is proposed that the metabolism of cancer cells, and indeed all proliferating cells, is adapted to facilitate the uptake and

4

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

incorporation of nutrients into the biomass (e.g., nucleotides, amino acids, and lipids) needed to produce a new cell. Supporting this idea are recent studies showing that • several signaling pathways implicated in cell proliferation also regulate metabolic pathways that incorporate nutrients into biomass; and that • certain cancer-associated mutations enable cancer cells to acquire and metabolize nutrients in a manner conducive to proliferation rather than efficient ATP production [2]. In common terms, whatever it is that caused the mutation resulting in a tumor, the cell now has a different mission in lifedaccelerated proliferationdand that mission is sustained by a shift in metabolism that favors it. This book will present an annotated and edited selection of research and clinical publications in the general context of cancer metabolic theory and amino acid and other nutrient substance deprivation therapy.

1.3 Medical cancer treatment in the United Statesdnow and then A number of conventional medical cancer treatment options are now available to cancer sufferers in the United States. The National Cancer Institute (NCI) lists them on their website: • Surgery: When used to treat cancer, surgery is a procedure in which a surgeon removes cancer from the body. • Radiation therapy: Radiation therapy uses high doses of radiation to kill cancer cells and shrink tumors. • Chemotherapy: Chemotherapy uses drugs to kill cancer cells. • Immunotherapy: Immunotherapy helps the immune system fight cancer. • Targeted therapy: Targeted therapy targets the changes in cancer cells that help them grow, divide, and spread. • Hormone therapy: Hormone therapy slows or stops the growth of breast and prostate cancers that use hormones to grow. • Stem cell transplant: Stem cell transplants are procedures that restore blood-forming stem cells in cancer patients who have had theirs destroyed by very high doses of chemotherapy or radiation therapy. • Precision medicine: Precision medicine helps doctors select treatments that are most likely to help patients based on a genetic understanding of their disease [Weblink 2]. There are a number of other medical treatment strategies, but they are not presently in common practice. For instance, there are viruses that are observed

Introduction

5

to infect and kill tumor cells. Known as oncolytic viruses, this group includes viruses found in nature as well as viruses modified in the laboratory to reproduce efficiently in cancer cells without harming healthy cells. Since the late 1800s, it was noted that some patients with cancer are in remission, if only temporarily, after a viral infection. Today, several dozen virusesdand a few strains of bacteriadare being studied as potential cancer treatments according to research presented at an NCI-sponsored conference on using microbes as cancer therapies in 2017 [3]. For this reason, some researchers consider oncolytic viruses to be a form of immunotherapyda treatment that harnesses the immune system against cancer. But it is generally agreed that more studies are needed to learn how different oncolytic viruses work against cancer. Although the notion of using viruses in cancer therapy is not new, the science only began to move forward in the 1990s with advances in genetic engineering technology and according to M. Gromeier of the Duke Cancer Institute, who has led clinical trials of a genetically modified form of poliovirus, “Today, viruses are firmly established as a potential option to enhance and mediate immunotherapy” [Weblink 3]. Another approach is the use of bacteria in cancer therapy. Bacterial therapy for cancer was recognized a century ago: live, attenuated, or genetically modified obligate or facultative anaerobic bacterial species can colonize tumors and, by multiplying selectively inside the tumors, inhibit cancerous growths. The bacteria and their spores are used in the targetspecific therapies, delivering the prodrugs and the various proteins to the tumors. Bacterial treatment of cancer is providing new perspective in the treatment of disease, but the use of microorganisms to target tumors has certain complications including safety, genetic instability, and the confounded interaction of the bacteria with treatment drugs [4]. There was also a Tumor Hyperthermia treatment modality. Tumor therapy relied on the observation that high temperatures, as high as 113 F, can damage and even kill cancer cells, usually with minimal injury to normal tissues [Weblink 4] [5,6]. And, there were other treatment strategies including some nonconventional such as amygdalins/vitamin B17. Amygdalin is a cyanogenic glycoside compound found in the pits or seeds of apricots, apples, peaches, plums, red cherries, and other fruits. It is also found in bitter almonds. A partly man-made, purified form of amygdalin, known as laetrile, was patented in the 1950s and became a popular alternative cancer treatment during the 1960s and 1970s. It is now banned by the FDA and has not been available in the United States since 1980.

6

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

The WebMD website reports about amygdalin that “The way your intestines break it down makes cyanide, which supposedly kills harmful cancer cells” [Weblink 5]. That is only partly correct. Digestive enzymes do cause the formation of cyanide from amygdalin, but in sublethal doses, it is more likely that the subsequent formation of nitric oxide (NO) may account for any effects it may have shown against certain tumors. Nitric oxide is currently being investigated for its potential contribution to cancer therapy [7,8]. Medical applications of cyanide are not new. In the early 20th century, ca. 1918, investigators in Wisconsin, while looking into the safety of cough syrups containing cyanide reported the astonishing observations of restored episodes of normal cognition and memory in chronic, otherwise unresponsive, catatonic mental patients. Clinical doses of cyanide stimulates both systemic and cellular respiration and apparently brain cells fancy that [9]. Parenthetically, they were investigating that cyanide-containing cough syrup at the time of the Great Flu Epidemic. They may have been concerned that more folk were dying after consuming that cough syrup. It is difficult to determine exactly how successful medical cancer treatments are in the United States, and the statistics that describe its prevalence here are staggering.

1.4 The epidemiology of cancer Cancer is the second leading cause of death in the United States, exceeded only by heart disease. One of every four deaths in the United States is due to cancer, according to the Centers for Disease Control and Prevention (CDC). Overall, in 2016, the latest year for which incidence data are available, 1,658,716 new cases of cancer were reported, and 598,031 people died of cancer in the United States. For every 100,000 people, 436 new cancer cases were reported and 156 died of cancer [Weblink 6]. In 2020 for all sites, there will be an estimated 1,806,590 (48% males) new cancer cases and 606,520 (53% males) deaths [10].

1.4.1 Breast cancer Breast cancer is the second most common cancer among women in the United States; some kinds of skin cancer are the most common. Black women and white women get breast cancer at about the same rate, but black women die from breast cancer at a higher rate than white women, according to the CDC [Weblink 7]. Overall, in 2016, the latest year for

Introduction

7

which incidence data are available, 245,299 new cases of female breast cancer were reported, and 41,487 women died of female breast cancer in the United States. For every 100,000 women, 124 new female breast cancer cases were reported and 20 died of cancer (CDC) [Weblink 6]. In 2020, there will be an estimated 279,100 (99% females) new cases and 42,690 deaths (98% females) [10].

1.4.2 Colorectal cancer Excepting some kinds of skin cancer, colorectal cancer is the third leading cause of cancer-related deaths in the United States. It is the third most common cancer in men and in women [Weblink 8]. Overall, in 2016, the latest year for which incidence data are available, 141,270 new cases of colon and rectum cancer were reported, and 52,286 people died of colon and rectum cancer in the United States. For every 100,000 people, 37 new colon and rectum cancer cases were reported and 14 died of cancer according to the CDC [Weblink 6]. In 2020, there will be an estimated 147,950 (53% males) new cases and 53,200 deaths (54% males) [10].

1.4.3 Prostate cancer Excepting nonmelanoma skin cancer, prostate cancer is the most common cancer among men in the United States. It is also one of the leading causes of cancer death among men of all races and Hispanic origin populations (CDC) [Weblink 9]. Overall, in 2016, the latest year for which incidence data are available, 1,658,716 new cases of cancer were reported, and 598,031 people died of cancer in the United States. For every 100,000 people, 436 new cancer cases were reported and 156 died of cancer (CDC) [Weblink 6]. In 2020, there will be an estimated 191,930 new cases and 33,330 deaths [10].

1.4.4 Pancreatic cancer Overall, in 2016, the latest year for which incidence data are available, 49,093 new cases of pancreatic cancer were reported, and 42,757 people died of this disease in the United States. For every 100,000 people, 13 new pancreatic cancer cases were reported, and 11 died of cancer (CDC) [Weblink 6]. In 2020, there will be an estimated 57,600 (53% male) new cases and 47,050 (52% males) deaths [10].

8

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

1.4.5 Lung cancer Overall, in 2016, the latest year for which incidence data are available, 218,229 new cases of lung and bronchus cancer were reported, and 148,869 people died of lung and bronchus cancer in the United States. For every 100,000 people, 56 new lung and bronchus cancer cases were reported and 39 died of cancer (CDC) [Weblink 6]. In 2020, there will be an estimated 228,820 (51% males) new cases and 135,720 (53% males) deaths [10].

1.4.6 Leukemias Overall, in 2016, the latest year for which incidence data are available, 48,082 new cases of leukemia were reported, and 23,287 people died of leukemia in the United States. For every 100,000 people, 13 new leukemia cases were reported and 6 died of cancer (CDC) [Weblink 6]. In 2020, there will be an estimated 60,530 (59% males) new cases and 23,100 (58% males) deaths [10]. In 2020 also, acute lymphocytic (ALL) leukemia is expected to account for 10% of new leukemia cases and 6.5% of leukemia deaths, while acute myeloid leukemia (AML) is expected to account for 33% of new leukemia cases and 48% of leukemia deaths [10].

1.4.7 Kidney and renal pelvis cancers Overall, in 2016, the latest year for which incidence data are available, 63,639 new cases of kidney and renal pelvis cancer were reported, and 13,842 people died of kidney and renal pelvis cancer in the United States. For every 100,000 people, 17 new kidney and renal pelvis cancer cases were reported and 4 died of cancer (CDC) [Weblink 6]. In 2020, there will be an estimated 73,750 (62% males) new cases and 14,830 (66% males) deaths [10].

1.4.8 Glioblastoma Brain tumors are the eighth most common cancers overall among persons 40þ years old; eighth most common among men; and fifth most common among women in this age group. Brain tumors are the fifth leading cause of cancer-related death in men 40e59 years old. The five-year relative overall survival rate for adults older than 40 years diagnosed with a brain tumor is only 21.3% according to the National Brain Tumor Society [Weblink 10]. Currently, an estimated 700,000 people in the United States are living with a primary brain tumor, 30% of which are malignant (National Brain Tumor Society) [Weblink 10].

Introduction

9

Glioblastoma is the most commonly occurring primary malignant brain tumor (accounting for 14.6% of all tumors and 48.3% of all malignant tumors). Approximately 55.4% of malignant tumors occur in men and 44.6% in women (National Brain Tumor Society) [Weblink 10]. In 2020, it has been estimated that 87,240 new brain tumors will be diagnosed and that 25,800 of those will be found to be malignant. For the most common form of primary malignant brain tumors, glioblastoma, the five-year relative survival rate is only 6.8% (National Brain Tumor Society) [Weblink 10].

1.4.9 Head and neck cancers According to the American Society of Clinical Oncology (ASCO), head and neck cancer accounts for about 4% of all cancers in the United States. These cancers are known collectively as head and neck cancers and they usually begin in the squamous cells that line the moist, mucosal surfaces inside the head and neck (e.g., inside the mouth, the nose, and the throat). These squamous cell cancers are often referred to as squamous cell carcinomas of the head and neck. Head and neck cancers can also begin in the salivary glands, but salivary gland cancers are relatively uncommon, according to the NCI. An estimated 65,630 people (48,200 men and 17,430 women) will develop head and neck cancer this year, and it is estimated that 14,500 of them will die (10,760 men and 3740 women) (American Society of Clinical Oncology) [Weblink 11,12].

1.4.10 Multiple myeloma Overall, in 2016, the latest year for which incidence data are available, 25,286 new cases of multiple myeloma were reported, and 12,266 people died of this cancer in the United States. For every 100,000 people, 7 new myeloma cases were reported and 3 persons died [Weblink 6]. In 2020, there will be an estimated 32,270 (54% men) new cases and 12,830 (56% women) deaths [10]. The numbers are mind-boggling, and they do not hint at the suffering cancer causes much less the cost to our society in individual and community unproductiveness. But health authorities are telling us that treatments are constantly improving. In conclusion, given these data, there is little doubt that any new evidence-based medical treatment, even as an adjuvant treatment, that could lower these numbers would seem worth exploring.

10

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

1.5 References [1] Seyfried TN and Shelton LM. 2010. Cancer as a metabolic disease. Nutrition and Metabolism, Jan 27; 7: 7. DOI: https://doi.org/10.1186/1743-7075-7-7. [2] Vander Heiden MG, Cantley LC, and Thompson CB. 2010. Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science, May 22; 324(5930): 1029e1033. DOI: https://doi.org/10.1126/science.1160809. [3] Forbes NS, Coffin RS, Deng L, Evgin L, Fiering S, Giacalone M, Gravekamp C, Gulley JL, Gunn H, Hoffman RM, Kaur B, Liu K, Lyerly HK, Marciscano AE, Moradian E, Ruppel S, Saltzman DA, Tattersall PJ, Thorne S, Vile RG, Zhang HH, Zhou S, and McFadden G. 2018. White paper on microbial anti-cancer therapy and prevention. Journal for ImmunoTherapy of Cancer, Aug 6; 6(78). DOI: https://doi.org/10. 1186/s40425-018-0381-3. [4] Sarotra P and Medhi B. 2016. Use of Bacteria in Cancer Therapy. In: Walther W. (eds) Current Strategies in Cancer Gene Therapy. Recent Results in Cancer Research, vol 209. Springer, Cham. DOI: https://doi.org/10.1007/978-3-319-42934-2_8. [5] Atmaca A, Al-Batran S-E, Neumann A, Kolassa Y, Jäger D, Knuth A, and Jäger E. 2009. Whole-body hperthermia (WBH) in combination with carboplatin in patients with recurrent ovarian cancer - A Phase II Study. Gynecologic Oncology, Feb; 112(2): 384e388. DOI: https://doi.org/10.1016/j.ygyno.2008.11.001. [6] Dietzel F. 1983. Basic principles in hyperthermic tumor therapy. Recent Results in Cancer Research, 86: 177e190. DOI: https://doi.org/10.1007/978-3-642-82025-0_31. [7] Huerta S. 2015. Nitric oxide for cancer therapy. Future Science OA, Aug 15; 1(1): FSO44. DOI: https://doi.org/10.4155/fso.15.44. [8] Xu W, Liu LZ, Loizidu M, Ahmed M, and Charles IG. 2002. The role of nitric oxide in cancer. Cell Research, Dec; 12: 311e320. DOI: https://doi.org/10.1038/sj.cr. 7290133. [9] Loevenhart AS, Lorenz WF, Martin HG, and Malone JY. 1918. Stimulation of respiration by sodium cyanid [ibid,] in clinical application. Archives of Internal Medicine, 92: 109e129. [10] Siegel RL, Miller KD, and Jemal A. 2020. Cancer statistics, 2020. CA: A Cancer Journal for Clinicians, 70(1): 7e30. DOI: https://doi.org/10.3322/caac.21590.

1.5.1 Weblinks Weblink 1: https://www.nationalgeographic.com/news/2016/07/oldest-human-cancerdisease-origins-tumor-fossil-science/; accessed 5.9.20. Weblink 2: https://www.cancer.gov/about-cancer/treatment/types; accessed 5.10.20. Weblink 3: https://www.cancer.gov/news-events/cancer-currents-blog/2018/oncolytic-viruses-totreat-cancer; accessed 5.10.20. Weblink 4: https://www.cancer.gov/about-cancer/treatment/types/surgery/hyperthermia-factsheet; accessed 5.20.20. Weblink 5: https://www.webmd.com/cancer/amygdalin-cancer-treatment; accessed 5.10.20. Weblink 6: https://gis.cdc.gov/Cancer/USCS/DataViz.html; retrieved 5.10.20. Weblink 7: https://www.cdc.gov/cancer/breast/statistics/; retrieved 5.9.20. Weblink 8: https://www.cdc.gov/cancer/colorectal/statistics/; accessed 5.11.20. Weblink 9: https://www.cdc.gov/cancer/prostate/statistics/index.htm; accessed 5.11.20. Weblink 10: https://braintumor.org/brain-tumor-information/brain-tumor-facts/; retrieved 5.10.20. Weblink 11: https://www.cancer.gov/types/head-and-neck/head-neck-fact-sheet; accessed 5.10.20. Weblink 12: https://www.cancer.net/cancer-types/head-and-neck-cancer/statistics; retrieved 5.10.20.

Introduction

11

1.6 Suggested additional reading Dewhirst MW, Gibbs FA Jr, Roemer RB, and Samulski TV. 2000. Hyperthermia. In: Gunderson LL and Tepper JE, editors. Clinical Radiation Oncology. 1st ed. New York, NY: Churchill Livingstone. Falk MH and Issels RD. 2001. Hyperthermia in oncology. International Journal of Hyperthermia, Jan-Feb; 17(1): 1e18. DOI: https://doi.org/10.1080/02656730150201552. Hildebrandt B, Wust P, Ahlers O, Dieing A, Sreenivasa G, Kerner T, Felix R, and Riess H. 2002. The cellular and molecular basis of hyperthermia. Critical Reviews in Oncology/ Hematology, Jul; 43(1): 33e56. DOI: https://doi.org/10.1016/s1040-8428(01)00179-2. van der Zee J. 2002. Heating the patient: a promising approach? Annals of Oncology, Aug; 13(8): 1173e1184. DOI: https://doi.org/10.1093/annonc/mdf280. Wust P, Hildebrandt B, Sreenivasa G, Rau B, Gellermann J, Riess H, Felix R, and Schlag PM. 2002. Hyperthermia in combined treatment of cancer. The Lancet Oncology, Aug; 3(8): 487e497. DOI: https://doi.org/10.1016/s1470-2045(02)00818-5.

CHAPTER 2

The metabolic theory of cancer and its clinical implications

2.1 The treatment implications of the Warburg discovery Otto Warburg, a German scientist and Nobel laureate, made the astonishing discovery that normal cells become cancerous when they revert to a primitive form of extracting energy from fuel called anaerobic metabolismea form that does not use oxygenealso called “fermentation.” In contrast to cancerous cells, normal healthy cells from which those cancerous cells are derived do use oxygen to extract energy, a process called aerobic metabolism, which is also termed “respiration.” Normal, healthy cells utilize both of those metabolic pathways. The first step, the anaerobic one, takes place in the cytoplasm of the cell. The second step, the aerobic one, takes place in the mitochondria, which are cellular organelles that are very efficient in utilizing oxygen to metabolize foodstuffs. Warburg provided evidence that the reason cancerous cells revert to using only the primitive anaerobic (fermentation) process is that their mitochondria are functionally damaged, e.g., (a) in the way that they metabolize, (b) structurally/anatomically, and (c) by virtue of mutations in the sequence of the DNA of the mitochondria. This damage to the mitochondria has major consequences because most of the enzymes required to carry out aerobic metabolism (respiration) are located in the mitochondria. Warburg observed that the damage is extensive enough to prevent the mitochondria, and therefore the cell as a whole, from utilizing oxygen for metabolism. One of the few ways these damaged cells can manage to remain alive is to revert to the default metabolic pathway that began to be utilized by some of the earliest living organisms in the ancient past before oxygen became abundant in the atmosphere. That metabolic pathway is called glycolysis, which means the “lysing” or breakdown of glucose. That ancient pattern of metabolism suits these now-cancerous cells fairly well for their new “mission” in life which is to multiply out of control and, in Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression ISBN 978-0-12-824013-7 https://doi.org/10.1016/B978-0-12-824013-7.00004-3

© 2021 Elsevier Inc. All rights reserved.

13

14

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

the case of solid tumors, to assemble into large masses. However, those large masses tend to have very little oxygen in the core despite “angiogenesis,” their mechanism to induce growth of abundant blood vessels to serve their tissue mass. The process is, of course, quite complicated and this has led a number of investigators to conduct experiments that they claim refute Warburg’s hypothesis. They believe alternatively that mutations in the nuclear DNA can explain why the mitochondria have gone “rogue.” But to all extents and purposes, it does not really matter which school of thinking is correct because no one disputes that the mitochondria are, in fact, damaged and malfunctioning. The only dispute is about the reasons why mitochondria are damaged. What does matter is that this return to a glycolysis-only way of life fortuitously leaves quite a number of gaping holes in cancer cell defenses. What’s more, we can exploit these gaping holes, in a “translational medicine” approach to cancer treatment by using metabolic methods to kill the cancer cells. To give you a preliminary sense, here are two of the metabolic intervention processes to which Warburg’s observations lead us: • A ketogenic diet that deprives cancer cells of glucose but provides ample ketones to fuel the normal, healthy cells. • Dietary restriction of those nutrients that some cancer cells will metabolize by fermentation to stay alive when glucose is no longer available.

2.2 Cell metabolism 101. A brief review In order to fully appreciate the treatment options that Warburg’s discoveries open to possibility, it may help to first review some of the basic biochemical principles concerning how cells extract energydhow they metabolize their food. To that end, we provide here a brief review of some basics with two reasons in mind: First, when Warburg refers to “respiration” versus “fermentation,” he has a very specific meaning in mind, and understanding that meaning requires a fair degree of precision. Second, how many remember the exact point where Acetyl CoA enters the tricarboxylic acid cycle? Do you recall what Acetyl CoA is derived from? And is the entry of Acetyl CoA into the cells under the regulation of a transporter, analogous to the Glut3 transporter assembly for glucose entry?

The metabolic theory of cancer and its clinical implications

15

These questions are hardly “academic,” because reconsidering this little bit of biochemistry can open up a way to help your patients to starve their cancer cells to death and/or alternatively provoke those cancer cells to trigger apoptosis (programmed cell death). Cancer cells have a remarkable mechanism for evading “apoptosis” and that evasion enables them to live on and to propagate endlessly. Our cells undergo a two-step process in forming energy from foodstuffs, the first one being called fermentation or glycolysis (which can take place whether oxygen is present or not), the second one being called respiration (which can take place only when oxygen is present, and is thus termed aerobic). In fermentation, enzymes in the cytoplasm of the cells split glucose, which has a 6-carbon chain, into two 3-carbon molecules of lactic acid (lactate) or pyruvate, depending on whether oxygen was absent or present, respectively. This chemical process releases a small but reasonable amount of energy, and that energy is stored, battery-like, by coupling it through substrate phosphorylation to a series of high-energy phosphate-containing molecules, culminating in one called adenosine triphosphate (ATP). The second step in energy production is, as mentioned, aerobic respiration. This step, using oxygen to extract far more energy from glucose than can be achieved by fermentation alone, became possible with the incorporation into the cell of a primitive type of bacteria that we call mitochondria. This conferred a great advantage to those ancestral animal cells: the mitochondria contain a sequence of enzymes that can convert the 3carbon lactate (from glycolysis) step-wise, into single-carbon molecules (carbon dioxide). This process culminates in a bonanza of ATP molecules creating a vast amount of energy. Thus, the mitochondria and the animal cells enter into perfect symbiosis. The enzymes contained in the mitochondria function in a cyclical manner referred to as the tricarboxylic acid cycle (TCA) because the mitochondria start out with a three (“tri”) carbon molecule as described above. The TCA is also commonly referred to as the Krebs cycle, and those two terms are used interchangeably. This cycle promotes oxidation of carbohydrates, coupling the energy inherent in the carbon-carbon bonds into high-energy phosphates. The process is therefore referred to as oxidative phosphorylation (“OxPhos”). Cells that have this two-step metabolism gained an enormous advantage: an initial fermentation of fuel (which can be anaerobic), coupled to and directly feeding into aerobic metabolism (respiration, resulting in OxPhos). The advantage is that in periods of starvation, when glucose or other carbohydrate molecules are scarce, organisms with these “hybrid-engine” cells could obtain an alternate source of fuel by burning (metabolizing) their

16

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

own fat stores. The fat molecules are eventually broken down into ketones, such as AcAc and beta-hydroxybutyrate. Both of those ketones can diffuse freely into the cells as well as into the mitochondria, there being no transport mechanism required such as the one to transport glucose. Once inside the mitochondria, the ketones enter and drive the Krebs cycle (a bit downstream from where pyruvate would have kicked it off). Thus, ketones can power the Krebs cycle to keep on producing its large quantities of ATP even in the complete absence of glucose and glycolysis. In fact, each mole of ketone bodies produces more molecules of ATP than does each mole of glucose [1]. In fact, that confers a great survival advantage for any animal coping with starvation such as in winter months when food is scarce. They can “burn” their stored fat down to ketone bodies, and survive. That same phenomenon of being able to bypass glycolysis now also gives us a powerful weapon to starve cancer cells: in clinical practice with cancer sufferers, we could institute a diet rich in ketones and that is simultaneously extremely low in (a) carbohydrates (especially glucose) as well as in (b) anything else that the body can convert into glucose, such as glutamine. This is detailed in Sections V and VI that describe glucose deprivation and glutamine deprivation, respectively. Let’s take a look at Figure 2.1 to see exactly where it is that ketones enter the Krebs cycle and come to the rescue as a metabolic strategy.

Figure 2.1 Mitochondrial generation of ATP in a representative neuron. Acetyl-CoA undergoes oxidative degradation in the TCA (tricarboxylic acid) cycle with reduction of the electron carriers NAD + (nicotinamide adenine dinucleotide) and FAD (flavin adenine dinucleotide) to NADH and FADH2, respectively. NADH and FADH2 donate

The metabolic theory of cancer and its clinical implications

17

2.3 When mitochondrial genes and nuclear genes uncouple: trouble in paradise The mitochondria have their own DNA, RNA polymerase, transfer RNA, and ribosomes [2]. Thus, when a cell divides, its nuclear DNA and its mitochondrial DNA both get replicated, so that the daughter cells obtain their own full complement of mitochondria (genes and all). The mitochondrial genes encode for most of the enzymes of the Krebs cycle. However, some of those enzymes are encoded by genes in the nucleus. Most important, mitochondrial DNA can be damaged and mutated by the same general types of carcinogens that induce mutations in the nuclear DNA. As noted previously, Warburg had reported a number of defects in the mitochondria of cancerous cells. In a pivotal book on this general topic, Cancer as a Metabolic Disease, by Thomas N. Seyfried [3], Seyfried discusses the following issues, starting on page 77: • The morphological defects in tumor cell mitochondria • The proteomic abnormalities in tumor cell mitochondria • The lipidomic abnormalities in tumor cell mitochondria • Cardiolipin and abnormal energy metabolism in tumor cells • Mitochondrial uncoupling and cancer

=

electrons to the protein complexes I (NADH-ubiquinone oxidoreductase) and II (succinate-ubiquinone reductase) of the mitochondrial electron transport chain. Energy derived from the transfer of electrons down the electron transport chain to oxygen (O2) is used by complexes I, III (Q-cytochrome c oxidoreductase [cytochrome reductase]), and IV (cytochrome c oxidase) to pump protons (shown as Hþ) out of the matrix into the intermembrane space, thereby generating a proton motive force (pmf) between the intermembrane space and the matrix. The pmf (reflecting the electrochemical proton gradient across the mitochondrial inner membrane) drives protons back through F1 ATP synthase (sometimes called complex V), thereby providing the energy to produce ATP from ADP (adenosine diphosphate) and Pi (inorganic phosphate). Glucose transporter protein type 1 (GLUT1) enables glucose transfer across the blood-brain barrier (BBB), being expressed at high levels in endothelial cells of the BBB. Because GLUT1 is consistently expressed at high levels in brain microvessel endothelial cells, it could affect access of glucose to neurons nourished by such microvessels. Glucose transporter protein type 3 (GLUT3) is the principal glucose transporter isoform in adult brain, being preferentially situated in neurons. Transfer of the ketone bodies (BHB and AcAc) across cell membranes is enabled by monocarboxylate transporters (MCTs). Abbreviations: AcAc, acetoacetate; ACT, acetoacetyl-CoA thiolase; BHB, b-hydroxybutyrate; HBD, b-hydroxybutyrate dehydrogenase; OCT, 3-oxoacid-CoA transferase; PDH, pyruvate dehydrogenase multienzyme complex. (Diagram from VanItallie TB. 2008. Parkinson disease: primacy of age as a risk factor for mitochondrial dysfunction. Metabolism: Clinical and Experimental, 57(Suppl. 2): S50eS55. https://doi.org/10.1016/j.metabol.2008.07.015. Reprinted with permission from Metabolism: Clinical and Experimental.)

18

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

Figure 2.2 Role of the nucleus and mitochondria in the origin of tumors. This image summarizes the experimental evidence supporting a dominant role of the mitochondria in the origin of tumorigenesis as described previously [3,4]. Normal cells are depicted in green with mitochondrial and nuclear morphology indicative of normal respiration and nuclear gene expression, respectively. Tumor cells are depicted in red with abnormal mitochondrial and nuclear morphology indicative of abnormal respiration and genomic instability. 1. Normal cells beget normal cells. 2. Tumor cells beget tumor cells. 3. Delivery of a tumor cell nucleus into a normal cell cytoplasm begets normal cells despite the persistence of tumor-associated genomic abnormalities. 4. Delivery of a normal cell nucleus into a tumor cell cytoplasm begets tumor cells or dead cells but not normal cells. The results suggest that tumors do not arise from nuclear genomic defects alone and that normal mitochondria can suppress tumorigenesis. (From Seyfried, T.N. (2012). Mitochondria: The ultimate tumor suppressor. In Cancer as a Metabolic Disease: On the Origin, Management, and Prevention of Cancer (Hoboken, NJ: John Wiley & Sons), pp. 195e205.)

Seyfried also published a seminal journal article on mitochondrial defects with the intriguing title, “Cancer as a metabolic disease: implications for novel therapeutics” in the journal Carcinogenesis [4]. And, his findings are quite surprising. The source for Figure 2.2 is that publication. There is an important point made in Figure 2.2 that might be “understood” by a biochemist, but is not necessarily that obvious to the rest of us and therefore it warrants being made explicit: when the authors state that “cytoplasm” is being transferred, what they really mean is that mitochondria that are contained in the cytoplasm are being transferred. So, for clarity, here is the text in diagram number 4 rephrased as “Tumor Mitochondria þ Normal Nucleus.” Then the meaning becomes quite clear. One would imagine that transferring tumor nuclei with their oncogenes present into normal cells would transform them into cancerous cells. But at least in the conditions stated in the publication, that is not what happened. It is startling to learn that the unexpected is true: transferring tumor mitochondria into normal cells did in fact transform them into cancerous ones. And, if this were without clinical significance, it would be no more than that, a curiosity.

The metabolic theory of cancer and its clinical implications

19

Here is where the objective, open mind of the scientist is tested and needed: Warburg’s observations and hypotheses have generated much “heat” for almost a 100 years. Yet, it is not mandated that one choose a side in the debate over whether there is a nuclear origin of cancer (through oncogenes that we know in fact are there, and can be suppressed) or, instead, a mitochondrial origin (even though mitochondrial genes can be mutated by the same carcinogens that mutate nuclear genes, according to Seyfried). Rather, we think it more productive to acknowledge that there is abundant evidence that “something is wrong” with the mitochondria in cancerous cells. Whatever the origin of what it is that is wrong, the evidence is clear that cancerous cells cannot (with rare exceptions) engage efficiently (if at all) in respiration and therefore OxPhos. So, if mitochondria are confined by these limitations, why not exploit that? Likewise, a practitioner may be routinely prescribing chemotherapy agents and/or radiation therapy regimens. Suddenly, it may come to his/her attention that for at least a good many types of cancer, there is good evidence from peer-reviewed journals that their mitochondria have metabolic defects. Wouldn’t it make sense to exploit this information rather than to ignore it? If the practitioner can clear a patient of a highly malignant and highly drug-resistant tumor by starving the cancer cells, and if that intervention safely complements the conventional treatments, why not implement it? Among the things that the studies appearing later in this book emphasize is the many advantages of the metabolic treatment methods and in particular the ones that cause little or no toxicity to our healthy cells, in contrast to chemotherapeutic agents and radiation therapy treatments. Those two approaches are well known to be highly toxic and damaging to healthy cells, especially those that are rapidly dividing such as bone marrow, gut, etc. We wish also to emphasize a point that Dr. Seyfried made in the Conclusions section of his book (p. 405) about the negative effects of cellular fermentation alone (without OxPhos): I addressed in Chapter 10 the answer to the question of how abnormalities in chromosome number arise in tumor cells. Basically, the stability of chromosome number and the integrity of the genome are dependent on the integrity [of] OxPhos. Spindle assembly and the fidelity of chromosomal segregation during mitosis are dependent on the energy of OxPhos. Injury to cellular respiration with compensatory fermentation will cause genomic instability including aneuploidy and mutations. It is the efficiency of mitochondrial respiration that maintains cellular differentiation and prevents tumorigenesis and dedifferentiation. [We added Italics for emphasis.]

20

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

2.4 The principal current metabolic strategies The metabolic treatment methods that arose from Warburg’s observations include, but are by no means limited to, the following: 1. Diets intentionally deficient in select nutrients that would otherwise fuel the cancer. 1.1 Restriction of glucose and other carbohydrates is an important approach to starve cancer cells, and is achieved preferentially through a “restricted ketogenic diet” (RKD, also written KD-R) so that the brain and other organs can get adequate supplies of ketone bodies for function. Dr. Seyfried discusses the KD-R diet approach at length in his book, cited above, as a way to “metabolically stress” the cancer cells. This approach sharply restricts the amounts of glucose and other carbohydrates in the diet, while providing ample amounts of fat in general and ketone sources in particular. In addition to the valuable guidance to be found on this subject in Dr. Seyfried’s book, there are many other books available on ketogenic diets. There is also a KetoCalculator program that one can find online, at www. ketocalculator.com. There are a few things to emphasize: 1.1.1 The KD-R diet will not provoke the ketoacidosis that is seen when diabetes mellitus is out of control. Many physicians mistakenly think that the ketogenic diet will be likely to provoke ketoacidosis. It does not. 1.1.2 While it is relatively easy to get the ketone levels up into the desired range, it is more difficult for many patients to get the blood glucose levels down to the desired range. It can take a lot of work, but it can be very worthwhile. 1.1.3 Certain medications will make it extremely difficult to keep glucose levels down. Principal among them are the steroids often used in oncology, and other specialties, such as Decadron and prednisone. For patients on such medications, a ketogenic diet may not be practical. 1.1.4 Certain other prescription medications might make it somewhat easier to get blood sugar levels down toward the desired levels, in conjunction with a ketogenic diet. Seyfried’s book lists some of the medications discussed in this regard, such as metformin, 2-deoxyglucose (2-DG), and phenylbutyrate (PB). 1.1.5 Vigorous exercise will increase lactic acid production in the muscles and, through the Cori cycle, that 3-carbon lactate

The metabolic theory of cancer and its clinical implications

21

can be converted by gluconeogenesis back into glucose, blunting if not negating the efficacy of the KG diet. It may therefore be desirable to persuade athletically-inclined patients to not exercise too vigorously, and, most important, to not engage in anaerobic exercise (e.g., sprints), as that will provoke a prolonged increase in lactic acid levels (because it takes time to metabolically process the excess lactate through the Krebs cycle). 1.1.6 See Chapter V of this book for a review of the evidence of clinical efficacy for this particular metabolic-restriction approach. 1.2 Glutamine restriction is another example of a metabolic approach to starve cancer cells. This essential amino acid can be fermented, i.e., metabolized nonoxidatively, by cancer cells to derive energy as ATP. Therefore, when cancer cells are intentionally being stressed by deprivation of glucose and other carbohydrates, the cells can “escape” by switching over to the fermentation of glutamine. To block that escape hatch, the practitioner could restrict glutamine as well as the carbohydrates. That being said, s/he should be aware that (a) some tumors will take up large amounts of glutamine, and (b) glutamine is critically needed for maintenance and repair of tissues, particularly the gut. Therefore, one would have to carefully balance a patient’s dietary requirements on the one hand as opposed to starving the cancer on the other. See Chapter VI for further information. 2. Hyperbaric oxygen (HBO). We quote from Seyfried’s book, cited above: 2.1 “Many types of tumor cells are susceptible to the hyperoxia produced by HBO.. This would kill any tumor cell with marginal respiratory activity.. 2.2 “The influence of HBO on the tumor growth and vascularity is remarkably similar to the influence of dietary energy restriction (DER) on tumor growth. Like DER, hyperoxia targets tumor angiogenesis, while increasing tumor cell apoptosis. 2.3 “The restriction of glucose availability will downregulate the pentose phosphate pathway, which is glucose dependent.. This reduces NADPH production.. NADPH is essential for maintaining catalase activity. Catalase is needed to metabolize H2O2 to water and O2. Tumor cells generate excessive Reactive Oxygen Species (ROS) due to respiratory dysfunction. NADPH depletion

22

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

would therefore increase the vulnerability of tumor cells to ROS through linkage to catalase reduction. 2.4 “If HBO were combined with KD-R, HBO would elevate ROS, thus increasing ROS-induced death. Ketones protect against ROS damage in cells with normal respiration because ketone metabolism in mitochondria oxidizes the coenzyme Q couple, thus decreasing the Q semiquinone, a major source of radical production in cells. 2.5 “I predict that treatment of cancer patients with a combination of the KD-R and HBO could be a new and effective therapeutic strategy for destroying tumor cells without harming normal cells.. [and would give a] potentially powerful therapeutic action against all glycolytic tumors.” The following metabolic strategy echoes features of the Seyfried approach described above for HBO. 3. Generating free radicals that cancer cells can’t quench, but normal cells can. The inability of cancer cells to quench free radicals has long been exploited by oncologists, who utilize chemotherapy drugs such as doxorubicin that generate long-acting oxygen free radicals [5]. Cancer cells, being anaerobic, synthesize little or none of the oxygen-quenching enzymes such as catalase and superoxide dismutase (SOD), and are therefore damaged and jeopardized by these types of medications that generate free radicals. Normal healthy cells do, of course, synthesize those free radicalescavenging enzymes in abundance, and thereby are not destroyed by the free radicals. However, these drugs are highly toxic to those normal, healthy cells and therefore they induce many side effects. A new approach has emerged in the last dozen years or so, consisting of administering ascorbic acid (vitamin C) intravenously in ultrahigh doses. Here are key points to consider in connection with this approach: 3.1 Low doses of ascorbate will quench free radicals, which could actually interfere with a number of cancer treatments that generate free radicals, such as radiation therapy. 3.2 In contrast, ultrahigh doses of ascorbate will generate a large number of free radicals, destroying the cancerous cells that lack the enzymes to quench those radicals, while sparing the normal cells that do have them. This, therefore, synergizes with chemotherapy and radiation therapy [6]. Many practitioners are reluctant to administer intravenous ascorbic acid (IVAA), because they only think of ascorbate as an antioxidant. But at large-enough dosages, it becomes a strong pro-oxidant, but one that is not cytotoxic to normal, healthy cells [7e10].

The metabolic theory of cancer and its clinical implications

23

3.3 The ascorbate has to be administered intravenously in an hour-long slow drip in order to reach the high in vivo concentrations that are required to generate free radicals. It has to be administered through a PICC line, because ascorbate at this high a dose can sclerose smaller veins. 3.4 The doses generally used are in the vicinity of 60 g of vitamin C per day. It would not be feasible to administer 60 g per day orally. With the oral route, such a high a dose would provoke profound diarrhea. 3.5 This strategy is so effective at killing cancerous cells that it is advised to start with much lower doses for the first few days, to avoid tumor lysis syndrome (TLS). 3.6 This IVAA approach is in daily use at a number of major medical centers in the United States, including University of Kansas Medical Center and the Henry Ford Hospital. One particular clinical trial of note documented the efficacy of IVAA in treating pancreatic cancer [11]. 3.7 Yet another metabolic strategy recommended by clinical oncologists is to put cancer patients on a diet that mimics fasting in combination with administration of vitamin C. Studies have found that fastingmimicking diets delay tumor progression and sensitize a wide range of tumors to chemotherapy, and that these diets increase the anticancer activity of vitamin C [12].

2.5 References [1] Masino SA, Kawamura M Jr, Wasser CD, Pomeroy LT, and Ruskin DN. 2009. Adenosine, ketogenic diet and epilepsy: The emerging therapeutic relationship between metabolism and brain activity. Current Neuropharmacology, Sep; 7(3): 257e268. DOI: https://doi.org/10.2174/157015909789152164. [2] Carew JS and Huang P. 2002. Mitochondrial defects in cancer. Molecular Cancer, Dec 9; 1: 9. DOI: https://doi.org/10.1186/1476-4598-1-9. [3] Seyfried TN. 2012. Cancer as a Metabolic Disease: On the Origin, Management, and Prevention of Cancer. 1st Edition. New York: John Wiley. [4] Seyfried TN, Flores RE, Poff AM, and D’Agostino DP. 2014. Cancer as a metabolic disease: implications for novel therapeutics. Carcinogenesis, Mar; 35(3): 515e527. DOI: https://doi.org/10.1093/carcin/bgt480. [5] Benchekroun MN, Sinha BK, and Robert J. 1993. Doxorubicin-induced oxygen free radical formation in sensitive and doxorubicin-resistant variants of rat glioblastoma cell lines. FEBS Letters, May; 322(3): 295e298. DOI: https://doi.org/10.1016/00145793(93)81589-R. [6] Koch CJ and Biaglow JE. 1978. Toxicity, radiation sensitivity modification, and metabolic effects of dehydroascorbate and ascorbate in mammalian cells. Journal of Cell Physiology, Mar; 94(3): 299e306. DOI: https://doi.org/10.1002/jcp.1040940307.

24

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

[7] Chen Q, Espey MG, Krishna MC, and Levine M. 2005. Pharmacologic ascorbic acid concentrations selectively kill cancer cells: Action as a pro-drug to deliver hydrogen peroxide to tissues. Proceedings of the National Academy of Sciences USA, Sep 20; 102(38): 13604e13609. DOI: https://doi.org/10.1073/pnas.0506390102. [8] Chen Q, Espey MG, Sun AY, and Levine M. 2007. Ascorbate in pharmacologic concentrations selectively generates ascorbate radical and hydrogen peroxide in extracellular fluid in vivo. Proceedings of the National Academy of Sciences USA, May 22; 104(21): 8749e8754. DOI: https://doi.org/10.1073/pnas.0702854104. [9] Chen Q, Espey MG, Sun AY, and Levine M. 2008. Pharmacologic doses of ascorbate act as a prooxidant and decrease growth of aggressive tumor xenografts in mice. Proceedings of the National Academy of Sciences USA, Aug 12; 105(32): 11105e11109. DOI: https://doi.org/10.1073/pnas.0804226105. [10] Riordan NH, Riordan HD, Meng YL, and Jackson JA. 1995. Intravenous ascorbate as a tumor cytotoxic chemotherapeutic agent. Medical Hypotheses, Mar; 44(3): 207e213. DOI: https://doi.org/10.1016/0306-9877(95)90137-x. [11] Drisko JA, Serrano OK, Spruce LR, Chen Q, and Levine M. 2018. Treatment of pancreatic cancer with intravenous vitamin C: a case report. Anticancer Drugs, Apr; 29(4): 373e379. DOI: https://doi.org/10.1097/CAD.0000000000000603. [12] Di Tano M, Raucci F, Vernieri C, and Longo VD. 2020. Synergistic effect of fastingmimicking diet and Vitamin C against KRAS mutated cancers. Nature Communications, May 11; 11(1): 2332. DOI: https://doi.org/10.1038/s41467-020-16243-3.

2.6 Suggested further reading Liberti MV and Locasale JW. 2016. The Warburg Effect: How does it benefit cancer cells? Trends in Biochemical Sciences, Mar; 41(3): 211e218. DOI: https://doi.org/10.1016/ j.tibs.2015.12.001. Martinez-Outschoorn UE, Pestell RG, Howell A, Tykocinski ML, Nagajyothi F, Machado FS, Tanowitz HB, Sotgia F, and Lisanti MP. 2011. Energy transfer in "parasitic" cancer metabolism: Mitochondria are the powerhouse and Achilles’ heel of tumor cells. Cell Cycle, Dec 15; 10(24): 4208e4216. DOI: https://doi.org/10.4161/cc.10.24.18487. Vander Heiden MG, Cantley LC, and Thompson CB. 2009. Understanding the Warburg Effect: The metabolic requirements of cell proliferation. Science, May 22; 324(5930): 1029e1033. DOI: https://doi.org/10.1126/science.1160809. Vyas S, Zaganjor E, and Haigis MC. 2016. Mitochondria and cancer. Cell, Jul; 166(3): 555e566. DOI: https://doi.org/10.1016/j.cell.2016.07.002.

CHAPTER 3

Tumor starvation by L-arginine deprivation

3.1 Introduction It was shown in the early 1930s that mice receiving an L- arginine-enriched diet developed larger tumors, and they did so faster than did mice on a standard diet. In contrast, an L-arginine-deficient diet reduced tumor incidence and growth in another group of mice [1]. Subsequently, the journal Archives of Surgery reported the effects of dietary arginine on the growth of a murine colon tumor metastatic to the liver in a model of advanced neoplastic disease. Tumor growth was influenced by arginine both in vivo and in vitro: An arginine-supplemented diet stimulated tumor growth by 55% compared to controls. Conversely, an arginine-depleted diet inhibited tumor growth by 78% compared to controls. In vitro culture of both murine and human colon tumor cells confirmed that arginine was necessary for cell growth. In that study, flow-cytometric analysis using propidium iodide and bromodeoxyuridine suggested that colon tumor cells cultured without arginine enter a quiescent S-phase and depend on arginine for further growth and cell cycle progression. The investigators concluded that these findings support a potential role for selective dietary arginine modulation in cancer patients with advanced disease [2]. Arginine deprivation is emerging as a novel, effective strategy for the treatment of arginine-dependent cancers (auxotrophs) that exploit differential expression and regulation of key enzymes. In this context, “deprivation” does not have its common meaning of simply “withholding,” but it is implemented by the action of enzymes that effectively eliminate L-arginine. Arginine dependence of certain tumor cells, auxotrophy, has been considered to be the “Achilles heel” of tumor cells. The inability of certain Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression ISBN 978-0-12-824013-7 https://doi.org/10.1016/B978-0-12-824013-7.00005-5

© 2021 Elsevier Inc. All rights reserved.

25

26

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

tumor cells to proliferate in the absence of arginine can be targeted for their selective destruction by arginine-depriving enzymes as illustrated in clinical studies citied below. A number of enzyme-based anticancer therapies are currently undergoing clinical evaluation. For instance, it is encouraging that deprivation by PEGylated (PEG ¼ polyethylene glycol) arginine deiminase (ADI-PEG20), a novel anticancer enzyme that produces depletion of L-arginine, already has achieved considerable success in patients without causing detrimental side effects and with high tolerability. L-arginine auxotrophy occurs in certain tumor types and is usually caused by the silencing of the enzyme argininosuccinate synthetase 1 (ASS 1), a hallmark of that cancer. ASS 1 is one of the enzymes of the urea cycle, the metabolic pathway transforming neurotoxic amonia produced by protein catabolism into inocuous urea in the liver. It catalyzes the formation of arginosuccinate from aspartate, citrulline and ATP and together with ASL it is responsible for the biosynthesis of arginine in most body tissues. In the studies that follow, these cancers are cited as targets for L-arginine deprivation in clinical studies of tumor “starvation” strategies: • acute lymphoblastic lymphoma • acute myeloid leukemia • breast cancer • colon carcinoma • hepatocellular carcinomas • head and neck cancer • human glioblastoma • human HCC • malignant melanomas • malignant pleural mesothelioma • multiple myeloma • nasopharyngeal carcinomas • osteosarcoma • prostate cancer • renal cell carcinoma • small cell lung cancer • in addition, L-arginine deprivation protocols can “stage” tumor cells for combination therapy where cells have not been killed outright by deprivation. The following studies were chosen on the basis that they are reasonably representative of current publications and are comprehensive, and that their references may contribute significantly to bibliography for researchers and clinicians alike.

Tumor starvation by L-arginine deprivation

27

3.1.1 References [1] Gilroy E. 1930. The influence of arginine upon the growth rate of a transplantable tumour in the mouse. Biochemical Journal, 24(3): 589e595. DOI: https://doi.org/10. 1042/bj0240589. [2] Yeatman TJ, Risley GL, and Brunson ME. 1991. Depletion of dietary arginine inhibits growth of metastatic tumor. Archives of Surgery, Nov; 126(11): 137613e137681. DOI: https://doi.org/10.1001/archsurg.1991.01410350066010.

3.2 Kremer JC and Van Tine BA. 2017. Therapeutic arginine starvation in ASS1-deficient cancers inhibits the Warburg effect. Molecular Cellular Oncology, 4(3): e1295131. DOI: https://doi.org/10.1080/23723556. 2017.1295131. 3.2.1 Abstract Argininosuccinate Synthetase 1 deficiency induces dependence on extracellular arginine for continued cellular growth and survival. Arginine starvation inhibits the Warburg effect and diverts glucose into serine biosynthesis, while simultaneously increasing glutamine metabolism via the tricarboxylic acid cycle. Simultaneous arginine deprivation and inhibition of the subsequent metabolic adaptations induce synthetic lethality. Cancer metabolism represents the next wave of cancer therapeutics after immunotherapy. The alterations in cancer cell metabolism that are currently being investigated result from the dramatic metabolic reprogramming that occurs within cancer for the production of biomass. One of the original metabolic hallmarks was the “Warburg Effect,” whereby cancer cells preferentially generate energy through glucose fermentation into lactic acid rather than oxidation through the tricarboxylic acid (TCA) cycle and subsequent oxidative phosphorylation. Other common metabolic alterations in cancer include mutations in isocitrate dehydrogenase, overexpression of pyruvate kinase M2 (PKM2), deficiencies in succinate dehydrogenase and fumarate hydratase, and the loss of argininosuccinate synthetase 1 (ASS1) expression [1]. ASS1 is an enzyme involved in clearance of nitrogenous waste via the urea cycle and de novo arginine biosynthesis. Loss of ASS1 expression forces cells to rely on extracellular arginine for continued growth and survival; however, it appears to offer a tumorigenic advantage as loss of ASS1 expression has been shown to be a prognostic biomarker of reduced

28

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

metastasis-free and overall survival. ASS1-deficient cells undergo autophagy when exposed to the arginine-depleting agent PEGylated arginine deiminase (ADI-PEG20) and undergo cell death upon simultaneous pharmacological or genetic inhibition of autophagy [2,3]. Acquired resistance to ADI-PEG20 occurs via a c-Myc-dependent reexpression of ASS1 and renewed cellular arginine biosynthesis capabilities [4,5]. Characterization of the cellular consequences of arginine starvation is essential in order to identify and target the metabolic reprogramming that occurs to inhibit the ability of cancer cells to acquire resistance to ADI-PEG20. These findings are being used to develop a biomarker-driven multiagent metabolic therapy for ASS1-deficient cancers. Metabolic characterization of both short- and long-term ADI-PEG20 treatment has shown significant alterations in global metabolism [5,6]. One of the many alterations of metabolic pathways that occur upon short-term ADI-PEG20 treatment was the significant redirection of cellular glucose. While cancer cells typically ferment the majority of cellular glucose into lactic acid and oxidize lesser amounts via the TCA cycle, ADI-PEG20 caused a decrease in the flux of glucose-derived carbons into both lactate and TCA cycle intermediates. The decrease in glucose flux to lactate and citrate caused a decrease in the levels of lactate dehydrogenase A (LDHA) and its accompanying activating phosphorylation of Tyr10, and the decrease of pyruvate dehydrogenase subunit E1a (PDH1) and a simultaneous induction of Ser300 phosphorylation shown to inhibit catalytic activity [7,8]. Additionally, the terminal enzyme in glycolysis, pyruvate kinase M2 (PKM2), decreased upon ADI-PEG20 treatment, as did the inhibitory phosphorylation of Tyr105. PKM2 is an important enzyme in cancer metabolism and has been shown to be implicated in the regulation of the Warburg effect due to its low enzymatic activity and ability to increase the concentrations of glycolytic metabolites to be used for numerous biosynthetic reactions [9]. The changes in PKM2, LDHA, and PDH1 coincided with a redirection of glucose carbons into serine biosynthesis via upregulation of the ratelimiting enzyme of the pathway, phosphoglycerate dehydrogenase (PHGDH). The increase in serine biosynthesis and subsequent metabolism to glycine lead to increased sensitivity to the inhibition of these pathways with CBR-5884, a small molecule inhibitor of PHGDH, and methotrexate, an antifolate analog that inhibits single carbon metabolism, including but not limited to the conversion of serine to glycine [10].

Tumor starvation by L-arginine deprivation

29

Indeed, co-treatment with ADI-PEG20 and CBR-5884 induced a synthetic lethal response that lead to significantly higher levels of cell death than treatment with either drug individually. Long-term ADI-PEG20 treatment was shown to induce sensitivity to both CBR-5884 and methotrexate. The ADI-PEG20-induced decrease in glucose flux into the TCA cycle would require cells to find another carbon source for the anaplerotic reactions needed to replenish TCA cycle intermediates removed for biosynthetic reactions. Stable isotope tracings of U-13C glutamine identified an upregulation of glutamine uptake and anaplerosis as a result of ADI-PEG20 treatment, with much of the glutamine-derived carbon ending up in aspartate and asparagine. The increase in glutamine metabolism and the biosynthesis of aspartate and asparagine were associated with an ADI-PEG20-induced upregulation of glutaminase (GLS), glutamate dehydrogenase (GDH), and asparagine synthase (ASNS). The dependence on glutamine metabolism upon arginine starvation was supported with in vitro and in vivo studies identifying pharmacological or genetic inhibition of GLS as another synthetic lethal interaction with ADI-PEG20. Changes in lactic acid fermentation and TCA cycle metabolism upon ADI-PEG20 treatment resulted in an overall increase of mitochondrial respiration. Extracellular flux analyses measuring extracellular acidification and oxygen consumption after ADI-PEG20 treatment provided additional data illustrating decreased fermentation and increased mitochondrial activity after arginine deprivation.Additionally, long-term ADI-PEG20 treatment was shown to significantly decrease the amount of adenosine triphosphate (ATP) generated via aerobic glycolysis and significantly increase the sensitivity to the oxidative phosphorylation inhibitor oligomycin. Together, these results show that arginine deprivation results in a significant shift in metabolic phenotype and an inhibition of the Warburg effect. Characterization of the metabolic changes that occur upon ADI-PEG20mediated arginine deprivation, summarized in Figure 3.2.1, allowed for the identification of compensatory pathways necessary for cellular adaptation to short- and long-term ADI-PEG20 treatment. The increase in glutamine metabolism and glucose-dependent serine biosynthesis and subsequent metabolism upon arginine starvation resulted in the emergence of multiple potential therapeutic targets, inhibition of which is capable of inducing a synthetic lethal response when paired with ADI-PEG20 treatment. Importantly, these results show the potential for synthetic lethal cancer therapies that are based on targeting the metabolism of cancer cells that do not express ASS1.

30

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

ADI-PEG20

Glucose

Arginine

Serine

Glutamine

PHGDH Inhibitor PHGDH

ATP

Glucose

Lactate

Glutamine

THF Folate Cycle

Antifolate GLS Inhibitor

meTHF

GLS

Glycine OxPhos ATP

TCA

Glutamate

O2

Figure 3.2.1 Metabolic effects of ADI-PEG20 induced arginine starvation on ASS1deficient cells. The metabolic effects of PEGylated arginine deiminase (ADI-PEG20) treatment of argininosuccinate synthetase 1 (ASS1)-deficient cells include the upregulation of serine biosynthesis, glutamine utilization, and oxidative phosphorylation upon arginine deprivation. Arrows and enzymes colored in green denote upregulation upon treatment with ADI-PEG20. The text in red denotes therapeutics that can be used in combination in order to induce a synthetic lethal cellular response. Abbreviations: THF - tetrahydrofolate. meTHF- 5, 10-methylene tetrahydrofolate. GLS - glutaminase. TCA - tricarboxylic acid cycle. OxPhos - oxidative phosphorylation. PHGDH - phosphoglycerate dehydrogenase.

3.2.2 References [1] DeBerardinis RJ, Lum JJ, Hatzivassiliou G, and Thompson CB. 2008. The biology of cancer: Metabolic reprogramming fuels cell growth and proliferation. Cell Metabolism, Jan; 7(1): 11e20. DOI: https://doi.org/10.1016/j.cmet.2007.10.002. [2] Bean GR, Kremer JC, Prudner BC, Schenone AD, Yao JC, Schultze MB, Chen DY, Tanas MR, Adkins DR, Bomalaski J, Rubin BP, Michel LS, and Van BA. 2016. A metabolic synthetic lethal strategy with arginine deprivation and chloroquine leads to cell death in ass1-deficient sarcomas. Cell Death and Disease, Oct 13; 7(10): e2406. DOI: https://doi.org/10.1038/cddis.2016.232. [3] Szlosarek PW, Klabatsa A, Pallaska A, Sheaff M, Smith P, Crook T, Grimshaw MJ, Steele JP, Rudd RM, Balkwill FR, and Fennell DA. 2006. In vivo loss of expression of argininosuccinate synthetase in malignant pleural mesothelioma is a biomarker for susceptibility to arginine depletion. Clinical Cancer Research, Dec 1; 12(23): 7126e7131. DOI: https://doi.org/10.1158/1078-0432.CCR-06-1101. [4] Tsai WB, Aiba I, Long Y, Lin HK, Feun L, Savaraj N, and Kuo MT. 2012. Activation of ras/pi3k/erk pathway induces c-myc stabilization to upregulate argininosuccinate synthetase, leading to arginine deiminase resistance in melanoma cells. Cancer Research, Mar 15; 72(10): 2622e2633. DOI: https://doi.org/10.1158/0008-5472.CAN-11-3605.

Tumor starvation by L-arginine deprivation

31

[5] Long Y, Tsai WB, Wangpaichitr M, Tsukamoto T, Savaraj N, Feun LG, and Kuo MT. 2013. Arginine deiminase resistance in melanoma cells is associated with metabolic reprogramming, glucose dependence, and glutamine addiction. Molecular Cancer Therapeutics, Nov 13; 12(11): 2581e2590. DOI: https://doi.org/10.1158/1535-7163. MCT-13-0302. [6] Kremer JC, Prudner BC, Lange SE, Bean GR, Schultze MB, Brashears CB, Radyk MD, Redlich N, Tzeng SC, Kami K, Shelton L, Li A, Morgan Z, Bomalaski JS, Tsukamoto T, McConathy J, Michel LS, Held JM, and Van Tine BA. 2017. Arginine deprivation inhibits the warburg effect and upregulates glutamine anaplerosis and serine biosynthesis in ass1-deficient cancers. Cell Reports, Jan 24; 18(4): 991e1004. DOI: https://doi.org/10.1016/j.celrep.2016.12.077. [7] Fan J, Hitosugi T, Chung TW, Xie J, Ge Q, Gu TL, Polakiewicz RD, Chen GZ, Boggon TJ, Lonial S, Khuri FR, Kang S, and Chen J. 2011. Tyrosine phosphorylation of lactate dehydrogenase a is important for nadh/nad(þ) redox homeostasis in cancer cells. Molecular and Cell Biology, Dec; 31(24): 4938e4950. DOI: https://doi.org/10. 1128/MCB.06120-11. [8] Rardin MJ, Wiley SE, Naviaux RK, Murphy AN, and Dixon JE. 2009. Monitoring phosphorylation of the pyruvate dehydrogenase complex. Analytical Biochemistry, Jun 15; 389(2): 157e164. DOI: https://doi.org/10.1016/j.ab.2009.03.040. Epub 2009 Mar 31. [9] Chaneton B and Gottlieb E. 2012. Rocking cell metabolism: Revised functions of the key glycolytic regulator pkm2 in cancer. Trends in Biochemical Sciences, Aug; 37(8); 309e316. DOI: https://doi.org/10.1016/j.tibs.2012.04.003. [10] Mullarky E, Lucki NC, Beheshti Zavareh R, Anglin JL, Gomes AP, Nicolay BN, Wong JC, Christen S, Takahashi H, Singh PK, Blenis J, Warren JD, Fendt SM, Asara JM, DeNicola GM, Lyssiotis CA, Lairson LL, and Cantley LC. 2016. Identification of a small molecule inhibitor of 3-phosphoglycerate dehydrogenase to target serine biosynthesis in cancers. Proceedings of the National Academy of Sciences USA, Feb 16; 113(7): 1778e1783. DOI: https://doi.org/10.1073/pnas.1521548113.

Reprinted full text and figure with permission from Molecular Cellular Oncology.

3.3 Patil MD, Bhaumik J, Babykutty S, Banerjee UC, and Fukumura D. 2016. Arginine dependence of tumor cells: targeting a chink in cancer’s armor. Oncogene, Sep 22; 35(38): 4957e4972. Published online 2016 Apr 25. DOI: https://doi.org/10.1038/onc.2016.37. (NIHMSID: NIHMS857065). 3.3.1 Abstract Arginine, one among the 20 most common natural amino acids, has a pivotal role in cellular physiology as it is being involved in numerous cellular metabolic and signaling pathways. Dependence on arginine is diverse for both tumor and normal cells. Because of decreased expression of argininosuccinate synthetase and/or ornithine transcarbamoylase, several types of tumor are auxotrophic for arginine. Deprivation of arginine exploits a significant vulnerability of these tumor cells and leads to their rapid demise. Hence, enzyme-mediated arginine depletion is a potential strategy for the selective destruction of tumor cells. Arginase, arginine deiminase and arginine decarboxylase are potential enzymes that may be used for arginine deprivation therapy. These arginine

32

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

catabolizing enzymes not only reduce tumor growth but also make them susceptible to concomitantly administered anti-cancer therapeutics. Most of these enzymes are currently under clinical investigations and if successful will potentially be advanced as anti-cancer modalities.

3.3.2 Introduction Amino acids play a major role in regulating important cellular events in both normal and malignant cells. Besides their role in the synthesis of hormones and peptides, amino acids also function as cell signaling molecules, playing a modulatory role in gene expression [1]. Amino acids regulate RNA synthesis by diverse mechanisms ranging from regulating transcription factors assembly [2], to total mRNA turnover [3,4]. Amino acids are major determinants of a normal cellular physiology, therefore potential signaling pathways such as amino acid response (AAR) pathway sense their altered metabolism (Figure 3.3.1). Hence, amino acid levels in the body are critical for important cellular functions [5e9]. Phosphorylated eIF2a binds more tightly to eIF2b, inhibiting the exchange of GDP for GTP. Inhibition of GDP exchange for GTP further inhibits the binding of eIF2 complex to methionine aminoacyl tRNA, leading to inhibition of translational initiation [197]. Recently, SLC38A9 Arginine

Arginine induced interaction of SLC38A9 with v-ATPase

Activated Rag GTPases

Recruitment of mTOR to the Lysosomal surface

Rhcb localization to the lysosomal surface

Low intracellular arginine

Increase in uncharged t-RNA pool

Activation of GCN2 Kinase

Inhibition of translational initiation Activation of mTOR

Figure 3.3.1 AAR pathway. Restriction of essential amino acids activates the general control nondepressible protein 2 (GCN2) kinase by increasing uncharged tRNA pool [196]. Activated GCN2 kinase phosphorylates the translation initiation factor eIF2a.

Tumor starvation by L-arginine deprivation

33

has been identified as an upstream positive regulator of the mTOR pathway. Amino acids activate the RAG GTPases, which then recruit mTOR to the lysosomal surface. Rheb also localizes to lysosomal membrane. mTOR activation occurs only when both RAG GTPases and Rheb are active. Upon amino acid deprivation, tuberous sclerosis complex translocates to lysosomal surface and promotes GTP hydrolysis by Rheb and thereby inhibiting mTOR complex [164]. There is a significant difference between the metabolism of normal and malignant cells [10]. For instance, bio-energetic requirements for homeostasis in normal cells are fulfilled by catabolic metabolism. On the other hand, the majority of the tumor cells alter their metabolic program (‘metabolic remodeling’) and consume additional nutrients in order to maintain a balance between elevated macromolecular biosynthesis [11] and adequate levels of ATP for survival [12,13]. However, the endogenous supply of nutrients becomes inadequate during intense growth. Thus tumor cells depend on exogenous nutrients in their microenvironment to fulfill the elevated energy requirements, that is, they become auxotrophic for nutrient and energy sources [14e16]. Deprivation of amino acids results in growth inhibition or death of tumor cells by the modulation of various signaling cascades [6e9,17,18]. Exogenously incorporated enzymes that deprive amino acids could be a novel strategy for the treatment of auxotrophic tumors. The first Food and Drug Administration approved heterologous enzyme for the treatment of cancer was Escherichia coli L-asparaginase [19]. L-asparaginase exploits the differences on their dependence of normal and leukemic cells toward L-asparagine [20]. L-asparaginase has been proven to be a promising agent for the treatment of L-asparagine auxotrophic T-cell acute lymphoblastic lymphoma (T-ALL). Use of L-asparaginase in T-ALL opened up new windows of “amino acid-depriving therapy.” Currently, there is a resurgence of interest in enzyme-mediated amino acid deprivation as a new therapeutic approach for cancer treatment [6,7,21,22]. For example, arginine depletion can inhibit tumor cell proliferation and induce cell death pathways. Here we endeavor to provide a basic understanding of the role of arginine in normal and tumor cell with emphasis on current knowledge and developments in the application of enzyme-mediated arginine-depriving therapy as a potential anti-cancer approach.

3.3.3 Enzyme-mediated arginine deprivation: a potential anti-cancer approach Arginine is involved in the regulation of various molecular pathways and thus the availability of arginine can modulate key metabolic,

34

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

immunological, neurological and signaling pathways of the cells (Figures 3.3.2 and 3.3.3) [23,24]. Auxotrophy toward arginine by certain tumor cells (particularly that of hepatocellular carcinoma and melanoma) has been well characterized [25,26]. Normal cells, when deprived of arginine, undergo cell cycle arrest at Go/G1 phase and become quiescent. If reinstated with arginine, the majority of the normal cells recover to their normal proliferation status. However, arginine deprivation in tumor cells does not arrest cell cycle at G1 phase and continue to be in a cell cycle, leading tumor cells to undergo unbalanced growth and eventually lead to the activation of apoptotic pathways [27,28]. Owing to the involvement of arginine in a plethora of cellular pathways, arginine dependence of tumor cells has rapidly emerged as a potential target for cancer [29]. However, dietary restriction results in the reduction of only 30% of plasma arginine [30]. Thus, arginine degrading enzyme-mediated arginine deprivation has been proposed as a potential anti-cancer therapy by various research groups [27e35]. Enzymes that can be used for arginine deprivation therapy (ADT) include arginine deiminase (ADI), arginase and arginine decarboxylase (ADC) as discussed below (Figure 3.3.3). Cell signalling

Cell proliferation

Nucleotides Agmatine Proline & Glutamate synthesis

Amino acids

Polyamines

Arginine

Hormones

Creatine

Muscle component

Immunity Nitric oxide

Insulin, Growth hormone, Prolactin synthesis

Neurotransmission, Vasodilation

T-cell functioning

Figure 3.3.2 Involvement of arginine in human physiology. Arginine is a dibasic, cationic amino acid and is considered as ‘conditionally essential’ amino acid. Arginine plays a crucial role in innate and adaptive immunity. For example, increased role of arginine in myeloid-derived suppressor cells results in the impairment of T-cell proliferation and function [190]. Arginine has been identified as the sole physiological precursor for NO, a key performer in many cellular regulatory functions. Arginine also is a precursor of two important amino acids, proline and glutamate [198]. One of the most important roles of arginine is its implication in the synthesis of polyamines through the diversion from NO synthesis pathway. Polyamines are known to promote tumor growth, invasion and metastasis [199]. Arginine also has a vital role in the synthesis of nucleotides, creatine, agmatine and hormones such as insulin and prolactin [200].

35

Tumor starvation by L-arginine deprivation

Polyamine synthesis

Cell signaling

e la hin oy nit m O r r b a C) ca OT ( se

ine e gin as Ar min i de

Ornithine

ns

de Ar ca gin rb in ox e yla se

tra

Ornithine

Agmatine

Citrulline

pyrroline-S-carboxylate Ornithine aminotranferase (OAT)

Arginase I Citrulline Arginine Citrulline

ide ox se Argininosuccinate c i a tr Ni ynth Lyase (ASL) s

Mitochondria

Argininosuccinate Synthetase (ASS)

Argininosuccinate

Figure 3.3.3 Arginine synthesis and homeostasis pathways. Arginine is synthesized as an intermediate in the urea cycle. Arginine homeostasis is mainly achieved by catabolism. In neonates, the gene expression of arginine anabolic enzymes such as 1-pyrroline-5-carboxylase, ASS and ASL is low. Thus, arginine is considered as an essential amino acid in neonates. After birth, the expression of ASS and ASL increases and expression of arginase is found undetectable at this stage [201]. Arginine can be degraded by arginase, ADC, ADI and NOSs (please note that ADI is not a mammalian enzyme). The products of arginine catabolism have important roles in tumor cell biology. For example, ornithine, the product of arginase, is diverted to polyamine synthesis via ornithine decarboxylase. NOSs degrade arginine into citrulline and NO. Citrulline is recycled to urea cycle, while NO is as a modulator of important metabolic and signaling cascades. Agmatine is synthesized by decarboxylation of arginine via ADC and has an important role in neurotransmission.

3.3.4 Arginine deiminase ADI (E.C.3.5.3.6) is a prokaryotic enzyme originally isolated from Mycoplasma, which catalyzes an irreversible deimination of the guanidine group of L-arginine to citrulline and ammonium ion [36]. Normal cells are able to convert citrulline into arginine through argininosuccinate synthetase (ASS) and ASL, expression of which are tightly regulated. However, the expression of ASS/ASL is downregulated in certain tumor cells by unknown mechanisms and these cells are unable to convert citrulline to arginine [30e33,37]. This makes the tumor cells auxotrophic for arginine for their growth and cellular functioning. ADI-mediated arginine deprivation leads to apoptotic cell death, selectively of arginine auxotrophic ASS () tumor cells sparing the ASS (þ) ADI resistant normal cells [38] (Table 3.3.1). Incidence of ASS deficiency varies depending on the tumor type and expression level of ASS has been proposed as a biomarker for identification of ADI sensitive tumors [24,25,39e42].

Table 3.3.1 Use of arginine catabolizing enzymes in ADT (Experimental studies)*. Enzyme used for deprivation

ADI

Cell line

Source and Cell type

Studies carried out

Reference

HSC-3 HSC-4

Human tongue squamous carcinoma

Cell growth inhibitory effect of ADI (purified from Mycoplasma infected cell lines) in comparison with arginase

[43]

CaSki C41 A549

T98G

Human cervix squamous Human carcinoma Human cervix squamous epithelium Human colon adenocarcinoma Human glioblastom

HeLa CHO FF9

Human cervix Chinese hamster ovary Fetal foreskin fibroblast

Concentration dependent effect of ADI on cell proliferation

[102]

HUVEC

Human umbilical vein endothelium Human stomach adenocarcinoma Mouse lymphoblastic leukemia

Anti-angiogenesis effect of ADI by inhibiting capillary-like tube formation Anti-proliferative effect and ADI induced cell cycle arrest and apoptosis Inhibition of cell division

[102]

Human mammary adenocarcinoma Human lung carcinoma

Effect of ADI on the regulation of cellular protein and polyamine synthesis

[85]

Human retinoblastoma

ASS expression related sensitivity of cells towards ADI

[204]

SCC

SNU-1 L5178Y MCF7 A549 SNUOT-Rb1 Y79

[202] [203]

ADI-PEG20

CWR22Rv1* A2058 SK-Mel-2 HUVE SaOS WAC2 Y-79 Meth AC 14 SK-LC-13* SW1271 NCI-H82 A375 SK-mel-2* SK-mel-28* SK-hep-2* SK-hep-3* HEP3B A2058*

Human prostate Human melanoma

Autophagy and caspase independent apoptosis Combination effect of ADI and TRAIL

[71] [66]

Human umbilical vein endothelium Human osteosarcoma Human neuroblastoma Human retinoblastoma Human sarcoma Human small cell lung

Cell cycle progression and apoptosis

[78]

Human melanoma

Inhibition of NO using PEGylated ADI

Effect of ADI-PEG20-mediated arginine deprivation on the production of NO ASS expression related sensitivity of cells towards PEG-ADI, induction of autophagy and caspaseindependent apoptosis Specificity of ADI for degradation of arginine and other amino acids; ASS expression dependent sensitivity of HCC and melanomas towards ADI

[103]

Involvement of Ras/PI3K/ERK pathway in induction of c-Myc stabilization and up-regulation of ASS

[61]

[39]

[47]

Human HCC

Human melanoma

SK-MEL-2 MDA-MB-231 Karpas-422

Human breast Human B-cell lymphoma

MyLa

Human T-cell lymphoma

Correlation between ASS methylation status and sensitivity of the cells towards ADI [26] Continued

Table 3.3.1 Use of arginine catabolizing enzymes in ADT (Experimental studies)*.dcont’d Enzyme used for deprivation

Cell line

Source and Cell type

Studies carried out

Reference

OEC-M1 SCC-15 HONE-1 A375 Sk-Mel2 A2058 MEL-1220 MlA-PaCa-2* PANC-1 Capan-1 HPAF II

Human head and neck cancer

Potential clinical correlation between ASS expression and tumor prognosis

[205]

Human melanoma

The role of ASS gene expression in ADI response/resistance

[72]

Human pancreatic cancer

The role of ASS gene expression in ADI response/ resistance

[74]

L1210

Murine lymphocytic leukemia Human cervical adenocarcinoma Human osteogenic sacroma Human melanoma

Cell proliferation and non-recoverable cell death of malignant cells on restoration of arginine Cell proliferation and ASS expression dependent

[187]

Pig intestinal porcine epithelial cells -I Human prostate

LPS- induced cell damage involving mTOR and TLR4 pathways Expression levels of ASS and OCT, rhArginase I-mediated modulations in mTOR signaling pathway

SeaX

Bovine liver arginase

HeLa SAos-2 A375 MEWO rh-Arginase I

IPEC-1 PC-3 DU-145

[132] recycling of citrulline to arginine [206] [156]

LNCap A375* SK-MEL-2 SK-MEL-28 B16-F0 L1210 HeLa rhArginase I-PEG5000mw

HEP-3B* Huh7 PLC/PRF/5 SK-HEP-1 SK-MEL-28 CCRF-CEM* Jurkat Molt-3 HepG2* Hep3B* HepG2* PLC/PRF/5*

Human melanoma

Mouse melanoma Murine lymphocytic leukemia Human cervical adenocarcinoma Human HCC

Human liver adenocarcinoma Human melanoma Human T-ALL

Human HCC Human HCC

Proliferation and cell cycle progression of melanoma cells, modulations in the cell cycle and apoptosis-related genes

[139]

Rescue of the arginase treated cells by norvaline (arginase inhibitor)

[207]

Gene expression profiling of ASS and OTC, Synergistic effect of PEGylated rhArginase I with 5Fluorouracil on cell growth inhibition [131]

Combination effect of PEGylated rhArginase I with Cytarabine (Ara-C) on expression of cyclins

[172]

Effect of PEGylated rhArginase I on its anti-tumor efficacy, immunogenicity and circulation half-life Cell cycle progression and transcriptional modulations of cyclins and/or CDKs

[133] [140] Continued

Table 3.3.1 Use of arginine catabolizing enzymes in ADT (Experimental studies)*.dcont’d Enzyme used for deprivation

Cell line

Source and Cell type

Studies carried out

Reference

CCRF-CEM* Molt-4 H9 Lousy Jurkat HPB-ALL KOPTK1

Human T-ALL

Global arrest in protein synthesis; Central role of phosphor-elF2a signaling and the kinases (GCN2 andvPERK) in the induction of T-ALL cell apoptosis by rhArginase I-PEG5000mw,

[173]

HepG2* Panc-1*

Human HCC Human pancreatic carcinoma

Effect of Co2þ substitution of the Mn2þ on catalytic activity and stability of human arginase I

[135]

Hep3b A375

Human HCC Human melanoma

Effect of Co2þ substitution of the Mn2þ on cytotoxicity

[134]

Hep3B

Bioengineered human arginase 1

Abbreviations: ADT, arginine deprivation therapy; ADI, arginine deiminase; ASS, argininosuccinate synthetase; NO, nitric oxide; T-ALL, T-cell lymphoblastic lymphoma. * Indicates tumor xenograft experiments.

41

Tumor starvation by L-arginine deprivation

US FDA granted orphan drug status to pegylated recombinant Mycoplasmal ADI for the treatment of HCC and malignant melanomas

Role of arginase in macrophagemediated tumor cell cytotoxicity [Ref.215] Elevated requirement of arginine by tumor cells revealed

1947

1965

Use of Ox liver arginase for regression of carcinoma [Ref.214]

1990

1992

Growth inhibition of Mycoplasma infected human tumor cells was observed

ADI-PEG20 is currently undergoing Phase III trial in patients with advanced HCC

A bio-engineered form of human arginase I was developed by the co-factor replacement

Growth inhibitory activity of ADI in human melanoma cells

1978

Pegylation of Bioengineered Co2+arginase to enhance circulation persistence

1993

1999

2005

Mycoplasma arginini ADI was successfully pegylated

2010

2011

2013

2014

Phase II study of ADIPEG20 for nonresectable and metastatic HCC

EMEA granted orphan drug status to ADI-PEG20 for the treatment of HCC

Phase I clinical trials of pegylated recombinant human arginase I completed in patients with advanced HCC

Figure 3.3.4 Timeline of important advancement in arginine deprivation therapy of cancer.

In 1990, Miyazaki et al. [43] were the first to report the growth inhibition of Mycoplasma infected human tumor cells. The cause of growth inhibition of human tumor cell lines was identified as a ADI produced by Mycoplasma. In vitro growth-inhibitory dose of Mycoplasmal ADI appeared to be 1000 times lower than that of bovine liver arginase. Subsequently in 1992, growth-inhibitory activity of ADI was demonstrated in ASSdownregulated human melanoma cells [44]. These pioneering studies established ADI as a potential anti-cancer enzyme (Figure 3.3.4).

3.3.5 PEGylated ADI Being microbial in origin, ADI has serious disadvantages of eliciting strong antigenicity and rapid plasma clearance (half-life of 4 h). To circumvent these limitations, several studies have aimed to extend the plasma half-life of ADI and to minimize its antigenicity. In 1993, Takaku et al. [45] addressed these problems for the first time by polyethylene glycol (PEG) modification. Remarkably, PEGylation of Mycoplasma arginini ADI enhanced its cytotoxic potential in vivo and once a week intravenous injection of PEG-ADI at a dose of 5 U per mouse (10 mg protein per kg) depleted plasma arginine to an undetectable level at least for a week, whereas native enzyme required 10 daily injections to achieve similar effects. Nevertheless, PEGylation of

42

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

Mycoplasma hominis ADI also resulted in significant enhancement of arginine lowering potential of native M. hominis ADI [46,47]. Recently, PEGylation and pharmacological properties of an engineered ADI originated from Pseudomonas plecoglossicida have been studied. PEGylated P. plecoglossicida ADI remarkably improves the stystemic half-life (by 11-folds) and found to exhibit superior efficacy than native ADI in depleting plasma arginine [48]. PEG-ADI has also shown promising outcomes for the treatment of human malignancies. In March 1999, ADI-PEG20, PEGylated recombinant Mycoplasmal ADI was approved as an orphan drug by the US Food and Drug Administration for the treatment of hepatocellular carcinoma (HCC) and malignant melanomas. Subsequently in July 2005, European Agency for the Evaluation of Medicinal Products granted orphan drug status to ADI-PEG20 for the treatment of HCCs [49]. ADI-PEG20 is currently undergoing clinical investigation as a randomized double-blind phase III trial in patients with advanced HCC (NCT 01287585), phase II studies in patients with ASS-negative metastatic melanoma (NCT 01279967) and phase II studies in patients with relapsed small-cell lung cancer (NCT 01266018) [50] (Table 3.3.2). Outcomes of the previous clinical studies were also encouraging, achieving response rates of 25 and 47% in melanoma and HCC, respectively (Table 3.3.2). Moreover, grades III and IV toxicities have not been observed in clinical investigations involving ADI-PEG20 in metastatic melanoma and HCC patients [51,52]. Therefore, clinicians are looking forward to the establishment of ADI-PEG20 as a potent anti-cancer modality.

3.3.6 Tumor sensitivity toward ADI The auxotrophicity of tumors toward arginine and their sensitivity toward it can be attributed to the lack or reduced expression of ASS in tumors [25,37e39,53]. Notably, numerous tumor cells that are deficient in ASS expression, are sensitive toward ADI treatment (Table 3.3.1). Transfection of an expression plasmid containing human ASS cDNA in HCC and melanoma cells confers severe resistance to ADI treatment compared with ASS-negative cells [47]. Till date, most promising targets for ASS expression-dependent ADT identified are human melanoma and HCCs. Other promising targets include malignant pleural mesothelioma, renal cell carcinoma, prostate cancer, T-ALL and osteosarcoma [50].

Table 3.3.2 Clinical investigations involving Phase of a clinical trial Enzyme Cancer type

ADI-PEG20

arginine depriving enzymes. Number of patients

Clinical outcomes

II

71

SD: 31% (22/ 71) DCR: 31% (22/71)

ASS (-) melanoma

I

17

PR: 23.5% (4/ 17) SD: 29.4% (5/17) CBR: 52.9% (9/17)

HCC

I/II

19

CR: 11% (2/19) PR: 37% (7/19) SD: 37% (7/19)

MM

I/II

24

OR: 25% (6/24) SD: 25% (6/24)

< 2 mM

[63]

Undetectable

[25]

< 2 mM

[208]

< 2 mM

[52]

Continued

43

Hypersensitivity/skin rash, local tissue reaction at injection site, hyperuricemia, pruritus, fatigue, hyperammonemia, fever, diarrhea Mild/moderate discomfort at the intramuscular injection site, neutropenia and thrombocytopenia, anaemia, fatigue Occasional elevation in serum lipase, bilirubin and amylase levels, hyperuricemia, mild pain at the site of injection, increase in fibrinogen Mild pain at the site of injection, hyperuricemia, elevated serum lipase, bilirubin, amylase and LDH, decreased hemoglobin, platelet and WBC count

Reference

Tumor starvation by L-arginine deprivation

HCC

Common side effects

Post-treatment levels of plasma argininea

Table 3.3.2 Clinical investigations involving arginine depriving enzymes.dcont’d

HCC

II

76

OR: 3% (2/76) SD: 61% (50/76)

MM

II

36

Melanoma

I/II

31

MPM

II

39

HCC

III

NonHodgkin’s Lymphoma SLCL

II

ORþSD: 28% (10/36) SD: 31% (9/29) PMR: 27% (8/29) PMR: 46% (18/ 39) SD: 31% (12/39) Ongoing (NCT01287585) Ongoing (NCT01910025)

II

Clinical outcomes

Ongoing (NCT01266018)

Common side effects

Transient and reversible encephalopathy, skin irritation, or discomfort at the site of injection combined with low-grade fever, decreased serum sodium, hemoglobin, albumin, fibrinogen levels, increased Potassium levels, uric acid and lipase Discomfort at the injection site Pain and rash at injection site, nausea, anorexia, pruritus, arthralgia Skin injection site reactions, neutropenia, anaphylactoid reactions, serum sickness

Post-treatment levels of plasma argininea

Reference

Undetectable

[51]

[209] Undetectable

[210]

2 mMb

[64,211]

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

Cancer type

Number of patients

44

Enzyme

Phase of a clinical trial

MM

I

Ongoing (NCT01665183)

ADI-PEG20 plus Cispilatin and Pemetrexed

Arginine auxotrophic tumors such as MPM and NSCLC

I

Ongoing (NCT02029690)

ADI-PEG 20 plus Docetaxel

Solid Prostate and NSCLC tumors

I

ADI-PEG20 Plus Doxorubicin

HER2 (-) Breast Cancer

I

Peg-rhArgI

HCC

I

Peg-rhArgI plus Oxaliplatin and Capecitabine

HCC

II

18

PR: 6% (1/18) SD: 33% (6/18)

Undetectable

[212,213]

< 8 mM

[141]

Ongoing (NCT01948843) 15

SD:26.7% (4/15)

Abdominal pain, diarrhea, nausea, elevated ALT, AST, GGT & bilirubin

Ongoing (NCT02089633)

45

Continued

Tumor starvation by L-arginine deprivation

ADI-PEG20 plus Cispilatin

Cancer type

Peg-rhArgI (the secondline therapy after sorafenib)

HCC

II

Number of patients

Clinical outcomes

Common side effects

Post-treatment levels of plasma argininea

Reference

Ongoing (NCT02089763)

Abbreviations: ALT, Alanine Transaminase; AST, Asparate Transaminase; CBR, Clinical benefit rate; CR, Complete response; DCR, Disease-control rate (complete/ partial responseþstable disease); GGT, Gamma-glutamyl transferase; HER2, Human epidermal growth factor receptor 2; MM, Metastatic melanoma; MPM, Malignant Pleural Mesothelioma; NSCLC, Non-Small Cell Lung Cancer; OR, Overall response (Completeþpartial response); OS, Overall survival; Peg-rhArg1, Pegylated recombinant human arginase 1; PR, Partial response; PMR, partial metabolic response; SD, Stable disease; SLCL, Small Cell Lung Cancer. a Basal (Pre-treatment) level of arginine was w 130 mM. b Basal (Pre-treatment) level of arginine was w 63 mM.

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

Enzyme

Phase of a clinical trial

46

Table 3.3.2 Clinical investigations involving arginine depriving enzymes.dcont’d

Tumor starvation by L-arginine deprivation

47

However, molecular mechanisms underlying tumor sensitivity toward ADI treatment, by downregulation of ASS expression in tumor cells, are still elusive. Promoter hypermethylation-dependent silencing of ASS gene is an endorsed mechanism of ASS gene repression [37,54e56]. Methylation frequency of the ASS promoter upto 50e80% level at the CpG loci is documented across a broad range of lymphomas. In contrast, normal lymphoid samples were found unmethylated [26]. Treatment of ADIPEG20 to ASS-methylated lymphoma cell lines revealed dramatic decrease in the proliferation rate and viability count, by inducing caspase-dependent apoptosis, without affecting normal lymphoblastoid cell lines. Demethylation-induced resistance to ADI-PEG20 treatment has also been confirmed in cutaneous T-cell lymphoma cell lines, as their incubation with 5-Aza-dC (demethylating agent) for 8 days which resulted in partial demethylation, followed by transcriptional activation and synthesis of ASS protein [26]. Recently, Rabinovich, Adler L, Yizhak et al. [57] have confirmed that proliferation of the osteosarcoma cells is supported by downregulation of ASS, by facilitating pyrimidine synthesis via activation of CAD (carbamoylphosphate synthase 2, aspartate transcarbamylase and dihydroorotase) complex. As cytosolic aspartate serves as a substrate for both ASS and for CAD complex, ASS downregulation can enhance aspartate availability for CAD for the synthesis of pyrimidine nucleotides to promote proliferation. Thus, aspartate transport can be exploited as an additional therapeutic target in tumors with ASS downregulation, especially in those ones which develop resistance to arginine-depriving enzymes.

3.3.7 Tumor resistance toward ADI ASS-deficient tumors are sensitive to ADI treatment; however, arginine deprivation eventually upregulates ASS expression in tumor cells and thereby confers resistance toward ADI [25,58]. Transcriptional induction of ASS expression and increase in ASS mRNA level is reported in human embryonic kidney cells and melanoma cells during arginine starvation [59,60]. Transcription factors such as c-Myc and HIF-1a are involved in the upregulation of ASS expression under arginine-depleted conditions [60]. E-box and GC-box are the important sequences located between 85 and 35 nucleotides in the ASS promoter region that modulate ASS expression through their interactions with c-Myc and HIF-1a. Under the normal concentrations of arginine, HIF-1a (but not c-Myc) binds to E-box and thus acts as a negative regulator of ASS expression.

48

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

Under the conditions of arginine depletion, HIF-1a is degraded and replaced by up-regulated c-Myc, which directly binds to E-box; thus, c-Myc acts as a positive regulator of ASS expression (Ref. [60]; Figure 6). Recently reported in melanoma cells, inhibition of ubiquitin-mediated protein degradation is a molecular mechanism responsible for the stabilization and accumulation of c-Myc [61]. Furthermore, various cellular pathways, such as Ras and its downstream ERK/PI3K/AKT kinase cascade are associated with the post-translational modifications of c-Myc, leading to its phosphorylation and stabilization during ADI-PEG20-mediated arginine deprivation conditions. Involvement of Ras/PI3K/ERK signaling pathway in the development of resistance toward ADI treatment suggests that combination of ADI with Ras/ERK, PI3K/AKT inhibitors is a potential therapeutic strategy to improve the anti-cancer response [62,63]. Development of anti-drug neutralizing antibodies is another possible mechanism of resistance toward ADI-PEG20 treatment [64]. Arginine concentrations were recovered up to pre-treatment levels in a patient with malignant pleural mesothelioma and in Asian patients with advanced hepatocellular carcinoma following the ADI-PEG20 treatment. This recovery in arginine concentration was found concomitant with an increase in antieADI-PEG20 antibody titer [65]. These studies suggest the involvement of drug-associated resistance i.e. anti-drug neutralizing antibodies, rather than tumor-related factors as another possible mechanism of resistance of some tumor cell types toward ADI-PEG20 treatment [62,63].

3.3.8 Anti-tumor mechanisms of ADI treatment 3.3.8.1 Role of autophagy and apoptosis in ADI-mediated arginine deprivation therapy Due to the involvement of arginine in numerous cellular pathways (Figure 3.3.2), the exact anti-proliferative mechanisms of ADI treatment, besides that of arginine depletion, are still elusive. One of the potential pathways involved in the cytostatic and cytotoxic potential of ADI is TRAIL (tumor necrosis factor-related apoptosis-inducing ligand) [66e68]. TRAIL has an important role in the cleavage of Beclin-1 (Atg6) and Atg5 in arginine-deprived melanoma cells [69]. Beclin-1 and Atg5 are essential for the formation of autophagosomes and thus crucial for autophagy. Since autophagy serves as a mean to evade apoptosis in arginine-depleted cells, TRAIL induced cleavage of Beclin-1 and Atg5 leads to decreased autophagy, thereby increasing apoptosis [69]. In addition, these two drugs (ADI and TRAIL) complement each other by activating the intrinsic apoptosis

Tumor starvation by L-arginine deprivation

49

pathways. ADI-PEG20 increases cell surface receptors DR4/5 for TRAIL thereby binding TRAIL to these death receptors. As a result, caspase-8 or 10 are activated [66]. ADI-PEG20 treatment also modulates different autophagic pathways involved in the cell survival. Adenosine 50 -monophosphate-activated protein kinase and ERK pathways are activated in ADI-treated prostate cancer cells; while AKT, mTOR and S6K pathways are attenuated. ADI-PEG20 treatment to CWR22Rv1 prostate cancer cells induced autophagy, as revealed by the appearance of LC-II only after 30 min exposure continues its persistence after 24 hours following ADI-PEG20 treatment [70,71]. Additionally, inhibition of autophagy by chloroquine, a clinically approved anti-malarial agent which inactivates lysosomal functions, accelerates the ADI-induced apoptotic cell death of prostate cancer [70,71] and small-cell lung cancers [39]. Thus autophagy has been proposed as a pro-survival mechanism of tumor cells during arginine deprivation [71]. ADI-mediated arginine deprivation is also known to induce caspasedependent apoptotic pathways in many of the tumor cells types. ADIPEG20 treatment activates caspase-3 in ASS-methylated malignant lymphoma cells, whereas ASS-positive normal lymphoblastoid cells are resistant to it [26]. Similarly, cell death has been attributed to caspases activation in glioblastoma [54], melanoma [38,72], leukemia [73] and pancreatic cancer cells [74]. Moreover, all these studies indicate that inhibition of autophagy leads to further advancement in the ADI-PEG20mediated demise of tumor cells, suggesting the induction of autophagy as a mechanism of tumor resistance to ADI-PEG20 treatment. Cumulative pieces of evidence suggest that the activation of caspases is not a sole decisive phenomenon in programmed cell death pathways. Caspase-dependent apoptosis is a major mode of cell death, but in its absence or failure, there are other pathways which can also execute cell death [75e77]. ADI-PEG20 treatment to small-cell lung cancer, leukemia, retinoblastoma and prostate cancer cells induces apoptotic cell death pathways; however, without activation of caspases, suggesting the role of caspase-independent apoptosis as a cell death pathway [33,39,69,70,78]. The inter-membrane space of mitochondrion contains proteins such as apoptosis-inducing factor (AIF) and endonuclease G (EndoG), which can induce apoptotic cell death in a caspase-independent fashion [79].

50

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

EndoG is one of the predominant endonucleases that are involved in the regulation of cellular functions such as mitochondrial biogenesis, DNA synthesis and repair. AIF is an FAD-containing flavoprotein which plays an important role in the stability of an electron transport chain [80]. Nutrient deficiency-mediated stress signals induce mitochondrial outer membrane permeabilization, which consequently releases inter-membrane space proteins such as AIF, EndoG and cytochrome c. AIF has a role of central mediator in caspase-independent cell death pathway [81]. AIF, once released into the cytosol, interacts with EndoG and cyclophilin A before its translocation into the nucleus [82]. Subsequently after translocation into the nucleus, it triggers cell death either directly, through interaction with DNA, or indirectly, through the production of reactive oxygen species [73,74,79,80]. Mitochondrial outer membrane permeabilization promotes both, caspase-dependent and caspase-independent apoptotic pathways, but with different kinetics [83]. Although, the upstream signaling stimulus for both, a caspase-dependent and caspase-independent pathway is the same, that is, via induction of mitochondrial outer membrane permeabilization, their downstream pathways are different. Moreover, nuclear alterations and the changes occurring in mitochondrial trans-membrane potential during caspase-independent pathways are different than those observed in a caspase-dependent apoptotic pathway [84]. To summarize, growing evidence suggests that autophagy is a prevailing cell survival mechanism in tumor cells undergoing ADI-mediated arginine deprivation. The overall cellular response to ADI-mediated arginine deprivation in different tumor cells operates through a complex cascade, initiating with induction of autophagy and followed by the activation of either caspase-dependent or caspase-independent cell death pathways. It is worth emphasizing that the discrepancy of cellular responses of tumor cells to ADI-mediated arginine depletion in activation of either caspasesdependent or caspases-independent cell death pathways can vary depending on tumor cell type [38,39,70,71,74]. As a result, the precise mechanisms of tumor cell deathdconsequential of cellular response to ADI-mediated arginine depletiondappear to be complex and variable, and need to be further elucidated.

Tumor starvation by L-arginine deprivation

51

3.3.9 Inhibition of de novo protein synthesis by ADI-mediated arginine deprivation Inhibition of de novo protein synthesis is another mechanism, which can be attributed to the anti-tumor potential of ADI. As extracellular arginine pool is responsible for 40% of de novo protein synthesis, ADI treatment to human lung carcinoma cells results in an anti-proliferative effect, mediated by inhibition of protein synthesis [85]. Arginine is present in various compartments such as extracellular, intracellular and citrulline-arginine regeneration, that is, cytosolic compartment and it is known to regulate various cellular pathways differently. Protein synthesis mainly utilizes arginine either from the intracellular pool or the citrulline-arginine regeneration mechanism, while polyamines synthesis largely utilizes arginine pool from the intracellular origin [86,87]. Polyamines are synthesized through the methionine salvage pathway via decarboxylation of S-adenosylmethionine. S-adenosylmethionine is a donor metabolite necessary for the transfer of methyl group to DNA and proteins. Human colon cancer (HCT116) cells treated with short hairpin CD44 RNA interference showed a decrease in the total amount of methionine-pool metabolites including polyamines, suggesting the role of polyamines in cancer proliferation [88]. ADI treatment toward human mammary adenocarcinoma and lung carcinoma cells differently modulates polyamine synthesis and the global protein synthesis. Interestingly, inhibition of protein synthesis has been correlated with the ASS-mediated regeneration of arginine. Cells expressing low levels of ASS (A549) result in decreased protein synthesis (without affecting polyamine synthesis) and those expressing higher ASS levels (MCF-7) are resistant to ADI treatment, as the decreased arginine levels can be replaced by citrulline-arginine regeneration pathway [85].

3.3.10 Anti-angiogenic effects of ADI-mediated arginine deprivation As a tumor grows beyond a certain size (2 mm in diameter for most solid tumors), available vasculature within the tumor becomes inadequate to supply sufficient quantities of essential nutrients for their growth [89]. This results in the generation of hypoxic tumor microenvironment and leads to the development of new blood vessels (angiogenesis) as a colossal requisite of the developing tumors [90]. Accordingly, neovascularization can be stated as one of the decisive phenomena during tumor growth and metastasis [91].

52

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

Emerging studies now indicate that not only molecular signals but also metabolic mechanisms regulate angiogenesis [92]. Under stress conditions such as hypoxia, tumor cells secrete angiogenic factors such as vascular endothelial growth factor (VEGF) [93]. Increased levels of VEGF activate VEGF receptor 2 (VEGFR2) signaling in the quiescent endothelial cells which in turn initiate angiogenesis [94e96]. Endothelial cells produce 85% of their total amount of ATP via glycolysis. Addiction of endothelial cells on anaerobic rather than aerobic pathway enables them for the formation of vascular sprouts in hypoxic areas [97,98]. Metabolism of tumor endothelial cells resembles that of highly activated endothelial cells because of the tumor induced switch from quiescence to proliferation due to metabolically regulated migration during sprouting [99,100]. Besides ADI’s role in modulation of apoptotic pathways, it has an antiangiogenic activity that contributes to its anti-tumor potential. The growth, migration and differentiation of human umbilical vein endothelial cells are strongly impaired in a medium containing recombinant ADI [101]. As a consequence; it results in decreased tube formation with intermittent and incomplete microvascular network. Similarly, Park et al. [102] found that E. coli ADI inhibits angiogenesis by inhibiting tube formation of endothelial cells and neovascularization in Chick Chorioallantoic membrane and Matrigel plug assay. Suppression of nitric oxide (NO) generation is also another possible mechanism for anti-angiogenic activity of ADI. Since L-arginine is required for nitric oxide synthases (NOSs) to generate NO, the depletion of arginine by ADI suppresses NO synthesis [102]. Potential role of ADI-mediated arginine depletion in inhibition of NO synthesis has been reported [103,104]. We and others have previously reported that NO promotes tumor growth through the stimulation of angiogenesis [105e107] and regulates cellular interaction by controlling adhesion molecule expression and ultimately cell adhesion [108,109]. NO directly, or indirectly through NO-mediated reactive nitrogen species, induces the activation of certain angiogenic signaling pathways in the endothelial cells [110]. NO acts as an autocrine mediator in endothelial cell functioning and as a final modulator in VEGF-stimulated angiogenesis [109,111]. NO not only mediates angiogenesis but also subsequent vessel maturation [112,113]. Moreover, NO is known to inhibit angiostatin and thrombospondin-1, two main inhibitors of angiogenesis [114]. Owing to the important role of NO in angiogenesis, ADI inhibits tumor growth not only by draining the

Tumor starvation by L-arginine deprivation

53

supply of arginine but also by its anti-angiogenic activity via suppression of NO generation. To summarize, certain tumor cell types such as, HCCs and metastatic melanomas are invariably deficient in ASS expression and can be specifically targeted by ADI-mediated ADT. It is worth noting that more than one pathway may be attributed to the cytotoxic potential of ADI-mediated ADT (Figure 3.3.5). The anti-tumor potential of ADI may not only be simply accredited to its action as arginine degrading enzyme but also to several other mechanisms important in the cellular functioning of tumor cells. Induction of apoptotic pathways, inhibition of angiogenesis and inhibition of de novo protein synthesis are the important mechanisms attributed to the cytotoxic potential of ADI. Moreover, studies have revealed the ADI-mediated modulations in tumor cell cycle. The fundamental difference of cell cycle modulations in normal and malignant cells should be exploitable as a means of selective demise of tumor cells and ADI, in combination with other anti-cancer chemotherapeutic agents, which can be a potential strategy to improve chemosensitization against tumor cells [115e118].

Arginine deiminase

Arginine

Arginine decarboxylase Arginase

Arginine deprivation Blood vessels

Tumor cell growth Anti-angiogenesis

Cell cycle arrest

Autophagy (?) Apoptosis

Inhibition of global protein synthesis

Tumor cell death

Tumor growth arrest

Figure 3.3.5 Schematic representation of cytostatic and cytotoxic pathways involved in arginine deprivation therapy. ADT can potentially modulate numerous cellular and signaling pathways rendering their cytotoxic and cytostatic pathways. Induction of apoptotic pathways, inhibition of angiogenesis and inhibition of de novo protein synthesis are the important mechanisms attributed to the cytotoxic potential of ADT. Moreover, ADT-mediated modulations in tumor cell cycle can be exploited as a means of tumor growth arrest.

54

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

3.3.11 Arginase Arginase (E.C.3.5.3.1) is a mammalian enzyme which catalyzes the conversion of arginine to ornithine and urea. Arginase is considered as an enzyme responsible for the cyclic nature of urea cycle, since only the organisms containing arginase are able to carry out the complete urea cycle [119]. Two distinct isoforms of mammalian arginase have been identified that are encoded by two separate genes [120]. Type I arginase (arginase I) is located in the cytosol and is mainly expressed in liver. Type II arginase is located in the mitochondrial matrix and is expressed in extra-hepatic tissues [121,122]. Intracellular regulation of arginase expression is of immense importance as it has crucial implications for the synthesis of essential cellular metabolites [123], For example, cytosolic co-localization of arginase I with ornithine decarboxylase (ODC) preferentially utilizes ornithine for the biosynthesis of polyamine. On the other hand, due to its co-localization with ornithine aminotransferase in the mitochondria, arginase II directs ornithine for the production of proline and glutamine [124,125].

3.3.12 PEGylated recombinant human arginase I Elevated requirements of arginine by tumor cells were first identified in 1947 and preferential utilization of arginine by tumor bearing animals was revealed in 1953 [126,127]. The use of bovine and murine arginase in ADT was prevailing until the advent of recombinant DNA technology [128e130], followed by the pervasive use of recombinant human arginase in subsequent decades [131,132]. Arginase from bovine and murine sources has been extensively used for the ADT in vitro. However, limited success was achieved in vivo because of its alkaline optimum pH and very low affinity for the substrate. Human arginase I also has a serious limitation of very short circulatory half-life (w30 min). To extend plasma half-life of arginase, PEGylation has been applied successfully. PEGylated recombinant human arginase I (rhArg-Peg5000mw) had efficient catalytic activity at physiological pH with improved in vivo half-life of 3 days. Furthermore, rhArg-Peg5000mw was found to have significant tumor inhibitory activity in BALB/c nude mice bearing HCC xenografts [131]. Notably, these results were consistent with those demonstrated by Tsui et al. [133]. Recently, a bioengineered form of human arginase I was developed by the co-factor replacement, the replacement of two Mn2þ ions by Co2þ ions. The modified Co2þ-arginase I resulted in 10-fold increase in the catalytic activity and five-fold greater stability at the

Tumor starvation by L-arginine deprivation

55

physiological pH. Nevertheless, IC50 values for killing human HCC and melanoma cell lines were lowered by 12e15 folds [134]. More recently, modifications in bioengineered Co2þ-arginase I were performed by conjugating 5-kDa PEG to enhance plasma half-life. This modified version of bioengineered arginase I (Co-hArgIePEG) was proven to be cytotoxic by significantly increasing the expression of caspases-3 in HCC and pancreatic carcinoma tumor xenografts [135]. Lately, the cytotoxic potential of Co-hArgIePEG was identified in acute myeloid leukemia and glioblastoma cells. Acute myeloid leukemia cell lines were found sensitive toward Co-hArgIePEG-mediated arginine deprivation with very low (58e722 PM) IC50 values, suggesting a very high potential of Co-hArgIePEG-mediated arginine depletion in acute myeloid leukemia cells [136]. Moreover, Co-hArgIePEG-mediated arginine deprivation has been demonstrated to induce caspase-independent, non-apoptotic cell death in human glioblastoma cells [137]. Alternative method to extend the plasma half-life of recombinant human arginase also has been established. Plasma half-life of a fusion protein form of a recombinant human arginase (rhArgFc, constructed by linking rhArg to the Fc region of human immunoglobulin IgG1), was evidenced to significantly extend upto w 4 days [138]. In addition, rhArg-Fc was confirmed to conspicuously inhibit the cell growth of human HCC cells in vitro and in vivo [138]. Past decade has evidenced a prevalent use of recombinant human arginase-mediated ADT in numerous cancer cell types, mainly metastatic HCC and melanomas [131,139,140]. Currently, PEGylated derivative of recombinant human arginase I is undergoing clinical trials for the treatment of human HCC [141,142]. Moreover, initiatives are now being taken to overcome the possible problem of accumulation of PEGylated products in the liver by impending approaches such as fusion proteins [138].

3.3.13 Anti-tumor mechanisms of arginase-mediated arginine deprivation Selective starvation of L-arginine in tumor cells, which are auxotrophic for L-arginine, is one of the most important anti-tumor mechanisms of ADT. Arginase can render its cytostatic effect as a result of modulations in the cell cycle proteins, whereas, cytotoxic effects rendered by arginase I-mediated arginine deprivation have been proposed as a result of induction of potential cell death pathways namely apoptosis and probably by “autophagic cell death.” Summarized below is the current understanding of the molecular mechanisms of cytostatic and cytotoxic effects rendered by arginase-mediated ADT.

56

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

3.3.14 Role of autophagy in arginase-mediated arginine deprivation Autophagy is a key sensing and regulatory mechanism of cells in nutrient deprived conditions. Under stress conditions, autophagy functions as a bio-energy management system by recycling cell organelles and damaged and/or long-lived proteins [143]. Although autophagy seems to be a survival mechanism of the cells, there is a growing evidence of accumulation of autophagosomes and other autophagic markers in dying cells unable to process apoptosis, raising the term “autophagic cell death” [144e147]. However, the term “autophagic cell death” is based on morphological features rather than the causative role of autophagy in cell death. New definition of “autophagic cell death” has been proposed, implying that cell death must occur without the involvement of apoptotic machinery, (caspase activation) but with an increase in autophagic flux [148,149]. Mammalian target of rapamycin (mTOR) is a key regulator of coupling cell growth and nutritional status of the cell [150,151]. Autophagy is induced by the inhibition of mTOR-signaling pathway [152]. During nutrient affiuent conditions, mTOR is involved in the negative regulation of Atg1 (autophagy-related gene 1) which inhibits autophagy [153,154]. Arginasemediated arginine deprivation leads to decreased levels of ATP, which in turn activates the adenosine 50 -monophosphate-activated protein kinase. Activated adenosine 50 -monophosphate-activated protein kinase eventually inhibits the mTOR-signaling pathway, manifested by the reduced phosphorylation of key downstream molecules, such as 4E-BP1 (eukaryotic translation initiation factor 4E-binding protein-1). Dephosphorylation of 4EBP1 is observed in Chinese hamster ovary (CHO), human melanoma cells and human prostate cancer cells following their exposure to recombinant human arginase I [65,155,156]. Phagosome/lysosome activity is also significantly increased following an incubation of human tumor cells in L-arginine-deficient medium [157]. Additionally, studies carried out by Hsueh et al. [156] evidenced no significant induction of apoptotic mechanisms in prostate cells after their exposure to rhArgI, suggesting the role of autophagic cell death, rather than apoptosis, as an alternative cell death mechanism. In addition, autophagy has often accompanied damaged mitochondria and higher levels of reactive oxygen species [158,159]. Acute generation of reactive oxygen species has been attributed to causing severe damages to the cellular macromolecules, which in

Tumor starvation by L-arginine deprivation

57

consequence, leads to necrosis of the tumor cells [160,161]. Overall, arginase leads to deprivation of arginine, in consequence, it inhibits mTOR pathway during the deprivation and thus forcing tumor cells to undergo ‘autophagic cell death’ pathway [162]. SLC38A9, a member 9 of the solute carrier family 38, has been recently identified as an integral component of the lysosomal machinery that controls amino acid-induced mTOR activation [163,164]. Amino acid starvation in human embryonic kidney (HEK293T) cells with stable expression of SLC38A9 has been shown to activate mTOR in a sustained manner. Moreover, shRNA-mediated silencing of SLC38A9 results in a reduction of arginine-induced mTOR activation. Also, depletion of SLC38A9 impaired mTOR activation induced by cycloheximide [ibid.] (a protein synthesis inhibitor which induces accumulation of intracellular amino acids), further suggests the role of SLC38A9 in mTOR activation at the lysosomal rather than at the plasma membrane. These studies have demonstrated that SLC38A9 acts as an upstream positive regulator mTOR functioning and thereby modulating autophagy in arginine-deprived tumor cells. Although some studies have advocated autophagy as a cell death mechanism of arginase-mediated ADT [156,157], many groups have explained it as a pro-survival mechanism; mainly by postponing the activation of apoptosis [38,161]. Thus, understanding the exact role of autophagy in arginase-mediated cell death pathways is complicated [162,165]. Therefore, much need to be elucidated about these new findings related to “autophagic cell death” and caution must be taken to assign autophagy as a cell death pathway in arginase-mediated ADT.

3.3.15 Role of apoptosis in arginase-mediated arginine deprivation The role of autophagy, either in cell survival or in cell death, depends on many factors such as cell type, nature and severity of the stimuli and so on [166]. If the attempt of the cells to survive through autophagy fails, apoptotic pathways take over and ultimately cause cell death [143]. Inhibition of autophagy in amino acid deficient conditions induces tumor cell death, mainly because of further exacerbation of energy dearth [167,168]. Also, longer persistence of autophagy is proposed to eventually lead the activation of caspase-dependent cell death pathways, as autophagy and apoptotic cell death pathways are interconnected and also share some common pathways through the induction of the membrane permeability transitions [169e171]. Induction of apoptotic pathways is another

58

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

consequence of arginine depletion and anti-tumor mechanism of arginase I-mediated arginine deprivation. Involvement of apoptosis as a cell death mechanism in arginasemediated ADT has been illustrated in various literature reports. Annexin V is known to selectively stain the cells, which are destined for apoptosis or in the process of apoptosis. 33% of human melanoma cell population was destined for apoptotic cell death following rhArg treatment [139]. Arginase I-mediated arginine deprivation led to the transcriptional upregulation of caspase 3, the intrinsic mitochondrial pathway of apoptosis, which is marked by the change in mitochondrial membrane potential [172]. Recently, an anti-leukemic potential of PEGylated-arginase has been attributed to kinases general control nonderepressible 2 (GCN2)-mediated induction of apoptosis in T-ALL cells [173]. Cell cycle arrest by arginase-mediated arginine deprivation and combination approaches rhArg-Peg5000mw-mediated arginine deprivation in various HCC cells results in their cell cycle arrest at G2/M phase, by decreased expression levels of cyclin B1 and cdc2, or in S phase, by a transcriptional upregulation of cyclin A1 [140]. rhArg-Peg5000mwmediated arginine depletion was witnessed to impair the expression of cyclin D3 in T-ALL cells, which was followed by an arrest of the cells in the G0-G1 phase of the cell cycle and induction of apoptosis [172]. Recent investigations of rhArg-Fc-mediated arginine deprivation in human HCC cells exhibited cell cycle arrest at S phase [138]. The exact mechanisms of these findings are still elusive, but the possible reasons seem to be the increased expression of cyclin A and declined transcription levels of p27 and p21 (the key cyclin kinase inhibitors). Owing to the evidence of cell cycle arrest, a combination of arginase and other cell-cycle specific anti-cancer chemotherapeutics as potential anti-tumor approaches have been established. Synergistic effects of rhArgPeg5000mw with 5-fluorouracil (5-FU, uracil analog which interferes with RNA and DNA synthesis) and cytarabine (Ara-C, anti-metabolic chemotherapeutic agent) have been investigated on the inhibition of proliferation of HCC and T-ALL cells, respectively [131,172]. Treatment of either rhArg-Peg5000mw or Ara-C alone induces a heterogeneous antitumor effect in vivo, whereas, combined treatment of rhArg-Peg5000mw and Ara-C induces a homogenous prevention of spleen growth, leading to the prolonged survival in all of the T-ALL bearing mice [172]. Moreover, combined treatment of PEGylated recombinant human arginase I and oxaliplatin has been demonstrated to synergize the inhibiting

Tumor starvation by L-arginine deprivation

59

effect on tumor growth and enhanced overall survival probability as compared with PEGylated recombinant human arginase I or oxaliplatin treatment alone [174]. Altogether, arginase has an advantage over ADI that it is efficacious in both ASS-negative and ornithine transcarbamoylase (OTC) negative tumors [59], whereas ADI is efficacious only in ASS-negative tumors. The tumor cell types expressing ASS are resistant to arginine deprivation treatment by ADI [25,26,54,61,131]. Even though arginase has been considered as a potential drug candidate over a period of six decades, low substrate specificity (high km of 2e4 mM), short plasma life and optimum alkaline pH (pH 9.3) limit in vivo applications of arginase [131,140]. In addition, robust homeostatic mechanisms in the body allow faster restoration of plasma-free arginine, making in vivo arginine deprivation by arginase more difficult. Most of the scientific efforts nowadays pay attention to these limiting characteristics of arginase [134,175,176].

3.3.16 Arginine decarboxylase ADC (E.C. 4.1.1.19) metabolizes arginine to agmatine, one of the minor metabolic products of arginine. ADC is mainly found in plants, bacteria and mammalian liver and brain membranes [177,178]. The mammalian ADC is different from other sources and distinct but related to ODC [179]. Although, arginine decarboxylation by ADC is a minor metabolic route, its product i.e. agmatine has a significant role in numerous cellular pathways [180]. Agmatine modulates the polyamine metabolism through its negative interaction with ODC [181]. Agmatine also confers an inhibitory effect on intracellular polyamine content by inhibiting polyamine uptake [182] and probably by increased polyamine catabolism [183]. Mayeur et al. [184] has reported the effect of agmatine accumulation on polyamine metabolism, cell proliferation and cell cycle distribution in human colon adenocarcinoma epithelial cell lines. Because of the agmatine-mediated reduction in polyamine synthetic capacity of the cells, agmatine markedly inhibits the cell proliferation of HT-29 and Caco-2 cells in a dose dependent manner, without affecting cell membrane integrity. Moreover, agmatine modulates the cell cycle progression by decreasing ODC activity and expression [181,185]. As ODC plays an important role in the G1/S progression of the cells, agmatinemediated modulations in ODC expression lead to modifications in the cell cycle progression [186]. Additionally, agmatine also has been shown to

60

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

delay the expression of cyclins in tumor cells, leading to the modifications in the cell cycle progression [184]. ADC has been investigated for the enzymatic degradation of arginine in normal and malignant cell cultures [187]. Arginine deprivation in human diploid fibroblasts (normal cells), achieved using human recombinant ADC, resulted in the cell cycle arrest at G1/G0. While treatment of 0.1 unit ml1 ADC to HeLa (Human cervical cancer) cells resulted in cell cycle arrest with an initiation of cell death after 2 days [187]. Similar results were evidenced in the studies by Wheatley et al. [188] where 5 units per ml ADC was found as effective as arginase in the inhibition of HeLa cells and cell cycle arrest at G1 (quiescence) in fibroblasts. Although some research groups have exhibited ADC as a potential antitumor enzyme, only a few reports are available to support this fact (Table 3.3.1) [187,188]. Even though ADC possesses low Km and can degrade arginine very rapidly, the serious problem is related to its product, that is, agmatine. Agmatine is toxic to normal cells when its concentration reaches to millimolar level, particularly when free arginine levels are low. Additionally, agmatine is not converted back to arginine under normal physiological conditions, which may lead to its accumulation and toxicity to normal cells [189]. Though recombinant human ADC expressed in E. coli has been evidenced more active than Sigma enzymes prepared from other sources, its PEGylation has been shown to result in the loss of its entire activity [187,189]. To consider the further rational use of this prospective enzyme as potential anti-cancer modality, it clearly warrants further evaluation (Table 3.3.3).

Table 3.3.3 Properties of arginine depriving enzymes. Arginine deiminase (E.C. 3.5.3.6)

Arginase (E.C.3.5.3.1)

Arginine decarboxylase (E.C.4.1.1.19)

Main products are citrulline and NH3

Main products are ornithine and urea

Main products are agmatine and CO2

At physiological pH, Mycoplasmal ADI is 300x more effective than arginase at depleting arginine

Very high alkaline pH optimum (pH 9.3) and has little enzymic activity at physiological pH

Mammalian ADC has a basic pH optimum (pH 8.23)

Tumor starvation by L-arginine deprivation

61

Table 3.3.3 Properties of arginine depriving enzymes.dcont’d Arginine deiminase (E.C. 3.5.3.6)

Arginase (E.C.3.5.3.1)

Arginine decarboxylase (E.C.4.1.1.19)

Circulatory half-life of w4h

Very short circulatory half-life (Approx. 30 min)

Not reported

Very high affinity for arginine (Km of 0.1e1 mM)

Low affinity for arginine (Km of 2e4 mM)

High affinity for arginine (Km of w 1 mM)

Most normal cells and tissues are able to take up citrulline from the circulation

Ornithine can only be reconverted back into arginine in the liver and can cause toxicity to extra-hepatic tissues by inhibiting protein synthesis

Agmatine is not converted back to arginine under normal physiological conditions, may lead to its accumulation and toxicity to normal cells

Only found in microorganisms and is strongly antigenic in mammals Tumor sensitivity to ADI is dependent on ASS expression

Human enzyme, non-immunogenic

Found in plants, microbes and human brain

The sensitivity of tumors to rhArg is independent of ASS expression

Studied only in human cervical cancer (HeLa) cell lines

Efficacious only in ASSnegative tumors

Efficacious in both ASS-negative and OTC-negative tumors

No cofactor requirement

Mn2þ is essential for catalytic activity

Pyridoxal phosphate is a cofactor

Pegylation improves catalytic activity at physiological pH

Pegylation improves catalytic activity at physiological pH

PEGylation results in the total loss of catalytic activity

Abbreviations: ADT, arginine deprivation therapy; ADI, arginine deiminase; ASS, argininosuccinate synthetase; OTC, ornithine transcarbamoylase.

62

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

3.3.17 Concluding remarks Sufficient evidence has been accumulated indicating that arginine catabolic enzymes-based approaches may be an effective way to target malignant cells. These enzymes control tumor cell proliferation as well as make them highly vulnerable to cell-cycle-specific chemotherapeutic agents. This combinatorial approach is one of the potential strategies to maximize the efficacy to obliterate the tumor cells. Extensive research of the arginine metabolic pathways led to the establishment of arginine-depriving enzymes as a potential anti-cancer strategy against arginine auxotrophic tumors. However, many of these enzymes can be co-expressed in the cells, which results in complex interactions. For example, arginine is a common substrate for arginase as well as NOS. The specific role of NO, either in inhibition or induction of cell proliferation is dependent on numerous factors like its interaction with other free radicals, cellular makeup, tumor milieu, proteins present the cellular microenvironment and also upon the chemical and biological heterogeneity of NO. NO has been known to demonstrate bipolar cellular effects and often termed as “double-edged sword.” Although, NOS remains a viable candidate for cancer treatment, the precise role of NO in the tumor microenvironment is extremely complex and conflicting. Also, the preferential utilization of arginine by arginase and/or NOS pathway is not fully understood. Thus, many of these pathways warrant further research to understand the arginine metabolism at cellular and molecular levels involving upstream and downstream pathways of the enzymes involved. It should be noted that modulation of the immunological responses is one of the major roles of arginine availability. Arginine metabolism in myeloid-derived suppressor cells via arginase and/or NOS markedly impairs the T-cell responses that would eradicate and remove tumor cells [190]. Many excellent articles are available which focus on the role of arginine in immunological aspects of the tumors [191e194]. It would suffice to say here that the ADT may have further anti-tumor effect through restoration of anti-tumor immunity. Arginine dependence of the tumor cells has been considered as the “Achilles heel” of tumor cells [195]. Inability of tumor cells to proliferate in the absence of arginine can be targeted for their selective destruction by arginine-depriving enzymes. Large numbers of enzyme-based anti-cancer therapies are currently undergoing clinical evaluation. It is encouraging that arginase and ADI already have achieved considerable success, without causing detrimental side effects and with high tolerability [51,63,141]. The knowledge acquired about the PEGylation has helped in the generation of

Tumor starvation by L-arginine deprivation

63

adducts of potential value, overcoming the serious limitations of the anti-cancer enzymes of the non-human origin. The approach of enzymemediated ADT is highly challenging, however rewarding upon success because of the provision of overturning the cancer dogma.

3.3.18 References [1] Wu G. 2013. Functional amino acids in nutrition and health. Amino Acids, Sep; 45: 407e411. DOI: https://doi.org/10.1007/s00726-013-1500-6. [2] Loreni F, Mancino M, and Biffo S. 2014. Translation factors and ribosomal proteins control tumor onset and progression: how? Oncogene, Apr; 33: 2145e2156. DOI: https://doi.org/10.1038/onc.2013.153. Epub 2013 May 6. [3] Proud CG. 2014. Control of the translational machinery by amino acids. American Journal of Clinical Nutrition, Jan; 99: 231se236s. DOI: https://doi.org/10.3945/ajcn. 113.066753. [4] Luo J-Q, Chen D-W, and Yu B. 2013. Upregulation of amino acid transporter expression induced by l-leucine availability in L6 myotubes is associated with ATF4 signaling through mTORC1-dependent mechanism. Nutrition, Jan; 29: 284e290. DOI: https://doi.org/10.1016/j.nut.2012.05.008. Epub 2012 Sep 15. [5] Palii SS, Kays CE, Deval C, Bruhat A, Fafournoux P, and Kilberg MS. 2009. Specificity of amino acid regulated gene expression: analysis of genes subjected to either complete or single amino acid deprivation. Amino Acids. May; 37: 79e88. DOI: https://doi.org/10.1007/s00726-008-0199-2. Epub 2008 Nov 14. [6] Qie S, Liang D, Yin C, Gu W, Meng M, Wang C, and Sang N. 2012. Glutamine depletion and glucose depletion trigger growth inhibition via distinctive gene expression reprogramming. Cell Cycle, Oct; 11: 3679e3690. DOI: https://doi.org/10. 4161/cc.21944. [7] Agrawal V, Alpini SEJ, Stone EM, Frenkel EP, and Frankel AE. 2012. Targeting methionine auxotrophy in cancer: discovery and exploration. Expert Opinion on Biological Therapy, Jan; 12(1): 53e61. DOI: https://doi.org/10.1517/14712598.2012.636349. [8] Wu G, Wu Z, Dai Z, Yang Y, Wang W, Liu C, Wang B, Wang J, and Yin Y. 2013. Dietary requirements of “nutritionally non-essential amino acids” by animals and humans. Amino Acids. Apr; 44: 1107e1113. DOI: https://doi.org/10.1007/s00726-012-1444-2. [9] Fu YM, Yu Z-X, Li Y-Q, Ge X, Sanchez PJ, Fu X, and Meadows GG. 2003. Specific amino acid dependency regulates invasiveness and viability of androgen-independent prostate cancer cells. Nutrition and Cancer, 45: 60e73. DOI: https://doi.org/10. 1207/S15327914 NC4501_8. [10] Icard P and Lincet H. 2012. A global view of the biochemical pathways involved in the regulation of the metabolism of cancer cells. Biochimica et Biophysica Acta Reviews on Cancer, Dec; 1826: 423e433. DOI: https://doi.org/10.1016/j.bbcan.2012.07.001. [11] Cantor JR and Sabatini DM. 2012. Cancer cell metabolism: one hallmark, many faces. Cancer Discovery, Oct; 2: 881e898. DOI: https://doi.org/10.1158/2159-8290.CD-120345. [12] Locasale JW and Cantley LC. 2011. Metabolic flux and the regulation of mammalian cell growth. Cell Metabolism, Oct 5; 14: 443e451. DOI: https://doi.org/10.1016/j. cmet.2011.07.014. [13] Ferreira LMR, Hebrant A, and Dumont JE. 2012. Metabolic reprogramming of the tumor. Oncogene, Sep 6; 31: 3999e4011. DOI: https://doi.org/10.1038/onc.2011.576. [14] Yamamoto T, Takano N, Ishiwata K, Ohmura M, Nagahata Y, Matsuura T, Kamata A, Sakamoto K, Nakanishi T, Kubo A, Hishiki T, and Suematsu M. 2014. Reduced methylation of PFKFB3 in cancer cells shunts glucose towards the pentose phosphate pathway. Nature Communications, Mar 17; 5: 3480. DOI: https://doi.org/10.1038/ ncomms4480.

64

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

[15] Matthew G and Heiden V. 2011. Targeting cancer metabolism: a therapeutic window opens. Nature Reviews Drug Discovery, Aug 31; 10: 671e684. DOI: https://doi.org/10. 1038/nrd3504. [16] Ward PS and Thompson CB. Metabolic reprogramming: a cancer hallmark even Warburg did not anticipate. Cancer Cell, Mar 20; 21: 297e308. DOI: https://doi.org/ 10.1016/j.ccr.2012.02.014. [17] Gelb T, Pshenichkin S, Rodriguez OC, Hathaway HA, Grajkowska E, DiRaddo JO, Wroblewska B, Yasuda RP, Albanese C, Wolfe BB, and Wroblewski JT. 2015. Metabotropic glutamate receptor 1 acts as a dependence receptor creating a requirement for glutamate to sustain the viability and growth of human melanomas. Oncogene, May 21; 34: 2711e2720. DOI: https://doi.org/10.1038/onc.2014.231. Epub 2014 Jul 28. [18] Cetinbas N, Daugaard M, Mullen AR, Hajee S, Rotblat B, Lopez A, Li A, DeBerardinis RJ, and Sorensen PH. 2015. Loss of the tumor suppressor Hace1 leads to ROSdependent glutamine addiction. Oncogene, Jul 23; 34: 4005e4010. DOI: https://doi. org/10.1038/onc.2014.316. Epub 2014 Oct 6. [19] Graham ML. 2003. Pegaspargase: a review of clinical studies. Advance Drug Delivery Reviews, Sep 26; 55: 1293e1302. DOI: https://doi.org/10.1016/s0169-409x(03) 00110-8. [20] Durden DL and Distasio JA. 1981. Characterization of the effects of asparaginase from Escherichia coli and a glutaminase-free asparaginase from Vibrio succinogenes on specific cell-mediated cytotoxicity. International Journal of Cancer, Jan 15; 27: 59e65. DOI: https://doi.org/10.1002/ijc.2910270110. [21] Dodd KM and Tee AR. 2012. Leucine and mTORC1: a complex relationship. American Journal of Physiology Endocrinology and Metabolism, Jun1; 302: E1329eE1342. DOI: https://doi.org/10.1152/ajpendo.00525.2011. Epub 2012 Feb 21. [22] Takano N, Sarfraz Y, Gilkes DM, Chaturvedi P, Xiang L, Suematsu M, Zagzag D, and Semenza GL. 2014. Molecular Cancer Research, Oct 12; 12: 1398e1406. DOI: https:// doi.org/10.1158/1541-7786.MCR-14-0184. [23] Wu G, Bazer FW, Davis TA, Kim SW, Li P, Rhoads JM, Satterfield MC, Smith SB, Spencer TE, and Yin Y. 2009. Arginine metabolism and nutrition in growth, health and disease. Amino Acids, May; 37: 153e168. DOI: https://doi.org/10.1007/s00726008-0210-y. [24] Appleton J. 2002. Arginine: clinical potential of a semi-essential amino acid. Alternative Medicine Review, Dec; 7: 512e522. PMID: 12495375. [25] Feun LG, Marini A, Walker G, Elgart G, Moffat F, Rodgers SE, Wu CJ, You M, Wangpaichitr M, Kuo MT, Sisson W, Jungbluth AA, Bomalaski J, and Savaraj N. 2012. Negative argininosuccinate synthetase expression in melanoma tumors may predict clinical benefit from arginine-depleting therapy with pegylated arginine deiminase. British Journal of Cancer, Apr 24; 106: 1481e1485. DOI: https://doi.org/10. 1038/bjc.2012.106. [26] Delage B, Luong P, Maharaj L, O’Riain C, Syed N, Crook T, Hatzimichael E, Papoudou-Bai A, Mitchell TJ, Whittaker SJ, Cerio R, Gribben J, Lemoine N, Bomalaski J, Li C-F, Joel S, Fitzgibbon J, Chen L-T, and Szlosarek PW. 2012. Promoter methylation of argininosuccinate synthetase-1 sensitises lymphomas to arginine deiminase treatment, autophagy and caspase-dependent apoptosis. Cell Death and Disease, Jul 5; 3: e342. DOI: https://doi.org/10.1038/cddis.2012.83. [27] Wheatley DN. 2004. Controlling cancer by restricting arginine availability-arginine catabolizing enzymes as anticancer agents. Anti-Cancer Drugs, Oct; 15: 825e833. DOI: https://doi.org/10.1097/00001813-200410000-00002. [28] García-Navas R, Munder M, and Mollinedo F. 2012. Depletion of L-arginine induces autophagy as a cytoprotective response to endoplasmic reticulum stress in human T lymphocytes. Autophagy, Nov; 8: 1557e1576. DOI: https://doi.org/10.4161/auto. 21315.

Tumor starvation by L-arginine deprivation

65

[29] Synakiewicz A, Stachowicz-Stencel T, and Adamkiewicz-Drozynska E. 2014. The role of arginine and the modified arginine deiminase enzyme ADI-PEG 20 in cancer therapy with special emphasis on phase I/II clinical trials. Expert Opinion on Investigational Drug, Nov; 23: 1517e1529. DOI: https://doi.org/10.1517/13543784.2014. 9348 08. Epub 2014 Jun 26. [30] Dillon BJ, Holtsberg FW, Ensor CM, Bomalaski JS, and Clark MA. 2002. Biochemical characterization of the arginine degrading enzymes arginase and arginine deiminase and their effect on nitric oxide production. Medical Science Monitor, Jul; 8: BR248eBR253. PMID: 12118186. [31] Qui F, Huang J, and Sui M. 2015. Targeting arginine metabolism pathway to treat arginine-dependent cancers. Cancer Letter, Aug 1; 364(1): 1e7. DOI: https://doi.org/ 10.1016/j.can let.2015.04.020. [32] Phillips MM, Sheaff MT, and Szlosarek PW. 2013. Targeting arginine-dependent cancers with arginine-degrading enzymes: Opportunities and challenges. Cancer Research and Treatment, Dec; 45: 251e262. DOI: https://doi.org/10.4143/crt.2013. 45.4.251. [33] Noh EJ, Kang SW, Shin YJ, Choi SH, Kim CG, Park IS, Wheatley DN, and Min B-H. 2004. Arginine deiminase enhances dexamethasone-induced cytotoxicity in human T-lymphoblastic leukemia CCRF-CEM cells. International Journal of Cancer, Nov 10; 112: 502e508. DOI: https://doi.org/10.1002/ijc.20435. [34] Stasyk OV, Boretsky YR, Gonchar MV, and Sibirny AA. 2015. Recombinant arginine-degrading enzymes in metabolic anticancer therapy and bioanalytics. Cell Biology International, Mar; 39: 246e252. DOI: https://doi.org/10.1002/cbin.10383. [35] Feun LG, Kuo MT, and Savaraj N. 2015. Arginine deprivation in cancer therapy. Current Opinion in Clinical Nutrition and Metabolic Care, Jan; 18: 78e82. DOI: https:// doi.org/10.1097/MCO.0000000000000122. [36] Shirai H, Blundell TL, and Mizuguchi K. 2001. A novel superfamily of enzymes that catalyse the modification of guanidine groups. Trends in Biochemical Science, Aug; 26: 465e468. DOI: https://doi.org/10.1016/s0968-0004(01)01906-5. [37] Huang H-Y, Wu H-Y, Wang Y-H, Wang J-W, Fang F-M, Tsai J-W, Li S-H, Hung H-C, Yu S-C, Lan J, Shiue Y-L, Hsing C-H, Chen L-T, and Li C-F. 2013. ASS1 as a novel tumor suppressor gene in myxofibrosarcomas: aberrant loss via epigenetic DNA methylation confers aggressive phenotypes, negative prognostic impact, and therapeutic relevance. Clinical Cancer Research, Jun 1; 19: 2861e2872. DOI: https://doi. org/10.1158/1078-0432.CCR-12-2641. [38] Savaraj N, You M, Wu C, Wangpaichitr M, Kuo MT, and Feun LG. 2010. Arginine deprivation, autophagy, apoptosis (AAA) for the treatment of melanoma. Current Molecular Medicine, Jun; 10: 405e412. DOI: https://doi.org/10.2174/ 156652410791316995. [39] Kelly MP, Jungbluth AA, Wu B-W, Bomalaski J, Old LJ, and Ritter G. 2012. Arginine deiminase PEG20 inhibits growth of small cell lung cancers lacking expression of argininosuccinate synthetase. British Journal of Cancer, Jun 17; 106: 324e332. DOI: https://doi.org/10.1038/bjc.2011.524. [40] Manca A, Sini MC, Izzo F, Ascierto P, Tatangelo F, Botti G, Gentilcore G, Capone M, Mozzillo N, Rozzo C, Cossu A, Tanda F, and Palmieri G. 2011. Induction of arginosuccinate synthetase (ASS) expression affects the antiproliferative activity of arginine deiminase (ADI) in melanoma cells. Oncology Reports, Jun; 25: 1495e1502. DOI: https://doi.org/10.3892/or.2011.1220. [41] Jungbluth AA, Tassello J, Frosina D, Hanson N, Ritter G, Wu B-W, and Old LJ. 2010. Expression pattern of argininosuccinate-synthetase (ASS) in normal and tumor tissue as a marker for susceptibility to arginine-deiminase (ADI) therapy. Laboratory Investigation, 90: 387A.

66

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

[42] Savaraj N, Wu C, Li YY, Wangpaichitr M, You M, Bomalaski J, He W, Kuo MT, and Feun LG. 2015. Targeting argininosuccinate synthetase negative melanomas using combination of arginine degrading enzyme and cisplatin. Oncotarget, Mar 20; 6(8): 6295e6309. DOI: https://doi.org/10.18632/oncotarget.3370. [43] Miyazaki K, Takaku H, Umeda M, Fujita T, Huang W, Kimura T, Yamashita J, and Horio T. 1990. Potent growth inhibition of human tumor cells in culture by arginine deiminase purified from a culture medium of a Mycoplasma-infected cell line. Cancer Research, Aug 1; 50: 4522e4527. PMID: 2164440. [44] Sugimura K, Ohno T, Kusuyama T, and Azuma I. 1992. High sensitivity of human melanoma cell lines to the growth inhibitory activity of mycoplasmal arginine deiminase in vitro. Melanoma Research, Sep; 2: 191e196. DOI: https://doi.org/10.1097/ 00008390-199209000-00007. [45] Takaku H, Misawa S, Hayashi H, and Miyazaki K. 1993. Chemical modification by polyethylene glycol of the anti-tumor enzyme arginine deiminase from Mycoplasma arginini. Japanese Journal of Cancer Research, Nov; 84: 1195e1200. DOI: https://doi. org/10.1111/j.1349-7006.1993.tb02821.x. [46] Holtsberg FW, Ensor CM, Steiner MR, Bomalaski JS, and Clark MA. 2002. Poly (ethylene glycol) (PEG) conjugated arginine deiminase: effects of PEG formulations on its pharmacological properties. Journal of Controlled Release, Apr; 80: 259e271. DOI: https://doi.org/10.1016/s0168-3659(02)00042-1. [47] Ensor CM, Holtsberg FW, Bomalaski JS, and Clark MA. 2002. Pegylated arginine deiminase (ADI-SS PEG20,000mw) inhibits human melanomas and hepatocellular carcinomas in vitro and in vivo. Cancer Research, Oct; 62: 5443e5450. PMID: 12359751. [48] Zhang L, Liu M, Jamil S, Han R, Xu G, and Ni Y. 2015. PEGylation and pharmacological characterization of a potential anti-tumor drug, an engineered arginine deiminase originated from Pseudomonas plecoglossicida. Cancer Letter, Feb; 357: 346e354. DOI: https://doi.org/10.1016/j.canlet.2014.11.042. [49] Ni Y, Schwaneberg U, and Sun ZH. 2008. Arginine deiminase, a potential antitumor drug. Cancer Letter, Mar; 261: 1e11. DOI: https://doi.org/10.1016/j.canlet. 2007.11.038. [50] Yoon J, Frankel AE, Feun LG, Ekmekcioglu S, and Kim KB. 2013. Arginine deprivation therapy for malignant melanoma. Clinical Pharmacology, 5: 11e19. DOI: https://doi.org/10.2147/CPAA.S37350. [51] Glazer ES, Piccirillo M, Albino V, Di Giacomo R, Palaia R, Mastro AA, Beneduce G, Castello G, De Rosa V, Petrillo A, Ascierto PA, Curley SA, and Izzo F. 2010. Phase II study of pegylated arginine deiminase for nonresectable and metastatic hepatocellular carcinoma. Journal of Clinical Oncology, May; 28(13): 2220e2226. DOI: https://doi.org/10.1200/JCO.2009.26.7765. [52] Ascierto PA, Scala S, Castello G, Daponte A, Simeone E, Ottaiano A, Beneduce G, De Rosa VB, Izzo F, Melucci MT, Ensor CM, Prestayko AW, Holtsberg FW, Bomalaski JS, Clark MA, Savaraj N, Feun LG, and Logan TF. 2005. Pegylated arginine deiminase treatment of patients with metastatic melanoma: results from phase I and II studies. Journal of Clinical Oncology, Oct 20; 23(30): 7660e7668. DOI: https:// doi.org/10.1200/JCO.2005.02.0933. [53] Kobayashi E, Masuda M, Nakayama R, Ichikawa H, Satow R, Shitashige M, Honda K, Yamaguchi U, Shoji A, Tochigi N, Morioka H, Toyama Y, Hirohashi S, Kawai A, and Yamada T. 2010. Reduced argininosuccinate synthetase is a predictive biomarker for the development of pulmonary metastasis in patients with osteosarcoma. Molecular Cancer Therapeutics, Mar; 9(3): 535e544. DOI: https://doi.org/10.1158/1535-7163. MCT-09-0774. [54] Syed N, Langer J, Janczar K, Singh P, Lo Nigro C, Lattanzio L, Coley HM, Hatzimichael E, Bomalaski J, Szlosarek P, Awad M, O’Neil K, Roncaroli F, and Crook T. 2013. Epigenetic status of argininosuccinate synthetase and argininosuccinate lyase modulates autophagy and cell death in glioblastoma. Cell Death and Disease, Jan 17; 4(1): e458. DOI: https://doi.org/10.1038/cddis.2012.197.

Tumor starvation by L-arginine deprivation

67

[55] Nicholson LJ, Smith PR, Hiller L, Szlosarek PW, Kimberley C, Sehouli J, Koensgen D, Mustea A, Schmid P, and Crook T. 2009. Epigenetic silencing of argininosuccinate synthetase confers resistance to platinum-induced cell death but collateral sensitivity to arginine auxotrophy in ovarian cancer. International Journal of Cancer, Sep 15; 125(6): 1454e1463. DOI: https://doi.org/10.1002/ijc.24546. [56] Szlosarek PW, Klabatsa A, Pallaska A, Sheaff M, Smith P, Crook T, Grimshaw NJ, Steele JP, Rudd RM, Balkwill FR, and Fennell DA. 2006. In vivo loss of expression of argininosuccinate synthetase in malignant pleural mesothelioma is a biomarker for susceptibility to arginine depletion. Clinical Cancer Research, Dec 1; 12(23): 7126e7131. DOI: https://doi.org/10.1158/1078-0432.CCR-06-1101. [57] Rabinovich S, Adler L, Yizhak K, Sarver A, Silberman A, Agron S, Stettner N, Sun Q, Brandis A, Helbling D, Korman S, Itzkovitz S, Dimmock D, Ulitsky I, Nagamani SCS, Ruppin E, and Erez A. 2015. Diversion of aspartate in ASS1-deficient tumours fosters de novo pyrimidine synthesis. Nature, Nov 19; 527(7578): 379e383. DOI: https://doi.org/10.1038/nature15529. [58] Feun L and Savaraj N. 2006. Pegylated arginine deiminase: a novel anticancer enzyme agent. Expert Opinion on Investigational Drugs, Jul; 15(7): 815e822. DOI: https://doi. org/10.1517/13543784.15.7.815. [59] Bobak YP, Vynnytska BO, Kurlishchuk YV, Sibirny AA, and Stasyk OV. 2010. Cancer cell sensitivity to arginine deprivation in vitro is not determined by endogenous levels of arginine metabolic enzymes. Cell Biology International, Nov; 34(11): 1085e1089. DOI: https://doi.org/10.1042/CBI20100451. [60] Tsai WB, Aiba I, Lee SY, Feun L, Savaraj N, and Kuo MT. 2009. Resistance to arginine deiminase treatment in melanoma cells is associated with induced argininosuccinate synthetase expression involving c-Myc/HIF-1a/Sp4. Molecular Cancer Therapeutics, Dec; 8(12): 3223e3233. DOI: https://doi.org/10.1158/1535-7163. MCT-09-0794. [61] Tsai WB, Aiba I, Long Y, Lin HK, Feun L, Savaraj N, and Kuo MT. 2012. Activation of Ras/PI3K/ERK pathway induces c-Myc stabilization to upregulate argininosuccinate synthetase, leading to arginine deiminase resistance in melanoma cells. Cancer Research, May 15; 72(10): 2622e2633. DOI: https://doi.org/10.1158/00085472.CAN-11-3605. [62] Long Y, Tsai W-B, Wangpaichitr M, Tsukamoto T, Savaraj N, Feun LG, and Kuo MT. 2013. Arginine deiminase resistance in melanoma cells is associated with metabolic reprogramming, glucose dependence and glutamine addiction. Molecular Cancer Therapeutics, Nov; 12(11): 2581e2590. DOI: https://doi.org/10.1158/15357163.MCT-13-0302. [63] Feun L, You M, Wu CJ, Kuo MT, Wangpaichitr M, Spector S, and Savaraj N. 2008. Arginine deprivation as a targeted therapy for cancer. Current Pharmaceutical Design, 14(11): 1049e1057. DOI: https://doi.org/10.2174/138161208784246199. [64] Szlosarek PW, Luong P, Phillips MM, Baccarini M, Ellis S, Szyszko T, Sheaff MT, and Avril N. 2013. Metabolic response to pegylated arginine deiminase in mesothelioma with promoter methylation of argininosuccinate synthetase. Journal of Clinical Oncology, Mar 1; 31(7): e111ee113. DOI: https://doi.org/10.1200/JCO.2012.42. 1784. [65] Yang TS, Lu SN, Chao Y, Sheen IS, Lin CC, Wang TE, Chen S-C, Wang J-H, Liao L-Y, Thomson JA, Wang-Peng J, Chen P-J, and Chen L-T. 2010. A randomised phase II study of pegylated arginine deiminase (ADI-PEG 20) in Asian advanced hepatocellular carcinoma patients. British Journal of Cancer, Sep 28; 103(7): 954e960. DOI: https://doi.org/10.1038/sj.bjc.6605856. [66] You M, Savaraj N, Wangpaichitr M, Wu C, Kuo TM, Varona-Santos J, Nguyen DM, and Feun L. 2010. The combination of ADI-PEG20 and TRAIL effectively

68

[67]

[68]

[69]

[70]

[71] [72] [73]

[74]

[75] [76]

[77] [78]

[79]

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

increases cell death in melanoma cell lines. Biochemical Biophysical Research Commuications, Apr 9; 394(3): 760e766. DOI: https://doi.org/10.1016/j.bbrc.2010.03.066. Feun LG, Wu G, Clark M, Bombalaski J, Holtsberg F, Wangpaijit M, Robles C, and Savaraj N. 2004. Mechanism of anti-tumor effect of arginine deiminase-polyethylene (ADI-PEG20) and the possible mechanism of resistance in melanoma. Proceeding of the American Association for Cancer Research, 45: 4565. Wangpaichitr M, Wu C, Bigford G, Theodoropoulos G, You M, Li YY, VeronaSantos J, Feun LG, Nguyen DM, and Savaraj N. 2014. Combination of arginine deprivation with TRAIL treatment as a targeted-therapy for mesothelioma. Anticancer Reearch, 34: 6991e7000. Dec; 34(12): 6991e6999. PMID: 25503125. You M, Savaraj N, Kuo MT, Wangpaichitr M, Varona-Santos J, Wu C, Nguyen DM, and Feun L. 2013. TRAIL induces autophagic protein cleavage through caspase activation in melanoma cell lines under arginine deprivation. Molecular and Cellular Biochemistry, Feb; 374(1e2): 181e190. DOI: https://doi.org/10.1007/s11010-0121518-1. Kim RH, Coates JM, Bowles TL, McNerney GP, Sutcliffe J, Jung JU, GandourEdwards RF, Chuang FYS, Bold RJ, and Kung H-J. 2009. Arginine deiminase as a novel therapy for prostate cancer induces autophagy and caspase-independent apoptosis. Cancer Research, Jan 15; 69(2): 700e708. DOI: https://doi.org/10.1158/ 0008-5472.CAN-08-3157. Kim RH, Bold RJ, and Kung HJ. 2009. ADI, autophagy and apoptosis: metabolic stress as a therapeutic option for prostate cancer. Autophagy, May; 5(4): 567e568. DOI: https://doi.org/10.4161/auto.5.4.8252. Savaraj N, Wu C, Kuo MT, You M, Wangpaichitr M, Robles C, Spector S, and Feun L. 2007. The relationship of arginine deprivation, argininosuccinate synthetase and cell death in melanoma. Drug Target Insights, 2: 119e128. PMCID: PMC3155228. Gong H, Zölzer F, Recklinghausen G, Havers W, and Schweigerer L. 2000. Arginine deiminase inhibits proliferation of human leukemia cells more potently than asparaginase by inducing cell cycle arrest and apoptosis. Leukemia, May; 14(5): 826-829. DOI: https://doi.org/10.1038/sj.leu.2401763. Bowles TL, Kim R, Galante J, Parsons CM, Virudachalam S, Kung HJ, and Bold RJ. 2008. Pancreatic cancer cell lines deficient in argininosuccinate synthetase are sensitive to arginine deprivation by arginine deiminase. International Journal of Cancer, Oct 15; 123(8): 1950e1955. DOI: https://doi.org/10.1002/ijc.23723. Surova O and Zhivotovsky B. 2013. Various modes of cell death induced by DNA damage. Oncogene, Aug 15; 32(33): 3789e3797. DOI: https://doi.org/10.1038/onc. 2012.556. Changou CA, Chen Y-R, Xing L, Yen Y, Chuang FYS, Cheng RH, Bold RJ, Ann DK, and Kung H-J. 2014. Arginine starvation-associated atypical cellular death involves mitochondrial dysfunction, nuclear DNA leakage, and chromatin autophagy. Proceedings of the National Academy of Sciences USA, Sep 30; 111(39): 14147e14152. DOI: https://doi.org/10.1073/pnas.1404171111. Kung H-J, Changou CA, Li C-F, and Ann DK. 2015. Chromatophagy: autophagy goes nuclear and captures broken chromatin during arginine-starvation. Autophagy, Feb; 11(2): 419e421. DOI: https://doi.org/10.1080/15548627.2015.1009789. Gong H, Zölzer F, Recklinghausen GV, Rössler J, Breit S, Havers W, Fotsis T, and Schweigerer L. 1999. Arginine deiminase inhibits cell proliferation by arresting cell cycle and inducing apoptosis. Biochemical and Biophysical Research Communications, Jul 22; 261(1): 10e14. DOI: https://doi.org/10.1006/bbrc.1999.1004. Lorenzo HK and Susin SA. 2013. Mitochondrial effectors in caspase-independent cell death. FEBS Letter, Jan 16; 557(1e3): 14e20. DOI: https://doi.org/10.1016/s00145793(03)01464-9.

Tumor starvation by L-arginine deprivation

69

[80] Polster BM. 2013. AIF, reactive oxygen species, and neurodegeneration: a “complex” problem. Neurochem International, Apr; 62(5): 695e702. DOI: https://doi.org/10. 1016/j.neuint.2012.12.002. [81] Norberg E, Orrenius S, and Zhivotovsky B. 2010. Mitochondrial regulation of cell death: processing of apoptosis-inducing factor (AIF). Biochemical and Biophysical Research Communications, May 21; 396(1): 95e100. DOI: https://doi.org/10.1016/j. bbrc.2010.02.163. [82] Zhu C, Wang X, Deinum J, Huang Z, Gao J, Modjtahedi N, Neagu MR, Nilsson M, Eriksson PS, Hagberg H, Luban J. Kroemer G, and Blomgren K. 2007. Cyclophilin A participates in the nuclear translocation of apoptosis-inducing factor in neurons after cerebral hypoxiaischemia. Journal of Experimental Medicine, Aug 6; 204(8): 1741e1748. DOI: https://doi.org/10.1084/jem.20070193. [83] Pradelli LA, Bénéteau M, and Ricc J-E. 2010. Mitochondrial control of caspasedependent and -independent cell death. Cellular and Molecular Life Sciences, May; 67(10): 1589e1597. DOI: https://doi.org/10.1007/s00018-010-0285-y. [84] Ulukaya E, Acilan C, and Yilmaz Y. 2011. Apoptosis: why and how does it occur in biology? Cell Biochemistry and Function, Aug; 29(6): 468e480. DOI: https://doi.org/ 10.1002/cbf.1774. [85] Shen LJ, Beloussow K, and Shen WC. 2006. Modulation of arginine metabolic pathways as the potential anti-tumor mechanism of recombinant arginine deiminase. Cancer Letters, Jan 8; 231(1): 30e35. DOI: https://doi.org/10.1016/j.canlet.2005.01. 007. [86] Mandal S, Mandal A, Johansson HE, Orjalo AV, and Park MH. 2013. Depletion of cellular polyamines, spermidine and spermine, causes a total arrest in translation and growth in mammalian cells. Proceedings of the National Academy of Sciences USA, Feb 5; 110(6): 2169e2174. DOI: https://doi.org/10.1073/pnas.1219002110. [87] Gerner EW and Meyskens FK Jr. 2004. Polyamines and cancer: old molecules, new understanding. Nature Reviews Cancer, Oct; 4(10): 781e792. DOI: https://doi.org/10. 1038/nrc1454. [88] Ohmura M, Hishiki T, Yamamoto T, Nakanishi T, Kubo A, Tsuchihashi K, Tamada M, Toue S, Kabe Y, Saya H, and Suematsu M. 2015. Impacts of CD44 knockdown in cancer cells on tumor and host metabolic systems revealed by quantitative imaging mass spectrometry. Nitric Oxide, Apr 30; 46: 102e113. DOI: https://doi.org/10. 1016/j.niox.2014.11.005. [89] Folkman J. 2002. What is the evidence that tumors are angiogenesis dependent? Journal of the National Cancer Institute, Jan 3; 82(1): 4e6. DOI: https://doi.org/10. 1093/jnci/82.1.4. [90] Jain RK. 2002. Tumor angiogenesis and accessibility: role of vascular endothelial growth factor. Seminars in Oncology, Dec; 29(6 Suppl 16): 3e9. DOI: https://doi.org/ 10.1053/sonc.2002.37265. [91] Cao Z, Shang B, Zhang G, Miele L, Sarkar FH, Wang Z, and Zhou Q. 2013. Tumor cell-mediated neovascularisation and lymphangiogenesis contrive tumor progression and cancer metastasis. Biochimica et Biophysica Acta Reviews Cancer, Dec; 1836(2): 273e286. DOI: https://doi.org/10.1016/j.bbcan.2013.08.001. [92] Stapor P, Wang X, Goveia J, Moens S, and Carmeliet P. 2014. Angiogenesis revisited e role and therapeutic potential of targeting endothelial metabolism. Journal of Cell Science, Oct 15; 127(Pt 20): 4331e4341. DOI: https://doi.org/10.1242/jcs. 153908. [93] Goel S, Duda DG, Xu L, Munn LL, Boucher Y, Fukumura D, and Jain RK. 2011. Normalization of the vasculature for treatment of cancer and other diseases. Physiological Reviews, Jul; 91(3): 1071e1121. DOI: https://doi.org/10.1152/physrev.00038.2010.

70

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

[94] Cantelmo AR, Brajic A, and Carmeliet P. 2015. Endothelial metabolism driving angiogenesis: emerging concepts and principles. Cancer Journal (Sudburry MA). JulAug; 21(4): 244e249. DOI: https://doi.org/10.1097/PPO.0000000000000133. [95] Chamorro-Jorganes A, Lee MY, Araldi E, Landskroner-Eiger S, Fernández-Fuertes M, Sahraei M, Quiles Del Rey M, van Solingen C, Jun Yu J, FernándezHernando C, Sessa WC, and Suárez Y. 2015. VEGF-induced expression of miR-17 w 92 cluster in endothelial cells is mediated by ERK/ELK1 activation and regulates angiogenesis. Circulation Research, Jan 8; 118(1): 38e47. DOI: https://doi. org/10.1161/CIRCRESAHA.115.307408. [96] Zhao D, Pan C, Sun J, Gilbert C, Drews-Elger K, Azzam DJ, Picon-Ruiz M, Kim M, Ullmer W, El-Ashry D, Creighton CJ, and Slingerland JM. 2015. VEGF drives cancerinitiating stem cells through VEGFR-2/Stat3 signaling to upregulate Myc and Sox2. Oncogene, Jun 11; 34(24): 3107e3119. DOI: https://doi.org/10.1038/onc.2014.257. [97] Kalucka J, Missiaen R, Georgiadou M, Schoors S, Lange C, De Bock K, Dewerchin M, and Carmeliet P. 2015. Metabolic control of the cell cycle. Cell Cycle, 14(21): 3379e3388. DOI: https://doi.org/10.1080/15384101.2015.1090068. [98] Moens S, Goveia J, Stapor PC, Cantelmo AR, and Carmeliet P. 2014. The multifaceted activity of VEGF in angiogenesis-Implications for therapy responses. Cytokine and Growth Factor Review, Aug; 25(4): 473e482. DOI: https://doi.org/10.1016/j. cytogfr.2014.07.009. [99] Eelen G, Zeeuw P, Simons M, and Carmeliet P. 2015. Endothelial cell metabolism in normal and diseased vasculature. Circulation Research, Mar 27; 116(7): 1231e1244. DOI: https://doi.org/10.1161/CIRCRESAHA.116.302855. [100] Zecchin A, Stapor PC, Goveia J, and Carmeliet P. 2015. Metabolic pathway compartmentalization: an underappreciated opportunity? Current Opinion in Biotechnology, Aug; 34: 73e81. DOI: https://doi.org/10.1016/j.copbio.2014.11.022. [101] Beloussow K, Wang L, Wu J, Ann D, and Shen W-C. 2002. Recombinant arginine deiminase as a potential anti-angiogenic agent. Cancer Letters, Sep 26; 183(2): 155e162. DOI: https://doi.org/10.1016/s0304-3835(01)00793-5. [102] Park I-S, Kang S-W, Shin Y-J, Chae K-Y, Park M-O, Kim M-Y, Wheatley DN, and Min B-H. 2003. Arginine deiminase: a potential inhibitor of angiogenesis and tumor growth. British Journal of Cancer, Sep 1; 89(5): 907e914. DOI: https://doi.org/10. 1038/sj.bjc6601181. [103] Thomas JB, Holtsberg FW, Ensor CM, Bomalaski JS, and Clark MA. 2002. Enzymic degradation of plasma arginine using arginine deiminase inhibits nitric oxide production and protects mice from the lethal effects of tumor necrosis factor a and endotoxin. Biochemistry Journal, May 1; 363(Pt 3): 581e587. DOI: https://doi.org/10. 1042/0264-6021:3630581. [104] Noh EJ, Kang SW, Shin YJ, Kim DC, Park I-S, Kim MY, Min Young Kim MY, Chun G, and Min BH. 2002. Characterization of Mycoplasma arginini deiminase expressed in E coli and its inhibitory regulation of nitric oxide synthesis. Molecules and Cells, Feb 28; 13(1): 1371e43. PMID: 11911465. [105] Fraisl P. 2013. Crosstalk between oxygen- and nitric oxide-dependent signaling pathways in angiogenesis. Experiental Cell Research, May 15; 319(9): 1331e1339. DOI: https://doi.org/10.1016/j.yexcr.2013.02.010. [106] Morbidelli L, Donnini S, and Ziche M. 2004. Role of nitric oxide in tumor angiogenesis. Cancer Treatment and Research, 117: 155e167. DOI: https://doi.org/10. 1007/978-1-4419-8871-3_11. [107] Jurasz P, Sawicki G, Duszyk M, Sawicka J, Miranda C, Mayers I, and Radomski MW. 2001. Matrix metalloproteinase 2 in tumor cell-induced platelet aggregation: regulation by nitric oxide. Cancer Research, Jan 1; 61(1): 376e382. PMID: 11196190.

Tumor starvation by L-arginine deprivation

71

[108] Carreau A, Kieda C, and Grillon C. 2011. Nitric oxide modulates the expression of endothelial cell adhesion molecules involved in angiogenesis and leukocyte recruitment. Experimental Cell Research, Jan 1; 317(1): 29e41. DOI: https://doi.org/10. 1016/j.yexcr.2010.08.011. [109] Fukumura D, Kashiwagi S, and Jain RK. 2006. The role of nitric oxide in tumour progression. Nature Reviews Cancer, Jul; 6(7): 521e534. DOI: https://doi.org/10. 1038/nrc1910. [110] Lee MY, Luciano AK, Ackah E, Rodriguez-Vita J, Bancroft TA, Eichmann A, Simons M, Kyriakides TR, Morales-Ruiz M, and Sessa WC. 2014. Endothelial Akt1 mediates angiogenesis by phosphorylating multiple angiogenic substrates. Proceedings of the National Academy of Sciences USA. Sep 2; 111(35): 12865e12870. DOI: https:// doi.org/10.1073/pnas.1408472111. [111] Fukumura D, Gohongi T, Kadambi A, Izumi Y, Ang J, Yun CO, Buerk DG, Huang PL, and Jain RK. 2001. Predominant role of endothelial nitric oxide synthase in vascular endothelial growth factor-induced angiogenesis and vascular permeability. Proceedings of the National Academy of Sciences USA, Feb 27; 98(5): 2604e2609. DOI: https://doi.org/10.1073/pnas.041359198. [112] Kashiwagi S, Izumi Y, Gohongi T, Demou ZN, Xu L, Huang PL, Buerk DG, Munn LL, Jain RK, and Fukumura D. 2005. NO mediates mural cell recruitment and vessel morphogenesis in murine melanomas and tissue-engineered blood vessels. Journal of Clinical Investigation, Jul; 115(7): 1816e1827. DOI: https://doi.org/10.1172/ JCI24015. [113] Kashiwagi S, Tsukada K, Xu L, Miyazaki J, Kozin SV, Tyrrell JA, Sessa WC, Gerweck LE, Jain RK, and Fukumura D. 2008. Perivascular nitric oxide gradients normalize tumor vasculature. Nature Medicine, Mar; 14(3): 255e257. DOI: https:// doi.org/10.1038/nm1730. [114] Roberts DD, Isenberg JS, Ridnour LA, and Wink DA. 2007. Nitric oxide and its gatekeeper thrombospondin-1 in tumor angiogenesis. Clinical Cancer Research, Feb 1; 13(3): 795e798. DOI: https://doi.org/10.1158/1078-0432.CCR-06-1758. [115] McAlpine JA, Lu H-T, Wu KC, Knowles SK, and Thomson JA. 2014. Downregulation of argininosuccinate synthetase is associated with cisplatin resistance in hepatocellular carcinoma cell lines: implications for PEGylated arginine deiminase combination therapy. BMC Cancer, Aug 28; 14: 621. DOI: https://doi.org/10.1186/ 1471-2407-14-621. [116] Liu J, Ma J, Wu Z, Li W, Zhang D, Han L, Wang F, Reindl KM, Wu E, and Ma Q. 2014. Arginine deiminase augments the chemosensitivity of argininosuccinate synthetase-deficient pancreatic cancer cells to gemcitabine via inhibition of NF-kB signaling. BMC Cancer, Sep 20; 14: 686. DOI: https://doi.org/10.1186/1471-240714-686. [117] Allen MD, Luong P, Hudson C, Leyton J, Delage B, Ghazaly E, Cutts R, Yuan M, Syed N, Lo Nigro C, Lattanzio L, Chmielewska-Kassassir M, Tomlinson I, Roylance R, Whitaker HC, Warren AY, Neal D, Frezza C, Beltran L, Jones LJ, Chelala C, Wu B-W, Bomalaski JS, Jackson RC, Lu Y-J, Crook T, Lemoine NR, Mather S, Foster J, Sosabowski J, Avril N, Li C-F, and Szlosarek PW. 2013. Prognostic and therapeutic impact of argininosuccinate synthetase 1 control in bladder cancer as monitored longitudinally by PET imaging. Cancer Research, Feb 1; 74(3): 896e907. DOI: https://doi.org/10.1158/0008-5472.CAN-13-1702. [118] Daylami R, Muilenburg DJ, Virudachalam S, Bold RJ. 2014. Pegylated arginine deiminase synergistically increases the cytotoxicity of gemcitabine in human

72

[119] [120] [121] [122]

[123]

[124] [125]

[126] [127] [128]

[129]

[130] [131]

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

pancreatic cancer. Journal of Experimental and Clinical Cancer Research, Dec 12; 33(1): 102. DOI: https://doi.org/10.1186/s13046-014-0102-9. Jenkinson CP, Grody WW, Cederbaum SD. 1996. Comparative properties of arginases. Comparative Biochemistry and Physiology. Part B. Biochemistry and molecular Biology, May; 114(1): 107e132. DOI: https://doi.org/10.1016/0305-0491(95)02138-8. Wu G and Morris SM Jr. 1998. Arginine metabolism: nitric oxide and beyond. Biochemistry Journal, Nov 15; 336(Pt 1): 1e17. DOI: https://doi.org/10.1042/ bj3360001. Morris SM Jr. 1992. Regulation of enzymes of urea and arginine synthesis. Annual Review of Nutrition, 12: 81e101. DOI: https://doi.org/10.1146/annurev.nu.12. 070192.000501. Gotoh T, Sonoki T, Nagasaki A, Tereda K, Takiguchi M, and Mori M. 1996. Molecular cloning of cDNA for nonhepatic mitochondrial arginase (arginase II) and comparison of its induction with nitric oxide synthase in a murine macrophage-like cell line. FEBS Letter, Oct 21; 395(2e3): 119-122. DOI: https://doi.org/10.1016/ 0014-5793(96)01015-0. Elms S, Chen F, Wang Y, Qian J, Askari B, Pandey D, Iddings J, Caldwell RB, and Fulton DJR. 2013. Insights into the arginine paradox: evidence against the importance of subcellular location of arginase and eNOS. American Journal of Physiology Heart and Circulation Physiology, Sep 1; 305(5): H651eH666. DOI: https://doi.org/10. 1152/ajpheart.00755.2012. Morris SM Jr. 2007. Arginine metabolism: boundaries of our knowledge. Journal of Nutrition, Jun; 137(6 Suppl 2): 1602Se1609S. DOI: https://doi.org/10.1093/jn/137. 6.1602S. Li H, Meininger CJ, Hawker JR Jr, Haynes TE, Kepka-Lenhart D, Mistry SK, Morris SM Jr, and Wu G. 2001. Regulatory role of arginase I and II in nitric oxide, polyamine, and proline syntheses in endothelial cells. American Journal of Physiology Endocrinology and Metabolism, Jan; 280(1): E75eE82. DOI: https://doi.org/10.1152/ ajpendo.2001.280.1.E75. Bach SJ and Lasnitzki I. 1947. Some aspects of the role of arginine and arginase in mouse carcinoma. Enzymologia, 12: 198e205. PMID: 18910560. Bach SJ and Maw GA. 1953. Creatine synthesis by tumor-bearing rats. Biochimica et Biophysica Acta, May; 11(1): 69e78. DOI: https://doi.org/10.1016/0006-3002(53) 90009-2. Koji T and Terayama H. 1984. Arginase as one of the inhibitory principles in the density-dependent as well as plasma membrane-mediated inhibition of liver cell growth in vitro. Experimental Cell Research, Dec; 155(2): 359e370. DOI: https://doi. org/10.1016/0014-4827(84)90196-4. Terayama H, Koji T, Kontani M, and Okumoto T. 1982. Arginase is an inhibitory principle in liver growth of various mammalian cells in vitro Plasma membranes arresting the growth of various mammalian cells in vitro. Biochimica et Biophysica Acta, Apr 29; 720(2): 1881e1892. DOI: https://doi.org/10.1016/0167-4889(82)90011-8. Huang M-H, Yang C-C, and Wang C-C. 1992. Inhibition of lymphocyte proliferation by liver arginase. Life Sciences, 51(22): 1725e1730. DOI: https://doi.org/10. 1016/0024-3205(92)90301-5. Cheng PNM, Lam TL, Lam WM, Tsui SM, Cheng AWM, Lo WH, and Leung Y-C. 2007. Pegylated recombinant human arginase (rharg-peg 5,000mw) inhibits the in vitro and in vivo proliferation of human hepatocellular carcinoma through arginine depletion. Cancer Research, Jan 1; 67(1): 309e317. DOI: https://doi.org/10.1158/ 0008-5472.CAN-06-1945.

Tumor starvation by L-arginine deprivation

73

[132] Wheatley DN and Campbell E. 2003. Arginine deprivation, growth inhibition and tumor cell death: 3. Deficient utilisation of citrulline by malignant cells. British Journal of Cancer, Aug 4; 89(3): 573e576. DOI: https://doi.org/10.1038/sj.bjc.6601134. [133] Tsui SM, Lam WM, Lam TL, Chong HC, So PK, Kwok SY, Arnold S, Cheng PNM, Wheatley DN, Lo W-H, and Leung Y-C. 2009. Pegylated derivatives of recombinant human arginase (rhArg1) for sustained in vivo activity in cancer therapy: preparation, characterization and analysis of their pharmacodynamics in vivo and in vitro and action upon hepatocellular carcinoma cell (HCC). Cancer Cell International, Apr 17; 9: 9. DOI: https://doi.org/10.1186/1475-2867-9-9. [134] Stone EM, Glazer ES, Chantranupong L, Cherukuri P, Breece RM, Tierney DL, Curley SA, Iverson BL, and Georgiou G. 2010. Replacing Mn2þ with Co2þ in human arginase I enhances cytotoxicity toward L-arginine auxotrophic cancer cell lines. ACS Chemical Biology, Mar 19; 5(3): 333e342. DOI: https://doi.org/10.1021/ cb900267j. [135] Glazer ES, Stone EM, Zhu C, Massey KL, Hamir AN, and Curley SA. 2011. Bioengineered human arginase I with enhanced activity and stability controls hepatocellular and pancreatic carcinoma xenografts. Translational Oncology, Jun; 4(3): 138e146. DOI: https://doi.org/10.1593/tlo.10265. [136] Tanios R, Bekdash A, Kassab E, Stone E, Georgiou G, Frankel AE, and Abi-Habib RJ. 2013. Human recombinant arginase I(Co)-PEG5000 [HuArgI(Co)-PEG5000]induced arginine depletion is selectively cytotoxic to human acute myeloid leukemia cells. Leukemia Research, Nov; 37(11): 1565e1571. DOI: https://doi.org/10. 1016/j.leukres.2013.08.007. [137] Khoury O, Ghazale N, Stone E, El-Sibai M, Frankel AE, and Abi-Habib RJ. 2015. Human recombinant arginase I (Co)-PEG5000 [HuArgI(Co)-PEG5000]-induced arginine depletion is selectively cytotoxic to human glioblastoma cells. Journal of Neurooncology, Mar; 122(1): 75e85. DOI: https://doi.org/10.1007/s11060-014-1698-5. [138] Li L, Wang Y, Chen J, Cheng B, Hu J, Zhou Y, Gao X, Gao L, Mei X, Sun M, Zhang Z, and Song H. 2013. An engineered arginase FC protein inhibits tumor growth in vitro and in vivo. Evididence-Based Complementary and Alternaternative Medicine, 2013: 2013: 423129. DOI: https://doi.org/10.1155/2013/423129. [139] Lam TL, Wong GKY, Chow HY, Chong HC, Chow TL, Kwok S-Y, Cheng PN, Wheatley DN, Lo W-H, and Leung Y-C. 2010. Recombinant human arginase inhibits the in vitro and in vivo proliferation of human melanoma by inducing cell cycle arrest and apoptosis. Pigment Cell and Melanoma Research, Apr; 24(2): 366e376. DOI: https://doi.org/10.1111/j.1755-148X.2010.00798.x. [140] Lam TL, Wong GKY, Chong HC, Cheng PNM, Choi SC, Chow TL, Kwok SY, Poon RTP, Wheatley DN, Lo W-H, and Leung YC. 2009. Recombinant human arginase inhibits proliferation of human hepatocellular carcinoma by inducing cell cycle arrest. Cancer Letters, May 8; 277(1): 91e100. DOI: https://doi.org/10.1016/j. canlet.2008.11.031. [141] Yau T, Cheng PNM, Chan P, Chan W, Chen L, Yuen J, Pang R, Fan ST, and Poon RT. 2013. A phase 1 dose-escalating study of pegylated recombinant human arginase 1 (Peg-rhArg1) in patients with advanced hepatocellular carcinoma. Investigational New Drugs, Feb; 31(1): 99e107. DOI: https://doi.org/10.1007/s10637-012-9807-9. [142] Yau CC, Chan P, Pang R, Chan W, Cheng PNM, and Poon R. 2010. A phase I study of recombinant human arginase I (rhArgI) for patients with advanced hepatocellular carcinoma. Journal of Clinical Oncology, (ASCO Annual Meeting Abstracts) May 20; 28(15_suppl): e13503-. [143] Ferraro E and Cecconi F. 2007. Autophagic and apoptotic response to stress signals in mammalian cells. Archives of Biochemistry and Biophysics, Jun 15; 462(2): 210e219. DOI: https://doi.org/10.1016/j.abb.2007.02.006.

74

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

[144] Jiang P and Mizushima N. 2014. Autophagy and human diseases. Cell Research, Jan; 24(1): 69e79. DOI: https://doi.org/10.1038/cr.2013.161. [145] Gozuacik D and Kimchi A. 2004. Autophagy as a cell death and tumor suppressor mechanism. Oncogene, Apr 12; 23(16): 2891e2906. DOI: https://doi.org/10.1038/sj. onc.1207521. [146] Macintosh RL and Ryan KM. 2013. Autophagy in tumour cell death. Seminars in Cancer Biology, Oct; 23(5): 344e351. DOI: https://doi.org/10.1016/j.semcancer. 2013.05.006. [147] Fulda S and Köge D. 2015. Cell death by autophagy: emerging molecular mechanisms and implications for cancer therapy. Oncogene, Oct 1; 34(40): 5105e5113. DOI: https://doi.org/10.1038/onc.2014.458. [148] Denton D, Nicolson S, and Kumar S. 2012. Cell death by autophagy: facts and apparent artefacts. Cell Death and Differentiation, Jan; 19(1): 87e95. DOI: https://doi. org/10.1038/cdd.2011.146. [149] Shen H-M and Codongo P. 2011. Autophagic cell death: Loch Ness monster or endangered species? Autophagy, May; 7(5): 457e465. DOI: https://doi.org/10.4161/ auto.7.5.14226. [150] Efeyan A, Zoncu R, and Sabatini DM. 2012. Amino acids and mTORC1: from lysosomes to disease. Trends in Molecular Medicine, Sep; 18(9): 524e533. DOI: https:// doi.org/10.1016/j.molmed.2012.05.007. [151] Beauchamp EM, and Platanias LC. 2013. The evolution of the TOR pathway and its role in cancer. Oncogene, Aug 22; 32(34): 3923e3932. DOI: https://doi.org/10. 1038/onc.2012.567. [152] Yan L and Lamb RF. 2012. Amino acid sensing and regulation of mTORC1. Semininars in Cell and Developmental Biology, Aug; 23(6): 621e625. DOI: https://doi. org/10.1016/j.semcdb.2012.02.001. [153] Ryter SW, Cloonan SM, and Choi AMK. 2013. Autophagy: a critical regulator of cellular metabolism and homeostasis. Molecules and Cells, Jul; 36(1): 7e16. DOI: https://doi.org/10.1007/s10059-013-0140-8. [154] Cui J, Gong Z, and Shen H-M. 2013. The role of autophagy in liver cancer: molecular mechanisms and potential therapeutic targets. Biochimica et Biophysica Acta, Aug; 1836(1): 15e26. DOI: https://doi.org/10.1016/j.bbcan.2013.02.003. [155] Hara K, Yonezawa K, Weng QP, Kozlowski MT, Belham C, and Avruch J. 1998. Amino acid sufficiency and mTOR regulate p70 S6 kinase and eIF-4E BP1 through a common effector mechanism. Journal of Biological Chemistry, Jun 5; 273(23): 14484e14494. DOI: https://doi.org/10.1074/jbc.273.23.14484. [156] Hsueh EC, Knebel SM, Lo W-H, Leung Y-C, Cheng PNM, and Hsueh CT. 2012. Deprivation of arginine by recombinant human arginase in prostate cancer cells. Journal of Hematology and Oncology, Apr 30; 5: 17. DOI: https://doi.org/10.1186/ 1756-8722-5-17. [157] Scott L, Lamb J, Smith S, and Wheatley DN. 2000. Single amino acid (arginine) deprivation: rapid and selective death of cultured transformed and malignant cells. British Journal of Cancer, Sep; 83(6): 800e810. DOI: https://doi.org/10.1054/bjoc. 2000.1353. [158] Wang Y, Nartiss Y, Steipe B, McQuibban GA, and Kim PK. 2012. ROS-induced mitochondrial depolarization initiates PARK2/PARKIN-dependent mitochondrial degradation by autophagy. Autophagy, Oct; 8(10): 1462e1476. DOI: https://doi.org/ 10.4161/auto.21211. [159] Li Z-Y, Yang Y, Ming M, and Liu B. 2011. Mitochondrial ROS. generation for regulation of autophagic pathways in cancer. Biochemical and Biophysical Research Communications, Oct 14; 414(1): 5e8. DOI: https://doi.org/10.1016/j.bbrc.2011.09. 046. [160] Chen Y, McMillan-Ward E, Kong J, Israels SJ, Gibson SB. 2008. Oxidative stress induces autophagic cell death independent of apoptosis in transformed and cancer

Tumor starvation by L-arginine deprivation

[161]

[162] [163]

[164]

[165] [166] [167] [168]

[169] [170] [171] [172]

[173]

75

cells. Cell Death and Differerentiation, Jan; 15(1): 171e182. DOI: https://doi.org/10. 1038/sj.cdd.4402233. Wang Z, Shi X, Li Y, Zeng X, Fan J, Sun Y, Xian Z, Zhang G, Wang S, Hu H, and Ju D. 2014. Involvement of autophagy in recombinant human arginase-induced cell apoptosis and growth inhibition of malignant melanoma cells. Applied Microbiology and Biotechnology, Mar; 98(6): 2485e2494. DOI: https://doi.org/10.1007/s00253-0135118-0. Wu WKK, Coffelt SB, Cho CH, Wang XJ, Lee CW, Chan FKL, Yu J, and Sung JJY. 2012. The autophagic paradox in cancer therapy. Oncogene, Feb 23; 31(8): 939e953. DOI: https://doi.org/10.1038/onc.2011.295. Rebsamen M, Pochini L, Stasyk T, de Araújo MEG, Galluccio M, Kandasamy RK, Snijder B, Fauster A, Rudashevskaya EL, Bruckner M, Scorzoni S, Filipek PA, Huber KVM, Bigenzahn JW, Heinz LX, Kraft C, Bennett KL, Indiveri C, Huber LA, and Superti-Furga G. 2015. SLC38A9 is a component of the lysosomal amino acid sensing machinery that controls mTORC1. Nature, Mar 26; 519(7544): 477e481. DOI: https://doi.org/10.1038/nature14107. Wang S, Tsun Z-Y, Wolfson RL, Shen K, Wyant GA, Plovanich ME, Yuan ED, Jones TD, Chantranupong L, Comb W, Wang T, Bar-Peled L, Zoncu R, Straub C, Kim C, Park J, Sabatini BL, and Sabatini DM. 2015. Lysosomal amino acid transporter SLC38A9 signals arginine sufficiency to mTORC1. Science, Jan 9; 347(6218): 188e194. DOI: https://doi.org/10.1126/science.1257132. Eng CH and Abraham RT. 2011. The autophagy conundrum in cancer: influence of tumorigenic metabolic reprogramming. Oncogene, Nov 24; 30(47): 4687e4696. DOI: https://doi.org/10.1038/onc.2011.220. Pérez E, Das G, Bergmann A, and Baehrecke EH. 2015. Autophagy regulates tissue overgrowth in a context-dependent manner. Oncogene, Jun; 34(26): 3369e3376. DOI: https://doi.org/10.1038/onc.2014.285. Tsujimoto Y and Shimizu S. 2005. Another way to die: autophagic programmed cell death. Cell Death and Differerentiation, Nov; 12 Suppl 2: 1528e1534. DOI: https:// doi.org/10.1038/sj.cdd.4401777. Wang Z, Shi X, Li Y, Fan J, Zeng X, Xian Z, Wang Z, Sun Y, Wang S, Song P, Zhao S, Hu H, and Ju D. 2014. Blocking autophagy enhanced cytotoxicity induced by recombinant human arginase in triple-negative breast cancer cells. Cell Death and Disease, Dec 11; 5(12): e1563. DOI: https://doi.org/10.1038/cddis.2014.503. Booth LA, Tavallai S, Hamed HA, Cruickshanks N, and Dent P. 2014. The role of cell signalling in the crosstalk between autophagy and apoptosis. Cellular Signalling, Mar; 26(3): 549e555. DOI: https://doi.org/10.1016/j.cellsig.2013.11.028. Fimia GM, Corazzari M, Antonioli M, and Piacentini M. Ambra1 at the crossroad between autophagy and cell death. Oncogene, Jul 11; 32(28): 3311e3318. DOI: https://doi.org/10.1038/onc.2012.455. Djavaheri-Mergny M, Maiuri MC, and Kroemer G. 2010. Cross talk between apoptosis and autophagy by caspase-mediated cleavage of Beclin 1. Oncogene, Mar 25; 29(12): 1717e1719. DOI: https://doi.org/10.1038/onc.2009.519. Hernandez CP, Morrow K, Lopez-Barcons LA, Zabaleta J, Sierra R, Velasco C, Cole J, and Rodriguez PC. 2010. Pegylated arginase I: a potential therapeutic approach in T-ALL. Blood, Jun 24; 115(25): 5214e5221. DOI: https://doi.org/10.1182/blood2009-12-258822. Morrow K, Hernandez CP, Raber P, Del Valle L, Wilk AM, Majumdar S, Wyczechowska D, Reiss K, and Rodriguez PC. 2013. Anti-leukemic mechanisms of

76

[174]

[175] [176]

[177] [178] [179] [180] [181]

[182]

[183] [184]

[185] [186] [187]

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

pegylated arginase I in acute lymphoblastic T-cell leukemia. Leukemia, DOI: https:// doi.org/10.1038/leu.2012.247. Chow AK, Ng L, Sing Li H, Cheng CW, Lam CS, Yau TC, Cheng PN, Fan ST, Poon RTP, and Pang RWC. 2012. Anti-tumor efficacy of a recombinant human arginase in human hepatocellular carcinoma. Current Cancer Drug Targets, Nov 1; 12(9): 1233e1243. DOI: https://doi.org/10.2174/156800912803988002. Marino T, Russo N, and Toscano M. 2013. What occurs by replacing Mn2þ with Co2þ in human arginase I: first-principles computational analysis. Inorganic Chemistry, Jan 18; 52(2): 655e659. DOI: https://doi.org/10.1021/ic301703t. Glazer ES, Kaluarachchi WD, Massey KL, Zhu C, and Curley SA. 2010. Bioengineered arginase I increases caspase-3 expression of hepatocellular and pancreatic carcinoma cells despite induction of argininosuccinate synthetase-1. Surgery, Aug; 148(2): 31031e31038. DOI: https://doi.org/10.1016/j.surg.2010.03.022. Regunathan S and Reis DJ. 2000. Characterization of arginine decarboxylase in rat brain and liver: distinction from ornithine decarboxylase. Journal of Neurochemistry, May; 74(5): 2201e2208. DOI: https://doi.org/10.1046/j.1471-4159.2000.0742201.x. Li G, Regunathan S, Barrow CJ, Eshraghi J, Cooper R, and Reis DJ. 1994. Agmatine: an endogenous clonidine-displacing substance in the brain. Science, Feb 18; 263(5149): 966e969. DOI: https://doi.org/10.1126/science.7906055. Zhu MY, Iyo A, Piletz JE, and Regunathan S. 2004. Expression of human arginine decarboxylase, the biosynthetic enzyme for agmatine. Biochimica et Biophysica Acta, Jan 22; 1670(2): 156e164. DOI: https://doi.org/10.1016/j.bbagen.2003.11.006. Molderings GJ and Haenisch B. 2012. Agmatine (decarboxylated l-arginine): physiological role and therapeutic potential. Pharmacology and Therapeutics, Mar; 133(3): 351e365. DOI: https://doi.org/10.1016/j.pharmthera.2011.12.005. Dudkowska M, Lai J, Gardini G, Stachurska A, Grzelakowska-Sztabert B, Colombatto S, and Manteuffel-Cymborowska M. 2003. Agmatine modulates the in vivo biosynthesis and interconversion of polyamines and cell proliferation. Biochimica et Biophysica Acta, Jan 20; 1619(2): 159e166. DOI: https://doi.org/10.1016/s03044165(02)00476-2. Satriano J, Matsufuji S, Murakami Y, Lortie MJ, Schwartz D, Kelly CJ, Hayashi S, and Blantz RC. 1998. Agmatine suppresses proliferation by frameshift induction of antizyme and attenuation of cellular polyamine levels. Journal of Biological Chemistry, Jun 19; 273(25): 15313e15316. DOI: https://doi.org/10.1074/jbc.273.25.15313. Choi YS and Cho YD. 1999. Effects of agmatine on polyamine metabolism and the growth of prostate tumor cells. Journal of Biochemistry and Molecular Biology, 32: 173e180. Mayeur C, Veuillet G, Michaud M, Raul F, Blottière H, and Blachier F. 2005. Effects of agmatine accumulation in human colon carcinoma cells on polyamine metabolism, DNA synthesis and the cell cycle. Biochimica et Biophysica Acta, Aug 15; 1745(1): 111e123. DOI: https://doi.org/10.1016/j.bbamcr.2004.12.004. Moinard C, Cynober L, and Bandt JPD. 2004. Polyamines: metabolism and implications in human diseases. Clinical Nutrition (Edinburg, Scotland), Apr; 24(2): 184e197. DOI: https://doi.org/10.1016/j.clnu.2004.11.001. Satriano J. 2004. Arginine pathways and the inflammatory response: interregulation of nitric oxide and polyamines. Amino Acids, Jul; 26(4): 321e329. DOI: https://doi.org/ 10.1007/s00726-004-0078-4. Philip R, Campbell E, and Wheatley DN. 2003. Arginine deprivation, growth inhibition and tumor cell death: 2. Enzymatic degradation of arginine in normal and malignant cell cultures. British Journal of Cancer, Feb 24; 88(4): 613e623. DOI: https://doi.org/10.1038/sj.bjc.6600681.

Tumor starvation by L-arginine deprivation

77

[188] Wheatley DN, Scott L, Lamb J, and Smith S. 2000. Single amino acid (arginine) restriction: growth and death of cultured HeLa and human diploid fibroblasts. Cell Physiology and Biochemistry, 10(1e2): 37e55. DOI: https://doi.org/10.1159/000016333. [189] Wheatley DN and Campbell E. 2002. Arginine catabolism, liver extracts and cancer. Patholology Oncology Research, 8(1): 18e25. DOI: https://doi.org/10.1007/bf03033696. [190] Bronte V and Zanovello P. 2005. Regulation of immune responses by L-arginine metabolism. Nature Reviews Immunology, Aug; 5(8): 641e654. DOI: https://doi.org/ 10.1038/nri1668. [191] Sikalidis AK. 2015. Amino acids and immune response: a role for cysteine, glutamine, phenylalanine, tryptophan and arginine in T-cell function and cancer? Pathology Oncology Research, Jan; 21(1): 9e17. DOI: https://doi.org/10.1007/s12253-0149860-0. [192] Raber P, Ochoa AC, and Rodriguez PC. 2012. Metabolism of L-arginine by myeloid-derived suppressor cells in cancer: Mechanisms of T-cell suppression and therapeutic perspectives. Immunological Investigations, 41(6e7): 614e634. DOI: https://doi.org/10.3109/08820139.2012.680634. [193] Peranzoni E, Marigo I, Dolcetti L, Ugel S, Sonda N, Taschin E, Mantelli B, Bronte V, and Zanovello P. 2008. Role of arginine metabolism in immunity and immunopathology. Immunobiology, 212(9e10): 795e812. DOI: https://doi.org/10.1016/j. imbio.2007.09.008. [194] Popovic PJ, Zeh HJ III, and Ochoa JB. 2007. Arginine and immunity. Journal of Nutrition, Jun; 137(6 Suppl 2): 1681Se1686S. DOI: https://doi.org/10.1093/jn/137. 6.1681S. [195] Wheatley DN, Campbell E, Lai PBS, and Cheng PNM. 2005. A rational approach to the systemic treatment of cancer involving medium-term depletion of arginine. Gene Therapy and Molecular Biology, 9: 33e40. https://www.gtmb.org/volumes/Vol9/05._ Wheatley_et_al,_33-40.pdf; accessed 4.22.20. [196] Kilberg MS, Balasubramanian M, Fu L, and Shan J. 2012. The transcription factor network associated with the amino acid response in mammalian cells. Advances in Nutrition (Bethesda, MD), May 1; 3(3): 295e306. DOI: https://doi.org/10.3945/an. 112.001891. [197] Jackson RJ, Hellen CU, and Pestova TV. 2010. The mechanism of eukaryotic translation initiation and principles of its regulation. Nature Reviews Molecular and Cell Biology, Feb; 11(2): 113e127. DOI: https://doi.org/10.1038/nrm2838. [198] Zang H, and Forman HJ. 2014. Glutathione synthesis and its role in redox signaling. Seminars in Cell and Devopmental Biology, Sep; 23(7): 722e728. DOI: https://doi.org/ 10.1016/j.semcdb.2012.03.017. [199] Weiger TM and Hermann A. 2014. Cell proliferation, potassium channels, polyamines and their interactions: a mini review. Amino Acids, Mar; 46(3): 681e688. DOI: https://doi.org/10.1007/s00726-013-1536-7. [200] Lind D. 2004. Arginine and cancer. Journal of Nutrition, Oct; 134(10 Suppl): 2837Se2841S. discussion 2853S. DOI: https://doi.org/10.1093/jn/134.10.2837S. [201] Kohler ES, Sankaranarayanan S, Van Ginneken CJ, Van Dijk P, Vermeulen JLM, Ruijter JM, Lamers WH, and Bruder E. 2008. The human neonatal small intestine has the potential for arginine synthesis; developmental changes in the expression of arginine-synthesizing and catabolizing enzymes. BMC Developmental Biology, Nov 10; 8: 107. DOI: https://doi.org/10.1186/1471-213X-8-107. [202] Kim JE, Kim SY, Lee KW, and Lee HJ. 2003. Arginine deiminase originating from Lactobacillus lactis ssp. lactis American Type Culture Collection (ATCC) 7962 induces G1 phase cellcycle arrest and apoptosis in SNU1 stomach adenocarcinoma cells. British Journal of Nutrition, Nov; 102(10): 1469e1476. DOI: https://doi.org/10.1017/ S0007114509990432.

78

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

[203] Gill P and Pan J 1970. Inhibition of cell division in L5178Y cells by argininedegrading mycoplasmas: the role of arginine deiminase. Canadian Journal of Microbiology, Jun; 16(6): 415e419. DOI: https://doi.org/10.1139/m70-071. [204] Kim JH, Kim JH, Yu YS, Kim DH, Min BH, and Kim KW. 2007. Anti-tumor activity of arginine deiminase via arginine deprivation in retinoblastoma. Oncology Reports, Dec; 18: 1373e1377. PMID: 17982619. [205] Huang CC, Tsai ST, Kuo CC, Chang JS, Jin YT, Chang JY, and Hsiao J-R. 2012. Arginine deprivation as a new treatment strategy for head and neck cancer. Oral Oncology, Dec; 48(12): 1227e1235. DOI: https://doi.org/10.1016/j.oraloncology. 2012.06.004. [206] Tan B, Yin Y, Kong X, Li P, Li X, Gao H, Li X, Huang R, and Wu G. 2010. LArginine stimulates proliferation and prevents endotoxin-induced death of intestinal cells. Amino Acids, Apr; 38(4): 1227e1235. DOI: https://doi.org/10.1007/s00726009-0334-8. [207] Wheatley DN, Philip R, and Campbell E. 2003. Arginine deprivation and tumor cell death: arginase and its inhibition. Molecular and Cellular Biochemistry, Feb; 244(1e2): 177e185. PMID: 12701828. [208] Izzo F, Marra P, Beneduce G, Castello G, Vallone P, De Rosa V, Cremona F, Ensor CM, Holtsberg FW, Bomalaski JS, Clark MA, Ng C, and Curley SA. 2004. Pegylated arginine deiminase treatment of patients with unresectable hepatocellular carcinoma: results from phase I/II studies. Journal of Clinical Oncology, May 15; 22(10): 181518e181522. DOI: https://doi.org/10.1200/JCO.2004.11.120. [209] Feun LG, You M, Wu C, Wangpaichitr M, Kuo MT, Marini A, Jungbluth A, and Savaraj N. 2010. Final results of phase II trial of pegylated arginine deiminase (ADIPEG20) in metastatic melanoma (MM) [abstract]. Journal of Clinical Oncology; 28(15_suppl): 8528e8528. https://ascopubs.org/doi/abs/10.1200/jco.2010.28.15_ suppl.8528; accessed 4.22.10. [210] Ott PA, Carvajal RD, Pandit-Taskar N, Jungbluth AA, Hoffman EW, Wu BW, Bomalaski JS, Venhaus R, Pan L, Old LJ, Pavlick AC, and Wolchok JD. 2013. Phase I/II study of pegylated arginine deiminase (ADI-PEG 20) in patients with advanced melanoma. Investigational New Drugs, Apr; 31(2): 425e434. DOI: https://doi.org/10. 1007/s10637-012-9862-2. [211] Szlosarek PW, Steele J, Sheaff M, Szyszko T, Ellis S, and Nolan L. 2013. A randomised phase II trial of pegylated arginine deiminase in patients with malignant pleural mesothelioma. 15th World Conference on Lung Cancer; Sydney, Australia. Abstr no. MO09.02. [See: https://pubmed.ncbi.nlm.nih.gov/27584578/; accessed 4.22.20]. [212] Tomlinson BK, Bomalaski JS, Diaz M, Akande T, Mahaffey N, Li T, Dutia TLP, Kelly K, Gong I-Y, Gandara DR, Pan C-X, and Lara P. 2013. Phase I trial of ADIpeg 20 plus docetaxel (DOC) in patients (pts) with advanced solid tumors. Journal of Clinical Oncology, (ASCO Meeting Abstracts) 2013; 31(15 Suppl): 2569. https:// ascopubs.org/doi/abs/10.1200/jco.2013.31.15_suppl.2569; accessed 4.22.20. [213] Tomlinson BK, Thomson JA, Bomalaski JS, Diaz M, Akande T, Mahaffey N, Li T, Dutia MP, Kelly K, Gong I-Y, Semrad T, Gandara DR, Pan C-X, and Lara PN Jr. 2015. Phase I trial of arginine deprivation therapy with ADI-PEG 20 plus Docetaxel in patients with advanced malignant solid tumors. Clinical Cancer Research, Jun 1; 21(11): 2480e2486. DOI: https://doi.org/10.1158/1078-0432.CCR-14-2610. [214] Bach SJ and Swaine D. 1965. The effect of arginase on the retardation of tumor growth. British Journal of Cancer, Jun; 19(2): 379e386. DOI: https://doi.org/10.1038/ bjc.1965.45. [215] Currie GA. 1978. Activated macrophages kill tumor cells by releasing arginase. Nature, Jun 29; 273(5665): 758e759. DOI: https://doi.org/10.1038/273758a0.

Reprint of entire text, figures and tables by permission of Oncogene.

Tumor starvation by L-arginine deprivation

79

3.4 Zou S, Wang X, Liu P, Ke C, and Xu S. 2019. Arginine metabolism and deprivation in cancer therapy. Biomedicine and Pharmacotherapy, Oct; 118: 109210. DOI: https://doi.org/10.1016/j.biopha.2019.109210. 3.4.1 Abstract Certain cancer cells with nutrient auxotrophy have a much higher nutrient demand compared with normal human cells. Arginine as a versatile amino acid has multiple biological functions in metabolic and signaling pathways. Depletion of this amino acid by arginine depletor is generally well tolerated and has become a targeted therapy for arginine auxotrophic cancers. However, the modulatory effect of arginine on cancer cells is very complicated and still controversial. Therefore, this article focuses on arginine metabolism and depletion therapy in cancer treatment to provide systemical review on this issue.

3.4.2 Introduction Arginine, a non-essential or semi-essential amino acid, plays important roles in a variety of biological functions such as cell proliferation, survival and protein synthesis. It is also a precursor associated with production of nitric oxide, polyamines, proline, creatinine and glutamate [1]. In recent years, researchers found that certain tumors were not capable to synthesize arginine independently, which seize a good opportunity to win the battle with cancer [2]. In fact, there are several amino acids with dysfunctional metabolism in cancer cells. Depletion of these amino acids, including asparagines, glutamine, serine and methionine, could be targeted therapies for auxotrophic cancers [3e7]. The best known is depletion of asparagine in treatment of acute lymphoblastic leukemia (ALL), which is a common kind of leukemia which occurs in children [7]. Latest study reveals that glutamine deprivation is of benefit to ovarian, pancreatic and breast cancer therapy [4,8,9]. Besides, methionine depletors showed the encouraging anti-cancer effects on corresponding auxotrophic cancer as well [6,10].

3.4.3 Metabolism of arginine 3.4.3.1 Endogenous synthesis of arginine Generally, arginine is a non-essential amino acid for adult human, but it would be essential under some physiological stress including burns, injury, small intestine and kidneys damaged [11,12]. The endogenous arginine is mainly biosynthesized via intestinal-renal axis in most mammals [13]. Citrulline is the precursor in arginine formation and plays a crucial role in arginine metabolism.

80

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

Figure 3.4.1 Arginine biosynthesis in mammals. Although arginine can be synthesized in the liver, there is almost no net production of arginine in this organ. The endogenous arginase, which is extremely abundant in liver, can immediately catalyze arginine to generate urea and ornithine in urea cycle. Indeed, the arginine level (0.03e0.1 mM) in liver cells is much lower than other amino acids (0.5e10 mM) [14]. Apart from arginine, the absorption of citrulline by hepatocyte is also negligible and the liver is inactive to extract arginine in urea cycle. Thus, almost 95% of gut derived arginine and citrulline bypass the liver in pigs [15]. And similar phenomena on arginine and citrulline metabolism are demonstrated in human as well [16].

3.4.3.2 Arginine degradation In adults, arginine turns over very rapidly with half-life of 1.06 hours [17]. Cationic amino acid transporters (CAT) are the major vehicles for arginine transportation in cellular activity [18]. There are several enzymes to degrade arginine including arginase, three types of nitric oxide synthase (NOS), arginine decarboxylase (ADC) and arginine: glycine amidinotransferase (AGAT) [11,13]. The products of arginine breakdown are mainly involved in ornithine, urea, nitric oxide (NO), glutamate, polyamines and proline [13]. Quantitatively, arginase is the most essential determinant in arginine degradation in mammals [1]. There are two isozymes that exist in mammals,

Tumor starvation by L-arginine deprivation

81

arginase 1 and arginase 2. Arginase 1, located in the cytoplasm, is highly expressed in liver cells and exerts key functions in urea cycle; while, arginase 2, mainly located in mitochondria, is found at relatively low expression in extrahepatic cells, such as neurons, nephrocytes, muscle cells and vascular cells [19,20]. The regulatory functions of arginase are mainly involved with the biosynthesis of NO and polyamines, as well as intracellular arginine/ citrulline level [21]. Accumulating evidences in the literature illustrated that the activity of arginase is associated with cellular proliferation and polyamine synthesis in many types of cell including macrophages, endothelial cells and aortic smooth muscle cell [22e24]. Interestingly, arginase expression is absent in enterocytes of suckling piglets [25]. This metabolic strategy guarantees the maximum supply of arginine from maternal milk to support the rapid growth of neonates. And in this setting proline oxidase was highly expressed to provide ornithine for further synthesis of polyamines, which are in great need for cell proliferation [26,27]. NOS is a vital enzyme and has a variety of functions in cellular activity. NOS1, NOS2, and NOS3 are three isoforms and play different roles in different organs [28]. NOS and arginase are competitive for their common substrate arginine. Increasing evidence has revealed that arginase could repress the transcription of inducible NOS and promote the activity of endogenous inhibitor of NOS, asymmetric dimethyl-L-arginine [29]. Wu et al. found that the intracellular arginine level (0.05e10 mM) increased the NO production in dose dependent manner [30]. At the same time, arginine could also elevate the mRNA level of NOS2 in astrocytes [31]. Thus, there is a sophisticated compensation mechanism for arginine degradation in mammals and it is necessary to intake dietary arginine to keep homeostasis for a healthy body [11]. 3.4.3.3 Regulation of arginine metabolism Many factors account for the regulation of arginine metabolism including hormones, cytokines, endotoxins and dietary components [30,32,33]. The amount of lysine uptake in diet could affect the arginine metabolism, because it competes for the same transporter to get into cells. Thus, the dietary arginine/lysine proportion should be less than 2.5, which is recommended in normal diet [13].

82

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

Findings in literature demonstrated that glucocorticoids could promote the arginine metabolism through stimulating arginase expression in enterocytes and hepatocytes [34]. In contrast, NO production was suppressed via inhibiting NOS activity and (6R)-5,6,7,8-tetrahydrobiopterin (BH4) biosynthesis [35]. And the surging glucocorticoids also induced arginase and the pyrroline-5-carboxylate (P5C) synthase, leading to arginine degradation as well as citrulline and polyamine overproduction [36,37]. Cytokines, lipopolysaccharide (LPS) and cAMP could significantly stimulate expression of arginase and ornithine decarboxylase (ODC) in a variety of cell types [1]. At the same time, endotoxins could induce NOS2 levels in nearly all cells [35]. Thus, all these substances promote arginine metabolism to produce ornithine, urea, NO and polyamines in specific ways. For instance, the NO levels were increased at least 10 folds within 24 hours in response to LPS stress [38]. 3.4.3.4 Function of arginine in health and disease As a versatile amino acid, arginine play an important role in body health. Arginine is involved in spermatogenesis, fetal development, neonatal growth, tissue injury and chronic metabolic disease [11]. Infertility is a big problem for society and individuals. It has been reported that arginine affects reproduction in mammals. Dietary supplementation with arginine could promote the sperm motility by 8% and increase the sperm number by about 20% [39]. Besides, pregnant rats with arginine free diets were at higher risk in intrauterine growth restriction and perinatal mortality compared with control [40]. The application of arginine in renal injury depends on the content of NO, since excessive NO contributs to renal injury. For example, adding 1% arginine in drinking water for 7 days enhanced the mesangial cell injury in rat model with glomerulonephritis [41]. While, most of trials demonstrated that arginine supplementation could be of great benefit to renal disease patients [42].

3.4.4 Arginine and cancer therapy A feature of alignant tumor is extremely high metabolic phenotype. Tumor cells have to expand nutritional needs (including amino acid) to meet intensive growth [43]. Arginine is a nutritionally essential amino acid and plays multiply functions in cell activities [44]. Certain tumors lose the ability

Tumor starvation by L-arginine deprivation

83

to synthesize arginine thus making them arginine auxotrophic. Therefore, arginine depletion was described as the Achilles heel in cancer treatment for arginine auxotrophic tumors [45]. Nowadays, it is held that there are mainly 5 enzymatic agents catabolizing free arginine (NOS, glycine amidinotransferase (GAT), ADC, arginase and arginine deiminase (ADI)) [2]. Considering the arginine depletion efficacy, immunogenicity, stability and potential byproducts, only modified arginase and ADI were applied to treatment for arginine auxotrophic tumors. 3.4.4.1 Enzymatic agents on arginine depletion Arginine deiminase (ADI), a metabolizing enzyme extracted from Mycoplasma [46], catalyzes arginine to its precursor citrulline. Due to the instability, strong immunogenicity and short half-time (5 hours) in the native form, ADI was conjugated with 20 kDa polyethylene glycol (ADIPEG20) in order to reduce the antigenicity and prolong half-time in serum [47]. Accumulating evidences in research reports have illustrated that ADIPEG20 showed anticancer effects in various tumors including prostate cancer, hepatocellular carcinoma, melanoma, leukemia as well as small cell lung cancer [48e51]. Although there are some opposing views [52], mainstream researchers still believe the metabolic enzymes in urea cycle account for the response to arginine depletors [53,54]. Argininosuccinate synthase (ASS1), which converted citrulline to argininosuccinate, was known as the rate-limiting enzyme in arginine generation. The expression of ASS1 was generally high in normal tissues, but showed heterogeneity in tumors (Figure 3.4.2). Tumors with deficient or low expression of ASS1 were much more sensitive to ADI-PEG20 therapy in vitro [53,55,56]. Besides, silencing of ASS1 in SW1222 cell line (ASS1 positive) could increase sensitivity to ADI-PEG20 treatment in colon carcinoma [51]. Besides the enzymatic function, ASS1 also has other biological effects such as tumor inhibition. For example, gene set enrichment analysis showed that down-regulation of ASS1 was relevant to lymphatic dissemination in esophageal adenocarcinoma [57]. Decreased ASS1 level was also markedly associated with the postoperative pulmonary metastasis and poor clinical outcomes in patients with osteosarcoma, whereas, tumor growth was inhibited by over-expression of ASS1 in preclinical trial [58]. In addition, about 50% of nasopharyngeal carcinomas cases were discovered with ASS1 deficiency, which was greatly correlated with advanced tumor status and

84

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

C

er e H bra ip l Undetected Low p o co c a rte m x C Ca p u Pa Th e r e ud s ra yro b e ate th id l l u y A d roi gla m r e d g nd n a la Bo Apl gl nd ne p e a n Ly m nd d m ar i x p h ro no w T de Sk Hea S o n s p r i e Sm let t m l e e l ooal m usc n th u le s m cl us e N as c op L u le h Br ar n g y on n ch x G a l L us lb iv l a Sa Pa d e r liv nc d e O ary rear ra g s l E s mu lan o p co d s S ha a Sm Du tomgus al ode ac li n h nt u es m t C ine R olo U ec n rin ar K i tum y dn bl e ad y d T er P Se E ro est m pid sta is Fa ina idy te llo l ve mi pl s s an icl t e Br ube U ea te En rine Vig st do c i n a m erv et ix riu Pl Ova m a So c ry e ft nt tis a su Sk e in

Medium

High

(A)

(B) Estimated protein expression log10 (ppm). -2 -1

0

1

2

3

4

5

T-cell leukemia, Jurkat Myeloid leukemia, K5G2 Lymphobiostic leukemia,CCRF-CEM Brain cancer, U251 Brain cancer, GAMG Bone cancer, U2OS Kidney, HEK293 Liver cancer, HuH-7 Liver cancer, HepG2 NSC lung cancer, NCI-H4GO Lung cancer, A549 Kidney cancer, RXF393 Colan cancer, RKO Colan cancer, Colo205 Melanoma, M14 Breast cancer, LCC2 Breast cancer, MCF7 Pancreas cancer Ovarion cancer, SKOV3 Prostate cancer, LnCap Prostate cancer, PC3 Cervical cancer, HeLa S3 Cervical cancer, HeLa

Figure 3.4.2 Estimated ASS1 protein level in human normal tissues and cancer cells. The expression of ASS1 in human normal tissue (A) and cancer cells (B). The figure was excerpted from human protein atlas (https://www.proteinatlas.org) and gene card (http://www.genecards.org) databases.

was a predictive biomarker of poor disease free survival [59]. At the same time, a similar phenomenon was observed in patients with myxofibrosarcoma [60]. Nicholson et al. illustrated that methylation of ASS1 could significantly shorten the OS and relapse-free survival in patients with ovarian cancer [61]. Consistent with these results, around 40% of bladder tumors were ASS1 deficiency, which accounted for a worse prognosis. And epigenetic silencing of ASS1 could indeed stimulate the tumor cell proliferation and migration [62]. Moreover, transfection with ASS1 hindered

Tumor starvation by L-arginine deprivation

85

the tumor angiogenesis and proliferation in myxofibrosarcoma, further indicating that ASS1 was a tumor suppressor gene [60]. Notably, the role of ASS1 in clinical outcomes might differ in various kinds of cancers. It has been illustrated that ASS1 might contribute to poor prognosis and outcomes in human gastric cancer [63,64]. Tsai et al. demonstrated that ASS1 promoted gastric cancer cell invasion by stabilizing active b-catenin, Snail, and Twist through autophagy inhibition, moreover ectopic expression of ASS1 protected gastric cancer cells from chemotherapeutical drugeinduced apoptosis through AKT and mTOR activation [63]. Huang et al. also demonstrated that high level of ASS1 predicted the unfavorable disease-free survival for patients with head and neck carcinoma [65]. The dual role of ASS1 in cancer therapy is still unclear. The expression of ASS1 in myxofibrosarcoma [60], bladder cancer [62], and osteosarcoma [58] was relatively low, while it was over-expressed in gastric cancer. Therefore the basal expression of ASS1 in tumor cells might account for opposite clinical outcomes [63]. But further in-depth studies are urgently needed to explore the underlying mechanisms. Arginase is a metabolizing enzyme hydrolyzes arginine to ornithine and urea. As expected from the metabolic pathway, arginase has two predominant impacts on cellular process. First, over-production of arginase consumes the supply of arginine which is the precursor of nitric oxide. Second, excessive byproduct ornithine results in vascular structural problem and nervous system dysfunction [66]. Unlike ADI, arginase presents in most of mammal and endogenous arginase as two isoforms, arginase 1 and arginase 2. 1 Compared with arginase 2, arginase 1 received more attentions because it was more stable and effective. Actually, these two subtypes shared approximately 60% similarity on amino acid sequence, and 100% homology in regions vital to enzymatic functions [67]. However, the functions of these two isoforms may not completely overlap. Mounting evidence from reports in recent years reveals that arginase 1 has a close relationship with M2 macrophages and has been served as M2 marker in macrophages, while arginase 2 can regulate M1 macrophages [68]. Furthermore, the increasing level of arginase 1 was observed in mouse model of ischemic stroke, without arginase 2 presence [69]. The role of arginase in cancer therapy has also been studied. There was no marked relationship between arginase and NOS in a wide range of breast cancer cell lines. But, the cells with high expression of arginase

86

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

exhibited rapid growth rate, which was explained by the accumulating of polyamines. Suppressing arginase activity by n-hydroxyarginine could lower the polyamine content and proliferation rate, leading to apoptosis [70]. Similar results were also observed in colon cancer [71]. However, the anticancer effect of arginase was not associated with human lung cancer. It was demonstrated that arginase 2 was indeed expressed in lung cancer cells (particularly in SCLC), but it did not affect the disease progression [72]. The antigenicity of arginase is negligible, but the optimal pH is 9.6 and the Km value is 10.5 mM [73] which is ineffective in cancer therapy in humans. Therefore, arginase 1 is conjugated with polyethylene glycol so as to improve its therapeutic efficiency. Luckily, pegylated arginase (especially recombinant human arginase 1, rhArg) showed powerful ability to deplete arginine level. RhArg (0.39 IU/mL) effectively consumed arginine to an undetectable level in 24 hours and accompanied with ornithine accumulation [74]. In fact, there are mainly two types of pegylated arginase used in preclinical trial. The most popular and well-studied agent is called BCT100, which is recombinant mutant human arginase 1 pegylated with 5 kDa polyethylene glycol (rhArgI-PEG-5000mw). The modification of wild-type arginase 1 sequence is focused on cysteine residues at site 45, 168 and 303, which results in an improved catabolic activity of BCT-100 for arginine (Km 6.0 mmol/L). Conjugating with PEG at cysteine residues 303 site could slow the clearance rate and prolong the half-life greatly [2]. The wild-type of arginase needs a cluster of Mn2þ ions for hydrolization of arginine by regulating the nucleophilic attack on guanidinium carbon of arginine [75]. The second type of pegylated arginase includes substitution of Mn2þ to Co2þ which could shift the optimal pH of arginase from 8.5 to 7.5, improve the catalytic activity and increase the half-life [76]. Even though the obvious improvement in efficacy, the off-target cytotoxicity of this kind of pegylated arginase in primary noncancerous cells was observed, which restricts its clinical application [2]. Since arginase and ADI have different roles in arginine metabolism (Figure 3.4.3), the strategy of using these two enzymatic agents on cancer treatment varies. ADI directly catalyzes arginine to citrulline and ammonia. Recycling accumulated citrulline to arginine requires argininosuccinate synthase (ASS1) and argininosuccinate lyase (ASL). Arginase hydrolyses arginine to ornithine, while conversion of ornithine to citrulline needs one additional enzyme, ornithine transcarbamylase (OTC), besides ASS1 and

Tumor starvation by L-arginine deprivation

87

Figure 3.4.3 Arginine depletion and urea cycle.

ASL (Figure 3.4.3). Therefore, ADI-PEG20 might be effective on tumors with ASS1 deficiency. And pegylated arginase might show anticancer effect on tumor with either ASS1 or OTC deficiency. Indeed, several studies have illustrated the great therapeutic effect of pegylated arginase on OTC negative tumors with ASS1 positive [54,77e79]. These findings indicate that ADI-PEG20 and pegylated arginase exert the same roles in arginine deprivation therapy but their applications in clinical should be based on genetic profiles of patients, especially ASS1 and OTC expression. Arginine decarboxylase (ADC) transforms arginine into agmatine and carbon dioxide (CO2) and this is a minor arginine metabolism pathway in mammals. The product agmatine has multiple roles in cellular activity, such as regulating polyamine catabolism, cell cycle distribution and neurotransmitter systems [80]. ADC is inactivated by its own product agmatine, which could be an anti-cancer agent. The cytotoxic effects of agmatine threaten tumor cells as well as adjacent normal cells. In addition, agmatine cannot be transformed to arginine in physiological conditions, which further potentiates the cytotoxic effect to normal cells [81]. Therefore, ADI-PEG and pegylated arginase are the two practical options for arginine depletion therapy.

88

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

3.4.4.2 Functions of arginine metabolites in cancer development Since different arginine depletors have different metabolism pathways, they lead to various metabolites in arginine depletion therapy. It is necessary to consider the role of different metabolites in cancer development. 3.4.4.2.1 Polyamines Polyamines, including putrescine, spermine and spermidine, are important products in arginase metabolism. Increasing evidence confirms that polyamines could promote tumor cell proliferation and metastasis [82]. The increased polyamines metabolites were correlated with ADI-PEG20 resistance in ASS1 deficient mesothelioma cells. Inhibition of polyamines biosynthesis could selectively potentiate the lethality on ASS1 negative tumors [83]. Indeed, the ornithine decarboxylase (ODC), an enzyme catalyzes ornithine to putrescine, plays key role in polyamine metabolism. Difluoromethylornithine (DFMO), the ODC inhibitor, has been demonstrated its anticancer effects on neuroblastoma and lung cancer. [84,85]. Thus, the function of polyamines in arginine deprivation therapy is essential and deserves more attentions. 3.4.4.2.2 Nitric oxide Nitric oxide (NO), a versatile signaling molecule, is involved in a series of physiological and pathological processes [86]. The effects of NO on cell proliferation are dependent on many factors including cell type, microenvironment, exposure time and concentration. For example, low concentration of NO could stimulate tumor growth, while high level of NO exhibited cytotoxic effect on chemoresistant cells through inducing DNA damage and apoptosis [87]. Nitric oxide could trigger the mitochondrial dependent apoptosis via increasing mitochondrial membrane permeability, which further leads to apoptotic biomarkers, such as apoptosis inducing factor (AIF) and cytochrome c release from mitochondria to cytoplasm and activate certain caspase events [88,90]. Moreover, it has been illustrated that NO was associated with HIF-1a and p53 in cancer treatment. HIF-1a is a transcription factor and its aberrant activation is related with cancer progression [90]. Increased concentration of NO could repress prolyl hydroxylases activity and result in HIF-1a accumulation in normoxia condition [91]. P53, acts as a well-known transcription factor and tumor suppressor, is an important regulator involved in DNA repair, energy metabolism and cell proliferation. The cancer cells with wildtype [53] were much sensitive to cytotoxicity induced by NO than variant p53 phenotypes [92]. The role of NO in

Tumor starvation by L-arginine deprivation

89

cardiovascular disease was mainly involved in anti-inflammation, antithrombosis and vasodilatory effect. And arginine intake might be beneficial to improve the outcome in heart disease [93]. However, there is no definitive conclusion on the effect of nitric oxide on angiogenic process in tumor. Some studies held that NO could increase angiogenesis via stimulating vascular endothelial growth factor (VEGF, angiogenic factor) in tumor, while other studies hold the different views [94]. Findings in the literature have demonstrated that NS1, a photoactive inhibitor of NOS, could decrease NO level accompanied with inhibition of VEGF and angiogenesis in melanoma cells [94]. Moreover, axitinib, a selective inhibitor of VEGF, suppressed the tumor growth and inducible NOS expression in melanoma [96]. On the contrary, the repression of NF-kB activity mediated by NO could sensitize tumor cells to chemotherapy. Besides, increased level of NO decreased angiogenesis and triggered cell death via inhibiting epithelialemesenchymal transition (EMT) process and PI3K/AKT signaling pathways [87,97]. At the same time, the anti-angiogenic effect of arginine has been manifested in colon cancer, and arginine supplementation decreased the tumor progression via derivative NO production [98]. 3.4.4.2.3 Agmatine Agmatine also called decarboxylated arginine, is derived from arginine by ADC (Figure 3.4.4). It has been shown that agmatine abolished ODC

Glutamate P5CS Proline

PO

Citrulline

P5CD

P5C

OCT OAT

Glutamate

Cytosol

P5CR Proline

AGAT Arginase II

Ornithine D-KG

Mitochondria

P5C

Guanidinoacetate

Arginine

Urea

H2O

Urea

H2O

Ornithine

Arginase I

ODC

Agmatine ADC CO2

NADPH

Arginine

BH4 NOS

NADP+ NO

ASL

CO2 Putrescine

AS

ASS1

Citrulline

SPDS Spermidine

Spermine SPMS

Figure 3.4.4 Systematic arginine metabolism in mammals. a-KG, a-ketoglutarate; P5CD, pyrroline-5-carboxylate dehydrogenase; P5CR, pyrroline-5-carboxylate reductase; P5CS, pyrroline-5-carboxylate synthase; SPDS, spermidine synthase; SPMS, spermine synthase.

90

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

protein expression and polyamines biosynthesis to induce caspase dependent apoptosis [99]. Regunathan et al. found that agmatine inhibited NO production mainly through depressing iNOS activity [100]. In addition, there are several pathways of agmatine on polyamines regulation. First, inhibiting ODC reduces the source of polyamines. Second, increasing the activity of polyamines degrading enzyme (polyamine-N-acetyltransferase) promotes its metabolism. Third, competing with its transport system reduces the intracellular level of polyamines [101]. The mechanism of agmatine was involved in cell cycle as well. As ODC is known as related to G1/S phase, agmatine could inhibit ODC expression or directly inactive cyclins expressions to induce cell cycle arrest [101]. However, the cytotoxic effect of agmatine to normal cell as mentioned above cannot be overlooked. Therefore, the clinical application of agmatine still needs further explorations. 3.4.4.2.4 Glutamine and proline Glutamine and proline, derived from ornithine breakdown by ornithine aminotransferase (OAT), are two derivatives in arginine metabolism and exert different functions in cancer development [102]. Although glutamine is a nonessential amino acid, it is still in highly demand for rapid tumor growth. The role of glutamine in tumor including biosynthesis of nucleotide and other amino acids, modulation of reactive oxygen species (ROS) content, upregulation of oncogenes such as MYC [103]. Proline, known as “stress substrate” under stress condition, has become a potential target for cancer treatment in recent years. Indeed, the proline and D1-pyrroline-5carboxylate (P5C) cycle (Figure 3.4.4) was be of help to provide much energy, redox homeostasis, and biosynthesis of protein and nucleotide in tumor cells [33,104]. Although the main source of glutamine and proline was derived from metabolic pathways and dietary intake rather than arginine metabolism, it is necessary to understand the role of these metabolites for therapeutic purposes.

3.4.5 Preclinical studies on arginine depletion in cancer treatment The anti-cancer effects of arginine depletors including ADI-PEG20 and pegylated arginase have been manifested in wide range of tumors [105e108]. However, the mechanism of these depletors varied in tumors.

Tumor starvation by L-arginine deprivation

91

Autophagy is an important biological event in cell growth and metabolism which is the principal catabolic response to nutritional starvation [109]. It is necessary to degrade the aged or damaged organelles to maintain normal function, but excessive autophagy is cytotoxic and can trigger apoptosis. Both ADI-PEG20 and pegylated arginase could induce autophagy in various tumors including mesothelioma, melanoma, glioblastoma, prostate cancer, breast cancer and leukemia [54,110e114]. As a cellular protective mechanism, autophagy induced by arginine depletors was acquired resistant mechanism to amino acid deprivation. Thus, the autophagy inhibitor chloroquine (CQ) or 3-methyladenine (3-MA) could enhance the anti-proliferative effects when combined with arginine depletors in vitro and in vivo [54,113,115]. Notably, Changou et al. demonstrated that ADI-PEG20 induced chromatin autophagy, which required nuclear membrane remodeling and DNA leakage [116]. It provides a novel cell death mechanism associated with arginine depletion treatment. Many studies revealed that ROS and autophagy have a close relationship in response to anticancer therapy [117,118]. Mitochondrion, an important organelle in eukaryotic organisms, serves as energy production in cellular activity. It is also one of essential intracellular ROS sources [119]. ADI-PEG20 could induce excessive ROS production and further lead to mitochondrial respiratory function damaged in ASS1 absent breast cancer cells [109]. Moreover, high ROS level impaired the mitochondrial function and DNA integrity in prostate cancer cells as well, and the ROS scavenger N-acetyl cysteine (NAC) could abolish the appearance induced by ADI-PEG [116]. Cell division process consists of interphase and mitosis, which is under strict regulation. As the uncontrollable regulation of cell division in cancer cells, cell cycle checkpoint has been served as a targeted therapy on clinic [120,121]. The arginine metabolism derivative agmatine could inhibit cyclins and ODC expression, which is associated with arrest in G1/S phase. Indeed, cell cycle arrest was observed in arginine depletors treatment including hepatocellular carcinoma, mesothelioma and melanoma [79,106,122]. Interestingly, the subtype of cell cycle arrest triggered by arginine depletors varied in different tumor cells. ADI-PEG20 caused G1 arrest in lymphatic leukemia [105,123], while pegylated arginase could induce G2/M phase arrest in melanoma A375 cells and HCC Hep3B cells, but S phase arrest in HCC HepG2 cells, even though they shared the same origin HCC [79,122]. The cytotoxic effects of arginine depletors on tumor cells finally lead to cellular apoptosis. However, apoptosis pathways in arginine depletion

92

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

depend on type of tumors. For example, ADI-PEG20 could induce caspase-dependent apoptosis in melanoma and lymphomas [124,125], while it could induce caspase-independent apoptosis in prostate cancer, glioblastoma and SCLC [51,113,126]. At the same time, pegylated arginase presented caspase-dependent apoptosis in lymphoma and NSCLC [115,127] and caspase-independent apoptosis in breast cancer and glioblastoma [54,112]. ADI-PEG20 and pegylated arginase showed different pathways on lymphoma, which indicated the mechanisms of these two agents on cancer therapy were dissimilar.

3.4.6 Clinical studies on arginine depletion in cancer treatment ADI-PEG20 has completed different clinical trials on patients with HCC, metastatic melanoma and mesothelioma with promising results [128e132]. For example, a phase II clinical trial conducted in the UK on milignant mesothelioma patients showed the mPFS in ADI-PEG20 treatment arm (3.2 months) was prolonged compared with control group (2.0 months), with the hazard ratio 0.56 (95% CI, 0.33-0.96). The OS in ADI-PEG20 arm was 15.7 months versus 12.1 months in control group, and the adverse events of grade 3/4 in treatment and control group were 25% and 17%, respectively [128]. Moreover, the lower expression of ASS1 the more benefits patients obtained from arginine depletion therapy, the hazard ratio in patients with ASS1 loss exceeding 75% was 0.25 (95% CI, 0.09-0.70) versus 0.72 (95% CI, 0.34e1.49) in patients with ASS1 deficiency of 50%e75% [133]. In detail, a phase I/II clinical study enrolled 19 patients with HCC showed 47% of patients were response to treatment, and the mOS of all patients in this trial was no less than 410 days including 4 patients over 680 days [129]. Another randomized phase II clinical trial enrolled 76 patients with metastatic HCC administered ADI-PEG20 (80 or 160 IU/m2 weekly) for 6 months. The mOS for all subjects was 474 days and the side effect mainly involved in grade 2 toxicities was acceptable [132]. However, in a randomized phase II clinical trials on advanced HCC conducted in Asia, the disease control rate was 31%, but the mOS was only 7.3 months due to most of patients experienced liver cirrhosis [131]. Despite the phase I/II clinical trials of ADI-PEG20 received encouraging results, the phase III clinical trial of ADI-PEG20 on HCC failed [134]. 635 patients were recruited in this study and randomized 2:1 into two groups

Tumor starvation by L-arginine deprivation

93

(ADI-PEG20, 18 mg/m2 VS placebo). The mOS in ADI-PEG20 arm was 7.8 months, while it was 7.4 months in control. Besides, there was no difference in mPFS in both groups (2.6 months VS 2.6 months). Post-hoc analysis reveals that some patients came off study early due to progression, which could no longer contribute to the pharmacodynamic analysis, might lead to selection bias. Although anti-drug antibodies were tested, neutralizing antibodies were not. The correlation between neutralizing antibodies and arginine levels was generally undetermined, and the possibility of current clinical efficacy despite the presence of neutralizing antibodies for ADI-PEG 20 has been noted [131,135]. Potential strategies to improve ADI-PEG20 efficacy induced arginine deprivation include: (1) an increased dose of ADI-PEG20 (36 mg/m2) [128], (2) combination with cytotoxic agents which may blunt the immune response to ADI-PEG20, and (3) developing a novel ADI that cannot be so quickly neutralized by antibodies. Therefore, further studies are needed to optimize the strategy for ADI-PEG20 treatment on patients with HCC. In a phase I/II clinical study conducted in Italy, 24 patients with melanoma were enrolled and given ADI-PEG20 160 U/m2 weekly. Six patients were responsive to treatment with one complete response, and no grade 3/4 toxicity events was reported [130]. However, another clinical study on patients with melanoma did not receive positive clinical response, only 31% (9/29) of patients achieved stable disease [107]. ADI-PEG20 was well tolerated with low toxicity in almost all studies. However further studies of ADI-PEG20 on wide range of ASS1 deficient tumors are warranted. Compared to ADI-PEG20, the clinical trials of pegylated arginase on cancer patients were very limited. The clinical study of pegylated arginase on advanced HCC has been completed. In this study, there were totally 15 patients recruited and were administered drug at a dosage of 1600 IU/kg per week. About 13% (2/15) of patients achieved stable disease and the mOS and mPFS was 5.2 months and 1.7 months respectively. The PFS in patients received adequate arginine depletion (serum arginine concentration < 8 mM) over 2 months was 6.4 months, while it was only 1.7 months in patients with adequate arginine depletion less than 2 months. Most of adverse events were involved in grade 1/2 such as neutropenia, fatigue, nausea and liver damage [136]. Further basic and clinical studies were required to provide more scientific ground and information for arginine depletion therapy.

94

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

ASS1 and OTC are two essential enzymes in arginine regeneration and also considered as biomarkers for arginine deprivation therapy [137,138]. There was a trend toward improved overall survival in patients with ASS1 negative HCC compared to those with ASS1 high expression [131]. For patients with advanced melanoma, the clinical benefit rate was associated with deficiency of ASS1. Only 10% (1/10) patients with ASS1 positive tumors had stable disease, while approximately 24% (4/17) patients achieved partial response and 29% (5/17) had stable disease [135]. Besides, the relation between negative basal ASS1 expression and longer PFS was observed. The mPFS for ASS1 negative patients was 3.6 months compared with 1.8 months for ASS positive patients. The mOS was also prolonged in negative patients (14.6 VS 9.3 months), although it was not statistically significant [135]. The potential of employing ASS1 expression as a predictive biomarker for arginine deprivation therapy should be further investigated in a larger patient population study. Some clinical trials of arginine depletors (ADIPEG20 and rhArg) on cancer therapy were summaried as followed (Table 3.4.1). Table 3.4.1 Clinical studies on arginine depletion in cancer treatment. Disease

Drug

Mesothelioma

ADIPEG20

36.8 mg/m

HCC

ADIPEG20 ADIPEG20 ADIPEG20

160 U/m2

HCC

ADIPEG20

18 mg/m2

Melanoma

ADIPEG20 ADIPEG20 rhArg

40,80,160 IU/m2 18 mg/m2

HCC HCC

AML HCC

mPFS (m)

Dosage

80, 160 IU/m2 160, 320 IU/m2

1600 U/kg

2

OS (m)

No. of Sub.

Ref.

3.2 VS 2.0 NR

15.7 VS 12.1

68

[128]

13.67

19

[129]

NR

16.1 VS 15.4 6.2 VS 8.4

76

[132]

71

[131]

7.8 VS 7.4

635

[139]

NR

31

[107]

1.8

3.5

43

[140]

1.7

5.2

20

[136]

1.9 VS 1.8 2.6 VS 2.6 NR

AML, acute myeloid leukemia; HCC, hepatocellular carcinoma; mPFS, median progression-free survival; OS, overall survival; NR, not reported.

Tumor starvation by L-arginine deprivation

95

3.4.7 Drug resistance in arginine deprivation therapeutics Multi-drug resistance (MDR) is one of the biggest obstacles in cancer therapeutics. It is defined as cancer cells acquired resistance to one drug leading to resistance to other agents which might have different structures or mechanisms [141]. The anticancer efficiency of chemicals would be dramatically decreased to MDR phenotype. Therefore, it is urgent to illuminate the role of MDR in cancer therapeutics including arginine deprivation treatment. The potential mechanisms of drug resistance to arginine depletors have been studied. In general, the current known mechanisms are mainly involved in ASS1 re-expression, enhanced glycolysis and antibody production. As the basal expression of ASS1 in malignant melanoma was undetectable, melanoma cells usually showed responsive to ADI-PEG20 treatment. However, one third (7/21) ADI-PEG20 sensitive cell lines displayed significantly elevated protein expressions of ASS1 after treatment, and they became much resistant to ADI-PEG20 treatment with high IC50 value compared with parental cell lines [142]. As reported by Tsai et al. ADIPEG20 could activate the Ras/PI3K/ERK pathway which stabilizes transcription factor c-Myc. Accumulated c-Myc could induce ASS1 expression through interacting with ASS1 promoter, which accounted for arginine depletion resistance in melanoma [143,144]. Moreover, the demethylating agent 5-Aza-dC could increase the level of ASS1 and abate the ADIPEG20 therapeutics activity in lymphoma cells [124]. The metabolic reprogramming was also involved in drug resistance to arginine depletion. Long et al. found that ADI-PEG20 resistant cells derived from melanoma exhibited high level of lactate dehydrogenase A and glucose transporter 1, meanwhile presented low level of pyruvate dehydrogenase, indicating the activation of glycolytic pathway [145]. Although ADI was modified by conjugating with polyethylene glycol to reduce the immunogenicity, the anti-ADI antibody was also observed in patients recruited in clinical trials. In a randomized phase II clinical study on HCC patients, the anti-ADI antibody was detected and reached a plateau at 70-day after initiation of treatment. At the same time, the content of arginine rebounded to baseline level [132]. Similarly, in a phase I/II study of ADI-PEG20 on advanced melanoma, about 32.3% (10/31) patients developed antibody against ADI-PEG20 since therapy initiation. The antibody titer was increased with ADI-PEG20 exposure time [107]. The mechanisms of multi-drug resistance on arginine depletion mainly focused on ADI-PEG20 with melanoma. Very limited data could be referred to

96

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

discuss the resistance on pegylated arginase treatment. The latest finding reveals that endogenous arginase 2 might be a potential biomarker for pegylated arginase therapy in squamous-cell lung carcinoma [146].

3.4.8 Conclusions It is increasing clear that arginine has mounts biological functions in cellular activity including tumor pathogenesis. Arginine deprivation therapy opens up a new realm of precision and targeted cancer treatment for certain cancer patients. More and more biomarkers for sensitivity and resistance to arginine deprivation will be indentified to provide scientific ground for future clinical development.

3.4.9 References [1] Morris SM Jr. 2007. Arginine metabolism: boundaries of our knowledge. The Journal of Nutrition, Jun; 137(6): 1602Se1609S. DOI: https://doi.org/10.1093/jn/137.6.1602S. [2] Fultang L, Vardon A, De Santo C, and Mussai F. 2016. Molecular basis and current strategies of therapeutic arginine depletion for cancer. International Journal of Cancer, Aug; 139(3): 501e509. DOI: https://doi.org/10.1002/ijc.30051. [3] Geck RC and Toker A. 2016. Nonessential amino acid metabolism in breast cancer. Advances in Biological Regulation, Sep; 62(2016): 11e17. DOI: https://doi.org/10.1016/ j.jbior.2016.01.001. [4] Furusawa A, Miyamoto M, Takano M, Tsuda H, Song YS, Aoki D, Miyasaka N, Inazawa J, and Inoue J. 2018, Ovarian cancer therapeutic potential of glutamine depletion based on GS expression. Carcinogenesis, Jun; 39(6): 758e766. DOI: https:// doi.org/10.1093/carcin/bgy033. [5] Maddocks OD, Berkers CR, Mason SM, Zheng L, Blyth K, Gottlieb E, and Vousden KH. 2013. Serine starvation induces stress and p53-dependent metabolic remodelling in cancer cells. Nature, 493(7433): 542e546. DOI: https://doi.org/10.1038/nature11743. [6] Strekalova E, Malin D, Good DM, and Cryns VL. 2015. Methionine deprivation induces a targetable vulnerability in triple-negative breast Cancer cells by enhancing TRAIL Receptor-2 expression. Clinical Cancer Research, Jun 15; 21(12): 2780e2791. DOI: https://doi.org/10.1158/1078-0432.CCR-14-2792. [7] Asselin BL. 1999. The three asparaginases. Comparative pharmacology and optimal use in childhood leukemia. Advances in Experiental Medicine and Biology, 457: 621e629. PMID: 10500842. [8] Gwangwa MV, Joubert AM, and Visagie MH. 2019. Effects of glutamine deprivation on oxidative stress and cell survival in breast cell lines. Biologial Research, Mar; 52(1): 15. DOI: https://doi.org/10.1186/s40659-019-0224-9. [9] Nishi K, Suzuki M, Yamamoto N, Matsumoto A, Iwase Y, Yamasaki K, Otagiri M, and Yumita N. 2018. Glutamine deprivation enhances acetyl-CoA carboxylase inhibitorinduced death of human pancreatic cancer cells. Anticancer Research, Dec; 38(12): 6683e6689. DOI: https://doi.org/10.21873/anticanres.13036. [10] Agrawal V, Alpini SE, Stone EM, Frenkel EP, and Frankel AE. 2012. Targeting methionine auxotrophy in cancer: discovery & exploration. Expert Opinion on Biological Therapy, Jan; 12(1): 53e61. DOI: https://doi.org/10.1517/14712598.2012.636349.

Tumor starvation by L-arginine deprivation

97

[11] Wu G, Bazer FW, Davis TA, Kim SW, Li P, Marc Rhoads MJ, Satterfield MC, Smith SB, Spencer TE, and Yin Y. 2009. Arginine metabolism and nutrition in growth, health and disease. Amino Acids, 37(1): 153e168. DOI: https://doi.org/10.1007/ s00726-008-0210-y. [12] Jahani M, Noroznezhad F, and Mansouri K. 2018. Arginine: challenges and opportunities of this two-faced molecule in cancer therapy. Biomedicine and Pharmacotherapy, June; 102: 594e601. DOI: https://doi.org/10.1016/j.biopha.2018.02.109. [13] Wu G and Morris SM Jr. 1998. Arginine metabolism: nitric oxide and beyond. Biochemical Journal, Nov; 336(1): 1e17. DOI: https://doi.org/10.1042/bj3360001. [14] Li P, Kim SW, Li X, Datta S, Pond WG, and Wu G. 2009. Dietary supplementation with cholesterol and docosahexaenoic acid affects concentrations of amino acids in tissues of young pigs. Amino Acids, Oct 30; 37(4): 709e716. DOI: https://doi.org/10. 1007/s00726-008-0196-5. [15] Mateo RD, Wu G, Moon HK, Carroll JA, and Kim SW. 2008. Effects of dietary arginine supplementation during gestation and lactation on the performance of lactating primiparous sows and nursing piglets. Amino Acids, Oct; 86(4): 827e835. DOI: https://doi.org/10.1007/s00726-008-0196-5. [16] van de Poll MC, Siroen MP, van Leeuwen PA, Soeters PB, Melis GC, Boelens PG, Deutz NE, and Dejong CH. 2007. Interorgan amino acid exchange in humans: consequences for arginine and citrulline metabolism. American Journal of Clinical Nutrition, Jan; 85(1): 167e172. PMID: 17209193. [17] Wu G, Bazer FW, Cudd TA, Jobgen WS, Kim SW, Lassala A, Li P, Matis JH, Meininger CJ, and Spencer TE. 2007. Pharmacokinetics and safety of arginine supplementation in animals. Journal of Nutrition, Jun; 137(6 Suppl. 2): 1673Se1680S. DOI: https://doi.org/10.1093/jn/137.6.1673S. [18] Hatzoglou M, Fernandez J, Yaman I, and Closs E. 2004. Regulation of cationic amino acid transport: the story of the CAT-1 transporter. Annual Review of Nutrition, 24: 377e399. DOI: https://doi.org/10.1146/annurev.nutr.23.011702.073120. [19] Li H, Meininger CJ, Hawker JR Jr, Haynes TE, Kepka-Lenhar D, Mistry SK, Morris SM Jr, and Wu G. 2001. Regulatory role of arginase I and II in nitric oxide, polyamine, and proline syntheses in endothelial cells. American Journal of Physiology Endocrinology and Metabolism, 280(1): E75eE82. PMID: 11120661. [20] Orlando GF, Wolf G, and Engelmann M. 2008. Role of neuronal nitric oxide synthase in the regulation of the neuroendocrine stress response in rodents: insights from mutant mice. Amino Acids, Jun; 35(1): 17e27. DOI: https://doi.org/10.1007/s00726-0070630-0. [21] Morris SM Jr. 2002. Regulation of enzymes of the urea cycle and arginine metabolism. Annual Review of Nutrition, 22: 87e105. DOI: https://doi.org/10.1146/annurev.nutr. 22.110801.140547. [22] Kepka-Lenhart D, Mistry SK, Wu G, and Morris SM Jr. 2000. Arginase I: a limiting factor for nitric oxide and polyamine synthesis by activated macrophages? American Journal of Physiology - Regulatory Integrative and Comparative Physiology, 279(6): R2237eR2242. PubMed ID: 11080091. [23] Wei LH, Wu G, Morris SM Jr, and Ignarro LJ. 2001. Elevated arginase I expression in rat aortic smooth muscle cells increases cell proliferation. Proceedings of the National Academy of Sciences of the USA, Jul; 98(16): 9260e9264. DOI: https://doi.org/10.1073/ pnas.161294898. [24] Li H, Meininger CJ, Kelly KA, Hawker JR Jr, Morris SM Jr, and Wu G. 2002. Activities of arginase I and II are limiting for endothelial cell proliferation, American Journal of Physiology. Regulatory, Integrative and Comparative Physiology, 282(1): R64eR69. DOI: https://doi.org/10.1152/ajpregu.2002.282.1.R64. [25] Wu G, Knabe DA, Flynn NE, Yan W, and Flynn SP. 1996. Arginine degradation in developing porcine enterocytes. American Journal of Physiology, Nov; 271(5 Pt 1): G913eG919. PMID: 8944707.

98

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

[26] Wu G. 1997. Synthesis of citrulline and arginine from proline in enterocytes of postnatal pigs. American Journal of Physiology, Jun 1; 272(6 Pt 1): G1382eG1390. DOI: https://doi.org/10.1152/ajpgi.1997.272.6.G1382. [27] Wu G, Bazer FW, Datta, Johnson GA, Li P, Satterfield MC, and Spencer TE. 2008. Proline metabolism in the conceptus: implications for fetal growth and development. Amino Acids, Nov; 35(4): 691e702. DOI: https://doi.org/10.1007/s00726-008-0052-7. [28] Alderton WK, Cooper CE, and Knowles RG. 2001. Nitric oxide synthases: structure, function and inhibition. Biochemistry Journal, Aug; 357(Pt 3): 593e615. DOI: https:// doi.org/10.1042/0264-6021:3570593. [29] Durante W, Johnson FK, and Johnson RA. 2007. Arginase: a critical regulator of nitric oxide synthesis and vascular function. Clinical and Experimental Pharmacology and Physiology, Apr; 34(9): 906e911. DOI: https://doi.org/10.1111/j.1440-1681.2007.04638.x. [30] Wu G and Meininger CJ. 2002. Regulation of nitric oxide synthesis by dietary factors. Annual Review of Nutrition, Jul; 22: 61e86. DOI: https://doi.org/10.1146/ann urev. nutr.22.110901.145329. [31] Lee J, Ryu H, Ferrante RJ, Morris SM Jr, and Ratan RR. 2003. Translational control of inducible nitric oxide synthase expression by arginine can explain the arginine paradox. Proceedings of the National Academy of Sciences USA, Apr; 100(8): 4843e4848. DOI: https://doi.org/10.1073/pnas.0735876100. [32] Wei LH, Yang Y, Wu G, and Ignarro LJ. 2006. IL-4 and IL-13 upregulate ornithine decarboxylase expression by PI3K and MAP kinase pathways in vascular smooth muscle cells, American Journal of Physiology Cell Physiology, May; 294(5): C1198eC1205. DOI: https://doi.org/10.1152/ajpcell.00325.2007. [33] Phang JM, Donald SP, Pandhare J, and Liu Y. 2008. The metabolism of proline, a stress substrate, modulates carcinogenic pathways. Amino Acids, Nov; 35(4): 681e690. DOI: https://doi.org/10.1007/s00726-008-0063-4. [34] Flynn NE, Bird JG, and Guthrie AS. 2009. Glucocorticoid regulation of amino acid and polyamine metabolism in the small intestine. Amino Acids, May; 37(1): 123e129. DOI: https://doi.org/10.1007/s00726-008-0206-7. [35] Shi W, Meininger CJ, Haynes TE, Hatakeyama K, and Wu G. 2004. Regulation of tetrahydrobiopterin synthesis and bioavailability in endothelial cells. Cell Biochemistry and Biophysics, Nov; 41(3): 415e434. DOI: https://doi.org/10.1385/CBB: 41:3:415. [36] Flynn NE, Meininger CJ, Kelly K, Ing NH, Morris M Jr, and Wu G. 1999. Glucocorticoids mediate the enhanced expression of intestinal type II arginase and argininosuccinate lyase in postweaning pigs. Journal of Nutrition, Apr; 129(4): 799e803. DOI: https://doi.org/10.1093/jn/129.4.799. [37] Wu G, Meininger CJ, Kelly K, Watford M, and Morris SM Jr. 2000. A cortisol surge mediates the enhanced expression of pig intestinal pyrroline-5-carboxylate synthase during weaning. Journal of Nutrition, Aug; 130(8): 1914e1919. DOI: https://doi.org/ 10.1093/jn/130.8.1914. [38] Wu G, Flynn NE, Flynn SP, Jolly CA, and Davis PK. 1999. Dietary protein or arginine deficiency impairs constitutive and inducible nitric oxide synthesis by young rats. Journal of Nutrition, 129(7): 1347e1354. PMID: 10395597. [39] Flachowsky G. 2007. Animal Production and Animal Science Worldwide, WAAP Book of the Year 2006, Rosati A, Tewolde A, and Mosconi C (Eds.), Wageningen Academic Publishers (2007), cloth binding, 370 pp., Price:V 99/US $133.00, ISBN: 978-90-8686-034-0. (2008): 346e347. [40] Greenberg SS, Lancaster JR, Xie J, Sarphie TG, Zhao X, Hua L, Freeman T, Kapusta DR, Giles TD, and Powers DR. 1997. Effects of NO synthase inhibitors, argininedeficient diet, and amiloride in pregnant rats. American Journal of Physiology, Sep 1; 273(3 Pt 2): R1031eR1045. DOI: https://doi.org/10.1152/ajpregu.1997.273.3. R1031.

Tumor starvation by L-arginine deprivation

99

[41] Peters H, Border WA, and Noble NA. 1999. L-Arginine supplementation increases mesangial cell injury and subsequent tissue fibrosis in experimental glomerulonephritis. Kidney International, Jun; 55(6): 2264e2273. DOI: https://doi.org/10.1046/j.15231755.1999.00462.x. [42] Bellinghieri G, Santoro D, Mallamace A, Di Giorgio RM, De Luca G, and Savica V. 2006. L-arginine: a new opportunity in the management of clinical derangements in dialysis patients. Journal of Renal Nutrition, Jul; 16(3): 245e247. DOI: https://doi.org/ 10.1053/j.jrn.2006.04.004. [43] Sullivan LB, Gui DY, and Heiden MGV. 2016. Altered metabolite levels in cancer: implications for tumour biology and cancer therapy. Nature Reviews Cancer, Oct; 16(11): 680e693. DOI: https://doi.org/10.1038/nrc.2016.85. [44] Stasyk OV, Boretsky YR, Gonchar MV, and Sibirny AA. 2015. Recombinant arginine-degrading enzymes in metabolic anticancer therapy and bioanalytics. Cell Biology International, 39(3): 246e252. DOI: https://doi.org/10.1002/cbin.10383. [45] Feun L, You M, Wu CJ, Kuo MT, Wangpaichitr M, Spector S, and Savaraj N. 2006. Arginine deprivation as a targeted therapy for cancer. Current Pharmaceutical Design, Apr; 14(11): 1049e1057. DOI: https://doi.org/10.2174/138161208784246199. [46] Takaku H, Matsumoto M, Misawa S, and Miyazaki K. 1995. Anti-tumor activity of arginine deiminase from Mycoplasma argini and its growth-inhibitory mechanism. Japanese Journal of Cancer Research, Sep; 86(9): 840e846. DOI: https://doi.org/10.1111/ j.1349-7006.1995.tb03094.x. [47] Holtsberg FW, Ensor CM, Steiner MR, Bomalaski JS, and Clark MA. 2002. Poly(ethylene glycol) (PEG) conjugated arginine deiminase: effects of PEG formulations on its pharmacological properties. Journal of Controlled Release, Apr; 80(1e3): 259e271. DOI: https://doi.org/10.1016/S0168-3659(02)00042-1. [48] Kim RH, Bold RJ, and Kung HJ. 2009. ADI, autophagy and apoptosis: metabolic stress as a therapeutic option for prostate cancer. Autophagy, May; 5(4): 567e568. DOI: https://doi.org/10.4161/auto.5.4.8252. [49] Ensor CM, Holtsberg FW, Bomalaski JS, and Clark MA. 2002. Pegylated arginine deiminase (ADI-SS PEG20,000 mw) inhibits human melanomas and hepatocellular carcinomas in vitro and in vivo. Cancer Research, Oct; 62(19): 5443e5450. PMID: 12359751. [50] Miraki-Moud F, Ghazaly E, Ariza-McNaughton L, Hodby KA, Clear A, AnjosAfonso F, Liapis K, Grantham M, Sohrabi FG, Cavenagh J, Bomalaski JS, Gribben JG, Szlosarek PW, Bonnet D, and Taussig DC. 2015. Arginine deprivation using pegylated arginine deiminase has activity against primary acute myeloid leukemia cells in vivo. Blood, June; 125(26): 4060e4068. DOI: https://doi.org/10.1182/blood2014-10-608133. [51] Kelly MP, Jungbluth AA, Wu BW, Bomalaski J, Old LJ, and Ritter G. 2012. Arginine deiminase PEG20 inhibits growth of small cell lung cancers lacking expression of argininosuccinate synthetase. British Journal of Cancer, Jan; 106(2): 324e332. DOI: https://doi.org/10.1038/bjc.2011.524. [52] Bobak YP, Vynnytska BO, Kurlishchuk YV, Sibirny AA, Stasyk OV. 2010. Cancer cell sensitivity to arginine deprivation in vitro is not determined by endogenous levels of arginine metabolic enzymes. Cell Biology International, Nov; 34(11): 1085e1089. DOI: https://doi.org/10.1042/CBI20100451. [53] Delage B, Fennell DA, Nicholson L, McNeish I, Lemoine NR, Crook T, and Szlosarek PW. 2010. Arginine deprivation and argininosuccinate synthetase expression in the treatment of cancer. International Journal of Cancer, Jun; 126(12): 2762e2772. DOI: https://doi.org/10.1002/ijc.25202.

100

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

[54] Wang Z, Shi X, Li Y, Fan J, Zeng X, Xian Z, Wang Z, Sun Y, Wang S, Song P, Zhao S, Hu H, and Ju D. 2014. Blocking autophagy enhanced cytotoxicity induced by recombinant human arginase in triple-negative breast cancer cells. Cell Death and Disease, Dec; 5: e1563. DOI: https://doi.org/10.1038/cddis.2014.503. [55] Liu J, Ma J, Wu Z, Li W, Zhang D, Han L, Wang F, Reindl KM, Wu E, and Ma Q. 2014. Arginine deiminase augments the chemosensitivity of argininosuccinate synthetase-deficient pancreatic cancer cells to gemcitabine via inhibition of NF-kappaB signaling. BMC Cancer, Sep; 14: Article 686. DOI: https://doi.org/10.1186/14712407-14-686. [56] Bowles TL, Kim R, Galante J, Parsons CM, Virudachalam S, Kung HJ, and Bold RJ. 2008. Pancreatic cancer cell lines deficient in argininosuccinate synthetase are sensitive to arginine deprivation by arginine deiminase. International Journal of Cancer, Oct; 123(8): 1950e1955. DOI: https://doi.org/10.1002/ijc.23723. [57] Lagarde SM, Ver Loren van Themaat PE, Moerland PD, Gilhuijs-Pederson LA, Ten Kate FJ, Reitsma PH, van Kampen AK, Zwinderman AH, Baas F, and van Lanschot JJ. 2008. Analysis of gene expression identifies differentially expressed genes and pathways associated with lymphatic dissemination in patients with adenocarcinoma of the esophagus. Annals of Surgical Oncology, Dec; 15(12): 3459e3470. DOI: https://doi.org/ 10.1245/s10434-008-0165-y. [58] Kobayashi E, Masuda M, Nakayama R, Ichikawa H, Satow R, Shitashige M, Honda K, Yamaguchi U, Shoji A, Tochigi N, Morioka H, Toyama Y, Hirohashi S, Kawai A, and Yamada T. 2010. Reduced argininosuccinate synthetase is a predictive biomarker for the development of pulmonary metastasis in patients with osteosarcoma. Molecular Cancer Therapeutics, Mar; 9(3): 535e544. DOI: https://doi.org/10.1158/1535-7163. MCT-09-0774. [59] Lan J, Tai HC, Lee SW, Chen TJ, Huang HY, and Li CF. 2014. Deficiency in expression and epigenetic DNA Methylation of ASS1 gene in nasopharyngeal carcinoma: negative prognostic impact and therapeutic relevance. Tumor Biology, Jan; 35(1): 161e169. DOI: https://doi.org/10.1007/s13277-013-1020-8. [60] Huang HY, Wu WR, Wang YH, Wang JW, Fang FM, Tsai JW, Li SH, Hung HC, Yu SC, Lan J, Shiue YL, Hsing CH, Chen LT, and Li CF. 2013. ASS1 as a novel tumor suppressor gene in myxofibrosarcomas: aberrant loss via epigenetic DNA methylation confers aggressive phenotypes, negative prognostic impact, and therapeutic relevance. Clinical Cancer Research, Jun; 19(Issue 11): 2861e2872. DOI: https://doi. org/10.1158/1078-0432.CCR-12-2641. [61] Nicholson LJ, Smith PR, Hiller L, Szlosarek PW, Kimberley C, Sehouli J, Koensgen D, Mustea A, Schmid P, and Crook T. 2009. Epigenetic silencing of argininosuccinate synthetase confers resistance to platinum-induced cell death but collateral sensitivity to arginine auxotrophy in ovarian cancerInternational. Journal of Cancer, Sep 15; 125(6): 1454e1463. DOI: https://doi.org/10.1002/ijc.24546. [62] Allen MD, Luong P, Hudson C, Leyton J, Delage B, Ghazaly E, Cutts R, Yuan M, Syed N, Lo Nigro C, Lattanzio L, Chmielewska-Kassassir M, Tomlinson I, Roylance R, Whitaker HC, Warren AY, Neal D, Frezza C, Beltran L, Jones LJ, Chelala C, Wu BW, Bomalaski JS, Jackson RC, Lu YJ, Crook T, Lemoine NR, Mather S, Foster J, Sosabowski J, Avril N, Li CF, and Szlosarek PW. 2014. Prognostic and therapeutic impact of argininosuccinate synthetase 1 control in bladder cancer as monitored longitudinally by PET imaging. Cancer Research, Feb; 74(3): 896e907. DOI: https:// doi.org/10.1158/0008-5472.CAN-13-1702. [63] Tsai CY, Chi HC, Chi LM, Yang HY, Tsai MM, Lee KF, Huang HW, Chou LF, Cheng AJ, Yang CW, Wang CS, and Lin KH. 2018. Argininosuccinate synthetase 1 contributes to gastric cancer invasion and progression by modulating autophagy. FASEB Journal, May; 32(5): 2601e2614. DOI: https://doi.org/10.1096/fj. 201700094R.

Tumor starvation by L-arginine deprivation

101

[64] Shan YS, Hsu HP, Lai MD, Yen MC, Luo YP, and Chen YL. 2015. Increased expression of argininosuccinate synthetase protein predicts poor prognosis in human gastric cancer. Oncology Reports, Jan; 33(1): 49e57. DOI: https://doi.org/10.3892/or. 2014.3556. [65] Huang CC, Tsai ST, Kuo CC, Chang JS, Jin YT, Chang JY, and Hsiao JR. 2012. Arginine deprivation as a new treatment strategy for head and neck cancer. Oral Oncology. Dec; 48(12): 1227e1235. DOI: https://doi.org/10.1016/j.oraloncology. 2012.06.004. [66] Caldwell RB, Toque HA, Narayanan SP, and Caldwell RW. 2015. Arginase: an old enzyme with new tricks. Trends in Pharmacological Sciences, 36(6): 395e405. DOI: https://doi.org/10.1016/j.tips.2015.03.006. [67] Vockley JG, Jenkinson CP, Shukla H, Kern RM, Grody WW, and Cederbaum SD. 1996. Cloning and characterization of the human type II arginase gene. Genomics, Dec; 38(2): 118e123. DOI: https://doi.org/10.1006/geno.1996.0606. [68] Yang Z and Ming XF. 2014. Functions of arginase isoforms in macrophage inflammatory responses: impact on cardiovascular diseases and metabolic disorders. Frontiers in Immunology, Oct; 533. DOI: https://doi.org/10.3389/fimmu.2014.00533. [69] Quirie A, Demougeot C, Bertrand N, Mossiat C, Garnier P, Marie C, and PrigentTessier A. 2013. Effect of stroke on arginase expression and localization in the rat brain. European Journal of Neuroscience, Apr; 37(7): 1193e1202. DOI: https://doi.org/ 10.1111/ejn.12111. [70] Singh R, Pervin S, Karimi S, Cederbaum S, and Chaudhuri G. 2000. Arginase activity in human breast cancer cell lines: N(omega)-hydroxy-L-arginine selectively inhibits cell proliferation and induces apoptosis in MDA-MB-468 cells. Cancer Research, Jun; 60(12): 3305e3312. PMID: 10866325. [71] Buga GM, Wei LH, Bauer PM, Fukuto JM, and Ignarro LJ. 1998. NG-hydroxy-Larginine and nitric oxide inhibit Caco-2 tumor cell proliferation by distinct mechanisms. American Journal of Physiology - Regulatory Integrative and Comparative Physiology, Oct; 275(4): R1256eR1264. PMID: 9756558. [72] Rotondo R, Mastracci L, Piazza T, Barisione G, Fabbi M, Cassanello M, Costa R, Morandi B, Astigiano S, Cesario A, Sormani MP, Ferlazzo G, Grossi F, Ratto GB, Ferrini S, and Frumento G. 2008. Arginase 2 is expressed by human lung cancer, but it neither induces immune suppression, nor affects disease progression. International Journal of Cancer, Sep 1; 123(5): 1108e1116. DOI: https://doi.org/10.1002/ijc.23437. [73] Dillon BJ, Holtsberg FW, Ensor CM, Bomalaski JS, and Clark MA. 2002. Biochemical characterization of the arginine degrading enzymes arginase and arginine deiminase and their effect on nitric oxide production. Medical Science Monitor, Jun; 8(7): BR248eB253. PMID: 12118186. [74] Cheng PN, Lam TL, Lam WM, Tsui SM, Cheng AW, Lo WH, and Leung YC. 2007. Pegylated recombinant human arginase (rhArg-peg5,000mw) inhibits the in vitro and in vivo proliferation of human hepatocellular carcinoma through arginine depletion. Cancer Research, Jan; 67(1): 309e317. DOI: https://doi.org/10.1158/0008-5472. CAN-06-1945. [75] Stone EM, Chantranupong L, and Georgiou G. 2010. The second-shell metal ligands of human arginase affect coordination of the nucleophile and substrate. Biochemistry, Nov; 49(49): 10582e10588. DOI: https://doi.org/10.1021/bi101542t. [76] Stone EM, Glaze ES, Chantranupong L, Cherukuri P, Breece RM, Tierney DL, Curley SA, Iverson BL, and Georgiou G. 2010. Replacing Mn(2þ) with Co(2þ) in human arginase i enhances cytotoxicity toward l-arginine auxotrophic cancer cell lines. ACS Chemical Biology, Jan; 5(3): 333e342. DOI: https://doi.org/10.1021/cb900267j.

102

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

[77] Hernandez CP, Morrow K, Lopez-Barcons, Zabaleta J, Sierra R, Velasco C, Cole J, and Rodriguez PC. 2010. Pegylated arginase I: a potential therapeutic approach in T-ALL. Blood, Jun; 115(25): 5214e5221. https://doi.org/10.1182/blood-2009-12-258822. [78] Hsueh EC, Knebel SM, Lo WH, Leung YC, Chen PN, and Hsueh CT. 2011. Deprivation of arginine by recombinant human arginase in prostate cancer cells. Journal of Hematology and Oncology, Apr; 5: 17. DOI: https://doi.org/10.1186/1756-8722-5-17. [79] Lam TL, Wong GK, Chow HY, Chong HC, Chow TL, Kwok SY, Cheng PN, Wheatley DN, Lo WH, and Leung YC. 2011. Recombinant human arginase inhibits the in vitro and in vivo proliferation of human melanoma by inducing cell cycle arrest and apoptosis. Pigment Cell and Melanoma Research, Oct; 24(2): 366e376. https://doi. org/10.1111/j.1755-148X.2010.00798.x. [80] Patil MD, Bhaumik J, Babykutty S, Banerjee UC, and Fukumura D. 2016. Arginine dependence of tumor cells: targeting a chink in cancer’s armor. Oncogene, Apr; 35(38): 4957e4972. DOI: https://doi.org/10.1038/onc.2016.37. [81] Wheatley DN and Campbell E. 2002. Arginine catabolism, liver extracts and cancer. Pathology Oncology Research, Mar; 8(1): 8e25. DOI: https://doi.org/10.1007/ BF03033696. [82] Dai F, Yu W, Song J, Li Q, Wang C, and Xie S. 2017. Extracellular polyaminesinduced proliferation and migration of cancer cells by ODC, SSAT, and Akt1mediated pathway. Anticancer Drugs, Apr; 28(4): 457e464. DOI: https://doi.org/10. 1097/CAD.0000000000000465. [83] Locke M, Ghazaly E, Freitas MO, Mitsinga M, Lattanzio L, Lo Nigro C, Nagano A, Wang J, Chelala C, Szlosarek P, and Martin SA. Inhibition of the polyamine synthesis pathway is synthetically lethal with loss of argininosuccinate synthase 1. Cell Reports, Aug; 16(6): 1604e1613. DOI: https://doi.org/10.1016/j.celrep.2016.06.097. [84] Samal K, Zhao P, Kendzicky A, Yco LP, McClung H, Gerner E, Burns M, Bachmann AS, and Sholler G. 2013. AMXT-1501, a novel polyamine transport inhibitor, synergizes with DFMO in inhibiting neuroblastoma cell proliferation by targeting both ornithine decarboxylase and polyamine transport. International Journal of Cancer, Mar; 133(6): 1323e1333. DOI: https://doi.org/10.1002/ijc.28139. [85] Samoilenko AO, Milinevska OA, Karnaushenko OV, Shlyakhovenko VA, and Zaletok SP. 2015. Effect of polyamine metabolism inhibitors on Lewis lung carcinoma growth and metastasis. Experimental Oncology, 37(2): 151e153. https://dspace.nbuv. gov.ua/handle/123456789/145472. [86] Karavaev BI. 1998. Current trends in the development of monitoring observation in a resuscitation and intensive therapy department. Anesteziologiia Reanimatologiia, Jun; 4: 57e58. PMID: 9770823. [87] Bonavida B and Garban H. 2015. Nitric oxide-mediated sensitization of resistant tumor cells to apoptosis by chemo-immunotherapeutics. Redox Biology, Dec1; 6: 486e494. DOI: https://doi.org/10.1016/j.redox.2015.08.013. [88] Tengan CH and Moraes CT. 2017. NO control of mitochondrial function in normal and transformed cells. Biochimica et Biophysica Acta e Bioenergetics, Aug; 1858(8): 573e581. DOI: https://doi.org/10.1016/j.bbabio.2017.02.009. [89] Sarti P, Forte E, Giuffre A, Mastronicola D, Magnifico MC, and Arese M. 2012. The chemical interplay between nitric oxide and mitochondrial cytochrome c oxidase: reactions, effectors and pathophysiology. International Journal of Cell Biology, 2012 (Article 571067). DOI: https://doi.org/10.1155/2012/571067. [90] Zhong H, De Marzo AM, Laughner E, Lim M, Hilton DA, Zagzag D, Buechler P, Isaacs WB, Semenza GL, and Simons JW. 1999. Overexpression of hypoxia-inducible factor 1alpha in common human cancers and their metastases. Cancer Research, Nov; 59(22): 5830e5835. PMID: 10582706. [91] Brune B and Zhou J. 2007. Nitric oxide and superoxide: interference with hypoxic signaling. Cardiovascular Research, Jul; 75(2): 275e282. DOI: https://doi.org/10.1016/j. cardiores.2007.03.005.

Tumor starvation by L-arginine deprivation

103

[92] Chang CF, Diers AR, and Hogg N. 2015. Cancer cell metabolism and the modulating effects of nitric oxide. Free Radical Biology and Medicine, Feb; 79: 324e336. DOI: https://doi.org/10.1016/j.freeradbiomed.2014.11.012. [93] Bahadoran Z, Mirmiran P, Tahmasebinejad Z, and Azizi F. 2016. Dietary L-arginine intake and the incidence of coronary heart disease: Tehran lipid and glucose study. Nutrition and Metabolism (Lond), Marr; 13: 23. DOI: https://doi.org/10.1186/s12986016-0084-z. [94] Yarlagadda K, Hassani J, Foote IP, and Markowitz J. 2017. The role of nitric oxide in melanoma. Biochimica et Biophysica Acta - Reviews on Cancer, Dec; 1868(2): 500e509. DOI: https://doi.org/10.1016/j.bbcan.2017.09.005. [95] Rouaud F, Romero-Perez M, Wang H, Lobysheva I, Ramassamy B, Henry E, Tauc P, Giacchero D, Boucher JJ, Deprez E, Rocchi S, and Slama-Schwok A. 2014. Regulation of NADPH-dependent Nitric Oxide and reactive oxygen species signalling in endothelial and melanoma cells by a photoactive NADPH analogue. Oncotarget, 5(21): 10650e10664. DOI: https://doi.org/10.18632/oncotarget.2525. [96] Zhang X, Fang X, Gao Z, Chen W, Tao F, Cai P, Yuan H, Shu Y, Xu Q, Sun Y, and Gu Y. 2014. Axitinib, a selective inhibitor of vascular endothelial growth factor receptor, exerts an anticancer effect in melanoma through promoting antitumor immunity. Anti-Cancer Drugs, 25(2): 204e211. DOI: https://doi.org/10.1097/CAD. 0000000000000033. [97] Baritaki S, Huerta-Yepez S, Sahakyan A, Karagiannides I, Bakirtzi A. Jazireh A, Bonavida B, and Kugel JF. 2010. Mechanisms of nitric oxide-mediated inhibition of EMT in cancer: inhibition of the metastasis-inducer Snail and induction of the metastasis-suppressor RKIP. Cell Cycle, Dec; 9(24): 4931e4940. DOI: https://doi. org/10.4161/cc.9.24.14229. [98] Yeh CL, Pai H, Li CC, Tsai YL, and Yeh SL. 2010. Effect of arginine on angiogenesis induced by human colon cancer: in vitro and in vivo studies. Journal of Nutritional Biochemistry, Jun; 21(6): 538e543. DOI: https://doi.org/10.1016/j.jnutbio.2009.03. 005. [99] Wolf C, Bruss M, Hanisch B, Gothert M, von Kugelgen I, and Molderings GJ. 2007. Molecular basis for the antiproliferative effect of agmatine in tumor cells of colonic, hepatic, and neuronal origin. Molecular Pharmacology, 71(1): 276e283. DOI: https:// doi.org/10.1124/mol.106.028449. [100] Regunathan S and Piletz JE. 2003. Regulation of inducible nitric oxide synthase and agmatine synthesis in macrophages and astrocytes. Annals of the New York Academy of Sciences, Jan; 1009(1): 20e29. DOI: https://doi.org/10.1196/annals.1304.002. [101] Satriano J. 2004. Arginine pathways and the inflammatory response: interregulation of nitric oxide and polyamines: review article. Amino Acids, Apr; 26(4): 321e329. DOI: https://doi.org/10.1007/s00726-004-0078-4. [102] Phang JM, Liu W, Hancock CN, and Fischer JW. 2015. Proline metabolism and cancer: emerging links to glutamine and collagen. Current Opinion in Clinical Nutrition and Metabolic Care, Jan; 18(1): 71e77. DOI: https://doi.org/10.1097/MCO. 0000000000000121. [103] Altman BJ, Stine ZE, and Dang CV. 2016. From Krebs to clinic: glutamine metabolism to cancer therapy, Nature Reviews Cancer, Sep; 16(10): 619e634. DOI: https:// doi.org/10.1038/nrc.2016.71. [104] Tanner JJ, Fendt SM, and Becker DF. 2018. The proline cycle as a potential Cancer therapy target. Biochemistry, Jun; 57(25): 3433e3444. DOI: https://doi.org/10.1021/ acs.biochem.8b00215. [105] Mussai F, Egan S, Higginbotham-Jones J, Perry T, Beggs A, Odintsova E, Loke J, Pratt G, Lo KPUA, Ng M, Kearns P, Cheng P, and De Santo C. 2015. Arginine dependence of acute myeloid leukemia blast proliferation: a novel therapeutic target. Blood, Apr 9; 125(15): 2386e2396. DOI: https://doi.org/10.1182/blood-2014-09-600643.

104

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

[106] Lam SK, Li YY, Xu S, Leung LL, Zheng KPUYF, Cheng PN, and Ho JC. 2017. Growth suppressive effect of pegylated arginase in malignant pleural mesothelioma xenografts. Respiratory Research, May; 18(Article 80): 80. DOI: https://doi.org/10. 1186/s12931-017-0564-3. [107] Ott PA, Carvajal RD, Pandit-Taskar N, Jungbluth AA, Hoffman EW, Wu BW, Bomalaski JS, Venhaus R, Pan L, Old LJ, Pavlick AC, and Wolchok JD. 2013. Phase I/II study of pegylated arginine deiminase (ADI-PEG 20) in patients with advanced melanoma. Investigational New Drugs, Apr; 31(2): 425e434. DOI: https://doi.org/10. 1007/s10637-012-9862-2. [108] Walts AE, Bomalaski JS, Ines D, and Orsulic S. 2015. Argininosuccinate synthetase (ASS) deficiency in high-grade pulmonary neuroendocrine carcinoma: an opportunity for personalized targeted therapy. Journal of Cancer Research and Clinical Oncology, Dec; 141(8): 1363e1369. DOI: https://doi.org/10.1007/s00432-014-1904-z. [109] Qiu F, Chen YR, Liu X, Chu CY, Shen LJ, Xu J, Gaur S, Forman HJ, Zhang H, Zheng S, Yen Y, Huang J, Kung HJ, and Ann DK. 2014. Arginine starvation impairs mitochondrial respiratory function in ASS1-deficient breast cancer cells. Science Signaling, Apr 1; 7(319): ra31. DOI: https://doi.org/10.1126/scisignal.2004761. [110] Battisti S, Valente D, Albonici L, Bei R, Modesti A, and Palumbo C. 2012. Nutritional stress and arginine auxotrophy confer high sensitivity to chloroquine toxicity in mesothelioma cells. American Journal of Respiratory Cell and Molecular Biology, Apr; 46(4): 498e506. DOI: https://doi.org/10.1165/rcmb.2011-0195OC. [111] Wang Z, Shi X, Li Y, Zeng X, Fan J, Sun Y, Xian Z, Zhang G, Wang S, Hu H, and Ju D. 2014. Involvement of autophagy in recombinant human arginase-induced cell apoptosis and growth inhibition of malignant melanoma cells. Applied Microbiology and Biotechnology, Mar; 98(6): 2485e2494. DOI: https://doi.org/10.1007/s00253-0135118-0. [112] Khoury O, Ghazale N, Stone E, El-Sibai M, Frankel AE, and Abi-Habib RJ. 2015. Human recombinant arginase I (Co)-PEG5000 [HuArgI (Co)-PEG5000]-induced arginine depletion is selectively cytotoxic to human glioblastoma cells. Journal of Neuro-Oncology, 122(1): 75e85. DOI: https://doi.org/10.1007/s11060-014-1698-5. [113] Kim RH, Coates JM, Bowles TL, McNerney GP, Sutcliffe J, Jung JU, GandourEdwards R, Chuang FYS, Bold RJ, and Kung HJ. 2009. Arginine deiminase as a novel therapy for prostate cancer induces autophagy and caspase-independent apoptosis. Cancer Research, Jan 15; 69(2): 700e708. DOI: https://doi.org/10.1158/ 0008-5472.CAN-08-3157. [114] Garcia-Navas R, Munder M, and Mollinedo F. 2012. Depletion of L-arginine induces autophagy as a cytoprotective response to endoplasmic reticulum stress in human T lymphocytes. Autophagy, Nov; 8(11): 1557e1576. DOI: https://doi.org/ 10.4161/auto.21315. [115] Shen W, Zhang X, Fu X, Fan J, Luan J, Cao Z, Yang P, Xu Z, and Ju D. 2017. A novel and promising therapeutic approach for NSCLC: recombinant human arginase alone or combined with autophagy inhibitor. Cell Death and Disease, Mar; 8(3): e2720. DOI: https://doi.org/10.1038/cddis.2017.137. [116] Changou CA, Chen YR, Xing L, Yen Y, Chuang FY, Cheng RH, Bold RJ, Ann DK, and Kung HJ. 2014. Arginine starvation-associated atypical cellular death involves mitochondrial dysfunction, nuclear DNA leakage, and chromatin autophagy. Proceedings of the National Academy of Sciences USA. Sep; 111(39): 14147e14152. DOI: https://doi.org/10.1073/pnas.1404171111. [117] Garg AD, Dudek AM, Ferreira GB, Verfaillie T, Vandenabeele P, Krysko DV, Mathieu C, Agostinis P. 2013. ROS-induced autophagy in cancer cells assists in evasion from determinants of immunogenic cell death. Autophagy, Sep; 9(9): 1292e1307. DOI: https://doi.org/10.4161/auto.25399.

Tumor starvation by L-arginine deprivation

105

[118] Ni Z, Wang B, Dai X, Ding W, Yang T, Li X, Lewin S, Xu L, Lian J, and He F. 2014. HCC cells with high levels of Bcl-2 are resistant to ABT-737 via activation of the ROS-JNK-autophagy pathway. Free Radical Biology and Medicine, May; 70: 194e203. DOI: https://doi.org/10.1016/j.freeradbiomed.2014.02.012. [119] Monsalve M, Borniquel S, Valle I, and Lamas S. 2007. Mitochondrial dysfunction in human pathologies. Frontiers in Bioscience, 12(3): 1131e1153. DOI: https://doi.org/ 10.2741/2132. [120] Lin AB, McNeely SC, and Beckmann RP. 2017. Achieving precision death with cellcycle inhibitors that target DNA replication and repair. Clinical Cancer Research, Jul; 23(13): 3232e3240. DOI: https://doi.org/10.1158/1078-0432.CCR-16-0083. [121] Medema RH and Macurek L. 2012. Checkpoint control and cancer. Oncogene, May; 31(21): 2601e2613. DOI: https://doi.org/10.1038/onc.2011.451. [122] Lam TL, Wong GKY, Chong HC, Cheng PN, Choi SC, Chow TL, Kwok SY, Poon RTP, Wheatley DN, Lo WH, and Leung YC. 2009. Recombinant human arginase inhibits proliferation of human hepatocellular carcinoma by inducing cell cycle arrest. CancerLetters,May;277(1):91e100.DOI:https://doi.org/10.1016/j.canlet.2008.11.031. [123] Gong H, Zolzer F, von Recklinghausen G, Havers W, and Schweigerer L. 2000. Arginine deiminase inhibits proliferation of human leukemia cells more potently than asparaginase by inducing cell cycle arrest and apoptosis. Leukemia, 14(5): 826e829. DOI: https://doi.org/10.1038/sj.leu.2401763. [124] Delage B, Luong P, Maharaj L, O’Riain C, Syed N, Crook T, Hatzimichael E, Papoudou-Bai A, Mitchell TJ, Whittaker SJ, Cerio R, Gribben J, Lemoine N, Bomalaski J, Li CF, Joel S, Fitzgibbon J, Chen LT, and Szlosarek PW. 2012. Promoter methylation of argininosuccinate synthetase-1 sensitises lymphomas to arginine deiminase treatment, autophagy and caspase-dependent apoptosis. Cell Death and Disease, Jul; 3: e342. DOI: https://doi.org/10.1038/cddis.2012.83. [125] You M, Savaraj N, Kuo MY, Wangpaichitr M, Varona-Santos J, Wu C, Nguyen, DM, and Feun L. 2013. TRAIL induces autophagic protein cleavage through caspase activation in melanoma cell lines under arginine deprivation. Molecular and Cellular Biochemistry, Feb; 374(1e2): 181e190. DOI: https://doi.org/10.1007/s11010-012-1518-1. [126] Syed N, Langer J, Janczar K, Singh P, Lo Nigro C, Lattanzio L, Coley HM, Hatzimichael E, Bomalaski J, Szlosarek P, Awad M, O’Neil K, Roncaroli F, and Crook T. 2013. Epigenetic status of argininosuccinate synthetase and argininosuccinate lyase modulates autophagy and cell death in glioblastoma. Cell Death and Disease, Jan; 4(1): e458. DOI: https://doi.org/10.1038/cddis.2012.197. [127] Zeng X, Li Y, Fan J, Zhao H, Xian Z, Sun Y, Wang Z, Wang S, Zhang G, and Ju D. 2013. Recombinant human arginase induced caspase-dependent apoptosis and autophagy in non-Hodgkin’s lymphoma cells. Cell Death and Disease, Oct; 4: e840. DOI: https://doi.org/10.1038/cddis.2013.359. [128] Szlosarek PW, Steele JP, Nolan L, Gilligan D, Taylor P, Spicer J, Lind M, Mitra S, Shamash J, Phillips MM, Luong P, Payne S, Hillman P, Ellis S, Szyszko T, Dancey G, Butcher L, Beck S, Avril NE, Thomson J, Johnston A, Tomsa M, Lawrence C, Schmid P, Crook T, Wu BW, Bomalaski JS, Lemoine N, Sheaff T, Rudd RM, Fennell D, and Hackshaw A. 2017. Arginine deprivation with pegylated arginine deiminase in patients with argininosuccinate synthetase 1-deficient malignant pleural mesothelioma: a randomized clinical trial. JAMA Oncology, Jan 1; 3(1): 58e66. DOI: https://doi.org/10.1001/jamaoncol.2016.3049. [129] Izzo F, Marra P, Beneduce G, Castello G, Vallone P, De Rosa V, Cremona F, Ensor CM, Holtsberg FW, Bomalaski JS, Clark MA, Ng C, and Curley SA. 2004. Pegylated arginine deiminase treatment of patients with unresectable hepatocellular carcinoma: results from phase I/II studies. Journal of Clinical Oncology, 22(10): 1815e1822. DOI: https://doi.org/10.1200/JCO.2004.11.120.

106

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

[130] Ascierto PA, Scala S, Castello G, Daponte A, Simeone F, Ottaiano A, Beneduce G, De Rosa V, Izzo F, Melucci MT, Ensor CM, Prestayko AW, Holtsberg FW, Bomalaski JS, Clark A, Savaraj N, Feun LG, and Logan TF. 2005. Pegylated arginine deiminase treatment of patients with metastatic melanoma: results from phase I and II studies. Journal of Clinical Oncology, 23(30): 7660e7668. DOI: https://doi.org/10. 1200/JCO.2005.02.0933. [131] Yang TS, Lu SN, Chao Y, Sheen IS, Lin CC, Wang TE, Chen, SC, Wang JH, Liao LY, Thomson JA, Wang-Peng J, Chen PJ, and Chen LT. 2010. A randomised phase II study of pegylated arginine deiminase (ADI-PEG 20) in Asian advanced hepatocellular carcinoma patients. British Journal of Cancer, Sep 28; 103(7): 954e960. DOI: https://doi.org/10.1038/sj.bjc.6605856. [132] Glazer ES, Piccirillo M, Albino V, Di Giacomo R, Palaia R, Mastro AA, Beneduce G, Castello G, De Rosa V, Petrillo A, Ascierto PA, Curley SA, and Izzo F. 2010. Phase II study of pegylated arginine deiminase for nonresectable and metastatic hepatocellular carcinoma. Journal of Clinical Oncology, May 1; 28(13): 2220e2226. DOI: https://doi.org/10.1200/JCO.2009.26.7765. [133] Burki TK. 2016. Arginine deprivation for ASS1-deficient mesothelioma. Lancet Oncology, Oct 1; 17(10): e423. DOI: https://doi.org/10.1016/S1470-2045(16) 30446-6. [134] Abou-Alfa G, Qin S, Ryoo BY, Lu SN, Yen CJ, Feng YH, Lim HY, Izzo F, Colombo M, Sarker D, Bolondi L, Vaccaro G, Harris WP, Chen Z, Hubner RA, Meyer T, Sun W, Harding JJ, Hollywood EM, Ma J, Wan PJ, Ly M, Bomalaski J, Johnston A, Lin CC, Chao Y, and Chen LT. 2018. ADI-PEG 20 plus best supportive care versus placebo plus best supportive care in patients with advanced hepatocellular carcinoma. Annals of Oncology, Jun; 29(6): 1402e1408. DOI: https://doi.org/10. 1093/annonc/mdy101. [135] Feun LG, Marini A, Walker G, Elgart G, Moffat F, Rodgers SE, Wu CJ, You M, Wangpaichitr M, Kuo T, Sisson W, Jungbluth AA, Bomalaski J, and Savaraj N. 2012. Negative argininosuccinate synthetase expression in melanoma tumours may predict clinical benefit from arginine-depleting therapy with pegylated arginine deiminase. British Journal of Cancer, Apr 24; 106(9): 1481e1485. DOI: https://doi.org/10.1038/ bjc.2012.106. [136] Yau T, Cheng PN, Chan P, Chen L, Yuen J, Pang R, Fan ST, Wheatley DN, and Poon RT. 2015. Preliminary efficacy, safety, pharmacokinetics, pharmacodynamics and quality of life study of pegylated recombinant human arginase 1 in patients with advanced hepatocellular carcinoma. Investigational New Drugs, Apr; 33(2): 496e504. DOI: https://doi.org/10.1007/s10637-014-0200-8. [137] Long Y, Tsai WB, Wang D, Hawke DH, Savaraj N, Feun LG, Hung MC, Chen HH, and Kuo MT. 2017. Argininosuccinate synthetase 1 (ASS1) is a common metabolic marker of chemosensitivity for targeted arginine- and glutamine-starvation therapy. Cancer Letters, Mar 1; 388(1): 54e63. DOI: https://doi.org/10.1016/j.canlet.2016.11.028. [138] Xu S, Lam SK, Cheng PN, and Ho JC. 2018. Recombinant human arginase induces apoptosis through oxidative stress and cell cycle arrest in small cell lung cancer. Cancer Science, Nov; 109(11): 3471e3482. DOI: https://doi.org/10.1111/cas.13782. [139] Abou-Alfa GK, Qin S, Ryoo BY, Lu SN, Yen CJ, Feng YH, Lim HY, Izzo F, Colombo M, Sarker D, Bolondi L, Vaccaro G, Harris WP, Chen Z, Hubner, RA, Meyer T, Sun W, Harding JJ, Hollywood EM, Ma J, Wan PJ, Ly M, Bomalaski J, Johnston A, Lin CC, Chao Y, and Chen LT. 2018. Phase III randomized study of second line ADI-PEG 20 plus best supportive care versus placebo plus best supportive care in patients with advanced hepatocellular carcinoma. Annals of Oncology, Jun 1; 29(6): 1402e1408. DOI: https://doi.org/10.1093/annonc/mdy101. [140] Tsai HJ, Jiang SS, Hung WC, Borthakur G, Lin SF, Pemmaraju N, Jabbour E, Bomalaski JS, Chen YP, Hsiao HH, Wang C, Kuo CY, Chang H, Yeh SP, Cortes J, Chen LT, and Chen TY. 2017. A phase II study of arginine deiminase (ADI-PEG20) in relapsed/refractory or poor-risk acute myeloid leukemia patients. Scientific Reports, Sep; 7(1): 11253. DOI: https://doi.org/10.1038/s41598-017-10542-4.

Tumor starvation by L-arginine deprivation

107

[141] Fojo A, Hamilton TC, Young RC, and Ozols RF. 1987. Multidrug resistance in ovarian cancer. Cancer, Oct 15; 60(Suppl. 8): 2075e2080. DOI: https://doi.org/10. 1002/1097-0142(19901015)60:8þ3.0.CO;2-F. [142] Manca A, Sini MC, Izzo F, Ascierto PA, Tatangelo F, Botti G, Gentilcore G, Capone M, Mozzillo N, Rozzo C, Cossu A, Tanda F, and Palmieri G. 2011. Induction of arginosuccinate synthetase (ASS) expression affects the antiproliferative activity of arginine deiminase (ADI) in melanoma cells. Oncology Reports, Mar 18; 25(6): 1495e1502. DOI: https://doi.org/10.3892/or.2011.1220. [143] Tsai WB, Aiba I, Lee SY, Feun L, Savaraj N, and Kuo MT. 2009. Resistance to arginine deiminase treatment in melanoma cells is associated with induced argininosuccinate synthetase expression involving c-Myc/HIF-1alpha/Sp4. Molecular Cancer Therapeutics, Dec; 8(12): 3223e3233. DOI: https://doi.org/10.1158/1535-7163. MCT-09-0794. [144] Tsai WB, Aiba I, Long Y, Lin HK, Feun L, Savaraj N, and Kuo MT. 2012. Activation of Ras/PI3K/ERK pathway induces c-Myc stabilization to upregulate argininosuccinate synthetase, leading to arginine deiminase resistance in melanoma cells. Cancer Research, May 14; 72(10): 2622e2633. DOI: https://doi.org/10.1158/00085472.CAN-11-3605. [145] Long Y, Tsai WB, Wangpaichitr M, Tsukamoto T, Savaraj N, Feun LG, and Kuo MT. 2013. Arginine deiminase resistance in melanoma cells is associated with metabolic reprogramming, glucose dependence, and glutamine addiction. Molecular Cancer Therapeutics, Nov; 12(11): 2581e2590. PMID: 23979920. [146] Lam SK, Yan S, Xu S, Cheng KPUPN, and Ho JC. 2019. Endogenous arginase 2 as a potential biomarker for PEGylated arginase 1 treatment in xenograft models of squamous cell lung carcinoma. Oncogenesis, Feb; 8(3): 18. DOI: https://doi.org/10. 1038/s41389-019-0128-0.

Reprint of full text, figures and tables with permission from Zou S, Wang X, Liu P, Ke C, and Xu S. 2019. Arginine metabolism and deprivation in cancer therapy. Biomedicine and Pharmacotherapy, Oct; 118: 109210. DOI: https://doi.org/10.1016/ j.biopha.2019.109210. Elsevier Masson SAS. All rights reserved. Note: Acknowledgments and grant support source(s) may be found in the original journal publication.

3.5 Riess C, Shokraie F, Classen CF, Kreikemeyer B, Fiedler T, Junghanss C, and Maletzki C. 2018. Arginine-depleting enzymes e An increasingly recognized treatment strategy for therapy-refractory malignancies. Cellular Physiology and Biochemistry, 51: 854e870. DOI: https://doi.org/10.1159/000495382. 3.5.1 Synopsis This study from Clinic III-Hematology/Oncology/Palliative Care, Rostock University Medical Center, University of Rostock, Germany, reports that aginine auxotrophy, present in certain tumor types, is due to the silencing of

108

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

argininosuccinate synthetase 1 or arginine lyase genes. Although such tumors are often associated with intrinsic chemoresistance and have, therefore, a poor prognosis, arginine auxotrophy on the other hand makes them vulnerable to treatment with arginine-degrading enzymes. The investigators report that bacterial arginine deiminases (ADI) is a frequently applied arginine-degrading agent. They further report that the anti-cancer effects of ADI derived from different bacteria have been extensively studied in numerous preclinical cell culture and xenograft models. Of those, mycoplasma-derived ADI-PEG20 is the most commonly used and is currently under clinical investigation as a single therapy agent as well as in combination with different antineoplastic compounds. ADI has been shown to reduce the metabolic activity in tumor cells, thus contributing to autophagy, senescence and apoptosis in arginine auxotrophic cells. The investigators outline recent experimental ADI-based treatment approaches and their translation into clinic application and they summarize new insights into the molecular mechanisms underlying the anti-cancer effects of ADI that might facilitate the refinement of ADI-based combination therapy approaches. This is a very detailed report and we regret that we could not reprint it here in full with figures due to cost considerations. However, we suggest that you go to https://doi.org/10.1159/000495382 to access this review for detailed information.

3.6 Burki TK. 2016. Arginine deprivation for ASS1deficient mesothelioma. Lancet Oncology, Oct 1; 17(10): e423. DOI: https://doi.org/10.1016/S14702045(16)30446-6. 3.6.1 Abstract Arginine deprivation therapy improves progression-free survival in patients with argininosuccinate synthetase 1 (ASS1)-deficient malignant pleural mesothelioma, according to a new study. The investigators in the phase 2 trial screened 201 patients with mesothelioma from centres in the UK. 97 (48%) of 201 patients showed ASS1 deficiency. 68 patients were randomly assigned to receive either the arginine-lowering agent pegylated arginine deiminase (ADI-PEG20; 36$8 mg/m2 per week for up to 6 months) plus best supportive care (n¼44), or best supportive care alone (n¼24).

Tumor starvation by L-arginine deprivation

109

Median progression-free survival in the ADI-PEG20 arm was 3$2 months versus 2$0 months in the best supportive care group (hazard ratio [HR] 0$56; 95% CI 0$33e0$96, p¼0$03). Abstract reprinted with permission from Lancet. Oncology.

3.7 Qiu F, Chen YR, Liu X, Chu CY, Shen LJ, Xu J, Gaur S, Forman HJ, Zhang H, Zheng S, Yen Y, Huang J, Kung HJ, and Ann DK. 2014. Arginine starvation impairs mitochondrial respiratory function in ASS1-deficient breast cancer cells. Science Signaling, Apr 1; 7(319): ra31. DOI: https://doi.org/10.1126/scisignal.2004761. 3.7.1 Abstract Autophagy is the principal catabolic response to nutrient starvation and is necessary to clear dysfunctional or damaged organelles, but excessive autophagy can be cytotoxic or cytostatic and contributes to cell death. Depending on the abundance of enzymes involved in molecule biosynthesis, cells can be dependent on uptake of exogenous nutrients to provide these molecules. Argininosuccinate synthetase 1 (ASS1) is a key enzyme in arginine biosynthesis, and its abundance is reduced in many solid tumors, making them sensitive to external arginine depletion. We demonstrated that prolonged arginine starvation by exposure to ADI-PEG20 (pegylated arginine deiminase) induced autophagy-dependent death of ASS1-deficient breast cancer cells, because these cells are arginine auxotrophs (dependent on uptake of extracellular arginine). Indeed, these breast cancer cells died in culture when exposed to ADI-PEG20 or cultured in the absence of arginine. Arginine starvation induced mitochondrial oxidative stress, which impaired mitochondrial bioenergetics and integrity. Furthermore, arginine starvation killed breast cancer cells in vivo and in vitro only if they were autophagy-competent. Thus, a key mechanism underlying the lethality induced by prolonged arginine starvation was the cytotoxic autophagy that occurred in response to mitochondrial damage. Last, ASS1 was either low in abundance or absent in more than 60% of 149 random breast cancer biosamples, suggesting that patients with such tumors could be candidates for arginine starvation therapy. Reprinted with permission from Science Signaling.

110

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

3.8 Kim RH, Bold RJ, and Kung H-J. 2009. ADI, autophagy and apoptosis: Metabolic stress as a therapeutic option for prostate cancer. Autophagy, May 20; 5(4): 567e568. DOI: https://doi.org/10.4161/auto.5.4.8252. (https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC4294541/). 3.8.1 Abstract Prostate cancer, the leading incidence of cancer in American males, is a disease in which treatment of nonlocalized tumors remains largely unsuccessful. These cancers lose expression of an arginine synthesis enzyme, argininosuccinate synthetase (ASS), and are susceptible to arginine deprivation by arginine deiminase (ADI). We show CWR22Rv1 prostate cancer cells are susceptible to ADI in a caspase-independent manner in vitro and in a xenograft model in vivo. We demonstrate that single amino acid deprivation by ADI is able to trigger autophagy. Inhibition of autophagy by chloroquine and siRNA enhances and accelerates ADI-induced cell death, suggesting that autophagy is a protective response to ADI, at least in the early phases. In addition, the co-administration of docetaxel, a caspase-dependent chemotherapy, with ADI inhibits tumor growth in vivo. Thus, targeting multiple cell death pathways, either through autophagy modulation or non-canonical apoptosis, may find expanded use as adjuvant chemotherapies, providing additional avenues for cancer treatment. Reprinted with permission from Autophagy. Note: Regrettably, we could not obtain permission to quote the following report.

3.9 Wheatley DN, Campbell E, Lai PBS, and Cheng PNM. 2005. A rational approach to the systemic treatment of cancer involving medium-term depletion of arginine. Review Article. Gene Therapy and Molecular Biology, Mar; 9: 33e40. 3.9.1 Synopsis This report details the ability of arginine catabolizing enzymes not only to inhibit tumor cell proliferation, but to kill them. The authors tell us that selectivity of action is based on the inability of many tumour cells to

Tumor starvation by L-arginine deprivation

111

circumvent arginine deprivation by utilizing (recycling) various precursors available through the urea cycle. While this may confer an immediate opportunity to treat melanomas and hepatocellular carcinomas in particular, in vitro treatment could customized so that even those tumour cell lines with intact urea cycles can be targeted, making the protocol more generally applicable. Because there are now convincing in vitro studies of the efficacy of arginine degrading enzymes, this treatment has moved on into clinical and veterinary trials. There are now encouraging initial findings suggesting that treatment could be effective with many tumour types, from leukemias to melanomas. Furthermore, arginine deprivation protocols can “stage” tumour cells for combination therapy where cells have not been killed outright by deprivation. This is also selective because deprived normal cells will have become quiescent but soon recover on restitution of the missing nutrient, whereas tumour cells in cycle can be hit by low doses of cycledependent cytotoxic drugs. For details of this study we suggest that you go to: https://www.gtmb. org/volumes/Vol9/05._Wheatley_et_al%2C_33-40.pdf; accessed 9.9.20.

CHAPTER 4

L-canavanine deprives tumors of L-arginine

4.1 Introduction The studies in the previous section detailed tumor cell starvation by the action of certain enzymes that degrade arginine. One in particular, arginine deiminase (ADI), is a microbial enzyme derived from mycoplasma. In the pegylated form, ADI-PEG20 (Polaris Pharmaceuticals), it is shown experimentally and clinically to result in tumor depletion of L-arginine of value particularly in connection with auxotrophs. This section addresses a different approach, tumor cell starvation by the inhibition of L-arginine uptake. That inhibition is effected by the nonprotein amino acid, L-canavanine. L-canavanine is shown to inhibit absorption of L-arginine by • pancreatic cancer cells, • breast cancer cells, and • cancerous plasma cells in multiple myeloma. Canavanine is an antimetabolite of L-arginine that is stored by many leguminous plants where it apparently serves for defense especially as protection of their seeds. It is naturally abundant in commercially available alfalfa (Medicago sativa (L)) sprouts and in 10 varieties of the seeds, and it is also available as a dietary supplement in the form of leaf extract that has a lesser concentration of L-canavanine. At the end of this section, the three reports (numbers followed by “a”) concern L-canavanine administration to enhance the effectiveness of chemotherapy or radiation treatment rather than just to displace L-arginine. Finally, we are cautioned that excessive ingestion of L-canavanine may compromise immune function, and can even lead to lupus. Several references to that effect are given.

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression ISBN 978-0-12-824013-7 https://doi.org/10.1016/B978-0-12-824013-7.00006-7

© 2021 Elsevier Inc. All rights reserved.

113

114

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

4.2 Rosenthal GA and Nkomo P. 2000. The natural abundance of L-canavanine, an active anticancer agent in alfalfa, medicago sativa (L.). Pharmacological Biology, 38(1): 1e6. DOI: https://doi.org/10.1076/ 1388-0209(200001)3811-BFT001. 4.2.1 Abstract L-Canavanine, a potentially toxic antimetabolite of L-arginine that is stored by many leguminous plants, has demonstrative antineoplastic activity against a number of animal-bearing carcinomas and cancer cell lines. This investigation evaluated the natural abundance of this anti-cancer compound in commercially available sprouts, and in ten varieties of the seed of alfalfa, Medicago Sativa (L.). Canavanine abundance in commercially grown sprouts varied according to the source; the young plant stored appreciable canavanine that ranged from 1.3 to 2.4% of the dry matter. Alfalfa seeds were also rich in this nonprotein amino acid as the canavanine content varied from 1.4 to 1.8% of the dry matter. On average, the tested seeds contained 1.54  0.03% canavanine. Alfalfa seed canavanine content was comparable to the levels found in the seeds of representative members of the genus Canavalia, which are amongst the more abundance sources of this antimetabolite. Reprinted with permission of Pharmacological Biology.

4.3 Rosenthal GA. 1977. The biological effects and mode of action of L-canavanine, a structural analogue of L-arginine. The Quarterly Review of Biology, Jun; 52(2): 155e178. DOI: https://doi.org/10.1086/409853. 4.3.1 Introduction Many of the 200 or so non-protein amino acids synthesized by higher plants are related structurally to the constituents of common proteins. L-Canavanine, the guanidinooxy structural analogue of L-arginine, is representative of this group. It has provided valuable insight into the biological effects and the mode of action of non-protein amino acids which act as analogues of the protein amino acids. The arginyl-tRNA synthetases of numerous canavanine-free species charge canavanine, and canavanine is subsequently incorporated into the nascent polypeptide chain. Production of canavanine-containing proteins ultimately can disrupt critical reactions of RNA and DNA metabolism as well as protein synthesis. Canavanine also

L-canavanine deprives tumors of L-arginine

115

affects regulatory and catalytic reactions of arginine metabolism, arginine uptake, formation of structural components, and other cellular processes. In these ways, canavanine alters essential biochemical reactions and becomes a potent antimetabolite of arginine in a wide spectrum of species. These deleterious properties of canavanine render it a highly toxic secondary plant constituent that probably functions as an allelochemic agent that deters the feeding activity of phytophagous insects and other herbivores. Reprinted with permission of The Quarterly Review of Biology.

4.4 Jacobi B, Stroeher L, Leuchtner N, Echchannaoui H, Alexander Desuki A, Kuerzer L, Habermeier A, Antunes E, Amann E, John Bomalaski J, Ellen C, Matthias Theobald M, and Munder M. 2015. Interfering with arginine metabolism as a new treatment strategy for Multiple Myeloma. Blood, 126(23): 3005. DOI: https://doi.org/10.1182/blood. V126.23.3005.3005. 4.4.1 Abstract Starvation of tumor cells from the amino acid arginine has recently gained particular interest because of the downregulation of the rate-limiting enzyme argininosuccinate synthethase 1 (ASS1) in various cancer entities. ASS1-deficient cells cannot resynthesize arginine from citrulline and are therefore considered arginine auxotrophic. The arginine depleting enzyme arginine deiminase (ADI-PEG20, Polaris Pharmaceuticals) is currently tested in phase I-III clinical trials for different arginine auxotrophic cancers. The natural arginine analogue canavanine can compete with arginine for arginyl-tRNA-binding sites and consequently be incorporated into nascent proteins instead of arginine. Canavanine could therefore potentially further disturb intracellular protein homeostasis, especially under arginine deprivation. The sensitivity of myeloma cells towards arginine depletion strategies has not been analyzed so far.

4.4.2 Methods Human myeloma cell lines and CD138-sorted primary human myeloma cells from patient bone marrow were screened for ASS1 expression by western blotting (WB). The cells were cultured in arginine free medium and assessed for proliferation and metabolic activity (CFSE/MTT assays), apoptosis

116

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

(caspase-3 cleavage) and cell death (annexinV/propidium iodide). Canavanine was supplied in both arginine-sufficient and -deficient conditions. The level of intracellular protein stress was determined by WB and/or flow cytometry analysis for ubiquitinated proteins, phosphorylated eukaryotic initiation factor 2a (peIF2a) and the spliced isoform of the X-Box binding protein 1 (Xbp1s). Repetitive ADI-PEG20  canavanine application i.p. were tested in vivo in an U266 myeloma xenograft model in NOD/SCID/IL2Rcg-/- (NSG) mice. Arginine and canavanine levels in plasma were determined by HPLC. Tumor growth was measured, mice were assessed for survival, weight and side effects. Tumor tissues were analyzed for caspase-3 cleavage and Ki67 expression by immunehistochemistry.

4.4.3 Results Five of 6 myeloma cell lines were negative for ASS1. Also, ASS1 was either not or only weakly expressed in the majority of primary CD138þ myeloma patient samples. Arginine starvation induced an arrest of cell proliferation and/or metabolic activity of primary myeloma cells and myeloma cell lines after 18-24 h. Addition of citrulline could only rescue ASS1 positive myeloma cells due to the intracellular resynthesis of arginine. Arginine starvation alone led to delayed induction of apoptosis (e.g. 35% cell death of NCI-H929 cells after 72 h of treatment). Addition of 100 mM canavanine strongly increased cell death specifically in the context of arginine deficiency (e.g. cell death in NCI-H929 cells: 87% after 24 h, 100% after 48h) while it was non-toxic and had no effect on cell viability under physiological arginine conditions. Co-application of canavanine induced ubiquitination of cellular proteins and led to the prolongation of a fatal unfolded protein response (UPR) as measured by markedly elevated Xbp1s levels. Prolonged UPR ultimately led to the induction of apoptosis as reflected by annexin V binding and caspase-3 cleavage. In an U266 myeloma NSG xenograft model, systemic arginine depletion by ADI-PEG20 suppressed tumor growth in vivo and significantly prolonged median survival of mice when compared with the control group (223 vs. 153 days). Canavanine treatment alone had no influence on viability (130 days). However, the combination of ADI-PEG20 and canavanine demonstrated the longest median survival (277days). Histological examination of explanted tumors showed the highest rates of caspase-3 cleavage in the ADI-PEG20/canavanine group.

L-canavanine deprives tumors of L-arginine

117

4.4.4 Conclusion Myeloma cells are mostly arginine auxotrophic and can be selectively targeted by arginine starvation. Combination of arginine depletion with the arginine analogue canavanine leads to a highly efficient and specific tumor cell eradication and should be further optimized in multiple myeloma preclinical models. Reprinted with permission from Blood.

4.5 Windschmitt J, Jacobi B, Bülbül Y, Sester L, Tappe J, Hiebel C, Behl C, Theobald M, and Munder M. 2018. Arginine depletion in combination with Canavanine supplementation induces massive cell death in Myeloma cells by interfering with their protein metabolism and bypassing potential rescue mechanisms. Blood, 132(Supplement 1): 3205. DOI: https://doi.org/10.1182/blood-2018-99-113396. 4.5.1 Introduction Although the therapeutic armamentarium against multiple myeloma has tremendously increased in recent years, it still remains an incurable disease. A highly promising novel anti-tumoral treatment strategy is to target specific non-redundant metabolic achilles heels of individual cancer entities. The semi-essential amino acid arginine can be synthesized from citrulline in most physiological tissues due to expression of the rate-limiting enzyme argininosuccinate synthetase 1 (ASS1). Various tumor entities do not express ASS1, therefore depend on the exogenous availability of arginine and pharmacological approaches to systemically deplete arginine are in phase IIII clinical development for such arginine-auxotrophic cancers. Cell death induction by arginine depletion can be dramatically enhanced by co-application of the arginine analogue canavanine. Canavanine can be used by the respective aminoacyl tRNA synthetase instead of arginine during protein translation and this leads to a highly toxic intracellular accumulation of misfolded proteins. In preliminary work we have seen that myeloma cells are largely arginine-auxotrophic and can be killed by arginine depletion and canavanine supplementation within hours, while ASS1 expressing cells are completely protected by their endogenous arginine rescue capability. Encouraging results of tumor control have already been seen in a murine myeloma model.

118

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

4.5.2 Methods Human myeloma cell lines (NCI-H929_A2 and FD50 developed in our laboratory) were cultured and treated in RPMI-1640 medium with or without arginine. Protein levels were determinded by western blot analysis. Cell viability was measured by propidium iodide staining and flow cytometry analysis. RNA quantification was done by qRT-PCR. For autophagosome and aggresome quantification we used immunofluorescence staining (IF) and laser scanning microscopy (LSM).

4.5.3 Results Arginine depletion and canavanine supplementation led to misfolded protein accumulation which was followed by massive apoptotic cell death. Both processes were further enhanced by co-treatment with the proteasome inhibitor bortezomib, indicated by an increase in intracellular polyubiquitinated proteins as well as higher cleaved caspase 3 levels and propidium-iodide positive cells after only 8-12 h in both tested cell lines. Unexpectedly, the endoplasmic reticulum (ER)-stress response was activated only very moderately. Expression of CHOP, a pro-apoptotic transcription factor that is highly translated under toxic ER stress, was not altered compared to control conditions. Tunicamycin-mediated induction of enhanced ER stress significantly improved the viability of arginine-starved and canavanine treated cells. This suggests that protein accumulation mainly takes place in the cytoplasm rather than the ER and tunicamycin might alleviate cell death by reduction of total protein translation. Despite severe arginine deficiency and induction of misfolded protein stress, the cells were not able to respond by an adequate upregulation of macroautophagy, as determined by an altered LC3 metabolism. The autophagic flux was significantly reduced compared to control conditions after 4-8 h of treatment. There was a strong induction of BAG3 and p62 proteins, which are both associated with chaperone-assisted autophagy as well as aggresome formation and are normally cleared via macroautophagy. Cytoplasmic aggresome formation was not detectable until onset of apoptosis. Also, no relevant modulation of phosphorylation of the autophagy inducer mTORC and the downstream kinase p70S6K1 was noted upon arginine depletion and canavanine co-treatment. Finally, ER stress induction via tunicamycin did not improve autophagic protein turnover, as determined by IF staining, LSM and western blot.

L-canavanine deprives tumors of L-arginine

119

4.5.4 Conclusions Arginine starvation in combination with canavanine supplementation induces fast and highly efficient cell death in arginine-auxotrophic myeloma cells. This novel strategy interferes with myeloma cellular metabolism by induction of misfolded protein accumulation. A relevant upregulation of potentially protective cellular strategies like ER stress responses, aggresome formation and autophagy are either not detectable or they remain insufficient. We hypothesize that our novel metabolic anti-tumor strategy is either too potent or too fast for the tumor cells to cope with its consequences. Reprinted with permission from Blood.

4.6 Rosenthal GA. 1998. L-Canavanine: A potential chemotherapeutic agent for human pancreatic cancer. (Ph.D. Thesis, 1997, University of Kentucky). Laboratory of Biochemical Ecology. University of Kentucky, Lexington, KY 40506. Published online: https://www.uky.edu/wgarose/cancerrev.htm; accessed 11/5/19. Abstract also in Pharmaceutical Biology, Sept; 36(3); 194e201. DOI: https://doi.org/10. 1076/phbi. 36.3.194.6340. 4.6.1 Abstract L-Canavanine, the principal nonprotein amino acid of certain leguminous plants, is a potent L-arginine antimetabolite. This natural product has demonstrative antineoplastic activity against a number of human cancers. Recent studies with MIAPaCa-2 and CFPAC have established canavanine’s potential anticancer potential against these human pancreatic adenocarcinomas. Canavanine has promise as a lead compound in the development of a chemotherapeutic agent for the treatment of human pancreatic carcinoma, but it has not been adequately investigated. Greater study of canavanine and its derivatives is needed to fully realize the experimental and therapeutic value of this naturally-occurring non-protein amino acid, and to obtain a chemotherapeutic agent of clinical value in treating human carcinomas.

4.6.2 Introduction L-Canavanine is a potent arginine antimetabolite that bears strong structural analogy to its protein amino acid counterpart:

120

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

O

O O

HO

N H

NH2 L-CANAVANINE

NH

NH2

H N

HO NH2

H

NH2

NH2

L-ARGININE

Replacement of the terminal methylene group of arginine with oxygen produces this nonprotein amino acid distinguished by a guanidinooxy moiety that has a pKa value of 7.04 and an isoelectric point near neutrality (Boyar and Marsh, 1982). The pKa of 10.48 for the guanidino group of L-arginine produces a much more basic amino acid with Io ¼ 12.48 (Greenstein and Winitz, 1963). Therefore, at physiological conditions, arginine is essentially fully protonated whereas canavanine is not. As a subtle structural mimic of L-arginine, canavanine can function in all enzymic reactions for which arginine is a substrate (Rosenthal, 1977). Therefore, canavanine potentially can inhibit any enzyme-directed reaction employing arginine as the preferred substrate. Arguably, canavanine’s most adverse effect results from its activation and aminoacylation to the cognate tRNAArg by the arginyl-tRNA synthetase of canavanine-sensitive organisms (Allende and Allende, 1964; Mitra and Mehler, 1967). Incorporated into the nascent polypeptide chain, the decreased basicity of canavanine relative to arginine can affect residue interaction and thereby disrupt the tertiary and/or quaternary interactions essential for establishing the requisite three dimensional conformation of a given protein.

4.6.3 Biochemical basis for canavanine antimetabolic properties To determine the biochemical and biological consequence of canavanyl protein formation, a number of canavanine-containing proteins, obtained from various insect sources, were examined in detail. In the initial study, canavanine was provided to gravid females of the locust, Locusta migratoria migratorioides [Orthoptera], whose ovarian mass had been removed surgically on the day of adult emergence. This surgical procedure resulted in a pronounced accumulation of vitellogenin, an important hemolymph storage protein. Analysis of canavanyl vitellogenin, purified from hemolymph of these canavanine-treated

L-canavanine deprives tumors of L-arginine

121

locusts, shows that on average 18 of the 200 arginyl residues of vitellogenin or about 1 in 225 amino acids are replaced by canavanine. This modest level of replacement is nevertheless sufficient to elicit a dramatic alteration in protein structure that is shown most effectively by electrophoretic analysis. The altered fragmentation pattern of canavanyl vitellogenin, relative to the native macromolecule, undoubtedly reflects the profound change in vitellogenin structure resulting from canavanine incorporation (Rosenthal, Lambert, and Hoffmann, 1989a). These disparate forms of vitellogenin were also treated with chemicals able to react with surface-exposed amino acid residues to form novel amino acids. For example, treatment with cyanate carbamylates reactive lysyl residues and converts them to homocitrullyl. About one quarter of the lysyl residues of native vitellogenin do not react with cyanate; presumably, these are inaccessible and buried within the interior of the macromolecule. These residues are readily carbamylated in canavanyl vitellogenin. A similar chemical approach establishes that nearly twice as many surfaceexposed tyrosine residues are acetylated as compared to the native protein. These experiments establish that canavanine incorporation into vitellogenin alters the three dimensional conformation of the protein, a property essential for normal function (Rosenthal, Lambert, and Hoffmann, 1989a). Microbial infection or mechanical injury to larvae of the fly, Phormia Terranovae [Diptera] induces a family of protective, antibacterial proteins known trivially as the diptericins. (Keppi, Zachary, Robertson et al., 1986). If canavanine is provided at the time of mechanically injury, it is incorporated into all of these protective proteins (Rosenthal, Reichhart, and Hoffmann, 1989b). This assimilation causes a total loss of detectable antibacterial activity for nearly all the diptericins; only a single diptericind diptericin A retains demonstrable biological activity. This insectan investigation provides compelling experimental evidence that canavanine incorporation into a protein can impair protein function. Canavanine-mediated impairment in function was also demonstrated with a catalytic protein, lysozyme (EC 3.2.1.17), that is induced by injection of fragments of the cell wall of Micrococcus lutea into larvae of the tobacco hornworm, Manduca sexta [Sphingidae]. If these larvae are provided canavanine when challenged, it is readily incorporated into de novosynthesized lysozyme. The ratio of canavanine to arginine in this aberrant

122

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

lysozyme is 1:3.8; assay of canavanyl lysozyme discloses a 48% loss in catalytic activity. This study provided the first demonstration of the ability of canavanine, through aberrant protein formation, to adversely affect the catalytic activity of an enzyme. The importance of aberrant protein formation in the expression of canavanine’s antimetabolic effect is also supported convincingly by a radically different experimental approach that determines how frequently canavanine replaces arginine in the proteins of various insects, i.e. the substitution error frequency (SEF). Manduca sexta larvae incorporate about 3.5% of the administered radiolabeled canavanine into hemolymphic proteins after 24 h. This results in a SEF that varies according to the insectan tissue but is about I in 2 for proteins of the larval body wall and musculature (Rosenthal et al., 1987). The bruchid beetle, Caryedes brasiliensis [Coleoptera], an inhabitant of the neotropic forests of Costa Rica, develops from larva to adult in the canavanine-laden seeds of Dioclea megacarpa [Fabaceae]. The woody pericarp of D. megacarpa houses seeds that can contain as much as 13% canavanine by dry weight (Rosenthal, 1983). The weevil, Sternechus tuberculatus [Curculionidae], oviposits on the pericarp of Canavalia brasiliensis [Fabaceae]; larvae, foraging within the fruit, are sustained by seeds that also store appreciable canavanine. These two seed predators are examples of insects that are adapted biochemically to canavanine (Bleiler, Rosenthal, and Janzen, 1988) and can therefore tolerate, even flourish with this normally poisonous natural product (Rosenthal, 1983). The SEF for C. brasiliensis is 1 in 365 while that of S. tuberculatus is 1 in 500-1,000. The tobacco budworm, Heliothis virescens [Noctuidae], does not consume canavanine-containing plants; however, it is naturally resistant to this potentially toxic allelochemical (Berge, Rosenthal, and Dahlman, 1986; Melangeli, Rosenthal, and Dahlman, 1997). This aggressive generalist herbivore exhibits a SEF of 1 in 65. Thus, canavanineadapted and canavanine-resistant insects minimize or avoid canavanyl protein formation while canavanine-sensitive insects readily incorporated this arginine antagonist.

4.6.4 Article of human diet At present, few canavanine-containing seeds are part of the human diet, but this is changing as the worldwide demand for reasonably priced, high quality protein increases. Jack bean seeds, Canavalia ensiformis (L.) DC.

L-canavanine deprives tumors of L-arginine

123

[Fabaceael, which contain about 2.5% canavanine by dry weight, and sword bean, Canavalia gladiata [Fabaceae], storing about 1.4% canavanine by dry weight (Rosenthal and Nkomo, 1997), are important table legumes, particularly in Asia and parts of the tropics and Africa. In North America, the most heavily consumed canavanine-containing plant is alfalfa, Medicago sativa [Fabaceae]. Canavanine is the preponderant nonprotein amino acid of the seed where it accounts for 1.5% of the dry matter; the sprout is also canavanine-rich since it can accumulate as much as 2.4% canavanine by dry weight (Rosenthal and Nkomo, 1997).

4.6.5 Canavanine antineoplastic activity In 1958, Kruse and McCoy reported that canavanine competed with arginine in meeting the growth requirements of Walker carcinosarcoma 256 cells. In a subsequent study, Kruse et al. (1959) demonstrated that canavanine was incorporated into the proteins of these cancer cells, and that the diminution in the amount of arginine in the protein hydrolysate equaled the canavanine content. This report established for the first time that canavanine specifically replaced arginine in de novo-synthesized tumor proteins. Schachtele and Rogers (1965) employed the same experimental approach to demonstrate the incorporation of canavanine into the proteins of Escherichia coli. These initial observations were confirmed by experiments conducted with M. sexta larvae that were injected with L-[guanidinooxy-14C]canavanine (Dahlman and Rosenthal, 1976). Enzymatic degradation of the radiolabeled canavanine, isolated from the insectan hydrolysate, demonstrated unequivocally that the hydrolysate 14carbon was derived from radiolabeled canavanine. Numerous descriptive studies have documented that exposure of a particular organism to canavanine adversely affected a basic property or functional parameter of one or more of its enzymes. For example, the normally soluble b-galactosidase of E. coli exhibited diminished activity, and sedimented readily with the 10,000 xg pellet after exposing the bacterium to canavanine (Prouty, Karnovsky, and Goldberg, 1972; Rosenthal, 1977). Several studies of canavanine’s antineoplastic activity have been conducted. Naha, Silcock, and Fellows (1980) demonstrated that canavanine selectively inhibited DNA replication in epithelial monkey kidney cells

124

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

possessing a transformed phenotype, as compared to their counterpart that remained “contact-inhibited” with a temperature change from 33 to 39.5 C. A detailed study of canavanine’s antineoplastic activity was conducted with mice bearing L1210 leukemic cells (Green, Brooks, Mendelsohn et al., 1980). These workers reported that DNA synthesis was reduced to only 9% of the control level, as assessed by [3H]thymidine incorporation, after 12-hourly i.p. injections of 20 mg canavanine per injection. A dramatic attenuation in DNA synthesis of 86% of the control level was also achieved when the infected mice were infused s.c. with canavanine continuously for 1 d at a rate of 20 mg h-1 after an initial i.p. injection of 20 mg canavanine. At an optimal dose of 18 g kg-1, the median life span of the cancerous mice increased by 44%. This investigation was of great significance because it demonstrated that canavanine could mediate its toxic effect not only at the level of protein function, but also through its ability to disrupt DNA replication. In an important follow-up study, Green and Ward (1983) reported that canavanine enhanced significantly the efficacy of g-irradiation of cultured HT-29 cells, a human tumor cell line. The lethal effect of this radiation was augmented both when canavanine was provided prior to as well as after g -irradiation. These workers provided convincing experimental evidence for their contention that canavanine’s lethal effect was manifested preferentially in rapidly proliferating cellsda property often essential for chemotherapeutic efficacy. The experimental efforts of Green and his collaborators did not distinguish between the possibility that canavanine affected nucleic acid turnover directly as compared to its acting by affecting the activity of one or more proteins essential to maintaining DNA replication. Canavanine also affected the growth of a rat colonic carcinoma in male Fischer rats (Thomas, Rosenthal, Gold et al., 1986). This tumor, which became palpable in 8-10 d, had a doubling time of 3 to 4 d. Canavanine, administered by s.c. injection into the flank opposite the tumor site, was provided initially after the tumor attained a volume of 500-1,000 m3 (Figure 1). Providing 2.0 g kg-1 canavanine for 5 d produce a tumor vs. control of 23%; after 9 d, this value fell to 14%. Canavanine’s efficacy was enhanced when the dose was increased to 3.0 g kg-1. At this dose, the percentage of regression was e13% after 5 d, and e8% after 9 d. These negative values reflect tumor regression. The loss in tumor volume,

L-canavanine deprives tumors of L-arginine

125

expressed as percentage of regression, was 22% for the 3 g kg-1 daily for 5 d, and 60% in the 3 g kg-1 daily for 9 d treatment groups. These promising findings with canavanine had the drawback that the treated rats lost weight. Canavanine’s cumulative toxicity resulted in about a 15% diminution in body weight after 5 treatment days. This finding led to an examination of the relationship of caloric deprivation to tumor growth reduction, and established that canavanine-directed curtailment of tumor growth was not caused by reduced food intake. Most importantly, canavanine-dependent weight loss was fully reversible. This investigation instigated efforts to develop canavanine derivatives with an enhanced therapeutic index while diminishing body weight loss.

4.6.6 Canavanine cytotoxic effect on human pancreatic cells Toxicological study of canavanine metabolism in the male Fischer rat revealed that L-[guanidinooxy-14C]canavanine was assimilated preferentially by proteins of the pancreas. Radiolabeled canavanine incorporation into these proteins was 10-times that of liver, brain, and muscle tissues, and 5-times that of the proteins of most other body organs (Thomas and Rosenthal, 1987a). Given the pronounced assimilation of canavanine into pancreatic proteins, we examined the effect of canavanine on the growth of MIAPaCa-2, a human pancreatic adenocarcinorna cell line (Swaffar, Ang, Desai et al., 1994). When these cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) containing 0.4 mM arginine, the 50% inhibitory concentration (IC50) for canavanine was 2 mM. Most importantly, the IC50 value fell precipitously to 10.0 mM when the arginine content of the culture medium was reduced to 0.4 mM. This study was important because it demonstrated canavanine’s marked cytotoxic activity against a human pancreatic cancer cell line. Moreover, it provided an excellent experimental system for evaluating novel canavanine derivatives on a small scale prior to conducting whole animal studies. Swaffar, Ang, Desai et al. (1994) also demonstrated that canavaninemediated inhibition of MIAPaCa-2 cell growth was reversible by arginine up to 12 hr post treatment, but then became irreversible. This finding, while not presently explicable, was important because it demonstrated that conditions existed where canavanine’s anti-cancer effect would not be reversed.

126

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

4.6.7 Combination therapy Having established canavanine’s antineoplastic activity against MIAPaCa-2 cells, Swaffar, Choo, Desai et al. (1995) asked the salient question of how canavanine functioned in combination with 5-fluorouracil (5-FU), the preferred drug presently utilized for treating pancreatic carcinoma. In this companion study, canavanine’s efficacy as a chemotherapeutic agent was assessed in combination with 5-FU. At a fixed molar ratio of 1:1, over the range of 0.06 to 1.0 mM, both drugs exhibited greater cytotoxic effects against MIAPaCa-2 cells relative to single administration of these drugs. The IC50 values for canavanine and 5-FU were 0.060 and 0.363, respectively; in combination, the IC50 value fell to 0.021. As part of this investigation, the interaction of canavanine and 5-FU was also assessed with the colonic carcinoma of adult male Fischer rats described above (Swaffar, Choo, Desai et al., 1995). Providing canavanine, either at 1.0 g kg-1 or 2.0 g kg-1 daily for 5 d in combination with 5-FU increased significantly the anti-tumor activity of either drug alone (Figure 2). These investigations demonstrated the value of employing canavanine in combination with drugs presently employed in treating pancreatic and colonic cancer. Thus, the development of a canavanine derivative as an effective drug is not limited only to its independent use, but also may prove more valuable when provided in combination with an established anticancer drug.

4.6.8 Canavanine accumulative toxicity Analysis of canavanine catabolism in the adult rat demonstrated that hepatic argjnase (EC 3.4.1.5) fostered the hydrolysis of L-canavanine to yield L-canaline and urea, this reaction pathway was the principal basis for canavanine catabolism in this mammal (Thomas and Rosenthal, 1987b). O

O NH2

O HO

N H

NH2 L-CANAVANINE

O

Arginase

NH2

O

HO NH2

H

NH2 L-CANALINE

H 2N

NH2 UREA

Thus, it is reasonable to propose that administration of L-canavanine to a human would result in the formation of L-canaline, a highly toxic nonprotein amino acid that is a powerful inhibitor of pyridoxal phosphatedependent enzymes. Rahiala, Kekomaki, Janne et al. (1971) were the

L-canavanine deprives tumors of L-arginine

127

first to recognize that a direct reaction occurred between canaline and the vitamin B6 moiety of an enzyme by postulating: “...that canaline probably inhibits pyridoxal phosphate-containing enzymes by its nonenzymic, irreversible and stoichiometric binding with pyridoxal phosphate.” The ability of canaline to inactivate an enzyme by forming a stable, covalently-linked oxime was demonstrated’ directly with L-[U-14C] canaline which reacted with ornithine aminotransferase (EC 2.6.1.13) of larval Manduca sexta to yield an enzyme-bound, radiolabeled canalinepyridoxal phosphate oxime (Rosenthal and Dahlman, 1990). In addition, canaline’s facile ability to form oximes means that canaline can scavenge such essential 2-oxo-containing metabolites as pyruvate, oxaloacetate, and 2-oxoglutarate to deplete carboxylic acid reserves and carbon skeleton required for amino acid synthesis. Finally, canaline also possesses significant antineoplastic and antiproliferative properties (Rosenthal, 1998).

4.6.9 Canavanine derivatives as chemotherapeutic agents The intrinsic toxicity of canavanine would be decreased significantly if it failed to function as a substrate for hepatic degradation via the action of arginine. Production of a simple ester of canavanine, such as methyl-Lcanavanine, provided a derivative that was not an effective substrate for rat arginase and therefore could not elicit the adverse biological effects caused by canaline. In addition, this ester possessed enhanced hydrophobicity relative to canavanine, and might therefore enjoy enhanced cellular uptake. Intracellular esterases should release the parent compound through hydrolysis and thereby increase canavanine’s bioavailability. To test this hypothesis, the methy, ethyl, n-propyl, isopropyl, n-butyl, and n-octyl esters of L-canavanine were synthesized and evaluated for cytotoxicity against MIAPaCa-2 cells (NaPhuket, Trifonov, Crooks et al., 1997). While the methyl, ethyl n-propyl, and isopropyl esters of canavanine exhibited slightly improved growth-inhibitory activity against MIAPaCa-2 cells, the n-butyl and n-octyl esters of canavanine dramatically attenuated cellular growth (Figure 3). This conclusion was firmly confirmed by bioevaluation of the growth inhibiting properties of these canavanine derivatives in terminal instar M. sexta larvae (Rosenthal and Nkomo, 1997). The increased potency of these latter esters may reflect their enhanced lipophilicity and greater membrane penetrational properties. None of the

128

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

free alcohols, except n-octanol, possessed significant growth-inhibiting activity. Only at concentrations above 2.5 mM did octanol exhibit significant activity against MIAPaCa-2 cells.

4.6.10 Conclusions The experimental efforts outlined in this review detail studies that demonstrate the significant antineoplastic activity of canavanine. Recently completed experiments probing 14C-radiolabeled canavanine uptake (0.26% of the administered dose) by MIAPaCa-2 cells revealed that 70% of the cellular canavanine was found in the proteins, the remainder was accounted for in the cytosol; 95% of the 14carbon found in proteins of the treated cells was canavanine (NaPhuket, 1997). In contrast, only 58% of the 14carbon of the cytosol was canavanine. The author raises for thought the interesting possibility that canavanine’s antineoplastic activity might arise from the formation of structurally aberrant, dysfunctional, canavanine-containing proteins by the cancer cells and that they may be unique to these cells. The findings outlined in this review support the contention that canavanine has marked potential as a lead compound in the development of a chemotherapeutic agent for the treatment of human pancreatic carcinoma. Particularly noteworthy are certain ester derivatives of canavanine, which might provide an efficacious drug capable of eliciting little if any body weight loss while enhancing the therapeutic index. Esterification of the carboxyl group of canavanine with longer-chained alcohols such as butanol and octanol represent structural modification of canavanine that augments significantly the growth-inhibiting properties of the parent compound against MIAPaCa-2 cells. Further study of canavanine and its derivatives could lead to chemotherapeutic agents of clinical value in treating human carcinomas. Additional experimental effort is warranted to realize the experimental and therapeutic value of this unusual amino acid antimetabolite of higher plants.

4.6.11 References Allende CC and Allende JE. 1964. Purification and substrate specificity of arginylribonucleic acid synthetase from rat liver. Journal of Biological Chemistry, Apr; 239: 1102e1106. PMID: 14165914. Berge MA, Rosenthal GA, and Dahlman DL. 1986. Tobacco budworm, Heliothis virescens [Noctuidae] resistance to L-canavanine, a protective allelochemical. Pesticide Biochemistry and Physiology, 25: 319e326.

L-canavanine deprives tumors of L-arginine

129

Bleiler J, Rosenthal GA, and Janzen DH. 1988. Biochemical ecology of canavanine-eating seed predators. Ecology, Apr; 69(2): 427e433. DOI: https://doi.org/10.2307/1940441. Boyar A and Marsh RE. 1982. L-Canavanine, a paradigm for the structures of substituted guanidines. Journal of the American Chemical Society, Apr 1; 104(7): 1995e1998. DOI: https://doi.org/10.1021/ja00371a033. Dahlman DL and Rosenthal GA. 1976. Further studies on the effect of L-canavanine on the tobacco hornworm, Manduca sexta (L.) (Sphingidae). Journal of Insect Physiology, 22(2): 265e271. DOI: https://doi.org/10.1016/0022-1910(76)90035-4. Green MH, Brooks TL, Mendelsohn J, and Howell SB. 1980. Antitumor activity of L-canavanine against L1210 murine leukemia. Cancer Research, Mar; 40(3): 535e537. PMID: 7471074. Green MH and Ward JF. 1983. Enhancement of human tumor cell killing by L-canavanine in combination with g-radiation. Cancer Research, Sep; 43(9): 4180e4182. Greenstein JP and Winitz M. 1963. Chemistry of the Amino Acids, Vols 1-3, Wiley & Sons, New York. Keppi E, Zachary D, Robertson M, Hoffmann D, and Hoffmann J. 1986. Induced antibacterial proteins in the hemolymph of Phormia terranovae [Diptera]. Purification and possible origin of one protein. Insect Biochemistry, 16: 395e399. Kruse PF Jr and McCoy TA. 1958. The competitive effect of canavanine on utilization of arginine in growth of Walker carcinosarcoma 256 cells in vitro. Cancer Research, Apr; 18(3): 279e282. PMID: 13523592. Kruse PF Jr, White PB, Carter HA, and McCoy TA. 1959. Incorporation of canavanine into protein of Walker carcinosarcoma 256 cells cultured in vitro. Cancer Research, 19: 122e125. Melangeli C, Rosenthal GA, and Dahlman DL. 1997. The biochemical basis for the tolerance of Heliothis virescens to L-canavanine. Proceedings of the National Academy of Sciences USA, Mar 18; 94(6): 2255e2260. DOI: https://doi.org/10.1073/ pnas.94.6.2255. Mitra SK and Mahler AH. 1967. The arginyl transfer ribonucleic acid synthetase of Escherichia coli. Journal of Biological Chemistry, 242: 5490e5494. Naha PM, Silcock JM, and Fellows L. 1980. An experimental model for selective inhibition of proliferating cells by chemotherapeutic agents. Cell Biology International Reports, 4: 155e166. NaPhuket S. 1997. L-Canavanine and its Derivatives as Potential Chemotherapeutic Agents for Pancreatic Cancer: Synthesis, Structure-Activity Relationships and Metabolic Studies. Ph.D. Thesis, University of Kentucky. NaPhuket S, Trifonov LS, Crooks PA, Rosenthal GA, Freeman JW, and Strodel WE. 1979. Synthesis and structure-activity relationships of some antitumor congeners of L-canavanine. Drug Development Research, 40: 325e332. Prouty WF, Karnovsky MJ, and Goldberg AL. 1972. Degradation of abnormal proteins in Escherichia coli. Journal of Biological Chemistry, Feb 10; 250(3): 1112e1122. PMID: 1089651. Rahiala E-L, Kekomaki MK, Janne J, Raina A, and Raiha NCR. 1971. Inhibition of pyridoxal enzymes by L-canaline. Biochimica et biophysica Acta, Feb; 227(2): 337e343. DOI: https://doi.org/10.1016/0005-2744(71)90065-9. Rosenthal GA. 1977. The biological effects and mode of action of L-canavanine, a structural analogue of L-arginine. Quarterly Review of Biology, Jun; 52(2): 155e178. DOI: https:// doi.org/10.1086/409853. Rosenthal GA. 1983. The adaptation of a beetle to a poisonous plant. Scientific American, 249: 164e171. Rosenthal GA and Dahlman DL. 1990. Interaction of L-canaline with ornithine aminotransferase of the tobacco hornworm, Manduca sexta [Sphingidae]. Journal of Biological Chemistry, Jan 15; 265: 868e873.

130

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

Rosenthal GA and Nkomo P. 1997. Unpublished data. Rosenthal GA, Lambert J, and Hoffmann D. 1989a. L-Canavanine incorporation into protein can impair macromolecular function. Journal of Biological Chemistry, Jun 15; 264(17): 9768e9771. Rosenthal GA, Reichhart J-M, and Hoffmann JA. 1989b. L-Canavanine incorporation into vitellogenin and macromolecular conformation. Journal of Biological Chemistry, Aug 15; 264: 13693e13696. Rosenthal GA, Berge MA, Bleiler JA, and Rudd TP. 1987. Avoidance of aberrant protein production and an organism’s ability to utilize or tolerate L-canavanine. Experientia, May 15; 43(5): 558e561. DOI: https://doi.org/10.1007/bf02143585. Schachtele CF and Rogers P. 1965. Canavanine death in Escherichia coli. Journal of Molecular Biology, 34: 843e860. Swaffar DS, Ang CY, Desai PB, and Rosenthal GA. 1994. Inhibition of the growth of human pancreatic cancer cells by the arginine antimetabolite, L-canavanine. Cancer Research, Dec 1; 54(23): 6045e6048. PMID: 7954443. Swaffar DS, Choo YA, Desai PB, Rosenthal GA, Thomas DA, John WJ, and Crooks PA. 1995. Combination therapy with 5-flurouracil and L-canavanine: in vitro and in vivo studies. Anti-Cancer Drugs, Aug; 6(4): 586e593. DOI: https://doi.org/10.1097/ 00001813-199508000-00012. Thomas DA, Rosenthal GA, Gold DV, and Dickey K. 1986. Growth inhibition of a rat colon tumor by L-canavanine. Cancer Research, Jun; 46: 2898e2903. https://cancerres. aacrjournals.org/content/canres/46/6/2898.full.pdf. Thomas DA and Rosenthal GA. 1987a. Toxicity and pharmacokinetics of the nonprotein amino acid L-canavanine in the rat. Toxicology and Applied Pharmacology, Dec; 91(3): 395e405. DOI: https://doi.org/10.1016/0041-008x(87)90061-5. Thomas DA and Rosenthal GA. 1987b. Metabolism of L-[guanidinooxy-14C]-canavanine in the rat. Toxicology and Applied Pharmacology, 91: 406e414.

Reprinted with permission from Rosenthal GA. 1998. Laboratory of Biochemical Ecology. University of Kentucky, Lexington, KY 40506. Note: The figures do not appear on the website. Figure 1. The effect of canavanine on tumor growth in male Fischer 344 rats. The rats were administered canavanine, 2.0 ( ), 3-0 ( ) g kg-1 for 9 d. Control animals ( ) received 0.95% (w/v) NaCl. The standard error bar was omitted if it fell within the area occupied by the data point, n¼5 þ SEM. Figure 2. Evaluation of the combined effect of canavanine and 5-FU on colonic tumor growth in male Fischer rats. Tumor growth was evaluated after 5 daily s.c. injections of: 1.0 g kg-1 canavanine ( ) 2.0 g kg-1, canavanine ( ) 35 mg kg-1 5-FU ( ) 35 mg kg-1 5-FU þ 1.0 g kg-1 canavanine ( ) and 35 mg kg-1 5-FU þ 2.0 g kg-1 canavanine ( ). The control animals ( ) received 0.95% (w/v) NaCl. Each value is the mean (n ¼ 5) þ SEM. Figure 3. Comparison of the IC50 value for canavanine and some of its esters.

L-canavanine deprives tumors of L-arginine

131

4.7 Swaffar DS, Ang CY, Desai PB, and Rosenthal GA. 1994. Inhibition of the growth of human pancreatic cancer by the arginine antimetabolite, L-canavanine. Cancer Research, Dec 1; 54(23): 6045e6048. PMID: 7954443. 4.7.1 Summary In this study, the authors report that L-Canavanine (CAV), the L-2-amino4-guanidinooxy structural analogue of L-arginine (ARG), a potent ARG antagonist which occurs in the jack bean, Canavalia ensiformis, is active against L1210 murine leukemia and a solid colonic tumor in the rat. Their studies show that CAV exhibits a 50% inhibitory concentration of approximately 2 mM against the human pancreatic adenocarcinoma cell line, MIA PaCa-2, when these cells are grown in Dulbecco’s modified Eagle’s medium containing 0.4 mM ARG. When the ARG concentration is reduced to 0.4 microM, the 50% inhibitory concentration for CAV falls precipitously to 0.01 mM. The increased ability of CAV to inhibit MIA PaCa-2 cell growth at the lower ARG concentration is thought due to enhanced CAV competition with ARG for incorporation into newly synthesized cellular proteins. At 0.4 microM ARG, 30 mM CAV almost completely inhibits cell growth by 6 h. In contrast, with 0.4 mM ARG, complete inhibition does not occur until after 48 h. A dramatic reversal of growth inhibition caused by a very high concentration of CAV was observed when cells treated with CAV were replenished with a high concentration of ARG. The authors conclude that CAV could lead to the development of analogues with enhanced activity against human pancreatic cancer.

4.8 Ding Y, Matsukawa Y, OhtaniFujita N, Kato D, Dao S, Fujii T, Naito Y, Yoshikawa T, Sakai T, and Rosenthal GA. 1999. Growth inhibition of A549 human lung adenocarcinoma cells by L-canavanine is associated with p21/WAF1 induction. Japanese Journal of Cancer Research, Jan; 90(1): 69e74. DOI: https://doi.org/10. 1111/j.1349-7006.1999.tb00667.x. 4.8.1 Abstract L-Canavanine (CAV) is a higher plant nonprotein amino acid and a potent L-arginine antimetabolite. CAV can inhibit the proliferation of tumor cells

132

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

in vitro and in vivo, but little is known regarding the molecular mechanisms mediating these effects. We demonstrated that the treatment of human lung adenocarcinoma A549 cells with CAV caused growth inhibition; G1 phase arrest is accompanied by accumulation of an incompletely phosphorylated form of the retinoblastoma protein, whose phosphorylation is necessary for cell cycle progression from G1 to S phase. In addition, CAV induces the expression of p53 and subsequent expression of a cyclin-dependent kinase inhibitor, p21/WAF1. The p53edependent induction of p21/WAF1 and the following dephosphorylation of the retinoblastoma protein by CAV could account for the observed CAV-mediated G1 phase arrest. Reprinted with permission from the Japanese Journal of Cancer Research. Note: We were not granted permission to include the references in this report. They are very comprehensive. If you would like to see them, go to DOI: https://doi.org/10.1111/j.1349-7006.1999.tb00667.x. Note: In the in vitro and in vivo studies cited below, L-canvanine is investigated as a potential adjunct to other forms of cancer treatment.

Canavanine combination: enhancing chemotoxicity or radiation effects Chemo-sensitizer in breast cancerd

4.9a Nurcahyanti ADR and Wink M. 2017. L-Canavanine potentiates cytotoxicity of chemotherapeutic drugs in human breast cancer cells. Anti-cancer Agents in Medicinal Chemistry, 17(2): 206e211. DOI: https://doi. org/10.2174/1871520616666160 223111551; https:// www.ingentaconnect.com/contentone/ben/acamc/ 2017/00000017/00000002/art00007. 4.9a.1 Abstract Due to the high level of argininosuccinate synthase (ASS), a key enzyme for the formation of arginine from citrulline, human breast cancers are often resistant to arginine deprivation therapy. An antimetabolite, Lcanavanine (L-CAV), can be incorporated into proteins in the place of arginine, disturbing protein conformation and leading to cellular death.

L-canavanine deprives tumors of L-arginine

133

Objectives This study was designed to investigate the potential of L-CAV to enhance the toxicity of chemotherapeutic drugs in the human breast cancer cell line MCF-7, and determine the most favorable drug combination to exert synergistic interaction in the presence or absence of arginine in the medium.

Method Combination experiment based on the median-effect principle and massaction law was conducted using constant ratios of cytotoxic agents as developed by Chou (2006).

Results We observed that L-CAV enhanced the toxicity of cisplatin (CIS) and vinblastine (VIN) in MCF-7, even in the presence of L-ARG. On the other hand, L-CAV potentiated the toxicity of doxorubicin (DOX), paclitaxel (PTX), 5- fluoruracil (5-FU), and amphotericin-B (AMP-B) in cells grown in arginine deprived media.

Conclusion We conclude that the combination of L-CAV with CIS or VIN can potentiate the toxicity for breast cancer cells. Thus this report presents a new possibility for treating human breast cancers known to be resistant to arginine deprivation. This initial study requires further investigation in in vivo experiments and exploration of the molecular mechanism of cellular response in human breast cancer.

Reference Chou TC. 2006. Theoretical basis, experimental design, and computerized simulation of synergism and antagonism in drug combination studies. Pharmacological Review, Sep; 58(3): 621e681. DOI: https://doi.org/10.1124/pr.58.3.10.

Reprinted with permission from Anti-cancer Agents in Medicinal Chemistry.

134

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

4.10a Vynnytska-Myronovska B, Bobak Y, Garbe Y, Dittfeld C, Stasyk O, and Kunz-Schughart LA. 2012. Single amino acid arginine starvation efficiently sensitizes cancer cells to canavanine treatment and irradiation. International Journal of Cancer, May 1; 130(9): 2164e2175. DOI: https://doi.org/10.1002/ijc. 26221. 4.10a.1 Abstract Single amino acid arginine deprivation is a promising strategy in modern metabolic anticancer therapy. Its potency to inhibit tumor growth warrants the search for rational chemo- and radio-therapeutic approaches to be co-applied. In this report, we evaluated, for the first time, the efficacy of arginine deprivation as anticancer therapy in three-dimensional (3D) cultures of human tumor cells, and propose a new combinatorial metabolicchemo-radio-treatment regime based on arginine starvation, low doses of arginine natural analog canavanine and irradiation. A sophisticated experimental setup was designed to evaluate the impact of arginine starvation on four human epithelial cancer cell lines in 2D monolayer and 3D spheroid culture. Radioresponse was assessed in colony formation assays and by monitoring spheroid regrowth probability following single dose irradiation using a standardized spheroid-based test platform. Surviving fraction at 2 Gy (SF2Gy) and spheroid control dose50 (SCD50) were calculated as analytical endpoints. Cancer cells in spheroids are much more resistant to arginine starvation than in 2D culture. Spheroid volume stagnated during arginine deprivation, but even after 10 days of starvation, 100% of the spheroids regrew. Combination treatment, however, was remarkably efficient. In particular, pretreatment of cancer cells with the arginine-degrading enzyme arginase combined with or without low concentration of canavanine substantially enhanced cell radioresponse reflected by a loss in spheroid regrowth probability and SCD50 values reduced by a factor of 1.5e3. Our data strongly suggest that arginine withdrawal alone or in combination with canavanine is a promising antitumor strategy with potential to enhance cancer cure by irradiation. Reprinted with permission from International Journal of Cancer.

L-canavanine deprives tumors of L-arginine

135

4.11 Cautionary note A number of sources voice concerns about the safety of L-canavanine in various Applications. We cite here a small sample of the reports that address those concerns. Montanaro A and Bardana EJ Jr. 1991. Dietary amino acid-induced systemic lupus erythematosus. Rheumatic Diseases Clinics of North America, May; 17(2): 323e332. PMID: 1862241. No authors given.* No date given.* Is Canavanine Harmful to Humans? University of Kentucky Webstite. https://www.uky.edu/wgarose/ link109.htm; accessed 10.26.19. Krakauer J, Long Y, Kolbert A, Thanedar S, and Southard J. 2015. Presence of L-canavanine in Hedysarum alpinum* seeds and its potential role in the death of Chris McCandless**. Wilderness and Environmental Medicine, Mar; 26(1): 36e42. DOI: https://doi.org/10.1016/ j.wem.2014.08.014.

Additional information about substances safety is available on the National Library of Medicine, TOXNET website. https://www.nlm.nih.gov/toxnet/ index.html; accessed 4.28.20.

CHAPTER 5

Glucose deprivation and fasting strategies

5.1 Introduction Some cancers play a classic parasite trick. They take advantage of something in the metabolic routine of the host and subvert it for their own purpose because as cancer cells, they need all the energy they can muster to drive their out-of-control growth, and it is driven primarily by glucose. In fact they consume so much glucose that, reportedly, one method for imaging cancer is simply to look for areas of extreme tissue glucose consumption. And, since increased regional glucose metabolism raises local tissue temperature above that of surrounding tissues, it can, in certain instances such as breast cancer, be detected by surface contact thermography [1]. And so it is said that where there is unexpectedly elevated glucose consumption, there is cancer. A University of Colorado Cancer Center study published in the journal Cancer Cell shows that leukemia manages to undermine the ability of normal cells to consume glucose, thus leaving more glucose available to feed its own growth (Ye, Adane, Khan et al., 2018). In the case of leukemia, the cells accomplish this by insidiously engineering a diabetes-like condition that reduces glucose going to normal cells, and as a consequence, there is more glucose available for them. So, they are stealing glucose from normal cells to drive their own growth. The cancer cells strategy here depends on subverting, insulin secretion and utilization. Healthy cells need insulin to enable glucose to be actively transported into the cells. In diabetes, either the pancreas underproduces insulin or insulin receptors are insensitive to it, and so excess glucose lingers in the blood. The current understanding is that leukemia goes about creating similar conditions of glucose buildup in two ways. As shown in the study cited below, first, tumor cells trick fat cells into overproducing a protein called IGFBP1. This protein makes healthy cells Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression ISBN 978-0-12-824013-7 https://doi.org/10.1016/B978-0-12-824013-7.00009-2

© 2021 Elsevier Inc. All rights reserved.

137

138

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

less sensitive to insulin, meaning that when IGFBP1 is high, it takes more insulin to use glucose than it does when IGFBP1 is low. Unless the supply of insulin goes up, high IGFBP1 means that the glucose consumption of healthy cells goes down. (This protein may also be a link in the chain connecting cancer and obesity: The more fat cells, the more IGFBP1, and the more glucose is available to the cancer.) Cancer also has a second strategy that ensures insulin production does not go up to meet the need created by increased IGFBP1. In fact, cancers turn insulin production down. In large part, they do this by altering the gut microbiome that ordinarily plays an important role in controlling glucose production and utilization by normal cells. The study cited below found that the guts of leukemic individuals lacked Bacteroides, a particular kind of bacteria involved in producing short-chain fatty acids that in turn feed the health of cells lining the gut. Without adequate Bacteroides, gut health suffers in ways that specifically help cancer including the loss of hormones called incretins. When postprandial blood glucose rises, the gut releases incretins which reduces blood glucose back into the normal range. But through its action in the gut, leukemia apparently inactivates these incretins allowing blood glucose to remain higher than it ordinarily would be. Leukemia also alters the activity of serotonin which is essential for the production of insulin in the pancreas. And, by attacking serotonin, leukemia reduces insulin production and glucose metabolism. The consequence of reduced insulin secretion and reduced insulin sensitivity is that cancer undercuts insulin availability to healthy cells and less insulin use by healthy cells leaves more glucose for the cancer cells. The Colorado study also details how the investigators were able to reregulate or recalibrate this process to restore glucose regulation and slow the growth of leukemia cells. The leukemia study and the other examples below provide a fascinating look at the disordered machinations of cancers that live in a parallel reality, with a mission different from ours, one that subverts ours. The following publications describe strategies that aim principally to attack tumors by depriving them of glucose. These studies primarily cite leukemia, breast cancer, and pancreatic cancer.

Glucose deprivation and fasting strategies

139

5.1.1 Reference [1] Fried R. 2015 (filing date). Reversible Thermochromatic Liquid Crystal Contact Means for Self-Detection of Subdermal Abnormal Cell Metabolism. United States Patent Application 20150297095.

5.2 Muti P, Quattrin T, Grant BJB, Krogh V, Micheli A, Schünemann HJ, Ram M, Freudenheim, JL, Sieri S, Trevisan M, and Berrino F. 2002. Fasting glucose is a risk factor for breast cancer. A prospective study. Cancer Epidemiology, Biomarkers and Prevention, Nov; 11(11): 1361e1368. PubMed ID: 12433712. 5.2.1 Abstract There is some evidence that glucose and other factors related to glucose metabolism, such as insulin and insulin-like growth-factors (IGFs) may contribute to breast cancer development. The present study analyzed the hypothesis that serum glucose, insulin levels, and IGF-I pattern are associated with breast cancer using a nested case-control study. Between 1987 and 1992, 10,786 women ages 35e69 were recruited in a prospective study in Italy. Women with history of cancer and on hormone therapy were excluded at baseline. At recruitment, blood samples were collected after a 12-h fast between 7:30 and 9:00 a.m. from all of the study participants. After 5.5 years, 144 breast cancer cases were identified among the participants of the cohort. Four matched controls were chosen for each breast cancer case from members of the cohort who did not develop breast cancer during the follow-up period. In premenopausal women, glucose was associated with breast cancer risk: the age, body mass index, and reproductive variable adjusted relative risk (RR) for the highest quartile of serum glucose versus the lowest was 2.8 [95% confidence interval (CI), 1.2e6.5], and P for trend was 0.02. Insulin showed a weaker association with breast cancer, the adjusted RR of the highest quartile versus the lowest was 1.7 (95% CI, 0.7e4.1), and P for trend was 0.14, whereas the adjusted RR of the highest quartile of IGF-I was 3.1 (95% CI, 1.1e8.6), and P for trend was 0.01. Increased levels of insulin-like growth factor binding protein-3 (IGFBP)-3 were related to breast cancer risk: the adjusted RR for the highest quartile was 2.1 (95% CI, 0.95e4.75), and P for trend was 0.02.

140

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

In postmenopausal women, the associations of glucose, insulin, and IGF-1 pattern were associated with breast cancer risk in heavier subjects characterized by a body mass index higher than 26. These results indicate that chronic alteration of glucose metabolism is related to breast cancer development. Reprinted with permission from Cancer Epidemiology.

5.3 Ye H, Adane B, Khan N, Alexeev E, Nusbacher N, Minhajuddin M, Stevens BM, Winters AC, Lin X, Ashton JM, Purev E, Xing L, Pollyea DA, Lozupone CA, Serkova NJ, Colgan SP, and Jordan CT. 2018. Subversion of systemic glucose metabolism as a mechanism to support the growth of leukemia cells. Cancer Cell, Oct 8; 34(4): 659e673. DOI: https://doi. org/10.1016/j.ccell.2018.08.016. 5.3.1 Summary From an organismal perspective, cancer cell populations can be considered analogous to parasites that compete with the host for essential systemic resources such as glucose. Here, we employed leukemia models and human leukemia samples to document a form of adaptive homeostasis, where malignant cells alter systemic physiology through impairment of both host insulin sensitivity and insulin secretion to provide tumors with increased glucose. Mechanistically, tumor cells induce high-level production of IGFBP1 from adipose tissue to mediate insulin sensitivity. Further, leukemia-induced gut dysbiosis, serotonin loss, and incretin inactivation combine to suppress insulin secretion. Importantly, attenuated disease progression and prolonged survival are achieved through disruption of the leukemia-induced adaptive homeostasis. Our studies provide a paradigm for systemic management of leukemic disease. Previous studies have shown that cell intrinsic mechanisms for glucose uptake/utilization are critical for growth of many cancer cell types. However, these studies do not consider cancer from an organismal perspective, where tumor cells comprise a relatively small proportion of host mass, and competition with normal tissues for finite amounts of systemic glucose may be a critical component of tumor growth. Our findings indicate that leukemic tumors gain a competitive advantage by co-opting multiple mechanisms to induce a diabetes-like physiologic condition in

Glucose deprivation and fasting strategies

Serotonin GLP-1

SCFAs

Insulin secretion

Normal tissues

Adipose tissue

Microbiota

Gut

141

IGFBP1

Insulin sensitivity

Glucose

Leukemia cells

Figure 5.3.1 Graphical abstract.

the host, and thereby subvert systemic glucose metabolism to facilitate disease progression. Further, our studies demonstrate that restoration of normal glucose regulation may be a feasible strategy to suppress systemic growth of malignant cell types.

5.3.2 Introduction It is well established that cancer cells consume more glucose than normal cells, a component of malignant cell bioenergetics that has been documented in numerous studies (Hay, 2016). To date though, studies have mainly focused on cell intrinsic mechanisms by which glucose is preferentially utilized, such as activation of glucose transporters and glycolysis. In the present study, we sought to approach glucose metabolism from a holistic perspective and to consider how cancer cells manage their need for glucose in the context of an entire mammalian organism. We note that, relative to the total tissue mass of an organism, the volume of tumor is generally low. Therefore, we hypothesized that activation of intrinsic pathways alone may not be sufficient to provide adequate glucose to drive robust cancer cell growth. Indeed, to successfully compete for finite amounts of systemic glucose, malignant cells need to reduce the glucose utilization of normal

142

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

tissues, such as adipose tissue and muscle, both of which are major sites of glucose consumption. Hence, the focus of this study was to investigate the concept that the tumor cell populations may alter utilization of glucose through perturbation of normal physiology. This type of systemic rebalancing of biological processes has previously been reported, where stimuli such as aging or environmental stress induce the expansion or contraction of the homeostatic range, a process termed adaptive homeostasis (Davies, 2016). Due to insulin resistance (IR) or/and abnormal insulin production, obese and diabetic patients reside in a state where their glucose utilization is hindered, a condition that causes increased levels of glucose in peripheral circulation. Therefore, in a diabetic macroenvironment, malignant cells would presumably have access to increased glucose. Furthermore, a common characteristic of type II diabetes is elevated insulin (hyperinsulinemia) (Shanik, Xu, Skrha et al., 2008), a condition in which insulin increases proliferation of some cancer cells (Gallagher and LeRoith, 2010). Additionally, obesity/diabetes-associated chronic inflammation also acts to promote the spread and survival of tumors (Deng, Lyon, Bergin et al., 2016). Further supporting a role for diabetic conditions in tumor pathogenesis, several studies have demonstrated the anti-cancer effects of anti-obesity/ diabetic drugs (Dowling, Niraula, Stambolic et al., 2012; Seguin, Carvalho, Bastos et al., 2012). Of particular interest, drugs like metformin can have direct anti-cancer effects but may also reduce hyperinsulinemia (Dowling, Niraula, Stambolic et al., 2012), which could in turn modulate glucose metabolism and/or reduce direct stimulation of tumor cells. Together, these studies raise the possibility that anti-diabetic therapies may act to restore more normal systemic glucose metabolism and suppress growth of malignant cells. Interestingly, obese populations are at a higher risk for certain types of solid tumors and leukemias (Basen-Engquist and Chang, 2011; Lichtman, 2010), suggesting a systemic link between tumor growth and metabolism. Indeed, previous studies have demonstrated that leukemic tumors alter multiple aspects of normal homeostasis. For example, we recently showed that inflammatory cytokines produced by leukemia cells elevate the lipolysis rate from adipose tissue, leading to elevated serum free fatty acids (FFAs) (Ye, Adane, Khan et al., 2016). Leukemic tumors also influence tissue metabolism by inducing hypoxia (Benito, Shi, Szymanska et al., 2011). Previous studies in non-cancer models have also shown that inflammation, fatty acids, and hypoxia are potent inducers of IR (Rasouli, 2016; Samuel and Shulman, 2012). Based on the studies outlined above, we therefore

Glucose deprivation and fasting strategies

143

sought to test the hypothesis that systemic adaptations caused by leukemia would result in IR and consequently reduce glucose utilization in normal tissues.

5.3.3 Results 5.3.3.1 Leukemia induces IR and reduces serum insulin level To address the questions posed in this study, we first employed two independent mouse syngeneic models, generated by either the combination of BCR/ABL and NUP98/HOXA9 fusions (hereafter termed “BN” model), or the MLL/AF9 fusion (hereafter termed “MLL” model). Both systems provide models of aggressive primary acute myeloid leukemia (AML). Using systems of this type, we have previously demonstrated that adipose tissue serves as a reservoir for leukemia cells. Notably, AT resident leukemia cells are highly pro-inflammatory and induce both systemic secretion of inflammatory cytokines as well as lipolysis (Ye, Adane, Khan et al., 2016). Further, adipose tissue represents a non-hematopoietic microenvironment where leukemia cells are preferentially shielded from effects of chemotherapy. Thus, in our studies we examined leukemia burden in the disease-initiating environment of bone marrow (BM) as well as in the extramedullary context of adipose tissue. The inflammatory cytokines and increased FFAs noted in our mouse models have the potential to act as inducers of IR. We therefore directly examined insulin responsiveness using insulin tolerance tests (ITTs). These studies demonstrated that the effect of insulin on leukemic mice was significantly impaired in the BN and MLL models (Figures 5.3.2A and S1A). To further evaluate systemic glucose metabolism in leukemic mice, we examined adipose and muscle tissues, which are the major glucose utilization sites in mammalian organisms and play critical roles in the development of IR. We found that both gonadal adipose tissue (GAT) and soleus muscle displayed a lower rate of basal and insulin-stimulated glucose utilization in leukemic mice compared with normal mice (Figures 5.3.2B and S1B). Additionally, a significantly weaker induction of p-Akt was observed in GAT from leukemic mice challenged with insulin (Figure 5.3.2C), suggesting a tissue intrinsic failure to mediate insulin signaling. Together, these data support the hypothesis that leukemic mice are insulin resistant. Elevated levels of peripheral glucose are usually seen in individuals with IR. Therefore, we hypothesized that the induction of IR would provide leukemia cells with more glucose. Notably, blood glucose levels in

144

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

Figure 5.3.2 Leukemia induces IR and reduces serum insulin level. (A) ITTs performed on normal and BN mice (n ¼ 8); (B) Glucose utilization by GAT from normal and BN mice in the basal and insulin-stimulated conditions (n ¼ 3); (C) Starved normal and BN mice were treated with insulin for 30 min and GAT were harvested for detection of pAkt; (D) Fasting blood glucose levels in normal and BN mice (n ¼ 4); (E) Glucose utilization in normal hematopoietic cells, and leukemia and non-leukemia cells from BN BM; (F) Sorted BN leukemia cells were treated with BSA, insulin (1 ng/mL), IGFBP1 (200 ng/mL), or insulin (1 ng/mL) plus IGFBP1 (200 ng/mL) for 30 min. Cells were harvested for detection of indicated protein; (G) Fasting serum insulin levels in normal, BN, and MLL mice (n ¼ 5); (H) GAT leukemic burden in type-1 diabetic BN mice (n ¼ 5); (I and J) BN mice were treated with insulin. GAT and BM leukemic burden (I), and serum FFAs (J) were examined (n ¼ 6). Data are represented as mean  SD. See also Figure S1.

leukemic mice were actually lower than in normal mice (Figure 5.3.2D), suggesting that, in balance, consumption of glucose by leukemia cells is so great that, even with increased availability, net glucose in circulation is reduced. This concept is supported by the studies shown in Figure 5.3.2E,

Glucose deprivation and fasting strategies

145

where glucose utilization by normal hematopoietic cells or non-leukemia cells from leukemic BM were compared with leukemia cells. The data demonstrate that leukemia cells have dramatically increased glucose consumption. Notably, glucose utilization in normal hematopoietic cells and non-leukemia cells was enhanced by insulin; however, leukemia cells did not respond to insulin, as reflected by their glucose utilization and p-Akt level (Figures 5.3.2E and F), indicating that the systemic insulin-resistant phenotype observed in leukemic mice will not affect glucose utilization by leukemic cells. Of note, in contrast to some studies suggesting that insulin promotes certain types of cancer growth (Gallagher and LeRoith, 2010), our findings indicate that, at least in leukemias, insulin does not function as a promoter for glucose utilization and cell proliferation (Figure S1C). A compensatory increase in serum insulin (i.e., hyperinsulinemia) is usually seen during the development of IR. Hyperinsulinemia has been suggested to be one of the mechanisms for higher risk for some cancer types due to its role in promoting cell proliferation (Arcidiacono, Iiritano, Nocera et al., 2012). However, in both murine leukemia models, serum insulin levels were significantly reduced (Figures 5.3.2G and S1D), a phenomenon that is seen in type 1 diabetic patients. Therefore, we hypothesized that modulations of insulin levels would affect disease progression. To test this hypothesis, we first employed the streptozotocin (STZ)-induced type 1 diabetes model, a system in which pancreatic b cells are damaged by administration of STZ, thereby inducing hypoinsulinemia and hyperglycemia (Figure S1E). Leukemia generated in these mice showed increased tumor burden in GAT and higher levels of serum FFAs than non-diabetic leukemic mice (Figures 5.3.2H and S1F). We did not observe increased BM leukemic burden in diabetic mice (Figure S1G). This is likely due to impairment of the BM microenvironment induced by STZ (Motyl and McCabe, 2009), as we observed reduced hematopoietic stem/progenitor cells (linScaþc-kitþ) in non-leukemic diabetic mice (Figure S1H). Next, we asked whether insulin supplementation would benefit leukemic mice. As shown in Figure 5.3.2I, insulin treatment significantly decreased leukemic burden in GAT and BM. Reduced lipolysis was achieved by insulin treatment as well (Figure 5.3.2J). Additionally, leukemia-induced body weight loss as well as atrophy of GAT was alleviated by insulin treatment (Ye, Adane, Khan et al., 2016) (Figures S1I and S1J). Insulin also

146

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

reduced BM leukemic burden in MLL mice, while its effect on GAT leukemic burden was minimal (Figure S1K). Collectively, our data suggest that the physiologic state induced by leukemia results in IR and loss of circulating insulin, both of which increase the glucose availability for the growth of leukemic tumors. 5.3.3.2 Adipose-derived IGFBP1 induces the development of IR in leukemia Having established that leukemic mice are insulin resistant, we next addressed the mechanism by which this condition is induced. Adipose tissue, through its endocrine function, regulates systemic homeostasis and is known to be involved in the development of IR (Bjorndal, Burri, Staalesen et al., 2011). To investigate the potential role of adipose tissue in leukemiainduced IR, we compared the endocrine function between normal and leukemic adipose tissue. Adipokine arrays were performed on conditioned medium (CM) from normal and leukemic GAT. We observed that one of the adipokines, IGFBP1, was highly elevated in CM from leukemic GAT relative to normal GAT (Figure 5.3.3A). We validated this finding and found that IGFBP1 in the leukemic GAT CM was approximately 50-fold higher than in the control (Figure 5.3.3B). Notably, mRNA and protein levels of IGFBP1 were significantly increased in both GAT and inguinal adipose tissue (IAT) from leukemic mice relative to normal mice (Figures 5.3.3C and S2A). In normal physiology, IGFBP1 is primarily produced by the liver. While some elevation of IGFBP1 is evident in the liver of leukemic mice (Figure 5.3.3C), the majority of aberrant IGFBP1 is produced in the adipose tissue. Previous studies have indicated that IGFBP1 may have a role in the development of IR (Lewitt, Dent, and Hall, 2014). Therefore, to explore a potential in vivo role for IGFBP1, we first examined the serum level of IGFBP1 in leukemic mice and observed the IGFBP1 level at more than 100 times the level detected in normal mice (Figure 5.3.3D). Further, in monitoring the development of leukemic disease, a gradual elevation of serum IGFBP1 was noted (Figure 5.3.3E). Elevated IGFBP1 was detected as early as day 6 after leukemic transplantation when disease burden was below 1% in both BM and GAT (Figure S2B). Elevated IGFBP1 was also evident in MLL mice (Figure S2C), and we confirmed that adipose tissues were the source of the aberrant IGFBP1 (Figures S2D and S2E). Together, these data suggest that the aberrant high level of IGFBP1 is a prevalent systemic feature of leukemic disease.

Glucose deprivation and fasting strategies

147

Figure 5.3.3 Adipose-derived IGFBP1 induces the development of IR in leukemia. (A) Adipokine arrays on CM from normal and BN GAT. Red circle indicates IGFBP1; (B) IGFBP1 levels in CM from normal and BN GAT; (C) IGFBP1 protein levels in GAT, IAT, and liver from normal and BN mice. Recombinant (RB) mouse IGFBP1 protein and liver protein extracts were used as positive controls; (D) Serum IGFBP1 levels in normal and BN mice (n ¼ 5). (E) Serum IGFBP1 levels in BN mice at different time points after leukemic transplantation (n ¼ 4); (F) 3T3-L1 adipocytes were serum starved for 1 hr and then treated with RGD peptides for 30 min. Cells were then treated with insulin (1 ng/mL) and IGFBP1 (200 ng/mL) for 30 min for detection of indicated protein. (G) ITT performed on normal mice treated with IGFBP1 (n ¼ 8); (H) Serum IGFBP1 levels in BN mice treated with insulin (n ¼ 6); (I and J) Serum IGF1 levels in normal and BN mice were detected by ELISA (n ¼ 6; I) and immunoblot (J); Data are represented as mean  SD. *p < 0.05; **p < 0.005; ***p < 0.0005; ****p < 0.00005. See also Figure S2.

Having established that leukemia induces a prominent level of peripheral IGFBP1, we examined the role of IGFBP1 in the development of IR. We treated 3T3-L1 adipocytes with IGFBP1 and found that insulin effects on these adipocytes were partially blocked by IGFBP1 (Figure 5.3.3F). Next, we treated normal GAT explants with IGFBP1 in the presence or absence of insulin.

148

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

As shown in Figure S2F, high doses of IGFBP1 partially impaired insulin effects on GAT. IGBFP1 functions as a ligand for integrin receptors due to its arginine-glycine-aspartic acid (RGD) motif in the C terminus (Wang, Wei, Krzeszinski et al., 2015). Additionally, the integrin receptor-mediated pathway has been implicated in the development of IR (Kang, Mokshagundam, Reuter et al., 2016). Therefore, we hypothesized that IGFBP1-mediated IR occurred at least partially through the integrin receptor-mediated pathway. Indeed, an RGD peptide restored the insulin effect on IGFBP1-treated 3T3-L1 adipocytes, as shown by increased p-Akt and p-IR (insulin receptor) levels (Figure 5.3.3F). To further examine IGFBP1 function in vivo, normal mice were treated with high doses of IGFBP1 for 2 weeks. Reduced insulin sensitivity was observed in IGFBP1treated mice (Figure 5.3.3G). These data suggest that high doses of IGFBP1 induce IR. Several previous studies have indicated an interplay between insulin and IGFBP1 (Lewitt, Dent, and Hall, 2014). We found that insulin treatment decreased serum IGFBP1 in leukemic mice (Figure 5.3.3H). Further, IGFBP1 level was elevated in diabetic leukemic mice (Figure S2G). These data indicate that serum IGFBP1 is negatively correlated with serum insulin in the context of leukemia. IGFBP1 binds to and affects IGF1 biological functions (Firth and Baxter, 2002). We observed a significant decrease in serum IGF1 in BN mice compared with normal mice (Figures 5.3.3I and J). IGF1 plays a similar role in glucose metabolism as insulin (Schiaffino and Mammucari, 2011). Indeed, IGF1 treatment induced phosphorylation of Akt in 3T3-L1 adipocytes and this effect was attenuated by IGFBP1 (Figure S2H). Therefore, functional loss of IGF1 may contribute to the insulin-resistant phenotype in leukemic mice as well. Together, the above data suggest that leukemic tumors induce highlevel production of IGFBP1 from adipose tissue, which in turn acts to impair insulin/IGF1 function and induce an insulin-resistant condition. 5.3.3.3 Modulation of IGFBP1 mediates leukemia growth in vivo To further confirm that IGFBP1 contributes to leukemia-induced IR, leukemic mice were treated with an IGFBP1 neutralizing antibody. As shown in Figures 5.3.4A, B, and S3A, insulin sensitivity and serum insulin levels were partially restored by blockage of IGFBP1, indicating that IGFBP1 not only mediates insulin sensitivity but is also involved in

Glucose deprivation and fasting strategies

149

Figure 5.3.4 Modulation of IGFBP1 mediates leukemia growth in vivo. (A) ITT performed on BN mice treated with anti-IGFBP1 antibody (n ¼ 8); (BeD) Fasting serum insulin (B), GAT leukemic burden (C), and serum FFAs (D) in BN mice treated with antiIGFBP1 antibody (n ¼ 5); (E) Example images and quantification of femur trabecular bone mass examined by micro-computed tomography (n ¼ 4); (F and G) BM and GAT leukemic burden (F) and fasting serum insulin (G) in BN mice preconditioned with IGFPB1 (n ¼ 5). Data are represented as mean  SD. *p < 0.05; **p < 0.005; ***p < 0.0005. See also Figure S3.

regulating serum insulin levels. Importantly, GAT leukemic burden was significantly reduced by IGFBP1 blockage (Figure 5.3.4C), and BM leukemic burden was mildly but significantly reduced (Figure S3B). Further, lipolysis rate, atrophy of GAT, and overall body weight loss were all mitigated by IGFBP1 blockage (Figures 5.3.4D, S3C, and S3D). IGFBP1 has recently been identified as a promoter for osteoclasts through activation of the Erk signaling pathway (Wang, Wei, Krzeszinski et al., 2015). Given the fact that leukemia pathogenesis can induce severe bone loss (Frisch, Ashton, Xing et al., 2012), we hypothesized that IGFBP1 blockage would reduce bone loss in leukemic mice. Indeed, increased trabecular bone mass was observed in leukemic mice that received antiIGFBP1 treatment (Figures 5.3.4E and S3E). These data suggest that an adipose-bone axis is activated by leukemia during disease progression. Next, we tested whether IGFBP1-preconditioning would facilitate disease progression. As shown in Figures S3F and 3F, IGFBP1preconditioned mice had higher serum IGFBP1 and higher BM leukemic burden. Additionally, early body weight loss as well as atrophy of GAT was found in IGFBP1-preconditioned mice (Figures S3G and S3H).

150

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

Further, serum insulin was reduced in IGFBP1-preconditioned mice (Figures 5.3.4G and S3I). While GAT leukemic burden did not significantly vary between preconditioned and control groups, due to logistical constraints, the analysis was performed at relatively low disease burden, making detection of differences more challenging. Besides its systemic effects, IGFBP1 also functions as a signaling molecule in hematopoietic cells (Wang, Wei, Krzeszinski et al., 2015), an activity that may have a direct effect on leukemia cell proliferation. Indeed, increased proliferation was observed in leukemia cells treated with IGFBP1 (Figure S1C). The mitogenic effect of IGFPB1 on leukemia cells was probably due to activation of the Erk signaling pathway, which was blocked by IGF1 (Figures 5.3.2F and S3J). Additionally, phosphorylated p38 was reduced in IGFBP1-treated leukemia cells (Figure 5.3.2F), suggesting that IGFBP1 also reduced cellular stress and might promote survival of leukemia cells. Together, these data suggest that leukemic tumors induce production of IGFBP1, which in turn facilitates disease progression by modulating host insulin sensitivity and peripheral insulin levels as well as by affecting leukemia cell proliferation and survival. 5.3.3.4 Loss of active GLP-1 and serotonin contributes to inhibition of insulin secretion in leukemia pathogenesis Reduced serum insulin further exacerbates IR, a scenario that is seen in late-stage type 2 diabetic patients. Thus, we next investigated the mechanisms for reduced serum insulin in leukemic mice. Immunofluorescent (IF) staining was performed on pancreas from normal and leukemic mice to examine insulin production. As shown in Figure 5.3.5A, a comparable staining intensity of insulin was observed between normal and leukemic pancreas. Further, glucosestimulated insulin secretion (GSIS) assays showed even stronger insulin release in leukemic mice challenged with a high dose of glucose (Figures 5.3.5B and S4A). These data suggest that insulin synthesis in pancreatic cells was comparable between normal and leukemic mice; however, insulin secretion under physiologic glucose levels was impaired in leukemic mice. Interestingly, even with the strong induction of insulin caused by glucose bolus in leukemic mice and a much higher glucose utilization by leukemia cells (Figure 5.3.2F), glucose tolerance tests (GTTs) demonstrated that the clearance of glucose was comparable between normal and leukemic mice (Figure 5.3.5C), further corroborating the fact that leukemic mice are insulin resistant.

Glucose deprivation and fasting strategies

151

Figure 5.3.5 Increased DPP4 and loss of active GLP-1 contribute to the inhibition of insulin secretion in leukemia pathogenesis. (A) Immunofluorescent staining for insulin in pancreas from normal, BN, and MLL mice; (B) GSIS performed on normal, BN, and MLL mice (n ¼ 4). (C) GTT performed on normal and BN mice (n ¼ 6); (D and E) Fasting serum DPP4 (D) and active GLP-1 levels (E) in normal and BN mice (n ¼ 5). BV/TV. (FeI) Fasting serum insulin (F), serum IGFBP1 (G), serum FFAs (H), and BM and GAT leukemic burden (I) in exenatide-treated BN mice (n ¼ 6). Data are represented as mean  SD. *p < 0.05; **p < 0.005; ***p < 0.0005. See also Figure S4.

To further investigate leukemia-induced suppression of insulin secretion, we examined peripheral levels of incretins and serotonin, both of which act to stimulate insulin release under various conditions (Kim and Egan, 2008; Sugimoto, Kimura, Yamada et al., 1990). Incretins such as GLP-1 regulate both insulin secretion and insulin sensitivity (Kim and Egan, 2008). We observed that DPP4, an enzyme known to inactivate incretins, was elevated in leukemic serum (Figure 5.3.5D), and that active GLP-1 was reduced (Figure 5.3.5E). To further evaluate the involvement

152

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

of GLP-1 in insulin secretion in our model, leukemic mice were treated with saxagliptin, a DPP4 inhibitor, or exenatide, a GLP-1 receptor agonist. We found that treatment with either saxagliptin or exenatide partially restored insulin levels in leukemic mice (Figures 5.3.5F, S4B, and S4C). Additionally, IGFBP1 and lipolysis was suppressed in treated mice (Figures 5.3.5G, H, S4D, and S4E). Importantly, leukemic burden in both BM and GAT was reduced by saxagliptin and exenatide treatments (Figures 5.3.5I and S4F). Together, our data indicate that leukemia-induced production of DPP4 acts to suppress insulin secretion, which thereby promotes disease progression. Next, we investigated the role of serotonin in leukemia-induced inhibition of insulin secretion. We found serum serotonin was drastically reduced in both models of leukemia (Figures 5.3.6A and S5A). Peripheral serotonin is primarily derived from gut. Expression of TPH1, encoding the rate-limiting enzyme for serotonin synthesis, was significantly reduced in the colon tissue of leukemic mice (hereafter termed “leukemic colons”)

Figure 5.3.6 Loss of serotonin leads to the inhibition of insulin secretion in leukemia pathogenesis. (A) Serotonin levels in normal and BN mice (n ¼ 4); (B) Expression of gene involved in serotonin metabolism in the colon tissues from normal and BN mice. (C) GI transition time in normal and BN mice (n ¼ 5); (DeG) Fasting serum insulin (D), serum IGFBP (E), GAT and BM leukemic burden (F), and serum FFAs (G) in serotonintreated BN mice (n ¼ 5); (H) Serum serotonin levels in anti-IGFBP1 antibody-treated BN mice (n ¼ 5). Data are represented as mean  SD. *p < 0.05; **p < 0.005; ***p < 0.0005. See also Figure S5.

Glucose deprivation and fasting strategies

153

(Figure 5.3.6B). We also noticed reduced expression of MAOA, which encodes an enzyme to catabolize serotonin, probably due to a rescue effect for serotonin loss. Serotonin regulates gastrointestinal (GI) transition (Yano, Yu, Donaldson et al., 2015). We observed that the average GI transition time for leukemic mice was significantly increased (Figure 5.3.6C), corroborating a functionally relevant reduction in serotonin in leukemic mice. To examine the role of serotonin in insulin secretion in our models, leukemic mice were supplemented with serotonin. As expected, serum insulin levels were partially restored by serotonin supplementation (Figures 5.3.6D and S5B). Further, IGFBP1 levels were suppressed and both BM and GAT leukemic burden were significantly reduced by serotonin treatment (Figures 5.3.6E and F). Additionally, the lipolysis rate, GAT atrophy, and overall body weight loss were suppressed by serotonin treatment (Figures 5.3.6G, S5C, and S5D). Intriguingly, IGFBP1 blockage by anti-IGFBP1 antibody mildly restored serotonin levels in leukemic mice (Figure 5.3.6H), while IGFBP1-preconditioning decreased serotonin levels (Figure S5E), suggesting that the benefits of anti-IGFBP1 treatment on leukemic mice were at least partially acting through a serotonin-mediated pathway and that there was interplay between these two molecules. Collectively, these data suggest that impaired serotonin production is a central systemic feature of leukemia pathogenesis, which acts to suppress insulin secretion and thereby promotes disease progression.

5.3.4 Leukemia-associated microbiota facilitates disease progression The loss of serotonin and increased GI transition time noted above indicate the presence of pathological changes in the gut of leukemic mice. Indeed, the expression of several inflammatory genes was upregulated in leukemic colons (Figure S6A). Interestingly, dysregulated expression of antimicrobial genes was observed in leukemic colons (Figure 5.3.7A), indicating the colonic microenvironment was altered in leukemic mice. This type of change would be predicted to result in dysbiosis. Notably, it has recently been shown that gut the microbiota serves as a central regulator for metabolism, where metabolites from microbiota influence gut homeostasis and have a profound impact on systemic homeostasis (Nicholson, Holmes, Kinross et al., 2012). Thus, we postulated that the microbiota could be an important component of the overall adaptive homeostasis we saw in leukemia.

154

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

Figure 5.3.7 Leukemia-associated microbiota facilitates disease progression. (A) Expression of anti-microbial genes in the colon tissues from normal and BN mice; (B) 16S-rRNA-sequencing was performed on the fecal materials collected from normal and BN mice (day 13 after leukemic transplantation). Circle indicates Bacteroidales S24-7; (CeE) BM and GAT leukemia burden (C), serum IGFBP1 (D), and fasting serum insulin (E) in BN mice transplanted with fecal materials from normal or BN mice (n ¼ 7). n.s., not significant; (F) BM and GAT leukemic burden in BN mice treated with Abx (n ¼ 5); (G) Serum IGFBP1 levels in BN mice treated with Abx (n ¼ 5); (H) ITT performed on non-leukemic mice transplanted with normal or BN fecal materials (n ¼ 8); (I) Fasting blood glucose and fasting serum insulin levels in non-leukemic mice transplanted with normal or BN fecal materials (n ¼ 7). Data are represented as mean  SD. *p < 0.05; **p < 0.005; ***p < 0.0005; ****p < 0.00005. See also Figure S6 and Table S1.

To compare the gut flora between normal and leukemic mice, 16SrRNA sequencing was performed on fecal materials from leukemic and normal mice. As shown in Figures 5.3.7B and S6B and Table S1, the composition of gut flora in leukemic mice differed from that of normal mice. To test the functional potential of leukemia-induced microbiota, we performed fecal materials transfer experiments. As outlined in Figure 5.3.7C, normal recipient mice were pretreated with an antibiotic cocktail (Abx) (Hill, Hoffmann, Abt et al., 2010) and then transplanted with fecal materials from normal or leukemic mice.

Glucose deprivation and fasting strategies

155

Following 2 weeks to allow gut colonization, animals were then transplanted with leukemia cells. As shown in Figure 5.3.7C, compared with animals transplanted with fecal materials from normal mice, GAT leukemic burden was significantly higher in mice transplanted with fecal materials from leukemic mice. Additionally, serum IGFBP1 was significantly increased and insulin was reduced in leukemic mice receiving leukemic fecal materials (Figures 5.3.7D, E, and S6C). Together, these data suggest that leukemia-associated microbiota contribute to the overall systemic perturbations observed in leukemia and act to facilitate disease progression. To further demonstrate the role of microbiota in disease progression, the gut microbiota of a cohort of mice was ablated using Abx and leukemic progression was monitored. We found that BM leukemic burden was mildly reduced, and GAT leukemic burden was remarkably decreased by Abx treatment (Figure 5.3.7F). Further, a striking reduction in serum IGFBP1 was observed in Abx-treated leukemic mice (Figure 5.3.7G). Additionally, serum insulin levels were partially restored and serum FFAs were reduced in Abx-treated animals (Figures S6D and S6E). Collectively, these data confirm that the microbiota plays a role in leukemia progression. Next, we asked whether the leukemic microbiota was linked to the regulation of insulin sensitivity. Previous studies have shown that transferring microbiota from obese or insulin-resistant donors induces similar phenotypes in recipients (Baothman, Zamzami, Taher et al., 2016). As shown in Figure 5.3.7H, transfer of leukemic fecal materials impaired insulin sensitivity in recipient mice. Additionally, fasted blood glucose and insulin levels were significantly higher in mice receiving leukemic fecal materials (Figures 5.3.7I and S6F). Together, these data suggest that the leukemia-associated microbiota contribute to the leukemia-induced insulin-resistant phenotype. Collectively, these findings indicate leukemic tumors induce dysbiosis and that leukemia-associated microbiota facilitate leukemia progression at least partially through nduction of IR.

5.3.5 Microbiota-derived short-chain fatty acids impede leukemia progression and are reduced in leukemic mice The microbiota is a key regulator of gut homeostasis, and dysbiosis contributes to pathological changes of gut. Indeed, we observed that leukemic

156

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

mice had shortened colons and that there was a clear loss of gut epithelia integrity (Figures S7A and 5.3.8A). Additionally, genes that are directly involved in the regulation of epithelial integrity were dysregulated in leukemic colon (Figure S7B). Further, we observed morphological changes in leukemic colons and invasion of leukemia cells into the colon epithelia, which may contribute to the loss of epithelial integrity (Figure 5.3.8B). Given the multiple signs of gut dysfunction and dysbiosis noted above, we asked whether products of the microbiota were altered in leukemia mice. Specifically, short-chain fatty acids (SCFAs) are secreted by specific types of gut bacteria and have major impacts on gut biology, such as regulating gut epithelial integrity (Kelly, Zheng, Campbell et al., 2015). Interestingly, 16S-rRNA sequencing showed that the proportions of several bacterial taxa with members capable of making SCFAs were significantly reduced in the fecal materials from leukemic mice, including the Lachnospiraceae and Bacteroidales S24-7 families (Evans, LePard, Kwak et al., 2014; Ormerod, Wood, Lachner et al., 2016), and the butyrate producing genus Anaerostipes (Figures 5.3.7B and S6B and Table S1). Indeed, fecal samples from leukemic mice had lower levels of the SCFAs butyrate and propionate (Figure 5.3.8C). To test the functional relevance of this observation, we treated leukemic mice with tributyrin or propionate and observed a partial rescue of GI epithelial integrity (Figures 5.3.8D and S7C). Strikingly, tributyrin suppressed BM and GAT leukemic burden and propionate suppressed GAT leukemic burden (Figures 5.3.8E and S7D). Additionally, serum IGFBP1 levels were reduced and insulin levels were increased by tributyrin and propionate treatments (Figures 5.3.8F, G, and S7EeS7G). Together, these data suggest that dysbiosis in leukemic mice results in loss of microbiotaderived SCFAs, which contributes to the leukemia-induced adaptive homeostasis by influencing gut epithelial integrity and modulating IGFBP1 and insulin levels.

5.3.6 Restoring systemic glucose metabolism provides survival benefits Our findings strongly suggest that tumor progression can be significantly affected by systemic glucose metabolism. Indeed, the data in Figures 5.3.6D, E, 5.3.8F, and G show that serotonin and tributyrin supplementation, through distinct mechanisms, were effective means to restore serum insulin and IGFPB1 levels. Therefore, combined supplementations with both serotonin and tributyrin (hereafter termed “Ser-Tri”) were

Glucose deprivation and fasting strategies

157

Figure 5.3.8 Microbiota-derived SCFAs impede leukemia progression and are reduced in leukemic mice. (A) Serum fluorescein isothiocyanate (FITC)-dextran levels in normal and BN mice (n ¼ 5); (B) H&E (left two images) staining and GFP (for leukemia cells, right two images) staining of colon tissues from normal and BN mice; (C) The amount of butyrate and propionate in fecal samples from normal and BN mice (n ¼ 8); (DeG) Serum FITC-dextran (D), BM and GAT leukemic burden (E), serum IGFBP1 (F), and fasting serum insulin (G) in tributyrin-treated BN mice (n ¼ 6); (H) BM and GAT leukemic burden in BN mice treated with combination of tributyrin and

158

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

selected to treat leukemic mice. As shown in Figure 5.3.8H, leukemia burden was strongly suppressed in both BM and GAT by the Ser-Tri treatment. More importantly, Ser-Tri therapy provided leukemic mice with significant survival benefits (Figure 5.3.8I). To confirm that the benefits of Ser-Tri therapy are due to modulations of systemic glucose metabolism, glucose utilization in leukemia cells and GAT was assessed. As shown in Figure 5.3.8J, in vivo analyses demonstrated that leukemia cells had a significant reduction in glucose utilization after Ser-Tri therapy treatment, whereas glucose utilization in GAT was increased (Figure 5.3.8J). Interestingly, in vitro analyses showed that the differences in glucose utilization between leukemia cells isolated from vehicle-treated and Ser-Tri therapy-treated leukemic mice were minimal (Figure 5.3.8J), indicating that the Ser-Tri therapy exerted its anti-leukemic effects by redirecting the systemic glucose flow from leukemia cells to host tissues, not by directly affecting leukemia cell glucose metabolism. To further investigate the effects of Ser-Tri therapy on systemic glucose metabolism, we employed positron emission tomography (PET)-computed tomography (CT) to visualize the uptake of glucose. As shown Figures 5.3.8K and S7H, glucose uptake was already dramatically increased in the spleen and BM and conversely decreased in the muscle tissue from BN mice compared with normal ones. Ser-Tri therapy significantly reduced glucose uptake in both BM and spleen and promoted glucose uptake in muscle from BN mice (Figures 5.3.8K and S7H). Interestingly, we also observed that glucose uptake in the liver of BN mice was increased compared with normal mice and Ser-Tri treatment also reduced BN liver glucose uptake (Figure 5.3.8K). Our previous studies (data not shown) indicate strong leukemic infiltration of liver in the BN model, suggesting that liver is one of the many organs in which glucose

=

serotonin (Ser-Tri) (n ¼ 7); (I) Survival curve for BN mice treated with Ser-Tri therapy (tributyrin and serotonin) (n ¼ 8); (J) A cartoon showing the experimental procedure and the glucose utilization in vehicle-treated and Ser-Tri therapy-treated (one-time treatment) leukemic mice (n ¼ 4); (K) Representative images of 18Ffluorodeoxyglucose (FDG) uptake by PET-CT scanning in normal mice, and untreated and Ser-Tri-treated BN mice. Normalized uptake values (NUV) fold changes examined by FDG-PET are presented in the graph (n ¼ 4); (L) Schematic representation of the treatment protocol and the survival curve for BN mice treated with chemotherapy alone or chemotherapy combined with the Ser-Tri therapy (n ¼ 9). Data are represented as mean  SD. *p < 0.05; **p < 0.005; ***p < 0.0005; ****p < 0.00005. See also Figure S7.

Glucose deprivation and fasting strategies

159

metabolism should be altered. Together, these results strongly support the concept that leukemic progression can be significantly attenuated by modulations of systemic glucose metabolism alone. Next, we asked whether modulation of the systemic glucose metabolism could be an adjuvant therapy for chemotherapy. Leukemic mice were treated with chemotherapy alone (Ye, Adane, Khan et al., 2016) or chemotherapy with the Ser-Tri therapy. As shown in Figure 5.3.8L, chemotherapy administered with the Ser-Tri therapy significantly increased the survival of leukemic mice compared with chemotherapy alone. Collectively, our findings suggest that, either alone or functioning as an adjuvant therapy, management of systemic glucose metabolism provides significant survival benefits.

5.3.7 Human leukemia induces an insulin-resistant phenotype To examine whether the findings in murine models can be recapitulated in human leukemia, we first compared the serum IGFBP1 level between normal controls, myelodysplastic syndrome (MDS) patients, and AML patients. As shown in Figure 5.3.9A, an approximately 20-fold higher level of IGFBP1 was observed in serum samples from AML patients compared with normal controls. Notably, serum IGFBP1 levels in MDS patients were elevated compared with normal controls but less than AML patients (Figure 5.3.9A), consistent with a progressive increase in IGFBP1 production as pathogenesis proceeds from chronic to acute phase disease. Additionally, we found that serum IGFBP1 level was significantly correlated with leukemic blast counts in AML patients (Figure S8A). Further, BM aspirates from leukemia patients contained significantly higher IGFBP1 than in BM aspirates from normal controls (Figure 5.3.9B). To further confirm the correlation between IGFBP1 level and disease burden, we compared BM aspirate IGFPB1 levels between paired diagnostic, remission, and relapsed AML samples. As shown in Figure 5.3.9C, the IGFPB1 level was significantly reduced in remission samples compared with diagnostic samples but rebounded in relapsed samples. Similar results were observed in BN mice that underwent chemotherapy-induced remission followed by relapse (Figure S8B). Collectively, our results suggest that an aberrant elevation of IGFBP1 may function as a biomarker for disease burden.

160

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

Figure 5.3.9 Human leukemia induces an insulin-resistant phenotype. (A) Serum IGFBP1 levels in normal controls (n ¼ 6), MDS (n ¼ 10), and AML (n ¼ 13) patients; (B) IGFBP1 level in normal (n ¼ 5) and AML (n ¼ 5) BM aspirate; (C) IGFBP1 levels in paired BM aspirates from newly diagnostic, remission, and relapsed AML patients; (D) Cytokine arrays performed on serum samples from normal controls and AML patients; (E) Serum FFA levels in normal controls, and MDS and AML patients; (F) ITT performed on NSG mice transplanted with a human primary leukemia sample (n ¼ 8); (G) Serum IGFBP1 levels in normal NSG mice and NSG mice transplanted with a human primary leukemia sample (n ¼ 8); (H) Serum serotonin levels in normal controls, and MDS and AML patients; (I) Serotonin levels in paired BM aspirates from newly diagnostic, remission, and relapsed AML patients; (J and K) Serum insulin levels (J) and serum glycated protein levels (K) in normal controls, and MDS and AML patients; (L) Schematic summary of system-wide perturbations leading to altered glucose utilization in leukemia. Data are represented as mean  SD. *p < 0.05; **p < 0.005; ***p 125 mg/dl), in comparison with hyperglycemic ones [8]. Ben, Xu, Ning et al. (2011) provided evidences in support of the intimate association between diabetes and pancreatic cancer with risk factor approximately 1.94, wherein the highest risk was observed in those patients diagnosed within one year [9]. Campbell, Deka, Jacobs et al. (2010) uncovered the association between colorectal cancer and type 2 diabetes mellitus or insulin use in men [10]. El-Serag, Hampel, and Javadi (2006) provided evidences in support of the association between diabetes and hepatocellular carcinoma through epidemiologic evidence [11]. Regarding breast cancer, Michels, Solomon, Hu et al. (2003) displayed type 2 diabetes and subsequent incidence of breast cancer in the Nurses’ Health Study [12], which was consolidated by the concomitant metaanalysis performed by Larsson, Mantzoros, and Wolk (2007) [13] and Wolf, Sadetzki, Catane et al. (2005) [14]. Consistent with the clinical

170

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

observations, an array of in vitro evidences addressed the potential impacts of high glucose on tumor biology of breast cancer. Flores-Lopez, MartinezHernandez, Viedma-Rodriguez et al. (2016) showed high glucose and insulin enhanced uPA expression, invasiveness and ROS formation in breast cancer-derived cells [15]. Takatani-Nakase, Matsui, Maeda et al. (2014) demonstrated that high glucose level promoted migrative behavior of breast cancer cells via zinc and its transporters [16]. Wei, Duan, Wang et al. (2017) proposed that high glucose and high insulin conditions promoted MCF-7 cell proliferation and invasion by upregulating IRS1 and activating the Ras/Raf/ERK pathway [17]. The above-mentioned results indicated the diverse mode-of-action underlying the contributions of high glucose to breast cancer, which prompted us to investigate this phenotype while focusing on the possible negative regulation on potential tumor suppressors.

5.5.3 Method and materials 5.5.3.1 Cell culture Human breast cancer cell lines MCF-7 and MDA-MB-231 were received from the American Type Culture Collection (VA, U.S.A.) and grown in RPMI-1640 medium (Hyclone, ThermoFisher, MA, U.S.A.) containing 10% fetal bovine serum (FBS, HyClone) and 1% penicillin/streptomycin (Gibco, MA, U.S.A.). Cells culture was maintained in humidified CO2 (5%) incubator. Cell transfection was performed with DharmaFECT Transfection Reagents (ThermoFisher, MA, U.S.A.) according to the manufacturer’s instruction. 5.5.3.2 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay To generate viability curves, 500 cells/100 ml were seeded into 96-well plate and cultured in regular medium for 24 h. The medium was replaced by MTT reagent (Sigma, MO, U.S.A.) dissolved in phosphate buffered saline and incubated for 3 h at 37 C. The resultant formazan was dissolved by 200 ml of dimethyl sulfoxide and absorbance at 570 nm was measured with ELx800 microplate reader (BioTek Instruments, VT, U.S.A.).

5.5.4 Cell counting The indicated cells were plated into 6-well plate at a density of 5  104 cells/2 ml and grown for 72 h. Cells were harvested and digested into

Glucose deprivation and fasting strategies

171

single-cell solution and subjected to trypan blue staining. The viable cells were counted under the light microscope using hemocytometer.

5.5.5 Transwell assay Transwell chambers (Corning, NY, U.S.A.) were used to measure migrative capacity. Indicated cells (5  105 cells/well in serum-free medium) were plated in the upper chamber. The lower compartment was filled with 0.75 ml complete medium plus extra 5% FBS. Cells were cultured for 10 h at 37 C in CO2 incubator. Non-migrated cells were carefully swiped with cotton swabs. Cells on the polycarbonate membrane were stained with 0.2% crystal violet. Cells were counted in five random fields to calculate the migration. For cell invasive assay, the chamber was pre-coated with Matrigel (0.25 mg/ml; BD Biosciences, CA, U.S.A.) and followed by the same procedure as described above. 5.5.5.1 Soft agar assay The 6-well plate was pre-coated with 0.75% agarose. The single-cell suspension was prepared in 0.25% low melting agarose (2  104 cells/ml) and cautiously laid onto the supporting layer. Cells were cultured for 3 weeks in the CO2 chamber. The formed colonies were stained and counted under a light microscope (Olympus, Tokyo, Japan). 5.5.5.2 Western blot Cells were harvested and lysed in radioimmunoprecipitation assay buffer (Beyotime, Nantong, China). Equal amounts of proteins were resolved by 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred onto methanol pre-activated polyvinylidene difluoride membrane on ice. 5% non-fat milk was used for blocking purpose. Incubation with specific primary antibody [anti-Angiotensinogen (AGT), #79299, 1:1000; anti-actin, #4967, 1:1000, Cell Signaling Technology, MA, U.S.A.] was performed at 4 C overnight and followed by horseradish peroxidaseconjugated secondary antibody (anti-rabbit, #7074, 1:5000, Cell Signaling Technology, MA, U.S.A.) hybridization. The bands were visualized by ECL (Millipore, CA, U.S.A.). 5.5.5.3 Real-time polymerase chain reaction Total RNA extraction was performed with BiooPure RNA Isolation Reagent (Bioo Scientific, TX, U.S.A.). Reverse transcription was prepared by the RevertAid First Strand cDNA Synthesis Kit (Fermentas, MD,

172

Starving Cancer Cells: Evidence-Based Strategies to Slow Cancer Progression

U.S.A.). Real-time polymerase chain reaction (PCR) was employed the SYBR Green MasterMix (Applied BioSystems, CA, U.S.A.) and followed the manufacturer’s instructions. Relative quantitation was calculated by the 2-DDCt method and normalized to b-actin. The primer sequences were provided as follows: AGT forward, 50 -CCCCAGTCTGAGATGGCTC-30 AGT reverse, 50 -GACGAGGTGGAAGGGGTGTA-30 b-actin forward, 50 -ACAGAGCCTCGCCTTTGCCGAT-30 b-actin reverse, 50 -CTTGCACATGCCGGAGCCGTT-30 5.5.5.4 Statistical analysis Statistical analysis was performed using SPSS 22.0. Statistical comparison was analyzed by Student’s t-test or one- or two-way ANOVA followed by a post hoc test. P-values less than 0.05 were considered significantly different. All results were acquired from at least three independent experiments.

5.5.6 Results 5.5.6.1 The influence of high glucose on cell proliferation We first set out to evaluate the potential influence of high glucose on proliferative index in breast cancer cells. Cell proliferation was determined by cell counting under normal culture conditions with different concentration of glucose ranging from 5.5, 25 to 50 mM. Our data clearly showed that supplementation with 25 mM of glucose significantly promoted proliferation of both MCF-7 and MDA-MB-231 cells, in comparison with 5.5 mM glucose in the normal culture medium (Figure 5.5.1A, B), which implicated a critical role of high blood glucose in the tumor biology of breast cancer. We further confirmed this phenotype in both cell lines using the MTT assay. As shown in Figure 5.5.1C, D, cell viability was measured at different time points upon high glucose treatments, and the result was definitely in support of the pro-proliferative effect of high glucose on cell growth. The anchorage-independent growth of breast cancer cells was analyzed by soft agar assay as well. The colony number of MCF-7 and MDA-MB was evidently increased in high glucose medium in comparison with low glucose counterparts (Figure 5.5.1EeH), and the representative images were provided. Our results unambiguously suggested that high glucose critically contributed to the growth of breast cancer cells.

Glucose deprivation and fasting strategies

173

Figure 5.5.1 Influence of high glucose on cell proliferation. (A, B) Proliferation of MCF7 (A) and MDA-MB-231 (B) cells exposed to medium with glucose concentrations varying from 5.5 to 50 mM was determined by cell counting assay. (C, D) Proliferation of MCF7 (C) and MDA-MB-231 (D) cells exposed to medium with glucose concentrations of 5.5 or 25 mM was determined by MTT assay. (E, F) Proliferation of MCF7 cells exposed to medium with glucose concentrations of 5.5 or 25 mM was determined by soft agar assay. (G, H) Proliferation of MDA-MB-231 cells exposed to medium with glucose concentrations of 5.5 or 25 mM was determined by soft agar assay. Data are shown as mean  S.D. **P